https://repository.seku.ac.ke/handle/123456789/86 2026-03-19T08:23:47Z https://repository.seku.ac.ke/handle/123456789/8284 Title: Spatio-temporal variations in physicochemical water quality and the impact of land use/land cover change In River Athi Basin, Kenya. Authors: Ogbonna, Vincent A. Abstract: The growing impacts of natural processes and human activities on water quality at global, regional, and local scales are raising concerns. The River Athi Basin natural gravitational flow toward lower elevations, ultimately reaching the Indian Ocean, facilitates waste disposal practices among the residence located along the river bank. This study sought to determine the Spatiotemporal Variations in Physicochemical Water Quality and the Impact of Land Use/Land Cover Change in the Mid Reaches of River Athi Basin, Kenya by; 1) examining the influence of land use/land cover change in river Athi Basin from 2015 to 2023. (2) Determining seasonal variation in the physicochemical water quality of the river Athi Basin. (3) Assessing the spatial variation in physicochemical water quality of the river Athi Basin. The study examined the influence of LULC changes from 2015 to 2023 using Landsat 8 imagery, GIS, remote sensing, and GPS technologies for data extraction, image processing, and LULC analysis. Pearson correlation analysis assessed spatial differences of land use land cover impacts on water quality across six sampling stations in the basin. Interview survey was used to supplement the water quality dataset. The studies applied multivariate analysis for spatial and temporal reduction of the multidimensional dataset and identification of pollution sources. Seasonal variations in the physico-chemical water quality of River Athi Basin was determined using eight physicochemical parameters (pH, EC, TDS, NO₃, K, PO₄, BOD, and COD) and two heavy metals (Cd and Cr). Data collection was carried out during two distinct seasonal periods: the short dry and rainy season (August-September for dry, November-December for rainy) in 2023, and the long dry and rainy season (January-February for dry, April-May for rainy) in 2024. An independent T-test was used to compare the mean levels of water quality parameters between dry and rainy seasons. The study assessed spatial variation in the physicochemical water quality of the river basin, covering six sampling stations. One way analysis of variance (ANOVA) compared the mean values of variables of the sampling stations. Multiple linear regression tested the influence of pH, EC, TDS, NO3, K, and PO4 on BOD and COD (oxidation parameters) and cadmium and chromium (heavy metals). The Pearson Product-Moment Correlation Coefficient (PPMCC) assessed the relationships between water physicochemical parameters and heavy metals in seasonal and spatial variations in water quality of the river basin. The findings on LULC analysis show notable shifts in land use from 2015 to 2023. Between these periods, the overall built-up increased to 0.29%, bare-lands declined by 7.06%. Farmlands, forests, and grasslands were elevated by 0.52%, 4.54%, and 2.77%, with decline in open waters by 1.24%. Spatial LULC difference with correlation analysis reveal higher amounts of EC, TDS, Cd, Cr, NO3, and PO4 influencing water quality. Interview survey revealed settlements, agriculture, and climate conditions as the main causes of degradation of water quality. Seasonal finding reveal significant fluctuations in pollution, with the dry season exhibiting higher pollution levels. February demonstrate proliferated temporal pollution, characterized by high concentrations of EC, TDS, BOD, and COD. Spatial finding demonstrated significantly higher pollution signatures in Athi River Town, Stony Athi, and NYS stations, while NYS contribute to higher levels of nutrients, organic pollutants, and heavy metals. In contrast, the control station and Kibwezi Bridge station demonstrates effective self-purification processes. Multivariate analysis revealed pollution sources over time and space in the River Basin. The stable pH levels over time and space was influenced by the buffering capacity. Multiple regression analysis indicates that physicochemical parameters, such as pH, EC, TDS, NO₃, K, and PO₄, explain 62% of BOD variation and 70% of COD variation, as well as 36% of both Cd and Cr variations. Pearson correlation analysis shows strong links between EC, TDS, BOD, and heavy metals (Cd, Cr), with significant associations among nutrients and other water quality indicators. Natural and anthropogenic activities are pivotal drivers of the water quality degradation of River Athi Basin over time and space. This study recommends that the National Environmental Management Authority (NEMA) and the National Environmental Policy (NEP) strengthen regulations on environmental management, water