Carbon,nitrogen and phosphorus coupling relationships and their influencing factors in the critical zone of Dongting Lake wetlands,China

Wetland is a transition zone between terrestrial and aquatic ecosystems, and is the source and sink of various biogenic elements in the earth’s epipelagic zone. In order to investigate the driving force and coupling mechanism of carbon (C), nitrogen (N) and phosphorus (P) migration in the critical zone of lake wetland, this paper studies the natural wetland of Dongting Lake area, through measuring and analysing the C, N and P contents in the wetland soil and groundwater. Methods of Pearson correlation, non-linear regression and machine learning were employed to analyse the influencing factors, and to explore the coupling patterns of the C, N and P in both soils and groundwater, with data derived from soil and water samples collected from the wetland critical zone. The results show that the mean values of organic carbon (TOC), total nitrogen (TN) and total phosphorus (TP) in groundwater are 1.59 mg/L, 4.19 mg/L and 0.5 mg/L, respectively, while the mean values of C, N and P in the soils are 18.05 g/kg, 0.86 g/kg and 0.52 g/kg. The results also show that the TOC, TN and TP in the groundwater are driven by a variety of environmental factors. However, the concentrations of C, N and P in the soils are mainly related to vegetation abundance and species which influence each other. In addition, the fitted curves of wetland soil C-N and C-P appear to follow the power function and S-shaped curve, respectively. In order to establish a multivariate regression model, the soil N and P contents were used as the input parameters and the soil C content used as the output one. By comparing the prediction effects of machine learning and nonlinear regression modelling, the results show that coupled relationship equation for the C, N and P contents is highly reliable. Future modelling of the coupled soil and groundwater elemental cycles needs to consider the complexity of hydrogeological conditions and to explore the quantitative relationships among the influencing factors and chemical constituents.