EXTENT OF USING NON-STRUCTURAL FLOOD CONTROL MEASURES AND THEIR IMPACT ON HOUSEHOLD WELFARE IN BUNYALA SUB-COUNTY, BUSIA COUNTY, KENYA

Abstract

Floods are the most destructive water-related disasters considered to have dire consequences to the livelihoods of the affected population. Kenya experiences both riverine and flash floods in various parts of the country and various measures have been devised to deal with floods. Structural measures; dykes, canals, reservoirs, and non-structural measures: early warning systems, insurance schemes, awareness campaigns, and land-use planning have been implemented as mitigation measures to help in coping with this disaster. Structural measures often require large investments hence these measures are mainly implemented by the government. Non-structural measures can however be implemented by various parties such as individuals and non-governmental organizations in their various capacities. This study focused on four non-structural measures namely; Flood forecasting and Early Warning Systems (FEWS), land-use planning with zoning, rainwater harvesting, and self-insurance, implemented in flood-prone Bunyala, Busia County in Kenya. Although these measures are in use, there is no clear evidence of the impact it has on the livelihoods of vulnerable households. To fill this gap, the main objective of this study was to contribute to the welfare of households in Bunyala Sub-County. The specific objectives were to determine the household perception, factors influencing the extent of use, and welfare impact of non-structural flood control measures. A multistage sampling procedure was used to obtain the desired sample of 384 with 256 participants and 128 non-participants. Descriptive statistics were used to analyze the household perception. Multivariate Multiple Regression model was used to determine factors influencing extent of use and Propensity Score Matching (PSM) was used to determine the welfare impact of non-structural measures. The descriptive results indicated that land-use planning with zoning was perceived more useful compared to the other measures due to its ease of application. Multivariate regression showed that Early Warning duration, flood frequency, flood effect, land regulation awareness, credit access, climate information access, receiving early warning, saving for floods, and water harvesting initiatives were factors found to be jointly significant in influencing the extent of using the non-structural flood control measures at p<0.05. PSM results revealed that participants spent Ksh. 3860.7 less consumption expenditure compared to participants. It was therefore concluded that the flood control measures had a positive impact on welfare. Policy implications included improving early warning technology and building awareness. Households need to come up with initiatives aimed at creating awareness and sharing the right information on floods and non-structural measures.