These Redon

MUSEUM NATIONAL D’HISTOIRE NATURELLE
Ecole Doctorale Sciences de la Nature et de l’Homme – ED 227
Année 2008 N°attribué par la bibliothèque

|_|_|_|_|_|_|_|_|_|_|_|_|
THESE
Pour obtenir le grade de
DOCTEUR DU MUSEUM NATIONAL D’HISTOIRE NATURELLE

Spécialité : ECOLOGIE
Présentée et soutenue publiquement par
Louis de REDON de COLOMBIER

Le 12 décembre 2008
INTERETS ECOLOGIQUES DES BORDS DE ROUTE

EN MILIEU AGRICOLE INTENSIF
Sous la co-direction de : Nathalie MACHON, Professeur,

et Frédéric JIGUET, Maître de conférences
Composition du jury :
Pr. Patrick BLANDIN Professeur MNHN, Paris, FRANCE Président
Pr. Thierry DUTOIT Professeur, IUT d’Avignon, FRANCE Rapporteur
Dr. Françoise ROZE Maître de conférences, HDR, Université de Rennes 1, FRANCE Rapporteur
Dr. Sabine BIELSA Chargée d'études Milieux naturels - Faune – Flore, SETRA, Paris, FRANCE Examinatrice
Dr. Päivi TIKKA Research coordinator, Université d’Helsinki, FINLANDE Examinatrice
Pr. Nathalie MACHON Professeur MNHN, Paris, FRANCE Co-directrice de Thèse
Dr. Frédéric JIGUET Maître de conférences, HDR, MNHN, Paris, FRANCE Co-directeur de thèse
THESE
DE DOCTORAT
DU
MUSEUM NATIONAL D’HISTOIRE NATURELLE
SPÉCIALITÉ : ÉCOLOGIE
INTERETS ECOLOGIQUES DES BORDS DE ROUTE EN MILIEU AGRICOLE INTENSIF
Présentée par
Pour obtenir le grade de

DOCTEUR EN SCIENCE DU MUSEUM NATIONAL D’HISTOIRE NATURELLE
Soutenue le 12 décembre 2008 devant un jury composé de :
Patrick BLANDIN Professeur MNHN Président

Thierry DUTOIT Professeur Université d’Avignon Rapporteur
Françoise ROZE Maître de conférences Université de Rennes Rapporteur
Sabine BIELSA Docteur SETRA Examinatrice
Päivi TIKKA Docteur University of Helsinki Examinatrice
Frédéric JIGUET Maître de conférences MNHN Directeur de thèse
Nathalie MACHON Professeur MNHN Directrice de thèse
Equipe d’accueil : U.M.R. 5173 « Conservation des Espèces, Restauration et Suivi des Populations »
Ecole doctorale : E.D. 227 « Sciences de la Nature et de l'Homme »
Ph.D. THESIS
OF THE
FRENCH NATIONAL MUSEUM OF NATURAL HISTORY
DISCIPLINE : ECOLOGY
ECOLOGICAL INTERESTS OF ROADSIDE VERGES IN INTENSIVE AGRARIAN LANDSCAPES
Defended by
To obtain the academic degree of:

PHILOSOPHIÆ DOCTOR OF THE FRENCH NATIONAL MUSEUM OF NATURAL HISTORY
Defended on December 12th, 2008 in front of a scientific jury composed by:
Patrick BLANDIN Professor MNHN Président

Thierry DUTOIT Professor University of Avignon Reviewer
Françoise ROZE Assistant professor University of Rennes Reviewer
Sabine BIELSA PhD Doctor SETRA Examiner
Päivi Marjatta TIKKA PhD Doctor University of Helsinki Examiner
Frédéric JIGUET Assistant professor MNHN Thesis supervisor
Nathalie MACHON Professor MNHN Thesis supervisor
Research Lab: U.M.R. 5173 « Conservation des Espèces, Restauration et Suivi des Populations »
College: E.D. 227 « Sciences de la Nature et de l'Homme
Gravures issues de Mitchell-Jones et al. (1999)
« Nous avouerons que notre héros était

fort peu héros en ce moment. Toutefois, la
peur ne venait chez lui qu’en seconde
ligne ; il était surtout scandalisé de ce
bruit qui lui faisait mal aux oreilles ».
STENDHAL
LA CHARTREUSE DE PARME (1839)
-I-
PREAMBULE
Cette thèse de doctorat a été réalisée sous la codirection de Nathalie MACHON (Pr.) et de
Frédéric JIGUET (Maître de conférences) au sein de l’U.M.R. 5173 M.N.H.N/C.N.R.S./
U.P.M.C. « Conservation des Espèces, Restauration et Suivis des Populations » dirigée par
Denis Couvet (Pr.)
Le financement a été assuré par convention signée dans le cadre de la Stratégie Nationale
pour la Biodiversité (www.ecologie.gouv.fr/-Strategie-nationale-pour-la-.html) entre le
Muséum national d’Histoire naturelle et la Direction Général des Routes (D.G.R.) du
Ministère de l’Ecologie, de l’Energie, du Développement Durable et de l’Aménagement du
Territoire (M.E.E.D.D.A.T.)
L’ensemble des ressources d’information géographiques et cartographiques utilisées a été

fourni par l’Institut d'Aménagement et d'Urbanisme de la Région Île-de-France (I.A.U.R.I.F.),
la Direction Départementale de l’Equipement de Seine-et-Marne (D.D.E. 77) et la Direction
Générale des Routes (D.G.R.)
Dans le cadre de l’Atlas de la biodiversité de Seine-et-Marne réalisé par le laboratoire, le

Conseil Général 77 a fourni l’aide nécessaire à la réalisation des travaux de terrain aux côtés
de la D.D.E. 77 (balisage et sécurisation des travaux en bords de route).
- II -
REMERCIEMENTS
Mes remerciements vont tout d’abord à l’ensemble des membres de mon jury : à mes deux
rapporteurs qui ont accepté cette lourde tâche, Thierry DUTOIT et Françoise ROZE, et aussi à
mes examinateurs Sabine BIELSA et Patrick BLANDIN et Päivi TIKKA.
Il est difficile de penser à tous ceux qui ont permis l’écriture de cette thèse après tant d’années
d’étude. Sans rechercher aucunement l’exhaustivité, avec l’excuse de derniers moments de
rédaction toujours difficiles auprès des personnes omises, je tiens à remercier spécialement :
Mes deux co-directeurs de thèse, Frédéric JIGUET et Nathalie MACHON, ainsi que Denis
COUVET, directeur du laboratoire, qui sont les trois personnes qui m’ont le plus entouré lors
de mes travaux de recherches avec les conseils et les critiques toujours opportuns.
Les membres de mon comité de thèse pour l’attention portée à mes travaux durant trois
années et les conseils donnés : Sabine BIELSA, Sébastien FILOCHE, Marie-Elise ILHAT, Jane
LECOMTE et Christophe PINEAU.
Les responsables de l’Ecole Doctorale 227, Madiha ABADA, Mahjouba FASSA et Guillaume
LECOINTRE qui m’ont permis de mener à bien cette thèse mais aussi mon M2 de droit en
l’intégrant à mon parcours de doctorant.
Mes professeurs de classes préparatoires qui ont su avec toute la pédagogie nécessaire
m’emmener là où je ne serai jamais parvenu sans leur précieuse aide : Bruno ANSELME, Remi
DEJEAN de LA BATIE, Martine GINESTET, Eric PERILLEUX, Annick ROUGEON et Denise
TOBAILEM, en particulier.
Mes professeurs de droit qui m’ont permis de réaliser le M2 de droit sur deux années et qui
ont encadré mon mémoire de fin d’étude sur « L’impact des enjeux liés à la protection de la
biodiversité sur l’évolution du droit de la propriété foncière non-bâtie » : Yves JEGOUZO,
Jacqueline MORAND-DEVILLER, Jacques-Henri ROBERT et François-Guy TREBULLE.
- III -
Aurélie GARNIER, Jane LECOMTE (re-) et Agnès RICROCH pour l’encadrement de mon stage de
M2 qui m’a initié à la recherche et m’a permis de décrocher la thèse.
Tous les stagiaires que j’ai encadrés durant ces trois années et sans qui autant de travail
n’aurait jamais pu être accompli : Jérémy CASTELLI, Marion CHARME, Elise CONTAN, Amélie
DELERUE, Rose-Line PREUD’HOMME, Johanna ROILLET, Hélène SCHERNBERG, Noémie
VARET et Arnaud WINTHER.
Toute l’équipe du C.E.R.S.P. avec qui j’ai vécu trois années fort riches, studieuses mais aussi
sympathiques : Jean-Claude (regard noir ?), Cindy (combien de sous me restent-ils ???),
Karine (et mon programme de saisie semi-automatique ???), Yves (depuis Fénelon quand
même…), Patricia (sans certificat administratif), François C (le roi du confeti !), Guillaume
(futur thésard sur l’étude de l’effet « pieds carrés » sur les résultats du P.S.G. ?), Joanne (que
je n’ai pas encore vu danser malgré la cotise !), Jocelyne (sur ma convention),Olivier
(Argggggggg !), Florence (vigilance s…paf !), Vincent (invisible man in Montpellier),
Aggeliki (malgré le refus de partenariat entre pélicans et bords de route…), Cécile (qui sait
donner le top du déjeuner, ça compte !), Ondine (l’ambiance de la salle), Denis G (KVO in
memoriam), Pierre (que je dois encore plaquer pour la forme), Sophie (promis, je m’inscris au
module R de Marseille), Axelle (spécialiste émérite de l’épreuve désormais olympique du
parcours administratif dont elle est la détentrice du record de vitesse), Romain J (avec ses
interventions télévisées qui me permettaient d’expliquer où je bossais…), Christian (Mister
JECONNAISTOUT-ENTAXONOMIE, impressive 4 ever), Grégoire (seigneuralement),
Romain LORILLIERE (pétanque ? fléchettes ? bière ?), Isabelle (je ne sais par où commencer !),
Harold (à l’orée d’un article qui va tout faire sauter !), Laeticia (en ex-voisin), Jean-Baptiste
(et le module G’TEM ? non B’DEM !!! Aie aie aie : inconscient !), Jean-Pierre (fournisseur
officiel de super-virus !), Audrey (je te proclame « reine des carex »), Alzira (chegarei a falar
portugues), Emmanuelle (comment réécrire une communication de thésard à la peine en
article Nature déjà ?), Maëlle (faudra bien se faire promenade en canasson à la fin !),
Alexandre (qui m’a enlevé toute illusion sur le Codex…), Céline T (on se comprend à la
pétanque), Fabien (malgré le test à deux balles sur les bourdons !), Maxime (et les bonnes
bouteilles), et sans oublier… Mélissandre et Auxence (et la pie) :o)
- IV -
Tous mes amis qui ont su rendre, malgré tout, Paris plus supportable, ou m’ont toujours gardé
une petite place chez eux : Adrien et Camille , Alexeï (un vrai chercheur !), Béné et Antoine
(entre pois et bd), Bini (évidement), Boli (faut toujours faire Pâque à montrésor !), Clèm et
Clémentine (de chevalier à chevalier), Dimitrios (une passion commune : la grande bleue),
Emmanuel (mes pas dans les tiens), Fabouchou et Yaëlle (les Bretons au soleil ???), Gaëlle
(voisine version pic-nic), Gigi et Sophie (pour une dent avec toi, je ferai n’importe quoi…),
Giovanni (à sao paolo promis !), Hindounette (princesse des mille et une nuits), K-Kate (qui
pue la f… à plein nez), Julio (mlle ?), Loulou et Jojo (je n’aurai pas fais que des c… dans ma
vie), Laurette (que dire ?), Loulou H (et ses 110 kg de bagages dans la 106 dont 5 paires de
chaussures pour un roadtrip de 10 jours dans les Balkans…), Marilia et Rodrigo (il faudra
remettre Torres), Matthias (malgré le coup du roquefort), Matthias (Night fever &
Roquefort !), Maxou et Armelle (entre artistes), Nico L (de coup de boule en coup de boule on
en arrive là), Nico et Véro (et tutur le rocker !), Olive (dit le super-cousin), Pakou et Blanche
(soounement je te remercie), PH et Marion (« ph » comme... ?), Rico et Erika (pouvaient pas
faire mieux), Sabine et Filou (Berry power in paris), Seb et Stéph (les ptis greffons), Tom
(voleur !), Tonio (depuis la Roumanie ? douce plaine verte…), Yoyo (fifon) et Yves (tu me
masseras, promis !).
Enfin, ma famille pour le soutien affectif, financier (dure vie de thésard !), logistique et
incandescent : François, Boubou, Maman, Mamie, Mimine, Moumoune, Papa et Sonia.
Sans oublier Riri, Fifi et Loulou (re-) !!! Pas ceux des fameuses Mystérieuses Cités d’Or
même si Esteban, Tao et Zia conservent malgré tout une place à part dans mon cœur… Mais
au fait ! Elle (il ?) est où Loulou ? Loulou ?! Loulou !!!
-V-
SOMMAIRE
p. III PREAMBULE
p. IV REMERCIEMENTS
p. VII SOMMAIRE
p. 1 INTRODUCTION
p. 2 § A. INFRASTRUCTURES ROUTIERES ET CRISE DE LA BIODIVERSITE

p. 2 (1) Ampleur et réalité de la crise
p. 3 (2) Infrastructures routières et causes de la crise
p. 4 § B. ETUDE DE LA NATURE ORDINAIRE ET BIOLOGIE DE LA CONSERVATION

p. 4 (1) Conservation de la biodiversité et Nature ordinaire
p. 5 (2) Développement de méthodes de suivis de la Nature ordinaire
p. 6 § C. LES BORDS DE ROUTE, ZONES CREATRICES D’HABITAT ET DE CONTINUITE POUR

LA BIODIVERSITE ?
p. 7 (1) Bords de route et création d’habitats en paysage agricole intensif
p. 8 (2) Bords de route et création de continuités en paysage agricole intensif
p. 9 § D. IMPACTS DES MODES DE GESTION ET DE LA STRUCTURE DES BORDS DE ROUTE

p. 9 SUR LA BIODIVERSITE
p. 10 (1) Fauchage des bords de route et diversité végétale
(2) Structure des bords de route et dynamique d’un campagnol (M. arvalis)
p. 10 § E. ANNEXES A LA THESE
p. 11 (1) Manuscrits complémentaires
p. 11 (2) Encadrés et Guide technique
p. 12 MATERIELS & METHODES
p. 13 § A. SITES D’ETUDE
p. 13 (1) Occupation des sols en Ile-de-France (2003)
p. 14 (2) Sites d’étude en bords de route (Seine-et-Marne)
p. 16 § B. STRUCTURES ET EMPRISES DES ROUTES AU SEIN DES PAYSAGES AGRICOLES

p. 16 (1) Importance du réseau routier en Ile-de-France
p. 17 (2) Bords de route en Ile-de-France
p. 19 § C. RELEVES DE VEGETATION
p. 19 (1) Quadrats de relévé et réplication
p. 20 (2) Base de donnée GALIUM
- VI -
p. 21 § D. RELEVES FAUNISTIQUES
p. 21 (1) Pot-pièges « Barber »
p. 22 (2) Réplication des relevés
p. 23 § E. ANALYSES DE SOL
p. 23 (1) Echantillonnage du sol
p. 23 (2) Eléments analysés
p. 24 PARTIE I
BIODIVERSITE DES BORDS DE ROUTE EN MILIEU AGRICOLE INTENSIF :

INTERETS LOCAUX ET IMPACTS SUR LES MILIEUX ADJACENTS
p. 25 § A. MANUSCRIPT N°1:
Plant diversity in agrarian landscapes:
Ecological roles of roadside verges as refugia and corridors;
Submitted to Acta Oecologica.
p. 43 § B. MANUSCRIPT N°2:
Road network in intensive agricultural landscape:
Refuges, corridors or barriers for small mammals?;
Submitted to Landscape Ecology.
p. 58 PARTIE II
STRUCTURES ET MODES DE GESTION DES ACCOTEMENTS ROUTIERS :

IMPACTS SUR LA BIODIVERSITE
p. 59 § A. MANUSCRIPT N°3:
Effect of delayed single mowing on roadsides vegetation communities:
A three years study;
Submitted to Journal of Environnemental Management.
p. 71 § B. MANUSCRIPT N°4:
Possible effects of roadside verges on vole outbreaks
in an intensive agrarian landscape;
In revision, Mammalian Biology.
p. 77 DISCUSSION ET CONCLUSION GENERALES
p. 78 § A. EFFETS DES PROTOCOLES UTILISES ET DES GROUPES SUIVIS

p. 78 (1) Relevés floristiques
p. 80 (2) Relevés faunistiques
p. 82 § B. IMPACTS NEGATIFS DES ROUTES SUR L’ENVIRONNEMENT

p. 82 (1) Destruction et fragmentation des habitats
p. 84 (2) Dispersion d’espèces exotiques
- VII -
Ecologie des bords de route
p. 84 § C. IMPORTANCE DES BORDS ROUTES EN MILIEU AGRICOLE INTENSIF

p. 85 (2) Mise en place de continuités biologiques
p. 85 (1) Mise en place de zones habitats et refuges
p. 86 § D. IMPACTS DES STRUCTURES ROUTIERES

p. 87 (1) Effets de la largeur de l’emprise
p. 89 (2) Effets des plantations
p. 90 § E. IMPACTS DES MODES DE GESTION

p. 91 (1) Impacts de la fauche
p. 93 (2) Impacts de l’exportation des déchets
p. 94 PERSPECTIVES
p. 95 § A. POURSUITE L’ETUDE DES EFFETS DES ROUTES SUR LA BIODIVERSITE
p. 95 § B. COMMUNICATION ET VULGARISATION DES SAVOIRS SCIENTIFIQUES
p. 97 ANNEXES
p. 98 § A. MANUSCRIPT N° 5:
Effects of local environment and human activities on plant communities
of field marginsin an intensive agrarian landscape;
In preparation.
p. 111 § B. MANUSCRIPT N° 6:
A three-year study of weed management
on plant community in field margins in an openfield landscape;
Submitted to Weed Research.
p. 127 § C. MANUSCRIPT N° 7:
Plant and spider communities benefit differently
from the presence of planted hedgerows in highway verges;
Published in Biological Conservation, June 2008.
p. 139 § D. ENCADRES
p. 140 (1) Encadré n°1 : Etendue de l’impact écologique des routes
p. 142 (2) Encadré n°2 : ComDyn logiciel de dynamique des communautés
p. 144 (3) Encadré n°3 : Stratégie Nationale pour la Biodiversité
p. 145 BIBLIOGRAPHIE GENERALE
- VIII -
INTRODUCTION
“Two roads diverged in a wood, and I —

I took the one less travelled by,
And that has made all the difference”.
(« Deux routes se séparaient dans un bois,

Et moi, je pris la moins empruntée.
Cela fit toute la différence »).
ROBERT FROST
THE ROAD NOT TAKEN – 1916.
-1-
§ A. INFRASTRUCTURES ROUTIERES ET CRISE DE LA BIODIVERSITE
Pourquoi la biodiversité est-elle en crise ? Quelle est la part de responsabilité des routes
dans cette crise et en quoi cela est-il un problème si important en 2008 ? En ce début de
XXIème siècle, la question des routes devient un sujet de plus en plus important en écologie
du fait (a) de leur contribution majeure aux changements globaux qui affectent actuellement
les écosystèmes terrestres, et (b) de la prise de conscience collective des enjeux écologiques
matérialisée par l’intérêt croissant porté aux questions environnementales au sein de notre
société (Angold 1997, Spellerberg 1998, Forman 2000, Ashenden et al. 2003). En effet, les
infrastructures routières ont des impacts très importants sur l’environnement (Forman et
Alexander 1998, Trombulak et Frissell 2000, Forman et al. 2003, Coffin 2007) et sont par là-
même fortement impliquées dans les processus menant aux changements globaux, de
manière générale, et à la crise actuelle de la biodiversité, plus particulièrement.
(1) Ampleur et réalité de la crise

La diversité biologique est composée de trois niveaux : diversité des écosystèmes, diversité
spécifique et diversité génétique (ou intra-spécifique). La crise que traverse aujourd’hui la
biodiversité touche durement ses trois niveaux :
• Au niveau des écosystèmes : une homogénéisation biotique, tant sur le plan
taxonomiques que fonctionnel, induisant le remplacement progressif d’écosystèmes
riches et complexes par des écosystèmes simplifiés pouvant rendre, entre autre, de
plus faibles services écosystémiques (Chapin et al. 2000, Díaz et Cabido 2001) ;
• Au niveau des espèces : des taux d’extinction si élevés que devant l’ampleur du
phénomène, certains auteurs ont été amenés à évoquer une sixième grande crise
d’extinction (Leakey et Lewin 1995). Bien qu’un temps controversé (Mann 1991,
Lomborg 2001), ce phénomène est désormais largement admis au sein de la
communauté scientifique (Levin et Levin 2002) ;
• Au niveau génétique : une réduction des populations naturelles de nombreuses
espèces (Lacy 1997) induisant une baisse de la variabilité intra-spécifique et la perte
d’allèles qui pourraient jouer un rôle dans l’adaptation potentielle et future des espèces
aux nouvelles donnes environnementales.
-2-
(2) Infrastructures routières et causes de la crise

Les modifications majeures apportées aux paysages au cours du XXème sont étroitement liées
à la construction des routes, à leur utilisation et à leur gestion (Bennett 1991, Noss et
Coorperrider 1994). Les impacts des routes sur la biodiversité sont nombreux (Sherwood et al.
2002) :
• Les routes participent aux mécanismes menant à la surexploitation des milieux
car elles sont construites pour desservir ou promouvoir l’agriculture, les industries et
les commerces qui nuisent à la flore et à la faune locales (Van Dyke et al. 1986,
Karnefelt et Mattson 1989, Seibert 1993, Allan et Flecker 1993, Roth et al. 1996) ;
• Les routes participent à la dégradation des habitats naturels car elles
consomment des espaces (Sherbune 1985, Reijnen et Foppen 1994, Angold 1997). Ces
surfaces sont loin d’être négligeables à l’échelle du paysage. Les routes participent à la
fragmentation des paysages qui a de nombreux effets sur la biodiversité notamment
sur les dynamiques des populations végétales (Timmins et William 1990, Brothers et
Spingarn 1992) et animales (Swihart et Slade 1984, Manserfgh et Scott 1989, Rich et
al. 1994). Par ailleurs, les routes, soit suivent l’urbanisation des territoires, soit la
permettent en améliorant l’accessibilité de certains espaces (Seibert 1993) qui sont
alors consommés indirectement du fait du développement d’infrastructures de
transports (Cramer et Hopkins 1982, Harris et al. 1996). Dans les espaces naturels,
cette accessibilité des territoires améliorée permet aussi une augmentation de la
pression des chasseurs (Bennett 1991, Robinson et Bodmer 1999) et des visiteurs qui
endommagent les communautés végétales (Matlack 1993) et perturbent les animaux
(Witmer et DeCalesta 1985, Pedevillano et Wright 1987, Del Campo et al. 1990,
Czech 1991) ;
• Les routes favorisent les invasions biologiques car leurs accotements permettent
la mise en place de continuités biologiques au sein des paysages ; continuités qui
favorisent la dispersion d’espèces exotiques (Wester et Juvik 1983, Henderson et
Wells 1986, Tyser et Worley 1992, Wein et al. 1992, Greenberg et al. 1997, Parendes
et Jones 2000). Cela est aussi vrai pour certaines plantes maritimes qui se retrouvent
au cœur des territoires à cause des salages hivernaux (Scott et Davinson 1982). Des
expansions de maladies (Dawson et Weste 1985, Gad et al. 1986) et d’insectes
(Pantaleoni 1989) sont aussi observées ;
-3-
• Les routes peuvent provoquer des extinctions locales car elles constituent de
véritables barrières physiques à la dispersion de certaines espèces au sein des paysages
(fragmentation) : insectes (Bhattacharya et al. 2003), petits mammifères (Oxley et al.
1974), grands mammifères (Nellemann et al. 2001) et oiseaux (Devely et Stouffer
2001). Cet effet barrière peut notamment trouver sa source dans les nombreuses
collisions entre véhicules et faune (Harris et Scheck 1991, Mazerolle 2004). Les
collisions peuvent être considérées la première cause de mortalité chez certaines
espèces (Forman et Alexander 1998).
Par ailleurs, les routes sont aussi sources de rejets ayant des impacts sur la biodiversité
comme les métaux lourds (Goldsmith et al. 1976, Dale et Freedman 1982, Leharne et al.
1992), les molécules organiques (Benfenati et al. 1992), l’ozone (Flueckiger et al. 1984), les
nitrogènes (Carslaw 2005, Truscott et al. 2005) et les sels (Bogemans et al. 1989, Molles et
Gosz 1980, Hoffman et al. 1981, Peters et Turk 1981, Mattson et Godfrey 1994).
Tous ces polluants se répandent ensuite à travers le paysage et se retrouvent à
plusieurs niveaux dans les sols (Byrd et al. 1983, Indu et Choudhri 1991), dans les eaux
(Gjessing et al. 1984) ainsi que dans les tissus des plantes (Datta et Gosh 1985, Beslaneev et
Kuchmazonokova 1991) et des animaux (Collins 1984, Birdsall et al. 1986, Grue et al. 1986).
Les routes entrainent aussi l’émissions de poussières (Farmer 1993, Auerbach et al. 1997) et
de bruit (Brumm 2004), une compaction des sols (Helvey et Kochenderfer 1990) et une
modification de conditions microclimatiques locales (Asaeda et Ca 1993).
Par ces modifications importantes de l’environnement, les routes perturbent la
croissance des végétaux (Sarkar et al. 1986, Kammerbauer et al. 1986, Thompson et Rutter
1986, Spencer et Port 1988, Angold 1997, Viskari et Kerenlampi 2000, Singh et al. 2003) et
les taux de germination de leurs graines (Spence 1988). Les routes induisent des changements
comportementaux chez les animaux : insectes (Alstad 1982, Mader 1984), gastéropodes (Baur
et Baur 1990), batraciens (Van der Zande et al. 1980), petits mammifères (Adams et Geis
1983, Mader 1984, Bakowski et Kozakiewicz 1988, Merriam et al. 1989, Korn 1991, Burnett
1992), grands mammifères (Murphy et Curatolo 1987, Brody et Pelton 1989) et oiseaux
(Canaday 1996, Reijnen et al. 1997, Weiserbs et Jacob 2001, Forman et al. 2002).
Enfin, les effets de ces pollutions diminuent avec l’éloignement à la route (Gjessent et
al. 1984, Thompson et al. 1986, Post et Beeby 1996, Truscott et al. 2005), la avec la réduction
du trafic routier (Goldsmith et al. 1976, Dale et Freedman 1982, Leharne et al.1992) et avec le
temps (Byrd et al. 1983, Tong 1990) mais de manière plus relative (Vora 1988).
-4-
§ B. INTERET DES BORDS DE ROUTE POUR LA BIOLOGIE DE LA CONSERVATION
« Quelle biodiversité conserver ? », « Comment la quantifier ? », sont des questions

importantes sur lesquelles les écologues travaillent depuis de nombreuses années. En effet,
la lutte contre la crise de la biodiversité est un impératif pressant compte tenu de l’urgence de
la situation et de l’importance de sa conservation. Dans ce contexte, si la problématique liée à
la conservation d’espèces menacées d’extinction est relativement connue, celle liée à la
conservation de la « Nature ordinaire » est davantage ignorée alors que capitale en termes
notament d’ingénierie écologique.
(1) Conservation de la biodiversité et Nature ordinaire

Dès le XIXème siècle, une prise de conscience de la nécessité de protéger la nature apparaît et,
en 1932, le peintre américain George CATLIN propose la création d'un « parc contenant
hommes et bêtes dans toute la beauté sauvage de leur nature ». Cette approche romantique de
la Nature se traduit à deux niveaux : (a) par la création de parcs (le premier parc naturel est le
Yellowstone National Park en 1872) et (b) par la mise en place de législations protectrices
d’espèces considérée en danger. Les efforts de conservation se focaliseront donc sur certaines
zones et certaines espèces. De nombreux débats scientifiques porteront sur ce sujet
(Simberloff et Abele 1982, McNeill et Fairweather 1993, Virolainen et al. 1998).
Si les disparitions de milieux naturels et d’espèces constituent des pertes irréversibles,
et que les empêcher doit être une priorité, la conservation de la Nature dans sa globalité doit
aussi intégrer les préoccupations des scientifiques et des politiques. L’idée d’une nécessaire
conservation des milieux naturels et des espèces menacées, à travers la mise « sous cloche »
d’une partie seulement de la Nature, est aujourd’hui dépassée car trop étroite : changer les
modes de gestion au sein de nos paysages devient un impératif sous peine de ne pouvoir
régler les problèmes de la biodiversité à long terme dans un contexte où les changements
globaux vont conduire à une importante, et désormais inéluctable, réorganisation (Génot et
Barbault 2004) car des déplacements d’aire de répartition et des extinctions locales d’espèces
sont à prévoir (Parmesan et Yohe 2003, Thuiller et al. 2005). Dans ce contexte, les zones
protégées apparaissent comme des îlots au sein d’une matrice constituée d’écosystèmes
anthropisés, incapables de protéger efficacement la biodiversité dans sa globalité (Franklin
1993), qui peuvent être regroupés sous le terme de « Nature ordinaire » (Abadie 2008).
-5-
(2) Développement de méthodes de suivis de la Nature ordinaire

La loi du 10 juillet 1976 relative à la protection de la Nature impose certes la conservation
d’une nature « remarquable » mais, dans le même temps, elle invite aussi implicitement à une
réflexion sur les moyens d’organiser une gestion rationnelle du patrimoine naturel dans sa
globalité comprennant aussi la Nature ordinaire (Blandin 1986). « Que mesurer ? » et
« Comment mesurer ? » (Yoccoz et al. 2001) sont donc deux questions actuelles capitales en
écologie pour permettre une meilleure prise en compte de cette Nature ordinaire difficilement
perceptible par la société civile. Dans ce contexte, la recherche d’indicateurs est devenue une
des priorités des écologues (Pearson 1994, Ferris et Humphrey 1999, Lindenmayer 1999,
Lindenmayer et al. 2000, Duelli et Obrist 2003, Dauber et al. 2003).
En effet, la biodiversité a souvent été le parent pauvre des politiques
environnementales car regardée comme un concept trop vague pour être intégré au monde
économique réel et aux processus de décision. Cependant ce problème peut être corrigé si de
bons outils sont développés pour dépasser la complexité et les multiples niveaux de la
biodiversité en offrant une information synthétique et de qualité. Il s’agit de trouver des
indicateurs par l’étude de certains groupes taxonomiques bien définis (ex : oiseaux et projet
STOC) dont les qualités essentielles doivent être (a) une bonne connaissance de leurs
taxonomies, (b) une collecte facile des informations relatives à leurs communautés et
populations, et (c) une diversité réelle existante en leur sein (Sutton and Collins 1991). Les
indicateurs de biodiversité doivent fournir une information à la fois synthétique,
scientifiquement rigoureuse et socialement transparente de l’état de la biodiversité (Couvet et
al. 2005).
C. LES BORDS DE ROUTE, ZONES CREATRICES D’HABITAT ET DE CONINUITE POUR LA
BIODIVERSITE ?
Au sein des paysages d’agriculture intensive, les routes peuvent-elles jouer un rôle dans la
lutte contre l’érosion de la biodiversité à travers la création de zones d’habitats et de
continuités pour la nature ordinaire ? L’intensification de l’agriculture, issue des années 50
et caractérisée par la suppression des haies, le remembrement, la mécanisation, l’utilisation de
produits chimiques (Pain et Pienkowski 1997), a eu d’importants effets sur l’environnement
(Burel et al. 1998) et la biodiversité a beaucoup reculé au sein des paysages agricoles (Altieri
-6-
1999, Le Cœur et al. 2002). Cela a pu se traduire, par exemple, par des crashs dans les
populations d’oiseaux (Donald et al. 2000). Or le maintien de la biodiversité en paysage
agricole permet le maintien de nombreux services écosystémiques (Altieri 1999) comme la
stabilisation des populations d’insectes ravageurs (Andow 1991), le maintien de la fertilité et
de la qualité des sols (Hendrix et al. 1990, Paoletti et al. 1994), ou encore la limitation des
pollutions liées aux métaux lourds émis depuis les routes (Dochinger 1980, Greszta 1982).
(1) Bords de route et création d’habitats en paysage agricole intensif

Dans ce contexte, les bords de routes peuvent présenter une opportunité pour le maintien de la
biodiversité. Les dépendances vertes des infrastructures de transport, dont les surfaces
représentent des chiffres non négligeables à l’échelle du paysage, contiennent une biodiversité
importante (Bennett 1991, Sherwood et al. 2002) et, avec les bords de champs (Marshall et
Moonen 2002, Meek et al. 2002, Perkins et al. 2002, Vickery et al. 2002, Woodcock et al.
2005), peuvent qvoir un rôle à jouer en biologie de la conservation au sein des paysages
agricoles. En effet il a été montré que les bords de route peuvent constituer des zones
« habitat » ou « refuge » pour la biodiversité en paysage d’agriculture intensive (Akbar 1997,
Spooner et al. 2004, O’Farrell et Milton 2006). Cela est le cas pour de nombreuses espèces de
plantes (Way 1977, Hansen et Jensen 1972, Wester et Juvik 1983, Sykora et al. 1993),
d’insectes (Samways 1989), de mammifères (Adams 1984) et d’oiseaux (Newbey et Newbey
1987).
Q1. Quelle est l’importance relative des bords de routes comme habitat ou
zone refuge pour la faune et la flore au sein de la matrice agricole des
quelques îlots d’habitats semi-naturels restant ?
Comme un unique taxon ne peut pas porter suffisamment de renseignements pour connaître
l’état des autres (Lovell et al. 2007), il est intéressant de travailler sur plusieurs groupes
taxonomiques (Pearman et Weber 2007) comme les communautés végétales (Schaffers et al.
2002, Gelbard et Belnap 2003, Rentch et al. 2004), les communautés carabiques (Day et al.
1993, Lovei et Sunderland 1996, Atlegrim 1997), les communautés d’araignées (Miyashita
1999, Le Viol et al. 2008), ou les communautés d’oiseaux (Canterbury et Blockstein 1997,
Boulinier et al. 2001). Les approches multi-taxons sont par ailleurs rares alors que nécessaires
(Kotze et Samways 1999). Ainsi, afin de répondre à cette question, l’étude de la biodiversité
-7-
des bords de route et des milieux adjacents (1) a été réalisée à plusieurs niveaux
d’organisation du Vivant (populations et communautés) et (2) a porté sur différents taxons : la
flore vasculaire (Plantae : Spermatophyta), les carabes (Animalia, Insecta, Coleoptera :
Carabidae) et les petits mammifères (Animalia, Mammalia, Rodentia : Muridae, et
Soricophorma : Soricadae).
(2) Bords de route et création de continuités biologiques en paysage agricole intensif

Par ailleurs, si les risques liés aux bords de route comme corridors pour un certain nombre
d’espèces exotiques et/ou invasives sont bien connus (Johnson et Johnson 2004), les
connaissances scientifiques concernant le rôle des bords de route comme corridors pour les
espèces natives restent très fragmentaires et que des études sont fortement espérées sur le
sujet (Bailey 2007). Or il a été montré que dans des habitats naturels de taille réduite le
maintien de certaines populations peut être compromis (Andrén 1994) sauf si la mise en place
de continuités peut assurer l’induction de « rescue effect » (Burkey 1989) et la dispersion
d’individus malgré la matrice agricole (Tischendorf et Farig 2000, Thomas 2000, Bullock et
al. 2002).
Q2.A. Les bords de route permettent-ils la mise en place de continuités

biologiques entre des éléments d’habitats semi-naturels isolés au sein de
la matrice agricole et quelles espèces peuvent profiter de la mise en
place d’une telle connectivité au sein du paysage ?
Si la mise en place de telles continuités a été montrée pour certaines espèces de plantes
(Schmidt 1989, Wilcox 1989), d’insectes (Vermeulen 1994), de mammifères (Getz et al.
1978, Warner 1985) et de batraciens (Seabrook et Dettmann 1996), leurs impacts sur les
milieux adjacents restent relativement méconnus.
Q2.B. Quels peuvent être les impacts des effets « bords de route » sur la
biodiversité des milieux adjacents ? Les bords permettent-ils (1) un
maintien et un enrichissement des communautés adjacentes, ou (2), au
contraire, leur appauvrissement et leur homogénéisation ?
-8-
L’étude des infrastructures routières comme créatrices de continuités biologiques au sein du

payage a aussi été menée à différents niveaux d’organisation (populations et communautés) et
sur plusieurs taxons (plantes et petits mammifères). L’étude des impacts des bords de route
sur les milieux adjacents a quant à elle été réalisée sur la végétation, et plus particulièrement
sur la flore forestière à travers l’étude des communautés végétales de petits bois isolés au sein
de la matrice agricole ou connectés à la route.
La première partie de la thèse, intitulée « Biodiversité des bords de route en milieu

agricole intensif : Intérêts locaux et impacts sur les milieux adjacents », aura pour
objectif de répondre à ces trois questions (Q1., Q2.A. et Q2.B.) à travers deux articles
soumis à Acta Oecologia (Road network in intensive agricultural landscape: Refuges,
corridors or barriers for small mammals?) et à Landscape Ecology (Plant diversity in
agrarian landscapes: Ecological roles of roadside verges as refugia and corridors).
§ D. IMPACTS DES MODES DE GESTION ET DE LA STRUCTURE DES DES BORDS DE ROUTE
SUR LA BIODIVERSITE
Les bords de routes sont des espaces fortement anthropisés. L’homme intervient
régulièrement sur ces espaces pour les entrenir (ex : fauche de la végétation) ou les
restructurer (travaux d’élargissement). Quels sont les impacts de telles mesures de
gestion sur la biodiversité des dépendances vertes des bords de route ? En effet, beaucoup
d’études sur la biodiversité se sont concentrées, jusqu’à un passé récent, sur des inventaires
floristiques ou faunistiques en négligeant le développement de suivis annuels, multi-espèces
et multi-sites sur le long terme (Rich et Woodruff 1992). Des suivis de végétation et de petits
mammifères ont donc été mis en place en bords de route durant les trois années de thèse.
(1) Fauchage des bords de route et diversité végétale

Les bords de route sont gérés de manière intensive en Seine-et-Marne, i.e. la végétation y est
fauchée trois fois par an.
Q3. Quels sont les impacts de tels modes de gestion sur la diversité végétale
des bords de route ? Quelles seraient les conséquences d’un passage de
trois fauches annuelles à une seule en termes de diversité spécifique et
fonctionnelle pour les communautés végétales des bords de route ?
-9-
(2) Structure des bords de route et dynamique d’un campagnol (M. arvalis)
Les bords de route réprésentent des surfaces plus ou moins importante selon les sites étudiés.
L’importance du réseau routier et de ses dépendances vertes n’est cependant pas sans
conséquence sur la répartition et la dynamique de la biodiversité au sein du paysage agricole.
Il a ainsi été montré que les bords de route pouvaient partiper à la stabilisation des populations
de campagnols des champs (Microtus arvalis) (Meunier et al. 1997), petit rongeur connu pour
ses explosion démographiques cycliques (Briner et al. 2007) préjudiciables à l’agriculture
(Delattre et al. 1999).
Q4. Dans quelle mesure et à quelles conditions, les bords de route peuvent-
ils participer à un contrôle des fluctuations démographiques cycliques
de campagnols des champs au sein du paysage agricole ?
La deuxième partie de la thèse, intitulée « Structures et modes de gestion des

accotements routiers : Impacts sur la biodiversité », aura pour objectif de répondre à ces
deux dernière questions (Q3. et Q4.) à travers deux articles : un soumis à Journal of
Environmental Management (Effect of delayed single mowing on roadsides vegetation
communities: A three years study) et un autre en révision chez Mamalian Biology
(Possible effects of roadside verges on vole outbreaks in an intensive agrarian landscape).
§ F. ANNEXES A LA THESE
Quelle peut-être la portée de la thèse située dans un contexte scientifique plus large et
diffusée auprès de professionnels de la route ou de l’environnement ? La conservation de la
biodiversité en milieu agricole est un vaste sujet qu’il est difficile de traiter à travers une
simple thèse. Les annexes apportent quatre textes qui me semblent importants pour la
compréhension des enjeux liés à cette thématique de la nature ordinaire au sein des espaces
anthropisés : ouverture à la problématique des autoroutes et des bandes enherbées autour des
champs ; transmission des savoirs au plus grand nombre et sensibilisation à la conservation de
la biodiversité des non-scientifiques : professionnels de l’environnement, agriculteurs ou éco-
citoyens.
- 10 -
1) Manuscrits complémentaires
Enfin et afin d’enrichir la discussion, trois manuscrits sont proposés en annexes : deux articles
écrits à partir de mes travaux de Masters 2 Recherche, réalisés sein de l’U.M.R. 80 79
« Ecologie, Systématique et Evolution », sur la diversité végétale en milieu agricole et
l’impact des bords de champs et de leur gestion (Effects of local environment and human
activities on plant communities of field marginsin an intensive agrarian landscape et A three-
year study of weed control on plant community in field margins in an openfield landscape) ; et
un article d’Isabelle LE VIOL publié dans Biological Conservation (juin 2008), auquel j’ai été
associé, sur l’impact de la plantation de haies dans les talus autoroutiers sur les communautés
végétales et d’araignées (Plant and spider communities benefit differently from the presence
of planted hedgerows in highway verges).
(2) Encadrés et Guide technique

Les travaux de thèse qui sont ici présentés ont été financés par la Direction Générale des
Routes (http://www.route.equipement.gouv.fr) du Ministère de l’Ecologie, de l’Energie, du
Développement durable et de l’Aménagement du Territoire (http://www.developpement-
durable.gouv.fr) dans le cadre de la Stratégie Nationale de la Biodiversité (S.N.B., cf. Encadré
N°4) à travers une convention passée avec le Muséum national d’Histoire naturelle en 2005.
Cette étude inscrite dans le plan d’action infrastructures de transports terrestres avec pour
double-objectif de faire progresser la recherche scientifique sur la biodiversité (connaissance
de l’écosystème « bords de route », impacts des routes sur les milieux adjacents,
développement d’indicateurs et de méthodologies d’expertises et de suivis) et les politiques en
faveur de la prise en compte de l’Environnement dans la gestion des infrastructures routières
(construction et entretien plus respectueux de la biodiversité) conformément aux engagements
pris par l’Etat dans le cadre de la S.N.B. (http://www.ecologie.gouv.fr/-Strategie-nationale-
pour-la-.html).
Compte-tenu de ce financement, figure aussi en annexe, quatre « encadrés »
complémentaires détaillant certains points de l’introduction.
- 11 -
MATERIELS & METHODES
« Dans la vie, rien n'est à craindre, tout est
à comprendre ».
MARIE CURIE
- 12 -
§ A. SITES D’ETUDE
L’étude de la biodiversité des bords de route a été menée en Ile-de-France où plus de

trois cents sites ont été échantillonnés au sein de différents habitats présents en paysage
agricole intensif (bords d’autoroute, bords de route, bords de champ, champs et îlots
forestiers). L’échantillonnage des bords de route a été centré sur le département de Seine-et-
Marne dans la mesure où le laboratoire CERSP était impliqué dans la création d’un Atlas de
la Biodiversité financé par le Conseil Général de ce département. Chaque question
scientifique posée nécessitant un protocole d’échantillonnage adapté, seul sera présentée ici la
zone d’étude dans sa globalité. Les protocoles d’échantillonnage des sites d’étude sont donnés
dans les articles.
(1) Occupation des sols en Ile-de-France (2003)

Malgré 11,5 millions d’habitants (I.N.S.E.E.) en Ile-de-France, seulement 22.1% des sols
sont concernés par l’urbanisation avec des infrastructures routières représentant près de 10%
de ces espaces urbanisés (cf. Figure N°1 et Table N°1). La région se composent
essentiellement de grands ensembles agricoles (48.6%), ex : Brie et Beauce, et forestiers
(23.4%), ex : vallée de Chevreuse et forêt Fontainebleau (cf. Figure N°2).
Figure N°1 Table N°1

Surface
Occupation du sol (km2) Surface (%)
Zones cultivées 5,722.49 47.64
Zones boisées 2,806.26 23.36
Zones urbaines pavillonnaires 994 8.28
Zones d'activité 660.58 5.5
Zones en herbe 574.85 4.79
Parcs et jardins 469.77 3.91
Emprises routières (hors chemins) 259.16 2.16
Zones urbaines denses 156.1 1.3
Rivières, lacs et étangs 147.04 1.22
Equipements de plein air 111.25 0.93
Zones agricoles intensives 109.5 0.91
Occupation des sols en Ile-de-France : repartition en pourcenbtages

(Figure N°1 et Table N°1) et en surface (Table N°1)
- 13 -
Figure N°2
Occupation des sols en Ile-de-France
(2) Sites d’étude en bords de route (Seine-et-Marne)

Un total de 68 sites a été échantillonné en bord de route en Seine-et-Marne (cf. Figure N°3) à
l’aide du logiciel d’information géographique ArcView 9.0 (SIG) :
a- Pour l’étude des bords de route comme habitat et zones créatrices de continuité :
- 10 sites pour la flore à l’ouest du département (autour de Provins) ;
- 48 pour les petits mammifères sur l’ensemble du département ;
b- 10 sites pour l’étude des impacts de la fauche sur la diversité végétale ;
c- 45 sites pour l’étude des effets de l’importance de la densité des bords de route sur les
fluctuations démographiques de M. arvalis.
- 14 -
Figure N°3
Carte de 68 sites d’étude échantillonnés en Seine-et-Marne (77)
NB : Toutes les données exploitées lors des différentes études ne sont pas issues de ces
travaux de terrain, notamment en ce qui concerne les petits mammifères. Les
données complémentaires ont été obtenues auprès de Christian KERBIRIOU (relevés
en bords de champ et en plein champ), Ondine FILIPPI-CODACCIONI (idem), Olivier
SCHER (relevés en milieu forestier) et Isabelle LE VIOl (relevés en bords
d’autoroutes). Le protocole utilisé entre les différents acteurs était le même et issu
d’une réflexion commune.
- 15 -
§ B. STRUCTURES ET EMPRISES DES ROUTES AU SEIN DES PAYSAGES AGRICOLES
Que sont les bords de route ? Que représentent-ils au sein des paysages agricoles ? Les
infrastructures routières sont souvent accompagnées de dépendances dites « vertes ». Il s’agit
d’une zone, plus ou moins importante, comprise dans l’emprise routière mais recouverte par
de la végétation. Ces « dépendances vertes », ou « bords de route », ont été aménagées le long
comme des zones « tampon » entre l’infrastructure et le milieu adjacent. Ces dépendances,
dont l’importance a été longtemps ignorée, constituent aujourd’hui un véritable enjeu en
termes d’écologie au sein des milieux agricoles.
(1) Importance du réseau routier en Ile-de-France

Les infrastructures routières sont très denses en Ile-de-France (cf. Figure N°4) et représentent
plus 40,000 km (sans tenir-compte des chemins communaux et des voies privées représentant
plus de 45,000 km) pour une région de 12,000 km². La densité de la voirie principale en Ile-
de-France est donc 3.36 km de routes par km².
Figure N°4
Réseau routier principal d’Ile-de-France (2003)
- 16 -
Les infrastructures routières se répartissent ainsi :

• 887 km d’autoroutes, i.e. 2.2% du linéaire total pour une densité de 0.07 km.km2 ;
• 1,568 km de routes nationales, i.e. 3.9% & 0.13 km.km-2 ;
• 8,518 km de routes départementales, i.e. 21.1% & 0.71 km.km-2 ;
• 29,311 km de routes communales, i.e. 72.8% & 2.44 km.km-2.
Les impacts des routes ne se limitent cependant pas à leur consommation d’espaces en termes
de linéaire et à leur environnement proche. En effet, c’est une grande partie des territoires qui
sont impactés directement ou indirectement par les routes (cf. Encadré N°1).
(2) Bords de route en Ile-de-France

En Ile-de-France, la structuration d’une dépendance verte routière est très variable d’une route
à une autre en fonction de son importance (nombre de voies et trafic), de sa situation (zones
urbaines, agricoles et boisées), de sa période de construction et de la nature même de la route
(autoroutes, routes nationales, routes départementales, routes communales et chemins) ;
faisant varier son emprise de rien du tout à plusieurs dizaines de mètres. Cependant, de
manière générale, l’emprise est constituée de deux à quatre zones (cf. Figure N°5) :
• Une zone de sécurité large de 1 à 2 m, la gestion de la végétation y est intensive pour
permettre les arrêts d’urgence de véhicules et une bonne visibilité ;
• Un accotement, large de 1 à 5 m, il est fauché plus ou moins intensivement, de une à
trois fois par an ;
• Un fossé, large de 0.5 à 2 m, il est curé, a priori, de manière décennale ;
• Un talus, large de 1 m à plusieurs dizaines de mètres, il est généralement géré de
manière extensive, i.e. une fauche par an en fin d’été.
Les dépendances vertes des différentes infrastructures routières représentent d’importantes
surfaces (cf. Table N°2). Pour un département comme la Seine-et-Marne (5,915 km²), les
bords de route représentent ainsi près de 130 km² (i.e. 2.2% de l’occupation des sols) quand,
par exemple, les espaces naturels sensibles départementaux avoisinent seulement les 6.5 km²
(0.11%). Les contributions des différentes infrastructures routières sont plus ou moins
importantes et il est intéressant de noter que les bords de chemins et bords de routes
communales, concernant des acteurs très locaux (maires et agriculteurs), représentent plus de
la moitié de la surface totale des emprises (NB : sans même tenir compte des bords de
champs).
- 17 -
Figure N°5
Structure des bords de route en France
Table N°2
Densité des Emprise Emprise Occupation du

