Floristic Diversity and Distribution Patterns Along an Elevational Gradient in the Northern Part of the Ardabil Province Rangelands, Iran

This study evaluated the distribution pattern, species diversity index (richness, diversity, and evenness), importance value index (IVI), and family importance value (FIV) of all vascular plant species and various plant community life-form spectra along the Moghan Plain–Sabalan Mountain rangelands, Ardabil province, Iran. Sampling was conducted in 11 elevation classes at 300-m elevation intervals (100–3300 masl). In each elevation interval, 30 quadrats (1 × 1 m) were laid to collect vegetation data. Different vegetation attributes (density, frequency, and canopy cover) were measured for each quadrat. Regression analyses were employed to explore the interrelation of elevation with diversity, species richness, evenness, and IVI. In total, 251 species (143 genera, 38 families) were recorded across the study area. Poaceae, Fabaceae, and Asteraceae were most dominant families according to the FIV. According to the IVI, Poa bulbosa, Festuca ovina, Medicago minima, and Artemisia austriaca were the dominant species along the elevation gradient. Total diversity and species richness showed a normal distribution along the elevation gradient. Overall, hemicryptophytes and therophytes were the dominant life forms, while chamaephytes and geophytes were less frequent forms. The life-form patterns changed along elevation gradients. While the diversity and species richness values of therophytes and chamaephytes showed a decreasing trend, hemicryptophytes increased with increasing elevation. Diversity, species richness, and evenness of geophytes were not significantly different across the elevation gradient. Class I (highly palatable) and class II (mostly palatable) species declined as the elevation increased and then increased, while class III (hardly or unpalatable) species showed a reverse trend. The difference in the abundance and distribution of species in elevation classes could be related to resource availability, overlap of habitats, habitat patchiness, land area, degree of human influence, or biotic disturbances.


Introduction
The Moghan Plain-Sabalan Mountain rangelands (MPSMRs) in Ardabil province, Iran, supply livestock forage and play an essential role in food provision, groundwater enrichment, clean water production, wildlife, tourism, recreation, medicinal plant provision, beekeeping, and climate change mitigation (Ghorbani et al 2014Nazari Anbaran et al 2016;Ghafari et al 2017Ghafari et al , 2018. According to the literature, 396 plant taxa have been identified in the area. Of these, 25 plant taxa are endemic to Iran (Ghafari et al , 2018. Because of the variety of plant species and the climate, beekeeping is a good source of food in this area (330,000 kg/y honey products; Sharifi et al 2017). The MPSMRs are also rich in medicinal and commercially valuable plants, which require sustainable management to promote health, income, and employment . In recent decades, with the increasing population, low level of development, and high dependence of local communities on rangelands for their primary livelihood needs, as well as increasing livestock numbers as living standards have improved, the MPSMRs have been overused. Excessive and irregular grazing, destruction by livestock, disregard for seasonal unsuitability for grazing, conversion of rangelands, and road construction are the main causes of degradation in the MPSMRs (Ghafari et al , 2018. In the early 1980s, the rangelands covered 58.2% of the land area; however, by 2015, its coverage had been reduced to 53.9% (Kakehmami et al 2017). Understanding the capabilities and limitations of rangelands is essential for appropriate management at the regional level. The rapid change in elevation, even over short distances, makes the MPSMRs an ideal place for studying the vegetation dynamics associated with biotic and abiotic factors that vary with elevation (K€ orner et al 2011).
Few studies conducted in the MPSMRs (Ghorbani et al 2014; Nazari Anbaran et al 2016; Ghafari et al 2017Ghafari et al , 2018 have focused on vegetation structure and species richness. However, understanding diversity patterns is essential for developing best approaches to diversity conservation as well as the sustainable utilization and management of plant resources. Hence, the aim of this study was to analyze floristic composition, diversity, and structure along the elevation gradient of the MPSMRs. This is expected to provide baseline data along with an analysis that can be used to develop diversity conservation strategies and sustainable rangeland management plans for the study area. This study attempted to answer the following questions: (1) Do species composition, ecological importance, and structural components of vegetation vary significantly along the elevation gradient? (2) What is the relation between elevation and species richness? (3) Do the diversity index and ecological importance along the elevation gradient differ, and if so, how?

