The Effects of LongTerm Grazing Exclosures on Range Plants Haymana

Environ Manage (2007) 39:326-337 DOI 10.1007/s00267-005-0392-y The Effects of Long-Term Grazing Exclosures on Range Plants in the Central Anatolian Region of Turkey Hüseyin K. Fırıncıoğlu Æ Steven S. Seefeldt Æ Bilal Şahin Received: 15 December 2005 / Accepted: 3 August 2006
Springer Science+Business Media, Inc. 2007 Abstract Over the last fifty years, almost half of the steppe rangeland in the Central Anatolian Region of Turkey (CAR) has been converted to cropland without an equivalent reduction in grazing animals. This shift has led to heavy grazing pressure on rangeland vegetation. A study was initiated in June 2003 using 6 multiscale Modified-Whittaker plots to determine differences in plant composition between areas that have not been grazed in 27 years with neighboring grazed plant communities. A total of 113 plant species were identified in the study area with the ungrazed plots containing 32 plants more than the grazed plots. The major species were Astragalus acicularis, Bromus tomentellus, Festuca valesiaca, Genista albida, Globularia orientalis, Poa bulbosa, and Thymus spyleus ssp rosulans. Grazing impacts on forbs were more pronounced than for grasses and shrubs. Based on Jaccard’s index, there was only a 37% similarity of plant species between the two treatments. Our study led to four generalizations about the current grazing regime and long-term exclosures in the steppe rangeland around the study area: (1) exclosures will increase H. K. Fırıncıoğlu (&) The Central Research Institute for the Field Crops, P.O. Box 226, Ulus Ankara, Turkey e-mail: huseyin@tr.net ; hkansur@yahoo.com S. S. Seefeldt United States Department of Agriculture-Agriculture Research Service, Room 355, O’Neill Bldg, University of Alaska Fairbanks, SubArctic Agricultural Research Unit, Fairbanks, Alaska 99775, USA B. Şahin Department of Biology, Gazi University, Beş evler, Ankara, Turkey 123 species richness, (2) heavy grazing may have removed some plant species, (3) complete protection from grazing for a prolonged period of time after a long history of grazing disturbance may not lead to an increase in desirable plant species with a concomitant improvement in range condition, and (4) research needs to be conducted to determine how these rangelands can be improved. Keywords Biodiversity
Diversity
Exclosure
Grazing
Modified-Whittaker plots
Steppe rangelands
Central Anatolia Introduction Turkey has 13.1 million hectares of natural grazing land, 33% of which is located in the Central Anatolian Region (CAR) (Anonymous 2001). Over the last fifty years, vast rangelands have been converted to cropland in Turkey. It was estimated that in 1970, there was a total of 6.2 million hectares of grazed rangeland (Anonymous 1980); by 2001 this area had declined to 3.3 million hectares (Anonymous 2001). The reduction of rangeland has not been linked to an equivalent reduction in grazing animals, thus remaining rangelands are heavily grazed. Turkey is endowed with a rich diversity of families, genera, and species of plants (163 families, 1225 genera, 9000 species) with 3000 plant species endemic to the area (Tan 1998). As a significant source of the world’s genetic resources and plant biodiversity, the consequence of increased grazing pressure on Turkey’s plant diversity is of great interest. Grazing intensity and timing are important factors affecting plant diversity (Laycock 1967). Excessive and Environ Manage (2007) 39:326-337 heavy grazing may cause plant mortality, resulting in the loss of some individual plant species, through weakening plants and disrupting regeneration, and resulting in the encroachment of non-palatable plant species that can persist under heavy grazing. Therefore, implementation of sound grazing management is very important for the conservation of individual plant species, plant populations, and ecosystems. In some severely overgrazed areas, destocking may be necessary. However, complete protection from grazing is not a normal condition for range plants in Turkey, which evolved with grazing animals, both wild and domestic. Under complete protection from grazing, some undesirable changes may occur, such as an excessive litter accumulation that will change the habitat enough to reduce or eliminate many native species in the area (Michunas and others 1988, Bakır 1998, Adler and others 2004). In the CAR, the steppe rangelands contain many grass, forbs, and shrub species. This steppe vegetation, which is bordered by more woody vegetation of the Anatolian territory, has been exploited for millennia, especially through grazing and intensive agricultural activities (Akman and others 1984). Understanding the effects of grazing and non-grazing on the dynamics of the herbaceous communities of CAR is important in formulating rational management plans for both conservation and sustainable animal production. It is well recognized that comparisons between grazed and ungrazed grasslands have been an important tool in determining the effects of grazing (Weaver and Rowland 1952, Fensham and others, 1999, Harrison 1999, Stohlgren and others 1999, Safford and Harrison 2001). Additionally, comparison of plant diversity of grazed and ungrazed sites can yield important theoretical insights on the role of herbivory and competition in structuring plant communities (Harper 1969, McNaughton 1983, Belsky 1986). It has been recognized for many years that grazing animals have an impact on vegetation and vegetation recovery in arid environments can be quite slow (Guo 2004); however, because ranges have been more heavily and continuously grazed in recent decades in the CAR, the