Exploring nonlinear responses of lake nutrients and algal blooms to restoration measures: A three-dimensional flux network modelling approach

Excessive anthropogenic nitrogen (N) and phosphorus (P) inputs accelerate harmful algal blooms (HABs) and pose a significant global threat to lake ecosystems. Despite the adoption of extensive restoration measures to improve water quality, their effects on nutrient cycles and the underlying pathways remain unclear. In this study, a three-dimensional nutrient flux modelling approach was developed based on a validated Environmental Fluid Dynamics Code (EFDC) model and the Spatiotemporal Integral for Nutrient Networks (SINN) to compute nutrient storages and fluxes in lakes. We then conducted a scenario analysis, composed of 10 schemes, to assess the eight restoration measures in Lake Dianchi, China. The results showed that our model exhibited a high level of agreement with observed data and fluxes reported in previous studies. Of all the measures implemented, pollution interception by lakeshore wastewater treatment plants (WWTPs) accounted for the largest historical improvements in water quality, with reductions of total nitrogen (TN) by 9.3%, total phosphorus (TP) by 36.4%, and chlorophyll a (Chla) by 18.2%. Furthermore, combined implementation of transboundary water transfer and wastewater diversion contributed to a reduction of 13.3% in TN, 11.5% in TP, and 6.2% in Chla. To achieve greater improvements in water quality, efforts must be made to improve the emission standard of WWTPs and ensure compliance with water quality standard of inflow rivers. The cyanobacteria biomass catch can only be used as a supplementary approach to restrict the HAB peaks. The nonlinear responses of lakes were characterized by large loading abatement but slight changes in water quality. Through the SINN approach, we revealed that weakened internal cycling was the primary driver of water quality improvements in all restoration measures. Further analysis demonstrated that measures targeting reducing external loadings significantly weakened internal cycling processes. In contrast, measures targeting the internal loadings, such as sediment dredging, exhibited limited effectiveness in reducing the intensity of internal cycling and Chla in the lake. For eutrophic lakes with substantial legacy nutrients, the abatement of external loadings remains the principal strategy. The approach presented in this study can be used to assess and optimize lake restoration measures.