Journal of Coastal Research SI 56 39 - 43 ICS2009 (Proceedings) Portugal ISSN 0749-0258

Preliminary Results of Exoscopic Analysis of Quartz Grains Deposited

by a Palaeotsunami in Salgados Lowland (Algarve, Portugal)
P. J. M. Costa†, C. Andrade†, M. C. Freitas†, M. A. Oliveira† and J.-M.Jouanneau ‡

† Centro de Geologia & Departamento de Geologia, Faculdade de Ciências da Universidade de Lisboa, Lisboa, 1749-016, Portugal
‡Dép. Géologie et Océanographie, UMR5805,CNRS, Univ. Bordeaux I, Talence, France.
{ppcosta@fc.ul.pt; candrade@fc.ul.pt; cfreitas@fc.ul.pt; alexandraoliv@gmail.com; jm.jouanneau@epoc.u-bordeaux1.fr}

ABSTRACT

COSTA, P. J. M., ANDRADE, C., FREITAS, M. C., OLIVEIRA, M. A. and Jouanneau, J.-M., 2009. Preliminary results
of exoscopic analysis of quartz grains deposited by a palaeotsunami in Salgados lowland (Algarve, Portugal).
Journal of Coastal Research, SI 56 (Proceedings of the 10th International Coastal Symposium), 39 – 43. Lisbon,
Portugal, ISSN 0749-0258.

The recognition of coastal inundations by tsunamis beyond the historical record depends solely on the
identification of characteristic ancient deposits and the stratigraphical record can provide crucial information on
recurrence and patterns of coastal flooding and/or specificities of a particular event. A palaeotsunami deposit was
found in the Salgados lowland (south coast of the Algarve, Portugal) and correspondent to the AD 1755 event
was studied using a scanning electron microscope (SEM)-based analysis of quartz grain microtextures (e.g.
grooves, precipitation/dissolution, marks, fractures, relief, roundness, craters, upturned plates) to investigate the
existence of specific signatures and to compare tsunami-deposited sand particles with grains of possible sources
in coastal to nearshore depositional contexts. The preliminary results suggest that in this specific location, the
AD 1755 tsunami did not craft an obvious microtextural signature in sand particles. Instead, the grains deposited
by this event were essentially characterized by source-inherited features and we suggest that coastal foredunes
may have been a relevant source.

ADITIONAL INDEX WORDS: Tsunami deposits, provenance, microtextures, SEM

tsunami run-up and nature of source sediment (inner shelf to

INTRODUCTION littoral) in any coastal setting. Consequently, the possible
Over the last three decades the study of palaeotsunamis and variations in sedimentary processes and products during these
palaeostorms has increased astonishingly. This study tries to complex events remain poorly understood. Sediment deposited
contribute to this effort by analyzing the sedimentological during tsunami run-in is generally recognized as reworked shallow
signature of tsunamis in a coastal lagoon (Lagoa dos Salgados, marine or beach/foredune material found in a stratigraphically
Algarve, South Portugal). The focus of this research is a detailed narrow band embedded in a landward low-energy depositional
exoscopic analysis of quartz grains to discriminate paleotsunami environment Moreover, the subsequent backwash is also being
samples and to establish their provenance. Other techniques used regarded as a process of significant geomorphic and
include stratigraphic description, granulometry and dating sedimentologic consequence (DAWSON, 1994; DAWSON, 1999;
techniques (14C, 210Pb and 137Cs). HINDSON and ANDRADE, 1999; DAWSON and SHI, 2000).
A group of criteria to identify palaeotsunami deposits have been
summarized by GOFF et al. (1998) and DAWSON and STEWART
Palaeotsunami deposits (2007). However, the key-problem of distinguishing between
From a geological viewpoint, tsunamis are notable as short lived storm and tsunami deposits. relevant in reconstructing
but extremely powerful agents with a very complex pattern of chronologies of coastal hazards, is still a matter of debate and the
erosion and deposition, leaving large volumes of sediment on the studies addressing this problem (e.g. CLAGUE et al., 2000,
seafloor and coastal areas (BONDEVIK et al., 1997). Palaeotsunami NANAYAMA et al., 2000, GOFF et al., 2004, MORTON et al., 2007,
recognition depends on the identification of ancient tsunami KORTEKAAS and DAWSON, 2007) remain to a large extent site
deposits. Tsunami deposition is most commonly characterized by specific.
the re-deposition of coarse shallow marine or coastal sediments in
a terrestrial environment and recognition of these deposits is the
primary field method of measurement of run-up height and run-in Study Area
distance, although patterns of erosion and deposition by both A group of coastal areas, in the southwestern part of the
landward- and seaward-directed flow is complex (BONDEVIK et Algarve, Portugal, are being studied with the purpose of detecting
al., 1997; DAWSON and SHI, 2000). Understandably, the nature of and analyzing tsunami deposits. Although rarely subjected to
tsunami deposits varies greatly with coastal topography, height of extreme storms or tsunamis, the Algarve coast is the most
vulnerable area in terms of tsunami risk in Portugal, given its

