Ethnopharmacological study of native medicinal plants and the impact of pastoralism on their loss in arid to semiarid ecosystems of southeastern Iran

The purpose of this study was to gather ethnopharmacological information on plants used by the pastorals of southeastern Iran. The relationships between ecological value of the plant species and ethnobotanical indices were investigated. The loss of medicinal plants and its effective factors were also determined under nomadism and sedentary pastoralism. Ethnopharmacological information of plants was collected through interviews with 85 local people including nomads (43%) and sedentary pastorals (57%). Ethnobotanical indices including relative frequency of citation (RFC), relative importance (RI), cultural value (CV), and use value (UV) were estimated. Canopy cover and density of plant species were measured at 60 sampling plots in the exclosure, nomadic rangelands and sedentary pastorals rangelands. The Importance Value Index (IVI) and Relative Loss Index (RL) were estimated for both nomadic and sedentary pastoral rangelands. Pearson correlation coefficient was used to investigate the relationship between ethnobotanical indices and IVI of plant species. The Bayesian networks was used to investigate the relationship between ethnobotanical indices and plant species loss. In total, 156 medicinal plant species of 50 families were identified in the region by locals. Positive correlation was observed between ethnobotanical indices (RFC and RI) and ecological index (IVI). The mean decline of the ecological importance of medicinal species in sedentary pastoral rangelands was approximately three times higher than in nomadic rangelands. Bayesian networks showed that cultural value, seed exploitation and aerial parts exploitation had direct relationships with species loss in both nomadic and sedentary pastoral rangelands. Bunium persicum, an ecologically and socially important species, had been extensively harvested (more than 60%) in the both nomadic and sedentary pastoral rangelands, making it a priority species in future conservation programs. Medicinal plants had high ecological value and were severely exploited, threatening sustainability of arid and semiarid ecosystems. Local pastorals not only use medicinal plants as herbal remedies but also consider them as a source of income. Popular plants with multiple medicinal uses were more susceptible to loss. Higher medicinal knowledge of pastorals did not help to mitigate medicinal plant loss, requesting new plans to aware them to the circumstances that often leads to species removal from community. Given the importance, abundance and widespread use of medicinal plants, further studies can provide a basis for their conservation and for identifying new therapeutic effects of plants in the region.

where in the first factor, NUs the number of different uses cited for the species and NC is the total number of use-categories. FCs is the relative frequency of citation of the species in the second factor. Finally, the third factor is the sum of number of participants who mentioned each use of the specie (URui) divided by N 23 .
Relative importance index (RI). This index was calculated using the following formula.
where RFC s(max) is the relative frequency of citation over the maximum, i.e., and was measured with FCs divided by the maximum value in all the species of the survey. RNU s is the relative number of use-categories over the maximum and it was calculated with the number of uses of the species (NU s ) divided by the maximum value in survey 24 .
Ecological data of medicinal plants. Ecological data were collected in three adjacent sites (inside the park, outside the park including both nomad and sedentary pastoral rangelands). Sites were selected in homogeneous areas with similar topography and ecological properties. The vegetation survey was carried out using 180 nested quadrats, which randomly located in the sites to show the loss of plant species under different pastoralism types. As a constant quadrat size may not be appropriate or all plant species with different sizes, sixty 10 × 10 m quadrats were used to sample trees and shrubs and 1 × 1 m sub-quadrats were used to sample semi-shrubs and herbaceous plants in each site. Canopy cover and individual numbers of plant species were recorded in each plot. Importance value index (IVI) provides information about the ecological importance of a species in plant communities 25 . IVI was used to determine ecological value of medicinal plant in our study. Since species density, frequency and dominance comprised IVI are sensitive indicators to anthropogenic actives 26,27 . Reaction of species to human exploitation can be revealed with comparing IVI with and without exploiting 28,29 . Relative loss (RL) index was measured to assess ecological situation of medicinal plants under both nomadic and sedentary pastoral system using changes IVI.
Index of RL of species s was calculated using the following formula: where IVI is the importance value index. Vegetation composition was evaluated by analyzing the frequency, density, dominance and IVI, using the following formula 30,31 .
Relative frequency = Frequency of a species/frequency of all species * 100, Density = Total no: of individuals of a species/total no. of quadrats studied, Relative density = Number of individuals of a species/number of individuals of all species * 100, and Relative Dominance = Canopy cover of a species/Canopy cover of all the species * 100. Data analysis. Bayesian networks (BNs) were used to investigate the relationship between ethnobotanical indices and relative loss of species for both pastoralism rangelands. The variations of the probability of species loss was estimated under different scenarios.
BNs are a general way to find important paths in a network that are usually not easily estimated by mathematical equations. The calculations in the BNs are estimated using Bayes' theorem 32 . According to Bayes' theorem, a prior probability represents information about the initial uncertainty of a parameter. However, the posterior probability is estimated using the observed data and its likelihood function to update the uncertainty distribution of the parameters. Bayes' theorem updates probability of each factor in the network as follow 33 :

