Influence of Terrain on the Distribution of Indian Wild Ass, Equus hemionus khur, in the Great Rann of Kachchh, Gujarat, India

Technical Summary: Influence of Terrain on the Distribution of Indian Wild Ass in the Great Rann of Kachchh

Problem Statement
The Indian Wild Ass (Equus hemionus khur) is restricted to the Rann of Kachchh ecosystem in Gujarat, India. While the species is known to inhabit the broader landscape, field observations suggest a non-uniform distribution heavily skewed toward flat terrain. Despite existing documentation of the species' presence in the Little Rann, the specific influence of terrain structure (flat vs. undulating vs. rough) on habitat selection and spatial ecology remains poorly understood. This study addresses the hypothesis that the Indian Wild Ass utilizes all major terrain types uniformly, testing instead whether terrain complexity acts as a primary driver of habitat suitability and distribution.

Methodology
The research employed a two-phase field design across the Great Rann and adjoining parts of the Little Rann of Kachchh. The study area was categorized into three terrain types: Flat (saline mudflats and gentle plains), Undulating (elevated bets or islands), and Rough (irregular, rocky surfaces with strong micro-relief).

• Phase I (Exploratory): Systematic transects (n=41) and opportunistic sightings were conducted across three seasons (winter, summer, post-monsoon) to establish baseline occurrence patterns. This phase yielded 150 independent sightings. Data collection included GPS coordinates, terrain classification, vegetation, water availability, human disturbance, and herd size.
• Phase II (Hypothesis-Driven): A structured sampling design was implemented using 210 systematically distributed points (100 flat, 60 undulating, 50 rough) to test habitat differences. Presence/absence and herd size were recorded at each point.
• Data Analysis:
  • Association: Chi-square ($\chi^2$) tests were used to determine the statistical relationship between terrain type and species occurrence.
  • Selection Ratios: Manly's selection ratio ($W_i$) was calculated to quantify habitat preference ($W_i > 1$) versus avoidance ($W_i < 1$).
  • Predictive Modeling: Binomial logistic regression was applied with presence/absence as the response variable and terrain attributes (slope, terrain category) as predictors. Model performance was evaluated using the Area Under the Curve (AUC).

Key Results
The study results strongly rejected the hypothesis of uniform distribution across terrain types.

• Occurrence Probability: A significant disparity was observed in occurrence probabilities: Flat terrain (P = 0.82), Undulating terrain (P = 0.30), and Rough terrain (P = 0.20).
• Statistical Significance: Chi-square analysis confirmed a highly significant association between terrain type and occurrence ($\chi^2 = 61.44$, $p < 0.001$).
• Habitat Selection: Manly's selection ratios indicated a clear preference for flat terrain ($W_i = 1.72$), near-neutral use of undulating terrain ($W_i = 1.05$), and avoidance of rough terrain ($W_i = 0.84$).
• Herd Dynamics: Herd size correlated inversely with terrain complexity. Mean herd sizes were highest in flat areas (33.1 individuals) and declined significantly in rough terrain (13.5 individuals).
• Predictive Modeling: Logistic regression identified slope as a strong negative predictor of suitability ($\beta = -1.12$), while flat terrain showed a strong positive association ($\beta = +1.85$). The model achieved a high predictive accuracy (AUC = 0.89), suggesting terrain variables alone are sufficient to explain a substantial portion of the distribution pattern.
• Spatial Clustering: High-density records were concentrated in flat zones (e.g., Gadka bet, Bhojardo), while rough terrain locations (e.g., Dholavira, Tragadi bet) supported fewer individuals. No sightings were recorded in highly undulating or hilly zones.

Key Contributions

1. Quantification of Terrain Influence: The study provides empirical evidence that terrain structure is a primary ecological determinant for the Indian Wild Ass, moving beyond general habitat descriptions to specific terrain-based selection metrics.
2. Habitat Hierarchy Definition: It establishes a clear hierarchy where flat terrain constitutes the optimal niche, undulating terrain serves as marginal habitat, and rough terrain functions as suboptimal or avoided habitat.
3. Linking Morphology to Distribution: The findings link the species' cursorial adaptations (morphology and behavior) directly to landscape use, demonstrating how the single, compact hoof structure of equids favors firm, open substrates for efficient locomotion and predator detection.
4. Predictive Framework: The development of a logistic regression model with high AUC demonstrates that terrain complexity can reliably predict species presence without requiring additional environmental variables.

Significance and Claims
The paper claims that the Indian Wild Ass is a habitat specialist strongly dependent on open, flat landscapes within the Rann ecosystem. The study concludes that terrain acts as the primary ecological driver shaping the species' spatial ecology. The observed preference for flat terrain is attributed to the energetic efficiency of movement, enhanced visibility for predator detection, and the facilitation of coordinated herd behavior. Conversely, increased terrain complexity imposes energetic costs and physical constraints that limit both occurrence and social group size.

The authors assert that these findings reinforce the role of terrain-driven habitat selection as a critical factor in the conservation planning of the Indian Wild Ass. By identifying rough and highly undulating zones as ecological barriers rather than functional habitat, the study suggests that conservation efforts should prioritize the protection and management of flat saline plains and gentle slopes, which represent the species' primary ecological niche. The results align with broader ecological patterns observed in the genus Equus, confirming that open habitat specialization is a defining characteristic of these species.