resource conservation, sustainable land use, public health protection, irrigation control, forest preservation, and aquatic ecosystem conservation in order to support global efforts toward achieving Sustainable Development Goals (SDGs). Description: Doctor of Philosophy (Ph.D.) In Environmental Management 2025-01-01T00:00:00Z https://repository.seku.ac.ke/handle/123456789/7727 Title: Climate change perception, vulnerability and adaptation among smallholder farmers in Machakos County, Kenya Authors: Kalia, David Makau Abstract: The unprecedented changes resulting from rapid and intensifying climate change have created extremely uncertain conditions for agricultural production, especially in ASALs. Farmers should, therefore, build resilience through appropriate adaptation strategies to cope with the new and emerging impacts of climate change and variability. This study was conducted to understand the awareness and perception of climate change and variability, its impacts on local agriculture, and to identify the most appropriate adaptive strategies for smallholder farmers to enhance resilience building in Machakos County. The study employed a mixed methods approach in which both qualitative and quantitative techniques were used. Field observations and 400 household surveys were conducted with smallholder farmers in the County. Questionnaires, interviews, field observations, and desk research techniques and tools were used to generate the relevant data. In addition, focus group discussions (FGD) and key informant interviews (KII) were conducted in the area. Data from FGD and KII complemented the survey results. The data analysis was done by descriptive statistics and econometric model (Heckman’s sample selection model). The statistical package for social scientists (SPSS) program for windows (version 20) and STATA software (version 12) were used for qualitative and quantitative analysis. Descriptive statistics were used to investigate the impacts of climate change, analyse farmers' perceptions of climate change, and the potential response and adaptation strategies to climate impacts. The study adopted the indicator and systematic review approach to document the agricultural sector's vulnerability to climate change. Heckman’s probit regression model was used to analyse factors influencing the smallholder farmers’ perceptions of climate change and variability and the choice of response options (adaptations) to climate impact applied by the households. Descriptive statistics of key variables were computed, analysed, and presented in frequency distributions, percentages, tables and charts. The study results show significant impacts from climate change and variability in Machakos County, 88.8% of smallholder farmers reported low yields, 82.3%, loss of income, 81.1%, crop failure, 69.0%, livestock deaths, 65.2%, forage scarcity, 64.0%, water shortage, and 53.3% infrastructure damage. In addition, the results revealed high exposure to climate change and variability, high sensitivity, and low adaptive capacity. Concerning exposure, a warming trend was identified in the County, while climate model outputs projected enhanced warming and drying towards the end of the 21st century. The farming systems were highly sensitive to climate change and variability, as indicated by high incidences of droughts, high rural population density, a high percentage of smallholder farmers, and severe high susceptibility to land degradation. Adaptive capacity was generally low but exhibited by substantial social capital and highly diversified agricultural production. The study's results also revealed that 96% of all the smallholder farmers surveyed reported they were aware of climate change. Of those who perceived the climate to be changing, 87.3% reported changes in temperature, 96.8% in rainfall while 98% of farmers indicated that they had observed an increase in drought incidents, Age of the household head, gender, education, access to extension services, access to climate information, off-farm income, household size, distance to market, access to credit, access to insurance, distance to input seller, land size and group membership, influenced the farmers' perception of climate change and variability. The study showed that farmers most farmers (92.2%) adopted several practices to cope with the current climate risks with only 7.8% of the respondents using none. The study findings suggest that some climate change- related responses to agricultural distress are inadequate to cope with the current climate risks. This indicates that farmers may be unable to cope with increasing climate change and variability. Additionally, most farmers (92.5%) foresee a likelihood of impacts of climate change and variability and change worsening in the future. The study recommends an enhancement