Taille
Type d'infrastructure emprises moyenne totale sol (emprises)
(km) (km.km-2) (ha.km-1) (km2) (%)
Autoroutes 429 0.07 5.20 22.3 0.38
Routes nationales 699 0.12 0.90 6.3 0.11
Routes départementales 4,119 0.70 0.70 28.8 0.49
Routes communales 8,745 1.48 0.30 26.2 0.44
Total/Moy. routes 13,992 2.37 0.60 83.6 1.41
Chemins 22,687 3.84 0.20 45.4 0.77
Total/Moy.
36,679 6.21 0.35 129.5 2.18
toutes infrastructures
Importance du réseau routier et de ses dépendances vertes en Seine-et-Marne (2003)
- 18 -
§ C. RELEVES DE VEGETATION
Durant les trois années de thèse, un important travail de terrain a été réalisé avec plus
de 2,500 relevés floristiques (cf. Table N°3). L’échantillonnage a été réparti entre
différents terrains et mis en place autour de protocoles différents pour permettre d’apporter
les réponses aux questions posées par la thèse : rôle « habitat » et effet « connectivité » des
bords de route, impact sur la diversité végétale des habitats adjacents et effets de la fauche
sur les communautés végétales des bords de route.
Table N°3
Nombre Nombre
Groupes Années Nombre total
Intitulé du terrain de sites de relevés
étudiés de suivi de relevés
d'étude par site
PLANTES Plantes 2007 10 25 250
PETITS
Petits mammifères 2006 48 5 240
MAMMIFERES
2006, 2007
FAUCHE Plantes 10 5 150
et 2008
DYNAMIQUE DES
Petits mammifères 2006 et 2007 45 5 450
CAMPAGNOLS
Efforts de terrain 2006-2008
(1) Quadrats de relévé et réplication

Les relevés de végétation ont été réalisés de 2006 à 2008 de manière exhaustive au sein de
quadrats de 1m² (0.5 x 2 m) à l’aide d’une échelle de Braun-Blanquet corrigée (Cf. Table
N°4). Les quadrats ont été placés dans trois zones des bords de route : accotement, fossé et
talus.
Table N°4
Recouvrement Recouvrement
par espèce Indice moyen
Moins de 5% 1 2.5%
5 à 10% 2 7.5%
10 à 25% 3 17.5%
25 à 50% 4 37.5%
Plus de 50% 5 75.0%
Echelle de Braun-Blanquet corrigée
- 19 -
Cependant travailler à l’échelle des communautés peut se révéler difficile car les espèces
peuvent avoir des probabilités de détection différentes : certaines espèces sont plus difficiles à
détecter ou à déterminer que d’autres ce qui peut induire un biais dans les analyses. Celles-ci
doivent donc pouvoir tenir compte de ce biais potentiel et, éventuellement, le corriger ou le
contrôler à travers l’estimation de probabilités de détection associées aux espèces ou le calcul
de richesse estimées (Boulinier et al). L’intégration aux protocoles de « réplicats » (5 par
zone étudiée) permet d’obtenir de tels paramètres par des méthodes de « capture-recapture ».
Les données ainsi corrigées par la prise en compte de probabilités de détection permettent des
analyses fiables à l’échelle des communautés (cf. Encadré n°2).
(2) Base de donnée GALIUM

Travailler à l’échelle des communautés peut aussi se révéler difficile car certaines espèces
peuvent avoir des fonctions écologiques équivalentes : certaines espèces présentes les mêmes
traits fonctionnels et il devient alors difficile d’expliquer les compositions des communautés
en se fondant strictement sur de la taxonomie. Le développement de nouvelles méthodologies
d’étude des communautés, à travers les traits fonctionnels des espèces qui les composent,
constitue un sujet de très forte actualité en recherche compte tenu des enjeux qui y sont liés en
termes de politiques de conservation (nécessité de maintenir une diversité fonctionnelle au
sein des écosystèmes) pour le maintien de services écosystémiques.
Ainsi, l’étude des communautés à travers les traits fonctionnels (i.e. « des traits
morpho-physiologiques qui impactent indirectement la fitness d’un individu via leurs effets
sur la croissance, la reproduction et la survie », Violle et al. 2007) est aujourd’hui une
nécessité (McGill et al. 2006) car l’approche taxonomique ne suffit pas à décrire les
fonctionnements des écosystèmes (Eviner 2004) et que les traits fonctionnels constituent de
bons prédicteurs pour l’étude des communautés (Lavorel et Garbier 2002, Pywell et al.
2003). Les informations complémentaires apportées par les traits sont essentielles à la
compréhension des dynamiques des communautés (Noble et Slatyer 1980, Fukami et al.
2005, Roy et Blois (de) 2006, Endels et al. 2007, Diaz et al. 2007, Aubin et al. 2007) ainsi
qu’à leur éventuelle homogénéisation (Devictor et al. 2007). De plus l’analyse de la diversité
fonctionnelle des communautés permet de mieux comprendre le fonctionnement des
écosystèmes (Petchey et Gaston 2002), élément capital en termes de recherche pour la mise
en valeur et la quantification des services écosystémiques.
- 20 -
Afin de travailler sur les traits de vie, un groupe s’est mis en place au laboratoire autour de
Sophie GACHET afin de constituer une base de données sur le modèle BASECO (Gachet et al.
2005) pour les plantes de la région parisienne : GALIUM (cf. Figure N°6).
Figure N°6
Nombre d’espèces renseignées : 928 (avril

2008), espèces détectées lors des différents
terrains des participants au projet.
Traits renseignés : 12, Ex : phénologie, modes de pollinisation, modes de dispersion,

capacités, aptitude à la multiplication végétative, taille à floraison, etc.
Participants : Sophie Gachet (coordinatrice), Jean-Claude Abadie, Isabelle Le Viol,

Audrey Muratey, Maëlle Rambaud, et Louis de Redon.
Sources : Flore des champs cultivés de Philippe JAUZEIN, Flora Helvetica : Flore
illustrée de Suisse de Konrad LAUBER et Gerhart WAGNER, et Nouvelle flore de la
Belgique, du G.D. du Luxembourg, du Nord de la France et des régions voisines de
Jacques LAMBINON, Léon DELVOSALLE et Jacques DUVIGNEAUD.
Constitution de la base de données GALLIUM
§ D. RELEVES FAUNISTIQUES
Originellement pensé pour la capture de carabes, le protocole de piégeage a très
fonctionné pour les petits mammifères dont l’étude a donné de nombreux résultats. Au
total 480 pièges ont été posés entre 2006 et 2007 (cf. Table N°4).
(1) Pot-pièges « Barber »

Le piégeage daunistique a été effectué à l’aide de pot-pièges de type « Barber ». Chaque
piège, mesurant 10.2 cm de haut pour 7.5 cm de diamètre, a été enfoncé dans le talus à 5 m de
la route. Rempli d’environ 155 mL de solution (75 ML d’eau, 75 mL de monopropylene-
- 21 -
glycol, conservateur non-attractif, 5mL de surfactant et 15g de sel), chaque piège est couvert
par un toit en plastique carré de 15 cm de côté pour éviter que les précipitations ne le
remplissent (Cf. Figure N°7).
Le système est laissé sur place durant les quatre semaines de mai pour éviter un impact
trop fort des conditions climatiques (un mois en mai permet de rencontrer pluies et beau
temps).
Une fois les pièges récupérés, leur contenu a été transféré dans l’alcool pour permettre
une meilleure conservation des animaux piégés en attendant leur détermination (l’alcool
n’ayant pas été utilisé dès le piégeagé car très attractif). Les animaux ont été ensuite
déterminés à la loupe binoculaire : Carabidae et petits mammifères (mulots, musaraignes et
campagnols).
(2) Réplication des relevés

Comme pour les plantes, 5 replicats ont été posés sur chaque site d’étude chacun à une
distance de 20m ; distance qui permettait l’indépendance des pièges les uns des autres pour le
piégeage des insectes carabiques. Les pots-pièges ne peuvent cependant pas être considérés
comme indépendants pour le piégeage des petits mammifères.
Figure 7
Protocole « petits mammifères »
- 22 -
§ E. ANALYSES DE SOL
Les bords de route sont des milieux très pollués, des analyses de sol ont donc été
effectuées sur les sites d’étude afin de les intégrer comme variables de contrôle lors des
analyses. Les relevés ont été effectués en 2006 au niveau de l’accotement (à 1m de la
chaussée) et du talus (à 5m de la chaussée) de 48 sites en Seine-et-Marne.
(1) Echantillonnage du sol

Pour analyses, 100g de terre était prélevés dans les deux zones de bords de route
échantillonnées (accotement et talus) à 10cm de profondeur en 5 endroits distincts de chaque
site d’étude (au niveau des 5 réplicats de relevé de végétation). Les 5 prélèvements étaient
ensuite regroupés et mélangés afin d’obtenir un échantillon de terre de 500g par accotement et
par talus de chaque site. Les 96 échantillons de terre de 500g on ensuite été envoyés pour
analyses au Pôle d’Aspach de la SADEF (www.sadef.fr).
(2) Eléments analysés

Les analyses ont portées sur cinq paramètres : le pH (méthode NF ISO 10 390), les quantités
de Na2O échangeable (NFX 31-108), d’azote (NF ISO 13 878), de phosphore (P2O5, Méthode
d’Olsen : NF ISO 11263) et de plomb (NFX 31-147).
- 23 -
PARTIE I
BIODIVERSITE DES BORDS DE ROUTE EN MILIEU AGRICOLE INTENSIF :

INTERETS LOCAUX ET IMPACTS SUR LES MILIEUX ADJACENTS
LES MANGEUX D'TERRE (extraits)
[…] Si je m'souviens, […] Z'ont semé du blé su l'terrain

Voyons dans c'coin d'Beauce. Qu'i's r'tir'nt à ma route ; […]
Y avait dans l'temps un bieau grand ch'min Les épis baisser l'nez d'vant moué
[…] Comm' s'i's avaient honte!...
A c't'heur' n'est pas pus grand qu'ma main…
[…] Et j'f'rai ben r'culer vos mouéssons,
[…] Mang'rint on n'sait quoué ces gas-là, Ah! les mangeux d'terre ! …
l's mang'rint d'la marde !
Le ch'min c'était, à leu' jugé Y avait dans l'temps un beau grand ch'min,
D'la bonn' terr' pardue: […]
A chaqu' labour i's l'ont mangé A c't'heur' n'est pas pus grand qu'ma main…
D'un sillon d'charrue… J'pourrais bien l'élargir, demain !
[…] Mais l'pauv' chemin en est d'venu

Minc' comme eun' couleuve. Gaston COUTE (1980 – 1911)
- 24 -
MANUSCRIPT N°1
Submitted to Acta Oecologica
Plant diversity in intensive agrarian landscapes:

Ecological roles of roadside verges as refuge and dispersal corridors
Louis de REDON
Frédéric JIGUET
Isabelle LE VIOL
Sophie GACHET
Jeffrey B. JOY
&
Nathalie MACHON
- 25 -
MANUSCRIPT N°1
PRESENTATION
TITRE
Diversité végétale au sein des paysages agricoles intensifs : Rôles écologiques des bords routes,
Refuge et corridor pour la flore native.
RESUME
Au sein des paysages agricoles intensifs, les bords de route peuvent-ils constituer un réel habitat
pour la flore native ainsi qu’un corridor pour la dispersion des plantes forestières entre différents
fragments forestiers ?
Notre étude a porté sur une zone de 520 km² située en Seine-et-Marne (77) principalement
composée de champs et de bois. Dix sites ont été échantillonnés afin d’évaluer la richesse et la
diversité des communautés végétales dans différents milieux. Dans chaque site, des inventaires
floristiques ont été menés dans cinq sous-sites : un champ, un bois isolé dans le champ, un bois
adjacent à un bord de route, un bord de route adjacent au bois étudié et un bord de route adjacent au
champ échantillonné.
Seulement 2.5% de la diversité végétale détectée a été relevée dans les champs et plus de
44% dans les bois. Plus de 65% des espèces détectées ont été observées en bords de route (soit
environ 25% de la diversité végétale locale, sources : C.B.N.B.P.) Une majorité des espèces (51%)
ont seulement été relevées en bords de route où un nombre important d’espèces forestières ont aussi
été détectées (40). Il a aussi été montré une forte corrélation négative entre la proportion de plantes
forestières au sein des communautés végétales des bords de route et les distances aux bois les plus
proches (isolés et adjacents).
Ces résultats, couplés aux différences observées entre les communautés végétales des bois
connectés et celles des bois isolés, ont permis de démontrer le rôle des bords de route comme
corridor pour certaines espèces végétales entre les différents patchs d’habitats naturels répartis au
sein de la matrice agricole.
Cette étude a donc permis de démontrer, que dans un paysage d’agriculture intensive, les
bords de route peuvent offrir un refuge de qualité à la flore native et servir de corridor pour la
dispersion des plantes entre des fragments de bois isolés.
MOTS CLEFS
Biodiversité, Bois, Communautés végétales, Diversité α et β, Forêt, Homogénéisation biotique,
Impacts des routes.
- 26 -
MANUSCRIPT N°1
PLANT DIVERSITY IN AGRARIAN LANDSCAPES:
ECOLOGICAL ROLES OF ROADSIDE VERGES AS REFUGE AND CORRIDORS
Louis de REDON1*; Frédéric JIGUET1; Isabelle LE VIOL1; Sophie GACHET2; Jeffrey B.

JOY3 and Nathalie MACHON1.
KEYWORDS ABSTRACT
α-diversity We studied a 520 km2 area crossed by roads in central France to
β-diversity evaluate if roadside vegetation communities are valuable (1) as refuge
Biodiversity for native plant diversity and (2) as corridors for dispersal of forest
Biotic plant species in intensive agrarian landscapes. The area is primarily
homogenisation composed of agricultural fields and groves. Ten sites were sampled to
Forest evaluate species richness and diversity of roadside plant communities.
Groves At each site, we inventoried five subsites: agricultural field, grove
Plant communities adjacent to roads, grove isolated from roads and surrounded by fields,
Road impacts roadside adjacent to groves and roadside adjacent to agricultural fields.
Only 2.5% of the plant diversity was found in crop field subsites and
44% in groves. More than 65% of plants were observed in roadsides
(representing about 25% of the total regional plant diversity). A
majority of plants (51%) were only detected in this habitat where we
also found a large number of forest species (40). We showed strong
negative correlations between percentages of forest plants in roadsides
and distances to groves. Those relations coupled with the observation of
enhanced plant diversity in groves adjacent to roadsides compared to
isolated ones demonstrated the role of roadsides as corridors between
natural habitats for plant communities. Taken together, the results of
this study show that in intensive agricultural landscapes roadsides may
often serve as refuge for native plant diversity and act as corridors for
plant dispersal between isolated forest fragments.
I. INTRODUCTION also detrimentally alters biodiversity through

intensive cultivation of large areas of land
Roads are known to have deleterious effects (Burel et al., 1998). In agricultural
on plant diversity (Forman andAlexander, landscapes, plant diversity is generally low
1998): during construction they cause the consisting primarily of crop species growing
destruction of natural habitats (Forman, in agricultural fields. In a large part of
2000), they alter the surrounding European agricultural landscapes, plant
environments through pollution produced by diversity is concentrated at field boundaries,
vehicles (Trombulak and Frissell, 2000), they groves and roadsides. For example, 45 % of
promote the dispersion of exotic/invasive the plant diversity was found along roadsides
plant species (Parendes and Jones, 2000) and in the United Kingdom (Way, 1977), and 50%
facilitate the penetration and use of natural in the Netherlands (Sykora et al., 1993).
zones by humans (Seibert 1993). Agriculture Finally, in agricultural landscapes, despite the
* Corresponding author; Tel: (+33) 662 045 936; Fax: (+33) 140 79 38 35; E-mail: redon@mnhn.fr;
1: UMR 5173 Conservation des Espèces, Restauration et Suivi des Populations, Muséum National
d’Histoire Naturelle, 55 rue Buffon, F-75005 Paris, FRANCE; 2: UMR 7179 Mécanismes
Adaptatifs : des Organismes aux Communautés, Muséum National d’Histoire Naturelle, 4 rue du
Petit Château, F-91800 Brunoy, FRANCE; 3: Department of Biological Sciences, Simon Fraser
University, Burnaby, B.C., CANADA.
- 27 -
MANUSCRIPT N°1
negative influence of roads on environment, II. MATERIALS AND METHODS
roadsides could constitute one of the main
refuges for wild plant species (Bennett, 1991; Using the following design we evaluated
Akbar, 1997). three hypotheses:
Roadsides, may also favour the (1) By comparing plant species in roadsides
dispersion of plant species (Ross, 1986; with those in adjacent habitats, we asked if
Parendes et al., 2000) and be considered as roadside habitats harbour native plant species
corridors (Corbit et al., 1999). They create which are not found elsewhere in the
connections between isolated natural patches agricultural landscape.
of habitats (such as groves). Vehicle traffic on (2) We hypothesized that native plant
roads is known to increase seed dispersal diversity would be higher in groves adjacent
(Schmidt, 1989; Lonsdale and Lane, 1994). to roads when compared with isolated groves
Corridor effects of roadsides have been thanks to corridor effects of roadsides.
assessed for marine (Scott and Davinson, (3) To evaluate the roles of roadside habitats
1985; Wilcox, 1989) and meadow plants as corridors we also assessed the percentage
(Tikka et al., 2001), but they are generally of forest species in roadside verges as a
difficult to highlight without the use of function of the distance to the nearest isolated
genetic markers (Machon et al. 2003) and or adjacent grove, and compared the linear
studies are needed to assess the effects of trends obtained.
connectivity on biodiversity in fragmented
landscapes (Bailey, 2007). 1. Study area (Figure 1A)
In many countries, roadsides cover
large areas. The French main and secondary The study area is approximately 520 km2 and
road networks are approximately 1 000 000 is located in the centre of France in the
km long (i.e. 1.82 km of roads per km2, data eastern part of the department of Seine-et-
from year 2005). On both sides, verges cover Marne, about 60 km east of Paris (48°47’
5 100 km2 i.e. near 1% of the country area North - 3°18’ East, Figure 1). The area has a
(540 000 km2), more than the total area of typical western European temperate climate.
French National Parks (representing 3 400 It is an intensive agricultural zone of the
km2). Thus, such large areas could play a suburb of Paris. The area is mainly composed
significant role in plant conservation. of agricultural fields (79.7 %), groves (12.5
The aim of the present study is to %) and roadside verges (1.6 %). The
assess the role of roadside verges as refuges remaining 5% is composed of villages,
and corridors for wild plant species in meadows, rivers, lakes, roadways, railways,
agricultural landscapes. We studied a typical etc. The road network is 852 km long i.e. 1.64
intensive agrarian landscape in central France, km of roads per km2. Road verges cover
about 60 km from Paris composed almost approximately 8.32 km2, namely 1.6 % of the
exclusively of agricultural fields and little total study area.
woods noted as “groves” (usually smaller
than 5 ha), the latter being fragmented and
spread within the landscape. Some groves are 2. Sampling design (Figure 1B)
isolated (i.e. completely surrounded by crop
fields) while others border roadside verges. Our study comprised ten similar sites, each
We performed plant inventories in groves, mainly composed of agricultural fields (>
agricultural fields and roadsides to estimate 80%), crossed by a road, containing at least
(1) which plant species are harboured in two groves (woods from 1 to 4 ha and
roadsides; (2) if differences in species > 20), one adjacent to the roadside
richness and composition of plant
communities exist between isolated groves verge (noted Grove-R), the other isolated
and groves adjacent to roadsides; and (3) the from roads (noted Grove-F) i.e. situated
role of roadsides as corridor for forest plants among the fields (more than 100 m and less
between groves.
- 28 -
MANUSCRIPT N°1
FIGURE 1:
SAMPLING DESIGN
The study area is composed of ten sites situated in central France. Each site is divided into five
subsites representing major local habitats: one grove connected to a roadside (Grove-R), one grove
isolated in a crop field (Grove-F), one crop field (Field), one roadside adjacent to the grove (Road-
G) and one roadside adjacent to the crop field (Road-F). The five subsites are located in a circle
1,000 m in diameter and comprise five 1m2 plots (2 x 0.5 m) where an exhaustive vegetation
inventory has been conducted (using a corrected Braun-Blanquet scale).
A. Study area
than 1 000 m from any road). The two groves the road (Grove-R), i.e. connected to roadside,
at each site were at least 1 000 m apart. To (2) the grove among agricultural fields
avoid potential biases due to the influence of (Grove-F), ), i.e. isolated from roadside, (3)
heterogeneity in local environmental the roadside adjacent to the connected grove
conditions, the ten sites were selected within (noted Road-G), (4) a roadside adjacent to the
an area of less than 30 km diameter. field containing the isolated grove (noted
For each site, we performed vegetation Road-F) and the crop field containing the
inventories in five subsites: (1) the grove near isolated grove (noted Field). Each subsite will
- 29 -
MANUSCRIPT N°1
FIGURE 1:
SAMPLING DESIGN
B. Subsites sampling design
be recognized thanks to its code: the habitat bias in vegetation development. In each plot,
information (grove, road, field) followed by a we listed all plant taxa detected and evaluated
letter describing the adjacent habitat (G for their abundance with a corrected Braun-
grove and F for field), except for the crop Blanquet method (Braun-Blanquet, 1932;
field noted just as “field”. Westhoff and Van der Maarel, 1978); because
At each subsite, vegetation inventories vegetation could be layered, cover may
were performed in five systematically placed exceed 100%. Almost all the taxa were
plots (2 m x 0.5 m). Their locations within a identified to species level according to the
habitat were determined as follows: International Plant Names Index
(a) In the groves, two perpendicular transects (http://www.ipni.org). In some cases (Crepis
were defined through the whole grove and sp. for example), taxa were identified to genus
plots were located, one at the intersection of level due to difficulties in identification when
the two transects, and the four others at 20 roadside verges had been mowed. We did not
meters from the inside edge of the grove distinguish the different species of moss and
along transects. ferns (except Equisetum arvense), these taxa
(b) In the roadside habitat, five plots were were recorded as “moss” and “ferns”. We
regularly placed every 20 meters along the considered a species as a “forest species” if it
road in the bank perpendicular to the grove (1 was listed in the reference French forest flora
m from the adjacent habitat). (Rameau, 2005).
(c) In the crop fields, the five plots were
placed every 20 meters along a transect
linking the Road-F subsite to the isolated 4. Data analysis
grove (Grove-F). The first plot was located at
20 meters from the grove in the field. The floristic value of a community was
evaluated through the calculation of its
richness and diversity.
3. Vegetation inventories
a. Diversity index
Vegetation sampling was performed during For each plot, we calculated a Hill diversity
spring 2007 from April 3rd to 15th. The index (Hill, 1973) that we adapted to our plant
period was short enough to avoid temporal sampling. We used the median recovery value
- 30 -
MANUSCRIPT N°1
(Pyšek, 2004) of our Braun-Blanquet We used ϕ as a similarity index and further
corrected scale given by the Appendix 1. built a matrix of similarities. We also
calculated estimated species richness and
percentage (Q*FOR) of forest plants in each
, subsite.
In further analyses, we used a log
transformation for observed species richness
, (S) to meet normality assumptions.
c. Indices equalities between subsites

, To test for differences in species richness and
similarity equalities between subsites, we
performed student tests of mean equalities for
. paired observations between the six possible
pairs of subsites at each site (two by two). To
S is the number of species detected in a plot. test for differences in Hill corrected index
between subsites, we performed ANOVA
Rn is the median Braun-Blanquet recovery with subsites as explaining factors. We used
value of the nth species of a plot. the F statistic for variance equality tests.
pi’ is the relative abundance of ith species in a d. Species composition of subsites

plot. We performed a PCA on subsites and plant
species composition of subsites
To have a more intuitive diversity index, we (presence/absence data). To avoid bias due to
calculated (E has a maximum the presence of rare species, more difficult to
detect, we removed from the dataset species
value for a maximum of diversity, ).
recorded in less than seven subsites.
In subsequent analyses, we used an e. Road impacts on plant communities
exponential transformation for Hill corrected To test for effects of connection to a roadside
index (E) to meet normality assumptions. on plant communities, we created a variable
“connection” to each grove, taking a value of
b. Species richness and similarities 1 for Grove-R (i.e. subsites adjacent to
First, to calculate species richness (S) we roadsides) and 0 for Grove-F (i.e. groves
counted the number of species observed at isolated in fields). We then performed an
each subsite. But considering the five plots as ANOVA on S and E indices with controlled
replicates, we also used a capture-recapture variables (grove dimensions: area and
approach to account for heterogeneity in area/perimeter, land cover: percentages of
detection probability among plant species groves, fields, meadows, urban, etc., and
over the sampled sites. We ran the software logarithm of the distance to road, cf. Appendix
ComDyn (Hines et al., 1999) using spatial 2) and “connection” as explicative. For
replicates and we made the null hypothesis analysis of each subsite all replicates were
that vegetation should not differ from a plot to pooled in the tests to increase the degrees of
another in each subsite. We obtained freedom. Though his method may introduce a
estimated species richness S* for each subsite bias into our analysis we assume that this bias
and the estimated proportion of species found will be constant across all subsites because we
on pairs A and B of two subsites (noted ϕA,B). have exactly five replicates of each.
We could not run ComDyn for plots located Because Grove-R are connected to
in agricultural fields because the number of roadside habitats, the presence of ruderal
species found in that habitat was species could increase artificially the diversity
insufficiently large to do so. index (S* and E). To control for this possible
- 31 -
MANUSCRIPT N°1
artefact, we tested the difference of Q*FOR FIGURE 2:
values between the two kinds of groves. HABITATS RICHNESS
f. Roadsides as corridors for forest plant

species
To test if roadsides function as corridors for
forest plant species, we performed a linear
regression between Q*FOR values of Road-F
sites and distances to groves (Grove-R and
Grove-F) according to the method given by
Christen and Matlack (2006). We tested for
slope equality with an ANOVA between
Q*FOR values and the distances to connected
wood, Dc, and the distances to unconnected
wood, Dnc (Q*FO R ~ Dc + Dnc + Dc:Dnc),
looking at the interaction effect
All statistical analyses were performed using Species richness was recorded in each
R (Ihaka and Gentleman, 1996). habitat. The highest species richness was
found in roadside habitats (both roadside
adjacent to grove (Road-G) and to field
III. RESULTS (Road-F) are richer than groves and field):
102 taxa (68 exclusive) representing 53.1% of
1. Roadsides as habitat for plants the total diversity. Crop field was the most
species poor habitat with only 12 taxa
a. Species composition detected (5 exclusive). Grove habitats
Among the 250 vegetation plots, we detected contained 56 taxa (24 exclusive). Isolated
128 taxa (Appendix 3) i.e. 25.7 % of the total groves (Grove-F) contained 45 taxa (10
flora recorded within the study area by the exclusive) and were richer than connected
National Botanical Conservatory of the Paris ones (Grove-R) (40 taxa, 2 exclusive).
region (data downloaded from Roadsides connected to groves (S*Road-
http://cbnbp.mnhn.fr/cbnbp/observatoire/coll
G = 29.8 ± 9.85) have a higher species
TerrForm.jsp). We identified 103 taxa at the richness than roadsides adjacent to
species level (80.5%) and 23 at the genus agricultural fields (S*Road-F = 20.9 ± 8.02, t9 =
level only (18.0%). We also found moss and 2.80, P = 0.021) and groves adjacent to roads
ferns that we did not identify more precisely (S*Grove-R = 20.5 ± 14.9, t9 = 2.69, P = 0.025).
(except Equisetum arvense). We detected no Differences just failed to reach significance
invasive species and only one non-endemic level with groves adjacent to agricultural
(Veronica persica) in our sampling sites. fields (S*Grove-F = 20.3 ± 13.4, t9 = 2.18, P =
Roadsides hosted the largest number of 0.057).
species, with 102 taxa detected among the The floristic composition differed
100 inventoried plots (79.7% of the observed between these three habitats, as shown by the
diversity), while 68 of these taxa were PCA (Figure 3) performed on the list of taxa
exclusively found in this habitat. We detected detected within subsites. Axis 1 explained
56 taxa in groves (100 plots, 43.8% of the 34.4% of variance and discriminated between
diversity) and 24 of these exclusively in grove subsites and roadside subsites. Axis 2
groves. In crop fields (50 plots), we detected explained 17.6% of variance and
12 taxa (9.4% of the diversity) with five discriminated between groves and crop fields.
exclusive ones (Figure 2). Axis 1 identified grassland species found in
For each site and habitat, we compared roadsides subsites (e.g. Achillea millefolium,
estimated species richness (calculated using Lolium perenne, Plantago lanceolata or
the software ComDyn) between habitats. negative values of Axis 1) and forest species
- 32 -
MANUSCRIPT N°1
FIGURE 3: b. Diversity of communities
PRINCIPAL COMPONENT ANALYSIS ON Comparing the diversity index E between
SUBSITES FLORISTIC COMPOSITION habitats, roadsides adjacent to groves (ERoad-G
= 0.697 ± 0.052) constituted the most diverse
A. Subsites projection on Axis 1 and 2 habitats and agricultural fields (EField = 0.194
± 0.185) the least diverse. Roadsides adjacent
to crop fields (ERoad-F = 0.562 ± 0.080) were
richer than groves and crop field subsites.
There was no significant difference between
the two types of groves (EGrove-R = 0.511 ±
0.102 and EGF = 0.496 ± 0.160) which were
richer than crop field subsites (F1,223 = 66.33,
P = 2.10-16).
2. Road and roadside effects on plant

communities
B. Species projection on Axis 1 and 2 a. The effect of the distance to the next road
Among groves isolated within crop fields, the
richest and most diverse flora is found in
groves that are the largest distant from roads
since a strong positive correlation was found
between both the observed richness (rS =
0.523, F1,48 = 19.48, P = 5.738x10-5) and the
diversity index (rE = 0.548, F1,48 = 20.61, P =
3.790x10-5) with the log-transformed distance
to the nearest road (Figures 4A and 4B). The
floristic richness and diversity of groves are
correlated with the distance to the nearest
road according to control variables (grove
Their species composition shows that the five dimensions or landcover).
habitats have a different floristic composition.
Axis 1 (34.4%) discriminates roadsides from b. The effect of connexion to roadside on
crop fields and groves. Axis 2 (17.6%) plant communities
separates subsites with their adjacent habitat: The previous linear trend was not maintained
roadside, crop field or grove. if groves adjacent to roads (Grove-R) were
added because their richness and diversity
Ranunculus acris which are positioned on the were as high as in the most distant isolated
found in groves (e.g. Arum maculatum, edera groves.
helix,Ranunculus auricomus or Rosa canina ANOVAs performed on S and E
which can be observed on the right of the showed that the connection of one grove to a
PCA: positive values). Axis 2 discriminated roadside enhanced its floristic quality: for the
plants adapted to crop field habitats (e.g. Poa richness (S: F1,91 = 12.34, P = 7x10-4) and for
pratensis or Veronica persica are situated on the community diversity (E: F1,91 = 13.31, P =
negative values of Axis 2) and others (e.g. 4x10-4).
Arum maculatum, Hedera helix, Primula veris No significant differences of QF*
or Ranonculus Acris situated on the top of the values were observed between the two types
PCA) which are totally excluded from of groves (QF*-Grove-R = 0.706 ± 0.037 & QF*-
agricultural fields and only present in Grove-F = 0.652 ± 0.041, t = -0.321 & P =
roadsides or groves. 0.749).
- 33 -
MANUSCRIPT N°1
FIGURE 4: Hill index were higher in isolated groves
DIVERSITY INDICES VS. LOG [DIST TO (Grove-F) than in groves connected to
ROAD (M)] roadsides (Grove-R).
Comparing similarity indices between
A. Observed species richness paired sites, we found that connected groves
were more similar to each other (ϕGrove-R =
0.711 ± 0.233, ϕ is the similarity index) than
they were to isolated ones (ϕGrove-F = 0.641 ±
0.250): t90 = 1.662 and P = 0.00680.
These results could ensure from the
heterogeneous distribution of some rare
species (specialists) which are only present in
isolated groves (e.g. Convallaria majalis,
Listeria ovata, Paris quadrifolia, etc.) and to
corridor effects between connected groves
allowing dispersion of other (generalist)
species.
B. Corrected Hill index
3. Roadsides as corridors
The latest published French forest flora

(Rameau 2005) includes 56 of the 103 plants
species detected in our inventories (i.e. 54.4
%). We ran ComDyn to estimate the
proportion of forest species (Q*FOR ) for each
Road-F subsite. Performing linear
regressions, we found significant negative
correlations between (Q*FOR) in roadsides
adjacent to agricultural fields (Road-F) and
the distance to the nearest connected (Grove-
Diversity indices (S and E) in isolated groves R: R2Road-F→Grove-R = 0.581, P = 0.006) and
(Grove-F) are strongly correlated with isolated (GF: R2Road-F→Grove-F = 0.648, P =
shortest distance from the grove to road. 0.003) groves (Figure 5). Slopes of these two
Observed richness (S: r = 0.523, F1,48 = regression models differed significantly (F1,6
19.48, P = 5.738 x 10-5 and f(x) = 6.39 x - = 7.055 and P = 0.0377). This seems to show
11.43) and corrected Hill Index (E: r = 0.548, a better dispersion of forest species to
F1,48 = 20.61, P = 3.790 x 10-5 and f(x) = 0.34 roadsides from connected groves than from
x - 0.43) are positively correlated to this unconnected ones (slope (Road-F→Grove-R) = -
distance. 0.268 ± 0.064 > slope (Road-F→Grove-F) = -0.435
± 0.119).
c. Effect of roads on plant communities:
Using F-tests we found significant
differencesin variance of diversity indices for IV. DISCUSSION
the two types of grove (isolated and
connected to a roadside), specifically for 1. Roadsides as habitat for biodiversity
species richness (σ²Grove-R = 5.16 and σ²Grove-F
= 5.48, F1,50 = 0.368, P = 3x10-4) and In this study, we assessed the floristic
corrected Hill index (σ²Grove-R = 0.595 and diversity in an intensive agrarian landscape.
σ²Grove-F = 0.608, F1,50 = 0.406, P = 0.001) We sampled subsites in the three most
between Grove-F and Grove-R. Moreover, important habitats of the study zone: groves,
variances in species richness and corrected roadsides and crop fields (representing more
- 34 -
MANUSCRIPT N°1
FIGURE 5: rich habitats (groves) can host a high floristic
REGRESSION OF PERCENTAGE OF diversity.
FOREST SPECIES AND THE DISTANCE It is perhaps surprising that we
FROM ROADSIDE TO CONNECTED AND detected no invasive species in our sampling
UNCONNECTED WOODS plots. This may be due to the placement of
our sampled plots in the roadside bank which
is the less disturbed zone of roadside verges.
Further, Ulman et al. (1995) showed that the
distribution of exotic plant species is spatially
heterogeneous.
2. Impact of roads on plant communities
The presence of roads in a landscape appeared

to have a negative impact on plant
communities of surrounding areas, since we
observed that species richness and diversity of
groves increased with distance to roads.
Causes of this negative trend could be either
direct or indirect.
Significant negative correlations between the
Direct causes could derive from
percentage of forest plant species in
pollution by vehicles with high volumes of
roadsides adjacent to crop fields (Road-F)
car and truck traffic in our study area (average
and the distance to the nearest connected
420 vehicles?h-1 during the evening rushes).
(Grove-R: r2Road-F→Grove-R = 0.581, P = 0.006)
Another possible direct cause could lie in
and isolated (Grove-F: R2Road-F→Grove-F = pressure of urban walkers (stamping and
0.648, P = 0.003) groves. harvesting) that visit groves for recreational
activities (the study area is approximately 60
than 95 % of the landcover in the study area). km from the center of Paris, containing 11
We found that plant communities of million people). Their visitation rate is likely
roadside verges were at least as rich and higher in groves situated closer to roads than
diverse as those in the other habitats, though in the groves completely isolated in fields. All
community compositions differed. The those factors are known to lead to an
majority of plant species detected in roadsides impoverishment of plant communities
(68 for a total of 128 detected taxas, 53.2 %) (Forman and Alexander, 1998; Trombulak
were not observed in groves and crop fields in and Frissell, 2000).
the surrounding area. Thus, roadsides can be Indirect causes could be the loss of
considered as refuges zones for many plant many animal populations essential to plant
species in such intensive agricultural life (pollinators, seed dispersers, and
landscape where grasslands do not exist. predators of phytophageous insects) through
Using estimated species richness, we pollution or impacts with automobiles
supposed that more than 25% of all plant (Forman and Alexander 1998). Road
species inventoried in the area by the National casualties may have a strongly negative
Botanical Conservatory of the Paris region impact on wildlife in our study area because
should be found in our 20 roadside subsites of the high traffic volume. Predators are
(i.e. an area representing 0.02% of the study known to be more affected by pollution
area or 0.1% of local roadside verges). This (Morgan et al., 1983) and their local
proportion is very similar to results of Way extirpation can lead to explosions of
(1977) and Sykora (1993) who found high herbivore populations in roadside verges
plant diversity on roadside verges in U.K. and (Flückiger et al., 1978) and in adjacent
in the Netherlands. Furthermore, some habitats and thus affect plant communities.
portions of the roadsides, when adjacent to
- 35 -
MANUSCRIPT N°1
3. Roadsides as corridors for plant species and lead to more diverse but more fragile
assemblages.
Our results suggest a road corridor effect for We conclude thanks to points (1) and
plant species: (2) that roadsides act as corridors for forest
(1) Because forest plant species are plant species. This corridor function had been
found in roadsides for hundreds of meters previously hypothesized (Zink et al., 1995;
with proportions negatively correlated within Corbit, 1999; Tikka et al., 2001) but never
distances to groves, and because the slope of conclusively established (Sherwood et al.,
the regression with distance to isolated groves 2002).
(Grove-F) is stronger than the one with the Actually, all “forest” species detected
distance to adjacent groves (Grove-R), we in roadsides are not “strictly forest
assume that there is a facilitation of forest specialists” but “semi-forest species”. But we
plant dispersal by roadsides. It was indeed can consider them as sufficiently generalist to
probably the case for species such as the disperse at short scale (few hundred meters)
forest plants Galium aparine, Ranunculus into roadsides. A future study should be
auricomus, Stellaria holostea and Veronica carried out to investigate the possible effect of
hederifolia that have been found migrating roadside with edges as corridor for strictly
the furthest along verges and found in forest plants.
roadsides adjacent to field (Road-F) subsites.
This corridor effect probably depends on
dispersal abilities of species (Noss, 1987; 4. Roadsides and biotic homogenisation
Beier and Noss, 1998) and could thus be
different according to species considered. Groves adjacent to roadsides had indeed more
However, it is likely stronger for species such similar plant communities than isolated ones.
as Heracleum sphondylium, Lamium This suggests a negative impact of roads on
purpureum and Stellaria holostea that were plant communities (loss of species and
found in connected groves (Grove-R), diversity) that could be partly mitigated by
roadsides adjacent to groves (Road-G) and their role as corridors. We indeed observed a
roadsides adjacent to fields (Road-F) but not high alpha diversity in connected groves due
in isolated groves (Grove-F). Those species to roadsides acting as corridors. However, we
seem to be only able to disperse along also found low beta diversity among them.
roadsides but not through fields. Hence, This relationship between alpha and beta
species that are able to migrate through diversity is in accordance with Loreau (2000)
roadsides may be favoured in such intensive proving the importance to work at different
agricultural landscapes. scales to correctly assess biodiversity: even if
(2) Because groves adjacent to each connected grove is richer than most of
roadsides (Grove-R) host particular, rich and the unconnected ones, the diversity they
diverse communities of plants with no globally harbour is low because they share
significant difference of forest plant similar species. Increase in taxonomic
proportions with isolated groves, we also similarity is typically scale dependent (Olden
assume that plants from roadsides disperse in and Poff, 2003).
groves and impact them. According to Moreover, specialist species of
metapopulation theory (Levins, 1968), surrounding habitats seem to be negatively
population viability is enhanced when affected by roads. In roadside verges, we
populations are connected to each other, found 40 of the 56 forest plant species
leading to better community stability. We detected in the study but many of them do not
hypothesize that for this reason, the plant strictly depend on forest conditions. They
communities in groves adjacent to roadsides usually tolerate shade but in roadsides they
have a higher floristic quality. In isolated also tolerate higher light conditions. Those
groves, the turn-over of species by stochastic plants can be defined as more generalist than
extinction/recolonization could be more rapid other species, strictly found in forest habitats
- 36 -
MANUSCRIPT N°1
which are considered specialist. In contrast, hedgerow plantation, etc.) could maintain and
species that have narrower ecological improve their role. Anthropogenic
requirements were essentially detected in disturbances have strong negative impacts on
isolated groves and were totally absent from the plant populations of roadside verges
connected ones (Anemone nemorosa, (Spooner, 2004). Therefore roadside
Convallaria majalis, Paris quadrifolia or management should also integrate
Veronica montana for example). Hence, environmental objectives to maintain verges
connections to roadsides seem to lead to a as viable habitats and corridors for plant
homogenisation of plant community species, if it can be proved that the corridor
compositions of surrounding habitats. Often, effect do not threaten too much biodiversity at
an increase of local richness can lead to a a large scale. We assume that the best way to
decrease of global diversity through the protect local forest diversity is to avoid road
increase of similarity among communities constructions when it is possible. Else, if the
composition. It is usually the result of road has to be established for numerous
generalists or introduced species progressions reasons, we assume that it will be better to
that are more likely to invade disturbed build it as far as possible from groves. If this
ecosystems (Hobbs and Mooney, 1998). This not possible to have more than a distance of
non-random reshuffling of species pools leads 500 m between road and grove, then it should
to biotic homogenisation (McKinney and be better to build it very close to grove to
Lockwood, 1999) that is considered as one of connect them to roadside verges in order to
the major causes of biodiversity crisis due the mitigate as maximum as possible road
increase of human activities and disturbances impacts favouring roadside corridor effects.
(Olden and Rooney, 2006). While largely
described for other human activities
(McKinney, 2006, Smart et al., 2006), such ACKNOWLEDGEMENTS
impact of roads had never been mentioned.
Given the area covered by roads and their The research was supported financially by the
potential influence on biodiversity at a large Direction Générale des Routes (Roads
scale, we consider that this effect should be General Direction, D.G.R.) of the Ministère
more studied to verify if the corridor function de l’Ecologie, du Développement et de
of roads do really mitigate their negative l’Aménagement Durables (French Minister
effects. for Ecology, Sustainable Developpement and
Spacial Planning, M.E.D.A.D.).
We thank the Direction
V. CONCLUSIONS Départementale de l’Equipement of Seine-et-
Marne (Equipement Department Agency,
Presence of roadside verges doubles the local D.D.E. 77) and the Institut d'Aménagement et
plant richness (53.2% of taxa were only d'Urbanisme de la Région d’Ile-de-France
detected in roadside verges) in intensive (Institute for Urban Planning and
agrarian landscapes. Life history and dispersal Developpement of the Paris Ile-de-France
mechanisms of plant species play an essential Region, I.A.U.R.I.F.) who provided GIS data
role in determining the efficiency of roadside and technical support to protect on roadside
verges as habitat (Godfree, 2004) and as verges. We thank also Rose-Line
corridors between habitats (Primack and PREUD’HOMME and Elise CONTAN,
Miao, 1992). A better understanding of how master students who worked on the field to
plant life history characteristics (i.e. dispersal collect data. We vigorously thank Didier BAS
and reproductive traits) influence the and Vincent DEVICTOR for softwares
efficiency of roads as corridors (Roy and de facilities (ComDyn).
Blois, 2006) should help to determine how
roadside management (mowing planning,
- 37 -
MANUSCRIPT N°1
REFERENCES Godfree R., Lepschi B., Mallinson D., 2004.
Ecological filtering of exotic plants in
Akbar K.F., 1997. Aspects of the ecology and an Australian environement. Journal of
conservation value of roadside Vegetation Science 15, 227-236.
vegetation in northern England. Ph.D. Hill M.O., 1973. Diversity and eveness: a
Thesis, University of Bradford, U.K. unifying notation and its consequences.
Bailey S., 2007. Increasing connectivity in Ecology 54, 427-432.
fragmentaed landscapes: An Hines J.E., Boulinier T., Nichols J.D., Sauer
investigation of evidence for J.R., Pollock K.H., 1999. COMDYN:
biodiversity gain in woodlands. Forest software to study the dynamics of
Ecology and Management 238, 7-23. animal communities using a capture-
Beier P., Noss R.F., 1998. Do habitat recapture approach. Bird Study 46, 209-
corridors provide connectivity ? 217.
Conservation Biology 12, 1241-1252. Hobbs R.J., Mooney H.A., 1998. Broadening
Bennett A.F., 1991. Roads, roadsides and the extinction debate: population
wildlife conservation: a review. In: deletions and additions in California
Saunders DA, Hobbs RJ, eds. Nature and Western Australia. Conservation
conservation 2: the role of corridors. Biology 12, 271–83.
Surrey Beatty, Chipping Norton, New Ihaka R., Gentleman R., 1996. R: A Language
South Wales, Australia, pp. 99-117. for Data Analysis and Graphics. Journal
Braun-Blanquet J., 1932. Plant sociology : the of computational and graphical statistics
study of plant communities. Graw-Hill 5, 299-314.
Book Company. New York. Levins R., 1968. Evolution in changing
Burel F., Baudry J., Butet A., Clergeau P., Le environments. Princeton:
Cœur D., Dubs F., Morvan N., Delettre PrincetonUniversity Press.
Y., Paillat G., Petit S., Thenail C., Lonsdale W.M., Lane A.M., 1994. Tourist
Brune E., Lefeuvre J.C., 1998. vehicles as vectors of weed seeds in
Comparative biodiversity along a Kakadu National Park, Northern
gradient of agricultural landscapes. Acta Australia. Biological Conservation. 69,
Oecologica 19, 47-60. 277-283.
Christen D., Matlack G., 2006. The Role of Lorreau M., 2000. Are communities
Roadsides in Plant Invasions: a saturated? On the relationship between
Demographic Approach. Conservation α, β and γ diversity. Ecology Letters 3,
Biology 20, 385-391. 73-76.
Corbit M., Marks P.L., Gardescu S., 1999. Machon N., Bardin P., Mazer S.J., Moret J.,
Hedgerows as habitat corridors for Godelle B., Austerlitz F., 2003.
forest herbs in central New-York, USA. Relationship between genetic structure
Journal of Ecology 87, 220-232. and seed and pollen dispersal in the
Flückiger W., Oertli J.J., Baltensweiler W., endangered orchid Spiranthes spirali.
1978. Observations of an aphid New Phytologist 157, 677-687.
infestation on hawthorn in the vicinity McKinney M.L., Lockwood J.L., 1999. Biotic
of a motorway. Naturwissenschaften 65, homogenisation, a few winners
654-655 replacing many losers in the next mass
Forman R.T.T., Alexander L.E., 1998. Roads extinction. Trends in Ecology and
and their major ecological effects. Evolution 14, 450–453.
Annual review of Ecology Evolution McKinney M L., 2006. Urbanization as a
and Systematics 29, 207-231. major cause of biotic homogenization.
Forman R.T.T., 2000. Estimate the area Biological Conservation 127, 247–260
affected ecologically by the road system Morgan E., Porak W., Arway J., 1983.
in the United States. Conservation Controlling acidic-toxic metal leachates
Biology 14,31-35. from southern Appalachian construction
- 38 -
MANUSCRIPT N°1
slopes: mitigating stream damage. Seibert P., 1993. Vegetation and man in South
Transportation Research Record. Africa from a historical perspective.
948,10–16 Phytocoenologia 23, 457-493.
Noss R.F., 1987. Corridors in Real Sherwood B., Cutler D,. Burton J,. 2002.
Landscapes: A Reply to Simberloff and Wildlife and roads, The ecological
Cox. Conservation Biology 1, 159-164. impact. London: Imperial College Press.
Olden J.D., Poff N.L., 2003. Toward a Smart S.M., Thompson K., Marrs R.H., Le
mechanistic understanding and Duc M.G., Maskell L.C., Firbank L.G.,
prediction of biotic homogenisation. 2006. Biotic homogenization and
American Naturalist 162, 442-460. changes in species diversity across
Olden J.D., Rooney T.P., 2006. On defining human-modified ecosystems.
and quantifying biotic homogenisation. Philosophical Transactions of the Royal
Global Ecology and Biogeography 15, SocietyB 273, 2659–2665.
113–120. Spooner P.G., Lunt I.D., Briggs S.V., 2004.
Parendes L.A., Jones J.A., 2000. Role of light Spacial analysis of antropogenic
avaibility and dispersal in exotic plant disturbance regimes and roadside shrubs
invasion along roads and streams in the in a fragmentaed agricultural landscape.
H.J. Andrews experimental forest, Applied Vegetation Science 7, 61-70.
Oregon. Conservation Biology 14n 64- Sykora K.V., de Nijs L.J., Pelmsa H.M.,
75. 1993. Plantengemeenschappen van
Primack R.B., Miao S.L., 1992. Dispersal Can Nerderlandse. Natuurhistorische
Limit Local Plant Distribution. Vereninging.
Conservation Biology 6n 513-519. Tikka P.M., Högmander H., Koski P.S., 2001.
Pyšek P., Chocholoušková Z., Pyšek A., Road and railway verges serve as
Jarošík V., Chytrŷ M, Tichŷ L., 2004. dispersal corridors for grassland plants.
Trends in species diversity and Landscape Ecology 16, 659-666.
composition of urban vegetation over Trombulak S.C.,Frissel C.A., 2000. Review
three decades. Journal of Vegetation of ecological effects of roads on
Science 15n 781-788. terrestrial and aquatic communities.
Rameau J.C., Mansion D., Dumé G., Timble Conservation biology 14, 18-30.
J., Lecointre A., Supont P., Keller R., Ullmann I., Bannister P., Wilson J.B., 1995.
2005. Flore forestière française. Guide The vegetation of roadside verges with
écologique illustré. 1- Plaines et respect to environnemental gradients in
collines. Institut pour le développement southtern New Zealand. Journal of
forestier. Vegetation Science 6, 131-142.
Ross S.M., 1986. Vegetation change on Way J.M., 1977. Roadside verges and
highway verges in south-east Scotland. conservation in Britain: A review.
Journal of Biogeography 13, 109-113. Biological Conservation 12, 65-74.
Roy V,. de Blois S., 2006. Using functional Westhoff V., Van der Maarel E., 1978. The
traits to assess the role of hedgerow Braun-Blanquet approach. In: Whittaker
corridors as environmental filters for R H (ed.), Ordination and classification
forest herbs. Biological Conservation of plant communities. Junk. The Hague.
130, 592-603. Wilcox D.A., 1989. Migration and control of
Schmidt W., 1989. Plant dispersal by motor purple loosestrife (Lythrum salicaria L.)
cars. Vegetatio 80, 147-52 along highway corridors. Environmental
Scott N.E., Davison A.W., 1985. The management 13, 365-370.
distribution and ecology of coastal Zink T.A, Allen M.F., Heindl-Tenhunen B.,
species on roadsides. Plant Ecology 62, Allen E.B., 1995. The effect of a
433-440. disturbance corridor on an ecological
reserve. Restoration ecology 3, 304-
310.
- 39 -
MANUSCRIPT N°1
SUPPLEMENTARY DATA
APPENDIX 1:
CORRECTED BRAUN-BLANQUET SCALE
Taxa/species Median
Index
recovery recovery value
Less than 5% 1 2.50%
5 to 10% 2 7.50%
10 to 25% 3 17.50%
25 to 50% 4 37.50%
More than 50% 5 75.00%
APPENDIX 2:
SITES SAMPLED
Area / Distance
Subsites Aera Landcover *
Perimeter to road
(m²) (m) (m) (A) (B) (C) (D) (E)
GR01 29728 38.8 0 7.9% 90.5% 0.0% 1.3% 0.3%
GR02 19620 28.1 0 5.8% 91.5% 0.0% 0.0% 2.7%
GR03 37268 28.6 0 6.5% 93.5% 0.0% 0.0% 0.0%
GR04 15289 28.9 0 9.7% 88.8% 0.0% 0.9% 0.5%
GR05 10686 22 0 3.4% 87.1% 0.0% 0.0% 9.4%
GR06 17174 28.8 0 3.7% 90.0% 0.0% 0.0% 6.4%
GR07 17933 26.4 0 5.0% 91.2% 0.2% 0.0% 3.6%
GR08 29221 43.2 0 9.3% 90.7% 0.0% 0.0% 0.0%
GR09 10958 22.8 0 4.3% 88.5% 0.0% 0.0% 7.3%
GR10 26192 23.9 0 5.3% 88.2% 0.0% 0.4% 6.2%
GF01 18123 25.4 385 4.2% 94.6% 0.0% 0.0% 1.2%
GF02 24631 24 600 3.6% 96.4% 0.0% 0.0% 0.0%
GF03 30163 39.7 517 6.5% 89.8% 0.0% 0.0% 3.7%
GF04 18925 31.7 137 2.9% 95.6% 0.0% 0.0% 1.5%
GF05 17841 31.4 888 8.6% 87.9% 1.4% 1.5% 0.6%
GF06 19840 33.2 311 2.6% 95.3% 0.0% 0.0% 2.1%
GF07 16735 30.7 238 5.7% 91.2% 0.0% 0.8% 2.3%
GF08 10435 24.5 531 7.0% 91.5% 0.0% 0.0% 1.4%
GF09 20201 29.7 999 3.3% 96.4% 0.0% 0.0% 0.3%
GF10 21070 22.8 507 6.3% 87.8% 0.0% 0.5% 5.4%
* Landcover: (A) Woods, (B) Fields, (C) Water, (D) Meadows and (E) Others (urban, road, etc.)
- 40 -
MANUSCRIPT N°1
APPENDIX 3: Glechoma hederacea L. • GLEHED
TAXA IDENTIFIED Hedera helix L. • HEDHEL
Heracleum sphondylium L. • HERSPH
Taxas Code Himantoglossum hircinum Spreng. HIMHIR
Acer campestre L. • ACECAM Hypericum perforatum L. • HYPPER
Adoxa moschatellina L. • ADOMOS Lamium album L. LAMALB
Agrimonia eupatoria L. • AGREUP Lamium purpureum L. LAMPUR
Agrostis sp. • AGRSPX Leondonton sp. LEOSPX
Ajuga reptans L. AJUREP Leucanthemum vulgare Lam. • LEUVUL
Alium sp. ALISPX Ligustrum vulgare L. LIGVUL
Alopecurus pratensis L. ALOPRA Listera ovata (L.) R.Br. • LISOVA
Anemone nemorosa L. • ANENEM Lolium perenne L. LOLPER
Anthriscus cerefolium Hoffm. ANTCER Lonicera periclymenum L. • LONPER
Anthriscus sylvestris (L.) Hoffm. • ANTSYL Lotus corniculatus L. LOTCOR
Arthemisia sp. ARTSPX Luzula campestris (L.) DC. • LUZCAM
Arum maculatum L. • ARUMAC Matricaria sp. MATSPX
Bellis perennis L. BELPER Medicago lupulina L. MEDLUP
Brassica napus L. BRANAP Medicago minima (L.) L. MEDMIN
Brassica rapa L. BRARAP Orchis mascula L. • ORCMAS
Bromus arvensis L. BROARV Paris quadrifolia L. • PARQUA
Bromus erectus Huds. • BROERE Picris echioides L. PICECH
Cardamine pratensis L. • CARPRA Picris hieracioides L. PICHIE
Carex sp. CARSPX Picris sp. PICSPX
Centaurea jacea L. CENJAC Plantago lanceolata L. PLALAN
Centaurea nigra L. • CENNIG Plantago major L. PLAMAJ
Cerastium fontanum Baumg. CERFON Plantago media L. PLAMED
Cirsium sp. CIRSPX Platanthera bifolia (L.) Rich. PLABIF
Clematis vitalba L. • CLEVIT Poa annua L. POAANN
Conium maculatum L. CONMAC Poa pratensis L. POAPRA
Convallaria majalis L. • CONMAJ Poa sp. POASPX
Convolvulus arvensis L. CONARV Poa trivialis L. • POATRI
Crataegus laevigata DC. • CRALAE Polygonatum multiflorum All. • POLMUL
Crataegus monogyna Jacq. • CRAMON Populus sp. POPSPX
Crataegus sp. CRASPX Potentilla anserina L. POTANS
Crepis sp. CRESPX Potentilla reptans L. • POTREP
Cruciata laevipes Opiz • CRULAE Potentilla sterilis Garcke • POTSTE
Dactylis glomerata L.• DACGLO Primula elatior Hill • PRIELA
Daucus carota L. DAUCAR Primula veris L. • PRIVER
Equisetum arvense L. • EQUARV Prunus sp. PRUSPX
Fagus sylvatica L. • FAGSYL Quercus sp. QUESPX
Festuca arundinacea Schreb. FESARU Ranunculus acris L. RANACR
Fragaria vesca L. • FRAVES Ranunculus arvensis L. RANARV
Galium aparine L. • GALAPA Ranunculus auricomus L. • RANAUR
Galium mollugo L. • GALMOL Ranunculus ficaria L. • RANFIC
Galium verum L. • GALVER Ranunculus repens L. • RANREP
Geranium dissectum L. GERDIS Ranunculus sp. RANSPX
Geranium molle L. GERMOL Ribes rubrum L. • RIBRUB
Geranium pusillum L. GERPUS Rosa canina L. • ROSCAN
Geranium robertianum L. • GERROB Rubus sp. RUBSPX
Geranium rotundifolium L. GERROT Rumex obtusifolius L. • RUMOBT
Geum urbanum L. • GEUURB Rumex sp. RUMSPX
- 41 -
MANUSCRIPT N°1
Sambucus nigra L. • SAMNIG Vicia sativa L. VICSAT
Senecio jacobaea L. SENJAC Vicia sp. VICSPX
Sonchus arvensis L. SONARV Vinca minor L. • VINMIN
Sonchus oleraceus L. SONOLE Viola reichenbachiana Jord. ex Bor. VIOREI
Stachys sylvatica L. • STASYL Viola riviniana Rchb. • VIORIV
Stellaria holostea L. • STEHOL Viola sp. VIOSPX
Tamus communis L. • TAMCOM Ferns FERN
Taraxacum sp. TARSPX Moss MOSS
Teucrium scorodonia L. • TEUSCO
Trifolium campestre Schreb. • TRICAM
Trifolium dubium Sibth. TRIDUB
Trifolium pratense L. TRIPRA Classification is according to the
Trifolium repens L. TRIREP International Plant Names Index
Trifolium sp. TRISPX (http://www.ipni.org)
Urtica sp. URTSPX
Veronica agrestis L. VERAGR
•
Presence in the French forest flora (Rameau,
Veronica arvensis L. VERARV 2000).
Veronica hederifolia L. VERHED
Veronica montana L. • VERMON * Non-native plant species according to the
Veronica officinalis L. • VEROFF Conservatoire botanique national du Bassin
Veronica persica Poir.* VERPER parisien (French flora protection Agency for
Veronica sp. VERSPX Paris region), database: http://cbnbp.mnhn.fr.
Sorex minutus
- 42 -
MANUSCRIPT N°2
Submitted to Landscape Ecology
ROAD NETWORK IN INTENSIVE AGRICULTURAL LANDSCAPE:

HABITAT, CORRIDOR OR BARRIER FOR TWO SMALL MAMMALS SPECIES?
Louis de REDON
Isabelle LE VIOL
Frédéric JIGUET
Nathalie MACHON
Olivier SCHER
&
Christian KERBIRIOU
- 43 -
MANUSCRIPT N°2
PRESENTATION
TITRE
Infrastructures routières en paysage agricole: Habitat, corridor ou barrière pour deux
espèces de petits mammifères ?
RESUME
Nous avons étudié la répartition et l’abondance de petits mammifères dans une zone d’agriculture
intensive en Ile-de-France. 176 sites on été échantillonnés dans différents milieux : zones
marginales (bords d’autoroute, bords de route et bordures de champ), zones naturelles (bois) et
zones de culture (champs). Des pot-pièges de type Barber (n = 864) ont été posés pour capturer des
campagnols, mulots et musaraignes (418 individus identifiés).
Nous avons trouvé les abondances les plus élevés dans les bords de route et d’autoroute et
avons ciblé notre étude sur deux espèces : Microtus arvalis (rongeur) et Sorex coronatus
(insectivore). Nous avons montré que les campagnols et les musaraignes ont des patterns de
répartition et des dynamiques très différents : les deux étaient très abondants dans les différents
accotements mais les abondances de Microtus arvalis dans les champs étaient beaucoup plus
importantes que celles de Sorex coronatus dont les quantités étaient essentiellement résiduelles.
Nous avons aussi montré que les dynamiques des musaraignes au sein des accotements
étaient sensibles à la taille des emprises : les accotements les plus larges, i.e. ceux des bords
d’autoroute, peuvent constituer un habitat de qualité pour Sorex coronatus alors que ceux plus
étroits, i.e. ceux des bords de route, peuvent servir uniquement pour leur dispersion entre différents
milieux naturels adjacents. Nos résultats ont aussi montré l’effet « barrière » de la route qui ne
semble pas ou peu traversée par Sorex coronatus.
L’ensemble des résultats de cette étude montrent, dans des paysages d’agriculture intensive,
que les dépendances vertes des réseaux routiers et autoroutiers peuvent être considérées comme des
zones refuges, des habitats ou des corridors pour les petits mammifères en fonction des espèces
considérées et des caractéristiques des emprises routières.
MOTS CLEFS
Bords de route, Campagnols, Dispersion, Microtus arvalis, Musaraignes, Sorex coronatus, Zones
marginales.
- 44 -
MANUSCRIPT N°2
ROAD NETWORK IN INTENSIVE AGRICULTURAL LANDSCAPE:
HABITAT, CORRIDOR OR BARRIER FOR SMALL MAMMALS?
Louis de REDON1*; Isabelle LE VIOL1; Frédéric JIGUET1; Nathalie MACHON1; olivier

SCHER2 and Christian KERBIRIOU1.
KEYWORDS ABSTRACT
Dispersal Road network have negative impacts on environment but their verges
Marginal areas can act as a refuge and/or as ecological corridors in anthropogenic
Mice ecosystems. We studied the distribution and abundance of small
Microtus arvalis mammals in intensive agrarian landscape in central France. 176 sites
Roadside were sampled in different habitats: marginal zones (highway verges,
Road management roadside and field margin), natural areas (woods) and fields (at
Shrew different distances from margins) using non-attractive pitfall traps (n=
Sorex coronatus 864) to capture shrews, voles and mice (418 small mammals captured).
Vole We found higher abundances of small mammals in highway and road
verges, and further focused on the two most-captured species: Microtus
arvalis (rodent) and Sorex coronatus (insectivorous). We showed that
voles and shrews had different dynamics and distribution patterns: both
were very abundant in verges and margins and Microtus arvalis was
more abundant in crop fields, a habitat where numbers of Sorex
coronatus were residual and correlated to the distance to closest
natural habitats. We showed that shrew dynamics in verges were
function of margin widths: larger verges, constitute a habitat for Sorex
coronatus while narrower verges serve only as corridor for their
dispersion between natural connected habitats. Taken together, these
results show that in intensive agricultural landscapes, roadside and
highway verges serve as refuge, habitat or corridor for small mammals
depending on species and margin characteristics.
I. INTRODUCTION Alexander 1998; Trombulak and Frissel,

2000).
Road networks have expanded with human However, the potential biological
population over large areas (Watts et al. value of road verges has also long been
2007). In France, main and secondary road recognised (Way 1977) with eventual
networks are approximately 1,000,000 km contributions to conservation of indigenous
long (i.e. 1.82 km of roads per km2, data from flora (Spooner et al. 2004, O’Farrell and
2005) and their verges cover approximately Milton 2006) and fauna (Meunier et al. 1999,
5,100 km², i.e. nearly 1% of the country area. Bellamy et al. 2000, Ries et al. 2001). In
They are known to influence landscape human dominated areas such as intensive
structure and have to impact ecosystems agrarian landscapes, where non-agricultural
dynamics through habitat destruction, habitats (e.g. edges) are critical to
alteration and fragmentation (Forman and conservation of biological diversity and
* Corresponding author; Tel: (+33) 662 045 936; Fax: (+33) 140 79 38 35; E-mail: redon@mnhn.fr;
1: UMR 5173 Conservation des Espèces, Restauration et Suivi des Populations, Muséum National
d’Histoire Naturelle, 55 rue Buffon, F-75005 Paris, FRANCE; 2: Pôle-relais Mares et Mouillères de
France, Maison de l'Environnement de Seine-et-Marne, Etang de Moret, 26 route de Montarlot, F-
77250 Ecuelles, FRANCE.
- 45 -
MANUSCRIPT N°2
ecological processes (Burel 1996), road and management practices, in order to
verges can play a crucial role as a refuge optimise the role of road verges as a refuge
and/or as ecological corridors (Dawson 1994, for wild fauna.
2002, Tikka et al. 2001). Road verges may The aim of our study was therefore to
present few interests in large natural areas but assess the role of road verges as habitat,
may be crucial in intensive agrarian corridor and barrier for small mammal species
landscapes where they represent the last semi- in an agririan landscape. The study region is
natural habitats (Davies and Pullin 2007). As almost exclusively composed of crop fields. A
linear areas of semi-natural vegetation, road few woods constitute the major natural areas.
verges may provide habitat and thus create We sampled small mammals in highway and
refuges in hostile matrix and corridors for roadside verges, which are relative important
animals (Hansen & Jensen 1972, Merriam et marginal areas because they are remaining as
al. 1990, Bennett 1990, Hodkinson & quasi-unique open space habitats in the
Thompson 1997). landscape. To obtain informations about the
Actually, small-mammals are relative abundance of small mammals in
supposed to be major natural factors of verges compared to other habitats, we launch
agrarian ecosystems because they regulate trapping with the same methodology in
invertebrate populations in crop fields and woods, crop fields and field margins.
grasslands (Churchfield et al. 1991). Shrews Given the species we trapped, we
are well known as voracious predators decided to focus our study on two particular
(Churchfield and Brown 1987) and even species: Microtus arvalis (vole, i.e. a rodent)
small rodents predate some invertebrates and Sorex coronatus (shrew, i.e.
(Parmenter and MacMahon 1988): for insectivorous). We further inferred the
examples Sorex araneus populations can kill ecological function of verges for those two
more than 100,000 moths per ha and per species from their capture rates within and
month (Buckner 1969) and Apodemus between habitats. By comparing small
sylvaticus can destroy up to 50% of the mammal species trapped in marginal areas
overwintering cocoons of Diprion pini (Obrtel (highway verges, road verges and field
et al. 1978), a hymenoptera causing important margins) to those trapped in adjacent habitats
damages to European pines forests (Pasquier- (fields and woods), we asked if those areas
Barre 1999). Maintaining invertebrate can be considered as habitats, corridors or/and
predators is crucial for agricultural production barriers, and finally interpret these results in
(Schoener 1988, Spiller & Schoener 1990, terms of roadside management.
Dial & Roughgarden 1995.) because it means
low pest abundances (Maisonneuve & Rioux
2001) and thus lower inputs of pesticides.
Small mammals are thus important for II. MATERIALS AND METHODS
biological equilibriums in agrarian landscape.
Small mammals have been observed to Using the following design we evaluated
spread tens of kilometers along highways and three hypotheses:
roads verges (Getz et al. 1978). Roadsides are (1) By comparing plant species in roadsides
thus considered as effective corridors for with those in adjacent habitats, we asked if
mammal species (Bennett 1990, Suckling roadside habitats harbour native plant species
1984, Verkaar 1990). On the contrary, roads which are not found elsewhere in the
could act as significant barriers to dispersal agricultural landscape.
for many animals (Rico et al. 2007), (2) We hypothesized that native plant
particularly in the case of high traffic rates diversity would be higher in groves adjacent
(Harris & Silva-Lopez 1992). Roadside width to roads when compared with isolated groves
certainly plays probably an important role in thanks to corridor effects of roadsides.
determining the role of edges as habitat, (3) To evaluate the roles of roadside habitats
corridor or barrier, which we aim to study to as corridors we also assessed the percentage
promote biodiversity-friendly construction of forest species in roadside verges as a
- 46 -
MANUSCRIPT N°2
function of the distance to the nearest isolated Road network in the study area is 38,906 km
or adjacent grove, and compared the linear long, i.e. 3,23 km of highways and roads per
trends obtained. km2. Verges of this network are narrow strips
but represent 1.6% of the total area of the
region (highway verges: 0.4%, road verges:
1. Study area (Figure 1) 1.2%). In the same way, field margins
(margins comprised between two fields) also
The study area is in the Ile-de-France region represent 1.5% of the territory (estimations
located in the centre of France around Paris obtained using GIS data provided by the
(48° 51′ North - 2° 21′ East). The study site is Institute for Urban Planning and
the intensive agricultural zone of the suburb Developpement of the Paris Ile-de-France
of Paris. It is thus mainly composed of Region, I.A.U.R.I.F). Road network verges
agricultural fields used for intensive crop: (highway or roadside) and field margins are
wheat, sugar beet, rapeseed (49.0%). herbaceous strips mowed once or twice a year
Urban areas represent 18.2% and the and differ mainly in width (15.7 ± 6.6 m [10.0
main “natural” habitat is woodland - 21.5] for highway verges, HV, 7.2 ± 3.4 m
(representing about 23.7% of landcover; [4.3 - 14.6] for road verges, RV, and 4.0 ± 0.4
Appendix 1). m for [2.0 – 10.0] for field margin, FM).
FIGURE 1:
SAMPLING DESIGN
The study area is located in central France in the Ile-de-France region (Paris area). Non-attractive
traps (Barber pitfall traps) were installed in 297 sites of different habitats (marginal areas, woods
and fields) to capture small mammals during one month in May 2006.
- 47 -
MANUSCRIPT N°2
2. Sampling design (Figure 1) 4. Data analysis
To assess the potential habitat or corridor a. Verges and margins as habitat

roles of road verges, we sampled the two We tested the variation in the distribution of
main types of road network verges: 31 sites two main small mammals species across the
along highways (noted HV) and 48 sites along sampled habitats (HV, RV, FM, F1, F2 &
ordinary national roads (noted RV). On a site, WD), Microtus arvalis and Sorex coronatus,
a sampling unit is composed of five traps, with an ANOVA on species abundances
linearly placed every 20m along highways (average number of individuals by trap in
and roads in the middle of banks. Each trap each site)
was further referenced with a letter according
to its position in the linear design (A to E); b. Verges and margins as corridor
given the small mammal movements, the five We ran ANOVAs to test two hypotheses
traps were not considered as independent possibly proving corridor function of verges
sampling replicates in further analyses. and margins.
To measure the relative importance of (1) If small mammals use verges as corridor,
verges for small mammals in the landscape, they preferentially fall in the two extreme
we also sampled field margins (65 sites, noted traps (A & E). We thus tested the trap position
FM, with three traps along each field, and a effect on the two species (for HV & RV sites).
distance between traps of at least 100 meters); (2) S. coronatus is a territorial species (Neet
field borders (65 sites, noted F1, with two and Hausser 1990) and its populations
traps placed 25m into the field), central fields densities are about 10.8 individuals per ha
(56 sites, noted F2, with two traps placed 50m (Cantoni 1993), i.e. a territory can be
into the field) and woods (32 sites, noted WD, approximately represented by a disk of
with one trap randomly placed). A total of 926m². Because of their territorial behaviour,
864 traps were installed in May 2006 and if shrews move among their territories located
collected one month later. Some of the traps in natural habitats (meadows or woods)
were destroyed during the sampling period, so through the verges, they should more often
that a total of 813 traps were analysed from visit verge sites in the neighbourhood of
297 sites (Table 1). natural sites connected to the verges. Thus,
we calculated the distance between each verge
site (HV & RV sites) and the nearest adjacent
3. Small mammal captures and connected meadow or wood (distance noted
determinations DC, Figure 2), and further sorted verge sites
in two categories: sites close to a natural
We used Barber pitfall traps, as 450 mL habitat (sites A, i.e. closer than ten times a
plastic pots (10.2 cm height & 7.5 cm territory diameter – equivalent to 343m), and
diameter) inserted into the ground. Each trap sites far from a natural habitat (sites B, i.e.
contained a 155 mL solution composed with further distant than ten times a territory
75 mL of water, 75 mL of conservative diameter). We then ran ANOVAs to test
(monopropylene glycol), 5 mL of surfactant differences of species abundance between the
(dishware soap) and 15 g of NaCl. Traps were two categories of sites.
protected from rain by plastic transparent
covers (15 cm×15 cm) supported by wood c. Highway and roadside verges as barrier
sticks 10 cm above the ground. Sampling was for Sorex coronatus
carried out from 2 May to 4 June 2006. All If roads and highways act as barriers for S.
captured mammals were dissected and coronatus and if the previous hypothesis is
identified from morphological cranial verified (i.e. significant effect of distance to
characters, such as teeth structures (Erome & nearest connected habitat, DC, on shrews
Aulagnier 1982, Chaline et al. 1974). abundances), we should not find any effect of
- 48 -
MANUSCRIPT N°2
distance to the nearest natural habitat located animals i.e. 34.6% of the total), and an
on the other side of the road on the abundance insectivorous shrew, S. coronatus (178 trapped
of S. coronatus. Thus we calculated the animals i.e. 43.6% of the total). Given the
distance between each verge of RV and the sampling sizes obtained, we decided to focus
next natural habitat of the other side of the our analysis on these two species (Table 1).
road (distance noted DX, Figure 2). As
previously, we sorted verges in two categories
(A: <343m and B: >343m) and ran an ANOVA 1. Verges and margins as habitat
to test the difference in shrew abundances
between sites of the two categories. The ANOVA showed (Z = 1.058 and P =
0.290, Figure 3A) that M. arvalis is equally
FIGURE 2: abundant in highway verges and roadside
DISTANCES CALCULATION verges (HV: 0.24 ± 0.05 SE & RV: 0.30 ± 0.02
SE ind.trap-1).
In field habitats, we found lower vole
densities (FM: 0.10 ± 0.05 SE, F1: 008 ± 0.04
SE & F2: 0.07 ± 0.03 SE ind.trap-1). Even if
field margins are very similar habitats to
verges, M. arvalis was less captured there
than in highway verges (Z = 3.160 and P =
0.002) and roadside verges (Z = 4.255 and P <
10-3). No differences were observed between
field habitats (FM vs. F1: Z = -0.398 and P =
0.691; FM vs. F2: Z = -0.755 and P = 0.450;
& F1 vs. F2: Z = -0.367 and P = 0.713). M.
arvalis was not captured in woods (WD).
The distribution of S. coronatus
(Figure 3B) varied significantly among
habitats. This species was abundant in verges
Distances between capture sites and nearest especially in highway verges (HV: 0.63 ± 0.08
connected natural habitat (DC) or opposite SE) where S. coronatus was more captured
nearest natural habitat (DX). than in roadside verges (Z = -5.444 and p <
10-3, RV: 0.26 ± 0.05 SE ind.trap-1). It was
All statistical analyses were performed with detected in woods (WD: 0.16 ± 0.08 SE
the R software (Ihaka & Gentleman 1996) ind.trap-1) where its abundance did not differ
using ANOVAs and considering Poisson- from that in roadside verges (Z = 0.427 and P
distributed models. = 0.431).
In field habitats, shrew abundances
were very low. S. coronatus abundances in
field margins were less important than in
III- RESULTS roadside verges (Z = -3.337 and P < 10-3) or
in woods (Z = -2.802 and P =0.005).
A total of eight species were captured among Abundances appeared as a weak function of
the different habitats: three shrews (Crocidura distance to margins: a tendency was observed
russula, Sorex coronatus and S. minutus), but no differences were significant (Z = 1.689
three voles (Clethrionomys glareolus, and P = 0.091) between abundances in field
Microtus agrestis and Microtus arvalis) and margin (FM: 0.05 ± 0.02 SE ind.trap-1) which
two mice (Apodemus flavicollis and appeared to be higher and abundances in
Apodemus sylvaticus). Two of these species field-25 (F1: 0.01 ± 0.01 SE ind.trap-1). No
represented more than 78% of the collected any S. coronatus was captured in field-50m.
specimen: a rodent, M. arvalis (141 trapped
- 49 -
MANUSCRIPT N°2
TABLE 1:
Species detected in the different habitats are distributed between shrews (Croci1dura russula, S.
coronatus and S. minutus), voles (Clethrionomys glareolus, M. agrestis and M. arvalis) and mice
(Apodemus flavicollis and Apodemus sylvaticus). The table presents the number of individuals by
species captured in each habitat. Only recovered traps are considered (some traps were destroyed
during the experiment).
Trap by Traps mice voles shrews

Zones Habitats Sites Total
site recovered (1) (2) (3) (4) (5) (6) (7) (8)
Natural
Woods 32 1 32 - 1 2 2 - 2 6 0 13
areas
Highway
31 5 155 - 4 - - 37 13 98 6 158
verges
Marginal Road
48 5 240 5 8 - 4 71 22 63 - 173
areas verges
Field
65 3 187 - 8 - - 18 5 10 - 41
margins
Fields
65 2 110 - 5 - - 9 1 1 - 16
25m
Fields
Fields
56 2 89 - 3 - - 6 - - - 9
50m
Total 6 297 813 5 29 2 6 141 43 178 6 410
(1) Apodemus flavicolli, (2) Apodemus sylvaticus, (3) Clethrionomys glareolus, (4) Microtus
agrestis, (5) Microtus arvalis, (6) Crossidura russula, (7) Sorex coronatus & (8) Sorex minutus.
FIGURE 3:
HABITAT ABUNDANCES
A. M. arvalis abundances by habitats B. S. coronatus abundances by habitats
Key
Variations in the habitat distribution of (A) M. arvalis and (B) S. coronatus. Groups labelled with
different letters differed significantly.
- 50 -
MANUSCRIPT N°2
2. Verges and margins as corridors connected natural habitat (Z = -5.048 and P <
0.001).
a. Trap position in verges
We found no effect of trap position on M. 3. Roads as barrier for shrews
arvalis distribution in highway verges (Z =
0.268 and P = 0.788) and in roadside verges Because we did not found any effect of
(Z = 0.339 & P = 0.735; Figure 4A). connected distance (DC) on shrew
For the shrew, we found no effect of abundances in highway verges, we focused
trap position in highway verges (Z = 1.681 this analysis on road verges. We found no
and P = 0.093; Figure 4B), but an effect was correlation between shrew densities and the
detected in roadside verges (Z = -3.890 and P distance to the nearest connected natural
< 10-3; Figure 4B): more shrews were habitat situated on the opposite side of road
captured in traps located at the extremity of (Z = -0.708 and P = 0.479, Figure 5B).
the trap lines (A & E: 0.43 ± 0.11 SE ind.trap-
1
) compared to middle traps (C, D & E: 0.15 ±
0.03 SE ind.trap-1). IV- DISCUSSION
b. Distance to connected natural habitats First, confirming Lowell & Geis observations
(DC) for S. coronatus (1983), we found higher densities of small
As previously, results differed if considering mammals in verge habitats compared to
highway verges or road verges (Figure 5A). neighbouring habitats (woods, crop fields and
For highway verges, the distance to the field margins). This demonstrates the
nearest connected natural habitat had no importance of these marginal zones for small
significant effect on S. coronatus abundance mammals in intensive agrarian landscapes.
(Z = 0.892 and P = 0.372). For road verges, Nevertheless, whereas frequented by voles
abundance of S. coronatus was negatively (M. arvalis) and shrews (S. coronatus), such
correlated to the distance to the nearest margins have not similar roles for the two
FIGURE 4:
TRAP POSITION EFFECTS
A. M. arvalis B. S. coronatus
No effects of trap position on M. arvalis captures (RV: Z = 0.339 & P = 0.735, HV: Z = 0.268 & P
= 0.788) and on S. coronatus captures in highway verges (Z = 1.681 & P = 0.093). Traps are not
independent for the shrew in road verges (Z = -3.890 & P < 10-3). Groups labelled with different
letters differed significantly.
- 51 -
MANUSCRIPT N°2
species. These results confirm that the habitat degradation and agriculture
ecological function of different elements of intensification (La Peña (de) et al. 2003).
landscapes is strongly species dependant On one hand, road verges appeared as
(Noss 1987, Beier & Noss 1998). habitats for shrews, species potentially
important in term of ecosystem services for
agriculture (regulation of pest insects), but on
1. Verges as habitat for voles and shrews another hand, road verges appeared as
habitats for voles, species considered as pests
Because (1) the vole M. arvalis was detected for agriculture. A future outlook of this study
in margins and in fields and because (2) there could be to study how simultaneously favour
was no evidence of dispersion along roads or shrew populations and limit vole populations,
highways (no effect of trap position), we through specific managements or structure of
consider that voles may use verges as verges.
habitats, which is in accordance with the
biology of rodent species feeding in fields.
We assume that highway verges 2. Verges as corridors for shrews
provide an effective habitat for S. coronatus
because (1) we trapped a large number of We observed differences in abundance
individuals, and because (2) shrews were patterns of S. coronatus in road verges.
quasi-absent from crop fields. The suitability Actually their abundance was negatively
of highway verges as habitat for shrews is correlated to the distance to the nearest
confirmed by our observations on the adjacent natural habitat, and we found a
distribution of another shrew species, Sorex further effect of trap position demonstrating
minutus: considering all traps, we only found linear movements of shrews along the road,
this species in highway verges, while this while shrews were quasi-absent from crop
shrew is considered as highly sensitive to fields. We rather assume that roadside verges
FIGURE 5:
DISTANCE TO NATURAL HABITATS EFFECTS ON S. CORONATUS ABUNDANCES
A. B.
In highway verges, (A) we found no significant correlation between S. coronatus abundances and
distance (DC) between capture sites and the nearest connected natural habitat (HV: Z = 0.892 & P
= 0.372) but this correlation was significant for road verges (Z = -5.048 & P < 0.001) and (B)
disappeared (Z = -0.708 & P = 0.479) considering the distance (DX) between capture sites and the
nearest adjacent natural habitat. Groups labelled with different letters differed significantly.
- 52 -
MANUSCRIPT N°2
could be used as corridor for dispersion and used forests. For Swihart & Slade (1984),
are lower habitat quality than highway verges, roads were crossed exclusively by medium
at least for S. coronatus. These results are in size animals. They demonstrated that even a
accordance with the biology of that species, a dirt track may act as a major barrier to prairie
territorial animal which moves for various voles Microtus ochrogaster but that crossings
purposes. During the breeding season, males of this track by cotton rats Sigmodon hispidus
disperse to find mates (Cantoni 2002). And were density dependant and much lower than
because S. coronatus is a very active shrew expected. Brehme (2003) found that different
(Genoud 1984), its daily energy expenditure small mammals were running along roads
is important (Genoud 1985), which may push rather than across and that only juveniles were
them to forage in roadside verges as access to abserved to cross minor roads. Oxley et al.
food or food supply are increased on road (1974) reported that the smallest road
networks (Oxley et al. 1974). clearance (less than 3m) can inhibit some
Because Tichendorf & Wissel (1997) small mammal movements, a result further
found that corridor widths have strong effects confirmed by Merriam et al. (1989) who
on small mammals dispersion and because showed that mice movements across roads
highway verges and road verges are very were very infrequent although movements
similar excepting for their structures (highway adjacent to roads were important and longer
verges, 15.7 ± 6.6m, are larger than road than road width.
verges, 7.2 ± 3.4m), we assume that those
differences in verge uses between highways
(habitat) and roads (corridor) by S. coronatus 4. Management policies and small
may be explained by their widths. In mammals
accordance with this hypothesis, we also
noted that field margins have the smallest With the new European Common Agricultural
widths (4.0 ± 0.4m) and hold the lowest Policy (C.A.P.) from 2002 onwards,
densities of the species (see Figure 3). environmental objectives to preserve
biodiversity were introduced into the process
of agrarian subsidies distribution: farmers
3. Roadside verges as barrier for shrews have to choose between different
environmental practices to obtain some of the
In roadside verges, we found no relation subsidies. One of those practices is the set-up
between shrew abundance and the distance to of new field margins around crops. This
the nearest natural habitat adjacent to the practice is expected to increase biodiversity in
opposite side of the road, though such a agricultural landscape. Such policies should
correlation was positive when considering be maintained and strongly encouraged
natural habitats on the same side of the road. because they may also increased ecosystem
We interpret these contrasting results as services maintaining population of agriculture
indicating some barrier effect of roads for pest predators such as shrew which forage
shrews. mainly on beetle, caterpillar and slug (Bellocq
Kirby (1997) suggested that roads may & Smith 1994, Churchfield 1982, Pernetta
be greater barriers for small mammals than 1976, Maisonneuve & Rioux 2001) and
their width may suggest, because of their stabilising vole dynamics by avoiding cycling
particularly unfriendly habitat. This outbreaks (Delattre et al. 1999).
assumption has been verified in many studies Another actual policy is installation of
which demonstrated that small mammals are road underground passages for the small
not able to cross roads and tracks: Mader fauna. Because (1) we showed that all roads
(1984) reported how rarely yellow necked are barriers for small mammals, because (2)
mice Sylvaemus flavicolis and bank voles we found that shrews densities are highly
Clethriomys glareolus crossed a variety of correlated to the distance to connected natural
roads from busy highways down to weakly habitats, because (3) it has been proved that
- 53 -
MANUSCRIPT N°2
culverts have strong effects on shrew Developpement and Spacial Planning,
dispersal between the two sides of the roads M.E.E.D.D.A.T.).
(Yanes et al. 1995) and because (4) they limit
small mammal road casualties (Lodé 2000).
We therefore recommend to purchase such REFERENCES
politics of underground passage installations
in order to aid the maintenance of shrew Beier P, Noss RF (1998) Do habitat corridors
communities in verges. provide connectivity? Conservation
According to Meunier et al. (1999) Biology 12: 1241-1252.
and because verges represent important Bellamy PE, Shore RF, Ardeshir D, Treeek
conservation areas for small mammals, it JR, Sparks TH (2000) Road verges as
should be also interesting to encourage habitats for small mammals in Britain.
extensive policies of management of highway Mammals Revue 30: 131-139.
and roadside verges in intensive agrarian Bellocq I, Smith SM (1994) Arthropods
landscapes. preferred as food by Sorex cinereus
(masked shrew) and Peromyscus
mariculatus (deer mouse): an
ACKNOWLEDGEMENTS experimental approach. Mammalia 56:
391-396.
We thank the Institut d'Aménagement et Bennett AF (1990) Habitat corridors and the
d'Urbanisme de la Région d’Ile-de-France conservation of small mammals in a
(Institute for Urban Planning and fragmented landscape forest
Developpement of the Paris Ile-de-France environement. Landscape Ecology 4:
Region, I.A.U.R.I.F.) and the Direction 109-122.
Départementale de l’Equipement of Seine-et- Brehme CS (2003) Response of small
Marne (Equipement Department Agency, terrestrial vertebrates to roads in a
D.D.E. 77) who provided GIS data and coastal sage scrub ecosystem. PhD
technical support to protect the field work on thesis. Faculty of San Diego State
road verges. We thank also François Hervieu University.
from the Ministry of Agriculture and the Buckner CH (1969) Some aspects of the
SRPV (Service Régional de Protection des population ecology of the common
Végétaux Ile de France). We thank also shrew, Sorex araneus, near Oxford,
Jeremy CASTELLI, Elise CONTAN, Ondine England. Journal of Mammalogy 50:
FILIPPI-CODACCIONI, Grégoire LOIS and 326-332.
Rose-Line PREUD’HOMME who helped us Cantoni D (2002) Social and spacial
during field works. organization of free-ranging shrews,
Sorex coronatus and Neomys fodiens
(Insectivora, Mammalian). Animal
FUNDING behaviour 45: 975-995.
Chaline, J., Baudvin, H., Jammot, D., Saint-
The research was supported financially by Girons, M.C., 1974. Les proies des
Cofiroute, the Conseil général de Seine-et- rapaces. Ed. Doin, Paris, France.
Marne (Seine-et-Marne Department Churchfield F (1982) Food availability and
Assembly, C.G. 77) and the Direction the diet of the common shrew Sorex
Générale des Routes (Roads General araneus in Britain. Journal of Animal
Direction, D.G.R.) of the Ministère de Ecology 51: 15-28.
l’Ecologie, de l’Energie, du Développement et Churchfield S, Brown VK (1987) The trophic
de l’Aménagement Durables (French Minister impact of small mammals in
for Ecology, Energy, Sustainable successional grasslands. Biological
- 54 -
MANUSCRIPT N°2
Journal of the Linnean Society 31: Microtus pennsylvanicus. Journal of
273-290. Mammalogy 59: 208-212.
Churchfield S, Hollier J, Brown VK (1991) Hansen K, Jensen J (1972) The vegetation on
The effects of small mammal roadsides on Denmark: a qualitative
predators on grassland invertebrates, and quantitative comparison. Dansk
investigated by field exclosure Botanisk Arkiv 28: 21–61.
experiment. Oikos 60: 83- 290. Harris LD, Silva-Lopez G (1993) Forest
Davies ZG, Pullin AS (2007)Are hedgerows fragmentation and the conservation of
effective corridors between fragments biological diversity. In P. L. Fiedler
of woodland habitat? An evidence- and S. K. Jain, editors. Conservation
based approach. Landscape Ecology Biology. Chapman and Hall, New
22: 333-351. York. 171–196.
Dawson DG (1994) Are habitat corridors Hodkinson DJ, Ken Thompson K (1997)
conduits for animals and plants in a Plant dispersal: The role of man. The
fragmented landscape? A review of Journal of Applied Ecology. 34: 1484-
the scientific evidence. English Nature 1496
Research Report 94. Ihaka R, Gentleman R (1996) R: A Language
Dawson DG (2002) Roads and habitat for Data Analysis and Graphics.
corridors for animals and plants. In: Journal of computational and
Imperial Colledge Press. Wildlife and graphical statistics 5: 299-314.
roads, The ecological impacts: Kirby KJ (1997) Habitat fragmentation and
Sherwood D., Culter D. & Burton J. infrastructure: problem and research.
186-198. In: Canters K eds. Habitat
Delattre P, De Sousa B, Fichet-Calvet E, fragmentation and infrastructure.
Quéré JP, Giraudoux P (1999) Vole Delft: Ministry of Transport, Public
outbreaks in a landscape landcape Works and Water Management. 32-39.
context: Evidence from a six year La Peña (de) NM, Butet A, Delettre Y, Paillat
study of Microtus arvalis. Landscape G, Morant P, Le Du L, Burel F. (2003)
Ecology 14: 401-412. Response of the small mammals
Dial R, Roughgarden J (1995) Experimental community to changes in western
removal of insectivores from rain French agricultural landscapes.
forest canopy: Direct and indirect Landscape Ecology 18: 265-278.
effects. Ecology 76 : 1821-1834. Lodé T (2000) Effetcts of a mortorway on
Erome G, Aulagnier S (1982) contribution à mortality and isolation of wildlife
l’identification des proies des rapaces. populations. Ambio 29: 163-166.
Bièvre 4: 129-135. Lowell WA, Geis AD (1983) Effects of roads
Forman RTT, Alexander LE (1998) Roads on small mammals. Journal of Applied
and their major ecological effects. Ecology 20: 403-415.
Annual Review of Ecology and Mader HJ (1984) Animal habitat isolation by
Systematics 29: 207-231. roads an agricultural field. Biological
Genoud M (1984) Activity of Sorex coronatus Conservation 29: 81-96.
(Insectivore, Soricidae) in the field. Maisonneuve C, Rioux S (2001) Importance
Zeitschrift fuer Saeugetier 49: 74-78. of riparian habitats for small mammals
Genoud M (1985) Ecological energetics of and herpetofaunal communities in
two European shrews: Crocidura agricultural landscapes of southern
russula and Sorex coronatus Québec. Agriculture, Ecosystems and
(Soricidae: Mammalia). Journal of Environment 83: 165-175.
Zoology 207, 63-85. Merriam G, Kozakiewicz M, Tsuchiya E,
Getz LL, Cole FR, Gates DL 1978. Interstate Hawley K (1989) Barriers as
Roadsides as Dispersal Routes for boundaries for metapopulations and
- 55 -
MANUSCRIPT N°2
demes of Peromyscus leucopus in grassland. Journal of Animal Ecology
farm landscapes. Landscape Ecology 45: 899-912
2: 227-235. Ries L, Debinski DM, Wieland ML (2001)
Merriam G, Lanoue A (1990) Corridor use by Conservation Value of Roadside
small mammals: field measurement Prairie Restoration to Butterfly
for three experimental types of Communities Conservation Biology
Peromyscus leucopus. 15, 401-411.
Meunier FD, Corbin J, Verheyden C, Rico A, Kindlmann P, Sedlacek F (2007)
Jouventin P (1999) Effects of Barrier effects of roads on movements
landscape type and extensive of small mammals. Folia zoological
management on use of motorway 56, 1-12.
roadsides by small mammals. Schoener TW (1988) Leaf damage in island
Canadian Journal of Zoology 77, 108- buttonwood, Conocarpus erectus:
117. Correlations with pubescence, island
Neet CR, Hausser J (1990) Habitat selection area, isolation and the distribution of
in zones of parapatric contact between major carnivores. Oikos 53: 253-
the common shrew Sorex araneus and 266.
Millet’s shrew S. Coronatus. Journal Spille DA, Schoener TW (1990) A terrestrial
of Animal Ecology 59, 235-250. field experiment showing the impact
Noss RF (1987) Corridors in Real of eliminating top predators on foliage
Landscapes: A Reply to Simberloff damage. Nature 347: 469-472.
and Cox. Conservation Biology 1: Spooner PG Lunt ID, Briggs SV (2004)
159-164. Spatial analysis of anthropogenic
O’Farrell PJ, Milton SJ (2006) Road Verge disturbance regimes and roadside
and Rangeland Plant Communities in shrubs in a fragmented agricultural
the Southern Karoo: Exploring What landscapes. Applied Vegetation
Influences Diversity, Dominance and Science 7: 62-70.
Cover. Biodiversity and Conservation Suckling GC. (1984) Population ecology of
15, 921-938. the sugar glider, Petaurus breviceps,
Obrtel R, Zejda J, Holisova V, (1978) Impact in a system of fragmented habitats.
of small rodent predation on an Australian Wildlife Research 11: 49-
overcrowded population of Diprion 75.
pini during winter (pines). Folia Swihart RK, Slade NA (1984) Road crossing
Zoologica 27: 97-110. in sigmondin hispidius an Microtus
Oxley DJ, Fenton MB, Carmody GR (1974). ochrogaster. Journal of Mammalogy
The effects of roads on populations of 65: 2 357-360.
small mammals. Journal of Applied Tikka PM, Högmander H, Koski PS (2001)
Ecology 11: 51-59. Road and railway verges serve as
Parmenter RR & MacMahon, JA (1988) dispersal corridors for grassland
Factors limiting populations of arid- plants. Landscape Ecology 16: 659-
land darkling beetles (Coleoptera: 666.
Tenebrionidae): predation by rodents. Tischendorf L, Wissel C (1997) Corridors as
Animal Ecology 17: 280-286. conduits for small animals: attainable
Pasquier-Barre (1999) Diprion pini/pinus distances depending on movement
sylvestris relationship foliage pattern, boundary reaction and
characterizations and biological corridor width. Oikos 79: 603-611.
consequences. PhD thesis. University Trombulak SC, Frissell CA (2000) Review of
of Paris 06, France. ecological effects of roads on
Pernetta JC (1976) Diet of the shrew Sorex terrestrial and aquatic communities.
araneus and Sorex minutus in Wytham Conservation Biology 14: 18-30.
- 56 -
MANUSCRIPT N°2
Verkaar HJ (1990) Corridors as a tool for APPENDIX:
plant species conservation? Species LANDCOVER OF THE STUDY AREA
Dispersal in Agricultural Habitats. Eds Area
RGH Bunce & DC Howard,. 82– 97. Landcover Percentages
(km²)
Belhaven Press, London. Urban Buildings 2083 17.3%
18.2%
Watts RD, Compton RW, McCammon JH, areas Transports 104 0.9%
Rich CL, Wright SW, Owens T, Ouren Woods 2850 23.7%
Natural
DS (2007) Roadless Space of the areas
Meadows 587 4.9% 29.7%
Conterminous United States. Science Water 148 1.2%
316: 736-738. Highway
Marginal verges 46 0.4%
Way JM (1977) Roadside verge and 3.1%
areas Road verges 144 1.2%
conservation in Britain: A review. Field margins 180 1.5%
Biological Conservation 12: 65– 74. Fields 49.0%
5908 49.0%
Yanes M Velasco J, Suarez, F (1995)
permeability of roads and railways to Ile-de-France region is mainly composed by
vertebrates: The importance of fields, forests and urban areas. Marginal
culverts. Biological Conservation 71: areas as field margins, roadside and highway
217-222. verges represent non-negligible areas.
Sorex coronatus
- 57 -
PARTIE II
STRUCTURES ET MODES DE GESTION DES ACCOTEMENTS ROUTIERS :

IMPACTS SUR LA BIODIVERSITE
CORRESPONDANCES
La Nature est un temple où de vivants piliers

Laissent parfois sortir de confuses paroles :
L'homme y passe à travers des forêts de symboles
Qui l'observent avec des regards familiers.
Comme de longs échos qui de loin se confondent

Dans une ténébreuse et profonde unité
Vaste comme la nuit et comme la clarté,
Les parfums, les couleurs et les sons se répondent.
Il est des parfums frais comme des chairs d'enfants,

Doux comme les hautbois, verts comme les prairies,
Et d'autres, corrompus, riches et triomphants,
Ayant l'expansion des choses infinies,

Comme l'ambre, la muse, le benjoin et l'encens,
Qui chantent les transports de l'esprit et des sens.
Charles BAUDELAIRE, LES FLEURS DU MAL (1857)
- 58 -
MANUSCRIPT N°3
Submitted to Journal of Environnemental Management
Effect of delayed single mowing on roadsides vegetation communities:

A three years study
Louis de REDON
Frédéric JIGUET
Jeffrey B. JOY
&
Nathalie MACHON
- 59 -
MANUSCRIPT N°3
PRESENTATION
TITRE
Effets d’une fauche unique et tardive des accotements routiers sur les communautés
végétales des bords de route: Trois ans d’étude
RESUME
Au sein des paysages agricoles, les bords de route peuvent constituer un des rares refuges
pour la flore locale. Comme la gestion et l’entretien des dépendances vertes peuvent impacter
les communautés végétales des bords de route, nous avons étudié les modifications en termes
de diversité spécifique et fonctionnelle au sein de ces communautés durant trois années après
un changement du nombre de fauches réalisées par les gestionnaires de routes en région
parisienne (FRANCE).
La réduction à une fauche annuelle réalisée en fin de saison (septembre) a eu des effets
très forts sur les communautés végétales des bords de route. Premièrement, la diversité
végétale a augmenté fortement et rapidement. Deuxièmement, les communautés végétales ont
fortement répondu à ces nouvelles pratiques par de profonds changements au niveau de leur
organisation fonctionnelle : la gestion extensive des bords de route a par exemple favorisé les
plantes bisannuelles et zoochores.
Avec cette expérimentation, nous avons pu montrer comment les politiques de gestion
peuvent être conciliées avec des politiques de conservation de la diversité végétale tout en
permettant des économies financières aux autorités publiques. Cette expérimentation
nécessiteraient cependant d’être poursuivies afin de mesurer les effets à plus long terme de
telles pratiques de gestion extensives des bords de route.
MOTS CLEFS
ACP, Gestion des bords de route, Indices de Hill, Politiques environnementales, Traits
fonctionnels.
- 60 -
MANUSCRIPT N°3
EFFECT OF DELAYED SINGLE MOWING
ON ROADSIDES VEGETATION COMMUNITIES: A THREE YEARS STUDY
Louis de REDONa*, Frédéric JIGUETa,1, Jeffrey B. JOYb,2 and Nathalie MACHONa,3.
KEYWORDS SUMMARY
Environmental policies In intensive agrarian landscape, roadside verges constitute one
Functional traits of few refuges for wild plant species. Because roadside
Hill index management shapes plant communities on verges, we studied
PCA the species and functional diversity of roadside plant
Road verges management communities during the three years following a reduction in the
number of times roadsides were mowed in the region of Paris
(France). The reduction of mowing to one single late cut had a
strong effect on roadside plant communities. First, plant
diversity increased strongly and rapidly, though this effect was
variable between years and thus requires long term monitoring
to understand its significance to the plant community over time.
Second, plant communities responded with deep changes in
their structure with reduced mowing favouring apparition of
bisannuals plants and species whose seeds are animal dispersed.
With this experiment, we showed how management policies are
able to conciliate plant diversity protection with public finance
savings.
1. INTRODUCTION could play a significant role in plant

conservation. Roadsides are known
In intensive agrarian landscape, plant toconstitute one of the main refuges for
diversity is low and mainly concentrated in wild plant species in agrarian landscape
field margins and roadsides (Way, 1977; (Bennett, 1991; Akbar, 1997). Usually,
Sỳkora et al., 1993; Burel et al., 1998). vegetated roadsides are only managed for
However, roads are known to have traffic visibility and esthetical reasons; to
deleterious effects on nearby plant fulfill those objectives, the vegetation on
communities (Forman and Alexander, banks and verges are mowed regularly by
1998). The metropolitan French main and the road services. The number of cuts
secondary road networks comprise generally depends on availability of
approximately 1,000,000 km long (1.82 manpower and machines (Genard et al.,
km of roads per km2) and roadside verges 1994). In France, ordinary management of
cover approximately 5,100 km2, an area roadsides usually includes three cuts a
equivalent to 1% of the area of France year. We posit that roadside management
(544,000 km2), more than the total area of and in particular the number of cuts per
French National Parks which represent year has a strong effect on roadside plant
3,400 km2. If roadside vegetation communities (Parr and Way, 1988; Sỳkora
management combines visibility and et al., 2002).
environmental objectives, roadside verges
* Corresponding author, Tel: (+33) 662 045 936, Fax: (+33) 140 79 38 35, E-mail:
redon@mnhn.fr – a: UMR 5173 Conservation des Espèces, Restauration et Suivi des
Populations, Muséum National d’Histoire Naturelle, 55 rue Buffon, F-75005 Paris, France –
b: Department of Biological Sciences, Simon Fraser University, Burnaby, B.C., CANADA.
- 61 -
MANUSCRIPT N°3
In order to provide managers with FIGURE 1
the best advice to preserve biodiversity, we STUDY AREA
sought to investigate the potential impact
of a single light and delayed yearly
mowing on roadside communities which
functional diversity of plant communities
during three years with a reduction of the
number of cuts in roadsides, compared to
roadsides still subjected to three cuts. The
experiment was conducted in the region of
Paris (France), a typical intensive agrarian
landscape composed almost exclusively of
agricultural fields and groves.
2. MATERIALS AND METHODS
2.1. Study area
In France, roadside management followed

a departmental council policy. Each
department is divided into management
units. Our study area is situated in the
Fontainebleau management unit of the
Seine-et-Marne department, central France
about 70 km south-east from Paris (48° 24′ year (March, May and September). The
N, 2 42′ E). 51.2 % of the management bank is situated between the berm and the
unit area, 258.2 km², is located in a Natural adjacent habitat and is mown only once a
Regional Park (Parc naturel régional du year in September.
Gâtinais français). It is mainly composed From 2006 to 2008, the park and
of agricultural fields used for intensive the department council reduced the
crop production (33.6 % of landcover) and treatment of the berm to one single late cut
woods (54.0 %). Road networks (246 km in September (1C) on the southern
long, i.e. 2.34 km.km-2) represent surveyed sites. The other five (northern)
important linear structures in the landscape sites were given the usual, three cuts (3C) a
along with their verges (5.23 km², 2.02 % year; they are considered controls for the
of the total area of the department). experiment.
For each site, five plots of 1 m² (0.5
2.2. Sampling design x 2 m) were regularly placed every 20
meters along roads in the berms and in the
We sampled 10 sites: five roadsides in the banks. Vegetation was recorded in the
southern part of the natural regional park plots in early May 2006, 2007 and 2008.
(area 1) and five other roadsides in the We identified all plant taxa found in the
northern part of the park (area 2, Figure 1). plots and evaluated their abundance with a
The distance between site 1 and site 2 is corrected Braun-Blanquet method (Braun-
approximately 16 km, thus climate and Blanquet, 1932; Westhoff and Van der
local conditions are similar enough to Maarel, 1978). Almost all the taxa were
permit a comparison between the different identified to species level according to the
roadsides of the two areas. International Plant Names Index
Each roadside is divided into two (http://www.ipni.org). In some cases
zones: the berm and the bank. The berm, (Crepis sp. for example), taxa were
situated into the 2 to 3 first meters from the identified to genus level due to difficulties
pavement was usually mown three times a in identification after mowing.
- 62 -
MANUSCRIPT N°3
2.3. Data analysis 2.3.2. Changes in plant traits composition
2006-2008
2.3.1. Changes of diversity index 2006-
2008 Using data issued from three reference
books (Jauzein, 1998; Lambinon et al.,
For each plot, we calculated a corrected 2004; Lauber and Wagner, 2007), we ran
Hill diversity index (Hill, 1973) using the an analysis on five life history traits with
median recovery value (Pyšek 2004) of the their different modalities to: flowering
Braun-Blanquet scale, noted as E: period (FP<2: flowering period shorter
than two months; 2<FP<4: flowering
, period comprised between 2 and 4 months;
FP>4: flowering period longer than four
months), number of cotyledons (Monocots;
, Dicots), phenology (Annuals; Bisannuals;
Perennials), pollen dissemination (Anemo-
gamous; Autogamous; Entomogamous) and
, seed dispersal (Anemochorous; Autocho-
rous; Zoochorous) and we built a matrix
A: species vs. traits modalities. In each plot
. of berm zones, we calculated the total area
covered by each species and we built a
matrix B: plots vs. species covering. We
S is the number of species detected in a then calculated matrix B.matrix A to
plot, obtain a matrix C (plots vs. traits
Rn is the median Braun-Blanquet recovery modalities covering) needed to perform a
th
value of the n species of a plot, PCA on plots and their plant trait
th
pi’ is the relative abundance of i species compositions (Aubin et al. 2007) on the
in a plot. three years.
We tested the impacts of year
To have a more intuitive diversity (2006, 2007 and 2008) and treatment (1C
index, we calculated (Ey has vs. 3C) on plot coordinates on Axis 1 and
Axis 2 of the PCA (Model:
a maximum value for a maximum of AXIScoordinates ~ Year*Treatments).
diversity, and [2006, 2007, Using this method, we studied plant
2008]). community modifications following
We then ran ANOVAs to test the changes of mowing treatment in berms.
differences of E2007 and E2008 values We also investigated in 1C berms
between the two categories of berms (1 cut changes in the covering of some traits
vs. 3 cuts). Because northern and southern modalities (calculated for matrix C)
communities could not be considered according to the PCA results (traits
completely similar, samples of 2006 are modalities which seemed to be favoured by
used as reference data with E2006 (BERM) as delayed mowing). We tested the impacts of
control variables in each site (Model: Ey years on those changes (Model:
TraitModalityBERM ~ TraitModalityBANK +
(BERM) ~ + E2006 (BERM) + Year)
Treatments, ). We also 2.3.3. Changes in detection probalities
tested the differences of E2007 and E2008 2006-2008
values in banks between the two sites
(Model: Ey (BANK) ~ E2006 (BANK) + We calculated detection probabilities in all
Treatments, ). sites for the three years with ComDyn
- 63 -
MANUSCRIPT N°3
FIGURE 2
CHANGES OF CORRECTED HILL INDEX (E) DURING THREE YEARS
A. In berms B. In banks
software (Hines et al., 1999). For this significant between 3C and 1C berms (F3,49
calculation, we used the five local plots as = 10.134 and P = 0.003).
spatial replicates and we made the null
hypothesis that vegetation did not differ 3.2. Changes in plant traits composition
among plots in each site. 2006-2008
We verified that the observed
changes in plant communities (diversity PCA showed clear changes in the traits of
and traits composition) were not due to roadside plant communities according to
variation in detection probalities by testing their management during the three years of
the variation in detection probabilities investigation (Figure 3). PCA 1 explained
between years and mowing treatments. 33% of the total variance Figure 3) and
sites along this axis were highly correlated
All statistical analyses were performed with mowing treatments (t = -2.846 and P
with the statistical package R (Ihaka and = 0.0051). Significant relationships were
Gentleman, 1996). not found for PCA 2 (15.6%, t = 1.889 and
P = 0.0609) or PCA 3 (13.4, t = 1.025 and
P = 0.307).
3. RESULTS Some traits modalities seemed to be
favoured by the 1C treatment such as
We detected a total of 133 taxa (106 at the dicots, bisannuals, entomogamous and
species level; cf. Appendix 1in 300 zoochorous species (Table 1 and Figure 4).
vegetation sample plots during the three Plants with a long flowering period seemed
years of survey (NE (2006) = 86, NE (2007) = also to benefit from the management
73, NE (2008) = 85). changes.
Traits modalities covering the 1C
3.1. Changes in diversity index 2006-2008 berms changed significantly with
year.(Figure 5). We found a strong
In 2007, E values changed differently increase of bisannuals species in berms (t =
between the two categories of sites: they 2.491 and P = 0.0151), 2006: 11.9% ±
increased by 41 % in 1C berms and 0.04, 2007: 15.8% ± 0.03 and 2008: 24.4%
decreased by 4% in 3C berms (F3,49 = ± 0.04.
5.735 and P = 0.028; Figure 2). Zoochorous species also increased
In 2008, we observed a decrease of (t = 2.135 and P = 0.0362), 2006: 5.6% ±
E values in berms (3C sites: -12%; 1C 0.02, 2007: 6.9% ± 0.02 and 2008: 13.7%
sites: -5%). The difference of E values was ± 0.03).
- 64 -
MANUSCRIPT N°3
FIGURE 3 No significant differences were
PCA ON SITES AND COMMUNITIES TRAITS observed for entomogamous (t = 0.027 and
RICHNESS, SITE PROJECTION (AXIS 1 X 2) P = 0.979) or dicots (t = 1.398 and P =
A. 2006 0.166) species. Berm plants with long
flowering periods were also not affected by
the new mowing treatment (t = 0.060 and P
= 0.953).
2.3. Changes in detection probalities 2006-

2008
No changes in detection probabilities were

observed between years and mowing
treatments (t = -0.336 and P = 0.740).
TABLE 1
LIFE TRAITS COORDINATES ON PCA
A. Axis 1
B. 2007 Axis 1 (33.0%)
Coordinates Life traits
0,870 Dicots
0,785 Entomogamous
0,599 Bisanuals
0,497 Zoochorous
0,451 FP>5
0,359 Autochorous
0,175 Annuals
0,026 Autogamous
-0,068 FP<2
-0,440 2<FP<4
-0,536 Anemochorous
-0,631 Perenials
-0,802 Anemogamous
-0,871 Monocots
C. 2008 B. Axis 2
Axis 2 (15.6%)
Coordinates Life traits
0,708 2<FP<4
0,675 Autochorous
0,395 Perenials
0,317 Entomogamous
0,150 Autogamous
0,124 Dicots
0,015 Annuals
-0,125 Monocots
-0,174 Zoochorous
-0,226 FP<2
-0,315 Anemogamous
-0,390 Anemochorous
-0,436 Bisanuals
-0,627 FP>5
KEY: 3 cuts sites 1 cut sites
- 65 -
MANUSCRIPT N°3
FIGURE 4 Some trait modalities, like
PCA ON SITES AND COMMUNITY TRAIT entomogamous, dicots or long flowering
RICHNESS, TRAITS PROJECTION: AXIS 1 period, appeared correlated to PCA axis 1
(33.0%) VS. AXIS 2 (15.6%) in a way suggests they are favoured by the
light mowing treatments (Table 1).
Because no significant changes were
observed between years in 1C berms, and
because the observed changes are real and
not due to variations in detection
probability, those differences highlighted
by the PCA may be a result of site specific
differences between 1C and 3C sites which
are separated by approximately 16km.
The power of this multiple trait
approach could be limited because of the
non-independence (colinearity) of some
FIGURE 5
CHANGES IN TRAITS COMPOSITION
OF PLANT COMMUNITIES OF ONE
4. DISCUSSION CUT BERMS (1C)
The reduction of mowing to one single late cut
strongly influenced plant communities of the A. Bisannuals species
berms. First, plant diversity increased strongly
and rapidly in one year (by more than 40%),
representing a substantial gain in biodiversity
in these habitats. Such strong increases in
plant diversity maybe due to management
changes or a result of plant community
dynamics.
Further evaluation of the effects of the
1C management regime will prove informative
in determining the relative roles of
management and plant community dynamics.
Second, plant communities responded
with changes in their structure, and some trait
modalities were favoured by the management
changes. Notably, proportions of bisannuals B. Zoochorous species
and zoochorous plants within the communities
increased strongly during the three-year
experiment. The increase in bisannuals may be
explained by the fact that a single late cut may
allow them to finish their two year cycles,
whereas more intensive mowing treatments (3
cuts a year) may not. Zoochorous species
might be favoured because intensive mowing
treatments have strong negative impacts on
animal communities (Meunier et al., 1999).
This response is slower, continuous and very
important because functional responses at the
community level have a strong influence on
ecosystem processes (Violle et al., 2007).
- 66 -
MANUSCRIPT N°3
traits such as pollen dissemination and providing GIS data and technical support
seed dispersal. to protect on road verges.
Nevertheless, the multiple trait Jérémy CASTELLI, Elise
approach is of value because (1) while CONTAN, Amélie DELERUE, Rose-Line
some traits strongly covary among species, PREUD’HOMME and Noémie VARET,
others traits can show no correlation Master students, are thanked for their help
(Wardele et al., 1998) and (2) some during field works.
correlations between traits are only
seasonal (Eviner, 2004). Such
experimental management changes should REFERENCES
be carried out over a longer period to
observe the long term effect of extensive Aubin, I., Gachet, S., Messier, C.,
mowing of roadsides on plant Bouchard, A., 2007. How resilient
communities, as done for grazing are the northern hardwood forests to
experiments (Bullock et al., 2001). human disturbance? An evaluation
using a plant functional group
approach. Ecoscience 14, 259-271.
5. CONCLUSION Braun-Blanquet, J., 1932. Plant sociology :
the study of plant communities.
With this experiment, we highlighted the Graw-Hill Book Company. New
impact of management policies on local York.
biodiversity even in areas usually Bullock, J.M., Franklin, J., Stevenson,
considered as marginal for plant M.J., Silvertown, J., Coulson, S.J.,
conservation. In intensive agrarian Gregory, S.J., Tofts, R., 2001. A
landscape, management of roadsides could plant trait analysis of responses to
lead to higher species and trait diversity grazing in a long term experiment.
(Le Viol et al. 2008)and thus better Journal of Applied Ecology 38, 253-
ecosystem services (Couvet et al., 2006) 267.
for agriculture without additional cost. We Burel, F., Baudry, J., Butet, A., Clergeau,
thus strongly encourage public policies that P., Le Cœur, D., Dubs, F., Morvan,
are able to conciliate plant diversity N., Delettre, Y., Paillat, G., Petit, S.,
conservation with public finance savings. Thenail, C., Brune, E., Lefeuvre,
J.C., 1998. Comparative biodiversity
along a gradient of agricultural
ACKNOWLEDGEMENTS landscapes. Acta Oecologica 19, 47-
60.
The research was supported financially the Couvet, D., Dehorter, O., Henry, P.-Y.,
Direction Générale des Routes (Roads Jiguet, F., Julliard, R., 2006. La
General Direction, D.G.R.) of the biodiversité contre les maladies
Ministère de l’Ecologie, de l’Energie, du infectieuses ? Annales des Mines.
Développement et de l’Aménagement Responsabilité and Environnement
Durables (French Minister for Ecology, 41, 77-83.
Energy, Sustainable Developpement and Eviner, V.T., 2004. Plant traits that
Spacial Planning, M.E.E.D.A.T.). influence ecosystem processes vary
Institut d'Aménagement et independently among species.
d'Urbanisme de la Région d’Ile-de-France Ecology 85, 2215-2229.
(Institute for Urban Planning and Genard, M., Carsignol, J., Charaven, P.,
Developpement of the Paris Ile-de-France Cuenot, E., Coumoul, H., Desire, G.
Region, I.A.U.R.I.F.) and the Direction Lansiart, M., Martin, V., Payany, M.,
Départementale de l’Equipement of Seine- Thiollet, J.M., velluzzi, V., Brigeot,
et-Marne (Equipement Department V., 1994. La gestion extensive des
Agency, D.D.E. 77) are thanked for dependences vertes routières.
- 67 -
MANUSCRIPT N°3
SERRA, Ministère de l’Environ- term effects of cutting. Journal of
nement. Applied Ecology 25, 1073-1087.
Hill, M.O., 1973. Diversity and eveness: a Pyšek, P., Chocholoušková, Z., Pyšek, A.,
unifying notation and its Jarošík, V., Chytrŷ, M., and Tichŷ,
consequences. Ecology 54, 427-432. L., 2004. Trends in species diversity
Hines, J.E., Boulinier, T., Nichols, J.D., and composition of urban vegetation
Sauer, J.R., Pollock, K.H., 1999. over three decades. Journal of
COMDYN: software to study the Vegetation Science 15, 781-788.
dynamics of animal communities Rentch, J.S., Fortney, R.H., Stephenson,
using a capture-recapture approach. S.L., Adams, H.S., Grafton, W.N.,
Bird Study 46, 209-217. Anderson, J.T., 2005. Vegetation-site
Ihaka, R., Gentleman, R., 1996. R: A relationships of roadside plant
Language for Data Analysis and communities in West Virginia, USA.
Graphics. Journal of computational Journal of Applied Ecology 42, 129-
and graphical statistics 5, 299-314. 138.
Jauzein, P., 1995. Flore cultivée des Sỳkora, K.V., de Nijs, L.J., Pelmsa, H.M.,
champs. Sopra. INRA. 1993. Plantengemeenschappen van
Lambinon, J., Delvossalle, L., Duvigneaud, Nerderlandse. Natuurhistorische
J., 2004. Nouvelle Flore de la Vereninging.
Belgique, du Grand-Duché du Sỳkora, K., Kalwij, J.M., Keiser, P.J.,
Luxembourg, du Nord de la France 2002. Phytosociological and floristic
ème
et des Régions Voisines. 5 evaluation of 15-year ecological
Edition, Editions du Patrimoine du management of roadside verges in
jardin botanique national de the Netherlands. Preslia 74, 421-436.
Belgique. Way, J.M., 1977. Roadside verges and
Lauber, K., Wagner, G., 2007. Flora conservation in Britain: A review.
Helvetica.2ème Edition, BELIN. Biological Conservation 12, 65-74.
Le Viol, I., Julliard, R., Kerbiriou, C., Violle, C., Navas, M.L., Vile, D., Kazanou,
Redon (de), L., Carnino, N., E., Fortunel, C., Hummel, I., Garnier,
Machon, N., Porcher, E., 2008. E., 2007. Let the concept of trait be
Plant and spider communities functional! Oikos 116, 882-892.
benefit differently from the Wardle, D., Barker, G., Bonner, K.,
presence of planted hedgerows in Nicholson, K., 1998. Can
highway verges. Biological comparative approaches based on
Conservation 141, 1581-1590. plant ecophysiological traits predict
Meunier, F.D., Corbin, J., Verheyden, C., the nature of biotic interactions and
Jouventin, P., 1999. Effects of individual plant species in
landscape type and extensive ecosystems? Journal of Ecology 86,
management on use of motorway 405-420.
roadsides by small mammals. Westhoff, V., van der Maarel, E., 1978.
Canadian Journal of Zoology 77, The Braun-Blanquet approach. In:
108-117. Whittaker R H (ed.), Ordination and
Parr, T.W., Way, J.M., 1988. Management classification of plant communities.
of roadside vegetation: The long- Junk. The Hague.
APPENDIX 1
LIST OF DETECTED TAXAS
Achillea millefolium Agrostis stolonifera

Agrimonia eupatoria Allium sp.
Elymus repens Alopecurus myosuroides
- 68 -
MANUSCRIPT N°3
Anthriscus cerefolium Geum urbanum

Anthriscus communis Glechoma hederacea
Anthriscus vulgaris Hedera helix
Arhenaterum elatius Heracleum sphondylium
Artemisia vulgaris Hieracium sp.
Aster sp. Himenthoglossum hircinum
Bellis perennis Holcus mollis
Bromus erectus Hyacinthoides non-scripta
Bromus inermis Lamium purpureum
Bromus hordeaceus Lamium sp.
Bromus sp. Lathyrus pratensis
Bromus sterilis Lathyrus tuberosus
Capsella bursa-pastoris Ligustrum vulgare
Carex acutiformis Linaria vulgaris
Carpinus betulus Lolium perenne
Carex cuprina Lotus corniculatus
Centaurea jacea Medicago lupulina
Centaurea scabiosa Mentha suaveolens
Centaurea sp. Mercurialis sp.
Cerastium fontanum Ononis sp.
Cerastium sp. Papaver rhoeas
Chaerophyllum temulum Petroselinum sp.
Cichorium intybus Phleum pratense
Circium vulgare Phragmites sp.
Convolvulus arvensis Picris echioides
Conyza sumatrensis Picris hieracioides
Crepis sp. Plantago coronopus
Dactylis glomerata Plantago lanceolata
Daucus carotta Plantago major
Dipsacus fullonum Poa chaixii
Equisetum arvense Poa pratensis
Erophila verna Poa sp.
Euphorbia sp. Poa trivialis
Festuca arundinacea Polygonatum multiflorum
Festuca filiformis Potentilla reptans
Festuca rubra Primula veris
Festuca sp. Prunus avium
Ficaria sp. Prunus spinosa
Galium aparine Prunella vulgaris
Galium molugo Ranunculus acris
Galium parisiense Ranunculus arvensis
Galium sp. Ranunculus repens
Galium verum brassicées hyb-----
Geranium columbinum Roripa amphibian
Geranium dissectum Rubus fruticosus
Geranium molle Rubia peregrina
Geranium rotundifolium Rubus sp.
- 69 -
MANUSCRIPT N°3
Rumex acetosella Trisetum flavecens

Rumex crispus Trifolium pratense
Rumex sp. Trifolium repens
Senecio jacobaea Trifolium sp.
Senecio sp. Tussilago sp.
Senecio vulgaris Urtica sp.
Silene vulgaris Veronica arvensis
Silene latifolia Verbena officinalis
Silene sp. Vicia cracca
Solanum nigrum Vicia hirsuta
Sonchus asper Vicia sativa
Sonchus oleraceus Vicia sp.
Tanacetum sp. Vicia tetrasperma
Tanacetum vulgare Viola arvensis
Taraxacum officinalis Vulpia myuros
Taraxacum sp.
Almost all the taxa were identified to
Teucrium scorodonia species level according to the International
Tragopogon pratensis Plant Names Index (http://www.ipni.org).
Apodemus sylvaticus
- 70 -
MANUSCRIPT N°4
In revision, Mammalian Biology
Possible effects of roadside verges on vole outbreaks

in an intensive agrarian landscape
Louis de REDON
Christian KERBIRIOU
Nathalie MACHON
&
Frédéric JIGUET
- 71 -
MANUSCRIPT N°4
PRESENTATION
TITRE
Effets potentiels des bords de route sur les explosions démographiques de campagnols dans les
paysages d’agriculture intensive
RESUME
Notre étude a porté sur les populations de Microtus arvalis (campagnol des champs) en bord de
route pendant deux saisons sur 45 sites. Entre les deux années de relevés, le nombre d’individus
capturés a été multiplié par 2,6. Nous avons montré que l’explosion du nombre de M. arvalis n’a
pas été homogène dans le paysage mais était fortement dépendante de sa structure, en particulier du
maillage routier. Les populations les plus stables ont été observées dans les sites avec un important
réseau de dépendances vertes des bords de route alors que les explosions ont été notées dans les
sites avec les dépendances vertes les moins étendues.
MOTS CLEFS
Microtus arvalis; Route, Service écosystémique.
- 72 -
MANUSCRIPT N°4
POSSIBLE EFFECTS OF ROADSIDE VERGES ON VOLE OUTBREAKS
IN AN INTENSIVE AGRARIAN LANDSCAPE
Louis de REDON1,2; MACHON2; Christian KERBIRIOU2 and Frédéric JIGUET2.
KEYWORDS ABSTRACT
Microtus arvalis We sampled populations of Microtus arvalis (common vole) in 45
Population dynamics roadside verges during two consecutive years. The number captured
Road increased by a factor of 2.6 between the two years. We showed that
outbreak dynamics were not homogeneous over the landscape and were
highly correlated to landscape structure, especially linked to the spatial
extent of the roadsides. Populations were stable in sites where roadside
cover large areas while population outbreaks occurred in sites with low
roadside areas.
I. INTRODUCTION II. MATERIALS AND METHODS
Road verges are considered as refuge zone or Our study area was located in central France,
habitat for a number of small mammals in Seine-et-Marne department, about 50 km east
intensive agrarian landscapes (Bellamy et al., of Paris (48° 32′N, 2° 39′E). It was mainly
2000; Maisonneuve and Rioux, 2001; Pita et composed of agricultural fields used for
al. 2006) and also participate to the intensive crop production (60.7% of
conservation of some endangered species landcover). Road networks represented
(Santos et al.; 2007). This refuge zone effect important linear structures in the landscape
has been verified for Microtus arvalis (La (13,992 km long, i.e. 2.4 km.km-2) along with
Peña (de) et al., 2003), a rodent known to their verges (63.0 km², i.e. 1.07 % of the total
display regular population fluctuations area of the department). We sampled 45
(Briner et al., 2007), local outbreaks (Murray, roadside verges, each one with five traps
1965, Delattre et al., 1992). Such variations in placed linearly every 20 meters along the road
population size lead periodically to increased in the middle of the bank (for a total of 225
vole densities causing significant damages to traps each year). We used Barber pitfall traps:
crops, especially for vole densities over 200 450 mL plastic pot (10.2 cm height & 7.5 cm
individuals per hectare (Delattre et al., 1999). diameter) fully inserted into the ground and
We trapped voles in two successive containing a 155 mL solution composed of 75
springs (2006 and 2007) along roadsides to mL of water, 75 mL of conservative (ethylene
test and estimate the impact of roadside glycol), 5 mL of surfactant (dishware soap)
network densities on their inter-annual and 15 g of NaCl.
abundance variations. We found that if local Each trap was protected from rain by a
voles densities were affected by fields plastic transparent cover (15 cm × 15 cm)
network importance; their dynamics cycles supported by wood sticks 10 cm above the
between years were partly impacted by ground. The traps were set up for 28 days
roadsides structures. Such marginal areas twice during May 2006 and during May 2007.
possibly provide habitats for vole predators, Not any pitfall was found empty or dried due
and therefore help with regulating vole to evaporation, and then all pitfalls were
populations and decreasing the intensity of considered as efficient. This solution was
occasional outbreaks (Delattre et al., 1999). considered as un-attractive solution for
1: Corresponding author; Tel: (+33) 662 045 936; Fax: (+33) 140 79 38 35; E-mail:
redon@mnhn.fr; 2: UMR 5173 Conservation des Espèces, Restauration et Suivi des Populations,
Muséum National d’Histoire Naturelle, 55 rue Buffon, F-75005 Paris, FRANCE.
- 73 -
MANUSCRIPT N°4
some animals as invertebrates (Darren et al. FIGURE:
2001). IMPACT OF ROAD NETWORK
We calculated (1) the area covered by ON VARIATIONS OF M. ARVALIS
crop fields and (2) the roadside network area POPULATIONS
in an area of 500 m around each trapping site
using ArcGis 9.1TM software on landcover and
road network layers (buffers). We used a
geographical layer classifying roads into three
categories according to their status; RN:
national roads (0.9 ha of verges for 1 km of
road), RD: departmental roads (0.7 ha.km-1),
and RL:local roads (0.3 ha.km-1). We reported
the numbers of captured M. arvalis per group
of 5 traps, then we studied their inter-annual
variations with an ANOVA, according to
local roadside network area (noted as ARN,
ARN = 0.9 * RN length + 0.7 * RD length + 0.3
* RL length). We further classified variations
in four categories:
- ABS, Absence: M. arvalis was not detected

during either year of survey;
ABS: absence (sites excluded from analysis),
- SMI, Stable or Moderate Increase: the
SMI: no/moderate increase of captures
number of captures did not vary or increased
(<75%), COL: colonization in 2007, OBK:
by less than 75% between years;
- COL, Local Colonization: M. arvalis was All statistical analyses were performed with R
not detected in 2006 but was captured in
(Ihaka and Gentleman, 1996).
2007;
- OBK, Outbreak: captures increased by more III. RESULTS

than 100% between the years.
The trapping method worked very well
We chose the level of 100% for outbreak without any full trap (0 to five individuals by
classification according to Giraudoux et al. trap) and 167 M. arvalis trapped (47 in 2006
(1997) and Duhamel et al. (2000) studies. It and 120 in 2007). Excluding sites without any
should be added that we observed no local capture (6), the effect of roadside network
decrease and no increase between 75% and areas on vole dynamics was significant (F2,36
100%. We also analysed the log ratio of local = 3.346 & P = 0.0465): non-outbreaking
abundance of both years, as persisting populations (SMI) were observed in
log(N2007+2)/log(N2006+2), with a Generalized sites with the most important roadside
Linear Model, to test for linear relationships network densities (3.61 ha ± 0.75 SE).
beyond potential differences between Populations outbreaks (OBK) occurred in sites
categories. Because M. arvalis is a rodent with less developped roadside networks(2.16
species adapted to field habitats, we also km ± 0.33 SE), and local colonisation too
investigated the impact of crop local coverage (COL: 1.93 km ± 0.36 SE).
on its distribution (total captured and We found no effect of local crop cover
variation classes). We also ran a GLM to test on M. arvalis variation classes (F2,36 = 0.8781,
the possible correlation between the two P = 0.4243), though a positive effect was
variables used in this study: roadside network detected if considering the number of
areas and local field coverage. individuals trapped during the two years
- 74 -
MANUSCRIPT N°4
(F1,43 = 4.919, P = 0.0319). We found weak individuals trapped in all sites during the two
slight non-significant linear relationship years and the local field cover are positively
between yearly variation in local vole correlated. To conclude, we strongly
abundance and linear roadside length (the encourage maintenance and development of
latter log-transformed; t = -1.91, DF = 37 & P large verges along existing roads to prevent
= 0.063) or roadside network area (the latter vole outbreaks and avoid agrarian damages by
square-root transformed; t = -1.99, DF = 37 & stabilizing their population dynamics.
P = 0.054). We finally noticed that field cover
and roadside areas were not correlated (F1,45 =
2.658, P = 0.11 & R² = 0.036). ACKNOWLEDGEMENTS
Institut d'Aménagement et d'Urbanisme de la

IV. DISCUSSION Région d’Ile-de-France (Institute for Urban
Planning and Developpement of the Paris Ile-
We observed important differences in vole de-France Region, I.A.U.R.I.F.) and the
populations sizes between the two years of Direction Départementale de l’Equipement of
sampling in the studied agrarian landscape. Seine-et-Marne (Equipement Department
Traps did not appear to be attractive to voles. Agency, D.D.E. 77) are thanked for providing
However, as for all capture methods, a bias is GIS data and technical support to protect road
possible. Neverless we assume that if it exists, verges. Francesca Church is thanked for her
it is probably the same for the two sampling English corrections.
years or the different study sites. Thus it The research was supported
should not have affected our results. financially by the Direction Générale des
Because road verges are mowed Routes (Roads General Direction, D.G.R.) of
intensively, i.e. three times a year, the results the Ministère de l’Ecologie, de l’Energie, du
are in accordance with previous reports on the Développement et de l’Aménagement
effects of such mowing treatments on vole Durables (French Ministry for Ecology,
population variations (Meunier et al., 1997). Energy, Sustainable Developpement and
Moreover, we highlighted a positive effect of Spatial Planning, M.E.E.D.D.A.T.).
roadside areas on vole outbreak regulations:
variations in population size were not
spatially homogeneous, and large roadsides REFERENCES
seemed to stabilize vole populations, while
local colonization or outbreak occurred where Bellamy, P.E., Shore, R.F., Ardeshir, D.,
roadside verges were smaller. The weak slight Treweek, J.R., Sparks, T.H., 2000.
non-significant linear relationship between Road verges as habitats for small
yearly variation in local vole abundance and mammals in Britain. Mammal Review
linear roadside length may be due to the data 30, 131-139.
set size which could be considered as not big Briner, T., Favre, N., Nentwig ,W.,
enough important. Such results are in Airoldi, J.-P., 2007. Population
accordance with previous studies reporting dynamics of Microtus arvalis in a weed
that M. arvalis outbreaks in agrarian strip. Mammalian Biology 72, 106-115.
landscapes can be controlled by predation Darren, F., Ward, T., New, R., Yen, A.L.,
thanks to marginal areas providing habitats 2001.Effects of Pitfall Trap Spacing on
for predators such as cats (Hansson, 1988), the Abundance, Richness and
polecats (Rondinini et al., 2006) or raptors Composition of Invertebrate Catches
(Meunier et al., 2000). Journal of Insect Conservation 5: 47-53.
We also found strong impacts of local Delattre, P., Giraudoux, P., Baudry, J., Musard,
landcover (field cover) on M. arvalis P., Toussaint, M., Truchetet, D., Stahl,
distribution. In fact, field habitats seemed to P., Poule, M.L., Artois, M., Damange,
impact local abundances more than inter- J.P., Quéré, J.P., 1992. Land use
annual variations because the number of patterns of common vole (Microtus
- 75 -
MANUSCRIPT N°4
arvalis) population kinetics. Maisonneuve, C., Rioux, S., 2001.
Agriculture, Ecosystems & Importance of riparian habitats for
Environment 39, 153-168. communities in agricultural
Delattre, P., De Sousa, B., Fichet-Calvet, E., landscapes of southern Québec.
Quéré, J.P., Giraudoux, P., 1999. Agriculture, Ecosystem and
Vole outbreaks in a landscape Environment 83, 165-175.
context, Evidence from a six year Meunier, F.D., Corbin, J., Verheyden, C.,
study of Microtus arvalis. Landscape Jouventin, P., 1999. Effects of
Ecology 14, 401-412. landscape type and extensive
Duhamel, R., Quéré, J.P., Delattre, P., management on use of motorway
Giraudoux, P., 2000. Landscape roadsides by small mammals.
effects on the population dynamics of Canadian Journal of Zoology 77,
the fossorial form of the water vole 108-117.
(Arvicola terrestris Sherman). Meunier, F.D., Verheyden, C., Jouventin P.,
Landscape Ecology 15: 89–98. 2000. Use of roadsides by diurnal
Giraudoux, P., Delattre, P., Habert, M., raptors in agricultural landscapes
Quéré, J.P., Deblay, S., Defaut, Biological Conservation 92: 291-
R.,Duhamel, R., Moissenet, M.F., 298.
Salvi, D., Trebuchetet, D.,1997, Murray, K.F., 1965. Population changes
Population dynamics of fossorial during the 1957-1958 vole
water vole (Arvicola terrestris (Microtus) outbreak in California.
scherman: a land use and landscape Ecology 46: 163-171.
perspective. Agriculture, Ecosystems Pita, R., Mira, A., Beja, P., 2006. Conserving
& Environment 66: 47-60. the Cabrera vole, Microtus cabrerae,
Hansson, L., 1988.The domestic cat as a in intensively used Mediterranean
possible modifier of vole dynamics. landscapes. Agriculture, Ecosystems
Mammalia 52, 159-164. & Environment 115: 1-5.
Ihaka, R., Gentleman, R., 1996. R, a Rondinini, C., Ercoli, V., Boitani, L., 2006.
language for data analysis and Habitat use and preference by
graphics. Journal of Computational polecats (Mustela putorius L.) in a
and Graphical Statistics 5, 299-314. Mediterranean agricultural
La Peña (de), N.M., Butet, A., landscape. Journal of Zoology 269:
Delettre, Y., Paillat, G., Morant, P., 213-219.
Le Du, L., Burel, F., 2003. Response Santos, M., Da Luz Mathias, M., Mira, A.,
of the small mammal community to Simões, M., 2007. Vegetation
hanges in western French structure and composition of road
Agricultural landscapes. Landscape verge and meadow sites colonized by
Ecology 18, 265-278. Cabrera vole (Microtus cabrerae
Thomas). Polish Journal of Ecology:
481-493.
Microtus arvalis
- 76 -
DISCUSSION ET CONCLUSION GENERALES
ANALOGIE DE LA GUEPE ET DU TRAMWAY

ELECTRIQUE
Quelque chose de muet au repos et de

chanteur en action. Quelque chose aussi
d'un train court, avec premières et
secondes, ou plutôt motrice et baladeuse.
Et trolley grésilleur. Grésillante comme
une friture, une chimie (effervescente).
Et si ça touche, ça pique. Autre chose

qu’un choc mécanique : un contact
électrique, une vibration venimeuse.
Mais son corps est plus mou – c’est-à-dire

en somme plus finement articulé – son vol
plus capricieux, imprévu, dangereux que la
marche rectiligne des tramways
déterminée par les rails.
Francis PONGE
LA RAGE DE L'EXPRESSION (1952)
- 77 -
§ A. EFFETS DES PROTOCOLES UTILISES ET DES GROUPES SUIVIS
Cette étude de la biodiversité commune en milieu agricole intensif a été riche en

résultats : sur la biodiversité des bords de route eux-mêmes (effet « habitat », effets des
polluants) et sur la compréhension de l’écologie des milieux adjacents (homogénéisation
biotique, effets « barrière », effets « corridor », maintien de services écosystémiques
locaux). Ces résultats sont très intéressants car ils montrent l’intérêt de l’étude la
biodiversité dans sa globalité, i.e. en tenant compte de la Nature ordinaire, riche en
informations et permettant la mise en évidence de nombreux processus écologiques de
nos paysages agricoles modernes. Cependant si, au début de la thèse, certains groupes
ont été ciblés pour servir d’indicateurs, plantes et carabes, les différents protocoles
n’ont pas tous eu le succès escompté.
1) Relevés floristiques
Le protocole de relevés floristiques mis en place en bords de route s’est révélé être assez bon
permettant l’obtention de résultats probants (cf. Manuscrit N°1 et Manuscrit N°3) grâce à :
• Sa rapidité d’exécution (8 minutes en moyenne par relevé de 1m², temps
d’installation compris, i.e. 2 heures sur site par station de 15 relevés), ce qui était
capital pour un terrain important (48 stations à étudier en 2 semaines pour éviter un
biais temporel lié au décalage de floraison des plantes) et un terrain dangereux (en
bords de route, et parfois de routes très passantes !) ;
• Sa répliquabilité et sa simplicité, les quadras de 1 m² sont très faciles à placer et à
replacer le long d’une structure linéaire et, par leur petite taille, permettent des relevés
exhaustifs de la végétation quel que soit l’observateur ;
• Des taux de détection élevés (67.27% ± 1.10% ES en 2006) et une richesse
estimée corrélée à la richesse observée (F = 147.93, r² = 0.536 & P < 10-3, cf. Figure
N°8) pour une erreur d’estimation associée acceptable (22.53% ± 0.48% ES < 25 %) ;
• Une flore des bords de route très commune centrée autour de quelques dizaines
d’espèces seulement représentant une grande majorité du recouvrement total
(seulement 80 des 212 espèces détectées représentent 95 % du recouvrement total et à
peine 9 espèces en représentent plus de 50%, cf. Figure N°9) qui permet un
apprentissage rapide par les observateurs.
- 78 -
Figure N°8
Richesse floristique observée Vs. Richesse floristique estimée (Ne)
Figure N°9
Répartition du recouvrement végétal en bords de route par espèce
- 79 -
Le protocole s’est révélé adapté aux analyses qui ont été réalisées et a permis la mise en
évidence des effets de la fauche au sein des communautés végétales des bords de route, de
l’existence d’une homogénéisation biotique des milieux adjacents liés à la flore des bords de
route, ou encore des effets « corridor » (des analyses génétiques étant malgré tout nécessaires
pour confirmation). Parallèlement, la construction de la base de données GALIUM s’est aussi
révélée utile et adaptée aux analyses sur les traits fonctionnels réalisées.
Enfin les deux problèmes rencontrés ont été liés à des problèmes de détermination
inhérents au milieu étudié : les bords de route. En effet, (a) certaines plantes avaient des
morphologiques altérées (ex : gigantisme) pouvant être notamment dues à la pollution des
sols (Angold 1977, Spencer et al. 1988), et (b) certains individus étaient devenus
inidentifiables à cause de la fauche.
(2) Relevés faunistiques

Si le protocole pots-pièges de type « barber » pour invertébrés s’est révélé très intéressant
pour l’étude des araignées (cf. Manuscrit N°7), l’étude des carabes en bords de route est
apparue quant à elle décevante à plusieurs niveaux :
• La variabilité des captures entre 2006 et 2007 n’a pas permis de dégager des
résultats d’une année sur l’autre même si d’autres études ont été menées avec succès
sur les bordures de champ comme habitat (Woodcock et al. 2005, Griffiths et al. 2007)
ou corridor (Martin et al. 2001), des bords de route comme habitat (Eversham et
Telfer 1994, Noordijk 2008) et des pollutions aux métaux lourds auxquelles les
carabes sont sensibles (Emets et Zhulidov 1983) ;
• Le nombre d’espèces (ou taxons) capturées s’est avéré décevant, 27 au total,
apparaissant comme assez bas et en-dessous à ce que l’on pourrait s’attendre en bords
de route (Pr. Michel BAGUETTE, communication personnelle). Ces résultats pourraient
s’expliquer par le fait que certains auteurs considèrent parfois les bords de route
comme un puits pour les carabes (Koivula 2005) ;
• La capture et la détermination des carabes se sont révélées longues et laborieuses :
plusieurs passages sur sites nécessaires, gestion difficile de près de 250 pots avec des
transferts d’individus dans l’alcool, et longue détermination à la loupe binoculaire de
plus de 4.000 individus (1.823 en 2006 et 2.291 en 2007).
- 80 -
Le piégeage s’est cependant révélé très efficace pour les petits mammifères (cf. Manuscrit
N°2 et Manuscrit N°4). Les analyses ont montré que les petits mammifères (musaraignes et
campagnols dans notre étude) réagissent très fortement aux conditions écologiques du milieu
(connectivité, fragmentation, occupation des sols, qualité des habitats) et pourraient donc
constituer de bons indicateurs des milieux agricoles. Cependant et malgré cette réussite en
terme de piégeage et d’analyses, certaines limites conviennent d’être apportées à l’utilisation
d’un tel protocole pour des suivis de petits mammifères :
• Il existe une vraie question éthique compte tenu du fait que le piégeage est
destructeur. Il conviendrait, à la vue des résultats probants obtenus, de développer un
protocole non létal d’étude des petits mammifères ;
• L’attractivité des pièges n’a pas été testée rigoureusement et même si ceux-ci se
veulent non attractifs (éthylène glycol employé à la place de l’alcool) ; il pourrait
exister des effets non désirés comme l’accumulation d’insectes piégés dans les pièges
(en place pendant un mois) sur la capture des musaraignes qui sont des insectivores.
• Il a tout de même été testé l’effet « nombre de carabes piégés » sur la variable
« nombre de musaraignes piégées » au niveau des pots et les résultats se sont révélés
non significatifs (F1,240 = 0.001 et P = 0.977) ; cependant les carabes, insectes
déterminés et comptés, ne représentent qu’une partie des invertébrés piégés
(nombreux autres insectes, araignées et gastéropodes présents dans les pièges) ;
• Le nombre d’espèces (sept en bord de route et d’autoroute) et le nombre
d’individus piégés (souvent moins d’une dizaine au total par espèce sauf pour Sorex
coronatus, Crossidura russula et Microtus arvalis) sont faibles, et même si ces
chiffres enregistrés en bords de route sont plus importants que ceux obtenus dans les
autres milieux, ils n’étaient pas suffisants pour le fonctionnement de ComDyn (pour
lequel un minimum de sept espèces par site est nécessaire avec au moins un individu
par répliqua) ; ce qui n’a pas permis (a) de calculer des taux de détectabilité pour les
différentes espèces piégées et (b) de travailler sur des richesses corrigées (ce qui était
l’objectif recherché avec les cinq répliquas par site).
Si les résultats obtenus sont prometteurs quant à l’utilisation des petits mammifères comme
indicateur, la recherche d’un protocole plus adapté semble nécessaire : (a) plus respectueux
des individus piégés, (b) avec des pièges dont l’attractivité potentielle déterminée, (c) avec
des répliquas repensés par rapport à d’autres outils que ComDyn et (d) permettant une
meilleure capture des mulots.
- 81 -
§ B. IMPACTS NEGATIFS DES ROUTES SUR L’ENVIRONNEMENT
Si les effets des bords de route étudiés lors de ces travaux de recherches ont
essentiellement porté sur des aspects positifs, effets « habitat » et « corridor » pour les
plantes et les petits mammifères, il ne faut cependant pas oublier la part de
responsabilité importante des infrastructures de transport dans la crise de la
biodiversité.
(1) Destruction et fragmentation des habitats

Les effets négatifs des routes en termes de destruction et de fragmentation des habitats ont
peu été montrés lors de cette étude (sauf les effets « barrière » de la route pour les
musaraignes et « appauvrissement » pour les bois). Cela peut être principalement dû à deux
causes :
• L’étude s’est concentrée principalement en milieu agricole à cause des enjeux
importants qui y sont associés en termes de conservation de la biodiversité (cf.
Introduction). Or, depuis de nombreuses années, les haies, ou autres structures semi-
naturelles, ont largement disparu de ces espaces sous l’effet de l’intensification des
pratiques agricoles. Dans ce contexte, les constructions de route consomment peu
d’espaces naturels et participent de manière très relative à la fragmentation de milieux
qui le sont déjà fortement. Au contraire, les routes permettent la mise en place de
zones refuges et de continuités. Les résultats de cette étude ne sont donc
appréhendables que dans ce contexte et il est clair que les résultats seraient tout autres
si l’étude avait été menée sur les effets des routes au sein d’espaces naturels (Forman
et Alexander 1998, Coffin 2006) ;
• Les résultats présentés ne concernent que deux groupes : plantes et petits
mammifères. Des effets négatifs des routes et de leurs accotements sur les
communautés d’oiseaux ont été montrés (Van der Zande et al. 1980, Johnson 1996,
Reijnen et Foppen 1994, Canaday 1996). Par ailleurs, les quelques résultats obtenus
sur les carabes ont montré en 2006 : (a) l’existence d’une corrélation négative entre
indices de diversité (Hill) et teneurs en plomb des sols des bords de route (t = 3.304 et
P = 0.00193) avec quelques espèces malgré tout favorisées (peut-être par simple
relâchement de compétition) comme Pterostichus melanarius et Anchomenus dorsalis,
cf. rapport de stage d’Elise CONTAN, stagiaire de M1 (www2.mnhn.fr/cersp
- 82 -
/IMG/zip/Contan_2007.pdf.zip) ; et (b) l’existence d’un impact de la fragmentation du

milieu sur la taille moyenne des carabes piégés (P = 0.002) avec les espèces de taille
importante (Carabus auratus, Carabus monilis, etc.) disparaissant des milieux
présentant des indices de fragmentation élevés, cf. rapport de stage de Johanna
ROILLET, stagiaire de M1 (www2.mnhn.fr/cersp/IMG/zip/Roillet _2007.pdf.zip).
Ces résultats sur les carabes sont cependant à prendre avec prudence dans la mesure où, ces
mêmes analyses réalisées sur les données de 2007 n’ont pas confirmé ces tendances. Ces
différences pourraient s’expliquer par (a) la difficulté à étudier les insectes sur des pas de
temps courts, comme deux années, alors que les effectifs des populations de carabes sont
fluctuants d’une année sur l’autre et liés aux aléas climatiques (Baars et Van Dijk 1984,
Wallin 1985, Honek 1997) ; 2006 ayant été très différente de 2007 notamment en début de
printemps (mars « pluvieux », « froid » et « mal ensoleillé » en 2006 et « normalement
arrosé », « doux » et « ensoleillé » en 2007 en Seine-et-Marne, sources Météo France :
http://climatheque.meteo.fr); (b) les effets du protocole d’échantillonnage qui prélève par
destruction de nombreux individus (cf. Table N°5) au sein d’un groupe, les Carabidae, aux
capacités de dispersion limitées à court terme en milieu fragmenté, surtout au niveau des
espèces ne disposant pas d’ailes importantes (Den Boer 1970, Frampton et al. 1995).
Il est intéressant de noter, par ailleurs, que si les bords de route permettent une
augmentation locale de la biodiversité (diversité α), nous avons montré qu’elles induisaient
aussi une homogénéisation biotique du milieu (baisse de la diversité β), cf. Manuscrit N°1.
Table N°5 :
2006 2007
Carabidae
Nombre Erreur standard Nombre Erreur standard
Taxons 28.00 - 17.00 -
Taxons par pot 1.97 0.10 1.82 0.09
Individus 1823.00 - 2291.00 -
Individus par pot 7.50 0.95 9.16 1.58
Bilan des captures de carabes en bords de route (53 sites)
- 83 -
(2) Dispersion d’espèces invasives