opportunities to study ranges that have been ungrazed for a considerable period of time are scarce. A limited number of studies have documented the floristic composition and plant cover of the CAR (Nalbantlı 1964, Bakır 1970, Yılmaz 1977, Büyükburç 1991). A few researchers have measured changes in plant cover resulting from grazing exclusion. Alınoğlu (1971) and Büyükburç (1983) documented plant cover increases of 89% and 115% after six and eight years of grazing rest, respectively. To date, no research has 327 been conducted comparing vegetation changes in grazed and protected areas for the rangelands of this region. The objectives of this study were (1) to compare several attributes of range plant diversity in grazed and long-term ungrazed sites, and (2) to determine the effect of grazing on plant species similarity. Materials and Methods The study was carried out in the rangelands of the Field Crops Central Research Institute Experimental _ Station, 45 km southwest of Ankara near the Ikizce village range in 2003 (Figure. 1). Prior to the establishment of the Research Station in 1976, the range areas were being grazed heavily and there are no records of tillage for the area. During station establishment, its borders were fenced and some areas have been protected from sheep and cattle grazing for 27 years, whereas the village range area has been continuously grazed by these animals. The village farmers (personal communications) reported that before enclosure establishment, there were 1280 AU of small and large ruminants (1 AU equals 500 kg live weight) grazing 600 ha of range area (0.47 ha/AU), which is considered high grazing pressure. Currently, 270 AU of both small and large ruminants are grazing 150 ha of native pasture (0.55 ha/AU) outside the Fig. 1 An aerial photo of the study site; ungrazed sites U1, U2, and U3 are in exclosures, and grazed sites G1, G2, and G3 are in the village range area 123 328 fenced area. Over the last 30 years in the village, both range area and number of livestock have been reduced dramatically, but the grazing pressure has not been reduced. Grazing occurs year round with free access and no management practices are implemented. Limited rangeland forage in the winter is supplemented by allowing animals to graze stubble after cereal harvest on croplands and by feeding cereal straw, barley grain, and other supplements (Fırıncıoğlu et al 1995). The exclosures did not prevent herbivory from small mammals or insects (Bigger and Marvier 1998). Soils are clay-loam, slightly alkaline, low in organic matter and phosphorous, high in lime, and abundant in potassium. The CAR is characterized by its arid climate with 300 to 350 mm annual rainfall, cold winters, and dry summers. Rainfall data obtained from the Experimental Farm metrological station measured a long-term average of 377 mm rainfall with 360 mm falling in 2003. The rangelands are a typical steppe vegetation type, consisting of some perennial (Festuca valesiaca, Poa bulbosa, Bromus tomentellus) and annual (Bromus tectorum, Bromus japonicus) grass species and some perennial shrubs (Thymus spyleus ssp. rosulans, Globularia orientalis). Other important steppe plant species include Artemisia santonicum, Thymus squarrosus, Onobrychis armena, Poa bulbosa var. vivipara, Stipa lagascea, Aegilops ovata, Cynodon dactylon, Salvia cryptantha, Koleria cristata and Kochia prostrata. The steppe vegetation in Turkey is included in the class Astragalo-Brometea (Quézel 1973), which is represented in Anatolia by the order Onobrychido armenaeThymetalia leucostomi (Akman and others 1984). Three multiscale vegetation plots were placed at each of the exclosure sites (Figure. 1). The plots, which were randomly placed inside the exclosure, were paired with three plots outside the exclosures that had similar soils, slopes, and aspects. The first ungrazed plot (U1) and the first grazed plot (G1) were on relatively flat terrain. U2, U3, G2, and G3 were on a 20 to 30 degree NE slope. The Modified-Whittaker plot (Figure 2) was chosen as it samples a large area (0.1 ha) and is useful for detecting less abundant plant species (Stohlgren et al. 1998). The plots were sampled at the phenological maximum (peak biomass) during the third week of June. The Modified-Whittaker plot was placed with the long axis along the major elevation gradient (Stohlgren et al. 1995). Each 20 · 50 m plot (1000 m2) had nested in it one 5 · 20 m plot (10 m2), two 2 · 5 m plots (10 m2) and ten 0.5 · 2 m plots (1 m2; Fig. 2). In the ten 1-m2 plots, all plant species were identified (Appendix), and the basal cover of each species, percentage of bare ground, and non-plant 123 Environ Manage (2007) 39:326-337 Fig. 2 The schematic presentation of Modified-Whittaker plots components (rock and stone) were estimated to the nearest percent. In addition, all plant species found in the 10-m2, 100-m2, and 1000-m2 areas were recorded. These lists were used to produce a species area curve (species richness = mx +b), where m is the slope of the line, x is the log area, and b is the y intercept (Stohlgren et al. 1995). Plant species that could not be identified in the field were collected and identified at the herbarium of the Biology Department in the Ankara Gazi University. Four specimens were not identifiable due to phenological stage, but could be classified into plant category and life form, and seven specimens were identified to genus. All 11 specimens were counted as individual species. Jaccard’s coefficient (Krebs 1989) was used to compare vegetation similarity between ungrazed and grazed plots with data from 1000-m2 plot. Jaccard’s coefficient (J), which gives equal weight to all