Modeling Ecosystem Dynamics and Services in Abijata-Shalla Lakes National Park under Pressures from Land Use and Climate Change

• Integrated SAHQ modeling projects substantial habitat and biodiversity decline by 2054 • Species richness declines by 57.3% and ecosystem integrity (SAHQ) by 32.6% by 2054 in ASLNP • Very high-priority conservation zones are projected to disappear under future scenarios • Water yield shows strong spatial variability, suggesting shifts in hydrological patterns • SAHQ achieved 87% agreement (spatial coincidence) with independent validation data Land use and land cover change (LULCC) and climate variability are key drivers of ecosystem degradation in semi-arid protected areas, yet their combined impacts on biodiversity and ecosystem services are rarely assessed within an integrated framework. This study develops a Species-Adjusted Habitat Quality (SAHQ) index that combines species distribution modeling (MaxEnt), habitat quality (InVEST), and water yield modeling to evaluate ecosystem dynamics and conservation priorities in Abijata-Shalla Lakes National Park, Ethiopia. Historical LULC changes (1994–2024) and future projections (2054) were analyzed using a CA-Markov model, while species distributions for focal waterbird species were modeled under current and projected climate scenarios. Results indicate substantial habitat degradation and declining species richness, with land-use change accounting for the majority of impacts (68%), followed by climate change (22%), and their interaction (10%). Water yield exhibited pronounced spatial variability, reflecting the sensitivity of hydrological processes to vegetation and climate interactions. The SAHQ framework revealed a significant contraction of high-priority conservation areas, with near-complete loss projected under future conditions. Model performance and prioritization outputs were evaluated using a multi-metric validation framework, including Area Under the Curve (AUC), True Skill Statistic (TSS), omission rates, spatial cross-validation, field-based habitat assessments, and expert evaluation. Results demonstrate strong model reliability and spatial consistency across validation approaches. The proposed framework provides a transparent and reproducible tool for integrating biodiversity and ecosystem service assessments, supporting evidence-based conservation planning under coupled land-use and climate pressures in data-limited semi-arid regions. The SAHQ index demonstrated strong robustness to weighting scheme variations (Kendall's W = 0.89), confirming its reliability for conservation prioritization.