Land use modeling and habitat quality assessment under climate scenarios: A case study of the Poyang Lake basin

• The proportion of built-up land area increased from 1.58% to 3.32% in 1990–2023. • The expansion of built-up land primarily encroached upon forest and farmland. • Projections show that forest and farmland will continue to decrease through 2050. • Habitat quality is projected to deteriorate further from 2023 to 2050. • Habitat quality scores are strongly correlated with plant species richness. Habitat quality influences the capacity to preserve biodiversity, typically assessed through land use data. The Poyang Lake basin is highly biodiverse yet heavily affected by human activities. This study aims to investigate the spatial and temporal trends in land use and habitat quality within the Poyang Lake basin from 1990 to 2050. Land use, environmental, and socio-economic data from 1990 to 2020 were employed to calibrate the System Dynamics and PLUS models for predicting the land use patterns up to 2050. The InVEST model was employed to assess habitat quality based on both historical and projected land use data. Habitat quality assessment was validated using species richness of vascular plants estimated from field-sampled specimens. Our findings reveal a rapid increase in built-up land from 1990 to 2023, while forest and farmland areas have generally decreased. These land use trends are expected to continue from 2023 to 2050. Habitat quality was generally good from 1990 to 2023, but showed a declining trend. Projections suggest a further decline in habitat quality from 2023 to 2050, especially under the SSP5-8.5 scenario. Strong correlations were found between habitat quality scores and species richness of vascular plants (0.58 ≤ Pearson’s r ≤ 0.64), confirming the effectiveness of land use data for habitat quality assessment. Our results underscore that optimized land use planning—such as controlling the expansion of human-altered land and conserving natural vegetation—is crucial for biodiversity conservation, and also suggest that land use data can be a valuable tool for biodiversity monitoring.