plant species, is derived from: J ¼ A=ðA þ B þ CÞ where A is the number of species found in both grazed and ungrazed sites, B is the number of species found only in the grazed sites, and C is the number of species found only in the ungrazed sites (Krebs 1989). Jaccard’s coefficient varies from 1 (completely similar species) to 0 (no similar species) and is a good similarity measurement (Stohlgren et al. 1997). It is often expressed in percent as J · 100. We also compared species diversity between grazed and Environ Manage (2007) 39:326-337 329 ungrazed sites using Shannon’s index and Simpson’s index as described by Ludwing and Reynolds (1998). Shannon’s index (H’) for a sample, which is the average degree of uncertainty in predicting what species an individual chosen at random from a sample will be, is defined as H0 ¼  s X ðni=nÞ lnðni=nÞ i¼1 where ni is the cover of the ith species of S species in the sample and n is the total cover of all species in the sample. Simpson’s index (k) for a sample, which is the probability that two individuals selected at random will be the same species, is defined as k¼ s X ni ðni  1Þ=nðn  1Þ i¼1 The values from these indices were transformed in a method recommended by Ludwing and Reynolds (1988) and described by Hill (1973), to determine the abundant (N1) and very abundant species (N2). N1 was calculated as N1 ¼ eH , and N2 was calculated as N2 ¼ 1=k With the values from the above equations, a modified Hill’s ratio was then determined as a measure of evenness (E5; Hill 1973). E5 was calculated as E5 ¼ ð1=kÞ  1=eH  1 ¼ N2  1=N1  1 As E5 approaches zero, one species becomes more dominant in the total cover component. Higher values of E5 indicate a more even division of cover among the species in the sample area. A two-independent samples t-test was employed to test for differences between plants in the ungrazed and grazed areas. Statistical analyses were conducted with the Minitab statistical package and alpha = 0.05 was used to determine significance in all tests. Before the analysis, to satisfy assumptions of normality, all data after adding 1 were Log 10 transformed. Additional analyses were conducted after grouping the plant species into the following functional groups and/or life forms: annuals, perennials, shrubs (including sub shrubs and other woody plants), forbs, grasses, annual forbs, perennial forbs, annual grasses, and perennial grasses. Results There were a total of 113 species in the study area (Appendix), including 91 forbs, 11 grasses, and 11 shrubs. The majority of these species were perennials (88), although 25 annual species were identified in the study area. At the 1-m2 scale, 64% of the total species number that were in the study area were sampled, and there were more species in the ungrazed plots (61) than in the grazed plots (44) (Table 1). As the sampling area increased from 1-m2 to 1000-m2, the number of species also increased (Table 2). At the 1000-m2 scale, there Table 1 Species richness as a function of plot size and plant category of plant populations in ungrazed and grazed practices at the Field Crops Central Research Institute Experimental Station, Ikizce, Turkey Sample size 1-m 2 1000-m2 Category U G Total U G Total Species number Forbs Grasses Shrubs Annuals Perennials Annual forbs Perennial forbs Annual grasses Perennial grasses Total 48 9 4 13 48 12 36 1 8 61 32 7 5 10 33 10 23 1 6 44 58 10 6 17 58 15 44 2 8 74 76 10 8 20 74 18 58 2 8 94 48 7 7 12 50 11 37 1 6 62 91 11 11 25 88 22 69 3 8 113 U = ungrazed and G = grazed 123 330 Environ Manage (2007) 39:326-337 Table 2 The number of species at the area scales of 1, 10, 100, and 1000-m2 and regression equations of the species area curve in the ungrazed and grazed plots at the Field Crops Central Research Institute Experimental Station, Ikizce, Turkey Area Scale Ungrazed 1-m2 10-m2 100-m2 1000-m2 Equation r2 grazed 13** 11 22* 16 38** 19 63** 38 y = 6.5 log (area) + 12.5 y = 3.1 log (area) + 10.9 0.92 0.74 *Significant at P < 0.005 and **Significant at P < 0.001, within rows. were 94 species in the ungrazed plots and 62 species in the grazed plots. The regression equations of the species area curve for the ungrazed and grazed plots are given in Table 2. The slope of the species area curve for the ungrazed plots (6.5 ± 0.32 SE) was greater (P = 0.002) than the slope of the grazed plots (3.1 ± 0.32 SE), indicating that species numbers increased more rapidly in ungrazed plots as area sampled increased compared to the grazed plots (Table 2). The species area curves had a high coefficient of determination for the ungrazed and grazed plots (r2 = 0.92 and 0.74, respectively). Compared to the grazed plots, there were more plant species in the ungrazed plots at the 1-m2 (11 vs. 13 ± 0.5 SE, P = 0.004), 10-m2 (16 vs. 22 ± 1.6, P = 0.03), 100-m2 (19 vs. 38 ± 2.3, P = 0.004), and 1000m2 (38 vs. 63 ± 3.2, P = 0.005) sample areas in the study. Based on Jaccard’s coefficient, there was a 37% similarity in species composition between ungrazed versus grazed plots, with 42 species in common, and 52 and 19 species only in the ungrazed and grazed plots, respectively. There were no significant differences in Shannon’s diversity index, diversity (N1, N2), or evenness (E5) between the grazing treatments (Table 3). The percent basal cover of the major plant species, measured at the 1-m2 scale, is given in Table 4. Only plant species that were sufficiently abundant could be analyzed due to the requirements of normality. The cover of Astragalus acicularis was similar in both grazing treatments, with 0.84% in ungrazed and 0.95% in grazed plots. Similarly, Bromus tomentellus cover did not differ with 0.30% in the