تأثیر عوامل محیطی بر الگوی پراکنش Astragalus adscendens در استان لرستان با استفاده از مدل MaxEnt

Introduction

The distribution and long-term persistence of medicinal plant species in Iran are strongly influenced by a complex interaction between natural processes and anthropogenic pressures. In recent decades, unsustainable harvesting practices, extensive land-use changes, overgrazing, habitat fragmentation, and ecosystem degradation have led to a substantial decline in the population size and geographical range of many valuable medicinal plants. These pressures have been further intensified by the absence of comprehensive national strategies for the identification, conservation, and sustainable utilization of native species. Inadequate ecological datasets, limited long-term monitoring programs, and the lack of reliable spatial distribution maps have collectively constrained effective planning and management of medicinal plant resources throughout the country.

In addition to direct human impacts, climate change has emerged as a major driver of ecological transformation, altering temperature regimes, precipitation patterns, and the frequency of extreme climatic events such as prolonged droughts. These changes have significantly modified habitat suitability for many plant species, resulting in the degradation, contraction, or displacement of their ecological niches. Mountain ecosystems, despite their ecological importance and biodiversity value, are particularly vulnerable to such changes due to their narrow climatic tolerance ranges and strong dependence on seasonal precipitation. Astragalus adscendens Boiss., an endemic perennial species of the Central Zagros Mountains, represents one of the ecologically significant legumes adapted to high-elevation rangelands. This species plays a vital role in soil stabilization, erosion control, and the maintenance of rangeland structure and functionality. Through its growth form and root system, it contributes to improved soil physical properties, enhanced water infiltration, and reduced surface runoff. Given its ecological importance and increasing exposure to environmental stressors, understanding the factors controlling the distribution of A. adscendens is essential for developing effective conservation and sustainable management strategies. Therefore, the primary objective of this study was to identify the key environmental variables governing the spatial distribution of A. adscendens and to model its potential habitat suitability in Lorestan Province, Iran, using the Maximum Entropy (MaxEnt) modeling approach.



Methodology

The potential distribution of Astragalus adscendens was modeled using MaxEnt version 3.4.4, a widely applied machine-learning method designed for species distribution modeling based on presence-only data. This approach estimates the most probable spatial distribution of a species by relating known occurrence points to a set of environmental predictors. Species occurrence data were collected during the spring and summer of 2025 from four major habitats in Lorestan Province, including Aligudarz (Qalikuh and Tamandar), Azna, Alashtar, and Noorabad. Geographic coordinates of each occurrence were recorded using a Global Positioning System (GPS), and relevant ecological characteristics such as elevation, slope, aspect, soil type, and vegetation cover were documented through field observations.

Following data verification and filtering to reduce spatial bias and overfitting effects, a total of 23 confirmed occurrence points were retained for final modeling. To address multicollinearity among environmental variables, Pearson correlation analysis was conducted, and variables with correlation coefficients greater than 0.8 were excluded. Ultimately, thirteen environmental variables were selected, including five bioclimatic variables related to temperature and precipitation and eight topographic and environmental variables describing terrain characteristics. Model performance was evaluated using the Area Under the Curve (AUC) of the Receiver Operating Characteristic (ROC). The relative contribution and importance of each variable were assessed using the jackknife test. Habitat suitability maps were generated in a Geographic Information System environment and classified into five suitability classes ranging from very low to very high.



Results and Discussion

The results indicated that the MaxEnt model achieved high predictive accuracy in identifying suitable habitats for A. adscendens across Lorestan Province. Among all environmental predictors, winter precipitation emerged as the most influential variable affecting species distribution. The highest probability of occurrence was associated with winter precipitation levels ranging from approximately 100 to 140 mm, whereas habitat suitability declined sharply at precipitation values exceeding 160 mm. This pattern reflects the species’ adaptation to semi-arid mountainous environments, where moderate winter moisture supports establishment and growth while excessive precipitation may create unfavorable soil conditions.Slope was identified as the second most important factor influencing habitat suitability. Optimal conditions were observed on moderately steep slopes (approximately 30–40%), particularly on north- and northeast-facing aspects. These terrain features promote greater soil moisture retention, reduce evaporation rates, and prevent prolonged waterlogging, thereby creating favorable microclimatic conditions for plant development. Temperature-related variables, elevation, and soil texture also contributed to shaping the distribution pattern, although their effects were secondary compared to precipitation and slope.

Among temperature variables, seasonal temperature variation and mean diurnal temperature range were particularly influential. The optimal mean temperature range for A. adscendens occurrence was estimated to be between 13 and 14 °C, while temperatures exceeding 15 °C resulted in a noticeable decline in habitat suitability. Elevation further constrained the species to altitudes between 2400 and 2700 meters above sea level, where cooler temperatures and relatively higher soil moisture availability support vegetation growth. In contrast, land-use and soil-related variables showed limited influence, indicating a relatively broad tolerance of the species to different soil conditions. The habitat suitability map revealed that approximately 95% of Lorestan Province is unsuitable for this species, with only about 27.2 km² classified as moderately to highly suitable, mainly in the Qalikuh, Tamandar, and Ashtrankuh highlands.



Conclusion

This study demonstrates that the distribution of Astragalus adscendens in Lorestan Province is primarily controlled by climatic and topographic factors, with winter precipitation and slope exerting the greatest influence. The species achieves its highest growth potential in cold, relatively dry mountainous environments characterized by moderate winter precipitation and warm summers. Given the dominant role of climatic variables, future changes in temperature and precipitation patterns may substantially alter the extent and spatial configuration of suitable habitats. Therefore, proactive conservation planning, protection of existing high-elevation habitats, sustainable grazing management, regulation of land-use changes, and continuous climatic monitoring are essential to ensure the long-term conservation and ecological stability of this species.