Determination of the drivers and impacts of water diversion and abstraction in selected rivers in the upper Tana basin, Kenya

Abstract

The South West Upper Tana Basin, Kiama, Chania, Kimakia and Thika sub basins, located in the central region of Kenya has been experiencing increased water abstraction and river diversions. However, the causes of this increased water abstraction have not been fully investigated. Several factors that could be responsible to these abstractions; climate change, land cover/land use changes, agricultural practices and changes in streamflow. Previous studies in the region have not determined the contribution of each of these factors in causing water abstraction in the basin. Determination of the drivers of water abstraction is important for sustainable water resource development in the basin. Agricultural intensification in the basin is heavily reliant on water abstraction. This study therefore investigated the drivers and impacts of water abstraction and river diversion in the Basin. The main objective of the study was to determine the causes and impacts of water abstractions/diversions on streamflow in the Basin. Socio-economic data was collected through 120 questionnaires that were equally administered to households in Kiama, Kimakia, Thika and Chania sub basins. Hydrological data, mainly river discharges and rainfall data were obtained from Water Resources Authority (WRA) and Kenya Meteorological Department (KMD). Simple linear regression and logit regression analysis was used to test the relationships between variables. The dependent variable for the study was water abstraction while independent variables were rainfall variability, streamflow, household structure, education, occupation, water abstraction technologies and agricultural practices. The study established that the main drivers of water abstraction are climate change, government policies and strategies and agricultural development. The pressures underlying these drivers are the need for water for irrigation agriculture, domestic use and inter-basin water transfers. There is evidence of increased variability in rainfall due to climate change. The variability of rainfall and streamflow has a strong positive correlation with r value of 0.70 and R2 0.61. Influence of crop sales (coefficient = 0.78; odds ratio = 22.35; P = 0.01), farm size (coefficient = 0.85; odds ratio = 11.19; P = 0.04) and fertilisers used (coefficient = 0.65; odds ratio = 3.08; P = 0.02) on water abstractions was also significant. The household structure was found not to have a significant influence on water abstraction and streamflow decline. The increased water abstraction has also increased upstream and downstream impacts in the basin. The impacts of water abstraction and streamflow decline include the decline in water levels, changes in turbidity, improved crop yields, siltation and changes in channel morphology. The main responses to these impacts are water use restrictions, policies and legislations and societal guidelines. This study recommends intensified reforestation programs to improve forest cover in the basin since this can potentially improve water yield in the rivers. There is also need for improved monitoring of water abstractions, public awareness creation and the involvement of stakeholders especially water users in the basins. The later is important for sensitizing people on the impacts of over-abstractions. This study will inform the development of policies that will ensure sustainable water abstraction in the SWUT basin.