ungrazed and 0.25% in the grazed plots. The cover of Festuca valesiaca (Syn. F. ovina) was significantly different (P < 0.05) between the grazing treatments, with 8.0% in the ungrazed and 4.7% in the grazed plots. The shrub species, Genista albida had a greater (P < 0.01) plant basal cover in the ungrazed plots (1.8%), whereas Thymus spyleus ssp 123 Table 3 Indices of diversity from percent cover estimates of 1-m2 plots in the ungrazed (U) and grazed (G) plots at the Field Crops Central Research Institute Experimental Station, Ikizce, Turkey Index Treatment Value Probability Shannon’s index U G U G U G U G U G 1.25 1.22 0.35 0.37 4.44 3.54 4.15 3.15 0.78 0.80 0.958 Simpson’s index N1 N2 E5 0.636 0.111 0.238 0.713 rosulans had more (P < 0.01) basal cover in the grazed plots (8.3%). Though not significant, Globularia orientalis had 5.8% cover in the grazed and 12.9% in the ungrazed plots. As a perennial grass, Poa bulbosa basal cover was twice that observed in the grazed plots (0.19%) compared to ungrazed plots (0.08%) plots (P < 0.01). There were differences in total cover between the treatments of plant categories and life forms when analyzed by basal cover (Table 5). The cover of the perennial species was significantly different (P < 0.01) between the grazing treatments with 22% in the ungrazed compared to 16% in the grazed plots. In comparing plant categories, the forbs in ungrazed plots had twice as much cover (3.3%) as did the grazed plots (1.6%). The grass basal cover was the same in the ungrazed (8.1%) and grazed plots (5.4%). Shrub species had similar cover between the treatments. Total plant cover was significantly greater (P < 0.01) in the ungrazed (22%) than in the grazed (16%) plots. The transformed data for the annual-biennials (life form) and the annual-biennial forbs (life form and category) could not be analyzed since they did not satisfy the requirements of normality. For the life forms, there were two different groups, the annualbiennial and perennials. Most of the annual-biennial species were forbs and their total cover appeared to be greater in the ungrazed (0.34%) than in the grazed (0.06%) plots. Discussion and Conclusion The basic assumption made in this study was that if all other variables influencing vegetation were the same on both sides of the fence, then any difference must result from grazing. Although this is a small case study, we perceive it to be an important measure of Environ Manage (2007) 39:326-337 331 Table 4 Percent basal cover (actual and normalized with Log10 (X + l) transformation) of major plant species at the Field Crops Central Research Institute Experimental Station, Ikizce, Turkey in ungrazed (U) and grazed (G) plots Plant species Treatment n Actual cover (%) Normalized cover (X ± SEM) P-value Astragalus acicularis U G U G U G U G U G U G U G 13 24 30 30 27 30 16 5 20 29 19 7 9 28 0.84 0.95 0.30 0.25 7.99 4.66 1.77 0.43 3.62 8.26 12.89 5.76 0.08 0.19 0.22±0.06 0.24±0.04 0.11±0.01 0.09±0.01 0.86±0.06 0.69±0.05 0.38±0.06 0.15±0.02 0.57±0.06 0.91±0.04 0.98±0.10 0.78±0.09 0.03±0.01 0.07±0.01 0.751 Bromus tomentellus Festuca valesiaca Genista albida Thymus spyleus ssp rosulans Globularia orientalis Poa bulbosa 0.150 0.037 0.001 0.01 0.126 0.011 n = number of plots with plant species; SEM = standard error of the mean; bold P values = significant differences. vegetation change in CAR rangelands. In this study, the dual concept is used at an alpha level. The dual concept as described by Peet (1974) combines the number of species with the relative abundance of the species and alpha diversity is the assortment of plant species within a given habitat (Whittaker 1972). Grazing can influence the structure and organization of plant communities in different ways (Crawley 1983, Noy-Meir and others 1989). Sternberg and others (2000) explained that grazing caused two kinds of effects: a direct effect that occurs through the selective and differential removal of plant tissue or species, and an indirect effect that occurs on botanical composition and species diversity when selective grazing on dominant species reduces their vigor and presence, thus favoring the spread of less competitive but more tolerant species. The 3000-m2 ungrazed area in this study had increased species richness (94) compared to the 3000-m2 grazed area (62). Three questions arise concerning these differences in species richness: (1) Was the increase in the ungrazed area the result of a greater abundance of rare plants? (2) Was the increase in the ungrazed area the result of the re-invasion of locally extinct species? (3) Was the decrease in the grazed area the result of a continued decline of palatable species as grazing intensity increased? The first question can be answered using the Red Data Book of Turkish Plants (Ekim and others, 2000). The only threatened plant found in this study was Fritillaria flaschaeriana, which was identified in both grazed and ungrazed plots (10-m2). There were 18 other forb species that are listed as least threatened. Of these, seven were in both grazed and ungrazed plots and ten were only in the ungrazed plots. The ten additional plant species in the ungrazed plots is an indication that there was an increase in species that are at least of some concern in ungrazed plots. Another way to answer the first question is to compare the smallest scale at which plants common to both treatment areas were measured (1, 10, 100, or 1000-m2). There were six species that were more abundant, three in the ungrazed areas: Leontodon asperrimus (1 vs. 100-m2), Salvia cryptantha (1 versus 10-m2), and Scabiosa argentea (1 vs. 1000-m2), and three in the grazed area: (Carduus nutans 1 vs. 