Evaluation of the climate change impacts and Floodplain restoration   in the Kelani River basin, Sri Lanka: a hydraulic modelling approach

Abstract

The Kelani River basin (KRB) is the highest flooding conditions appeared river basin in Sri Lanka. Based on this, the study investigates the impact of climate resilience on rainfall patterns, temperature, wind speed, and streamflow dynamics of the Kelani River basin in Sri Lanka to enhance flood risk prediction and floodplain restoration planning. Rainfall, temperature, and wind speed data for the 2013 to 2024 time period were obtained from five stations: Angoda, Colombo, Hanwella, Maliboda, and Rathnapura. Long-term trends were detected using statistical approaches such as the Mann-Kendall test. Actual versus predicted streamflow data were used to determine the effectiveness of the hydrological model, along with error distribution analysis. There was some variation in the rainfall; Rathnapura peaked at 6000 mm in 2019, and Maliboda was almost constant throughout. Statistical tests indicated an increase of 3.12 mm per year in rainfall. Within the temperature trends, there was no long-term variation at the Colombo and Rathnapura stations, while in wind speed, a moderate decline without statistical significance was recorded. The hydrological model showed shortcomings in the streamflow forecast, mainly due to underestimating the high flows and overestimating the low flows. The residual analysis indicated a deficiency in the model's representation of extreme events, and the Q-Q plot indicated poor performance in predicting runoff extremes. These limitations highlight the need for model refinement using nonlinear-learning-based approaches to improve flood risk assessment and inform adaptive restoration strategies under changing climate conditions.