La connectivité mise en place par les dépendances vertes des bords de route en milieu
agricole permet certes la dispersion d’espèces invasives et éventuellement leur maintien au
sein des paysages (métapopulations et métacommunautés) mais elle permet aussi la
dispersion et la diffusion d’espèces exotiques et/ou invasives (cf. Introduction). Cette étude de
la biodiversité des bords de route n’a cependant pas révélé la présence de telles espèces dans
les relevés effectués, en particulier chez les plantes, malgré la présence avérée en Seine-et-
Marne de certaines d’entre elles (Sources : http://cbnbp.mnhn.fr) comme l’ailanthe (Ailanthus
altissima), la renouée du Japon (Reynoutria japonica), la solidage du Canada (Solidago
canadensis) et le séneçon du Cap (Senecio inaequidens). Cela pourrait être dû essentiellement
à deux causes :
• Un protocole non-adapté à la détection de ces espèces dans la mesure où leur
répartition se fait rarement sur un front continu mais plutôt sous forme de « tâches »
éparses en bords de route (Pysek et Hulme 2005) et, donc, nécessitant un ciblage pour
les échantillonner alors que nos relevés étaient tirés au hasard ;
• Une détermination difficile liée au fait que ces plantes invasives se trouvent
souvent en toute bordure de route, dans la zone dite « de sécurité », là où la fauche est
très intensive, précoce et systématique réduisant généralement les plantes à un certain
état de mutilation et à une taille de quelques centimètres.
L’existence et les impacts de ces plantes en bords de route ne doivent cependant pas être
oubliés à la lecture de ce manuscrit.
§ C. IMPORTANCE DES BORDS ROUTES EN MILIEU AGRICOLE INTENSIF
Si les routes ont des effets négatifs importants sur la biodiversité, cette étude a montré
que leurs dépendances vertes permettaient la création (a) de zones refuges, habitat pour
une biodiversité exclue de la matrice agricole, et (b) de continuités biologiques, tout en
favorisant le maintien de services écosystémiques. Il ne s’agit pas de dire que ces effets
peuvent en compenser d’autres. Il s’agit de montrer que, dans la mesure où les routes
existent, un certain nombre de mesures peuvent être mises en place pour la biodiversité.
- 84 -
(1) Mise en place de zones habitats et refuges

Si les bords de route créent des zones « refuge » ou « habitat », ce ne sont pas les seuls
éléments des paysages agricoles à offrir cela à la biodiversité. L’impact des bordures de
champ est aussi loin d’être négligeable (bandes enherbées avec ou sans haies) : elles peuvent
aussi représenter des surfaces importantes et constituer des zones « refuge » ou « habitat »
pour la biodiversité, participant aussi à son maintien au sein des paysages agricoles intensifs
(Marshall et Moonen 2002, Benton et al. 2003, Abadie 2008).
Des différences existent cependant entre bords de route et bordures de champ qui ne
sont pas en bords de route. Cela a été montré pour les micro-mammifères (cf. Manuscrit N°2)
mais cela est aussi vrai pour la flore. En effet, une étude (réalisée lors de mon stage de Master
2 sur la diversité végétale des bordures de champ) a montré que le premier facteur explicatif
de la composition des communautés végétales était le facteur « bordure de champ en bords de
route » (par rapport aux bordures situées entre deux champs et devant toute une série de
modes de gestion opérées par les agriculteurs : fauche, broyage, herbicides, etc.) : l’ACP
réalisée sur les stations et leur composition floristique a révélé une ségrégation entre bordures
de champ à l’interface champ/chemin, et bordures de champ à l’interface champ/route ; les
bordures à l’interface champ/route étant par ailleurs les plus riches en terme de diversité
spécifique (cf. Manuscript N°5). Ces différences entre bords de route (interface route/champ)
et bordures de chemin (interface champ/chemin) peuvent s’expliquer par un certain nombre
de paramètres : impacts des routes (pollution, trafic, conditions microclimatiques, etc.), de la
structure de l’emprise (largeur, présence de haies) et des modes de gestion (fauche, salage,
etc.) et montrent l’intérêt de travailler sur les deux types de structures pour une meilleure
appréhension des agro-écosystèmes.
(2) Mise en place de continuités biologiques

La mise en place de continuités biologiques par les bords de route a clairement été mise en
évidence au cours de cette étude. Nos résultats ont confirmé que la notion de « corridor », ou
de continuité biologique, était bien dépendante des espèces étudiées (Beier 1998) :
• L’homogénéisation biotique observée dans les bois connectés au niveau des
communautés végétales (cf. Manuscrit N°1) a montré que seules certaines plantes
bénéficiaient de la présence des bords de route pour leur dispersion (essentiellement
des plantes semi-forestières non totalement inféodées au milieu forestier, ex. :
Stellaria holostea, Cruciata laevipes ou Ranunculus auricomus). Afin d’affiner ces
- 85 -
résultats, il pourrait être intéressant de poursuivre le travail de recherche sur le plan

génétique au niveau de certaines populations de plantes au sein de bois connectés et
non-connectés aux bords de route à l’aide de marqueurs génétiques afin de pouvoir
affirmer avec certitude l’existence d’effets corridors ;
• Les patterns de piégeage différents entre micro-mammifères ont permis de
montrer que les bords de route pouvaient permettre la dispersion des musaraignes
mais pas celle des campagnols (cf. Manuscrit N°2). Un regret au sujet de cette étude
est de ne pas avoir déterminé le sexe des individus piégés en bords de route. En effet,
chez les musaraignes, animaux territoriaux, deux raisons peuvent provoquer le départ
temporaire du territoire : la recherche de nourriture pour les mâles comme les femelles
(Oxley et al. 1974), ou la recherche de partenaire sexuel pour le mâle uniquement
(Cantoni 2002). Une étude du sexe ratio aurait donc pu nous renseigner sur les raisons
pouvant expliquer la dispersion des musaraignes en bords de route. Enfin, une étude
complémentaire, par piégeage au mois de septembre, permettrait de savoir si les
jeunes musaraignes dispersent le long des bords de route à la recherche de nouveaux
territoires et ne traversent pas la structure routière à cette même fin, comme cela est le
cas pour les adultes.
La structuration des corridors est importante (Beier et Noss 1998) et cela a aussi pu être
observé lors de l’étude : la largeur de l’emprise semble être un élément capital du
fonctionnement de la structure pour la dispersion des musaraignes (cf. Manuscrit N°2).
§ D. IMPACTS DES STRUCTURES ROUTIERES
Si la route a de nombreux effets négatifs sur l’environnement, il convient donc, dans la

mesure où celles-ci existent, de favoriser les impacts potentiellement positifs de ses
dépendances vertes. Il est tout d’abord possible de travailler sur la structure même de
l’emprise. En effet, les différences observées pour la biodiversité au niveau des
différents types d’emprises, bords d’autoroutes, bords de route et bordures de champ,
montrent combien la structure de l’emprise est importante.
- 86 -
(1) Effets de la largeur de l’emprise

La largeur de l’emprise semble avoir un effet important sur l’environnement et la
biodiversité : bien qu’une certaine hétérogénéité puisse exister entre les modes de gestion
d’une emprise à une autre ; bords d’autoroutes (largeur, l = 15.7 ± 6.6 m), bords de route (l =
7.2 ± 3.4 m) et bords de champ (l = 4.0 ± 0.4 m) diffèrent fortement par la taille de leurs
emprises respectives.
Figure N°10
Résultats des analyses de sols de bords de route en Seine-et-Marne (58 sites)
Notre étude a montré, notamment pour Sorex coronatus, que les bords d’autoroute pouvaient
constituer un habitat suffisant, alors que les bords de route servaient davantage pour leur
dispersion et que la musaraigne était quasi-absente des bords de champs (cf. Manuscrits N°2).
Pour les populations de campagnols des champs, Microtus arvalis, l’importance des emprises
- 87 -
semblent avoir un effet important sur la régulation des populations sujettes à des explosions
démographiques préjudiciables aux cultures (cf. Manuscrit N°4). Enfin, les différences
observées entre bords de route et bordures de champ, au niveau des communautés végétales
(cf. Manuscrit N°5), pourraient peut-être s’expliquer par les pollutions émises depuis la route
et par les modes de gestion différents des emprises (cf. Introduction) ; mais ces deux facteurs
sont corrélés à la largeur de la structure.
En effet, premièrement, les deux premiers mètres des bords de route sont fauchés
intensivement par mesure de sécurité (visibilité, arrêts d’urgence, etc.) ; il apparaît donc
évident qu’un bord de route de moins de deux mètres de large ne sera pas de largeur
suffisante pour une politique de gestion prenant en compte la biodiversité. Deuxièmement, la
pollution des sols diminuent fortement avec l’éloignement à la route (cf. Introduction).
Nos résultats ont confirmé ces observations (cf. Figure N°10) : nos analyses de sols à
1 m et 5 m de la route ont révélé un fort effet de la distance à la route sur le pH (test t d'égalité
des espérances, observations pairées, t = 6.065 et P < 10-3), sur les teneurs en azote (t = 2.423
et P = 0.020), Na2O (t = 10.044 et P < 10-3) et en plomb (t = 8.470et& P < 10-3) dans les sols.
La largeur de l’emprise est donc déterminante pour la biodiversité en bords de route. Il est
important de le souligner car les récentes politiques prises en matière de développement de
bandes enherbées dans le cadre des Mesures Agri-Environnementales (M.A.E.) de la
Politique Agricole Commune (P.A.C.) ont imposé une largeur minimale de 5 mètres pour
l’attribution de subventions depuis 2002. Cette largeur semble, à la vue des résultats obtenus
au cours de cette thèse, une bonne base de travail car elle peut permettre une conservation
effective de la biodiversité et le maintien de services écosystémiques par :
• Le maintien des populations de musaraignes qui sont des prédateurs d’insectes et
de gastéropodes pouvant être des ravageurs des cultures (espèces listées d’intérêt
communautaire dans l’annexe III de la Convention de Berne, cf. Manuscrit N°2) ;
• La stabilisation des populations de campagnols par le maintien de zones
favorables à leurs prédateurs qui peuvent accèder aux campagnols dans les champs en
utilisant les bords de route et les bordures de champ (cf. Manuscrit N°4) ;
• Le maintien d’un habitat de qualité pour les insectes pollinisateurs en bordure de
champ : une gestion extensive des bords de route, possible au sein des emprises
importantes, entraîne une augmentation des plantes zoochores (cf. Manuscrit N°3)
pouvant permettre le maintien des communautés d’insectes grâce aux apports de
ressources en pollen et en nectar (Kohler et al. 2006) ;
- 88 -
• La création d’une zone tampon entre la route et les cultures (à vocation

essentiellement alimentaires) : notre étude a montré, par exemple, que les teneurs de
plomb détectées en bords la route dépassent 100 µg.g-1, valeur limite selon l’Arrêté du
8 janvier 1998 et entraînant une obligation de dépollution dans certains pays comme la
Suisse (www.admin.ch/ch/f/rs/814_12/app1.html), dans 72,1% des prélèvements à 1
mètre de la route et dans seulement 7.0 % des prélèvements à 5 mètres.
La viabilité de telles mesures est assurée à travers un système de subventions complexes et
que les attentes en terme d’évaluation de leurs impacts est un attendu fort et actuel compte
tenu des coûts générés : 450 € par hectare de bande enherbée. Dans le même temps, les pertes
en production liées à la réduction de l’espace cultivable sont d’environ 161 €.ha-1.an-1 pour
l’agriculteur (Sources : Chambre d’Agriculture de Rhône-Alpes www.rhone-
alpes.chambagri.fr/phytov3/pages/bande_herbe.htm), permettant ainsi en bout de chaîne un
bénéfice annuel total de 95 € par hectare de bande enherbée pour l’exploitant agricole selon
Desbois et Legris (2005), bénéfice financé par la P.A.C.
(2) Effets des plantations

L’étude a été centrée sur le réseau des routes nationales et départementales qui sont des
infrastructures anciennes et construites à une époque où les enjeux environnementaux
n’étaient pas encore connus, ou du moins, n’étaient pas une priorité. Aucun effort particulier
n’avait donc été porté à la mise en place de leurs dépendances vertes dont la végétation est
globalement issue de processus de colonisation depuis les milieux adjacents ou lds graines
apportées par les voitures. Il s’agit d’une végétation qui pourrait être qualifiée de
« spontanée ».
Cela n’est pas le cas des autoroutes dont les constructions ont commencé à la fin des
années soixante et se poursuivent actuellement. Compte-tenu de leur importance (en termes
d’emprises), de la rapidité de construction et des moyens financiers mis en place, les
dépendances vertes font l’objet de véritables aménagements. Ces aménagements ont été,
jusqu’à très récemment, dictés par des contraintes absolues d’économies financières
(construire le « moins chère possible » au lieu de construire le « mieux chère possible ») et
d’esthétisme (création de « paysages autoroutiers »). Les ensemencements à partir de
cocktails de graines et les plantations de haies constituent la structure de ces aménagements.
Il serait intéressant d’étudier leurs impacts sur la biodiversité afin de savoir s’ils peuvent aussi
participer à son maintien et à quelles conditions : faut-il des haies de manière continue le long
- 89 -
des routes ou laisser certains espaces ouverts ? Les cocktails de graines orientent-ils la
structure de certaines communautés ou l’influence du milieu adjacent est-elle le facteur
déterminant ?
Les réponses apportées à ces questions pourraient permettre de quantifier l’importance
de ces investissements pour ensuite, peut-être, essayer de les mettre en place au niveau de tout
le réseau routier s’il s’avère que les effets sont bénéfiques pour la biodiversité : plantation de
haies le long des routes et ensemencement des accotements après chaque dérasement (le
dérasement est une opération d’entretien décennale qui consiste à remettre les bords de route
à niveau pour permettre le bon écoulement des eaux de la chaussée).
S’il n’a donc pas été possible d’étudier ces effets, effet « haies » et effet « semis »,
durant ma thèse, il convient néanmoins de rappeler quelques résultats connus. Au niveau de la
végétation, si il a ainsi été prouvé que les semis avaient peu d’impacts, à terme, sur la
composition des communautés (De Cauwer et al. 2005), il a en revanche été montré que les
haies (a) permettent la conservation d’une flore plus diversifiée qu’en milieux ouverts sans
toutefois défavoriser certaines espèces (cf. Manuscrit N° 7), et (b) constituent de véritables
habitats pour les plantes forestières (McCollin et al. 2000).
Il pourrait être intéressant de réaliser les mêmes observations que celles effectuées
dans la première étude (cf. Manuscrit N°1) en présence de haies en bords de route afin de
savoir si l’effet « corridor » mis en évidence serait renforcé et limiterait, ou non, l’effet
homogénéisant des routes ; des effets « corridor » ayant par ailleurs déjà été montrés pour les
plantes forestières au niveau de haies (Davies et Pullin 2007)
Enfin, la présence de haies (a) assure la mise en place de communautés d’araignées
originales en bords de route, différentes de celles observées en leur absence, et (b) permet
ainsi la conservation de la diversité fonctionnelle au sein de l’écosystème (cf. Manuscrit N°7).
§ E. IMPACTS DES MODES DE GESTION
La structure de l’emprise n’est pas le seul paramètre ayant des effets sur la biodiversité,
les modes de gestion sont aussi importants. Si les traitements chimiques semblent avoir
disparus, du moins officiellement, la fauche intensive reste la règle et l’exportation des
déchets n’est pas encore d’actualité. Les agriculteurs interviennent aussi sur les bords
de route, même illégalement, puisque ceux-ci sont à l’interface de leurs champs.
- 90 -
(1) Impacts de la fauche

L’un des objectifs de cette thèse, sur le plan appliqué, était d’étudier les différentes modalités
de gestion afin de trouver les plus respectueuses de la biodiversité. La majorité du terrain a
été construit autour de cette question avec 43 stations échantillonnées à cet effet (calendriers
de gestion différents, âges des routes variables, intensités de trafic distincts).
Les effets de ces paramètres ont été très difficiles à détecter au niveau des
communautés végétales (cf. rapport de stage de Noémie VARET, stagiaire de M1 :
www2.mnhn.fr/ Crsp/IMG/zip/Varet_2007.pdf.zip) comme carabiques (résultats obtenus en
2006 non vérifiés en 2007), i.e. au niveau des deux groupes étudiés.
Une importante variabilité interannuelle a été observée pour les deux groupes : une
réduction importante du nombre d’espèces observées chez les végétaux (180 en 2006 contre
seulement 126 en 2007) avec cependant une relative stabilité des plantes communes (sur les
97 espèces représentant 95% du recouvrement en 2006, 79 ont été détectées à nouveau en
2007 et représentaient encore 92% du recouvrement) ; et un piégeage carabique qui a semblé
finalement assez aléatoire en 2007 malgré des résultats prometteurs en 2006 (effets négatifs
de la « fauche de fin de printemps », i.e. la deuxième fauche de l’année située entre mai et
juin selon les sites étudiés, sur les populations carabiques, nombre d’individus piégés, F1,43 =
4.146 et P= 0.0481, cf. rapport de M1 d’Elise CONTAN cité précédemment :
www2.mnhn.fr/cersp/IMG/zip/Contan_2007.pdf.zip).
Pour la végétation, il a tout de même été détecté un léger effet non-significatif du
calendrier de fauche au niveau des modes de dissémination du pollen avec un enrichissement
des communautés végétales en plantes entomogames associé à l’allongement des délais entre
aussi bien les deux premières fauches, i.e. entre celle de début de printemps en mars/avril et
celle de fin de printemps en mai/juin (F1,0 = 3.183 et P= 0.082), qu’entre les deux dernières,
i.e. entre celle de fin de printemps et celle d’été en août/septembre (F1,40 = 2.964 et P= 0.093).
Le trait « dissémination du pollen » est le seul à avoir montré une telle tendance face aux
modes de gestion des bords de route parmi tous ceux testés (« hauteur de floraison »,
« dispersion des graines », « mode de reproduction végétative », « durée de floraison », ou
encore « phénologie »).
Cette légère tendance à l’enrichissement des communautés végétales en plantes
entomogames liées à une extensification des pratiques de fauche est en accord avec les
résultats obtenus lors de l’expérimentation du « fauchage tardif » où là cependant, les
résultats deviennent significatifs (cf. Manuscrit N°3).
- 91 -
De manière générale, il semble donc que travailler sur le calendrier de fauche, sans changer
les pratiques en profondeur, ne soit pas suffisant pour permettre une modification des
communautés végétales vers des ensembles plus riches dotés d’une plus grande diversité
fonctionnelle.
Cela a été confirmé par les travaux réalisés sur les bordures de champ où l’on peut
voir (a) que le facteur déterminant pour les communautés végétales est la présence d’une
route à l’interface (cf. Manuscrit N°5) et, et (b) que les modes de gestion ont aussi un impact
relativement faible sur les communautés végétales quand ceux-ci sont finalement assez
semblables, i.e. variabilité de la gestion dans un système restant malgré tout intensif (cf.
Manuscrit N°6).
Ces résultats semblent en accord avec des études ayant montré le peu d’impact des
mesures de gestion sur les communautés végétales des bords de route/bords de champ (Kleijn
et Verbeek 2000) alors que finalement les dynamiques de plantes « adventices » sont
considérées comme chaotiques par certains auteurs (Gonzalez-Andujar 1996) ; ce qui est par
ailleurs critiqué (Freckleton et Watkison 2002).
Les résultats obtenus avec l’expérimentation « fauchage tardif » (cf. Manuscrit N°3)
d’une part, et l’absence de résultats avérés avec l’étude des effets du calendrier de fauche et
avec les études réalisées en bord de champs (cf. Manuscrits 5 et Manuscrit 6), d’autre part,
montrent clairement qu’une meilleure prise en compte de la diversité des bords de route passe
par un changement global des pratiques, comme le passage de trois à une fauche, et non par
une adaptation à la marge de celles-ci, comme l’avancement de deux à trois semaines de la
deuxième fauche de printemps, fauche qui semble être la plus critique.
Ces résultats sont malgré tout à prendre avec prudence et une poursuite de l’étude reste
nécessaire dans la mesure où :
• Le pas de temps d’étude est faible (deux à trois ans) : (a) pour les insectes, il est
trop court pour dégager de réelles tendances (résultats différents entre 2006 et 2007) ;
et (b) pour la végétation, il a été montré que les communautés répondaient fortement à
l’extensification des pratiques et étaient en changements en ne semblant pas
stabilisées au bout de trois ans. Il faudrait voir si la réorganisation des communautés
autour de nouveaux cortège floristiques n’entraîne pas, à terme, des communautés
différentes mais pas nécessairement plus riches (Parr et Way 1988, Sykora et al.
2002) ; un fauchage différencié pourrait être alors une solution au maintien de
différentes communautés en bords de route ;
- 92 -
• Seuls deux groupes ont été étudiés (plantes et carabes) alors que d’autres sont
aussi susceptibles de subir les changements des modes de gestion des bords de routes :
araignées, petits-mammifères, oiseaux, etc. (cf. Introduction).
(2) Impacts de l’exportation des déchets

Cette opportunité de travail peut s’organiser autour de l’étude d’autres facteurs qui n’ont pu,
faute de moyens, être étudiés au cours de ces travaux de recherche :
• L’exportation des déchets est désormais envisageable en Seine-et-Marne ce qui ne
l’était pas il y a trois ans alors qu’il s’agit d’un point critique pour l’étude de la
diversité végétale des bords de route (De Cauwer et al. 2005). En effet, les bords de
route constituent un lieu d’emmagasinement de nutriments notamment azotés, et
l’exportation des déchets organiques issus de la fauche peut participer au rééquilibrage
des sols (Berense et al. 1994, Schaffers et al. 1998) ;
• L’excavation des sols des bords de route serait aussi à étudier mais certainement
moins envisageable pour le moment compte-tenu des coûts très importants d’une telle
opération. Il n’en demeure pas moins que les sols des bords de route se sont révélés
être hautement pollués (cf. Partie § C. (1) de la Discussion) ; en accord avec les études
déjà menées sur le sujet (Thompson et al. 1986, Rutter et Thompson 1986, Carslaw
2005). De telles concentrations en azote, sels et métaux lourds ne sont pas sans
conséquences sur la biodiversité : microbes du sol (Post et Beeby 1996), plantes
(Akbar 2003, Singh et al. 2003, Johnston et Johnston 2004, Truscott 2005) et insectes
(Alstad et Edmunds 1982, Spencer et al. 1988, Throop et Lerdeau 2004).
La question des sols a aussi été abordée lors de notre étude mais malgré des premiers résultats
encourageants (cf. le rapport de M1 d’Amélie DELERUE pour les plantes : www2.
mnhn.fr/cersp/IMG/zip/Delerue_2008.pdf.zip, et cf. le rapport de M1 d’Elise CONTAN cité
précédemment pour les carabes :, www2.mnhn.fr/cersp/IMG/zip/ Contan_2007.pdf.zip ), il
n’a pas été possible de pousser des investiguations plus loin : tous les sols étant très pollués, il
était difficile de dégager des résultats plus intéressants.
- 93 -
PERSPECTIVES
« Le gain de la recherche, c'est la

recherche elle-même ».
GREGOIRE DE NYSSE
- 94 -
L’étude des effets écologiques des bords de route s’est révélée très intéressante. Elle
implique le travail avec de nombreux acteurs : politiques (Ministère et Conseil général),
gestionnaires des routes (Direction Départementale de l’Equipement, Service « Route »
du C.G., bureaux d’étude et agriculteurs de fait) et encore d’autres acteurs (associations
naturalistes, citoyens et scientifiques). L’attente est forte auprès de ces différentes
personnes au niveau du développement (a) de méthodes d’évaluation des impacts de la
construction et de la gestion des infrastructures, et (b) de méthodes de suivis sur le long
terme des mesures mises en place pour la biodiversité.
§ A. NECESSAIRE POURSUITE DE L’ETUDE DES EFFETS DES ROUTES SUR LA BIODIVERSITE
Il s’agit du moment difficile où l’impression du travail inachevé domine (études à poursuivre

ou à lancer, données à analyser, résultats à valoriser et articles à écrire) alors qu’il faut mettre
un point final à trois ans de travaux et passer à autre chose. Alors qu’il reste beaucoup à
chercher en écologie des routes.
En effet, de manière générale en écologie, et plus particulièrement au sein de l’U.M.R.
51 73, les travaux sur les routes sont nouveaux. Cette thèse, ainsi que celle d’Isabelle LE
VIOL, constituent une première étape dans ce champ d’étude de l’écologie assez proche de
l’écologie urbaine (il s’agit de milieux fortement anthropisés, perturbés et pollués). De
nombreuses questions se posent encore sur le sujet, et comme les enjeux liés (a) aux espaces
marginaux tels que les bords de route au sein de paysages agricoles intensifs, et (b) à la
compréhension des impacts des actions anthropiques sur la biodiversité, constituent des
enjeux majeurs en terme de recherche pour les années à venir ; la poursuite de ces travaux
semblent être une nécessité.
§ B. NECESSAIRE COMMUNICATION ET VULGARISATION DES SAVOIRS SCIENTIFIQUES
Si davantage de recherches sont nécessaires, il n’en reste pas moins que les connaissances
scientifiques sont aujourd’hui suffisantes pour affirmer qu’il existe des impacts très forts des
routes sur la biodiversité et qu’il est possible de les réduire par des politiques de construction
- 95 -
et de gestion appropriées. La prise en compte de la biodiversité dans toute politique est

désormais un enjeu majeur pour les années à venir.
Ce défi ne pourra être relevé qu’à travers la sensibilisation de la société à la
conservation de Nature, et de toute la Nature : Nature ordinaire comprise. Si la prise de
conscience sociétale autour de la question de la Nature protégée semble désormais en bonne
voie, la question de la Nature ordinaire est encore largement ignorée. Il est vrai qu’un sujet
sur les bords de route, « Intérêts écologiques des bords de route… », peut prêter à sourire,
malheureusement, chez beaucoup de personnes pourtant sensibilisés à la nécessaire
conservation de la biodiversité. Le sourire est à la mesure par lui-même du travail à accomplir
sur ce sujet.
Pour conclure ces travaux de thèse, je me permets de dire que cette situation m’a fait
penser à celle de Francis PONGE (1899-1988). Poète du parti pris des choses, il n’a eu de
cesse d’essayer et d’opérer un retournement complet de la poésie par rapport à une
conception romantique alors à la mode. Un texte, tout ce qui peut paraître de plus ordinaire à
la première lecture (à tel point que le lecteur s’en demande franchement l’utilité), lorsqu’il est
décrypté révèle sous une comparaison incongrue entre un tramway et la guêpe, toute l’utilité
de la Nature. Poète de l’ordinaire, loin de tout sentimentalisme, Francis PONGE révèle l’utilité,
et par là-même la beauté, des objets et des mots les plus ordinaires par leurs qualités
physiques et recourt volontiers au vocabulaire technique des sciences expérimentales ;
signalant même à plusieurs reprises sa dette envers Buffon. Une reconnaissance qui peut
sembler prémonitoire à l’heure où ce sont des disciples de Buffon, écologues de la rue du
même nom, qui tentent désormais de faire reconnaître à leurs contemporains la nature
ordinaire comme utile, nécessaire et donc belle ; bien que semblant sans intérêt au premier
abord.
L’étude de la Nature ordinaire est indispensable à l’écologie et à l’Homme tout
comme les mots ordinaires le sont à la pensée et à l’expression comme Francis PONGE le
montra à travers sa poésie prosaïque.
- 96 -
ANNEXES
MANUSCRITS COMPLÉMENTAIRES
&
GUIDE TECHNIQUE A DESTINATION

DU M.E.E.D.D.A.T.
« Quel peut donc être notre intérêt à

courir si nous nous trouvons sur la
mauvaise route ? »
PROVERBE
- 97 -
MANUSCRIPT N°5
MANUSCRIPT N°5
In preparation
Effects of local environment and human activities on plant communities of field margins
in an intensive agrarian landscape
Louis de REDON
Aurélie GARNIER
Jane LECOMTE
Robert HAICOURT
Jean-Michel DREILLAUX
Sandrine PIVARD
&
Agnès RICROCH
- 98 -
MANUSCRIPT N°5
PRESENTATION
TITRE
Impacts des conditions locales et des activités humaines sur les communautés végétales des bords
champ en paysage agricole intensif
RESUME
Nous avons étudié les impacts de l’organisation des exploitations agricoles et des pratiques des
agriculteurs qur les communautés végétales des bords de champ au sein d’un paysage agricole
intensif français. Des inventaires de végétation exhaustifs ont été menés au sein de 83 bordures de
champ et un total de 187 taxons identifiés. Nous avons sélectionné 11 paramètres pour expliquer la
richese et la diversité fonctionnelle des communautés végétales des bords de champ : des
caratéristiques de l’exploitation (taille, nombre d’employés, remembrement), la localisation des
bordures (adjacentes ou pas à des routes) et des pratiques de traitements des bordures (type de
traitements et calendrier). Il est apparu que les 16 agriculteurs contactés, lors d’une enquête que
nous avons menée, avaient des modes de gestion intensifs de leurs bordures de champ et que peu de
variabilité existait en fait entre les différentes modalités de traitement. Malgré ce, nos résultats ont
montré que les richesses spécifiques et fonctionnelles des communautés végétales des bords de
champ étaient affectées par quelques paramètres : les bordures adjacentes aux routes étaient plus
riches et le fauchage intensif favorisait les plantes à floraison tardive (automne).
MOTS CLEFS
ACP, Agriculteurs, Biodiversité, Bords de route, Politiques environnementales, Traits fonctionnels.
- 99 -
MANUSCRIPT N°5
EFFECTS OF LOCAL ENVIRONMENT AND HUMAN ACTIVITIES ON PLANT
COMMUNITIES OF FIELD MARGINS IN AN INTENSIVE AGRARIAN LANDSCAPE
Louis de REDON1,*, Aurélie GARNIER2, Jane LECOMTE2, Robert HAÏCOUR2, Jean-

Michel DREUILLAUX2, Sandrine PIVARD2 and Agnès RICROCH2,3.
KEYWORDS ABSTRACT
Biodiversity We studied the impacts of farm structure and farmer practices on
Environmental policies field margins plant communities in a French intensive agrarian
Farmers region. An exhaustive evaluation of species richness was carried
Functional traits out in 83 field margins and a total of 187 taxa were identified. We
PCA selected 11 parameters to explain richness and traits composition of
Road verges plant communities in field margins: some farm characteristics (size,
number of workers, land regrouping), field margin localisation
(adjacent to a road or a track) and weed control practices (weed
control type and calendar). It appeared that almost all the 16
farmers contacted during the survey were using very intensive weed
control on their field margins and that only few variations could be
detected between different types of intensive weed control.
However, our results showed that plant traits and community
richness were affected by some parameters: field margin adjacent
to roads were richer and mowing favoured flowering in autumn.
I. INTRODUCTION events generally depends on the availability

of manpower, machines, farm structure and
Biodiversity is becoming rarer within farmer sensibility to biodiversity protection.
intensive agrarian landscapes despite of its In France, ordinary weed control practices in
major roles in agro-ecosystem functioning FM usually includes from two to three
(Altieri 1999; Le Coeur et al. 2002). Field mowings a year and/or herbicide spraying
margins (FM) are crucial habitats for plant with glyphosate. In order to provide farmers
conservation in intensive agrarian landscapes with the best advice to preserve biodiversity
in such context. Indeed, as permanent non in this paper, we sought to investigate the
crop habitats, they are generally considered to potential impact of different parameters on
play a larger role than arable land in richness and composition of plant
maintaining wildlife diversity (Benton et al. communities.
2003). Plant diversity is low and mainly Because analysis of plant communities
concentrated in field margins (FM) (Burel et via their traits composition is now a priority
al. 1998; Le Coeur et al. 2002) and they are (McGill 2006) and because this allows a
known to constitute one of the main refuges good understanding of plant communities
for wild plant species in agrarian landscapes composition (Lavorel & Garnier 2002; Pywell
(Marshall & Moonen 2002). et al. 2003), sampled FM were studied
Farm characteristics and farmer according to some chosen life traits as
practices are very likely to affect plant flowering period, pollen and seed dispersal,
communities of FM. The number of mowing light tolerance, etc. We estimated the species
* Corresponding author; E-Mail: redon@mnhn.fr; Phone: 0033 662 045 936; Fax: 0033 140 793
835; 1- Laboratory of “Conservation des Espèces, Restauration & Suivi des Populations”; UMR 51-
73 CNRS-MNHN-UPMC; 55, rue Buffon; F-75005 Paris; FRANCE; 2- Laboratory of “Ecologie,
Systématique & Evolution”; UMR 80-79 ; CNRS-Université Paris Sud-AgroParisTech; Bat. 360-
362; F-91405 Orsay cedex; FRANCE; 3- AgroParisTech ; Department of “Sciences de la Vie &
Santé”; 16, rue Claude Bernard; F-75231 Paris cedex 05; FRANCE.
- 100 -
MANUSCRIPT N°5
richness and functional diversity of plant 1. Study area and site sampling
communities during spring 2005 in
different farms with different management The study area is located in the central
practices. The experiment was carried out region of Beauce which is a very intensive
in the region of Beauce (France), a typical agrarian zone of France.Thanks to other
intensive agrarian landscape composed studies (Pivard et al. 2008), informations
almost exclusively of agricultural fields. about landcover and farmers have been
previously registered within an area of
46 km² (Fig. 1A) around the village of
II. MATERIALS AND METHODS Selommes (47°45′ N, 1°11′ E). This area is
mainly composed by crop fields and their
Here, the term “field margin” (FM) is FM, by roads and tracks, and by small
adapted from Greaves and Marshall (1987) isolated farms.
and defined as the whole of the crop edge, We focused our study on the central
any margin strip present and the semi- part of this area and divided it into 28
natural habitat associated with the 1 km × 1 km plots (Fig. 1B). A small part
boundary. In this study, they are all of the last plot (plot G4) was not in the
adjacent to a road or to a track. previous studied area of 46 km² and its
FIGURE 1
STUDY AREA AND SAMPLING PROTOCOL
1A- Study area of 46 km² (roads and tracks are painted in black)
1B- Study area divided in to 28 (7x4) plots of 1 km²
1C- FM randomly chosen (3 by plot)
1D- Location of the 83 FM studied (sites are represented by a black point)
- 101 -
MANUSCRIPT N°5
dimensions were reduced to 0.7 km². In the fields adjacent to the studied FM to
each plot, 3 points were placed (excepting collect the following informations:
for plot G4 where only 2 points - Farm attributes: farm area (in hectares),
weresampled to take its size into account) work time unit (number of workers into
according to randomly chosen coordinates the barn corrected by their time
(x, y). The nearest FM to the point was presence), land regrouping (i.e. a value
then selected as a study site (Fig. 1C). between 1 and 5: 1 if fields where all
According to this sampling, 83 FMs have scattered around the study area and 5 if
been randomly selected into a 27.7 km² all fields where adjacent to the farm)
study area (Fig. 1D). and distance to farm buildings for each
We examined 30 FMs adjacent to FM (in meters);
an asphalt road (i.e. road in on side and - Weed control calendar: number of events
field in the other side of the FM) and 53 (between 1 and 4) and period of
FMs adjacent to a private grassy track (i.e. application (March/April, and/or
track in on side and field in the other side May/June, and/or July/August);
of the FM). This point is very important - Type of weed control: vegetation
because the State road agency may also mowing, and/or herbicide spraying.
take part in weed control performed in FM All FM were described by those ten
for those adjacent to a roads. variables and their localisation (i.e.
adjacent to a road vs. to a track). Those
eleven variables are referenced as “FM
2. Vegetation inventories characteristics”.
In each FM a 1 m × 25 m plot was

positioned along the FM. Vegetation 4. Data analysis
sampling (restricted to Angiosperm taxa)
was performed during one week in May To test the impacts of local environment
2005. The period was short enough to and management on FM plant
avoid temporal bias in vegetation communities, we made two groups of tests:
development. In each plot, we listed all (1) Richness (number of taxa, noted
plant taxa detected. Almost all the taxa N) was recorded. We then performed
were identified to species level according ANOVAs on N with the eleven FM
to the International Plant Names Index characteristics as explicative variables.
(http://www.ipni.org). In some cases Because those analyses represented a quite
(Crepis spp. for example), taxa were large amount of tests we reduced the
identified to genus level due to difficulties significant probability to 0.027 (instead of
in identification when FMs had been 0.05) according to False Discovery Rate
mown or sprayed. method (Benjamini et al. 1995):
We then established a file for six
with 1 = 0.05 and m = 11;
life traits (Phenology, Vegetative
reproduction, Flowering period, Prefered (2) We then calculated for each FM
soil, Shadow affinity & Raunkiaer type) the percentage of plants of each trait
and their different modalities (18 in total) modalities for the six life traits and we
for all the species detected (Table 1). For performed a PCA on sites and their traits
each species traits modalities were found composition (percentages). We then ran
in Fitter (1986). ANOVAs on the coordinates of sites on the
more important axes of the PCA (each axis
representing more than 10% of the total
3. Management practices of farmers and variance) with the eleven FM
farm structures characteristics as explicative. We used the
significant probably of 0.027 as previously
We contacted the 16 farmers who owned for the same reasons.
- 102 -
MANUSCRIPT N°5
TABLE 1: TABLE 2
LIST OF THE 18 TRAITS RESULTS OF ANOVAS PERFORMED
MODALITIES CHARACTERISTICS ON LOCAL PLANT RICHNESS (N)
DESCRIBING PLANT COMMUNITIES AND 11 VARIABLES DESCRIBING
FIELD MARGINS ENVIRONMENT
Traits Trait modalities Symbol AND MANAGEMENT
Annual A1
Phenology Variables describing Plant richness (N)
Bisannual A2 field margin (FM) t-value P Effect
Perennial A3
Vegetative Yes B1 Farm area -0.724 0.471
reproduction No B2
Work time
2.057 0.043 +
Spring C1 Farm unit
Flowering
Summer C2 attributes Land
period -0.930 0.355
Autumn C3 Regrouping
Poor D1 Distance to
1.094 0.278
Prefered farm buildings
soil Rich D2
Nb of weed
Fertilized D3 1.945 0.055 +
control events
Shadow Yes E1
affinity Weed March - April 0.151 0.880
No E2 control
Cryptophyte F1 calendar May - June 0.583 0.561
Raunkiaer Geophyte F2
type Hemicryptophyte F3 July - August 1.872 0.065 +
Nanophanerophyte F4
Type Mowing 1.150 0.254
of weed
Traits modalities from Fitter (1986). control Spraying -0.224 0.824
All statistical analyses were performed Field Margin Adjacent

5.609 <10-3 +
using R (Ihaka &Gentleman 1996). Localisation to road
III. RESULTS of them used glyphosate. If applied once,