1000-m2), Centaurea virgata (1 versus 100-m2), and Scutellaria orientalis (1 vs. 10-m2). Of these six species, only two are palatable. One of them, Scabiosa argentea, occurred more frequently in the ungrazed plots and the other, Scutellaria orientalis, more frequently in the grazed plots. This second answer to the first question is much less conclusive about the fate of rare plants. The second and third questions can only be answered when the starting vegetation is known. However, many of the 62 plant species found in the grazed area should be able to survive grazing pressure. In fact, of the 19 plant species specific to the grazed area, five (Astragalus plumosus var plumosus, Astragalus podporae, Astragalus wiedemannianus, Astragalus xylobasis var angustus, and Marrubium parviflorum) are noxious species and five (Erodium ciconium, Achillea wilhelmsii, Erysimum crassipes, Salvia sp, and Sisymbrium altissima) have little forage value. Only two species, (Sanguisorba minor and Artemisia santonicum), with reasonable feeding value for herbivores appeared in grazed areas, but not in the ungrazed areas. There is an expectation that many of the plant species specific to the ungrazed areas would be more 123 332 palatable. However, most of the forbs were noxious and some were poisonous. The two annual grasses that appeared (Bromus japonicus and Bromus tectorum) are invaders, one perennial grass (Agropyron cristatum) is considered to be a decreaser, and the other (Stipa lessingiana) is an increaser. Except for Acantholimon acerosum with a spiny stature, the other shrubs (Fumana procumbens, Genista sessilifolia, Gypsophila sphaerocephala) are all grazed by sheep and goats. The cessation of grazing for 27 years has resulted in increased species richness compared to the grazed plots. However, after 27 years of protection from grazing and only a 37% similarity of species, one would wonder if the plant communities are even comparable. Based on basal cover, the finding that the indices of diversity and evenness were the same at 1-m2 is a reflection of the structure of the plant community (Table 3). In both the ungrazed and grazed areas, two shrubs (Globularia orientalis and Thymus spyleus ssp rosulans) and a perennial grass (Festuca valesiaca) were dominant based on basal cover. Additionally, in the ungrazed areas, there was a third dominate shrub (Genista albida). All other plant species only made minor contributions to basal cover. Life history and morphology of dominant plant species are important attributes in the responses of the community to grazing intensity and timing (Noymeir and others 1989, Diaz and others 1994, McIntyre and others 1995, Lavorel and others 1997). The general hypothesis is that the sensitivity of plant communities to grazing depends on the frequency and strength of adaptations helping plants avoid or tolerate grazing animals (Diaz and others 2001, McIntyre and Lavorel 2001, Vesk and Westoby 2001). Adaptation to aridity, such as small stature, basal meristems, and drought-deciduous leaves, also proves advantageous in preventing or recovering from grazing (Coughenour, 1985). Adler and others (2004) specified two separate hypotheses: first the main control on the development of grazing-resistance traits is the evolutionary history of grazing and environmental factors related to aridity, and, second, because grazing resistance traits mediate plant-herbivore interactions, such traits are the direct determinants of ecosystem response to grazing. In the arid environment of the CAR rangelands, Festuca valesiaca, with its sod formation capability and greater drought resistance, maintained basal cover in both grazing treatments. The greater abundance of Poa bulbosa in grazed sites might be due to its better seed viability, poor quality and quantity of leaves, and/or its semi-prostrate growth habit, all of which may be effective defenses against herbivores. Neither treatment changed the 123 Environ Manage (2007) 39:326-337 cover of Bromus tomentellus. Perhaps with its conspicuous stature and copious production of high viability seed, it is able to exploit both situations equally. The cessation of grazing can increase the cover and frequency of shrubs (Schulz and Leininger 1990, Coughenour 1991). However, there are some conflicting results: Huges (1980, 1983) found higher shrub frequency on grazed sites in desert shrub communities in Arizona, and Smeins and others (1976) found no significant increase in shrub cover after 25 years of protection from grazing. The major shrub species, Genista albida, Globularia orientalis, and Thymus spyleus ssp rosulans, had considerable cover in both treatments. However, the cover of Globularia orientalis and Genista albida increased and the cover of Thymus spyleus ssp rosulans decreased when grazing was excluded. In a non-grazed environment, Thymus spyleus ssp rosulans may not be able to compete against the other two shrubs. Despite the long period of enclosure, the combined basal cover of all shrub species was similar in both grazing treatments (Table 5). Grazing resulted in a stronger selective pressure against forbs rather than for grass and shrub species. Grasses and shrubs were less affected and appeared to be more resilient to grazing pressure. Because shrubs and grass species accounted for 85 to 90% of basal plant cover, overlooking the forb component may lead to misinterpretation of the effects of grazing on species richness in these communities. Though the effect of overgrazing on soil erosion is not measured in this study, the considerable reduction in the plant cover of the grazed sites has exposed the soil to water and wind erosion. Avcıoğlu and Erekul (1996) pointed