Determining the drivers and rates of soil erosion on the Loess Plateau since 1901

Vegetation Restoration Increases the Drought Risk on the Loess Plateau

The extensive implementation of the 'Grain for Green' project over the Loess Plateau has improved environmental quality. However, it has resulted in a greater consumption of soil water, and its overall hydrological effects remain highly controversial. Our study utilized a coupled land-atmosphere model to evaluate the effects of vegetation changes resulting from revegetation or reclamation on the hydrology of the Loess Plateau. Revegetation was found to stimulate an increase in precipitation, evapotranspiration, and atmospheric water content. However, the increase in precipitation was insufficient to compensate for soil water loss driven by intensified evapotranspiration, resulting in a decrease in both runoff and soil water content. In contrast to revegetation, reclamation would reduce precipitation, although the reduction was less than the decrease in evapotranspiration. This could lead to an increase in both runoff and soil water content. The results provide an important scientific basis for the hydrological effects of vegetation changes on the Loess Plateau, which is particularly important for guiding current and future revegetation activities toward sustainable ecosystem development and water resources management.

Modeling soil loss under rainfall events using machine learning algorithms

Soil loss is an environmental concern of global importance. Accurate simulation of soil loss in small watersheds is crucial for protecting the environment and implementing soil and water conservation measures. However, predicting soil loss while meeting the criteria of high precision, efficiency, and generalizability remains a challenge. Therefore, this study first used three machine learning (ML) algorithms, namely, random forest (RF), support vector machine (SVM), and artificial neural network (ANN) to develop soil loss models and predict soil loss rates (SLRs). These soil loss models were constructed using field observation data with an average SLR of 1756.48 t/km2 from rainfall events and small watersheds in the hilly-gully region of the Loess Plateau, China. During training, testing and generalizability stages, the average coefficients of determination from the RF, SVM, and ANN models were 0.903, 0.860, and 0.836, respectively. Similarly, the average Nash-Sutcliffe coefficients of efficiency from the RF, SVM and ANN models were 0.893, 0.791 and 0.814, respectively. These results indicated that MLs have superior predictive performance and generalizability, and broad prospects for predicting SLRs. This study also demonstrated that the RF model outperformed better than the SVM and ANN models. Therefore, the RF model was used to simulate the SLR of each small watershed in the Chabagou watershed. Our results showed the four-year (2017-2020) average annual SLR of the small watersheds ranged from 0.73 to 1.63 × 104 t/(km2∙a) in the Chabagou watershed. Additionally, the results also indicated the SLR of small watersheds under the rainstorm event with a 100-year recurrence interval was 4.4-51.3 times that of other rainfall events.Furthermore, this study confirmed that bare land was the predominant source of soil loss in the Chabagou watershed, followed by cropland land and grassland. This study helps to provide the theoretical basis for deploying soil and water conservation measures to realize the sustainable utilization of soil resources in the future.

Ecological risk changes and their relationship with exposed surface fraction in the karst region of southern China from 1990 to 2020

Due to anthropogenic disturbances, the karst region in southern China is vulnerable to ecological problems such as soil erosion and surface exposure. However, limited studies on variations in large-scale ecological risk (ER) and their influencing factors, particularly the coupling/decoupling relationship with an exposed surface fraction (ESF), make ER regulations and ecological restoration challenging. The present study evaluates the ER of eight typical karst provinces in Southern China from 1990 to 2020 using the technique for order preference by similarity to an ideal solution (TOPSIS) model and ecosystem services (habitat quality, water yield, carbon storage, soil conservation, and food production), and extracts the contemporaneous ESF using Landsat satellite data in Google Earth Engine (GEE). The spatiotemporal change of ER and ESF are analyzed, and their coupling/decoupling relationship and driving mechanism are explored using coupling coordination degree (CCD) and multi-scale geographically weighted regression (MGWR) models. The results show that: (1) Over the past 30 years, the ER has increased until 2010 and subsequently declined, with an increasing mean value (0.463-0.503), except in Chongqing municipality. The ESF decreased significantly (the mean value dropped from 44.7% to 38.7%), except that in Sichuan province. (2) The average CCD between ER and ESF decreased with fluctuation of -0.017, with a decoupling relationship (58.18%). The coupling area is larger than the decoupling area in the Sichuan area, while other provinces are opposite. (3) The coupling/decoupling relationship in the study area is mainly driven by terrain (elevation, slope) and socio-economic (population density, per capita GDP) factors. More attention should be paid to the role of these factors in the continuous reduction and control of ESF and ER. This study can serve as a reference for similar studies in karst regions, such as risk assessment and surface monitoring, rocky desertification control, ecological engineering layout, and territorial planning.