Rangeland history and biophysical specificities of the area
The Ardabil province rangelands are located at the confluence of the 3 climatic areas: the Siberian from the north, the Mediterranean from the west, and the wet from the Caspian Sea. This, together with the Sabalan and Khoroslou mountains and hills (slopes 5% to more than 100%), affects climate and plant diversity in the region (Sharifi et al 2017). Various types of rangelands, such as summer rangelands (Sabalan, Baghru, and Buzghush mountains rangelands), spring and autumn rangelands (low mountainous areas, such as Kalantar-Khoroslou, to hilly areas, such as Gharesou river landscapes, also known as the Arshagh rangelands) and winter rangelands (Moghan Plain rangelands) are part of Ardabil province.
The second largest nomad tribe, the Shahsavan, belong to Ardabil province (Hosseinzadeh et al 2017a). The nomadic livestock husbandry system is one of the most important and oldest livestock husbandry systems in Iran. It strongly depends on natural rangelands. For centuries, the Moghan Plain has served as a grazing area for nomadic tribes, with agriculture being of secondary importance. Shahsavan nomads move their livestock to make the best use of available forage and water in a seasonal cycle on their allotted spring, summer, autumn, and winter rangelands (Hosseinzadeh et al 2017b). Shahsavan nomads use the Moghan rangelands for about 6 months of the year (Arjmand Karkazloo 2016).
The climate is favorable for agriculture and animal husbandry. Traditional and semi-industrial animal husbandry are practiced in villages (paddocks) and are dependent on rangeland forage and agricultural residues. In Ardabil province, 1600 pastoralist households (11,000 people) and their livestock (about 3,400,000 animal units/ sheep) contribute at least $57 million per year (milk, meat, dairy, and other by-products) to the province's economy (Sharifi et al 2017). Sheep are the dominant livestock (95%), followed by goats (3%) and cows (2%) (Nazari Anbaran et al 2016; Ghafari et al 2018).
Recently, more extensive irrigated farmlands and agroindustrial schemes have been introduced in the winter rangelands of the Moghan Plain, which, together with the expansion of farming, has further reduced the rangelands and increased the settlement rate among nomads, causing drastic social and economic changes in their way of life (Arjmand Karkazloo 2016). In spring (April to late June) and autumn (mid-September to end of November) the herds feed on the heavily grazed transition routes between winter and summer rangelands. In recent years, spring and autumn rangelands have been cultivated in residential areas, irrigated, dry-farmed, and used for horticulture. The small remaining area is rangeland (Ghafari et al 2018). While overall the rangelands of Ardabil province are now becoming better understood, little is known about its spring and autumn rangelands. At the end of April, as the snow recedes and rangeland becomes available, the nomads move to their summer Sabalan Mountain rangelands at higher elevations. Summer rangelands have prolonged snow cover in winter and comparatively high annual precipitation (400-700 mm). They mark the upper limit of agriculture and permanent human settlement. The harsh climate (extremely low temperature and strong winds), rugged terrain, and geological and hydrological conditions make the Sabalan Mountain rangelands particularly vulnerable to unsound practices of natural resource management and environmental degradation. Erosion rates are higher and the loss of fertility due to nutrient leaching is more pronounced than elsewhere. Because of the prevailing low temperatures, plant growth and soil formation are slower, and vegetation cover is significantly less abundant than in lowland areas (Sharifi et al 2017).

Study sites
The study was carried out in the area stretching from the Moghan Plains to Sabalan Mountain in Ardabil province, Iran, located at 47845 0 to 48823 0 E and 38818 0 to 39827 0 N ( Figure 1). Data collection sites have the following characteristics: (1) stratified at approximately 300 m elevation intervals; (2) located in the natural rangelands (depending on road accessibility corridors selected); and (3) a minimum distance of 300 m from the road, agricultural lands, and residential and other institutional compounds to avoid edge effects. According to the criteria used in selecting sites, the study sites were in the elevational range of 100-3300 masl (anthropogenic ecosystems, such as agricultural lands and residential and industrial areas, dominate elevations below 100 masl, and there are no access roads to sites above 3300 masl). For this study, we categorized the elevation zone between 100 and 3300 masl into 3 elevation categories: low elevation (100-900 masl), mid-elevation (900-2100 masl), and high elevation (2100-3300 masl). The climate at higher elevations is cold semiarid, with cold winters and temperate summers; however, lower elevations have a temperate semiarid climate and most of the precipitation is during winter. Higher elevations are covered with snow and ice for 3-4 months of the year (Tavosi and Delara 2011). More details about the study sites are presented in Table 1.