out that the deleterious effect of soil erosion was worsened by overgrazing in Turkey’s rangelands; this is especially true for water erosion in steeply sloped areas. Our results indicate that a more species-rich plant community can be developed with the complete exclusion of grazing for 27 years after a long history of grazing disturbance. However, long protection from grazing did not lead to an increase in the amount of desirable range plants species such as Bromus tomentollus, Agropyron cristatum, Koeleria cristata, and Onobrychis armena. The exclusion of grazing in areas with a long history of grazing can be considered a disturbance (Milchumas and others 1988), and either very heavy grazing or the absence of grazing can lead to alternative, less productive vegetative states (Harrison 1979). According to Akman and others’ (1984) classification, the rangeland vegetation in the study should be the Environ Manage (2007) 39:326-337 333 Table 5 Percent basal cover (actual and normalized Log10 (X + l)) of plant categories at the Field Crops Central Research Institute Experimental Station, Ikizce, Turkey, in ungrazed (U) and grazed (G) plots. Categories Treatment n Actual cover (%) Normalized cover (X ± SEM) P value Total cover U G U G U G U G U G U G U G U G U G 30 30 30 30 30 30 29 30 26 22 30 30 26 22 30 30 30 30 21.82 16.23 3.25 1.63 8.08 5.41 10.49 9.19 0.34 0.06 21.50 16.18 0.34 0.06 2.93 1.58 8.08 5.41 1.33 ± 0.03 1.21 ± 0.03 0.54 ± 0.05 0.37 ± 0.04 0.86 ± 0.06 0.75 ± 0.04 0.89 ± 0.08 0.93 ± 0.05 0.12 ± 0.02 0.02 ± 0.01 1 .32 ± 0.03 1.20 ± 0.03 0.12 ± 0.02 0.02 ± 0.01 0.47 ± 0.06 0.36 ± 0.04 0.86 ± 0.06 0.75 ± 0.04 0.005 Forbs Grasses Shrubs Annuals -biannuals Perennials Annual-biannual forbs Perennial forbs Perennial grass 0.013 0.163 0.916 — 0.009 — 0.120 0.163 n = number of plots with plant species; SEM = standard error of the mean; bold P values = significant differences; — = data not normalizable. Phlomido armeniaceae-Astragalion microcephali alliance, which is characterized by the presence of Phlomis armeniaca, Astragalus microcephalus, Teucrium chamaedrys, and Marrubium parviflorum. Of these four species, Astragalus microcephalus was not found in either treatment, Phlomis armeniaca and Teucrium chamaedrys were only found in the exclosure plots, and Marrubium parviflorum was only found in the grazed plots. According to historical vegetation typing, it appears that vegetation in both treatments is transitioning towards different vegetation types. Many mechanisms exist that may prevent an altered plant community from returning to its original composition once the disturbances that initiated the changes have been removed (Westoby and others 1989, Laycock 1991, Friedel 1991). However, as the exact pristine condition of this steppe vegetation is not known, it is not possible to determine which of these treatments has resulted in the plant community most similar to pre-settlement vegetation. Certainly, the continued grazing outside the exclosure is transitioning vegetation to another, more degraded state. Since the establishment of the exclosures, it is possible that there has been a slow transition towards a new state, but it seems quite different from historic plant communities. Stopping grazing for 27 years should not be regarded as a comprehensive range rehabilitation option in this environment, which evolved with grazing (Harrison 1979, Milchunas and other 1988). Our study reveals the importance of considering plant adaptation to grazing when planning restoration projects. This study led to four generalizations about current grazing regime and long-term exclosures in the rangeland around the study area: 1. 2. 3. 4. Exclosures will increase species richness. Heavy grazing may have removed some plant species from the vegetation community. Complete protection from grazing for a lengthy period of time after a long history of grazing disturbance may not lead to an increase in desirable plant species, with a concomitant improvement in range condition. Research needs to be conducted to determine how these rangelands can be improved. In conclusion, these grazing exclosures have provided important insights on vegetation patterns that mere comparisons of grazed sites would have never generated. Acknowledgments The authors wish to thank to the administration (Dr. Hüseyin Tosun, Director, and Dr. Aydan Ottekin, Assistant Director) of the Central Research Institute for Field Crops for funding the research, to Prof. Dr. Mecit Vural for helping plant identification, Öztekin Urla and Dr. Ali Mermer for providing the aerial photo of study area, and anonymous reviewers who have provided valuable advice and comments. 123 334 Environ Manage (2007) 39:326-337 Appendix Appendix Species list, functional groups (forb, grass, shrub) and life forms (annual, biannual, perennial), and smallest plot (l-m2, 10-m2, 100-m2, 1000-m2) in which the species occurred of plants identified in the study Annual-biennial 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 Perennial forbs 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 123 Ungrazed Grazed forbs Alyssum minus. var minus Alyssum pateri ssp pateri Anagallis arvensis Androsace maxima Centaur eadepressa Crupina crupinastrum Erodium ciconium Helianthemum ledifolium Minuartia hamata Orobanchae sp Scabiosa rotata Sideritis montana Trigonella velutina Glaucium corniculatum Sisymbrium altissima Reseda lutea Carduus nutans Cram betataria Tragopogon dubius Verbascum sp Thesium billardieri Brassica elongata 1 1 100 1 1 1 — 1 1 100 1 1 — 100 — 1 1000 10 1 — 1 100 1 1 — 1 — — 1 1 1 — — 1 1000 — 1 — 1 — — 10 — — Achillea wilheImsii Allium pseudoflavum Anthemis cretica