Results
Respondents' socio-demographics. All of the 85 participants who randomly selected in the study (55 men and 30 women) were locals. The participants were belonged to two pastoralism types i.e. nomadism (43%) and sedentary pastoralism (57%). The focus was on older generations as the holders of traditional knowledge. Therefore, 98% of the participants were more than 45 years old and 60% were over 60 years old (Table 1).

Medicinal plants.
In total, 156 medicinal species of 50 families with medicinal uses were identified by pastorals ( Table 2). The families Asteraceae, Lamiaceae, Apiaceae, and Fabaceae were the most abundant families in terms of medicinal species in the region (Fig. 1). plant parts used. The plant parts used by pastorals for treatment include stems, flowers, seeds, fruits, roots, gums, leaves, and whole aerial parts, of which aerial parts (25%), leaves (24%), and fruits (24%,) were the most abundant parts used (Table 3). Medicinal uses of the species were divided into 16 different categories, of which the uses for digestive system (50% species), immune system (23% species), sedative (20% species) had the most frequencies (Table 3). the relationships between ecological and ethnopharmacological indices. In terms of importance value index, A. sieberi (IVI = 0.55), B. persicum (IVI = 0.54) and R. pallasii (IVI = 0.54) had the highest values in the study area (Fig. 2). Importance value index was significantly correlated with RFC and RI indices (Fig. 3, p < 0.01).
The values of relative loss index in the rural area varied between 0.01 and 0.84, and Z. tenuir (RL = 0.84), B. persicum (RL = 0.83), and Z. clinopodioides (RL = 0.76) had the highest values (Fig. 2). However, the values of RL index in the nomadic rangelands varied from 0.03 to 0.75, which T. fedtschenkoi (RL = 0.75), M. sativa (RL = 0.73) and Z. tenuir (RL = 0.68) had the most relative loss index. (Fig. 2). There were significant differences between pastoralism types (sedentary pastoralism and nomadism) in the relative loss of medicinal plant species (Fig. 4). The relative loss indices of medicinal species were 0.648 ± 0.222 and 0.223 ± 0.212 in sedentary pastoral and nomadic rangelands respectively, indicating more species are removed under the sedentary pastoralism.    (Fig. 5). There are 15 nodes in the network. The probabilities that relative loss of species being high and low were 87.2% and 12.8% in sedentary pastoralism, respectively. The probabilities that relative loss of species being high and low were 71.9% and 28.1% in nomadism, respectively. Sensitivity analysis revealed the most important variables affecting the relative loss index ( Table 4). The CV, seed, aerial part and RI were the most important variables affecting the relative loss index whose effects on probabilities related to the relative loss were investigated under five different scenarios.
Under scenario 1, by changing the probability of CV index, the effect of increasing this variable on probability of relative loss index was investigated. The results showed that the probability of relative loss was increased to 97.86 and 88.45 for sedentary pastoralism and nomadism respectively, indicating a direct relationship between the relative loss and CV indices. Therefore, harvesting plant organs is one of the main causes of plant loss in both pastoralism types (Table 5). By changing the probability of seed, aerial part, and RI index, the effects of these variables increase on probability of relative loss index was investigated under the scenarios 2 to 4. The probabilities of relative loss index were respectively increased 95.23%, 90.04% and 89.03% in sedentary pastoralism and 83.03%, 78.12% and 75.82% in nomadism under scenarios 2 to 4 (Table5), indicating a direct relationship between the relative loss index and seed, aerial part, and IR index.