Journal of Coastal Research, Special Issue 56, 2009

39
 Costa et al.

Figure 1A, 1B and 1C. Location of study area (Lagoa dos Salgados). 1D- Nearshore (NS) sample locations. 1E- Surface samples, core
locations and thickness of the tsunami deposit (SG14 – core sample; B – berm; BF – beach face; C – channel; D – dune; DC – dune crest).

development, location relative to the Azores-Gibraltar plate channel) were also collected. Visual description of sediment was
boundary and neotectonic- and gravity-induced sea-floor activity conducted focusing in color, lithology, texture, erosional features
offshore the Gulf of Cadiz. The mean tidal range in the area is and macrofossil content. Logs were plotted and correlations were
approximately 2.1 m and ca. 3 m during spring tides, but during established. A thin layer of coarse sediment was detected within
extreme tides the high water level may exceed 2 m above MSL Late Holocene low-energy sediments of Salgados lowland with
(ANDRADE and HINDSON, 1999). The wave climate in the Algarve many of the stratigraphical, textural and compositional diagnostic
is one of low-energy. On average, once every winter a SW storm criteria used to recognize tsunami deposits (Figure 1); in a smaller
raises waves with significant height of about 3 m and mean period group of cores from this lowland, a second and younger high-
of 7-8 s (CAPITÃO, 1992). energy event layer with a controversial origin was also detected.
The Salgados lowland is located in the bay between Armação de
Pêra and Galé (FIGURE 1A, 1B, 1C). This coast features a 6 km- Exoscopy
long intermediate-reflective sand beach backed by a continuous, 3 The use of scanning electron microscope (SEM) in sedimentary
to 17 m-high vegetated foredune, the latter covering cemented studies was initiated by BIEDERMAN (1962) and KRINSLEY et al.
Pleistocene-Holocene aeolianites and beachrock (PINTO et al., (1964) and developed to study microtextures in quartz grains by
2001). The continuity of the beach-dune system is interrupted by KRINSLEY and DOORNKAMP (1973). Since then, a considerable
the ephemeral inlets of two infilled lagoons, which developed in number of studies have been produced in this subject. However,
relation with the outlet of intermittent streams: the Alcantarilha this technique has not been consensual in the scientific community
lagoon and stream, and the Salgados lagoon and Espiche stream, due to a certain degree of subjectivity in the visual analysis of
which drain, respectively, 204 and 41 km2 watersheds, mostly microtextures and conflicting interpretations of the surface
developed in Early Miocene limestone and Late Miocene siltstone features. The use of exoscopic analysis has been focusing in
and sandstone. The Salgados Lagoon extended across some 1.5 establishing the provenance of the quartz grains (e.g. NIETER and
km2 but about ½ of this surface has been reclaimed and landfilled KRINSLEY, 1976; LE RIBAULT, 1977; MAHANEY, 1988; MAZZULLo
for a golf course, which occupies its eastern section (FIGURE 1C). et al., 1988; MORAL CARDONA et al., 1997; BRUZZI and PRONE,
The remnant surface is a flat-floored depression developing about 2000; MAHANEY, 2002; PARIS et al; 2007) and all the studies on
1.1 to 1.7 m above mean sea level, collecting water and muddy quartz grains microtextures were based on visual description, thus
sediment from the adjacent watershed and is usually flooded by increasing the degree of subjectivity. So far, only WILLIAMS et al.
about 1 m of water. Water addition by rain and marine overwash (1998) proposed a quantitative method for characterizing and
may rapidly increase the water level and expand the wet perimeter classifying the shape and features of quartz grains by
until the dune line and the inlet is artificially opened to prevent mathematical analysis of surface textures.
damaging of the golf facilities; seasonal drought conditions
leading to eutrophication, may also require additional re-opening
of the inlet to allow for water renewal. In any case, the inlet closes METHODS
naturally in a matter of weeks by wave action. Twelve grains from the tsunami layer referred above (here
Hundred and forty three short sediment cores were obtained, in represented by sample SG TSU obtained in core SG 14) and 12
Lagoa dos Salgados (Figure 1), using hand-operated gauge and grains from the controversial deposit (referred to in this study as
Edelman augers, extension rods and a Van der Staay suction corer. SG_X) were analyzed with SEM. In addition, 490 quartz grains,
Moreover, 30 surface samples (nearshore, beach, dune and inlet from 5 different sedimentary environments around and within