Methodology
The selected rangelands were sampled from April 2015 to July 2016. Within each of the elevation classes, three 100-m transects were set at 50-m intervals parallel to each other and perpendicular to the main slope. Along each transect, 10 quadrats (1 3 1 m 2 ) (at 10-m intervals) were set for systematic random sampling. Since diversity and species richness are quadrat size and abundance dependent (Magurran 2004), the same quadrat size was used for both. Vegetation attributes (density, canopy cover, and frequency), along with spatial coordinates (latitude, longitude, and elevation) using a global positioning system, were recorded for each quadrat.
Species were identified using literature such as Assadi et al (1988Assadi et al ( -2012, Davis (1965Davis ( -1988, Ghahreman (1979Ghahreman ( -2003, Komarov (1934Komarov ( -1954, and Rechinger (1963Rechinger ( -2010. The name of taxon authors was verified using IPNI (International Plant Names Index; and The Plant List (2018). The life forms were identified using Raunkiaer classification (Raunkiaer 1934). Species were sorted into palatability classes: highly palatable (class I)-forage plants, most preferred by domestic livestock; mostly palatable (class II)plants averagely preferred by livestock; hardly or unpalatable (class III)-plants least preferred or not grazed by livestock at any stage; and harmful and possibly toxic (Hussain and Durrani 2009). The palatability classes for each species were determined using literature such as FRWO (1982) and Ghafari et al (2017).
A complete picture of the ecological importance of each species in the community was obtained by calculating the importance value index (IVI) and the family importance value (FIV). To determine the IVI, the percentage values of relative density, relative dominance, and relative frequency were added together (Curtis and McIntosh 1950). The IVI was calculated for each species in each of the 11 elevation classes to determine the vegetation structure. The plant species with the highest importance value at each site were considered to be dominant species, and the plant species with the lowest importance value at each site were considered rare (Curtis and McIntosh 1950).
The FIV for botanical families was calculated by summing the IVI for different species of the same family (Mori et al 1983). IVI and FIV were calculated using the equations in Table 2.
The vegetation was analyzed for species diversity, richness, and evenness to reduce the complexity of data (Whittaker 1953). Species diversity (Shannon and Weaver 1963), richness (Menhinick 1964), and evenness (Pielou 1969) were determined using the equations in Table 2. Species diversity indices (species diversity, richness, and evenness) for the 4 life forms (hemicryptophyte, therophyte, chamaephyte, and geophyte) were obtained using Past 3.1 software (Øyvind Hammer, Natural History Museum, University of Oslo, Oslo, Norway, 2013). Regression models were fitted to assess the interrelations between elevation and the 3 diversity indices (diversity, richness, and evenness) and IVI of total plant and variety life form. For each model, the usual statistics such as R 2 and partial coefficients were obtained. Regression analyses were performed using Minitab 17 (Minitab Inc. State College, PA, USA, 2013).
Of the 251 species that were recorded, 44 species were class I, 55 species were class II, and 152 species were class III (  Table S1; https://doi.org/10.1659/MRD-JOURNAL-D-18-00089.1.S1). The IVI of species class I declined with increasing elevation and then increased, while species class III showed the reverse trend ( Figure 2).

Diversity index
Analyses of the diversity of species showed that diversity indices were significantly different across the elevation gradient ( Figure 3). Diversity and species richness showed similar trends. They increased significantly with increasing elevation and then declined (Figure 4). There was no significant difference in species evenness along the elevation gradient ( Figure 5).
The dominant species of hemicryptophytes with high IVI values were Festuca ovina and Colpodium parviflorum. Silene sp and Astragalus stevenianus were less abundant hemicryptophytes, with the lowest IVI in the study sites ( Figure 6; Supplemental material, Table S1; https://doi.org/10. 1659/MRD-JOURNAL-D-18-00089.1.S1). A statistically significant second-order polynomial interrelation between species richness, evenness, and IVI with elevation was found, but the interrelation between diversity and elevation was not significant. Species richness and evenness were normally distributed along the elevation gradient, while IVI increased with increasing elevation and showed J-shaped interrelations with elevation. The diversity of hemicryptophytes increased with elevation but was not significantly different across the elevation gradient (Figures 3-5).
The most abundant therophytes of the study area were Trifolium subterraneum and Medicago minima, with high IVI. The rarest therophytes recorded with minimum IVI were Linaria simplex and Malva nicaeensis ( Figure 6; Supplemental material, Table S1; https://doi.org/10.1659/MRD-JOURNAL-D-18-00089.1.S1). The overall pattern of therophyte IVI shows an inverse interrelation with increasing elevation. A statistically significant second-order polynomial interrelation between diversity, species richness, and evenness with elevation was found for therophytes. Diversity and species richness were normally distributed along the elevation gradient ( Figures  3-5).
