ML-SWAN-v1: a hybrid machine learning framework for the prediction of daily surface water nutrient concentrations

Nutrient data from catchments discharging to receiving waters are necessary to monitor and manage water quality, however, they are often sparse in time and space and have non-linear responses to environmental factors, making it difficult to systematically analyse long- and short-term trends and undertake nutrient budgets. To address these challenges, we developed a hybrid machine learning (ML) framework that first separated baseflow and quickflow from total flow, and then generated data for missing nutrient species, using relationships with hydrological data, rainfall, and temporal data. The generated nutrient data were then included as additional variables in a final simulation of tributary water quality. Hybrid random forest (RF) and gradient boosting machines (GBM) models were employed and their performance compared with a linear model, a multivariate weighted regression model and stand-alone RF and GBM models that did not pre-generate nutrient data. The six models were used to predict TN, TP, NH3, dissolved organic carbon (DOC), dissolved organic nitrogen (DON), and filterable reactive phosphorus (FRP) discharged from two study sites in Western Australia: Ellen Brook (small and ephemeral) and the Murray River (large and perennial). Our results showed that the hybrid RF and GBM models had significantly higher accuracy and lower prediction uncertainty for almost all nutrient species across the two sites. We demonstrated that the hybrid model provides a flexible method to combine data of varied resolution and quality, and is accurate for the prediction of responses of surface water nutrient concentrations to hydrologic variability.