Journal of Coastal Research, Special Issue 56, 2009

40
 Costa et al.

Figure 2. SEM images of quartz grains from Lagoa dos Salgados. Figure 3. 210Pb excess and 137Cs depth profile from core SG48.
A – SG_X controversial sample; B – SG TSU tsunami sample; C
– Dune sample; and D – Beach sample.
which is the near vicinity of the maximum nuclear testing in the
northern hemisphere.
Lagoa dos Salgados (e.g nearshore, beach face, beach berm, dune The statistical analysis of the microtextures using Principal
and dune crest), all sharing potential to constitute source for the Component Analyses indicate that two sets of characteristics can
tsunami deposit were also studied. A minimum of 12 quartz grains be identified (FIGURE 4) and that they are orthogonal. One set
per sample (FIGURE 2) were collected from the (previously sieved) consists of features associated with mechanical reworking of
125-500µm size-fraction and these grains were considered as grains and energy level (roundness, fresh surfaces, sharp edges,
representative in what concerns exoscopic properties. The bulk abrasion marks, upturned plates, linear fractures, conchoidal
samples were washed with tap water and wet-sieved with 0.5  fractures, parallel fractures and steps) and the other includes
interval. Quartz grains were analyzed under the binocular features related with chemical reworking (precipitation,
microscope and prepared for SEM. This involved mineralization dissolution, linear grooves, circular grooves, deep grooves and
of the grains (coating), either with gold or carbon. The grains were relief); however, a small number of mechanically-borne
them taken to the SEM lab and photographs obtained. Two SEM characteristics (craters, V-marks and crescent marks) plot together
were used: JEOL JSM 5200 LV (FCUL, Lisbon) and JEOL JSM- and directly correlate with the chemical features. The features
5910 LV (Univ. Clermont-Ferrand). After this, a visual analysis of adhering particles and cracked grains showed poor discriminating
each grain was performed including the careful inspection of potential.
microtextures in grain surfaces and their classification according No peculiar microtexture feature or association of features of
to established criteria. A list of characteristic microtextures was grains, exclusive of tsunami SG TSU sand was found, preventing
compiled based on selected references (e.g. MAHANEY, 2002) and clear-cut discrimination with other deposits. Similarities between
in the authors criteria; each grain was described using all samples were investigated using bivariate plots of features versus
microtextural features and the relative performance of each features and features versus sedimentary environments. The plot
characteristic was rank-ordered from “high” to “low”. In this of features versus a or b-axis of grains indicates that the former
classification scheme, roundness (according to Power’s scale) are not influenced by grain size, in agreement with the findings of
varies from 1 (well rounded) to 6 (very angular) and was BRUZZI and PRONE (2000). The plot of all test-samples from
evaluated by comparison with a reference chart; relief, dissolution distinct environments in bivariate diagrams are scattered and
and precipitation vary in the range of 1 (low) to 3 (high) and rank significant intersection occurs between fields representing
ordering depended on visual appreciation. All the remaining different depositional environments. FIGURE 5 presents one
characteristics (e.g. fresh surfaces, steps, fractures, upturned example showing that overlap between different environments
plates, grooves, craters, abrasion marks, crescent marks, v-marks) varies from app. 30% (e.g. dune) to app. 65% (e.g.
were classified according to the percentage of area occupied by channel).Moreover, it was noted that the SG-X and SG-TSU
features in the grain (e.g. 0 - not present to 5 - dominant >75% of samples do not systematically cluster with any other specific
the grain surface occupied by the microtexture analyzed). environment. In addition, significant scattering occurs when all
Average and median values for each variable characterizing each grains in one sample are plotted, instead of the representative
set of grains were calculated after normalization by the largest median value. However, the enigmatic SG_X sample plots close
value. to the tsunami sand in the majority of bivariate diagrams. The
features of the tsunami sample in bivariate plots are consistent
RESULTS with a wide range of possible sources: in fact, if a small number of
Age constraining of the coarse layer found within low-energy attributes is carefully selected, it is possible to suggest strong
deposits of Salgados was attempted using 210Pb and 137Cs affinity towards a specific sedimentary environment. However, if
(Universitée de Bordeaux). Results are consistent and coherent a different set of attributes is selected, grains from other
with the AD 1755 tsunami event, which is known to have environments may show stronger resemblance with the tsunami
extensively impacted this area (e.g. PEREIRA DE SOUSA, 1919). The sample.
accumulation rate derived from the vertical profile of 210Pb excess Notwithstanding the above limitations, our results suggest that
(2.6 mm/yr) suggests that low-energy deposits resting upon the the SG 14 tsunami sample exhibits the highest values of
coarse (tsunamigenic) layer are younger than AD 1840 (FIGURE dissolution and grooves, both deep and linear. Adhering particles
3). Moreover, if that rate is used to calculate the age of the were less represented in beach samples. Relief tends to be higher
maximum value of 137Cs in the core, the date obtained is AD 1961, in beach and palaeotsunami grains, which also showed higher and
more recent steps and fractures, just as grains from the inlet