spraying presented the advantage of a
1. Management practices of farmers and unique application per year (time savings).
farm structures Farmers did not discriminate weed control
in FM according to the type of adjacent
To control weeds in field margins (FM), crop. The period of herbicide application
68% of farmers (corresponding to 62 % of depended on vegetation height, Gramineae
FM) used mowing to counterpart state, time schedule, and equipment
disadvantages of chemical treatments. availability.
Cutting height varied from 0 to 50 cm and In order to avoid seed production of
the most cited reasons for performing FM weeds (and thereby seed dispersal),
mowing was preventing the colonisation farmers mowed FM in spring in March-
by perennial plants of crop fields when the April or May-June (when the vegetation
soil is nude. The second disadvantage of development is maximal), then during the
herbicide spraying was the cost of summer in July-August (when plants grow
chemical products, then the social pressure again). Some farmers mowed in July
for not using such products. Only 12.5% of (when plants are well developed), then in
farmers (corresponding to 7.6% of FM) August if necessary. Some farmers mowed
performed chemical treatments and 100 % twice before and after harvesting (in July-
- 103 -
MANUSCRIPT N°5
August). Some other farmers mowed once 3. Impacts of local environment and
in spring (in March-April or May-June). human management on plant life traits
in FM plant communities
2. Impacts of local environment and The PCA performed on sites and their
human activities on plant richness (N) of composition in plant life traits showed that
FM plant communities four axes represented each more than 10%
of the total variance (and altogether 68.9%
A total of 187 different taxa (Appendix 1) of the total variance). These PCA axes
were observed including 150 plants established strong discriminations between
identified at the species level (80.2%). FM according to their of plant
Only 30 taxa were recorded in more than communities composition so we focused
in 10% of the FM (Appendix 2). our tests on them (Fig. 2A and 2B).
The presence of a road adjacent to We did not detect any significant
the FM was the only characteristic with a effects of FM characteristics on the
significant positive effect on FM plant coordinates of FM along the first PCA axis
richness (t = 5.609; P < 10-3 – Table 2 ). 1 (Table 3). Some modalities of weed
We observed a positive tendency, control calendar had little influence on
(although insignificant), of three other those coordinates (insignificant
characteristics: “work time unit”, “number tendencies): negative effect of “number of
of weed control events” and “July/August weed control events” (t = -1.799; P =
weed control” (Table 2). 0.076) and of “March/April weed control”
TABLE 3
RESULTS OF ANOVAS PERFORMED ON COORDINATES OF STUDIED SITES ON
MAIN PCA AXIS (AXIS 1 TO AXIS 4) AND 11 VARIABLES DESCRIBING FIELD
MARGINS ENVIRONMENT AND MANAGEMENT.
- 104 -
MANUSCRIPT N°5
(t = -2.165; P = 0.033) and positive effects agricultural landscapes since they provide
of “May/June weed control” (t = 1.671; P a habitat or corridor for a large range of
= 0.099). plants and animals (Klejin & Snoeijing
We detected a strong and 1997). The species richness in FM can be
significant negative effect of the presence influenced by their structure and weed
an adjacent road to FM (t = -2.739; P = control performed by farmers in FM (Le
0.008) and a tendency for a positive effect Coeur et al. 1997). Because these authors
of spraying (t = 1.730; P = 0.087) on the found 224 species in hedgerow network
coordinates of FM along Axis 2 (Table 3). landscapes also located in western part of
We only found a significant France (16 km²), while we found 187
positive effect (t = 2.647; P = 0.01) of species in an openfield landscape; we
mowing on the coordinates of FM along assume that diversity in FM of an
Axis 3 (Table 3), but no significant effect openfield landscape is comparable to that
on the coordinates of FM along Axis 4 of FM in a hedgerow landscape is thus not
(Table 3). negligible.
IV. DISCUSSION
1. Farmer’s surveys
Field margins (FM) are composed by semi-
natural vegetation and constitute an Surveys we carried out allowed a better
important habitat area for plant population understanding of how farmers act on
persistence in landscapes dominated by landscapes in a practical manner, and could
intensive agriculture. Decreases in plant help to design future farming systems,
diversity in FM is generally linked to the integrating environmental goals. We had a
intensification of agricultural land use, pretty good return from our contacts to
causing a reduction of FM area, an farmers and most of them accepted to
increase of disturbances and input of agro- answer all our questions. The fact that only
chemicals from neighbouring arable fields few explicative factors had a significant
(Kleijn & Snoeijing 1997; De Snoo 1999). impact on plant diversity in FM could be
FM play an important role in the partially due to the two following facts:
FIGURE 2
PROJECTION OF THE 18 LIFE TRAITS ON AXIS N°1 TO N°4 OF THE PCA
A. Projection on Axis 1 vs. Axis 2 B. Projection on Axis 3 vs. Axis 4
- 105 -
MANUSCRIPT N°5
(1) All farmers were practicing a strong 3. Impacts of field margin management
FM management because they want to
control plants they describe as “weeds”. Mowing seemed to favour plants flowering
According to this observation, the type of in autumn and developing in rich soil (Axis
weed control actually performed by 3 – Fig. 2B & Table 3). This could be
farmers in the study area was generally explained by the fact that mowing occurs
intensive and therefore we did not have during spring and summer periods, and
contrasted weed control scenarios in the because mown material is not removed: its
studied FM. decomposition in FM results in a very rich
(2) The number of people working in soil potentially affecting plant
the farm and the existence of spraying communities (De Cauwer et al. 2005).
treatment performed on FM may have been Management calendar seemed to
underestimated in the data collected during slightly affect the composition of plant
the farmers’ survey because a part of the communities but this effect was
manpower might be undeclared workers insignificant (Axis 1– Fig 2A & Table 3).
and it is generally unpopular to admit using Plant richness seemed to be positively
chemical treatments. affected by three parameters “work time
unit”, “number of weed control events”
and “July – August weed control” (non
2. Impacts of local environment significant effect). In fact those three
parameters were not independent: “work
The strongest impact of local environment time unit” was strongly correlated to
on FM plant communities was the location “number of weed control events” (t =
of the FM: adjacent to an asphalt road or to 7.580; P < 10-3; R2 = 0.436) because when
a grassy. This could be explained by the the farm hires more workers, additional
fact that roads have strong direct impacts weed control events can be easily
on plants communities: pollution, performed.
fragmentation, light conditions, seed
dispersal by cars, etc. , which are known to
affect plant diversity and plant growth 4. Conclusion
rates in roadside verges (Formal &
Alexander 1998). FMs adjacent to a road This work showed the importance of field
are also managed by the local State agency margins as habitat for plant species of
for roads in addition to farmers (?) and are farming landscape (187 plant taxa
significantly wider than FM adjacent to a detected) and the relatively poor impact of
track (respective mean widths are 2.23 ± variations of weed control in a context of
0.13m and = 1.25 ± 0.05m, t = 8.223 & P < intensive farming practices. The results
10-3). Plant richness is larger in FM proved that considering different field
adjacent to a road and plants tolerating margins adjacent to roads, which are wide
shadow conditions to flower appeared to and where weed control is less intensive
be favoured there (Axis 2 – Fig. 2A). This and generally performed by State
result could be explained by a vegetation employees, it is possible to favour different
more developed in FM adjacent to a road and rich plant communities. Small
because (1) these FM are more extensively modifications in farmer practices in terms
managed (they are very wide and farmer’s of weed control would have little impact in
mowing device only covers 1 to 1.5 m); the attempt to make field margins become
and (2) microenvironmental conditions in a better refuge area for plant diversity in
FM adjacent to a road are very different intensive agrarian landscapes.
than that of FM adjacent to a track,
according to the soil composition that is
known to impacts plant growth (Angold
1997).
- 106 -
MANUSCRIPT N°5
ACKNOWLEDGMENTS De Snoo, G.R. (1999) Unsprayed field
margins: effects on environment,
We are grateful to the farmers of the region biodiversity and agricultural
of Selommes who allowed us to carry out practice. Landscape and Urban
experiments in their field margins. We Planning 46:151-160.
thank J Baudry for advice on the Fitter, R. (1986) Guide des fleurs sauvages
experiment design. The authors thank Ed Delachaux & Niestle Paris.
Mathilde Bouvron and Céline Robert for Forman, R.T.T., & Alexander, L.E. (1998)
surveys and data management respectively Roads and their major ecological
and Sovuthy Him-Thean for her assistance effects. Annual Review of Ecology
in maintaining and harvesting the Evolution and Systematics 29: 207-
experiment. The French Ministry of 231.
Research (2005-2006) provided financial Ihaka, R., & Gentleman, R. (1996) R: A
support for this study. Language for Data Analysis and
Graphics. Journal of
Computational and Graphical
REFERENCES Statistics 5: 299-314.
Klejin, D., & Snoeijing, G.I.J. (1997) Field
Altieri, M.A. (1999) The ecological role of boundary vegetation and the effects
biodiversity in agroecosystems. of agrochemical drift: botanical
Agriculture, Ecosystems and change caused by low levels of
Environment 74: 19-31 herbicide and fertilizer. Journal of
Angold, P.G. (1997) The impact of a road Applied Ecology 34:1413-1425.
upon adjacent heathland vegetation: Lavorel, S., & Garnier, E. (2002).
Effects on plant species Predicting changes in community
composition. Journal of Applied composition and ecosystem
Ecology 34: 409-417 functioning from plant traits
Benjamini, Y., & Hochberg, Y. (1995) revisiting the Holy Grail.
Controlling the false discovery rate: Functional Ecology 16: 545–556.
a practical and powerful approach Le Coeur, D., Baudry, J. & Burel, F.
to multiple testing. Journal of the (1997) Field margins plant
Royal Statistical Society, Series B assemblages: variation partitioning
(Methodological) 57: 289–300. between local and landscape
Benton, T.G., Vickery, J.A., & Wilson, factors. Landscape and Urban
J.D. (2003) Farmland biodiversity: Planning 37: 57-71.
Is habitat heterogeneity the key? Le Coeur, D., Baudry, J., Burel, F. &
Trends in Ecology and Evolution 4: Thenail, C. (2002) Why and how
182-188. we should study field boundary
Burel, F., Baudry, J., Butet, A., Clergeau, biodiversity in an agrarian
P., Le Cœur, D., Dubs, F., Morvan, landscape context. Agriculture
N., Delettre, Y., Paillat, G., Petit, Ecosystem and Environment 89:
S., Thenail, C., Brune, E. & 23-40.
Lefeuvre, J.C. (1998). Comparative Marshall, E. J. P. & Moonen, A.C. (2002)
biodiversity along a gradient of Field margins in northern Europe:
agricultural landscapes. Acta their functions and interactions with
Oecologica 19: 47-60. agriculture. Agriculture,
De Cauwer, B., Reheul D., D’hoogh, K., Ecosystems and Environment 89: 5-
Nijs, I. & Milbau, A. (2005) 21.
Evolution of the vegetation of McGill, B.J., Enquist, B.J., Weiher, E. &
mown field margins over their first Westoby, M. (2006) Rebuilding
3 years. Agriculture, Ecosystems & community ecology from functional
Environment. 109: 87–96. traits. Trends in Ecology &
Evolution 21:178-185.
- 107 -
MANUSCRIPT N°5
Pivard, S., Adamczyk, K., Lecomte, J., Applied Ecology 45: 476-485.
Lavigne, C., Bouvier, A., Deville, Pywell, R.F., Bullock, J.M., Roy, D.B.,
A., Gouyon, P.H. & Huet, S. (2008) Warman, L.I.Z, Walker, K.J. &
Where do the feral oilseed rape Rothery, P. (2003) Plant traits as
populations come from? A large- predictors of performance in
scale study of their possible origin ecological restoration. Journal of
in a farmland area. Journal of Applied Ecology 40: 65-77
1
2
3
4 SUPPLEMENTARY MATERIALS
5
6
7 APPENDIX 1
8
9 LIST OF THE 187 TAXA OBSERVED IN FIELD MARGINS
10
Taxa Family Capsella bursa-pastoris Brassicaceae
Acer platanoides Aceraceae Carduus nutans Asteraceae
Achillea millefolium Asteraceae Carex divulsa Cyperaceae
Adenostyles alliariae Asteraceae Carex muricata Cyperaceae
Agrimonia eupatoria Rosaceae Centaurea jacea Asteraceae
Agropyrum repens Poaceae Centaurea scabiosa Asteraceae
Agrostis ssp. Poaceae Centaurea ssp. Asteraceae
Agrostis stolonifera Poaceae Ceorynephorus canesciens Poaceae
Agrostis tenuis Poaceae Cerastium ssp. Caryophyllaceae
Ajuga reptans Lamiaceae Cerastium triviale Caryophyllaceae
Allium ssp. Lamiaceae Chenopodium album Chenopodiaceae
Alopecurus agrestis Poaceae Chrysantenum segetum Asteraceae
Alopecurus ssp. Poaceae Chrysanthemum leucanthemum Asteraceae
Amaranthus retroflexus Amaranthaceae Circium lanceolata Asteraceae
Anagallis arvensis Primulaceae Cirsium arvense Asteraceae
Anthriscus sylvestris Apiaceae Cirsium vulgare Asteraceae
Anthriscus vulgaris Apiaceae Convolvulus arvensis Convolvulaceae
Aphenes arvensis Rosaceae Conyza canadensis Asteraceae
Arabidopsis thaliana Brassicaceae Crataegus oxyacantha Rosaceae
Arctium lappa Asteraceae Crepis setosa Asteraceae
Arenaria serpyllifolia Caryophyllaceae Crepis ssp. Asteraceae
Arrhenaterum elatius Poaceae Dactylis glomerata Poaceae
Artemisia vulgaris Poaceae Daucus Carota Apiaceae
Atriplex ssp. Asteraceae Epilobium ssp. Onagraceae
Avena sativa Chenopodiaceae Erigeron cancolensis Asteraceae
Bellis perennis Asteraceae Erodium cicutarium Geraniaceae
Brachypodium pinnatum Chenopodiaceae Erophila verna Brassicaceae
Brachypodium sylvaticum Poaceae Ervum lens Fabaceae
Brassica napus Brassicaceae Eryngium campestre Apiaceae
Brassica nigra Brassicaceae Euonymus europaeus Celastraceae
Brassica ssp. Brassicaceae Euphorbia aparisias Euphobiaceae
Bromus arvensis Poaceae Euphorbia helioscopia Euphobiaceae
Bromus maximus Poaceae Fallopia convolvulus Polygonaceae
Bromus mollis Poaceae Festuca arumdinacea Poaceae
Bromus sterilis Cucurbitaceae Festuca pratense Poaceae
Bryonia dioica Lamianae Festuca rubra Poaceae
- 108 -
MANUSCRIPT N°5
Galium aparine Rubiaceae Poa pratensis Poaceae
Galium cruciata Rubiaceae Poa ssp. Poaceae
Galium mollugo Rubiaceae Poa trivialis Poaceae
Gallium ssp. Rubiaceae Polygonatum multiflorum Convallariaceae
Geranium dissectum Geraniaceae Polygonum aviculare Polygonaceae
Geranium molle Geraniaceae Polygonum convolvulus Polygonaceae
Geranium pirenaicum Geraniaceae Polygonum lapatifolium Polygonaceae
Geranium robertianum Geraniaceae Polygonum persicaria Polygonaceae
Geranium rotundifolium Geraniaceae Portulaca oleracea Portulacaceae
Geranium ssp. Geraniaceae Potentilla fragaria Rosaceae
Hedera helix Araliaceae Potentilla reptans Rosaceae
Heracleum ssp.ondylium Apiaceae Prunus avium Rosaceae
Hordeum murinum Poaceae Prunus spinosa Rosaceae
Hypericum ssp. Hypericaceae Pulmonaria sp Borraginaceae
Juncus conglomeratus Juncaceae Quercus robur Fagaceae
Kickxia ssp.uria Plantaginaceae Ranunculus acris Renonculaceae
Knautia arvensis Dipsacaceae Ranunculus bulbusus Renonculaceae
Lactuca scariola Asteraceae Ranunculus repens Renonculaceae
Lamium purpureum Labiaceae Ranunculus sp Renonculaceae
Lamium ssp. Lamiaceae Raphanus raphanistrum Brassicaceae
Lampsana communis Asteraceae Renunculus sardus Renunculaceae
Lathyrus pratensis Fabaceae Rosa sp. Rosaceae
Lathyrus ssp. Fabaceae Rubus sp Rosaceae
Lathyrus tuberosus Asteraceae Rumex acetosa Polygonaceae
Ligustrum vulgare Oleaceae Rumex crispus Polygonaceae
Linaria supinia Plantaginaceae rumex obtusifolius Polygonaceae
Linaria vulgaris Plantaginaceae Rumex sanguineus Polygonaceae
Lolium perenne Poaceae Rumex sp Polygonaceae
Lotus corniculata Fabaceae Sagina sp Caryophyllaceae
Lychnis dioica Caryophyllaceae Sanguisorba sp Crassulaceae
Malva neglecta Malvaceae Sedum telephium Crassulaceae
Malva rotondifolia Malvaceae Senecio vulgaris Asteraceae
Matricaria chamomilla Asteraceae Silene inflata Caryophyllaceae
Matricaria discoidea Asteraceae Silene latifolia Caryophyllaceae
Matricaria inodora Asteraceae Sinapis arvensis Brassicaceae
Matricaria ssp. Asteraceae Sisymbrium officinale Brassicaceae
Medicago lupulina Fabaceae Solanum nigrum Solanaceae
Mercurialis annua Euphobiaceae Sonchus asper Asteraceae
Myosotis arvensis Boraginaceae Sonchus oleraceus Asteraceae
Mysosotis ssp. Boraginaceae Stachys betonica Labiaceae
Ononis repens Fabaceae Stellaria graminea Caryophyllaceae
Origanum vulgaris Lamiaceae Stellaria media Caryophyllaceae
Orobanche ssp. Orobachaceae Tamus communis Dioscoreaceae
Papaver rhaeas Pavaveraceae Taraxacum dens-leonis Asteraceae
Papaver ssp. Papaveraceae Taraxacum sp Asteraceae
Pastinaca sativa Apiaceae Tragopogon dubius Asteraceae
Phleum pratense Poaceae Tricatum flavecens Fabaceae
Picris heracioides Asteraceae Trifolium minus Fabaceae
Picris ssp. Asteraceae Trifolium repens Fabaceae
Plantago coronopus Plantaginaceae Ulmus campestris Ulmaceae
Plantago lanceolata Plantaginaceae Urtica dioica Urticaceae
Plantago major Plantaginaceae Verbena officinialis Verbenaceae
Poa annua Poaceae Veronica arvensis Plantaginaceae
Poa nemoralis Poaceae Veronica chamaedris Plantaginaceae
Veronica hederifolia Plantaginaceae
- 109 -
MANUSCRIPT N°5
Veronica persica Plantaginaceae Viola arvensis Violaceae
Vicia cracca Fabaceae Viola sp Violaceae
Vicia hirsuta Fabaceae Vulpia myuros Poaceae
Vicia sativa Fabaceae 1
APPENDIX 2
LIST OF 30 SPECIES PRESENT ON
MORE THAN 10% OF FIELD MARGINS
Taxa (%)
Convolvulus arvensis 67%
Bromus mollis 59%
Bromus sterilis 57%
Poa annua 56%
Poa pratensis 53%
Geranium dissectum 51%
Dactylis glomerata 47%
Lolium perenne 39%
Galium aparine 38%
Plantago major 33%
Geranium molle 31%
Taraxacum dens-leonis 28%
Cirsium arvense 27%
Artemisia vulgaris 27%
Achillea millefolium 25%
Agropyrum repens 25%
Veronica persica 25%
Polygonum aviculare 24%
Plantago lanceolata 23%
Potentilla reptans 22%
Papaver rhaeas 21%
Senecio vulgaris 21%
Heracleum spondylium 20%
Brassica napus 18%
Arrhenaterum elatius 17%
Sonchus asper 17%
Veronica arvensis 17%
Polygonum convolvulus 16%
Capsella bursa-pastoris 14%
Matricaria discoidea 11%
(%): Percentage of field margins where

species were observed.
- 110 -
MANUSCRIPT N°6
MANUSCRIPT N°6
Submitted to Weed Research
A three-year study of weed management on plant community

in field margins in an openfield landscape
Nadia MICHEL
Aurélie GARNIER
Robert HAICOURT
Louis de REDON
Aude SOURISSEAU
&
Agnès RICROCH
- 111 -
MANUSCRIPT N°6
PRESENTATION
TITRE
Effets du contrôle des adventices sur les communautés végétales des bordures de champs dans
un paysage agricole intensif, trois ans d’étude.
RESUME
Nous avons étudié durant trois ans les effets du contrôle des adventices (par fauche et produits
chimiques) sur la composition et la richesse des communautés végétales de 27 bordures de champ
au sein d’un paysage d’agriculture intensive. Onze paramètres, incluant les modalités de
pollinisation, ont été définis pour décrire la flore des bordures de champ et leurs variations ont été
analysées à l’aide d’analyses multivariées et de tests d’égalité sur les observations pairées. Les
paramètres testés se sont révélés affectés les systèmes de gestion des agriculteurs ; dépendant des
modes de contrôle des populations d’adventices. La richesse spécifique observée a augmenté entre
2005 (102 espèces) et 2007 (134 espèces) coïncidant avec une extensification des pratiques de
gestion. Les récents changements de pratique de gestion, abandonnant la volonté de contrôle des
populations d’adventices dans les bordures de champ adjacents aux routes (à l’initiative des
nouvelles politiques environnementales françaises), a montré l’effectivité de certaines mesures pour
protéger la biodiversité et augmenter la qualité de espaces marginaux comme les bordures de champ
comme habitats. Dans ce contexte, l’augmentation observée d’espèces végétales entomophiles peut
procurer des ressources en pollen et en nectar aux populations de pollinisateurs permettant leur
maintien au sein des paysages agricoles intensifs. Le nombre d’espèces non-entomophiles a aussi
augmenté pendant les trois années d’étude ce qui a montré qu’elles pouvaient aussi se maintenir au
sein de bordures de champ perturbées dans les agro-écosystèmes.
MOTS CLEFS
Contôle des adventices, Zones non-cultivées, Pratiques de gestion, Agro-écologie & Plantes
entomophiles.
- 112 -
MANUSCRIPT N°6
A THREE-YEAR STUDY OF WEED CONTROL ON PLANT COMMUNITY
IN FIELD MARGINS IN AN OPENFIELD LANDSCAPE
Nathalie MICHEL1,2,3, Aurélie GARNIER1,2,3, Robert HAÏCOUR1,2,3, Louis de REDON4, Adia

SOURISSEAU1,2,3and Agnès RICROCH*,1,2,3,5.
KEYWORDS ABSTRACT
Weed control This study examined the effects of three years of weed control (mowing
Uncultivated areas and spraying) on the composition and richness of plant communities in
Farming practices 27 field margins (FM) in an openfield landscape. Eleven synthetic
Agro-ecology parameters, including pollination modes and life-form types, were
Insect-pollinated defined to describe the FM flora and their variations were analysed
species using multivariate analyses and pair-wise comparisons. The synthetic
parameters were affected by some components of the farming system.
They all depended on weed control, but differently according to the
year. The species richness increased from 2005 (102 species) to 2007
(134 species). This increase in species richness coincided with a
decrease in the number of mowed or sprayed FM. The recent adoption
of farming practices avoiding weed control in FM adjacent to roads
(validated by the French Departmental Commission for Agricultural
Policy) proved effective to preserve plant communities and rapidly
enhance the quality of FM habitats. For insect-pollinated species, their
persistence and their ability to flower provided pollen and nectar
resources for pollinators that could help ensure the persistence of the
pollinator community. Species that are not insect-pollinated increased
also, which showed that they were able to persist in the disturbed
habitats of the FM in an agro-ecosystem..
I. INTRODUCTION to generate great community variability in

space and time (Harper, 1977). Field margins,
At the local scales of the field and the farm, as permanent non crop habitats, are generally
improvements in crop management considered to play a larger role than arable
techniques have been adopted and they now land in maintaining wildlife diversity in
involve weed control by mowing and agricultural landscapes (Benton, Vickery and
herbicide spraying, increased use of fertilizer, Wilson 2003). Here, the term “field margin”
simplification of crop rotations, and (FM) is adapted from Greaves and Marshall
improvements in seed-cleaning techniques (1987) and defined as the whole of the crop
(Benton et al., 2003). Arable plant edge, any margin strip present and the semi-
communities are thus exposed to a frequently natural habitat associated with the boundary.
changing environment, which is in turn likely Any plant species that occurs in a FM is
*Corresponding author, Laboratoire Ecologie, Systématique & Evolution. Bâtiment 360. Université
Paris-Sud 11 91405 Orsay cedex France, Phone: +33-1 69 15 56 65; Fax: +33-1 69 15 46 97, Email
address: agnes.ricroch@u-psud.fr; 1, Université Paris-Sud. Faculté des Sciences d’Orsay. F-91405
Orsay cedex, FRANCE; 2, CNRS. Laboratoire Ecologie, Systématique & Evolution, Bâtiment 360.
F-91405 Orsay cedex, FRANCE; 3, AgroParisTech. Laboratoire Ecologie, Systématique &
Evolution, Bâtiment 360. F-91405 Orsay cedex, FRANCE; 4, Laboratoire Conservation des
Espèces, Restauration & Suivi des Populations, UMR 5173 CNRS-Muséum National d’Histoire
Naturelle. 55, rue Buffon. F-75005 Paris, FRANCE; 5, AgroParisTech. Département des Sciences
de la Vie & Santé, 16, rue Claude Bernard. F-75231 Paris cedex 05, FRANCE.
- 113 -
MANUSCRIPT N°6
generally considered as a weed species by and speciation. Indeed, modifications of the
farmers because it can cause major crop available local nectar resources could
losses. From a point of view of biodiversity influence the spatial patterns of insect-
preservation these weed species are however mediated gene flow at the landscape scale
important components of the agro-ecosystem because a lack of insect-pollinated species as
(Marshall et al., 2003). Indeed, reduced weed local food resources would make the
densities may result in fewer weed seeds pollinators travel over longer distances,
being available as food for wintering birds thereby favouring long-distance pollen
(Watkinson et al., 2000) and may reduce the dispersal (Kohler et al., 2007). Biesmeijer et
number of invertebrate herbivores, together al. (2006) showed that linked elements in
with their predators. The management of FM biological communities (i.e. specialist
is complex as FM are biodiversity refuges, pollinators and the obligatory outcrossed
and also potential relays for the spread of plants that they pollinate) are declining in
transgenes since they are the main habitat of tandem in Britain and in the Netherlands.
feral populations of OSR. GM feral plants can In this paper we present the first French
introduce transgenes into wild and weed openfield assessment of the impacts of
populations in FM. chemical and mechanical control options on
Since 1998 we studied the dynamics of the floristic composition in FM. The main
feral populations of oilseed rape (OSR; purpose of this study was the evaluation of (1)
Brassica napus L.) in an open field landscape the spatio-temporal variability of plant
in France (Pivard et al. 2008). Feral OSR community compositions in relation to its
populations mainly result from the escape of pollination mode in FM and (2) the impact of
OSR from cultivated crops and are mainly weed control on this variability in an
located in FM. These feral populations as a openfield farming landscape.
source for further transgene flow by pollen We thus performed a three-year field
and/or seed dispersal can persist several years survey in 27 FM within a 46 km² openfield
(Hüsken & Dietz-Pfeilstetter 2007). To limit landscape in France (2005-2006-2007). We
transgene escape via feral populations FM used several synthetic parameters describing
spraying and/or mowing should be more richness and diversity of plant communities in
intense (Garnier, Deville and Lecomte, 2006). these FM to examine the impacts of mowing
Spraying herbicide onto or mowing FM and herbicide spraying on plant community
may affect the biodiversity in the local agro- composition. We addressed three main
ecosystem in three main ways by: (i) a direct questions: (i) Does plant community
impact on plant community composition, (ii) composition in FM vary over time? (ii) Can
a direct impact on pollinators due to herbicide these variations be related to the type of weed
toxicity or mowing of flowering FM, and (iii) control? (iii) Do these variations differ among
an indirect impact on pollinators if categories of plant species, especially
nectariferous and/or polleniferous plants are concerning insect-pollinated vs. not insect-
affected by weed control. pollinated species?
Pollinator population dynamics and
persistence of plants according to their
reproductive systems should be better 2. MATERIAL AND METHODS
documented to understand the complex
interactions between cropped and non-crop 2.1. Sampling design of FM plant
areas. It is also important to understand the communities
pollination processes that generate landscape-
scale gene dispersal in plants, particularly in The field study was conducted in an intensive
crop plants with genetically modified openfield agricultural region of 46 km2
varieties. At the landscape scale in particular, around Selommes (47° 45′ 24″ N; 1° 11’
gene flow among populations maintains the 34’’E) in Loir-et-Cher region (France). This
cohesion of a species’ gene pool, thereby study site is located within the temperate
diminishing the potential for local adaptation marine climate zone. The selection of field
- 114 -
MANUSCRIPT N°6
FIGURE 1: Lambinon et al. (2004). Then we checked the
LOCATION OF THE 27 FIELD MARGINS species names on the Germplasm Resources
IN THE STUDY AREA Information Network (GRIN) Taxonomy for
Plants (http://www.ars-grin.gov). We referred
to plant taxa for either a species (for plants
identified at the species scale) or a genus (for
plants identified at the genus scale). Indeed,
the external damage resulting from mowing
and/or spraying prevented the identification at
the species scale of some plant samples.
Along with floristic surveys, the
external damage to plants was recorded by
eye to detect whether the plants had been
sprayed with herbicide or mown before the
sampling session.
The black points represent the FM that were

surveyed during the field study. Roads and 2.2. Statistical analyses
paths are painted in black.
2.2.a. Definition of synthetic parameters
margins (FM) for the survey was performed describing plant community diversity
was first divided into 1 km x 1 km quadrats
using a stratified sampling: the 46 km2 area Eleven synthetic parameters were computed
and then locations of FM were randomly from the floristic data to describe the
chosen within each quadrat, with the composition of the plant community and to
constraint that FM were equally distributed evaluate its variations among years and
between those adjacent to a road and those among the FM. These synthetic parameters
adjacent to a path. A total of n = 27 FM were are the following: species richness, number of
thus selected and located with a Trimble® botanical families, Shannon index, Simpson
GPS: 14 were adjacent to a path (52 %), 8 to a index, number of Eudicotyledons and
two-way road (30 %), and 5 to a one-way Monocotyledons species, number of Annuals,
road (18 %) (Fig. 1). Biennials and Perennials species and number
To allow inter-annual comparisons of species that are known to be insect-
between the FM plant communities, the pollinated or not. Not insect-pollinated
floristic composition (restricted to species designates either autogamous or wind-
angiosperms) was surveyed in the 27 FM in pollinated species. Data about life form and
May (i.e. the peak bloom period of plant type of pollination are only available for the
species in FM) in 2005, 2006 and 2007. Two taxa identified at the species scale.
complementary sampling methods were The Shannon and Simpson diversity
carried out in each FM: 1) an exhaustive indices were calculated using the ‘diversity’
determination of all the taxa that were present function of the ‘vegan’ library (Oksanen et
along a 25 m × 1 m-strip and 2) the measure al., 2007) of the R software (R Development
of cover percentage (using the Braun- Core Team, 2004). In 2007, Shannon and
Blanquet index - 1932) for each taxa observed Simpson diversity indices calculated with two
in two (resp. five) 1 m²-quadrats randomly quadrats were significantly smaller than the
selected in the 25 m2 strip in 2005 (resp. 2006 same indices calculated with five quadrats.
and 2007). The total plant cover was also The same result was obtained in 2006 for the
recorded in each FM. The 25 m2 strip is Shannon index. As a consequence, we chose
assumed to be representative of the FM since randomly two quadrats among the five
the weed control unit is the same in the whole quadrats for 2006 and 2007 so that diversity
FM (Baudry et al., 2000). The identification indices were calculated with two quadrats for
of species was completed according to all years.
- 115 -
MANUSCRIPT N°6
Both ubiquitous and rares species were year, we performed Mann-Whitney tests with
counted each year of the study. Ubiquitous StatisticaTM to compare the synthetic
species (resp. rare species) are defined by parameters according to the type of weed
Gabriel, Thies and Tscharntke (2005) as control (No Control, Mowing, and Spraying)
species occurring in more than 50 % (resp. in To test whether the different types of weed
less than 5 %) of the quadrats surveyed, that control would favour some reproductive
is to say in more than 13 FM (resp. in less strategies, we particularly focussed on five of
than 2 FM) in our study. the synthetic parameters that are associated
with two life traits: (1) type of life form
2.2.b. Multivariate analyses: global view and (annual or biennial species vs. perennial
evolution of the FM plant communities species) and (2) type of pollination (insect-
pollinated vs. not insect-pollinated species).
We performed a co-inertia analysis (Dolédec
et al., 1996) on three annual synthetic tables
that were build with the eleven synthetic 3. RESULTS
parameters measured in the 27 FM (one table
for each year studied). These analyses were 3.1. General view of the plant community
not directly performed at the taxa level,
because the number of columns of matrices, The land-use in fields adjacent to the FM
i.e. the number of taxa observed (185 – see followed crop rotations (including bare
Results) would have been too high. This type fallows and fallows) and the more common
of multivariate analysis is a two-table crops were winter cereals, conventional
ordination method based on a covariance oilseed rape, and pea
matrix that is calculated for pairs of A total of 185 plant taxa were
(contingency) matrices (i.e. pairs of annual observed in the 27 FM including 154 taxa
synthetic tables here). The correlation identified at the species scale and 29 at the
between pairs of tables was measured by the genus scale (“spp.”) (150 Eudicotyledons; 35
RV coefficient (Robert and Escoufier, 1976) Monocotyledons; 37 botanical families; 112
computed with ADE-4 SoftwareTM genus - Appendix S1). The large majority of
(Thioulouse et al., 1997). A Monte-Carlo the taxa identified at the species scale were
permutation test was then performed for each equally distributed between Annuals (70
pair of tables to test whether the percentage of species, 45.5 %) and Perennials (75 species,
co-variation between the two tables was 48.7 %); Biennials were less numerous (9
significant (number of permutations: 1000). species, 12.0 %). The ratio insect-/wind-
pollinated species was 2.6 (108 vs. 41 species,
2.2.c. Inter-annual comparisons of the i.e. 70.1 % vs. 26.6%). The remaining five
synthetic parameters species were strictly autogamous (3.2 %). All
Perennials we recorded had the ability to
We performed paired-samples Wilcoxon tests reproduce sexually (Fitter, 1986). All plants
of the eleven synthetic parameters using observed in FM were herbs except seedlings
StatisticaTM Software (StatSoft, Inc.) to draw of Crataegus laevigata, Ligustrum vulgare and
inter-annual comparisons of FM plant Prunus spinosa (tree species), Hedera helix
community richness and diversity. (liana), and Avena sativa, Brassica napus and
Triticum aestivum (feral crops plants). We
2.2.d. Impact of weed control on community observed the presence of one feral OSR plant
composition in FM in 7 FM, 4 FM and 1 FM in 2005, 2006 and
2007 respectively.
We evaluated the impact of weed control on Among the total richness (185 taxa; 37
the composition of plant communities in FM families), 60 taxa (19 families) were present
by comparing the plant community the three successive years, 53 taxa (25
composition according to the different types families) two of the three years and 72 taxa
of control performed in FM. Thus for each (25 families) only one year.
- 116 -
MANUSCRIPT N°6
Number of mowed or sprayed field margins (FM) and FM with no weed control for each of the 3
years of the study, and number and proportion of species according to life forms and types of
pollination.
The total richness in the 27 FM insect-pollinated species) showed a trend to

increased from 2005 to 2007. Indeed a total of increase over years (Fig. 3). This increase in
102 taxa were found in 2005, 122 in 2006, parameters of richness over years was
and 134 in 2007 (Fig. 2). The number of statistically significant for all parameters at
botanical families slightly increased over least between two years (Fig. 3), except for
years, from 34 to 39 families. Interestingly we Biennials due to a very low number of species
noted that this increase in richness coincided that prevented detection of any significant
with the decrease of the number of mowed or trend. Evolution of diversity indices was quite
sprayed FM: 21 FM in 2005 (78 % of FM), 15 different: as richness increase, both Shannon
in 2006 (56 %), and only 13 in 2007 (48 % - and Simpson diversity indices increased from
Fig. 2). Conversely, the relative proportions 2005 to 2006 whereas they decreased in 2007
of Monocotyledons vs. Eudicotyledons (79 % (Fig. 3c).
vs. 21 %), Annuals vs. Perennials (45 % for The major global differences were
both), and insect-pollinated vs. not insect- observed between 2005 and 2006 (8
pollinated species (70 % vs. 30 %) were all parameters were significantly different - Fig.
stable from year to year. As well as total 3), and between 2005 and 2007 (9
Richness, numbers of ubiquitous species, rare significantly different parameters). The
species, insect-pollinated species and not differences between 2006 and 2007 were
insect-pollinated species all increased over minor (4 significantly different parameters).
years.
3.3. Effects of weed control on the plant
3.2. Synthetic parameters of the plant community
community and their evolution over years
Some of the studied FM were not submitted
Co-inertia analysis showed that the synthetic to any weed control (no mowing nor
tables (11 parameters) of 2006 and 2007 were spraying): 6 FM (22 % of 27 FM), 12 FM
significantly correlated (RV = 36.85 %; (44 %), and 14 FM (52 %) in 2005, 2006, and
Monte-Carlo test: p = 0.01) whereas no 2007 respectively. The remaining FM were
correlations were found between the tables of either mown or sprayed. In 2007 the 8 FM
2005 and 2006 (RV = 17.10 %; Monte-Carlo adjacent to a two-way road (ca. 30 % FM)
test: p = 0.10), and between 2005 and 2007 were not submitted to any control by the State
(RV = 9.66 %; Monte-Carlo test: p = 0.44). agents who adopted farming practices
All the eight parameters of richness avoiding weed control in FM adjacent to
(Total Richness; Number of Eudicotyledons, roads (validated by the French Departmental
Monocotyledons, Annuals, Biennials, Commission for Agricultural Policy). Our
Perennials, Insect-pollinated species and Not results showed that the different types of
- 117 -
MANUSCRIPT N°6
FIGURE 3: control (either mowing or spraying) that had a
CHANGES OVER THE THREE YEARS OF significant effect and the synthetic parameters
THE SYNTHETIC PARAMETERS that were affected a given year sometimes
DESCRIBING THE PLANT COMMUNITY differ from those of the other two years. For
COMPOSITION. example, the pattern observed in 2007 (i.e. a
significant effect of spraying - Fig. 4) clearly
differed from the pattern observed in 2006
and described above (i.e. a significant impact
of mowing). More precisely, in 2007 Total
Richness and numbers of Eudicotyledons,
Monocotyledons and Perennials were
significantly lower in sprayed FM compared
to FM submitted to any control (no significant
difference was observed between mowed FM
and FM with no weed control).
3.4. Type of crops adjacent to FM surveyed

and effects on the plant community
Land-use in fields that were adjacent to the

studied FM was mostly cereal crops: 20 FM
(74 % of FM), 18 (67 %), and 18 (67 %) were
adjacent to a cereal crop in 2005, 2006 and
2007, respectively. Oilseed rape crops and pea
crops represented a smaller proportion of
adjacent crops (a few FM each year, about 10-
20% of FM). Bare fallow and fallows were
even rarer and only concerned one or two FM
each year. We did not detect any effect of
adjacent land-use on community composition.
Moreover no correlation was observed
between the type of weed control of FM and
For each of the eleven synthetic parameters the land-use in field.
and for each pair-wise comparison,
significant differences (i.e. p < 0.05) were
indicated using different letters (a, b, c). 4. DISCUSSION
weed control had no impact on total richness In this three-year study, we highlighted the
in 2005 whereas it had a significant negative high inter-annual variability in plant
effect on total richness in 2006 and 2007 (Fig. community composition in field margins
4). (FM) in an openfield farming landscape and
In 2006 mowing had a significant showed that this variability was mediated by
effect on several synthetic parameters: plant weed control (mowing and herbicide
communities in mowed FM indeed showed spraying) performed in FM. The adjacent
lower values in Richness, Shannon and land-use (i.e. the nature of the adjacent crop
Simpson Diversity, Insect-pollinated species, and indirectly the associated farming
and Not insect-pollinated species than in FM practices) however did not have any impact of
with no weed control. Conversely, no effect FM plant composition in our study, in
of spraying in FM was detected in 2006 (Fig. opposition to field studies carried out in
4). hedgerows farming landscapes (Baudry et al.
The effects of weed control however differed 2000).
largely from year to year. Indeed, the type of
- 118 -
MANUSCRIPT N°6
FIGURE 4:
SIGNIFICANT EFFECTS OF WEED CONTROLS ON THE SYNTHETIC PARAMETERS OF
THE PLANT COMMUNITY
For each year separately, results were presented only for the synthetic parameters for which at
least one Mann-Whitney test was significant. For those synthetic parameters, results were however
presented for all the three types of controls (NoWC, Mowing and Spraying) for better clarity. For
each pair-wise comparison, significant differences (i.e. p < 0.05) were indicated using different
letters (a, b, c). The numbers of FM that were mown, sprayed or not submitted to any weed control
each year are indicated on the top of the figures (“n=”).
4.1. Variability of plant community concerned rare species, which were present in
larger proportions in species-rich FM. Rarer
Our results indicated a high variability in species may have smaller niche width and
space and time of all the synthetic parameters thereby an increased sensitivity to changing
that were chosen to describe the richness and weed control, which would explain the high
diversity of the FM plant community within a inter-annual heterogeneity in plant
46-km² openfield farming landscape in communities observed.
western France. Indeed, these synthetic Variability of plant species richness
parameters presented each year a large range and diversity are traditionally related to local
of variation and therefore pointed out the site conditions such as nutrient and water
disparity between FM (i.e. spatial variability). availability (Gabriel, Thies and Tscharntke,
Moreover, we underlined inter-annual 2005) and on factors determining the regional
variations of these synthetic parameters: they species pool. These factors concern
all generally increased with time (globally: approaches based on Ellenberg species
2005 < 2006 ≤ 2007). This increase also indicator values and approaches based on the
- 119 -
MANUSCRIPT N°6
occurrence of species in different syntaxa in new practices occurring since 2007 aim at
the framework of the Braun-Blanquet system improving the quality of FM for fauna and
– for example the number of species available plant preservation. The effect of practical
for colonization (Huston 1999). Plant solution was immediate in our study as seven
populations occurring within disturbed synthetic parameters describing FM richness
habitats (particularly numerous in farming displayed higher values in FM adjacent to
landscapes) such as arable weeds in crop two-way roads (i.e. no weed control) than in
fields or plant species in FM are exposed to a FM adjacent to path or adjacent to one-way
frequently changing environment, i.e. in local road (i.e. possibly mown or sprayed). Insect-
management, abiotic site conditions, climate pollinated plants (in addition to other
and historical changes on arable plant parameters of richness and diversity)
communities. Such environmental benefited from this policy in 2007.
heterogeneity is in turn likely to generate Ubiquitous species were able to be
great variability in community structure in maintained in species-poor FM under drastic
space and time. conditions of weed control. For example,
Bromus hordeaceus subsp. Hordeaceus, B.
sterilis, Convolvulus arvensis, Geranium
4.2. Effects of weed control on community dissectum, Lolium perenne, and Poa annua
composition were found in the three FM characterized by a
very low vegetation cover (< 10 %) that
In our study, farmers mowed or sprayed only probably resulted from intensive and repeated
the FM that were adjacent to paths. State weed controls (herbicide sprayings and/or
agents were in charge of weed control of FM mowing).
adjacent to a road until 2006. The De Cauwer et al. (2005) showed that
management practices probably played a large early succession of newly created
role in the evolution of the plant community sown/unsown FM on ex-arable land, mown
structure and composition. State agents and twice a year (with or without removal of
farmers all preferentially mowed the FM they mown material), was characterised by the
were in charge of, but farmers also replacement of Annuals in favour of
occasionally sprayed FM when mowing did Perennials, a steady increase in the
not eradicate all weeds (De Redon pers. importance of Monocotyledons and a
comm.). Paths adjacent to FM were disturbed decrease in non-nitrogen-fixing
habitats due to intensive agricultural practices Eudicotyledons. We did not found any similar
(weed control, pesticide use, and passage of trend in this study: the increase of total
tractor). richness did not favour any particular
Our results showed that from 2005 to botanical group of plants, nor did it induce the
2007 all richness parameters increased. This replacement of any group in favour of another
increase probably mainly resulted from the group. Numbers of Eudicotyledons and
decrease in the proportion of FM that were Monocotyledons, Annuals and Perennials as
mown or sprayed from year to year. well as Insect-pollinated species, and Not
Chiverton and Sotherton (1991) showed that insect-pollinated species all increased with
biomass, density and diversity of plant species time, simultaneously with the global increase
were affected by applications of herbicides, of the total richness. Our data were thus not in
particularly non-selective herbicides such as accordance with the general observation that
glyphosate. Moreover inter-annual variations the perennial vegetation most often prevents
appeared to be mediated by weed control the establishment of annual species in the FM
practices. Indeed since 2007, 30 % of the FM and their spread into the crop (Marshall and
in our studied area were not submitted to any Moonen, 2002).
weed control according to the new application Diversity indices mathematically
of usual good agricultural practices (GDA) increased from 2005 to 2006 then decreased
validated by the French Departmental in 2007 because species frequencies in 2005
Commission for Agricultural Policy. These and 2006 were more balanced than in 2007
- 120 -
MANUSCRIPT N°6
and the total number of species observed persistence of the pollinator community
increased from 2005 to 2007. In 2007 (Kohler et al. 2007). The activity of honey-
ubiquitous species were indeed more bees (Apis mellifera L.) and bumble-bees
dominant than during the two previous years (Bombus terrestris L.) appears indeed to be
and rare species were less represented intense in the FM of the study area (Chifflet
relatively to other species (although their pers. comm.). The presence of local sources
numbers were greater compared to previous of pollen and nectar is likely to prevent long-
years). range foraging of pollinators and thereby to
According to Wilson (1992), a limit long-distance gene flow for FM insect-
potential factor for the decline of plant species pollinated species in the agro-ecosystem
in farming landscapes is the application of studied here. Species that are not insect-
chemicals. Comparisons between FM that pollinated increased also, which showed that
were submitted to different types of weed they were able to persist in the disturbed
control (none, mowing or spraying) habitats of the FM in an agro-ecosystem
confirmed this effect of weed control (Biesmeijer et al. 2006).
practices on the composition of the plant Concerning not-pollinated species (i.e.
community in FM. Indeed, most of the autogamous species and wind-pollinated
richness and diversity parameters were species), Regal (1982) showed that plant
affected, at least one year. Nevertheless these community structure of disturbed habitats is
effects varied among years and each type of characterised by higher proportions of self-
weed control had generally a significant effect pollinated plants compared to natural and
only one year but non-significant effects the semi-natural systems. Our studied area could
other two years of the study. An experimental not be considered as a disturbed habitat in the
trial with controlled parameters (number of sense of Regal (1982) since we observed only
applications of mowing and/or herbicide use, 3.2 % of strictly autogamous species.
height of mowing, concentration of
herbicides, and timing of application of weed
control) would help to understand more 4.4. Effects of local characteristics of the
precisely the effect of these agricultural field margins
practices on plant community composition.
Many studies have demonstrated the
importance of local site characteristics and
4.3. Persistence of plants according to their management practices for the occurrence of
reproductive strategies single species, community composition and
species richness (see reviews in Gabriel,
The global increase in plant species richness Thies and Tscharntke, 2005). The structure
concerned equally insect-pollinated species and heterogeneity of FM as disturbed areas
and those that are not: the numbers of both were characterized by the width of FM and
types increased while their relative the type of weed control. In Auestad et al.
proportions remained constant. The apparent (1999), FM that shelter the richest plant
persistence (and even the increase) of both communities are often wide, regularly mown,
types of species resulted from the combined adjacent to permanent grassland, and located
effects of local recruitment (via sexual in extensively managed agricultural
reproduction and/or vegetative landscapes. In our study area, the FM that
multiplication), plant survival (via specific were adjacent to a path were narrow and
vegetative forms adapted to over winter for therefore received larger amounts of herbicide
perennials) and seed survival (for seed- from the field edge than wider FM did. Given
bankers). that weeds within fields compete with the
For insect-pollinated species, their crop for space and resources, farmers regulate
persistence and their ability to flower their densities by using herbicides,
provided pollen and nectar resources for mechanical weeding and soil tillage within
pollinators that could help ensure the the field. In FM adjacent to path, individual
- 121 -
MANUSCRIPT N°6
species may decline as a result of habitat of research in 2005 and in 2006, and by ANR
degradation and interactions of species, as OGM ‘GMBIOIMPACT’ project in 2007.
well as additional, stochastic threats. Local
plant species composition was influenced by
landscape complexity (aggregation of REFERENCES
intensively managed arable fields) together
with farming system (increased use of Auestad I, Norderhaug A, & Austad I (1999)
pesticides and mineral fertilizers). Road verges-species rich habitats. A.
Appl. Biol. 54, 269–274.
Baudry J, Burel F, Thenail C, & Le Cœur D
4.5. Impacts on pollinators and gene flow in (2000) A holistic landscape ecological
agro-ecosystems study of the interactions between
farming activities and ecological
Studies on pollinators and persistence of patterns in Brittany, France. Landsc.
plants according to their reproductive systems Urban Plan. 50, 119-128.
should be more documented to understand Benton TG, Vickery JA, & Wilson JD (2003)
complex interactions between cropped and Farmland biodiversity: Is habitat
non-crop areas. It is also important to heterogeneity the key? Trends in Ecol.
understand the pollination processes that and Evol. 4, 182-188.
generate landscape-scale gene dispersal in Biesmeijer JC, Roberts SPM, Reemer M,
plants, particularly in crop plants with Ohlemuller R, Edwards M, Peeters T,
genetically modified varieties (Gallais and Schaffers AP, Potts SG, Kleukers R,
Ricroch 2006). At the landscape scale in Thomas CD, Settele J & WE Kunin
particular, gene flow among populations (2006). Parallel declines in pollinators
maintains the cohesion of a species’ gene and insect-pollinated plants in Britain
pool, thereby diminishing the potential for and the Netherlands. Science 313, 351-
local adaptation and speciation. 354.
A better understanding of the impact Braun-Blanquet J (1932) Plant Sociology (ed.
of weed control on pollinators and plants in McGraw-Hill Book Co.). New York.
openfield farming landscapes may allow an USA.
optimal design of field and margins Thus it Chiverton PA & Sotherton NW (1991) The
may allow the selection of management effects on beneficial arthropods of the
measures that preserves the biodiversity in exclusion of herbicides from cereal
FM adjacent to crops and conserve farmland crops. J. Appl. Ecol. 28, 1027–1039.
wildlife. These studies help to suggest Dailey TB & Scott PE (2006) Spring nectar
farming practices that ensure a sustainable sources for solitary bees and flies in a
protection of plant biodiversity and pollinator landscape of deciduous forest and
community in agro-ecosystems. agricultural fields: production,
variability, and consumption. J. Torrey
Bot. Soc. 133, 535-547.
ACKNOWLEDGEMENTS De Cauwer B, Reheul D, D’hoogh, K, Nijs I
& Milbau A (2005) Evolution of the
We are grateful to the farmers of the region of vegetation of mown field margins over
Selommes who allowed us to carry their first 3 years. Agr. Ecosyst. Env.
experiments in their field margins. We thank J 109, 87–96.
Baudry for advice on the experiment design. Doledec S, Chessel D, Ter Braak CJF &
We are grateful to JM Dreuillaux for species Champely S (1996) Matching species
identification. We acknowledge A Forchioni, traits to environmental variables: a
S Him-Thean, O Jonot, M de Miguel for new three-table ordination method.
assistance in the field, and S Nadot, S Pivard, Environ. Ecol. Stat. 3, 143-146.
and C Robert for data management. This Fitter R (1986) Guide des fleurs sauvages (ed.
work was supported by The French Ministry Delachaux & Niestle). Paris, France.
- 122 -
MANUSCRIPT N°6
Gabriel D, Thies C & Tscharntke T (2005) Marshall EJP, & Moonen AC (2002) Field
Local diversity of arable weeds margins in northern Europe: their
increases with landscape complexity. functions and interactions with
Persp. Plant Ecol., Evol. Systematics 7, agriculture. Agr. Ecosyst. Env. 89, 5–
85-93. 21.
Gallais A & Ricroch A (2006) Plantes Marshall EJP, Brown VK, Boatman ND,
transgéniques: faits et enjeux. Quae, Paris. Lutman PJW, Squire GR & Ward LK
Garnier A, Deville A & Lecomte J (2006) (2003) The role of weeds in
Stochastic modelling of feral plant supporting biological diversity within
populations with seed immigration and crop fields. Weed Res. 43, 77-89.
road verge management. Ecol. Oksanen J, Kindt R, Legendre P & O'Hara
Modelling 197, 373-382. RB (2007) Vegan: Community
Greaves MP & Marshall EJP (1987) Field Ecology Package Version 1.8–5 URL
margins: definitions and statistics. In: http://cran.r-project.org/.
Field Margins, Monograph No. 35 Pivard S, Adamczyk K, Lecomte J, Lavigne
(eds JM Way & PJ Greig-Smith), 3– C, Bouvier A, Deville A, Gouyon PH
10. British Crop Protection Council, & Huet S (2008) Where do the feral
Surrey, UK. oilseed rape populations come from?
Harper JL (1977) Population biology of plants A large-scale study of their possible
(ed. Academic Press). New York, USA. origin in a farmland area. J. Appl.
Hüsken A & Dietz-Pfeilstetter A Ecol. 45 (2) 476-4850
(2007) Pollen-mediated gene flow R Development Core Team (2004) R: A
from herbicide-resistant oilseed rape language and environment for
(Brassica napus L.). Transgenic Res. statistical computing. R Foundation
16, 557-569. for Statistical Computing, Vienna,
Huston MA (1999) Local processes and Austria;. URL: http://www.R-project.
regional patterns: appropriate scales org.
for understanding variation in the Regal PJ (1982) Pollination by wind and
diversity of plants and animals. Oikos animals: Ecology of geographic
86, 393-401. patterns. Ann. Rev. Ecol. Syst. 13, 497-
Kohler F, Verhulst J, Knop E, Herzog F, & 524.
Kleijn D (2007) Indirect effects of Robert P & Escoufier Y (1976) An unifying
grassland extensification schemes on tool for linear multivariate statistical
pollinators in two contrasting methods: the RV-coefficient. Appl.
European countries. Biological Stat. 25, 257–265.
Conservation 135, 302–307. Thioulouse J, Chessel D, Doleldec S &
Lambinon J, Delvosalle L & Duvigneaud J Olivier JM (1997) ADE-4: a
(2004) Nouvelle Flore de Belgique, du multivariate analysis and graphical
Grand-Duché de Luxembourg, du display software. Statistics and
Nord de la France et des régions Computing 7, 75-83.
voisines (Ptéridophytes et Turner MG (1980) Flora Europaea. Vol.1-5
Spermatophytes) 5e edition (ed. (ed. Cambridge University Press).
Patrimoine du jardin botanique Cambridge, UK.
national de Belgique). Meise,
Belgique.
- 123 -
MANUSCRIPT N°6
APPENDIX: Cerastium arvense L.
CEFT ssp. strictum (L.) Caryophyllaceae E
LIST OF THE 185 PLANT TAXA Ugborogho
RECORDED IN THE 27 FIELD MARGINS Cerastium
CESE Caryophyllaceae E
SURVEYED FROM 2005 TO 2007. semidecandrum L.
CERS Cerastium spp. Caryophyllaceae E
Taxa code E. vs. Chaerophyllum
Genus and species Family CHTE Apiaceae E
number M temulum L.
Achillea millefolium Chenopodium album
ACMM Asteraceae E CHAA Chenopodiaceae E
L. L.
Agrimonia eupatoria Cirsium arvense (L.)
AGEE Rosaceae E CIAR Asteraceae E
L. Scop.
AGCA Agrostis capillaris L.1 Poaceae M Cirsium vulgare (Savi)
COAR Asteraceae E
Ten. 7
AGSP Agrostis spp. Poaceae M Convolvulus arvensis
Ajuga reptans L. a CORC Convolvulaceae E
AJRE Lamiaceae E L. d, e, f
ALVI Allium vineale L. Alliaceae M Conyza canadensis
COCA Asteraceae E
Alopecurus (L.) Cronquist 8
ALMY Poaceae M Coronopus squamatus
myosuroides Huds.2 COSP Brassicaceae E
ALSP Alopecurus spp. Poaceae M (Forsk.) Ascherson c
Crataegus monogyna
Amaranthus CRMO Rosaceae E
AMRE Amaranthaceae E Jacq. b
retroflexus L. b
Crepis capillaris (L.)
AMSP Amaranthus spp. c Amaranthaceae E CRCA
Wallr. 9
Asteraceae E
ANAR Anagallis arvensis L. Myrsinaceae E CRSE Crepis setosa Haller f. Asteraceae E
ANSP Anthriscus spp. Apiaceae E CRSP Crepis spp. Asteraceae E
Anthriscus sylvestris Cruciata laevipes
ANSX Apiaceae E CRLA Rubiaceae E
(L.) Hoffmann3 Opiz 10, b
ATSP Anthyllis spp. b, c Fabaceae E DAGG Dactylis glomerata L. Poaceae M
ARLA Arctium lappa L. 4 Asteraceae E DACA Daucus carota L. Apiaceae E
Arrhenatherum elatius Elytrigia repens (L.)
AREL Poaceae M ELRE Poaceae M
(L.) P. Beauv. Desv. ex Nevski 11, f
ARVU Artemisia vulgaris L. Asteraceae E ELSP Elytrigia spp. Poaceae M
ATSP Atriplex spp. Chenopodiaceae E EPSP Epilobium spp. Onagraceae E
AVSA Avena sativa L. Poaceae M Erodium cicutarium
ERCI Geraniaceae E
BEPX Bellis perennis L. Asteraceae E (L.) L’Her.
Brachypodium Eryngium campestre
ERVE Apiaceae E
BRPI pinnatum (L.) P. Poaceae M L.
Beauv. b Euphorbia cyparissias
EUCY Euphobiaceae E
Brachypodium L. c
BASY sylvaticum (Huds.) P. Poaceae M Euphorbia helioscopia
EUHE Euphobiaceae E
Beauv. a L.
BRNO Brassica napus L. Brassicaceae E Fallopia convolvulus
FACO Polygonaceae E
(L.) A. Love 12
Brassica nigra (L.) W.
BRNI Brassicaceae E Festuca arundinacea
D. J. Koch FEAR Poaceae M
Schreb.
BRSP Brassica spp. b Brassicaceae E
FEOV Festuca ovina L. c Poaceae M
BRAV Bromus arvensis L. Poaceae M
Festuca pratensis
Bromus erectus Huds. FEPP Poaceae M
BRER b Poaceae M Huds.
FERR Festuca rubra L. Poaceae M
Bromus hordeaceus L.
BRHO Poaceae M FESP Festuca spp. Poaceae M
subsp. hordeaceus 5, d
Bromus rigidus Roth 6,
GAAP Galium aparine L. e, f Rubiaceae E
BRRI c Poaceae M
GAMO Galium mollugo L. Rubiaceae E
BRST Bromus sterilis L. d, e, f Poaceae M GASP Galium spp. Rubiaceae E
CASP Calendula spp. b Asteraceae E GAVE Galium verum L. Rubiaceae E
Capsella bursa- Geranium
CABU Brassicaceae E GECO Geraniaceae E
pastoris (L.) Medik. columbinum L.
CANU Carduus spp. c Asteraceae E GEDI Geranium dissectum L. d, e, f Geraniaceae E
Carex echinata GEMO Geranium molle L. Geraniaceae E
CAEC Cyperaceae M
Murray c
GEPU Geranium pusillum L. Geraniaceae E
CAML Carex muricata L. Cyperaceae M
Geranium pyrenaicum
CEJA Centaurea jacea L. Asteraceae E GEPY Geraniaceae E
Burm. F.
CENI Centaurea nigra L. Asteraceae E Geranium
GERT Geraniaceae E
CESP Centaurea spp. Asteraceae E rotundifolium L.
- 124 -
MANUSCRIPT N°6
GESP Geranium spp. Geraniaceae E Plantago lanceolata
PLLA Plantaginaceae E
Glechoma hederacea L. d, e
GLHE Lamiaceae E PLMM Plantago major L. Plantaginaceae E
L.
HEHH Hedera helix L. a Araliaceae E POAN Poa annua L. d, e, f Poaceae M
HEAN Helianthus annuus L. Asteraceae E POPR Poa pratensis L. d, e, f Poaceae M
HYSP
Heracleum
Apiaceae E POSP Poa spp. b Poaceae M
sphondylium L. POTR Poa trivialis L. Poaceae M
Hordeum murinum L.
HOMU c Poaceae M Polygonum aviculare
POAV Polygonaceae E
L.
Hordeum vulgare L.
HOVU Poaceae M Portulaca oleracea L.
subsp. vulgare 13, a POoL a Portulacaceae E
Hypericum perforatum
HYPE Hypericaceae E PORE Potentilla reptans L. Rosaceae E
L.
HYSP Hypericum spp. a Hypericaceae E POSP Potentilla spp. Rosaceae E
Hypochaeris radicata PRSP Prunus spinosa L. Rosaceae E
HYRA Asteraceae E
L. c RAAA Ranunculus acris L. Ranunculaceae E
Juncus conglomeratus Ranunculus ficaria L.
JUCO Juncaceae M RAFI Ranunculaceae E
L. a 21, b
Kickxia spuria (L.) RARE Ranunculus repens L. Ranunculaceae E