The dominant species of geophytes were Poa bulbosa and Colchicum trigynum, while the rare geophytes were Allium affine and Bongardia chrysogonum, which had the lowest IVI in the region ( Figure 6; Supplemental material, Table S1; https://doi. org/10.1659/MRD-JOURNAL-D-18-00089.1.S1). Species richness and evenness decreased with elevation, while diversity and IVI increased with increasing elevation; however, neither of these indicators was significantly different across the elevation gradient (Figures 3-5).

Interrelation between species composition and elevation gradient
The most species-rich families were Asteraceae, Poaceae, and Fabaceae. The dominance of these families is due to their high compatibility with the semiarid climate conditions (Ghafari et al 2018). Several studies ( 2018), which results in their dominance in all selected sites. However, members of other families were restricted to a particular elevation, suggesting FIGURE 2 Variation in IVI by palatability class along the elevation gradient of the Moghan Plain-Sabalan Mountain rangelands. Palatability classes: highly palatable (class I)-forage plants, most preferred by domestic livestock; mostly palatable (class II)-plants averagely preferred by livestock; hardly or unpalatable (class III)-plants least preferred or not grazed by livestock at any stage; and harmful and possibly toxic. ns indicates no significant difference.*Significant difference at P , 0.05. **Significant difference at P , 0.01.  that specific physiographic and elevation conditions support different species. Asteraceae was the most species-rich family, showing high compatibility with the difficult conditions (shallow soil depth, steep slopes, and high solar radiation) in cold arid and semiarid climates Taheri Niari et al 2018).
Some species were mainly distributed at lower elevations, such as Pimpinella kotschyana, Lepidium draba, Phlomis herbaventi, and Senecio glaucus, whereas Ficaria kochii, Agrostis canina, and Lolium rigidum were recorded at the higher elevations. Plant species composition gradually changed as a function of variation in environmental factors; therefore, species distribution and composition differ across elevation classes along an ecological gradient (Rahman et al 2016). The species distribution showed that no single species has a wide distribution along the whole elevational range. Poa bulbosa (300-3000 masl) and Festuca ovina (900-3300 masl) were the only species that had a wide range of compatibility over the study area. The occurrence of these species across the elevation gradient suggests their wider ecological amplitude and tolerance to biotic pressures, in addition to their good reproductive capacity (Gairola et al 2011).
The IVI of species differed along the elevation, which indicates the relative importance of each species in the spectrum of elevation classes (Pandey 2015). Among the species with a high IVI, Poa bulbosa, Festuca ovina, Medicago minima, Artemisia austriaca, and Bromus tectorum spanned the widest elevational range. This indicates that they are the most compatible with available resources. Some species showed no distinctive distribution patterns along the gradient; some were distributed over a restricted elevational range, whereas others showed wider elevational ranges. The number of rare species in this study confirms the commonly cited concept that most of the species in plant communities are rare, rather than dominant (Magurran 2004). The rarity of species may be due to (1) strong density dependence (population growth rates are regulated by the density of a population); (2) uneven distribution of a resource, which leads to occupation by species at various locations resulting in variation in abundance and distribution; (3) natural or anthropogenic disturbances; and (4) competition effects (Scherrer and K€ orner 2011;Kacholi 2014;K€ orner et al 2017).

Interrelation of diversity, species richness, and evenness with elevation gradient
Species richness and diversity significantly increased with increasing elevation to 1200-1500 masl but then start to decline to 3300 masl. The increase in diversity value and species richness at mid-elevations could be because of (1) appropriate environmental conditions, such as low thermal and water stress and high humidity; (2) overlapping habitats; (3) the wide range (larger area) and proportionally more habitat area availability; and (4) potential high productivity (Ren et al 2012).
Due to the optimal combination of resources, more species coexist in the mid-elevation ranges (Ahmad Dar and Sundarapandian 2016); thus, more species were found in this elevation range. A positive impact of increased precipitation combined with the negative impact of decreased temperature produces a complex double gradient resulting in an optimum climate at the mid-elevation that favors diversity and species richness. While the limitations of low precipitation at low elevations or low temperature at high elevations would lead to different low-diversity ecosystems, dominated respectively by species adapted to limited precipitation or low temperatures (Sanchez et al 2013;Moradi et al 2017;K€ orner and Paulsen 2017).
