Ecological and Statistical Evaluation of Genetic Algorithm (GARP), Maximum Entropy Method, and Logistic Regression in...

Area of Science:

• Ecology
• Environmental Science
• Computational Biology

Background:

• Understanding plant species distribution is vital for effective rangeland management and biodiversity conservation.
• Astragalus sp. plays a significant role in rangeland ecosystems, necessitating habitat assessment.
• Predictive modeling aids in identifying suitable habitats and informing conservation strategies.

Purpose of the Study:

• To evaluate potential habitats for Astragalus sp. using species distribution models.
• To identify key environmental factors influencing the spatial distribution of Astragalus sp. in the Savar-Abad basin rangelands.
• To compare the performance of three modeling techniques: Maximum Entropy (MaxEnt), Genetic Algorithm for Rule-Set Production (GARP), and logistic regression.

Main Methods:

• Field sampling of vegetation and soil properties across diverse rangeland types.
• Generation of soil maps using interpolation techniques.
• Application and comparison of MaxEnt, GARP, and logistic regression models for habitat prediction.

Main Results:

• The Maximum Entropy (MaxEnt) model demonstrated superior performance in predicting suitable habitats compared to GARP and logistic regression.
• Soil salinity, elevation, and soil acidity were identified as significant factors influencing Astragalus sp. distribution.
• Elevation and soil salinity were found to have the most substantial impact on habitat suitability, with soil texture playing a secondary role.

Conclusions:

• Species distribution modeling, particularly MaxEnt, is effective for assessing Astragalus sp. habitat suitability.
• Environmental factors like soil salinity and elevation are critical determinants of Astragalus sp. distribution.
• Findings provide valuable insights for optimizing rangeland management, conservation planning, and biodiversity enhancement efforts.