Quantifying spatiotemporal variations of cropland phosphorus runoff loss in China with machine learning algorithms

    Excessive applications of phosphate fertilizers have led to significant phosphorus (P) accumulation in agricultural soils. This surplus P is prone to being lost through surface runoff, thereby threatening downstream water bodies such as aquifers, streams, lakes, and oceans. Indeed, the runoff loss of P has led severe global environmental concerns, including proliferation of harmful algal blooms, onset of eutrophication, and expansion of anoxic dead zones in coastal marine ecosystems. Assessing the spatial distribution of total P (TP) runoff loss from croplands is essential for developing targeted mitigation strategies against the persistent issue of nonpoint source pollution. In this study, we compiled 812 datasets from 114 peer-reviewed papers for cropland P loss across China. We then developed machine learning (ML) approaches to estimate the temporal and spatial variations in P runoff loss across China from 1990 to 2020. Four prevalent ML models were considered, namely, multiple linear regression (MLR), random forest (RF), classification and regression trees (CART), and boosted regression trees (BRT). Among these four models, RF exhibited the highest predictive accuracy for both uplands (calibration: R² = 0.86, n = 293; validation: R² = 0.61, n = 96) and paddy fields (calibration: R² = 0.88, n = 137; validation: R² = 0.60, n = 44). According to RF, China’s croplands are estimated to have lost an average of 148 ± 27 Gg P yr⁻¹ from 1990 to 2020, with uplands and paddy fields contributing 114 ± 26 Gg P yr⁻¹ and 34 ± 4 Gg P yr⁻¹, respectively. The data showed a significant increase in upland TP runoff loss over the study period (p<0.001), whereas paddy field TP loss remained relatively constant. Regions in southern, eastern, and southwestern China, notably in Hainan, Guangxi, and Fujian provinces, were identified as hotspots of TP runoff loss. Scenario predictions suggest a 1.4-11.8% reduction in TP runoff loss under various conditions, most effectively when minimizing runoff depth. To effectively mitigate TP runoff loss in China, an integrated management approach involving water, soil, and fertilizer is recommended. Overall, this study enhances our quantitative understanding of cropland TP runoff loss in China, providing crucial insights for efficient cropland P management, which is key to managing nonpoint source pollution on a national level.