Exploring the driving forces of potential marsh wetlands formation and distribution in the Qilian Mountains of Qinghai, China

Abstract

The formation and distribution of marsh wetlands are influenced by both climate changes and human activities. Quantifying the driving factors of marsh wetlands is essential for understanding the sustainable utilization of wetland ecosystems. The current study simulated the potential marsh wetlands’ spatial distribution in the Qilian Mountains of Qinghai (QMQ) by a random forest (RF) model with a series of environmental indicators, including climate, vegetation, soil, topography, and human activity factors in 2020. Simultaneously, a structural equation modeling (SEM) was utilized to explore driving factors of formation and distribution of marsh wetlands. Results showed that: (1) The potential marsh wetlands area was determined to be 2502.01 km2 in the QMQ, mainly distributed in the watershed source and along the river. (2) Normalized difference vegetation index (NDVI), normalized difference water index (NDWI), digital elevation model (DEM), fraction vegetation coverage (FVC), and ground surface temperature (GST) were the most critical factors in marsh wetland simulation, with the relative importance index of 11.14 %, 9.70 %, 9.07 %, 8.39 %, and 6.50 %, respectively. (3) The topography acted on the potential marsh wetlands’ distribution by influencing the nutrient input function of potential marsh wetlands (indirect effect 0.28). The climate was the fundamental factor that directly controls the distribution of potential marsh wetlands (direct effect 0.44), and vegetation had a total effect of 0.24. The current study may provide information for policy decisions on wetland conservation and restoration.

Keywords

• Alpine marsh wetlands
• Random forest
• Driving factors
• The Qilian Mountains