Can the watershed non-point phosphorus pollution be interpreted by critical soil properties? A new insight of different soil P states.

The physicochemical properties of surface soil play a key role in the fate of watershed non-point source pollution. Special emphasis is needed to identify soil properties that are sensitive to both particulate P (PP) pollution and dissolved P (DP) pollution, which is essential for watershed environmental management. The Chaohu Lake basin, a typical eutrophic lake in China, was selected as the study site. The spatial features of the Non-point Source (NPS) PP loads and DP loads were calculated simultaneously based on the integration of sediment delivery distributed model (SEDD) and pollution loads (PLOAD) model. Then several critical physicochemical soil properties, especially various soil P compositions, were innovatively introduced to determine the response of the critical soil properties to NPS P pollution. The findings can be summarized: i) the mean PP load value of the different sub-basins was 5.87 kg, and PP pollution is regarded to be the primary NPS P pollution state, while the DP loads increased rapidly under the rapid urbanization process. ii) iron-bound phosphorus (Fe-P) and aluminum-bound phosphorus (Al-P) are the main components of available P and showed the most sensitive responses to NPS PP pollution, and the correlation coefficients were approximately 0.9. Otherwise, the residual phosphorus (Res-P) was selected as a sensitive soil P state that was significantly negatively correlated with the DP loads. iii) The DP and PP concentrations were represented differently when they were correlated with various soil properties, and the clay proportion was strongly negatively related to the PP loads. Meanwhile, there is a non-linear relationship between the DP loads and the critical soil properties, such as Fe and Total Nitrogen (TN) concentrations. Specifically, a strong inhibitory effect of TN concentration on the DP load was apparent in the Nanfei river (NF) and Paihe (PH) river basins where the R2 reached 0.67, which contrasts with the relatively poor relationship within the other five basins. In addition, the degree of correlation between the Fe and DP loads severely degraded in the basins that were mostly covered by construction land or those that underwent a rapid urbanization process. The findings indicate that land use/cover change (LUCC), especially the distribution of agricultural land and construction land, as well as the soil background information (TN, Fe and Soil organic matters, etc.) can be considered as factors that influence NPS P pollution.