ssp anatolica Anchusa leptophylla Anthemis tinctoria var tinctoria Astragalus acicularis Astragalus sp. Ajuga chamaepitys Astragalus densifolius ssp. densifolius Astragalus karamasicus Astragalus nitens Astragalus plumosus var plumosus Astragalus podporae Astragalus sigmoideus Astragalus sp Astragalus vulnerariae Astragalus wiedemannianus Astragalus xylobasis var angustus Unidentifiable Bungea trifida Centaurea carduiformis ssp carduiformis var carduiformis Centaurea drabifolia Cerinthe minor Centaurea urvillei Centaurea virgata Convolvulus holosericeus Crepis foetida ssp rhoedifolia Cuscuta sp. Dianthus anatolious Echinophora tenuifolia ssp sibthorpiana Echinophora tournefortii — 1 1 1 1 1 10 1 100 1 1 — — 1 — 1 — — — 1000 100 1 1000 10 100 1 1000 100 10 1 1000 10 — 1 — — 1 1000 — — 1 — 1000 1000 — 1 1 10 10 1 — — — — 10 1 1 1 — — 1 — Environ Manage (2007) 39:326-337 335 Appendix Continued 32 Eryngium campestre 33 Erysimum crassipes 34 Euphorbia macroclada 35 Ferulago pauciradiata 36 Fritillaria flaschaeriana 37 Galium incanum 38 Galium verum 39 Haplophyllum corniculatum 40 Hedysarum cappadocicum 41 Helianthemum nummularium 42 Hedysarum varium 43 Iris sp 44 Isatis tinctoria 45 Jurinea pontica 46 Unidentifiable 47 Lappula barbata 48 Leontodon asperrimus 49 Marrubium parviflorum 50 Minuartia anatolica 51 Moltkia aurea 52 Moltkia caerulea 53 Onobrychis armena 54 Onobrychis oxytropa 55 Onosma tauricum 56 Paronychia kurdica 57 Paracaryum racemosum var. racemosum 58 Phlomis armeniaca 59 Polygala pruinosa 60 Prangos meliocarpoides 61 Salvia cryptantha 62 Sanguisorba minor 63 Salvia sp 64 Scabiosa argentea 65 Scutellaria orientalis 66 Teucrium chamaedrys 67 Teucrium polium 68 Veronica multifida 69 Vinca herbacea Annual Grasses 1 Bromus japonicus 2 Bromus tec forum 3 Aegilops speltoides Perennial Grasses 1 Agropyron cristatum 2 Bromus tomentellus 3 Festuca valesiaca 4 Koelaria cristata 5 Poa bulbosa 6 Poa tmeleontis 7 Stipa holosericea 8 Stipa lessingiana Shrubs 1 Acantholimon acerosum 2 Artemisia santonicum 3 Unidentifiable 4 Fumana procumbens 5 Genista albida 6 Genista sessilifolia 7 Globularia orientalis 8 Gypsophila sphaerocephala 9 Noaea mucronata 10 Unidentifiable Ungrazed Grazed 1 — 1 100 10 1 1000 1000 1 1 1 1 100 1 1 1 1 — 1 1 1000 1 100 1 1 1 10 1 1 1 — — 1 10 1000 1 100 1 1 1 1 — 10 1 1000 — — — 1 — — 1 1 — 100 1 1 — — 1 — — 1 — — — — 10 10 10 1000 1 — 1 — — 1 10 — — — 1 1 1 1 1 1 1 1 1 — 1 1 1 1 1 1 — 1000 — — 10 1 1000 1 1 100 — — 1000 1 1 — 1 — 10 1 123 336 References Adler P. B., D. G. Milchunas, W. K. Lauenroth, O. E. Sala, I. C. Burke. (2004). Functional traits of graminoids in semiarid steppes: a test of grazing histories. J Appl Ecol 41:653–663 Akman Y., O. Ketenoğlu, P. Quézel Marseille, M. Demirors. (1984). A syntaxonomic study of steppe vegetation in Central Anatolia. Phytocoenologia 12:563–584 Alınoğlu N. (1971). Research on the effects of continuous grazing and various resting regimes on range vegetation. Ankara Grassland and Animal Research Institute Publication Number: 16, Ankara Anonymous (1980). Soil Statistical Bulletin, The Ministry of Rural Affairs and Cooperatives, the publication of The General Directorate of Soil-Water, Ankara Anonymous (2001). State Statistics Institution, The General Agriculture Census, Ankara Avcıoğlu R., O. Erekul (1996). Erosion and rangeland vegetation. The third Turkish National Meadow-Rangeland and Forage Crop Congress, 17–19 June 1996. Atatürk University, Agricultural Faculty, Erzurum, pp 32–50 Bakır Ö. (1970). A range vegetation survey in the range areas of Middle East. Technical University, Ankara University, Agricultural Faculty Publication number 382, Scientific Research and Studies Number: 232. Ofset Publication, Ankara Bakır Ö. (1998). Management systems for in-situ conservation of plants. In: J. Zencirci, Z. Kaya, Y. Anikster, W. T. Adams. (eds) The proceedings of International Symposium on In situ Conservation of Plant Genetic Diversity. Central Research Institute for Field Crops, Ankara, Turkey, pp 67–72 Belsky A. J. (1986). Does herbivory benefit plants? A review of the evidence. Am Naturalist 127:870–892 Bigger D. S., M. A. Marvier. (1998). How different would a world without herbivory be? A search for generality in ecology. Integrat Biol 1:60–67 Büyükburç U. (1983). Research on the rehabilitation possibilities of Ankara, Yavrucak village rangelands with the implementation of rest and fertilization. Ankara Grassland and Animal Research Institute, Publication Number: 79, Ankara Büyükburç U. (1991). A rehabilitation study on Karayavsan Village rangelands with the application of the different kinds and doses of fertilizer and resting. Republic University, Tokat Agricultural Faculty, Publication Number 8, Scientific research: 5, Tokat Coughenour M. B. (1985). Graminoid responses to grazing by large herbivores: adaptations, exaptations and interacting processes. Ann Missouri Bot Garden 72:852–863 Coughenour M. B. (1991). Biomass and nitrogen responses to grazing of upland steppe on Yellowstone’s northern winter range. J Appl Ecol 28:71–82 Crawley M. J. (1983). Herbivory: the dynamics of plant-animal interactions. Blackwell Scientific Publications, Oxford, UK Diaz S., A. Acosta, M. Cabido. (1994). Community structure in monatane grasslands of central Argentina in relation to land use. J Vege Sci 5:483–488 Diaz S., I. Noy-Meir, M. Cabido. (2001). Can grazing response of herbaceous plants be predicted from simple vegetative traits? J Appl Ecol 38:497–508. Ekim T., M. Koyuncu, M. Vural, H. Duman, Z. Aytaç, N. Adıgüzel. (2000). Red Data Book of Turkish Plants (Pteridophyta and Spermatophyta. Turkish Nature Conservation Association, Van Centenary