Journal of Coastal Research, Special Issue 56, 2009

41
 Costa et al.

Figure 5. Bivariate plot of abrasion marks vs dissolution of

Lagoa dos Salgados samples. Discrepancy between studied
samples and symbols is apparent due to overlapping.

significant particle reworking; in fact, these high-energy events

should essentially suspend sand particles during transport,
Figure 4. Diagram of characteristics of quartz grains from therefore reducing the effectiveness of mechanical grain to grain
Lagoa dos Salgados. interaction. In this case, the prevailing microtextural signature of
tsunami grains should match the source material and microtextural
signatures of the pre-transport setting or inherited from previous
channel. Sharp edges and craters present the higher values in
sedimentary cycles should be to a large extent preserved.
both palaeotsunami and dune samples.
Results are essentially characterized by ubiquity and this might
be a consequence of the Salgados deposit having been assembled
DISCUSSION from multiple sources. A careful revision of the results makes this
Results obtained did not allow the definition of any hypothesis tenuous: the grains of SG_TSU are distinct in the
unmistakable microtextural characteristic or set of characteristics neatness and abundance of chemical reworking features and rarity
exclusive of palaeotsunami deposits that could be associated with of adhering particles (including microfossil carapaces); this
the processes of particle entrainment and transport by a tsunami; suggests a long residence period with limited mobility prior to
furthermore, the bivariate plots and multiple correlation methods entrainment by the tsunami and only the dune and nearshore
used showed poor ability in discriminating grains deposited by (beyond the closure depth) environments fulfill this condition.
tsunami and also in separating distinct nearby coastal Actually, grains from the nearshore show resemblances with the
environments. This can be caused by methodological or tsunami particles but include abundant adhering microfossil tests,
sedimentary reasons. For instance, the representativeness of the whereas inshore particles are virtually free of these biological
number of grains analyzed in this study might not be sufficient to additions, as expectable, given the high wave-induced turbulence
reduce the effects of intra- and inter-sample variability of a given and mobility. The reasoning above implies significant preservation
deposit; the statistical tools may not be adequate to deal with the of the assemblage of inherited features; accordingly, we can
nature of the selected variables, which are in essence qualitative hypothesize that foredunes should have been a relevant (and
and were arbitrarily rank-ordered. Further analyses are being eventually, primary) source for the tsunami sand at Salgados.
conducted to overcome these issues, namely increasing the Given the discussion above and resemblance (mostly in chemical
number of test-grains per sample and switching to discriminating features) shown by the controversial sample it is reasonable to
analysis or to classification trees. accept that grains of the SG_X deposit were also remobilized from
At the present state of knowledge, “blind” numerical methods the dune, but a tsunamigenic mechanism of transport and
may conveniently be replaced by qualified analysis and emplacement cannot be readily deduced. The inferences presented
interpretation of results, which is what we attempt further below. above can be challenged in the case of extensive post-depositional
BRUZZI and PRONE (2000) investigated exoscopic features of weathering of the quartz grains. We believe that in the case
quartz grains transported by storm (Rhone delta) and tsunami studied here, the extension of chemical reworking observed in
waves (Boca do Rio, Algarve, AD 1755) and were extremely tsunami grains is hardly compatible with the short time-interval
cautious in concluding on the ability to discriminate between both elapsed since AD 1755; the study of precedence between
transport processes. They found microtextures associated with mechanical and chemical microtextures in grains bearing both
mechanical impacts (e.g. v-marks), exfoliation of parts of the grain features did not clearly support a post-depositional reworking
surface and orientation of groups of fractures in both high-energy model.
deposits; Paris (pers. comm.) studied samples from the 2004 Investigating the spatial variability in microtextures of grains
tsunami in Indonesia and also found some fresh microtextures deposited by the same tsunami event in different locations of the
characterizing these grains. However, the inexistence in both cases same coast and of grains from adjacent sources may contribute to
of comparison with possible source materials precludes the unravel the issues previously discussed. We compared quartz
assignment of the microtextures either to the stock previously grains from the 1755 event in two other locations of the Algarve
existing in the coastal zone or to mechanical reworking during coast (Martinhal and Boca do Rio) and preliminary results suggest
transport. The short-distance traveled by tsunami grains and that spatial variability in the microtextures of the source areas is
uniqueness of the event may explain the inability to produce retained by the tsunami-deposited grains.