KISS Plantaginaceae E
Dumort. 14, c
Ranunculus sardous
Knautia arvensis (L.) RASA Ranunculaceae E
KNAR Dipsacaceae E Crantz
Coult.
RASP Ranunculus spp. Ranunculaceae E
LASE Lactuca serriola L. 15 Asteraceae E
Raphanus
Lamium amplexicaule RARA Brassicaceae E
LAAM Lamiaceae E raphanistrum L.
L. a
Lamium hybridum RELU Reseda lutea L. c Resedaceae E
LAHY Lamiaceae E ROSP Rosa spp. Rosaceae E
Vill.
LAPU Lamium purpureum L. Lamiaceae E RUFR Rubus fruticosus L. Rosaceae E
Lapsana communis L. RUAC Rumex acetosa L. Polygonaceae E
LACO c Asteraceae E
RUOO Rumex acetosella L. c Polygonaceae E
LAPR Lathyrus pratensis L. Fabaceae E
RUCR Rumex crispus L. Polygonaceae E
LASP Lathyrus spp. Fabaceae E
RUSA Rumex sanguineus L. a Polygonaceae E
Leucanthemum
LEVU Asteraceae E RUSP Rumex spp. Polygonaceae E
vulgare Lam. 16, c
LIVU Ligustrum vulgare L. b Oleaceae E SASP Sagina spp. a Caryophyllaceae E
Lithospermum arvense SAPA Salvia pratensis L. Lamiaceae E
LIAR Boraginaceae E
L. 17 SEJA Senecio jacobaea L. Asteraceae E
LOPE Lolium perenne L. d, e Poaceae M SEVU Senecio vulgaris L. Asteraceae E
LOCO Lotus corniculatus L. Fabaceae E SEMO Seseli montanum L. c Apiaceae E
Malva neglecta Wallr. SHAR Sherardia arvensis L. Rubiaceae E
MANE 18 Malvaceae E
Silene dioica (L.)
Matricaria SIDI Caryophyllaceae E
MACH Asteraceae E Clauv. 22
chamomilla L.
Silene vulgaris
Matricaria discoidea SIVU Caryophyllaceae E
MADI Asteraceae E (Moench.) Garcke 23
DC
Matricaria perforata SIAR Sinapis arvensis L. a Brassicaceae E
MAPE Asteraceae E Sisymbrium officinale
Mérat 19 SYOF Brassicaceae E
MASP Matricaria spp. Asteraceae E (L.) Scop.
SONI Solanum nigrum L. c Solanaceae E
MESP Medicago spp. c Fabaceae E
Sonchus asper (L.)
MEAN Mercurialis annua L. Euphobiaceae E SOAS Asteraceae E
Hill
MYSP Mysosotis spp. Boraginaceae E SOOL Sonchus oleraceus L. Asteraceae E
ONRE Ononis repens L. Fabaceae E STGR Stellaria graminea L. Caryophyllaceae E
PARH Papaver rhoeas L. Papaveraceae E Stellaria media (L.)
STMM Caryophyllaceae E
PASP Papaver spp. Papaveraceae E Vill. a
PASA Pastinaca sativa L. b Apiaceae E Taraxacum officinale
TAOF Asteraceae E
aggr. 24
Persicaria maculosa
PEMA Polygonaceae E Torilis anthriscus (L.)
Gray 20, b TOAN Apiaceae E
C. C. Gmel. 25
PIHI Picris hieracioides L. Asteraceae E Tragopogon dubius
PISP Picris spp. Asteraceae E TRDU Asteraceae E
Scop.
PISA Pisum sativum L. c Fabaceae E TRAR Trifolium arvense L. Fabaceae E
Plantago coronopus Trifolium campestre
PLCO Plantaginaceae E TRCA Fabaceae E
L. a Schreb.
- 125 -
MANUSCRIPT N°6
TRDU
Trifolium dubium
Fabaceae E NB 1:
Sibth. 26
Trifolium incarnatum
E = Eudicotyledon & M = Monocotyledon
TRIN Fabaceae E
L.
TRRE Trifolium repens L. Fabaceae E NB 2:
TRFL
Trisetum flavescens
Poaceae M The taxa code number follows Flora Europaea
(L.) P. Beauv. c
(Turin et al. 1980). a rare species observed in
TRAE Triticum aestivum L. Poaceae M
2005, b in 2006, and c in 2007 ; d ubiquitous
ULMI Ulmus minor Mill. 27 Ulmaceae E
species observed in 2005, e in 2006 and f in
URDI Urtica dioica L. Urticaceae E
Valerianella locusta
2007.
VALO Valerianaceae E
(L.) Laterr. 28
VBOF Verbena officinalis L. Verbenaceae E NB 3:
1
VEAR Veronica arvensis L. Plantaginaceae E syn. A. tenuis Sibth.; 2 syn. A. agrestis L.; 3 syn.
VECH
Veronica chamaedrys
Plantaginaceae E Chaerophylum silvestre L.; 4 syn. Lappa major
L.
Veronica hederaefolia L.; 5 syn. B. mollis L.; 6 syn. Bromus maximus
VEHH
L.
Plantaginaceae E Desf.; 7 syn. Cirsium lanceolatum (L.) Scop.; 8
VEOF Veronica officinalis L. Plantaginaceae E syn. Erigeron canadensis L.; 9 syn. C. virens L.;
10
VEPE Veronica persica Poir. Plantaginaceae E syn. Galium cruciata Scop.; 11 syn. Agropyron
VICR Vicia cracca L. c Fabaceae E repens (L.) P. Beauv.; 12 syn. Polygonum
VISA Vicia sativa L. Fabaceae E convolvulus L.; 13 syn. Hordeum hexastichon L.;
14
VISP Vicia spp. 29 Fabaceae E syn. Linaria spuria L.; 15 syn. Lactuca
VIAR Viola arvensis Murray Violaceae E 16
scariola L.; syn. Chrysanthemum
Vulpia myuros (L.) C. 17
VUMY
C. Gmel.
Poaceae M leucanthemum L.; syn. Buglossoides arvensis
L.; 18 syn. Malva rotundifolia auct. non L.; 19
syn. Matricaria inodora L.; 20 syn. Polygonum
persicaria L.; 21 syn. Ficaria ranonculoides
Roth; 22 syn. Lychnis dioica; 23 syn. S. inflata
Sm. var. prostrata Gaudin; 24 syn. T. dens-
leonis Desf.; 25 syn. T. japonica (Houtt.) DC. or
Anthriscus caucalis M. Bieb.; 26 syn. T. minus
Sm.; 27 syn. Ulmus campestris auct. ; 28 syn. V.
olitoria (L.) Pollich; 29 syn. Ervum spp.
Microtus agrestis
- 126 -
MANUSCRIPT N°7
Published in Biological Conservation

(June 2008, Vol. 141, p. 1581-1590)
Plant and spider communities benefit differently

from the presence of planted hedgerows in highway
Isabelle LE VIOL
Romain JULLIARD
Christian KERBIRIOU
Louis de REDON
Nathalie CARNINO
Nathalie MACHON
&
Emmanuelle PORCHER
127
- 127 -
MANUSCRIPT N°7
PRESENTATION
TITRE
Communautés de plantes et communautés d’araignées bénéficient différemment de la
présence des plantations de haies au sein des dépendances vertes des bords d’autoroutes.
RESUME
Les bords de routes peuvent jouer un rôle crucial comme refuge pour la flore et la faune
native au sein des paysages anthropisés. Il apparaît que l’influence des choix de construction,
comme les plantations d’espèces forestières, sur la biodiversité des bords de route a peu été
étudiée alors même que la présence de haies, par exemple, pourrait jouer un rôle d’habitat
pour les espèces forestières. A l’aide d’une méthodologie standardisée, nous avons étudié
l’impact des plantations de haies sur deux groupes taxonomiques (les plantes et les araignées)
établis en bords de route au sein des paysages d’agriculture intensive. Nous avons examiné la
richesse spécifique et fonctionnelle des communautés au sein de sites avec et sans haies. Au
niveau des sites, la réponse des communautés de plantes et d’araignées ont répondu très
différemment : les haies été associées à une augmentation significative de la richesse en
plantes (plus grande diversité α) et à une même richesse en araignées. Les communautés de
plantes des sites sans haies sont apparues comme partie des communautés de plantes des sites
avec haies ; alors que pour les communautés d’araignées, celles des sites sans haies étaient
complémentaire de celles des sites avec haies (plus grande diversité β). La présence de haies
étaient aussi associée à une augmentation de la diversité taxonomique et des traits
fonctionnels au niveau du paysage (diversité γ) grâce à une augmentation de la diversité β
pour les plantes comme pour les araignées. Nos résultats suggèrent qu’un mélange d’habitats
avec haies et d’habitats ouverts est crucial pour le maintien de la biodiversité à l’échelle des
paysages. En procurant des informations pour les professionnels de la route et aux décideurs
politiques quant à leur influence potentielle sur la biodiversité, ces résultats ont des
implications directes pour la gestion des réseaux routiers et de leurs dépendances vertes.
MOTS CLEFS
Diversité α, diversité β, biodiversité, traits fonctionnels, gestion des bords de route,
revégétalisation & analyses RLQ.
128
- 128 -
MANUSCRIPT N°7
129
- 129 -
MANUSCRIPT N°7
130
- 130 -
MANUSCRIPT N°7
131
- 131 -
MANUSCRIPT N°7
132
- 132 -
MANUSCRIPT N°7
133
- 133 -
MANUSCRIPT N°7
134
- 134 -
MANUSCRIPT N°7
135
- 135 -
MANUSCRIPT N°7
136
- 136 -
MANUSCRIPT N°7
137
- 137 -
MANUSCRIPT N°7
138
- 138 -
ENCADRES
« Délaisse les grandes routes, prends les

sentiers ».
PYTHAGORE (-569/-494)
- 139 -
E N CA D R E N°1
ETENDUE DE L’IMPACT ECOLOGIQUE DES ROUTES
Les effets des routes ne se limitent pas à leurs emprises (destruction d’habitat et mise en
place de bordures vertes), 22% de la surface totale des U.S.A. est ainsi affectée
directement par le réseau routier, i.e. 1,73 millions de km² (Forman 2000).
Les effets des routes sont asymétriques et dépendent des milieux traversés (cf.
Figure N°6, zone orangée), du type de route considérée (autoroute, route principale,
route secondaire, etc.) et du trafic (Forman 2000, Forman et Deblinger 2000).
Figure N°6
Zones d’impact de la route (10 km Massachusetts, U.S.A.)
- 140 -
(a), (b), (i), (j), (m), (p) & (z) zones construites ; (c) couloir de vent ; (d) zone basse ;
(e), (g), (l), (n), (r) &(t) forêts ; (f), (h), (o) & (q) talus routiers ; (k) zones de collision ;
(s) & (y) collines ; (u) prairie ; (v) rivière ; (w) zone humide et (x) golf (Forman 2000).
E N CA D R E N °2
COMDYN LOGICIEL DE DYNAMIQUE DES COMMUNAUTES
L’estimation de paramètres, tels que la richesse spécifique d’une communauté ou la

similarité entre deux communautés, peut poser problème dans la mesure où ils sont issus
d’observations souvent incomplètes ou réalisées par différents observateurs ayant des
aptitudes différentes (connaissances, concentration, etc.). Il convient donc, non pas
d’éliminer ces aléas qui sont partie intégrante de tout travaux de terrain, mais de quantifier
les erreurs pour les corriger. L’utilisation de réplicats, spatiaux ou temporels, permet
l’estimation de probabilités de détection permettant la correction des données brutes.
Ainsi deux communautés composées, en partie, d’espèces différentes dont les probabilités
de détection sont faibles seront considérées comme relativement plus similaires entre elles
que deux autres communautés présentant les mêmes particularités mais dont les espèces
auront des probabilités de détection élevées.
Nous avons ainsi travaillé avec des protocoles intégrant nécessairement la mise en place
systématique de réplicats spatiaux (cinq) et avec un logiciel de capture-recapture ComDyn
(Hines et al. 1999) permettant l’estimation d’un certains nombre de paramètres et de leurs
erreurs associées : richesses estimées, probabilités de détection, similarités estimées, etc.
Exemple :
Trois communautés sont échantillonnées à partir de cinq réplicats et, bien qu’en termes de
richesse brute, ces communautés peuvent apparaître comme similaires (sept espèces
détectées pour chacune d’elles), elles apparaissent sont néanmoins très différentes.
- 141 -
(1) Dans la première communauté, aucune espèce n’est détectée plus de deux fois
seulement et de nouvelles espèces sont détectées à chacun des réplicats, la richesse des
réplicats est assez faible : Ne = 2 ou 3 ;
(2) Dans la deuxième communauté, certaines espèces sont détectées quasiment dans chaque
réplicat alors que d’autres ne sont détectées qu’une seule fois et progressivement au
cours des réplicats, la richesse des réplicats est moyenne : Ne = 3 à 6 ;
(3) Dans la troisième communauté, toutes les espèces sont détectées au moins quatre fois et
toute la richesse est échantillonnée avec uniquement les deux premiers relevés, la
richesse des quadrats est importante : Ne = 5 à 7.
- 142 -
Il apparaît donc évident que ces différentes communautés, ultra-simplifiées et

schématiques, ne sont pas équivalentes. Si l’ensemble des espèces composant la
communauté n°3 semblent avoir été détecté, cela ne semble pas être le cas pour la
communauté n°2 et encore moins pour la communauté n°1.
Table N°3
Résultats obtenus avec ComDyn
NeE-se Pdetec-se
Communauté NeO NeE NeE-se Pdetec Pdetec-se
/ NeE / Pdetec
n°1 7 8.6 1.24 14.4% 0.81 0.10 12.7%
n°2 7 7.8 1.37 17.6% 0.90 0.11 11.7%
n°3 7 7.0 0.00 0.0% 1.00 0.00 0.0%
Clef : NeO : Nombre d’espèce Observé – NeE : Nombre d’espèce Estimé – Pdetec :
Probibilté de Détection – se : Standard Error (erreur standard).
Les résultats obtenus avec ComDyn confirment ce sentiment avec des richesses corrigées
(NeE3<NeE2<NeE1) et des probabilités de détections finalement différentes d’une
communauté à l’autre (cf. Table N°3). Les rapports Valeurs estimées/Erreurs standards
sont acceptables (toujours inférieurs à 20%).
- 143 -
E N CA D R E N °3
STRATEGIE NATIONALE POUR LA BIODIVERSITE
De part sa signature, au Sommet de la Terre à Rio de Janeiro (Brésil, 1992), du traité

fondateur des politiques de conservation de la biodiversité, la Convention sur la
Diversité Biologique (http://www.cbd.int/convention/convention.shtml), et de part
l’engagement de l’Union européenne qui s’en est suivi au Sommet de Göteborg (Suède,
2001), la France a pris l’engagement d’arrêter l’érosion de la biodiversité sur son
territoire avant 2010 à travers le « Strategic Plan for the Convention on Biological
Diversity » (http://www.cbd.int/decisions/?dec=VI/26), plus connu du grand public
sous l’appellation « Countdown 2010 ».
Afin de d’honorer ses engagements internationaux, la

France s’est dotée en 2005 d’une Stratégie Nationale
pour la Biodiversité (SNB, http://www.ecologie.
gouv.fr/-Strategie-nationale-pour-la-.html).
La recherche d’indicateurs et la mise en place de suivis

sont des éléments essentiels à la mise en place des plans
d’action de la stratégie française.
Le financement de travaux de recherche sur le sujet, et en

particulier de thèses, sont aussi lancés.
- 144 -
BIBLIOGRAPHIE
« Si vous possédez une bibliothèque et un

jardin, vous avez tout ce qu'il vous faut ».
CICERON (-106/-43)
- 145 -
Abadie JC (2008) La nature ordinaire face aux pressions humaines : le cas des plantes
communes : Méthodes de suivis et évaluation de l’impact. Thèse de doctorat, MNHN.
Adams LW, Geis AD (1983) Effects of roads on small mammals. Journal of Applied
Ecology 20, 403–415.
Adams LW (1984) Small mammal use of an interstate highway median strip. Journal of
Applied Ecology 21, 175–178.
Akbar KF (1997) Aspects of the ecology and conservation value of roadside vegetation in
northern England. Ph.D. Thesis, University of Bradford, U.K.
Akbar KF, Ahmad Z, Shad MA, Ansari TM (2003) An ecological study of roadside
vegetation and soils in Sahiwal district. Journal of Biological Sciences 3, 627-634.
Allan JD, Flecker AS (1993) Biodiversity conservation in running waters. BioScience 43,
32-43.
Alstad DN, Edmunds GF (1982) Effects of air polluants on insect populations. Annual
review of Entomology 27, 369-384.
Altieri MA (1999) The ecological role of biodiversity in agroecosystems. Agriculture,
Ecosystems and Environment 74, 19–31.
Anderson MJ, Willis TJ (2003) Canonical analysis of principal coordinates: A useful
method of constrained ordination for ecology. Ecology 84, 511–525.
Andow DA (1991) Vegetational diversity and arthropod population response. Annual Review
of Entomology 36, 561–586.
Andren H (1994) Effects of habitat fragmentation on birds and mammals in landscapes with
different proportions of suitable habitat: A review. Oikos 71, 355-366.
Angold PG (1997) The impact of road upon adjacent heathland vegetation: effects on plant
species composition. Journal of Applied Ecology 34, 409–417.
Asaeda T, Ca VT (1993) The subsurface transport of heat and moisture and its effect on the
environment: A numerical model. Boundary-Layer Meteorology 65, 159-179.
Ashenden TW, Ashmore M, Bell JNB, Bignal K, Binnie J, Cape JN, Caporn SJM,
Carroll J, Davison A, Hadfield P, Honour S, Lawton K, Moore S, Power S,
Shields C (2003) Impacts of vehicle emissions on vegetation. In Sucharov LJ and
Brebbia CA (Eds), 313–322. Urban transport IX: urban transport and the environment
in the 21st century, Advances in transport 14. WIT Press, Southampton.
- 146 -
Atlegrim O, Sjoeberg K, Ball JP (1997). Forestry effects on a boreal ground beetle

community in spring: Selective logging and clear-cutting compared. Entomologica
Fennica 8, 19-26.
Aubin I, Gachet S, Messier C, Bouchard A (2007) How resilient are the northern hardwood
forests to human disturbance? An evaluation using a plant functional group approach.
Ecoscience 14, 259-271.
Auerbach NA, Walker MD, Walker DA (1997) Effects of road dust on substrate and
vegetation properties in arctic tundra. Ecological Applications 7, 218–235.
Auestad I, Norderhaug A., Austad I (1999) Road verges-species rich habitats. Aspects of
Applied Biology 54, 269–274.
Baars MA, van Dijk TS (1984) Population dynamics of two carabid beetles at a Dutch
heathland. I. Subpopulation fluctuations in relation to weather and dispersal. Journal
of Animal Ecology 53, 375-388.
Bailey S (2007) Increasing connectivity in fragmented landscapes: An investigation of
evidence for biodiversity gain in woodlands. Forest Ecology and Management 238: 7-
23.
Bakowski C, Kosakiewicz M (1988) Effects of a forest road on Bank Vole and Yellow-neck
populations. Global Ecology and Biogeography 72, 245–252.
Baudry J, Bunce RGH, Burel F (2000) Hedgerows: an international perspective on their
origin, function and management. Journal of Environmental Management 60, 7–22.
Baudry J, Burel F, Thenail C, Le Coeur D (2000) A holistic landscape ecological study of
the interactions between farming activities and ecological patterns in Brittany, France.
Landscape Urban Planning 50, 119-128.
Baur A, Baur B (1990) Are roads barriers to dispersal in the land snail Arianta arbustorum?
Canadian Journal of Zoology 68, 613–617.
Beals M (2006) Understanding community structure: A data-driven multivariate approach.
Oecologia 150, 484–495.
Bedford SE, Usher MB (1994) Distribution of arthropod species across the margins of farm
woodlands. Agriculture Ecosystems and Environment 48, 295–305.
Beier P, Noss RF (1998) Do habitat corridors provide connectivity? Conservation Biology
12, 1241-1252.
- 147 -
Bell JR, Bohan DA, Shaw EM, Weyman GS (2005) Ballooning dispersal using silk: world
fauna, phylogenies, genetics and models. Bulletin of Entomological Research 95, 69–
114.
Bellamy PE, Shore RF, Ardeshir D, Treeek JR, Sparks TH (2000) Road verges as habitats
for small mammals in Britain. Mammals Revue 30, 131-139.
Bellocq I, Smith SM (1994) Arthropods preferred as food by Sorex cinereus (masked shrew)
and Peromyscus mariculatus (deer mouse): An experimental approach. Mammalia 56,
391-396.
Benfenati E, Valzacchi S, Mariani G, Airoldi L, Fanelli R (1992) PCDD, PCDF, PCB,
PAH, cadmium and lead in roadside soil: Relationship between road distance and
concentration. Chemosphere 24, 1077-1083.
Benjamini Y, Hochberg Y (1995) Controlling the false discovery rate: a practical and
powerful approach to multiple testing. Journal of the Royal Statistical Society (Series
B) 57, 289–300.
Bennett AF (1990) Habitat corridors and the conservation of small mammals in a fragmented
landscape forest environement. Landscape Ecology 4, 109-122.
Bennett AF (1991) Roads, roadsides and wildlife conservation: a review. In: Saunders DA,
Hobbs RJ (Eds), 9-117 Nature conservation 2: the role of corridors. Surrey Beatty,
Chipping Norton, New South Wales, Australia.
Benton TG, Vickery JA, Wilson JD (2003) Farmland biodiversity: Is habitat heterogeneity
the key? Trends in Ecology and Evolotion 4, 182-188.
Berendse F, Oopmes MJM, Altena HJ, De Wisser W (1994) A comparative study of
nitrogen flows in two similar meadows affected by different groundwater levels.
Journal of Applied Ecology 31, 40-48.
Beslaneev VD, Kuchmazokova FA (1991) The effect of motorways on the accumulation of
toxic substances in walnuts. Sadovodstvo i Vinogradarstvo 5, 38.
Bhattacharya M, Primack RB, Gerwein J (2003) Are roads and railroads barriers to
bumblebee movement in a temperate suburban conservation area. Biological
Conservation 109, 37–45.
Bianchi FJJA, Booij CJH, Tscharntke T (2006) Sustainable pest regulation in agricultural
landscapes: A review on landscape composition, biodiversity and natural pest control,
Proceedings of the Royal Society B 273, 1715–1727.
- 148 -
Biesmeijer JC, Robert SPM, Reemer M, Ohlemuller R, Edwards M, Peeters T,

Schaffers AP, Potts SG, Kleukers R, Thomas CD, Settele J, Kunin WE (2006)
Parallel declines in pollinators and insect-pollinated plants in Britain and the
Netherlands. Science 313, 351-354.
Birdsall CW, Grue CE, Anderson A (1986) Lead concentrations in bullfrog Rana
catesbeiana and green frog Rana clamitans tadpoles inhabiting highway drainages.
Environmental Pollution (Series A) 40, 233 248.
Blandin P (1986) Bioindicateurs et diagnostic des systèmes écologiques. Bulletin d’Ecologie
17, 215-307.
Bochet E, Garcia-Fayos P, Tormo J (2007) Road slope revegetation in semiarid
mediterranean environments: Part I. Seed dispersal and spontaneous colonization,
Restoration Ecology 15, 88–96.
Bogemans J, Neirinckx L, Stassart JM (1989) Effect of deicing chloride salts on ion
accumulation in spruce (Picea abies (L.) sp.). Plant and Soil 113, 3-11.
Bonte D, Vandenbroecke N, Lens L, Maelfait JP (2003) Low propensity for aerial
dispersal in specialist spiders from fragmented landscapes, Proceedings of the Royal
Society of London (Series B) 270, 1601–1607.
Boulinier T, James, D, Nichols, John R, Sauer JR, Hines JE, Pollock KH (1998)
Estimating species richness: The importance of heterogeneity in species detectability.
Ecology 79, 1018–1028.
Boulinier T, Nichols JD, Hines JE, Sauer JR, Flather CH, Pollock KH (1998) Higher
temporal variability of forest breeding bird communities in fragmented landscapes,
Proceedings of the National Academy of Sciences of the United States of America 95,
7497–7501.
Boulinier T, Nichols JD, Hines JE, Sauer JR, Flather CH, Pollock KH (2001) Forest
fragmentation and bird community dynamics: inference at regional scales. Ecology
82, 1159–1169.
Bournérias M, Arnal G, Bock C (2001) Guide des groupements végétaux. Belin.
Boutin C, Jobin B, Belanger L, Choiniere L (2002) Plant diversity in three types of
hedgerows adjacent to cropfields. Biodiversity and Conservation 11, 1–25.
Braak (ter) CJF (1986) Canonical correspondence-analysis – a new eigenvector technique
for multivariate direct gradient analysis. Ecology 67, 1167–1179.
- 149 -
Braun-Blanquet J (1932) Plant sociology : The study of plant communities. Graw-Hill Book
Company. New York.
Brehme CS (2003) Response of small terrestrial vertebrates to roads in a coastal sage scrub
ecosystem. PhD thesis. Faculty of San Diego State University.
Briner T, Favre N, Nentwig W, Airoldi JP (2007) Population dynamics of Microtus arvalis
in a weed strip. Mammalian Biology 72, 106-115.
Brody AJ, Pelton MR (1989) Effects of roads on black bear movements in western North
Carolina. Wildlife Society Bulletin 17, 5–10.
Brothers TS, Spingarn A (2003) Forest fragmentation and alien plant invasion of central
Indiana old-growth forests. Conservation Biology 6, 91-100.
Brown GP, Phillips BL, Webb JK, Shine R (2006) Toad on the road: use of roads as
dispersal corridors by cane toads (Bufo marinus) at an invasion front in tropical
Australia. Biological Conservation 133, 88–94.
Brumm H (2004) The impact of environmental noise on song amplitude in a territorial bird.
Journal of Animal Ecology 73, 434-440.
Buckner CH (1969) Some aspects of the population ecology of the common shrew, Sorex
araneus, near Oxford, England. Journal of Mammalogy 50: 326-332.
Bullock JM, Franklin J, Stevenson MJ, Silvertown J, Coulson SJ, Gregory SJ, Tofts R
(2001) A plant trait analysis of responses to grazing in a long term experiment.
Burel F (1996) Hedgerows and their role in agricultural landscapes. Critical Reviews in Plant
Sciences 15, 169–190.
Burel F, Baudry J, Butet A, Clergeau P, Le Cœur D, Dubs F, Morvan N, Delettre Y,
Paillat G, Petit S, Thenail C, Brune E, Lefeuvre JC (1998) Comparative
biodiversity along a gradient of agricultural landscapes. Acta Oecologica 19, 47-60.
Burkey TV (1989) Extinction in nature reserves: the effect of fragmentation and the
importance of migration between reserve fragments. Oikos 55, 75-81.
Burnett SE (1992) Effects of a rainforest road on movements of small mammals:
mechanisms and implications. Wildlife Research19, 95–104.
Burnham KP, Overton WS (1979) Robust estimation of population-size when capture
probabilities vary among animals. Ecology 60, 927–936.
- 150 -
Byrd DS, Gilmore JT, Lea RH (1983) Effect of decreased use of lead in gasoline on the soil
of a highway. Environmental Science and Technology 17, 121-123.
Canaday C (1996) Loss of insectivorous birds along a gradient of human impact in
Amazonia. Biological Conservation 77, 63–71.
Canterbury GE, Blockstein DE (1997) Local changes in a breeding bird community
following forest disturbance. Journal of Field Ornithology 68, 537-546.
Cantoni D (2002) Social and spacial organization of free-ranging shrews, Sorex coronatus
and Neomys fodiens (Insectivora, Mammalian). Animal behaviour 45, 975-995.
Carslaw DC (2005) Evidence of an increasing NO2/NOX emissions ratio from road traffic
emissions. Atmospheric Environment 39, 4793–4802.
CETE de l’Ouest (1986) Bords de route : Quelques exemples de leur richesse en
Arthropodes. Division Interurbaine et Environnement.
Chakrabarty D, Subhas CS, Mukherjee A, Roy B, Das (1997) Status of road traffic noise
in Calcutta metropolis, India. The Journal of the Acoustical Society of America 101,
943-949.
Chaline J, Baudvin H, Jammot, D, Saint-Girons MC (1974) Les proies des rapaces. Doin
(Eds). Paris, France.
Chapin FS, Zavaleta ES, Eviner VT, Naylor RL, Vitousek PM, Reynolds HL, Hooper
DU, Lavorel S, Sala OE, Hobbie SE, Mack MC, Díaz S (2000) Consequences of
changing biodiversity. Nature 405, 234-242.
Chiverton PA, Sotherton (1991) The effects on beneficial arthropods of the exclusion of
herbicides from cereal crops. Journal of Applied Ecology 28, 1027–1039.
Christen D, Matlack G (2006) The role of roadsides in plant Invasions: A demographic
approach. Conservation Biology 20, 385-391.
Churchfield F (1982) Food availability and the diet of the common shrew Sorex araneus in
Britain. Journal of Animal Ecology 51, 15-28.
Churchfield S, Brown VK (1987) The trophic impact of small mammals in successional
grasslands. Biological Journal of the Linnean Society 31, 273-290.
Churchfield S, Hollier J, Brown VK (1991) The effects of small mammal predators on
grassland invertebrates, investigated by field exclosure experiment. Oikos 60, 83-290.
Coffin AW (2007) From roadkill to road ecology: A review of the ecological effects of roads.
Journal of Transport Geography 15, 396-406.
- 151 -
Collins JA (1984) Roadside lead in New Zealand and its significance for human and animal
health. New-Zealand Journal of Science 27, 93-98.
Collins HM (2002) The Third Wave of science studies: Studies of expertise and experience.
Social Studies of Science 32, 235-296.
Conseil Général de Seine-et-Marne (2003) Classeur de l’entretien de la route et de
l’exploitation en Seine-et-Marne. Service Routes et Transports, Les outils.
Contan E (2007) Etude des invertébrés des dépendances vertes des bords de route. Rapport
de stage de M1 Environnement, Université des Sciences et Technologies de Lille.
UMR 51/73 MNHN/CNRS/UPMC.
Corbit M, Marks PL, Gardescu S (1999) Hedgerows as habitat corridors for forest herbs in
central New-York, USA. Journal of Ecology 87, 220-232.
Costanza R, d’Arge R, de Groot R, Farber S, Grasso M, Hannon B (1997) The value of
the world’s ecosystem services and natural capital. Nature 387, 253–260.
Couvet D, Jiguet F, Julliard R, Levrel H (2005) Les indicateurs de Biodiversité. In
Barbault R et Chevassus-au-Louis B 2005. Biodiversité et changements globaux.
Enjeux de société et défis pour la recherche, 40-46. ADPF-Ministère des Affaires
Etrangères.
Couvet D, Dehorter O, Henry PY, Jiguet F, Julliard R (2006). La biodiversité contre les
maladies infectieuses ? Annales des Mines. Responsabilité and Environnement 41, 77-
83.
Cramer GH, Hopkins WC (1982) Effects of a dredged highway construction on water
quality in a Louisiana wetland. Transportation Research Record 896, 47–51.
Croxton PJ, Franssen W, Myhill DG, Sparks TH (2004) The restoration of neglected
hedges: a comparison of management treatments, Biological Conservation 117, 19–
23.
Czech B (1991) Elk behavior in response to human disturbance at Mount St. Helens National
Volcanic Monument. Applied Animal Behaviour 29, 269-277.
Dailey TB, Scott PE (2006) Spring nectar sources for solitary bees and flies in a landscape of
deciduous forest and agricultural fields: production, variability, and consumption.
Journal of the. Torrey Botanical Society 133, 535-547.
Dajoz R (2006) Précis d’écologie. 8ème édition, Dunod.
- 152 -
Dale JM, Freedman B (1982) Lead and zinc contamination of roadside soil and vegetation
in Halifax, Nova Scotia. Proceedings of the Nova Scotian Institute of Science 32, 327
336.
Darren F, Ward T, New R, Yen AL (2001) Effects of pitfall trap spacing on the abundance,
richness and composition of invertebrate catches. Journal of Insect Conservation 5,
47-53.
Dasnias P, Gaultier C, Le Bloch F (1997) « Programme de Recherche sur les Dépendances
Vertes des Autoroutes Françaises : Résultats et analyses des inventaires floristiques ».
Scétauroute et AFSA. Ecosphère, étude et aménagement de milieux naturels.
Scétauroute et AFSA. Ecosphère, étude et aménagement de milieux naturels.
Datta SC, JJ Ghosh (1985) A study of the distribution pattern of lead in the leaves of banyan
trees (Ficus benghalensis) from different traffic density regions of Calcutta.
Ecotoxicology and Environmental Safety 9,101 106.
Dauber J, Hirscha M, Simmeringb D, Waldhardtb R, Otteb A, Woltersa V (2003)
Landscape structure as an indicator of biodiversity: matrix effects on species richness.
Agriculture, Ecosystems and Environment 98, 321-329.
Davies ZG, Pullin AS (2007) Are hedgerows effective corridors between fragments of
woodland habitat? An evidence-based approach. Landscape Ecology 22, 333-351.
Dawson P, Weste G (1985). Changes in the distribution of Phytophthora cinnamomi in the
Brisbane Ranges National Park between 1970 and 1980–81. Australian Journal of
Botany 33, 309 315.
Dawson DG (1994) Are habitat corridors conduits for animals and plants in a fragmented
landscape? A review of the scientific evidence. English Nature Research Report.
Dawson DG (2002) Roads and habitat corridors for animals and plants. In: Imperial Colledge
Press. Wildlife and roads, The ecological impacts, 186-198.
Day KR, Marshall S, Heaney C (1993) Associations between forest type and invertebrates:
Ground beetle community patterns in a natural oakwood and juxtaposed conifer
plantations. Forestry 66, 37-50.
De Snoo GR (1999) Unsprayed field margins: effects on environment, biodiversity and
agricultural practice. Landscape and Urban Planning 46, 151-160.
- 153 -
De Cauwer B, Reheul D, D’Hoogh K, Nijs I, Milbau A (2005) Evolution of the vegetation

of mown field margins over their first 3 years. Agriculture, Ecosystems and
Environment. 109, 87–96.
Del Campo JC, Marquinez JL, Naves J, Palomero G (1990). The brown bear in the
Cantabrian Mountains. Aquilo. Serie Zoologica 27, 97 101.
Delattre P, Giraudoux P, Baudry J, Musard P, Toussaint M, Truchetet D, Stahl P, Poule
ML, Artois M, Damange JP, Quéré JP (1992) Land use patterns of common vole
(Microtus arvalis) population kinetics. Agriculture, Ecosystems and Environment 39,
153-168.
Delattre P, De Sousa B, Fichet-Calvet E, Quéré JP, Giraudoux P (1999) Vole outbreaks
in a landscape context, Evidence from a six year study of Microtus arvalis. Landscape
Ecology 14, 401-412.
Den Boer PJ (1970) On the significance of dispersal power for populations of carabid-beetles
(Coleoptera, Carabidae). Oecologia 4, 1-28
Desbois D, Legris B (2007) Prix et coûts de production de six grandes cultures : Blé, maïs,
colza, tournesol, betterave et pomme de terre. L’agriculture, nouveaux défis, 65-78.
INSEE.
Develey PF, Stouffer PC (2001) Effects of roads on movements by understory birds in
mixed-species flocks in central Amazonian Brazil. Conservation Biology 15, 1416–
1422.
Devictor V, Julliard R, Couvet D, Lee A, Jiguet F (2007) Functional homogenization effect
of urbanization on bird communities, Conservation Biology 21, 741–751.
Dial R, Roughgarden J (1995) Experimental removal of insectivores from rain forest
canopy: Direct and indirect effects. Ecology 76, 1821-1834.
Diamond J (1989) Overview of recent extinctions. In Western D and Pearl M (Eds), 37–41.
Conservation for the 21st Century, Oxford University Press, Oxford.
Díaz S, Lavorel S, Chapin FS, Tecco PA., Gurvich DE, Grigulis K (2007) Functional
diversity - At the crossroads between ecosystem functioning and environmental filters.
Terrestrial ecosystems in a changing world, Part B, Chapter 7, 81-91. Springer Berlin
Heidelberg.
Dı́az S, Cabido M (2001) Vive la différence: plant functional diversity matters to ecosystem
processes. Trends in Ecology and Evolution 16, 646-655.
- 154 -
Direction Départementale de l’Equipement de Seine-et-Marne, Conseil Général de

Seine-et-Marne (1994) Ecosystème des bords de route : Vers une gestion écologique
des dépendances vertes routières. Ecosphère, étude et aménagement de milieux
naturels.
Direction Générale des Routes (2003) La nature et la route. Ministère de l’Equipement, des
Transports, du Logement, du Tourisme et de la Mer.
Dochinger LS (1980) Interception of airborne particles by tree plantings. Journal of
Environmental Quality 9, 265-268.
Doherty PF, Sorci G, Royle JA, Hines JE, Nichols JD, Boulinier T (2003) Sexual
selection affects local extinction and turnover in bird communities. Proceedings of the
National Academy of Sciences of the United States of America 100, 5858–5862.
Doledec S, Chessel D, ter Braak CJF, Champely S (1996) Matching species traits to
environmental variables: a new three-table ordination method. Environmental and
Ecological Statistics 3, 143–166.
Donald PF, Green RE, Heath MF (2000) Agricultural Intensification and the Collapse of
Europe's Farmland Bird Populations. Proceedings: Biological Sciences 268, 25-29.
Downie IS, Coulson JC, Butterfield JEL (1996) Distribution and dynamics of surface-
dwelling spiders across a pasture–plantation ecotone. Ecography 19, 29–40.
Duelli P, Obrist MK (2003) Biodiversity indicators: the choice of values and measures.
Duhamel F, Fosse D, Trouvilliez J, Pechoux J (1982) Ecologie du bord de route, Pas-de-
Calais, Somme. Centre d’études techniques de l’équipement de Lille.
Duhamel R, Quéré JP, Delattre P, Giraudoux P (2000) Landscape effects on the
population dynamics of the fossorial form of the water vole (Arvicola terrestris
Scherman). Landscape Ecology 15, 89–98.
Emets VM, Zhulidov AV (1983) Peculiarities of lead accumulation in the body of imago of
Pterostichus melanarius. III. (Coleoptera, Carabidae) in different populations in
unpolluted and polluted territories. Rossiiskaya Akademiya Nauk. Doklady 269,
1278-1280.
Endels P, Adriaens D, Bekker RM., Knevel IC, Decocq G, Hermy M (2007) Groupings of
life-history traits are associated with distribution of forest plant species in a
fragmented landscape. Journal of Vegetation Science 499-508.
- 155 -
Entling W, Schmidt MH, Bacher S, Brandl R, Nentwig W (2007) Niche properties of

Central European spiders: shading, moisture and the evolution of the habitat niche.
Global Ecology and Biogeography 16, 440–448.
Erome G, Aulagnier S (1982) contribution à l’identification des proies des rapaces. Bièvre
4, 129-135.
ESRI (2000) ArcView 3.2. Redlands, California.
Eversham BC, Telfer MG (1994) Conservation value of roadside verges for stenotopic
heathland Carabidae: corridors or refugia? Biodiversity and Conservation 3, 538-545.
Eviner VT (2004) Plant traits that influence ecosystem processes vary independently among
species. Ecology 85, 2215-2229.
Farmer AM (1993) The effects of dust on vegetation – A review. Environmental Pollution
79, 63–75.
Ferris R, Humphrey JW (1999) A review of potential biodiversity indicators for application
in British forests. Forestry 72, 313-328.
Fitter R (1986) Guide des fleurs sauvages. Delachaux and Niestle (Eds). Paris.
Flückiger W, Oertli JJ, Baltensweiler W (1978) Observations of an aphid infestation on
hawthorn in the vicinity of a motorway. Naturwissenschaften 65, 654-655.
Flueckiger W, Fluckiger-Keller H, Braun S (1984) Untersuchungen ueber waldschaeden in
der Nordwestschweiz. Schweizerische Zeitschrift fuer Forstwesen 135, 389 444.
Forman RTT, Baudry J (1984) Hedgerows and hedgerow networks in landscape ecology.
Environmental Management 8, 495–510.
Forman RTT, Alexander LE (1998) Roads and their major ecological effects. Annual
Review of Ecology and Systematics 29, 207-231.
Forman RTT (2000) Estimate the area affected ecologically by the road system in the United
States. Conservation Biology 14, 31-35.
Forman RTT, Deblinger RD (2001) The Ecological Road-Effect Zone of a Massachusetts
(U.S.A.) Suburban Highway. Conservation Biology 14, 36-46.
Forman RTT, Reineking B, Hersperger AM (2002) Road traffic and nearby grassland bird
patterns in suburbanizing landscape. Environmental Management 29, 782–800.
Forman, RTT, Sperling D, Bissonette J, Clevenger A, Cutshall C, Dale V, Fahrig L,
France R, Goldman C, Heanue K, Jones J, Swanson F, Turrentine T, Winter T
(2003) Road Ecology: Science and Solutions. Island Press, Washington, DC.
- 156 -
Fouillet P (1991) Mortalité des Insectes causée par la circulation automobile dans le Parc
Naturel Régional des Vosges du Nord et influence du fauchage tardif des bords de
route. PNR des Vosges du Nord. Office pour l’Information Eco-Entomologique.
Frampton GK, Cilgi T, Fry GLA, Wratten SD (1995) Effects of grassy banks on the
dispersal of some carabid beetles (Coleoptera: Carabidae) on farmland. Biological
Conservation 71, 347-355.
Franklin JF (1993) Preserving Biodiversity: Species, Ecosystems, or Landscapes?
Ecological Applications 3, 202-205.
Freckleton RP, Warkinson AR (2002) Are weed population dynamics chaotic? Journal of
Gabriel D, Thies C, Tscharntke T (2005) Local diversity of arable weeds increases with
landscape complexity. Perspectives in Plant Ecology, Evolution and Systematics 7,
85-93.
Gachet S, Errol V, Tatoni T (2005) BASECO: a floristic and ecological database of
Mediterranean French flora. Biodiversity and Conservation 14, 1023–1034.
Gad AM, Feinsod FM, Allam IH, Eisa M, Hassan AN, Soliman BA, El Said S, Saah AJ,
El Said S, Sellers RF, Hoogstraal H (1986) A possible route for the introduction of
Rift Valley fever virus into Egypt during 1977. Journal of Tropical Medicine and
Hygiene 89, 233 236.
Gallais A, Ricroch A (2006) Plantes transgéniques: faits et enjeux. Quae, Paris.
Garnier A, Deville A, Lecomte J (2006) Stochastic modelling of feral plant populations with
seed immigration and road verge management. Ecological Modelling 197, 373-382.
Gelbard JL, Belnap J (2003) Roads as conduits for exotic plant invasions in a semiarid
landscape. Conservation Biology 17, 420-432.
Genard M, Carsignol J, Charaven P, Cuenot E, Coumoul H, Desire G, Lansiart M,
Martin V, Payany M, Thiollet JM, Velluzzi V, Brigeot V (1994) La gestion
extensive des dépendences vertes routières. SERRA, Ministère de l’Environnement.
Génot JC, Barbault R (2004). Quelle politique de conservation ? Paper presented at the
Biodiversité et changements globaux. Enjeux de société et défis pour la recherche.
Genoud M (1984) Activity of Sorex coronatus (Insectivore, Soricidae) in the field.
Zeitschrift fuer Saeugetier 49, 74-78.
- 157 -
Genoud M (1985) Ecological energetics of two European shrews: Crocidura russula and
Sorex coronatus (Soricidae: Mammalia). Journal of Zoology 207, 63-85.
Getz LL, Cole FR, Gates DL (1978) Interstate Roadsides as Dispersal Routes for Microtus
pennsylvanicus. Journal of Mammalogy 59, 208-212.
Giraudoux P, Delattre P, Habert M, Quéré JP, Deblay S, Defaut R, Duhamel R,
Moissenet MF, Salvi D, Trebuchetet D (1997) Population dynamics of fossorial
water vole (Arvicola terrestris Scherman: a land use and landscape perspective.
Godfree R, Lepschi B, Mallinson D (2004) Ecological filtering of exotic plants in an
Australian environement. Journal of Vegetation Science 15, 227-236.
Goldsmith CD, Scanlon PF, Pirie WR (1976) Lead concentrations in soil and vegetation
associated with highways of different traffic densities. Bulletin of Environmental
Contamination and Toxicology 16, 66 70.
Gonzalez-Andujar JL (1996) High control measures cannot produce extinction in weed
populations. Ecological Modelling 91, 293-294.
Greave MP, Marshall EJP (1987) Field margins: definitions and statistics. In Field Margins,
Monograph N° 35. Way JM and Greig-Smith PJ (Eds), 3–10. British Crop Protection
Council, Surrey, UK.
Greenberg CH, Crownover SH, Gordon DR (1997) Roadside soils: A corridor for invasion
of xeric scrub by nonindigenous plants. Natural Areas Journal 17, 99-109.
Greszta J (1982) Correlation between the content of copper, zinc, lead and cadmium in the
soil and the content of these metals in the seedlings of selected forest tree species.
Fragmenta Floristica et Geobotanica 28, 29-52.
Griffiths GJK, Winder L, Holland JM, Thomas CFG, Williams E (2007) The
representation and functional composition of carabid and staphylinid beetles in
different field boundary types at a farm-scale. Biological Conservation 135, 145–152.
Grue CE, Hoffman DJ, Beyer WN, Franson LP (1986) Lead concentrations and
reproductive success in European starlings, Sturnus vulgaris, nesting within highway
roadside verges. Environmental Pollution (Series A) 42, 157-182.
Hansen K, Jensen J (1972) The vegetation on roadsides on Denmark: a qualitative and
quantitative comparison. Dansk Botanisk Arkiv 28, 21–61.
- 158 -
Hansen MJ, Clevenger AP (2005) The influence of disturbance and habitat on the presence
of non-native plant species along transport corridors. Biological Conservation 125,
249–259.
Hansson L (1988) The domestic cat as a possible modifier of vole dynamics. Mammalia 52,
159-164.
Harper JL (1977) Population biology of plants (ed. Academic Press). New York, USA.
Harris LD, Scheck J (1991) From implications to applications: the dispersal corridor
principle applied to the conservation of biological diversity. In Sounders DA and
Hobbs RJ (Eds). Nature conservation 2: the role of corridors. Surrey Beatty, Sydney.
Harris LD, Silva-Lopez G (1993) Forest fragmentation and the conservation of biological
diversity. In Fiedler PL and Jain SK (Eds), 171–196. Conservation Biology. Chapman
and Hall, New York.
Harris LD, Hoctor TS, Gergel SE (1996) Landscape processes and their significance to
biodiversity conservation. In Population Dynamics in Ecological Space and Time.
Rhodes O, Chesser R, Smith M (Eds). Chicago: Univ. Chicago Press.
Helvey JD, Kochenderfer JN (1990) Soil density and moisture content on two unused forest
roads during first 30 months after construction. Research paper NE-629. U.S. Forest
Service, Northeast Forest Experiment Station, Broomhall, Pennsylvania.
Henderson L, Wells MJ (1986) Alien plant invasions in the grassland and savanna biomes.
In Macdonald IAW, Kruger FJ and Ferrar AA (Eds). The ecology and management of
biological invasions in southern Africa, 109 117. Oxford University Press, Capetown.
Hendrix PF, Crossley DA, Blair JM, Coleman DC (1990) Soil biota as components of
sustainable agroecosystems. In Edwards CA, Lal R, Madden P, Miller RH, House G
(Eds), 637–654. Sustainable Agricultural Systems. Soil and Water Conservation
Society, IA.
Hicks BJ, Hall JD, Bisson PA, Sedell JR (1991) Response of salmonids to habitat change.
In W. R. Meehan, Editor. Influences of forest and rangeland management on salmonid
fishes and their habitats, 483-518. Special publication 19. American Fisheries Society,
Bethesda, Maryland.
Hill M.O (1973) Diversity and Eveness: a unifying notation and its consequences. Ecology
54, 427-432.
- 159 -
Hines JE, Boulinier T, Nichols JD, Sauer JR, Pollock KH (1999) COMDYN: Software to
study the dynamics of animal communities using a capture-recapture approach. Bird
Study 46, 209-217.
Hobbs RJ, Mooney HA (1998) Broadening the extinction debate: population deletions and
additions in California and Western Australia. Conservation Biology 12, 271–83.
Hodkinson DJ, Ken Thompson K (1997) Plant dispersal: The role of man. The Journal of
Applied Ecology. 34, 1484-1496.
Hoffman RW, Goldman CR, Paulson S, Winters GR (1981) Aquatic impacts of deicing
salts in the central Sierra Nevada Mountains, California. Water Resources Bulletin 17,
280-285.
Honek A (1997) The effect of temperature on the activity of Carabidae (Coleoptera) in a
fallow field. European Journal of Entomology 94, 97–104.
Hüsken A, Dietz-Pfeilstetter A (2007) Pollen-mediated gene flow from herbicide-resistant
oilseed rape (Brassica napus L.). Transgenic Research 16, 557-569.
Huston MA (1999) Local processes and regional patterns: Appropriate scales for
understanding variation in the diversity of plants and animals. Oikos 86, 393-401.
Ihaka R, Gentleman R (1996) R: A Language for Data Analysis and Graphics. Journal of
Computational and Graphical Statistics 5, 299-314.
Indu B, Choudhri GN (1991) Impact of automobile effusion on plant and soil. International
Journal of Ecology and Environmental Sciences 17, 121-127.
Irwin A (2001) Constructing the scientific citizen: Science and democracy in the biosciences.
Public Understanding of Science 10, 1-18.
Jantunen J, Saarinen K, Valtonen A, Saarnio S (2007) Flowering and seed production
success along roads with different mowing regimes. Applied Vegetation Science 10,
285–292.
Jauzein P (1995) Flore cultivée des champs. Sopra. INRA.
Johnson NP (1990) Nesting Bald Eagles (Haliaeetus leucocephalus) in urban areas of
southeast Alaska: assessing highway construction and disturbance impacts.
Transportation Research Record 127, 960–968.
Johnston FM, Johnston SW (2004) Impacts of road disturbance on soil properties and on
exotic plant occurrence in subalpine areas of the Australian Alps. Arctic, Antarctic,
and Alpine Research 36, 201–207.
- 160 -
Kammerbauer H, Selinger H, Rommelt R, Ziegler-Jons A, Knoppik D, Hock B (1986)