University, Barıscan publisher, Ankara 123 Environ Manage (2007) 39:326-337 Fesnsham R. J., J. E. Holman, M. J. Cox. (1999). Plant species responses along a grazing disturbance gradient in Australian grassland. J Vege Sci 10:77–86 Friedel M. H. (1991). Range condition assessment and the concept of thresholds: a viewpoint. J Range Manage 44:422– 426 Frıncıoğlu H. K., S. Cristiansen, E. J. Lamont, S. Unal, M. Peskircioglu, SPS. Beniwal (1995). Village, farm, and pasture resource assessment in planning a forage, livestock, and range improvement project for the central highlands of Turkey. In: N. Haddad, R. Tutwiller, E. Thomson (eds) The proceedings of the regional symposium on integrated crop-livestock systems in the dry areas of West Asia and North Africa, held 6-8 November, Amman, Jordan, pp 39–50 Guo Q. (2004). Slow recovery in desert perennial vegetation following prolonged human disturbance. J Vege Sci 15:757– 762 Harrison G. W. (1979). Stability under environmental stress: resistance, resilience, persistence and variability. Am Naturalist 113:659–669 Harrison S. (1999). Native and alien species diversity at the local and regional scales in a grazed California grassland. Oecologia 121:99–106 Harper J. L. (1969). The role of predation in vegetational diversity. Brookhaven Symp Biol 22:48–62 Hill M. O. (1973). Diversity and evenness: A unifying notation and its consequences. Ecology 54:427–432 Hughes L. E. (1980). Six grazing exclosures with a message. Rangelands 2:17–18 Hughes L. E. (1983). Is no grazing really better than grazing? Rangelands 5:159–161 Krebs C. J. (1989). Ecological methodology. Harper and Row, New York Lavorel S., S. McIntyre, J. Landsberg, T. D. A. Forbes. (1997). Plant functional classifications: from general groups to specific groups based on responses to disturbance. Trends Ecol Evol 12:474–477 Laycock W. A. (1967). How heavy grazing and protection affect sagebrush-grass ranges. J Range Manage 20:206–213 Laycock W. A. (1991). Stable states and thresholds of range condition on North American rangelands: a viewpoint. J Range Manage 44:427–433 Ludwing J. A., J. F. Reynolds. (1988). Statistical ecology: a premier on methods and computing. John Wiley, New York McIntyre S., S. Lavorel, R. M. Tremont. (1995). Plant life-history attributes: their relationship to disturbance response in herbaceous vegetation. J Ecol 83:31–44 McIntyre S., S. Lavorel. (2001). Livestock grazing in subtropical pastures: steps in the analysis of attribute response and plant functional types. J Ecol 89:209–226 McNaughton S. J. (1983). Serengeti grassland ecology: the role of composite environmental factors and contingency in community organization. Ecol Monogr 53:291–320 Micchunas D. G., O. E. Sala, W. K. Lauenroth. (1988). A generalized model of the effects of grazing by large herbivores on grassland community structure. Am Naturalist 132:11–23 Nalbantlı G. M. (1964). Studies on the plant composition and grazing pressure of communally-used rangelands in the Central Anatolia, Turkey, unpublished PhD thesis, Ankara University, Ankara Noy-Meir I., M. Gutman, Y. Kaplan. (1989). Responses of Mediterranean grassland plants to grazing and protection. J Ecol 77: 290–310 Environ Manage (2007) 39:326-337 Peet R. K. (1974). The measurement of species diversity. Annu Rev Ecol System 5:285–307 Quézel P. (1973). Contributions to phytosociology studies in the Taurus Massif. Phytocoenologia 1:131–222. Safford H. D., S. P. Harrison. (2001). Grazing and substrate interact to affect native vs. exotic diversity in roadside grasslands. Ecol Appl 11:1112–1122 Schulz T. T., W. C. Leininger. (1990). Differences in riparian vegetation structure between grazing areas and exclosures. J Range Manage 43:295–299 Smeins F. E., T. W. Taylor, L. B. Merrill. (1976). Vegetation of a 25-year exclosure on the Edwards Plateau, Texas. J Range Manage 29:24–29 Sternberg M., M. Gutman, A. Perevolotsky, E. D. Ungar, J. Kigel. (2000). Vegetation response to grazing management in a Mediterranean herbaceous community: a functional group approach. J Appl Ecol 37:224–237 Stohlgren T. J., M. B. Falker, L. D. Schell. (1995). A ModifiedWhittaker nested vegetation sampling method. Vegetation 117:113–121 Stohlgren T. J., M. B. Coughenour, G. W. Chong, D. Binkley, M. Kalkhan, L. D. Schell, D. Buckley, J. Berry. (1997). Landscape analysis of plant diversity. Landscape Ecol 12:155–170 Stohlgren T. J., K. A. Bull, Y. Otsuki. (1998). Comparison of rangeland vegetation sampling techniques in the Central Grasslands. J Range Manage 51:164–172 337 Stohlgren T. J., L. D. Schell, B. Vanden Heuvel. (1999). How grazing and soil quality affect native and exotic plant diversity in Rocky Mountain grasslands. Ecol Appl 9:45– 64 Tan A. (1998). Current status of plant genetic resources conservation in Turkey. In N. Zencirci, Z. Kaya, Y. Anıkster, W. T. Adams (eds) The proceedings of International Symposium on In Situ Conservation of Plant Genetic Diversity. Central Research Institute for Field Crops, Ankara, pp 5–16 Vesk P. A., M. Westoby (2001). Predicting plant species’ responses to grazing. J Appl Ecol 38:897–900 Weaver J. E., N. W. Rowland. (1952). Effects of excessive natural mulch on development, yield, and structure of native grassland. Bot Gazette 114:1–19 Westoby M., B. Walker, I. Noy Meir. (1989). Opportunistic management for rangelands not at equilibrium. J Range Manage 24:266–274 Whittaker R. H. (1972). Evolution and measurement of species diversity. Taxonomy. 21:213–215 Yılmaz T. (1977). Studies on the botanic composition of some problematic range areas in Konya Province. Konya Regional Directorate of Soil-Water Research Institute, Publication Number: 46, Konya 123