Journal of Coastal Research, Special Issue 56, 2009

42
 Costa et al.

CONCLUSIONS HINDSON, R. A. and ANDRADE, C., 1999. Sedimentation and

This research analyzed the sedimentological signature of a hydrodynamic processes associated with the tsunami
tsunami deposit in a coastal lagoon (Lagoa dos Salgados, Algarve, generated by the 1755 Lisbon earthquake. Quaternary
South Portugal). The techniques used heavily relied upon detailed International, 56, 27-38.
exoscopic analysis of quartz grains microtextures. The purpose of KORTEKAAS, S. and DAWSON, A. G., 2007. Distinguishing tsunami
the research is to compare and distinguish different transport and and storm deposits: An example from Martinhal, SW Portugal.
depositional contexts and to sort a signature of each sedimentary Sedimentary Geology, 200, 3-4, 208-221.
event/deposit and environment relying upon SEM microtextures KRINSLEY, D. H. and DOORNKAMP, J. C., 1973. Atlas of quartz
of quartz grains. sand surface textures. Cambridge University Press, 91p.
The preliminary results for the Salgados lowland suggest that, KRINSLEY, D.; TAKAHASHI, T.; SILBERMAN, M. L. and NEWMAN,
with the methodology used and for this specific location, the W. S., 1964. Transportation of sand grains along the Atlantic
grains deposited by the AD 1755 tsunami were not clearly shore of Long Island, New York: An application of electron
differentiated from the source material, suggesting that the microscopy. Marine Geology, 2, 100-120.
tsunami did not craft an obvious microtextural sin-depositional LE RIBAULT, L. L., 1977. L’Exoscopie des Quartz. Ed. Masson,
signature and that coastal foredunes may have been a relevant 150p.
source. The study of grains deposited by the same tsunami MAHANEY, W. C., 2002. Atlas of sand grain surface textures and
elsewhere in the Algarve coast suggests that they are also applications. Oxford University Press, 237 p.
essentially characterized by source-inherited features. MAHANEY, W. C., 1998. Scanning electron microscopy of
Generalization of this result to every other coastal context may Pleistocene sands from Yamal and Taz peninsulas, Ob River
complicate our ability to distinguish multiple waves from multiple Estuary, northwestern Siberia. Quaternary International, 45-
source events in coastal sedimentary sequences. 46, 49-58.
The results stress the need for more research in this technique MAZZULLO, J.; LESCHAK, P. and PRUSAK, D., 1988. Sources and
and currently the dataset used in this study is being enlarged and distribution of late Quaternary silt in the surficial sediment of
compared with deposits from the Storegga tsunami, with deposits the northeastern continental shelf of the United States. Marine
from the 2004 tsunami and with storm deposits from the Hebrides Geology, 78, 241-254.
Islands, Scotland. MORAL-CARDONA, J. P.; BELLÓN, A. S.; LÓPEZ-AGUAYO, F. and
CABALLERO, M. A., 1996. The analysis of quartz grain surface
features as a complementary method for studying their
LITERATURE CITED provenance: the Guadalete River Basin (Cádiz, SW Spain).
BIEDERMAN, E. W., 1962. Distinction of shoreline environments in Sedimentary Geology, 106, 155-164.
New Jersey. Journal of Sedimentary Research, 32, 2; 181-200. MORTON, R. A.; GELFENBAUM, G. and JAFFE, B. E., 2007.