Toxic effects of exhaust emissions on spruce Picea abies and their reduction by the
catalytic converter. Environmental pollution (Series A) 42, 133–142.
Karnefelt I, Mattsson JE (1989) Cetraria cucullata and Cetraria nivalis, two vanishing
lichens from southernmost Sweden. International Journal of Mycology and
Lichenology 4, 299 306.
Keddy PA (1992) Assembly and response rules – 2 goals for predictive community ecology.
Journal of Vegetation Science 3, 157–164.
Kerbiriou C, Le Viol I, Jiguet F, Julliard R (2008) The impact of human frequentation on
coastal vegetation in a biosphere reserve. Journal of Environmental Management 88,
715-28.
Kirby KJ (1997) Habitat fragmentation and infrastructure: problem and research. In Canters
K (Eds). Habitat fragmentation and infrastructure. Delft: Ministry of Transport, Public
Works and Water Management, 32-39.
Kleijn D, Verbeek M (2000) Factors affecting the species composition of arable field
boundary vegetation. Journal of Applied Ecology 37, 256-266.
Klejin D, Snoeijing GIJ (1997) Field boundary vegetation and the effects of agrochemical
drift: botanical change caused by low levels of herbicide and fertilizer. Journal of
Applied Ecology 34, 1413-1425.
Kohler F, Verhulst J, Knop E, Herzog F, Kleijn D (2006) Indirect effects of grassland
extensification schemes on pollinators in two contrasting European countries.
Biological Conservation 135, 302-309.
Koivula MJ (2005) Effects of forest roads on spatial distribution of boreal carabid beetles
(Coleoptera: Carabidae). The Coleopterists Bulletin 59, 465–487.
Koivula M J (2006) Effects of Forest Roads on Spatial Distribution of Boreal Carabid
Beetles (Coleoptera: Carabidae) The Coleopterists Bulletin 59, 465–487.
Korn H (1991) Rapid repopulation by small mammals of an area isolated by roads.
Mammalia 55, 629–632.
Kotze DJ, Samways MJ (1999) Support for the multi-taxa approach in biodiversity
assessment, as shown by epigaeic invertebrates in an Afromontane forestarchipelago.
Journal of the Insect Conservation 3, 125-143.
- 161 -
La Peña (de) NM, Butet A, Delettre Y, Paillat G, Morant P, Le Du L, Burel F (2003)

Response of the small mammal community to changes in western French agricultural
landscapes. Landscape Ecology 18, 265-278.
Lacy R (1997) Importance of genetic variation to the viability of mammalian populations:
Perspectives on individual variation in mammals. Journal of Mammalogy 78, 320-
335.
Lambinon J, Delvossalle L, Duvigneaud J (2004) Nouvelle Flore de la Belgique, du Grand-
Duché du Luxembourg, du Nord de la France et des Régions Voisines. 5ème Edition,
Editions du Patrimoine du jardin botanique national de Belgique.
Lauber K, Wagner G (2007) Flora Helvetica.2ème Edition, Belin.
Lavorel S, Garnier E (2002) Predicting changes in community composition and ecosystem
functioning from plant traits revisiting the Holy Grail. Functional Ecology 16, 545–
556.
Le Coeur D, Baudry J, Burel F (1997) Field margins plant assemblages: variation
partitioning between local and landscape factors. Landscape and Urban Planning 37,
57-71.
Le Coeur D, Baudry J, Burel F, Thenail C (2002) Why and how we should study field
boundary biodiversity in an agrarian landscape context. Agriculture Ecosystem and
Environment 89, 23-40.
Le Viol I, Julliard R, Kerbiriou C, Redon (de) L, Carnino N, Machon N, Porcher E
(2008) Plant and spider communities benefit differently from the presence of planted
hedgerows in highway verges. Biological Conservation 141, 1581-1590.
Leakey RE, Lewin R (1995) The sixth extinction: patterns of life and the future of
humankind. Doubleday, New York, 271 p.
Lecointe A, Trihot C, Lesaulnier S (1990) Recherche sur la gestion naturelle des abords
routiers du Calvados : Utilisation de la végétation. DDE 14 and Laboratoire de
phytogéographie.
Legrand A, Keime MP, Genard M (1989) Entretien des bords de route : Influence sur la
flore et la faune. SRETIE, Secrétariat à l’Environnement, contrat n° 88-061.
Leharne S, Charlesworth D, Chowdhry B (1992) A survey of metal levels in street dusts in
an inner London neighbourhood. Environment International 18, 263 270.
- 162 -
Leibold MA, Holyoak M, Mouquet N, Amarasekare P, Chase JM, Hoopes MF, Holt RD,
Shurin JB, Law R, Tilman D, Loreau M, Gonzalez A (2004) The metacommunity
concept: a framework for multi-scale community ecology. Ecology Letters 7, 601-
613.
Lekve K, Boulinier T, Stenseth NC, Gjosaeter J, Fromentin JM, Hines JE, Nichols JD
(2002) Spatio-temporal dynamics of species richness in coastal fish communities.
Proceedings of the Royal Society of London (Series B) 269,1781–1789.
Leps J, Dolezal J, Bezemer TM, Brown VK, Hedlund K, Igual AM, Jorgensen HB,
Lawson CS, Mortimer SR, Peix-Geldart A, Rodriguez-Barrueco C, Santa-Regina
I, Smilauer P, van der Putten WH (2007) Long-term effectiveness of sowing high
and low diversity seed mixtures to enhance plant community development on ex-
arable fields. Applied Vegetation Science 10, 97–110.
Levins R (1968) Evolution in changing environments. Princeton: PrincetonUniversity Press.
Lindenmayer DB (1999) Future directions for biodiversity conservation in managed forests:
indicator species, impact studies and monitoring programs. Forest Ecology and
Management 115, 277-28.
Lindenmayer DB, Margules CR, Botkin DB (2000) Indicators of biodiversity for
ecologically sustainable forest management. Conservation Biology 14, 941-950.
Lodé T (2000) Effetcts of a mortorway on mortality and isolation of wildlife populations.
Ambio 29, 163-166.
Lomborg B (2001) The skeptical environmentalist: Measuring the real state of the world.
Cambridge: Cambridge University Press.
Lonsdale WM, Lane AM (1994) Tourist vehicles as vectors of weed seeds in Kakadu
National Park, Northern Australia. Biological Conservation 69, 277-283.
Lorreau M (2000) Are communities saturated? On the relationship between α, β and γ
diversity. Ecology Letters 3, 73-76.
Lovei GL, Sunderland KD (1996) Ecology and behavior of ground beetles (Coleoptera:
Carabidae). Annual Review of Entomology 41, 231-256.
Lovell S, Hamer M, Slotow R, Herbert D (2007) Assessment of congruency across
invertebrate taxa and taxonomic levels to identify potential surrogates. Biological
- 163 -
Lowell WA, Geis AD (1983) Effects of roads on small mammals. Journal of Applied
Ecology 20, 403-415.
MacArthur RH, Wilson EO (1967) The Theory of Island Biogeography. Princeton
University Press, Princeton, New Jersey.
Machon N, Bardin P, Mazer SJ, Moret J, Godelle B, Austerlitz F (2003) Relationship
between genetic structure and seed and pollen dispersal in the endangered orchid
Spiranthes spirali. New Phytologist 157, 677-687.
Mader HJ (1984) Animal habitat isolation by roads an agricultural field. Biological
Mahe C (2007) Gestion de la végétation spontanée sur la voirie et ses dépendances vertes en
France métropolitaine : Etat des lieux et propositions d’actions. Mémoire de stage.
ENESAD.
Maisonneuve C, Rioux S (2001) Importance of riparian habitats for small mammal and
herpetofaunal communities in agricultural landscapes of southern Québec.
Agriculture, Ecosystem and Environment 83, 165-175.
Mann CC (1991) Extinction: Are Ecologists Crying Wolf? Science 253, 736–738.
Mansergh IM, Scotts DJ (1989) Habitat continuity and social organization of the mountain
pygmy-possum restored by tunnel. Journal of Wildlife Management 53, 701–707.
Marc P, Canard A, Ysnel F (1999) Spiders (Araneae) useful for pest limitation and
bioindication. Agriculture Ecosystems and Environment 74, 229–273.
Marshall EJP, Moonen AC (2002) Field margins in northern Europe: their functions and
interactions with agriculture. Agriculture, Ecosystems and Environment 89, 5-21.
Marshall EJP, Brown VK, Boatman ND, Lutman PJW, Squire GR, Ward LK (2003)
The role of weeds in supporting biological diversity within crop fields. Weed
Research 43, 77-89.
Martina M, Bastardie F, Richard D, Burela F (2001) Studying boundary effects on animal
movement in heterogeneous landscapes: the case of Abax ater (Coleoptera:
Carabidae) in hedgerow network landscapes. Life Sciences 324, 1029–1035.
Matesanz S, Valladares F, Tena D, Costa-Tenorio M, Bote D (2006) Early dynamics of
plant communities on revegetated motorway slopes from southern Spain: is
hydroseeding always needed? Restoration Ecology 14, 297–307.
- 164 -
Matlack GR (1993) Sociological edge effects: spatial distribution of human impact in

suburban forest fragments. Environmental Management 17, 829 835.
Mattson MD, Godfrey PJ (1994) Identification of road salt contamination using multiple
regression and GIS. Environmental Management 18, 767 773.
Maudsley MJ (2000) A review of the ecology and conservation of hedgerow invertebrates in
Britain, Journal of Environmental Management 60, 65–76.
Mazerolle MJ (2004) Amphibian road mortality in response to nightly variation in traffic
intensity. Herpetologica 60, 45–53.
McCollin D, Jackson JI, Bunce RGH, Barr CJ, Stuart R (2000) Hedgerows as habitat for
woodland plants. Journal of Environmental Management 60, 77–90.
McGill BJ, Enquist BJ, Weiher E, Westoby M (2006) Rebuilding community ecology from
functional traits. Trends in Ecology and Evolution 21, 178–185.
McKinney ML, Lockwood JL (1999) Biotic homogenisation, a few winners replacing many
losers in the next mass extinction. Trends in Ecology and Evolution 14, 450–453.
McKinney ML (2006) Urbanization as a major cause of biotic homogenization. Biological
McNeill SE, Fairweather PG (1993) Single large or several small marine reserves? An
experimental approach with seagrass fauna. Journal of Biogeography 20, 429-440.
Meeka B, Loxton D, Sparksa T, Pywella R, Picketta H, Nowakowskic (2002) The effect
of arable field margin composition on invertebrate biodiversity. Biological
Melbourne BA (1999) Bias in the effect of habitat structure on pitfall traps: an experimental
evaluation. Australian Journal of Ecology 24, 228–239.
Merriam G, Kozakiewiez M, Tsuchiya E, Hawley K (1989) Barriers as boundaries for
metapopulations and demes of Peromyscus leucopus in farm landscapes. Landscape
Ecology 2, 227–235.
Merriam G, Lanoue A (1990) Corridor use by small mammals: field measurement for three
experimental types of Peromyscus leucopus. Landscape Ecology 4, 123-131.
Meunier FD, Corbin J, Verheyden C, Jouventin P (1999) Effects of landscape type and
extensive management on use of motorway roadsides by small mammals. Canadian
Journal of Zoology 77, 108-117.
- 165 -
Meunier FD, Verheyden C, Jouventin P (2000) Use of roadsides by diurnal raptors in

agricultural landscapes. Biological Conservation 92, 291–298.
Miller S (2001) Public understanding of science at the crossroads. Public Understanding of
Science 10, 115-120.
Mitchell-Jones AJ, Amori G, Bogdanowicz W, Krystufek B Reijnders PJH,
Spitzenberger F, Stubbe M, Thissen JBM, Vohralik V, Zima J (1999) The atlas of
European mammals. Pyser Natural History. T and AD Pyser for the Societas Europaea
Mammalogica.
Mitchell-Jones AJ, Reijnders P, Ziman JM (2002) The Atlas of European Mammals. T &
AD Poyser for Societas Europaea Mammalogica.
Miyashita T (1999) The effects of forest fragmentation on web spider communities in urban
areas. Biological Conservation 86, 357-364.
Miyashita T (2002) Life-history variation in closely related generalist predators living in the
same habitat: a case study with three Cyclosa spiders. Functional Ecology 13, 307-
314.
Molles MC, Gosz JR (1980) Effects of a ski area on the water quality and invertebrates of a
mountain stream. Water, Air and Soil Pollution 14, 187 205.
Morgan E, Porak W, Arway J (1983) Controlling acidic-toxic metal leachates from
southern Appalachian construction slopes: mitigating stream damage. Transportation
Research Record. 948, 10–16.
Murphy SM, Curatolo JA (1987) Activity budgets and movement rates of caribou
encountering pipelines, roads, and traffic in northern Alaska. Canadian Journal of
Zoology 65, 2483-2490.
Murray KF (1965) Population changes during the 1957-1958 vole (Microtus) outbreak in
California. Ecology 46, 163-171.
Neet CR, Hausser J (1990) Habitat selection in zones of parapatric contact between the
common shrew Sorex araneus and Millet’s shrew Sorex Coronatus. Journal of Animal
Ecology 59, 235-250.
Nellemann C, Vistnes I, Jordhoy P, Strand O (2001) Winter distribution of wild reindeer in
relation to power lines, roads and resorts. Biological Conservation 101, 351–360.
Newbey BJ, Newbey KR (1987) Bird dynamics of foster road reserve, near ongerup, western
Australia. nature conservation: the role of remnants of native vegetation. In Saunders
- 166 -
DA, Arnold GW, Burbridge AA, Hopkins AJM (Eds), 341–343. Surrey Beatty and
Sons Pty Ltd, Chipping Norton.
Nichols JD, Boulinier T, Hines JE, Pollock KH, JR (1998) Sauer, Inference methods for
spatial variation in species richness and community composition when not all species
are detected. Conservation Biology 12, 1390–1398.
Noble IR, Slatyer RO (1980) The use of vital attributes to predict successional changes in
plant communities subject to recurrent disturbances. Plant Ecology 43, 5-21.
Noordijk N, Schaffers AP, Sykora KV (2008) Diversity of ground beetles (Coleoptera:
Carabidae) and spiders (Araneae) in roadside verges with grey hair-grass vegetation.
European Journal of Entomology 105, 257–265.
Noss RF (1987) Corridors in Real Landscapes: A Reply to Simberloff and Cox. Conservation
Biology 1, 159-164.
Noss RF, Cooperrider AY (1994) Saving Nature's Legacy: Protecting and Restoring
Biodiversity. Washington (DC). Island Press.
Nyffeler M, Sunderland KD (2003) Composition, abundance and pest control potential of
spider communities in agroecosystems: a comparison of European and US studies.
Agriculture Ecosystems and Environment 95, 579–612.
O’Farrell PJ, Milton SJ (2006) Road verge and rangeland plant communities in the southern
Karoo: Exploring what influences diversity, dominance and cover. Biodiversity and
Oberg S, Mayr S, Dauber J (2008) Landscape effects on recolonisation patterns of spiders
in arable fields, Agriculture Ecosystems and Environment 123, 211–218.
Obrtel R, Zejda J, Holisova V (1978) Impact of small rodent predation on an overcrowded
population of Diprion pini during winter (pines). Folia Zoologica 27, 97-110
Oksanen J, Kindt R, Legendre P, O’Hara RB (2007) Vegan: Community Ecology Package
Version 1.8–5, URL: http://cran.r-project.org.
Olden JD, Poff NL (2003) Toward a mechanistic understanding and prediction of biotic
homogenisation. American Naturalist 162, 442-460.
Olden JD, Rooney TP (2006) On defining and quantifying biotic homogenisation. Global
Ecology and Biogeography 15, 113–120.
Oxley DJ, Fenton MB, Carmody GR (1974) The effects of roads on populations of small
mammals. Journal of Applied Ecology 11, 51-59.
- 167 -
Pain DJ, Pienkowski MW (1997) Farming and birds in Europe: the common agricultural
policy and its implications for bird conservation. Academic Press.
Pantaleoni RA (1989) Ways of invasion of a new area by Metcalfa pruinosa (Say, 1830)
(Auchenorrhyncha, Flatidae). Bollettino Dell'istituto Di Entomologia Della Universita
Degli Studi Di Bologna 43, 1-8.
Paoletti MG, Foissner W, Coleman D (1994) Soil Biota, Nutrient Cycling, and Farming
Systems. Lewis Publishers, Boca Raton.
Parendes LA, Jones JA (2000) Role of light avaibility and dispersal in exotic plant invasion
along roads and streams in the HJ Andrews experimental forest, Oregon. Conservation
Biology 14, 64-75.
Parmenter RR, MacMahon, JA (1988) Factors limiting populations of arid-land darkling
beetles (Coleoptera: Tenebrionidae): predation by rodents. Animal Ecology 17, 280-
286
Parmesan C, Yohe G (2003) A globally coherent fingerprint of climate change impacts
across natural systems. Nature 421, 37-42.
Parr TW, Way JM (1988) Management of roadside vegetation: The long-term effects of
cutting. Journal of Applied Ecology 25, 1073-1087.
Pasquier-Barre F (1999) Diprioni pini/Pinus sylvestris relationship foliage characterizations
and biological consequences. PhD thesis. University of Paris 06, France.
Pearman PB, Weber D (2007) Common species determine richness patterns in biodiversity
indicator taxa, Biological Conservation 138, 109–119.
Pearson DL (1994) Selecting indicator taxa for the quantitative assessment of biodiversity.
Philosophical Transactions. Biological Sciences 345, 75-79.
Pedevillano C, Wright RG (1987) The influence of visitors on mountain goat activities in
Glacier National Park, Montana. Biological Conservation 39, 1–11.
Perkins AJ, Whittingham MJ, Morris AJ, Bradbury RB (2002) Use of field margins by
foraging yellowhammers Emberiza citrinella. Agriculture, Ecosystems and
Environment 93, 413-420.
Pernetta JC (1976) Diet of the shrew Sorex araneus and Sorex minutus in Wytham
grassland. Journal of Animal Ecology 45, 899-912
Perry DA (1994) Forest Ecosystems. Johns Hopkins University Press, Baltimore, MD.
- 168 -
Petchey OL, Gaston KJ (2002) Functional diversity (FD), species richness and community
composition. Ecology Letters 5, 402–411.
Peters NE, Turk JT (1981) Increases in sodium and chloride in the Mohawk River, New
York, from the 1950s to the 1970s attributed to road salt. Water Resources Bulletin
17, 586 598.
Petriccione B (2004) Monitoring and Indicators of Forest Biodiversity in Europe. From Ideas
to Operationality. Marchetti M (Eds), 445–454. EFI Proceedings 51.
Pita R, Mira A, Beja P (2006) Conserving the Cabrera vole, Microtus cabrerae, in
intensively used Mediterranean landscapes. Agriculture, Ecosystems and Environment
115, 1-5.
Pivard S, Adamczyk K, Lecomte J, Lavigne C, Bouvier A, Deville A, Gouyon PH, Huet
S (2008) Where do the feral oilseed rape populations come from? A large-scale study
of their possible origin in a farmland area. Journal of Applied Ecology 45, 476-485.
Platnick NI (2007) The world spider catalog, Version 8. American Museum of Natural
History, New York, NY.
PNR Loire Anjou Touraine (2002) Les dépendances vertes routières, refuges à flore et à
faune. Les entretiens techniques et écologiques du parc naturel régional Loire Anjou
Touraine.
Post RD, Beeby AN (1996) Activity of the microbial decomposer community in metal-
contaminated roadside soils. Journal of Applied Ecology 33, 703-709.
Preud’Homme RL (2007) Inventaires floristiques et études des facteurs impactant la
diversité végétale des bords de route. Rapport de stage de M1 Biogéomédia, Paris 7
Denis Diderot. UMR 51/73 MNHN/CNRS/UPMC.
Primack RB, Miao SL (1992) Dispersal Can Limit Local Plant Distribution. Conservation
Biology 6, 513-519.
Pyšek P, Chocholoušková Z, Pyšek A, Jarošík V, Chytrŷ M, Tichŷ L ( 2004) Trends in
species diversity and composition of urban vegetation over three decades. Journal of
Vegetation Science 15, 781-788.
Pyšek P, Hulme PE (2005) Spatio-temporal dynamics of plant invasions: Linking pattern to
process. Ecoscience 12, 302–315.
- 169 -
Pywell RF, Bullock JM, Roy DB, Warman LIZ, Walker KJ, Rothery P (2003) Plant traits
as predictors of performance in ecological restoration. Journal of Applied Ecology 40,
65-77.
Pywell RF, James KL, Herbert I, Meek WR, Carvell C, Bell D, Sparks TH (2005)
Determinants of overwintering habitat quality for beetles and spiders on arable
farmland. Biological Conservation 123, 79–90.
R Developpment core team (2004) R: A language and environment for statistical
computing. R Foundation for Statistical Computing, Vienna, Austria. URL:
http://www.R-project.org.
Rameau JC, Mansion D, Dumé G, Timble J, Lecointre A, Supont P, Keller R (2005)
Flore forestière française. Guide écologique illustré. 1- Plaines et collines. Institut
pour le développement forestier.
Regal PJ (1982) Pollination by wind and animals: Ecology of geographic patterns. Annual
Reviews in Ecology and Systematics 13, 497-524.
Reijnen R, Foppen R (1994) The effects of car traffic on breeding bird populations in
woodland. I. Evidence of reduced habitat quality for willow warblers (Phylloscopus
trochilus) breeding close to a highway. Journal of Applied Ecology 31, 85–94.
Reijnen R, Foppen R, Veenbaas G (1997) Disturbance by traffic of breeding birds:
evaluation of the effect and considerations in planning and managing road corridors.
Biodiversity and Conservation 6, 567-581.
Rentch JS, Fortney RH, Stephenson SL, Adams HS, Grafton WN, Anderson JT (2004)
Vegetation–site relationships of roadside plant communities in West Virginia, USA.
Ribera I, Doledec S, Downie I.S, Foster GN (2001) Effect of land disturbance and stress on
species traits of ground beetle assemblages. Ecology 82, 1112–1129.
Rich TCG, Woodruff, ER (1992) Recording bias in botanical surveys. Watsonia 19, 73-95.
Rich AS, Dobkin DS Niles LJ (1994) Defining forest fragmentation by corridor width: the
influence of narrow forest-dividing corridors on forest-nesting birds in Southern New
Jersey. Conservation Biology 8, 1109–1121.
Rico A, Kindlmann P, Sedlacek F (2007) Barrier effects of roads on movements of small
mammals. Folia zoological 56, 1-12.
- 170 -
Ries L, Debinski DM, Wieland ML (2001) Conservation value of roadside prairie

restoration to butterfly communities. Conservation Biology 15, 401–411.
Robert P, Escouffier Y (1977) An unifying tool for linear multivariate statistical methods:
the RV-coefficient. Applied Statistics 25, 257–265.
Roberts M (1985) The Spiders of Great Britain and Ireland, Atypidae to Theridosomatidae
vol. 1. Harley Books, Colchester, Essex.
Roberts M (1987) The Spiders of Great Britain and Ireland, Linyphiidae vol. 2. Harley
Books, Colchester, Essex.
Robinson JG, Bodmer RE (1999) Towards wildlife management in tropical forests. Journal
of Wildlife Management 63, 1–13.
Roillet J (2007) Impact du paysage environnant et de l’entretien des dépendances vertes des
bords de route sur des carabes. Rapport de stage de M1 Sciences et Technologies du
Vivant, AgroParisTech. UMR 51/73 MNHN/CNRS/UPMC.
Rondinini C, Ercoli V, Boitani L (2006) Habitat use and preference by polecats (Mustela
putorius L.) in a Mediterranean agricultural landscape. Journal of Zoology 269, 213-
219.
Rosenzweig ML (2003) Win-win ecology: how the earth's species can survive in the midst of
human enterprise. Oxford University Press.
Ross SM (1986) Vegetation change on highway verges in south-east Scotland. Journal of
Biogeography 13, 109-113.
Roth NE, Allan JD, Erickson DL (1996) Landscape influences on stream biotic integrity
assessed at multiple spatial scales. Landscape Ecology 11, 141-156.
Roy V, Blois (de) S (2006) Using functional traits to assess the role of hedgerow corridors as
environmental filters for forest herbs, Biological Conservation 130, 592–603.
Rutter AJ, Thompson JR (1986) Salinity of Motorway Soils. III. Simulation of the Effects
of Salt Usage and Rainfall on Sodium and Chloride Concentrations in the Soil of
Central Reserves. Journal of Applied Ecology 23, 281-297.
Samways MJ (1994) Insect conservation biology. London: Chapman and Hall. Sanford MT.
Santos M, Luz-Mathias (da) M, Mira A, Simões M (2007) Vegetation structure and
composition of road verge and meadow sites colonized by Cabrera vole (Microtus
cabrerae Thomas). Polish Journal of Ecology, 481-493.
- 171 -
Sarkar RK, Banerjee A, Mukherji S (1986) Acceleration of peroxidase and catalase

activities in leaves of wild dicotyledonous plants, as an indication of automobile
exhaust pollution. Environmental pollution (Series A) 42, 289–295.
Schaffers AP, Vesseur MC, Sýkora KV (1998) Effects of delayed hay removal on the
nutrient balance of roadside plant communities. Journal of Applied Ecology 35, 349-
364.
Schaffers AP (2002) Soil, biomass, and management of semi-natural vegetation – Part II.
Factors controlling species diversity. Plant Ecology 158, 247–268.
Schaffers AP, Vesseur MC, Sýkora. KV (2002) Effects of delayed hay removal on the
nutrient balance of roadside plant communities. Journal of Applied Ecology 35, 349-
364.
Schmidt W (1989) Plant dispersal by motor cars. Vegetatio 80, 147-52.
Schmidt MH, Tscharntke T (2005) The role of perennial habitats for Central European
farmland spiders. Agriculture, Ecosystems and Environment 105, 235–242.
Schoener TW (1988) Leaf damage in island buttonwood, Conocarpus erectus: Correlations
with pubescence, island area, isolation and the distribution of major carnivores. Oikos
53, 253-266.
Scott NE, Davison AW (1982) De-icing salt and the invasion of road verges by maritime
plants. Watsonia 14, 41 52.
Scott NE (1985) The updated distribution of maritime species on British roadsides. Watsonia
15, 381–386.
Scott NE, Davison AW (1985) The distribution and ecology of coastal species on roadsides.
Plant Ecology 62, 433-440.
Seabrook WA, Dettmann EB (1996) Roads as activity corridors for cane toads in Australia.
The Journal of Wildlife Management 60, 363-368.
Seibert P (1993) Vegetation and man in South Africa from a historical perspective.
Phytocoenologia 23, 457-493.
Selmi S, Boulinier T (2003) Does time of season influence bird species number determined
from point-count data? A capture–recapture approach. Journal of Field Ornithology
74, 349–356.
SETRA (1980) Impacts des routes sur la végétation. Rapport du groupe de travail LCPC –
SETRA.
- 172 -
SETRA (1994) La gestion extensive des dépendances vertes routières. Ministère de

l’Environnement.
SETRA (1994) Routes et Environnement : Guide pratique. Rapport TWU 13.
Sherburne J (1985) Wildlife populations utilizing right-of-way habitat along interstate 95 in
northern Maine. Transportation Research Record 1016, 16-20.
Sherwood B, Cutler D, Burton J (2002) Wildlife and roads, The ecological impact. London:
Imperial College Press.
Simberloff D, Abele LG (1982) Refuge design and island biogeographic theory: Effects of
fragmentation. The American Naturalist 120, 41-50.
Singh RP, Tripathi RD, Dabas S, Rizvi SMH, Alic MB, Sinha SK, Gupta DK, Mishra S,
Rai UN (2003) Effect of lead on growth and nitrate assimilation of Vigna radiata (L.)
Wilczek seedlings in a salt affected environment. Chemosphere 52, 1245–1250.
Sitzia T (2007) Hedgerows as corridors for woodland plants: A test on the Po plain, northern
Italy. Plant Ecology 188, 235–252.
Sjödin A, Lenner M (1995) On-road measurements of single vehicle pollutant emissions,
speed and acceleration for large fleets of vehicles in different traffic environments.
Science of the Total Environment 169, 157-165.
Smart SM, Marrs RH, Le Duc MG, Thompson KEN, Bunce RGH, Firbank LG, Rossall
MJ (2006) Spatial relationships between intensive land cover and residual plant
species diversity in temperate farmed landscapes. Journal of Applied Ecology 43,
1128–1137.
Smart SM, Thompson K, Marrs RH, Le Duc MG, Maskell LC, Firbank LG (2006)
Biotic homogenization and changes in species diversity across human-modified
ecosystems. Philosophical Transactions of the Royal Society (B) 273, 2659–2665.
Southwood RE, Way MJ (1970) Ecological background to pest management. In Rabb RC,
Guthrie FE (Eds), 6–29. Concepts of Pest Management. North Carolina State
University, Raleigh, NC.
Spellerberg IF (1998) Ecological effects of roads and traffic: A literature review. Global
Ecology and Biogeography Letters 7, 317–333.
Spencer HJ, Port GR (1988) Effects of roadside conditions on plants and insects. II. Soil
conditions. Journal of Applied Ecology 25, 709–715.
- 173 -
Spencer HJ, Scott NE, Port GR, Davison AW (1988) Effects of roadside conditions on
plants and insects. I. Atmospheric conditions. Journal of Applied Ecology 25, 699–
707.
Spiller DA, Schoener TW (1990) A terrestrial field experiment showing the impact of
eliminating top predators on foliage damage. Nature 347, 469-472.
Spooner PG, Lunt ID, Briggs SV (2004) Spacial analysis of antropogenic disturbance
regimes and roadside shrubs in a fragmentaed agricultural landscape. Applied
Vegetation Science 7, 61-70.
Spooner PG, Lunt ID, Briggs SV, Freudenberger D (2004) Effects of soil disturbance
from roadworks on roadside shrubs in a fragmented agricultural landscape. Biological
Suckling GC (1984) Population ecology of the sugar glider, Petaurus breviceps, in a system
of fragmented habitats. Australian Wildlife Research 11, 49-75.
Sutton SL, Collins NM (1991) Insects and tropical forest conservation. The Conservation of
Insects and Their Habitats. Collins NM and Thomas JA (Eds), 405-424. Academic
Press, New York.
Swihart RK, Slade NA (1984) Road crossing in sigmondin hispidius an Microtus
ochrogaster. Journal of Mammalogy 65, 357-360.
Sỳkora KV, Nijs (de) LJ, Pelmsa HM (1993) Plantengemeenschappen van Nerderlandse.
Natuurhistorische Vereninging.
Sỳkora K, Kalwij JM, Keiser PJ (2002) Phytosociological and floristic evaluation of 15-
year ecological management of roadside verges in the Netherlands. Preslia 74, 421-
436.
Tallis H, Kareiva P, Marvier M, Chang A (2008) An ecosystem services framework to
support both practical conservation and economic development. Proceedings of the
National Academy of Science of the United States of America 105, 9457-9464.
Thioulouse J, Chessel D, Dpleldec S, Olivier JM (1997) ADE-4: A multivariate analysis
and graphical display software. Statistics and Computing 7, 75-83.
Thomas SR, Noordhuis R, Holland JM, Goulson D (2002) Botanical diversity of beetle
banks: Effects of age and comparison with conventional arable field margins in
southern UK. Agriculture, Ecosystems and Environment 93, 403–412.
- 174 -
Thompson JR, Rutter AJ (1986) The salinity of motorway soils. 4. Effects of sodium
chloride on some native British shrub species, and the possibility of establishing
shrubs on the central reserves of motorways. Journal of Applied Ecology 23, 299–313.
Thompson JR, Rutter AJ, Ridout PS (1986) The salinity of motorway soils II. Distance
from the carriageway and other sources of local variation in salinity. Journal of
ThroopHL, Lerdau MT (2004) Effects of nitrogen deposition on insect herbivory:
Implications for community and ecosystem processes. Ecosystems 7, 109–133.
Thuiller W, Lavorel S, Araujo MB, Sykes MT, Prentice IC (2005) Climate change threats
to plant diversity in Europe. Proceedings of the National Academy of Sciences of the
United States of America 102, 8245-8250.
Tikka PM, Högmander H, Koski PS (2001) Road and railway verges serve as dispersal
corridors for grassland plants. Landscape Ecology 16: 659-666.
Timmins SM, Williams PA (1990) Reserve design and management for weed control.
Alternatives to the chemical control of weeds. In Bassett C, Whitehouse LJ and
Zabkiewicz JA (Eds), FRI Bulletin 155, 133–138. Ministry of Forestry, Forest
Reseach Institute, Rotorua, New Zealand.
Tischendorf L, Wissel C (1997) Corridors as conduits for small animals: attainable distances
depending on movement pattern, boundary reaction and corridor width. Oikos 79,
603-611.
Tischendorf L, Fahrig L (2000) How should we measure landscape connectivity?
Landscape Ecology 15, 633–641.
Tong STY (1990) Roadside dusts and soils contamination in Cincinnati, Ohio, USA.
Environmental Management 14, 107-114.
Trombulak SC, Frissel CA (2000) Review of ecological effects of roads on terrestrial and
aquatic communities. Conservation biology 14, 18-30.
Truscotta AM, Palmer SCF, McGowana GM, Cape JN, Smart S (2005) Vegetation
composition of roadside verges in Scotland: the effects of nitrogen deposition,
disturbance and management. Environmental Pollution 136, 109-118.
Turner MG (1980) Flora Europaea, vol.1-5. Cambridge University Press (Eds). Cambridge,
UK.
- 175 -
Tyser RW, Worley CA (1992) Alien flora in grasslands adjacent to road and trail corridors
in Glacier National Park, Montana (U.S.A.). Conservation Biology 6, 253–62.
Ullmann I, Bannister P, Wilson JB (1995) The vegetation of roadside verges with respect to
environnemental gradients in southtern New Zealand. Journal of Vegetation Science
6, 131-142.
Van der Zande AN, ter Keurs WJ, Van der Weijden WJ (1980) The impact of roads on
the densities of four bird species in an open field habitat - Evidence of a long distance
effect. Biological Conservation 18, 299-321.
Van Dyke FB, Brocke RH, Shaw HG, Ackerman BB, Hemker TP, Lindzey FG (1986)
Reactions of mountain lions to logging and human activity. The Journal of wildlife
management 50, 95–102.
Varet N (2006) Inventaire de la diversité floristique et étude d’impact de la gestion des
dépendances vertes du réseau routier seine-et-marnais. Rapport de stage de M1
Sciences, technologie and Santé, Université de Rouen. UMR 51/73
MNHN/CNRS/UPMC.
Verkaar HJ (1990) Corridors as a tool for plant species conservation? Species Dispersal in
Agricultural Habitats. Bunce RGH and Howard DC (Eds), 82– 97. Belhaven Press,
London.
Vermeulen HJW (1994) Corridor function of a road verge for dispersal of stenotopic
heathland ground beetles (Carabidae). Biological Conservation 69, 339-349.
Vickery J, Carter N, Fuller RJ (2002) The potential value of managed cereal field margins
as foraging habitats for farmland birds in the UK. Agriculture, Ecosystems and
Environment 89, 41–52.
Violle C, Navas ML, Vile D, Kazanou E, Fortunel C, Hummel I, Garnier E (2007) Let
the concept of trait be functional! Oikos 116, 882-892.
Virolainen KM, Suomi T, Suhonen J, Kuitunen M (1998) Conservation of vascular plants
in single large and several small mires: Species richness, rarity and taxonomic
diversity. Journal of Applied Ecology 35, 700-707.
Viskari EL, Kärenlampi L (2000) Roadside scots pine as an indicator of deicing salt use –
A comparative study from two consecutive winters. Water, Air and Soil Pollution 122,
405-419.
- 176 -
Vora RS (1988) Potential soil compaction forty years after logging in northeastern
California. Great Basin Naturalist 48, 117-120.
Wallin H (1985) Spatial and temporal distribution of some abondant carabid beetles
(Coleoptera : Carabidae) in cereal fiels and adjacent habitats. Pedobiologia 28, 19-34.
Wardle D, Barker G, Bonner K, Nicholson K (1998) Can comparative approaches based
on plant ecophysiological traits predict the nature of biotic interactions and individual
plant species in ecosystems? Journal of Ecology 86, 405-420.
Warner RE (1985) Demography and movements of free-ranging domestic cats in rural
Illinois. Journal of Wildlife Management 49, 340-346.
Watts RD, Compton RW, McCammon JH, Rich CL, Wright SW, Owens T, Ouren DS
(2007) Roadless Space of the Conterminous United States. Science 316, 736-738.
Way JM (1977) Roadside verges and conservation in Britain: A review. Biological
Wein RW, Wein G, Bahret S, Cody WJ (1992) Northward invading non-native vascular
plant species in and adjacent to Wood Buffalo National Park, Canada. Canadian Field
Naturalist 106, 216-224.
Weiserbs A, Jacob JJ (2001) Is breeding bird distribution affected by motorway traffic
noise. Alauda 69, 483–489.
Wester L, Juvik JO (1983) Roadside plant communities on Mauna Loa, Hawaii. Journal of
Biogeography 10, 307-316.
Westhoff V, van der Maarel E (1978) The Braun-Blanquet approach. In Whittaker RH
(Eds). Ordination and classification of plant communities. Junk. The Hague.
Weyman GS, Jepson PC, Sunderland KD (1995) Do seasonal changes in numbers of
aerially dispersing spiders reflect population density on the ground or variation in
ballooning motivation? Oecologia 101, 487–493.
Wilcox DA (1989) Migration and control of purple loosestrife (Lythrum salicaria L.) along
highway corridors. Environmental management 13, 365-370.
Wilson DS (1992) Complex interactions in metacommunities, with implications for
biodiversity and higher levels of selection. Ecology 73, 1984-2000.
Witmer GW, DeCalesta DS (1985) Effect of forest roads on habitat use by Roosevelt elk.
Northwest Science 59, 122-125.
- 177 -
Woodcock BA, Westbury DB, Potts SG, Harris SJ, Brown VK (2005) Establishing field
margins to promote beetle conservation in arable farms. Agriculture, Ecosystems and
Environment 107, 255–266.
Yanes M, Velasco J, Suarez F (1995) Permeability of roads and railways to vertebrates: The
importance of culverts. Biological Conservation 71, 217-222.
Yoccoz NG, Nichols JD, Boulinier T (2001) Monitoring of biological diversity in space and
time. Trends in Ecology and Evolution 16, 446-453.
Zink TA, Allen MF, Heindl-Tenhunen B, Allen EB (1995) The effect of a disturbance
corridor on an ecological reserve. Restoration Ecology 3, 304-310.
- 178 -
RESUME / SUMMARY
Les routes sont des infrastructures essentielles de nos paysages et ont de nombreux impacts
sur l’environnement. Plusieurs études ont cependant montré que leurs dépendances vertes
pouvaient jouer un rôle bénéfique pour la biodiversité au sein des paysages agricoles où
l’intensification des pratiques depuis soixante ans a mené à sa forte raréfaction. Dans ce
contexte, nous avons étudié les rôles « habitat » et « corridor » des bords de route pour les
communautés de petits mammifères et de plantes. Nous avons tout d’abord retrouvé un
certains nombres d’effets négatifs associés aux routes, comme une homogénéisation biotique
des milieux adjacents, une pollution importante de l’environnement proche et un effet barrière
pour certaines espèces animales. Nous avons surtout montré que les effets « refuge pour la
biodiversité » et « corridors de continuités biologiques » étaient bien une réalité associée aux
dépendances vertes des bords de route en milieu agricole. Nous avons aussi montré que ces
effets dépendaient de paramètres comme les modes de gestion (fauche), les dimensions de
« l’emprise verte », ou encore, la présence de haies ; et qu’ils permettaient le maintien de
services écosystémiques locaux. L’intérêt écologique des bords de route en milieu agricole
intensif a donc été mis en évidence tout en montrant que certaines mesures politiques
pouvaient être instaurées pour une meilleure prise en compte de la biodiversité originale de
ces espaces marginaux représentant 2% de l’occupation de sols. Un document pratique de
conseils de gestion a été élaboré à destination des gestionnaires de routes et de leurs
dépendances vertes.
MOTS-CLEF : Agriculture intensive, Anthropisation des milieux, Biologie de la Conservation,

Changements globaux, Communautés végétales, Corridors, Homogénéisation biotique,
Nature ordinaire, Politiques de gestion, Traits fonctionnels, Services écosystémiques.
Roads are essential components of our landscapes and they have important negative effects
on biodiversity. However studies have shown that their verges may provide a positive effect
for biodiversity in agrarian landscape where management intensification of the last sixty
years leaded to the high rarefaction of biodiversity. We studied the possible functions of
“habitat” and “corridor” of roadside verges for small mammals and plant communities in this
agrarian context. First we rediscovered numbers of negative road effects as biotic
homogenization of adjacent habitats, pollution of environment and barrier effects for some
mammal species. We then showed that “refuge” and “corridor” effects associated to verges
were a reality in agrarian landscapes. We also showed that number of anthropogenic
parameters could impact those effects: roadside management (cutting), verge widths and
hedgerows. We found that biodiversity of roadsides was able to maintain some local
ecosystem services. Ecological interests of roadsides have been demonstrated with evidences
that management policies could protect biodiversity of such marginal areas representing more
than 2% of the landcover.
KEY-WORDS: Anthropization; Biotic homogenization; Common nature; Conservation

biology; Corridors; Ecosystem services; Functional traits; Global changes; Intensive
agriculture; Management policies; Plant communities; Reconciliation ecology.

These Redon

Transféré par

Droits d'auteur :

Formats disponibles

These Redon

Transféré par

Informations du document

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

These Redon

Transféré par

Droits d'auteur :

Formats disponibles

MUSEUM NATIONAL D’HISTOIRE NATURELLE

Ecole Doctorale Sciences de la Nature et de l’Homme – ED 227

Année 2008 N°attribué par la bibliothèque

Pour obtenir le grade de

DOCTEUR DU MUSEUM NATIONAL D’HISTOIRE NATURELLE

Présentée et soutenue publiquement par

Louis de REDON de COLOMBIER

INTERETS ECOLOGIQUES DES BORDS DE ROUTE

Sous la co-direction de : Nathalie MACHON, Professeur,

Pr. Patrick BLANDIN Professeur MNHN, Paris, FRANCE Président

Pr. Thierry DUTOIT Professeur, IUT d’Avignon, FRANCE Rapporteur

Dr. Päivi TIKKA Research coordinator, Université d’Helsinki, FINLANDE Examinatrice

Pr. Nathalie MACHON Professeur MNHN, Paris, FRANCE Co-directrice de Thèse

INTERETS ECOLOGIQUES DES BORDS DE ROUTE EN MILIEU AGRICOLE INTENSIF

Pour obtenir le grade de

Soutenue le 12 décembre 2008 devant un jury composé de :

Patrick BLANDIN Professeur MNHN Président

ECOLOGICAL INTERESTS OF ROADSIDE VERGES IN INTENSIVE AGRARIAN LANDSCAPES

To obtain the academic degree of:

Defended on December 12th, 2008 in front of a scientific jury composed by:

Patrick BLANDIN Professor MNHN Président

« Nous avouerons que notre héros était

L’ensemble des ressources d’information géographiques et cartographiques utilisées a été

Dans le cadre de l’Atlas de la biodiversité de Seine-et-Marne réalisé par le laboratoire, le

p. 2 § A. INFRASTRUCTURES ROUTIERES ET CRISE DE LA BIODIVERSITE

p. 4 § B. ETUDE DE LA NATURE ORDINAIRE ET BIOLOGIE DE LA CONSERVATION

p. 6 § C. LES BORDS DE ROUTE, ZONES CREATRICES D’HABITAT ET DE CONTINUITE POUR

p. 9 § D. IMPACTS DES MODES DE GESTION ET DE LA STRUCTURE DES BORDS DE ROUTE

p. 12 MATERIELS & METHODES

p. 16 § B. STRUCTURES ET EMPRISES DES ROUTES AU SEIN DES PAYSAGES AGRICOLES

BIODIVERSITE DES BORDS DE ROUTE EN MILIEU AGRICOLE INTENSIF :

STRUCTURES ET MODES DE GESTION DES ACCOTEMENTS ROUTIERS :

p. 77 DISCUSSION ET CONCLUSION GENERALES

p. 78 § A. EFFETS DES PROTOCOLES UTILISES ET DES GROUPES SUIVIS

p. 82 § B. IMPACTS NEGATIFS DES ROUTES SUR L’ENVIRONNEMENT

p. 84 § C. IMPORTANCE DES BORDS ROUTES EN MILIEU AGRICOLE INTENSIF

p. 86 § D. IMPACTS DES STRUCTURES ROUTIERES

p. 90 § E. IMPACTS DES MODES DE GESTION

p. 95 § A. POURSUITE L’ETUDE DES EFFETS DES ROUTES SUR LA BIODIVERSITE

p. 95 § B. COMMUNICATION ET VULGARISATION DES SAVOIRS SCIENTIFIQUES

p. 145 BIBLIOGRAPHIE GENERALE

“Two roads diverged in a wood, and I —

(« Deux routes se séparaient dans un bois,

§ A. INFRASTRUCTURES ROUTIERES ET CRISE DE LA BIODIVERSITE

(1) Ampleur et réalité de la crise

(2) Infrastructures routières et causes de la crise

§ B. INTERET DES BORDS DE ROUTE POUR LA BIOLOGIE DE LA CONSERVATION

« Quelle biodiversité conserver ? », « Comment la quantifier ? », sont des questions

(1) Conservation de la biodiversité et Nature ordinaire

(2) Développement de méthodes de suivis de la Nature ordinaire

C. LES BORDS DE ROUTE, ZONES CREATRICES D’HABITAT ET DE CONINUITE POUR LA

(1) Bords de route et création d’habitats en paysage agricole intensif

(2) Bords de route et création de continuités biologiques en paysage agricole intensif

Q2.A. Les bords de route permettent-ils la mise en place de continuités

L’étude des infrastructures routières comme créatrices de continuités biologiques au sein du

La première partie de la thèse, intitulée « Biodiversité des bords de route en milieu

§ D. IMPACTS DES MODES DE GESTION ET DE LA STRUCTURE DES DES BORDS DE ROUTE

(1) Fauchage des bords de route et diversité végétale

La deuxième partie de la thèse, intitulée « Structures et modes de gestion des

(2) Encadrés et Guide technique