Early warning and scenario simulation of ecological security based on DPSIRM model and Bayesian network: A case study of east Liaohe river in Jilin Province, China

Abstract

Ecological security early warning research is an effective tool for statistical unsustainable bottom line and watershed management. Recent research has shown that climatic elements influence the characteristics of ecological security and its variability and that landscape patterns constrain the basis of regional ecological security. Thus, there is a need to identify trends in ecological safety under the driving forces of climate and landscape pattern changes to enable early warning measures. In this study, a new ecological security early warning system (ESEW) was established, and a comprehensive index was constructed to assess the ESEW in the East Liaohe River Basin (ELRB), China during 2000–2020. The system used a geographic detector to reveal the internal influences, external drivers, and synergistic effects of ecological security warnings in the basin in different periods. A Bayesian network (BN) model was used to simulate the risk of different ecological security warnings occurring under different scenarios. The results showed that the ESEW level is higher in the northwestern part of the ELRB, and the area occupied by extreme warning decreased simultaneously as the area occupied by no warning decreased. The changes of ecological security warning situation in the ELRB are not only influenced by internal factors but also driven by external factors, and the synergy between the core factors and external factors can better explain the driving mechanism of ecological security alarm than a single factor. The BN model was more sensitive to the simulation of no warning and extreme warning. This study provided insights into the maintenance of the security and sustainable development of watershed ecosystems and new perspectives for watershed ESEW research.