Objective This study aims to construct a susceptibility risk prediction model for soil salinization in the middle reaches of the Heihe River and reveal the formation mechanisms of salinization under multi-level driving forces, providing quantitative support for precise regional salinization prevention and control. Methods A prediction model was developed by coupling the pressure-state-response (PSR) model with the random forest algorithm, incorporating regional mechanism corrections. A total of 14 factors related to meteorology, hydrology, and soil were collected to establish a three-level evaluation system. After correcting indicator weights using regional background values, risk levels were classified, and model accuracy was validated through field sampling. Results In the salinization risk evaluation system of the study area, the weights of pressure, state, and response layers were 51.3%, 34.9%, and 12.9%, respectively. The core driving factors were evapotranspiration, farmland irrigation water consumption, and groundwater depth. Engineering measures were more effective than agricultural measures in suppressing salinization. Spatially, salinization risk showed a spatial pattern of lower in the south and higher in the north, converging along water systems and low-lying areas, with extremely high-risk areas accounting for 11.4% of the total study area. The model achieved an overall accuracy of 87%, with a Kappa coefficient of 0.8, indicating reliable predictions. Conclusion This model achieves a systematic dynamic linkage among natural driving forces, anthropogenic pressures, and governance response, overcoming the limitations of traditional mechanistic models. It provides a quantitative framework for soil salinization prevention and control in arid and semi-arid regions of Northwest China and a scientific reference for land degradation prevention in the middle reaches of the Heihe River.
YangJ S, YaoR J, WangX P, et al. Research on salt-affected soils in China: history, status quo and prospect[J]. Acta Pedologica Sinica, 2022,59(1):10-27.
ChenY N, LiZ Q, XuJ H, et al. Changes and protection suggestions in water resources and ecological environment in arid region of northwest China[J]. Bulletin of Chinese Academy of Sciences, 2023,38(3):385-393.
ChengG D, XiaoH L, FUB J, et al. Advances in synthetic research on the Eco-hydrological process of the Heihe River Basin[J]. Advances in Earth Science, 2014,29(4):431-437.
WuJ, DingL. Evaluation and prediction of Water-economy-ecology harmonious development structure in Heihe River Basin based on harmony theory[J]. China Rural Water and Hydropower, 2023(1):38-44.
ZhaoX Y, XiH Y, ZhaoJ, et al. Inversion and spatial distribution characteristics of soil salinity in Alxa area, China[J]. Journal of Desert Research, 2023,43(1):27-36.
[15]
马登科.黑河中游绿洲土壤盐分空间格局及运移机制[D].北京:中国科学院大学,2023.
[16]
MaD K. Spatial pattern and transport mechanism of soil salt in oasis in the middle reaches of the Heihe River[D]. Beijing: University of Chinese Academy of Sciences, 2023.
MengX, ZhangJ, ZhengY, et al. Impacts of hydrological processes on soil salinity in the middle Heihe River basin[J]. Ecology and Environmental Sciences, 2015,24(7):1108-1112.
ZhaoY, FengQ, LiH Y. Distribution characteristics of soil salts in the lower reaches of Heihe River, northwest China[J]. Journal of Desert Research, 2017,37(6):1196-1203.
GaoZ J, RenX H, AnY H, et al. Analysis of soil salinity characteristics in salinized areas of northern minghua township in the middle reaches of Heihe River[J]. Northern Horticulture, 2020(21):71-79.
WangQ L, MaX X, LiuX X, et al. Mountain torrent disaster risk assessment method and application based on random forest[J]. Yellow River, 2022,44(4):63-66,73.
ZhiZ M, ChenQ, ZhouQ, et al. Evaluation of Geo-environmental carrying capacity of counties along the Sichuan-Tibet railway based on Random Forest[J]. Science Technology and Engineering, 2022,22(24):10451-10460.
[27]
KimY, EvansJ P, SharmaA. Multivariate bias correction of regional climate model boundary conditions[J]. Climate Dynamics, 2023,61(7):3253-3269.
[28]
WangN, XueJ, PengJ, et al. Integrating remote sensing and landscape characteristics to estimate soil salinity using machine learning methods: a case study from southern Xinjiang, China[J]. Remote Sensing, 2020,12(24):4118.
ZhangY, LiM R, KouL Q. Evaluation of water resources Security in Baoding central city based on PSR-PCA-ANFIS model[J]. China Rural Water and Hydropower, 2022(2):68-75.
SunY, LiuW Z, ShengY. Spatiotemporal differences and influencing factors of economic and ecological resilience of water resources in Xinjiang based on the PSR model[J]. Arid Land Geography, 2023,46(12):2017-2028.
[35]
BiauG. Analysis of a Random Forests Model[J]. Journal of Machine Learning Research, 2010,13(2):1063-1095.
[36]
MaxwellA E, WarnerT A, FangF. Implementation of machine-learning classification in remote sensing: an applied review[J]. International Journal of Remote Sensing, 2018,39(9):2784-2817.
[37]
BelgiuM, DrăguţL. Random forest in remote sensing: a review of applications and future directions[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2016,114:24-31.
JiangL, LiuX L, GuoS, et al. Evaluation of soil salinization susceptibility based on Logistic regression analysis: a case of Tarim irrigation area in southern Xinjiang[J]. Arid Land Geography, 2023,46(11):1858-1867.
[40]
RenD Y, WeiB Y, XuX, et al. Analyzing spatiotemporal characteristics of soil salinity in arid irrigated agro-ecosystems using integrated approaches[J]. Geoderma, 2019,356:113935.
LiuP F, ZhangG H, CuiS J, et al. Threshold value of ecological water table and dual control technology of the water table and its quantity in the salinized farmland around wetland in arid areas[J]. Hydrogeology&Engineering Geology, 2022,49(5):42-51.
XinX J, XiangG S. Study on reasonable groundwater exploitation based on the ecological water level in Zhangye Basin[J]. Yellow River, 2013,35(8):43-45,48.
MiaoJ X. Protection and treatment of groundwater resources in Linze county, Gansu province[J]. Water Resources Development and Management, 2023,9(2):10-13,18.
ZhouJian, LiX, WangG X, et al. The Spatio-temporal variation analysis of groundwater and response to land-use change in the middle reaches of the Heihe river basin[J]. Journal of Natural Resources, 2009,24(3):498-506.
XuY, GeZ, WangJ, et al. Study on relationship between soil salinization and groundwater table depth based on indicator Kriging[J]. Transactions of the Chinese Society of Agricultural Engineering, 2019,35(1):123-130.
ZhaoC Y, LiS B, JiaY H, et al. Dynamic changes of groundwater level and vegetation in water table fluctuant belt in lower reaches of Heihe River: coupling simulation[J]. Chinese Journal of Applied Ecology, 2008,19(12):2687-2692.
QiangX H, GaoW W, WangB, et al. Remote sensing-based risk assessment of soil salinization and its change over time[J]. Arid Zone Research, 2025,42(3):431-444.