BONDEVIK, S.; SVENDSEN, J. I. and MANGERUD, J., 1997. Tsunami Physical criteria for distinguishing sandy tsunami and storm
sedimentary facies deposited by the Storegga tsunami in deposits using modern examples. Sedimentary Geology, 200,
shallow marine basins and coastal lakes, western Norway. 184-207.
Sedimentology, 44, 1115-1131. NANAYAMA, F.; SHIGENO, K.; SATAKE, K.; SHIMOKAWA, K.;
BRUZZI, C. and PRONE, A., 2000. A method of sedimentological KOITABASHI, S.; MIYASAKA, S. and ISHII, M., 2000.
identification of storm and tsunami deposits: Exoscopic Sedimentary differences between the 1993 Hokkaido-nansei-
analysis, preliminary results. Quaternaire, 11, 3-4, 167-177. oki tsunami and the 1959 Miyakojima typhoon at Taisei,
CAPITÃO R. (1992). Wave climatology of the Portuguese coast. southwestern Hokkaido, northern Japan. Sedimentary
Clima de agitação marítima na costa portuguesa. Unp. Tech. Geology, 135, 255-264.
Rep. NATO PO-Waves TFOM 10/92, Instituto Hidrográfico, NIETER, W. M. and KRINSLEY, D. H., 1976. The production and
Lisboa, 14 p. recognition of aeolian features on sand grains by silt abrasion.
CLAGUE, J. J.; BOBROWSKY, P. T. and HUTCHINSON, I., 2000. A Sedimentology, 23, 5, 713.
review of geological records of large tsunamis at Vancouver PARIS, R.; LAVIGNE, F.; WASSMER, P. and SARTOHADI, J., 2007.
Island, British Columbia, and implications for hazard. Coastal sedimentation associated with the December 26, 2004
Quaternary Science Reviews, 19, 849-863. tsunami in Lhok Nga, west Banda Aceh (Sumatra, Indonesia).
DAWSON, A. G. and STEWART, I., 2007. Tsunami deposits in the Marine Geology, 238, 93-106.
geological record. Sedimentary Geology, 200, 166–183. PEREIRA DE SOUSA, F.L., 1919. O terremoto do 1º de Novembro de
DAWSON, A.G. and SHI, S., 2000. Tsunami deposits. Pure and 1755 em Portugal e um estudo demográfico: Distritos de
Applied Geophysics, 157, 875-897. Faro, Beja e Évora. TIPOGRAFIA DO COMÉRCIO, 61, 62 and 90p.
DAWSON, A. G., 1999. Linking tsunami deposits, submarine slides WILLIAMS, A. T.; WILTSHIRE, R. J. and THOMAS, M. C., 1998.
and offshore earthquakes. Quat. International, 60, 119-126. Sand grain analysis--image processing, textural algorithms and
DAWSON, A. 1994. Geomorphological effects of tsunami run-up neural nets. Computers & Geosciences, 24, 111-118.
and backwash. Geomorphology, 10, 1-12.
GOFF, J.; MCFADGEN, B. G. and CHAGUE-GOFF, C., 2004.
Sedimentary di¡erences between the 2002 Easter storm and the ACKNOWLEDGEMENTS
15th-century Okoropunga tsunami, southeastern North Island, This work has been supported by projects GETS (PTDC/CTE-
New Zealand. Marine Geology, 204, 235-250. GEX/65948/2006) financed by FCT (Fundação para a Ciência e
GOFF, J.R.; CROZIER, M.; SUTHERLAND, V.; COCHRAN, U.; SHANE, Tecnologia) and NEAREST (EU-037110-GOCE-2006). P. Costa
P., 1998. Possible tsunami deposit from the 1855 earthquake, benefited from a PhD grant from FCT. Thanks to R. Paris for
North Island, New Zealand. In: Stewart, I.S., Vita-Finzi, C. SEM facilities at Clermont-Ferrand, France, to S. Teixeira
(ed.), Coastal Tectonics. Geological Society of London, (CCDR-Algarve) for the nearshore samples and to R. Taborda and
Special Publications, Special Issue 146, 353–374. A. Silva (FCUL) for discussions of statistical interpretations.

Journal of Coastal Research, Special Issue 56, 2009

43