Detection of long-term land use and ecosystem services dynamics in the Loess Hilly-Gully region based on artificial intelligence and multiple models

Abstract

Balancing intensive development with biodiversity conservation is critical for ecological sustainability. In the Loess Hilly-Gully region, which has experienced significant environmental changes from 1980 to 2020, land use changes are vital for ecosystem service maintenance. However, these processes are susceptible to environmental stressors such as climate change, land degradation, and inadequate land management. Despite recognizing these stressors, their combined impact on land use and ecosystem services is not well understood. This study addresses this gap by utilizing multi-model land use simulations combined with artificial intelligence to analyze changes over the four-decade period. We focused on identifying ecological nodes essential for landscape connectivity, finding that these nodes significantly enhance regional ecosystem services. We also examined the relationship between land use changes and ecosystem services under various environmental stressors within the specified timeframe. Our analysis indicates that land use changes are a strong indicator of ecosystem evolution in these areas. We observed that under natural conditions, land use and ecosystem services interactions are governed by both stochastic and deterministic factors. Interestingly, multiple stressors intensified the stochastic nature of land use changes and the deterministic aspects of ecosystem services. Specifically, climate change, coupled with suboptimal land management, significantly affects this relationship, with climate change being the most influential factor on land use patterns during the studied period. This research offers important insights for improved environmental management and reducing impacts on aquatic ecosystems amid escalating climate change threats.