Discussion
The region had a rich medicinal flora, which mostly belonged to Lamiaceae and Asteraceae families. The rich flora can be related to the diverse topo-climatic zones in the region from warm low elevation plains in the south to snow caped mountains and cold high elevation plains in the north, providing habitat for different plant species. Lamiaceae and Asteraceae are two main medicinal families in Iran 35,36 . The dominance of these two families can be attributed to their widespread distribution in the area and their diverse traditional medicinal uses for the pastorals.
Decoction was the most common method of preparation in the region. Decoction is a powerful method for extracting the active ingredients of medicinal plants 37 , making it a pluralistic approach to faster and better treatment between local families 38,39 .
The medicinal plants were mostly used for digestive system between pastorals. Given the high incidence of gastrointestinal diseases among the population, there is more interest in treating it by locals 40 .
Local people mostly used leave of medicinal plants for health care. The leaves are usually easier to harvest than other plant organs and can be usually eaten directly as medicines. Leaves are rich in phytochemicals 41 , resulting in wide medicinal values 42,43 .
Based on ethnopharmacological indices, Z. multiflora, D. sophia, C. intybus, and B. persicum were the most important medicinal plant species for the pastorals. These species are completely known as herbs (Z. multiflora, D. sophia, C. intybus, and B. persicum) or spices (Z. multiflora and B. persicum) across the country 44,45 . Therefore, it is obvious they attract locals' attraction. ecological value of medicinal plants. Our resulted showed that popular medicinal plants had higher dominance and abundance in ecosystem. Past studies also found a positive relationship between ecological value and traditional medicinal use of plant species 46,47 . Common species are more mostly used as medicines than rare species 48 . A number of studies suggested a negative relationship between ecological important of species and its medicinal use 49 .  50 . Therefore, conservation of such species is more important in ecosystem management than other species.

Medicinal plants loss and effective factors.
Medicinal plants were strongly threated in arid and semiarid ecosystems of the region. About 15% of medicinal plants lost more than 50% of their ecological importance in pastoral rangelands due to overexploitation. Locals are mostly poor people whom their livelihood strongly   17 . Selling of medicinal plants has become an alternative and additive income source for pastorals, especially whom with forage deficiency resulted from recent droughts. We found medicinal plants with higher social value are particularly vulnerable to overexploitation. The harvest of species with high use popularity may encounter sustainability problems 51 . The plant part used and exploitation method can effect on medicinal species loss. Harvesting of plant seed and whole of aerial parts endangered sustainable use of medicinal plants 41,52 . The ecological importance of B. persicum as one of the medicinal plants with the highest CV, has declined 80% and 60% in the both pastoral rangelands, due to over-collecting of its seeds. Over-exploitation of aerial pats of Z. tenuir, T. fedtschenkoi, M. sativa and Z. clinopodioides has caused depletion of these plants in the both pastoral rangelands. These species were popular medicinal plants with relative importance more than 60% for social system. Reduction of photosynthetic ability linked with the loss of leaf area of plants can restrict plant growth in ecosystems 53      www.nature.com/scientificreports/ due to limited exploitation months compared to sedentary pastoralism in which all year round the rangelands are exploited 56 . Nomads are in fact the real rangelands dwellers who have historically adapted to the rangelands 17 and have more experiences and information about the medicinal properties on plants 57,58 . conclusions This study highlights the presence of ethnobotanical knowledge in southeastern Iran and the occurrence of native medicinal species as a key factor in their potential use and locals' attention. Given the abundance and widespread use of medicinal plants, further studies can provide a basis for identifying new therapeutic effects of plants in the region. Popular plants with multiple medicinal uses were more susceptible to loss. Higher medicinal knowledge of pastorals did not help to mitigate medicinal plant loss, requesting new plans to aware them to the circumstances that often leads to species removal from community. Given the abundance and widespread use of medicinal plants, further studies can provide a basis for identifying new therapeutic effects of plants in the region.

Data availability
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.
Received: 11 May 2020; Accepted: 28 August 2020 Scientific RepoRtS | (2020) 10:15526 | https://doi.org/10.1038/s41598-020-72536-z www.nature.com/scientificreports/ Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat iveco mmons .org/licen ses/by/4.0/.