Development and evaluation of a process-based model to assess nutrient removal in floating treatment wetlands.

Modelling is a useful tool for comprehensively describing the processes occurring in floating treatment wetlands (FTWs). However, temperature effects and phosphorus dynamics are not considered in the current FTW models. Therefore, a process-based model comprised of a plant growth submodel, a nitrogen dynamic submodel and a phosphorus dynamic submodel was developed to understand the complicated processes occurring in FTWs. The model was fully calibrated using a mesocosm FTW system operated for 168 days. Global sensitivity analysis revealed that nitrogen removal performance was predominantly sensitive to parameters representing plant characteristics and microbial activity. Because of the high concentration of organic matter, mineralization and sedimentation played important roles in nitrogen and phosphorus removal. In addition, the coprecipitation rate of phosphate also had a significant influence on phosphorus removal performance. When further investigation was applied to understand the behavior of the model, the ratio of nitrogen to phosphorus in plant tissue was found to be an indicator of the nutrient limitation in the water column. Furthermore, the model illustrated that both FTW operating conditions and plant characteristic parameters exerted an important influence on nitrogen removal and plant uptake contribution. Therefore, the selection of appropriate operating conditions and plant species can achieve high nutrients removal and make effective use of plants in FTWs. The model provides a useful tool for assessing the nutrients removal performance of FTWs and for evaluating strategies for them in design and operation.