Overlapping habitats and resources in mid-elevation areas could be partially responsible for the high diversity and species richness at this elevation range, which may indicate the existence of a transitional region for plants located at the mid-elevation (Senbeta et al 2014). Thus, in the midelevation a mix of high-elevation and low-elevation species give greater species richness than at higher or lower elevations (Ghafari et al 2018). The maximum diversity at mid-elevations might be attributed to the intermediate disturbance hypothesis. which suggests that the highest diversity will occur at levels of intermediate disturbance (Connell 1978;Mohamed Neji et al 2018). Most residential areas and agricultural lands are at low elevations, where there is a heavy disturbance. At higher elevations, degradation involves seasonal grazing and tourism activities; thus, disturbance is less than at low and mid-elevations. The reasoning is that diversity and species richness should be increased at median levels of disturbance (mid-elevation) because at high-disturbance levels (low elevation) most species will go locally extinct, and only the most resistant species will survive, whereas, at low levels of disturbance (high elevation), competitive exclusion increases and excludes other species (Zhang et al 2013). At the median disturbance, diversity increases because of disturbancetolerant species and because highly competitive species coexist (Lee and Chun 2016).
Usually, with an increase of elevation, the total habitat decreases (McCain and Grytnes 2010;Noroozi and K€ orner 2018). Hence, the wider area in the lower elevations would provide a larger area for species distribution (Lomolino 2001). Though the area was larger at the lower elevation gradient, because of high temperature, low humidity, high human population density, changes in land use, road construction, overgrazing, and conversion of rangelands for FIGURE 6 Variation in IVI by life form along the elevation gradient of the Moghan Plain-Sabalan Mountain rangelands. Life forms: He indicates hemicryptophyte; Th, therophyte; Ge, geophyte; Ch, chamaephyte. ns indicates no significant difference. *Significant difference at P , 0.05. **Significant difference at P , 0.01. agricultural purposes, only a few plant species are favored, with correspondingly decreased diversity and species richness (Nogu es-Bravo et al 2008;Noroozi et al 2018).
The effect of fewer surfaces with increasing isolation results in a decreasing number of species at higher elevations (Lomolino 2001). In addition, the reduction in diversity and species richness at higher elevations could be related to ecophysiological constraints and harsh climatic conditions (extremely low temperature and strong winds; increase in snow cover and solar radiation), nutrient availability, water holding capacity, short growing seasons, geographical obstacles, and low ecosystem productivity (Mahdavi et al 2013;Sanchez et al 2013;K€ orner and Paulsen 2017). These restrict the physiological development of species to adapt and survive (Namgail et al 2012). The degree of disturbance decreased as elevation increased because tourism density and agriculture activities decreased with increased elevations (Sharifi et al 2017). However, heavy snowfall in winter, high grazing intensity in early spring and summer, and uncontrolled collection of medicinal and fodder plants has decreased species diversity (Sharifi et al 2017). With increasing elevation, people are increasingly dependent on rangelands because access to goods and services decreases (Thapa et al 2016). Sources of income and occupations other than livestock husbandry in the MPSMRs are rare, and, because of this, the number of livestock per unit area held by each household has increased (Sharifi et al 2017).
The evenness index showed no significant difference along the elevation gradient. Although the total evenness of all plants showed some fluctuation, there was no clear peak. According to Huang et al (1997), variations in the evenness of plant communities could be independent of the environmental gradient.
Hemicryptophytes were dominant (the highest IVI) at higher elevations. The change in dominance of life forms with elevation is probably due to the greater frequency of some species along the elevation gradient. For instance, the abundance of Festuca ovina, Colpodium parviflorum, Alopecurus vaginatus, Trifolium medium, and Poa pratensis resulted in a large number of hemicryptophytes at 2100-3300 masl elevation. Mota et al (2017) and Moradi and Attar (2019) found that hemicryptophytes are the most dominant life forms at high elevations. The lower disturbance at higher elevations favors hemicryptophytes. Thus, elevation and human activities play an important role in determining the diversity, life-form, and biological distribution spectrum of a habitat (Ren et al 2012).
