Water conservation assessment and its influencing factors identification using the InVEST and random forest model in the northern piedmont of the Qinling Mountains

Abstract

The north piedmont of the Qinling Mountains (NPQM), Northwest China. By combining the Integrated Valuation of Ecosystem Services and Tradeoffs (InVEST) and random forest model, the current study assessed the water conservation in the NPQM, one of the most significant Earth's critical zone in China, and identified the key factors influencing it. The average annual water conservation in the NPQM was calculated to be 90.8 mm in 1980, 85.0 mm in 1990, 61.1 mm in 2000, 97.4 mm in 2010, and 121.3 mm in 2020, respectively. Water conservation was significantly higher in the southern mountainous region of the Qinling Mountains compared to the northern alluvial plain region of the Wei River. Key determinants affecting water conservation included annual precipitation, potential evapotranspiration, the proportion of forest and agricultural land, and soil texture Furthermore, precipitation, forest coverage rate, and sand content in topsoil were demonstrated to positively influence water conservation. Conversely, evapotranspiration, agricultural land, and clay content in topsoil were found to negatively affect water conservation. This study reveals the spatiotemporal characteristics of water conservation and identifies its key influencing factors in the NPQM. Additionally, the effects of these factors on water conservation are further discussed. The results and conclusions of this study could provide scientific insights for water security and ecosystem management in the NPQM region. • The spatiotemporal characteristics of water conservation in the NPQM have been revealed. • The random forest model identified the key factors influencing water conservation. • Precipitation, forest and sand content in topsoil positively affect water conservation. • Evapotranspiration, agricultural land and topsoil clay content negatively affect water conservation.