Investigating the Teleconnection Between Enso and Basin-scale Flooding: Case Study of Nzoia River Basin, Western Kenya

Abstract

Even though there is progress in enhancing flood forecasting in Kenya, floods still continue to cause multiple casualties and substantial damages to infrastructure and properties leading to massive economic losses in areas like the Nzoia River Basin within Western Kenya. This unfortunately might worsen in the future, considering factors like climate change, urbanization and increased vulnerability of populations. The main objective of the study was to establish the teleconnection between ENSO and basin scale floods within the Nzoia River Basin. The datasets used included daily rainfall observations, the Climate Hazards Group InfraRed Precipitation with Station data (CHIRPS), Potential Evapotranspiration (PET), Oceanic Niño Index (ONI) for ENSO analysis, and daily river discharge. To determine a comprehensive picture of rainfall across the Nzoia Basin, the rainfall data obtained was subjected to spatio-temporal analysis. Analysis entailed an identification of trends or patterns in the rainfall data, such as deviations in the intensity and frequencies of rainfall events over the years. From the results, there is an increasing trend of precipitation received over the basin from the western to the eastern part as evident from the mean annual precipitation ratio. Also, the basin receives consistently higher mean monthly rainfall from the months of March to May which represents the long rainy season in Kenya. The relationship between ENSO and daily rainfall within the Nzoia River Basin was determined through statistical analysis using the box and whisker plots. Box and whisker plots would represent the distributions of historical daily rainfall linked to ENSO over the Basin. To represent the relevant flood simulation processes within the Nzoia River Basin, the GR4J model, a lumped daily model, belonging to Soil Moisture Accounting Model (SMA) group was considered.