The high proportion of therophytes in the investigated area reflects the cold winters and dry summers (Aghaei et al 2013). The spread of therophytes in the study area, especially unpalatable weeds, is likely a response to overgrazing (Balpinar et al 2018;Gamoun et al 2018). The preponderance of therophytes at mid-elevations might be the result of agricultural activities, overgrazing, and human activities (Ahmadi et al 2013). The dominance of therophytes at lower elevation can be partly explained by low soil moisture. At the lower elevation, although temperatures and growing season length are suitable, soil moisture is low (Barone et al 2008). Regardless of the type of ecosystem, overgrazing, especially at low elevations, led to the predominance of invasive and weed therophytes, indicating degradation (Carey et  The higher percentage of chamaephytes in the midelevation may be related to the impact of human activities (Balpinar et al 2018). Chamaephytes affect other species associated with stress tolerance and a strong competitiveresponse ability (Arila and Gupta 2016). Additionally, most chamaephytes are relatively resistant to livestock grazing (Terradas 1991) because of protective or defensive organs (such as spines, hairs, thorns, bitter taste, and unpleasant odor) and unpalatability. The species that provide the greatest protection to the buds are compact cushion plants, such as Acantholimon, Astragalus, and Onobrychis (Balpinar et al 2018;Moradi and Attar 2019). The higher elevation is characterized by higher solar radiation, which is unfavorable for the development of chamaephytes (Traquillini 1979).
The low proportion of geophytes in the region indicates that they are not adapted to the climatic conditions. The dominance of geophytes at higher elevations results from the high average annual rainfall, lengthy wet period during the growing season, and long duration of snow cover (Naqinezhad and Esmailpoor 2017). According to Meç o et al (2018), geophytes are mostly associated with cold winters.

Plant palatability variation with elevation gradient
IVI species class I declined as the elevation increased (1800-2100 masl) and then increased, while species class III showed a reverse trend. At lower elevations, the rangeland conversion to agricultural land decreases rural livestock grazing intensity. Most of the rangeland is grazed by nomads' livestock. In addition, at low elevations, fodder crops, such as Medicago, Trifolium, and Onobrychis, are planted as livestock forage. Moreover, post-harvest crop residues are given to ranchers during the dry season for hay feeding, which provides part of the livestock needs in these areas. Therefore, in these regions, the presence of plants with good and medium palatability is higher than class III plants. In other words, although the diversity and species richness are low, rangeland degradation is medium at low elevation.
Optimal humidity conditions at mid-elevations and the high productivity in these habitats resulted in an optimal combination of resource availability (Hua 2004). Because of human population density, excessive grazing (rural and nomads' livestock), and conversion of rangelands for agricultural purposes, there is more degradation at midelevations. Although diversity and species richness increased, the presence of class III unpalatable plants was greater than the presence of plants with good and medium palatability at the mid-elevation. This is consistent with other studies that found that heavier grazing intensity was associated with higher species richness (Espinoza et al 2017). Moreover, many plants were found to be highly preferred, whereas some plants were avoided by the grazing livestock, leading these species to be widespread. This leads to the dominance of many unpalatable species, such as Astragalus microcephalus, Bromus arvensis, Bromus tectorum, Taeniatherum caput-medusae, and Artemisia austriaca as well as the disappearance of many other species that are less well adapted to unfavorable conditions and intense biotic influences. According to Renne and Tracy (2007), unpalatable plants increased in sites under grazing pressure. The major palatable species in the mid-elevation are Astragalus stevenianus, Astragalus pinetorum, Sanguisorba minor, Trifolium repens, and Trigonella arcuata, which are scarcely present at mid-elevations. Because they are highly palatable, these species have been greatly affected by grazing pressure. Most of the rare species present at midelevations were palatable and faced the excessive pressure of grazing. The grazing caused substitution of palatable plants by unpalatable plants, affecting abundance, dominance, plant frequency, and thus overall plant species composition (Espinoza et al 2017).
At higher elevations, the dominance of palatable species might be due to low human population density, lack of cultivation and agricultural activities, and limited seasonal grazing due to the shortened growing period (mid-May to mid-September). This rangeland is also only grazed by nomads' livestock. Nogu es-Bravo et al (2008) explained that the higher elevations probably harbor large numbers of species because anthropogenic activities and destruction decrease with increased elevation.

Conclusion
Results showed that species diversity in plant communities of the region was considerable. Species richness and diversity peaked at mid-elevations along the elevational gradient. However, relative richness and diversity of species within different life-form groups varied differently along the gradient. The IVI of species differed along the elevation, indicating the relative importance of each species in a spectrum of elevation classes. The results showed that the analysis of plants using diversity indicators did not provide sufficient information on the actual condition of the rangeland plants. We believe the present approach should be encouraged in future studies for better understanding the structural status of rangelands using both species diversity and ecological importance indices of plants. By considering the results of this study, a proper conservation strategy, multipurpose use, and comprehensive regional planning could be developed to promote the sustainable use of MPSMRs.