IMPACTS OF CLIMATE CHANGE ON THE HYDROLOGY OF OSUN RIVER BASIN,

Abstract

The likelihood changes in the hydrological processes under the impact of Climate change in Osun River basin in Nigeria were examined by comparing simulated future data with the current climate records. To achieve this, Mann Kendal trend test, Linear Regression coefficient, Standardized precipitation index, climate and hydrological model were employed based on 35 years historical and future climate data sourced from ERA 5 and Coordinated Regional Climate Downscaling Experiments (CORDEX) website respectively. According to the results, Mann Kendal indicates more decreasing trend in Historical annual rainfall with decreasing trend of -9.8mm/year but in contrast to Future years that indicates a lesser decreasing trend of -6.86mm/year but statistically significant at 5% significant level. Standardized precipitation index indicates 7 cycles of wet year in SPI 3, 5 cycles of wet in SPI 6, 8 cycles of wet for SPI 12, 6 cycles of wet in SPI 24 for the period of 1985-2019 while 6 cycle of wet experienced in SPI 3, 5 cycles of wet in SPI 6, 3 cycle of wet in SPI 12, 3 cycle of wet in SPI 24 for the observed future years (2021-2055). Based on the coefficient of variation results, both historical period of 19.32% and future period of 11.57% variability are classified as being low variability since their variability are less than 30%. Variability is associated with unstable weather and it may impact negatively on Agriculture and Hydropower generation. Furthermore, downscaled climate model data under representative concentration pathways (RCP) 8.5 scenario showed an increase trend in minimum, maximum and mean temperature for future years than the past. Hydrological model was used in the calibration and validation of the observed streamflow of the area. The results obtained during calibration and validation indicated that (r factor is 1.15 , p factor is 0.56), and (r factor is 0.91 , p factor is 0.45),which indicates that the model has captured the uncertainty in arcswat model and this further signify that the model is suitable for simulating streamflow in the watershed. According to the results, historical years represented the most varied year, while the future years represents the least varied years. It has been found that rainfall trend will increase in the future with low variability, also there is going to be increase in the stream flow which is likely to results to flood in the area of study. This result will be useful in the mitigation of natural hazard (flood).