and/or a shift to more resilient and environmentally sustainable strategies, which include: diversification of livelihoods, water harvesting/Irrigation, mixed farming, crop diversification, tree planting/agroforestry, improved pasture and fodder management, construction of sheds for livestock, improved crop varieties and livestock breeds, drought- tolerant crops and livestock breeds, conservation agriculture, and more use of weather and climate information. Description: Doctor of philosophy in agricultural economics, 2024 2024-01-01T00:00:00Z https://repository.seku.ac.ke/handle/123456789/7357 Title: Environmental and socio-economic predictors of anthrax spatial distribution in Kenya Authors: Otieno, Fredrick T. Abstract: Anthrax spatial distributions and the potential driving factors remain poorly understood worldwide and in Kenya. This study aimed at establishing environmental and social economic predictors of the spatial distribution of anthrax in Kenya through (1) determining the relationship between selected environmental and socio-economic factors on spatial distribution of anthrax through use of an ecological niche modelling framework; (2) predicting the effect of climate change on the future spatial distribution of anthrax; and (3) establishing the influence of socio-economic factors in vulnerability to anthrax. Ecological Niche Model (ENM) of boosted regression trees (BRT) algorithm was applied to predict the suitable spatial environments for anthrax under current and future climate scenarios in Kenya. The model fitted confirmed anthrax occurrences from three distinct sources of retrospective records (2011 to 2017), sporadic anthrax outbreaks (2017 to 2018) and active surveillance (2019 to 2020) against selected predictor variables to yield current and future anthrax risk maps. Finally, the underlying socio-economic vulnerability due to the risks of anthrax distribution was assessed by laying over socio-economic indicators in spatial multicriteria decision analysis to produce socio-economic vulnerability maps. The high-risk areas for anthrax outbreaks were identified predominantly in: regions around western Kenya bordering Uganda; southwestern regions bordering Tanzania and regions around central highlands of Kenya. Based on the current scenario, the number of humans affected was estimated at ~ 193,00,840 people/sq.km while that of livestock was at ~7,750,675 animals / sq.km. The important contributing predictor variables were predominantly cattle density, rain of the wettest month, monthly precipitations, soil clay, soil pH, soil carbon, longest dry season and temperature range. The anthrax highly suitable areas expanded from current to future climatic scenarios with current at 36131 km2, RCP 4.5, 40012 km2, and RCP 8.5, 39835 km2. Highly socio-economic vulnerable areas closely correlated with areas of high anthrax risk currently and into the future. At current vulnerability index > 75%, approximately 40,369,455 people were estimated to be at risk. This study results will guide risk-based surveillance and strategies for managing anthrax under One Health approach and also contributes to future research studies within Kenya and beyond. Description: Doctor of Philosophy in Environmental Management, 2022 2023-10-19T00:00:00Z https://repository.seku.ac.ke/handle/123456789/7354 Title: The impacts of agricultural sector devolution on delivery of agricultural extension services and agricultural productivity in Kitui County, Kenya Authors: Kyambo, Onesmus M. Abstract: Agriculture supports the livelihoods of rural people in developing countries, including Kenya. Agriculture is the mainstay and driver of the Kenyan rural economy. Despite the critical role of agriculture in Kenya, poor access to extension support services persists. The study was carried out to evaluate the impact of devolution of the agricultural sector on the provision of agricultural extension services and agricultural productivity in Kitui County. To achieve this objective, the following specific objectives were addressed, namely to: assess the influence of selected socio-economic factors on farmers’ awareness of the devolution of agricultural extension services; determine the factors influencing the delivery of extension services by the county governments; establish the interactions between agricultural extension functions run by county and national governments; and assess the impact of agricultural extension services on the farmers’ household income with respect to agricultural productivity and income as proxies for the years 2012 (before devolution) and 2016/2017 (after devolution). A total of 70 extension officers and 99 farmers were sampled from Kitui County using a stratified random sampling approach. Secondary information sources such as national and county ministries’ reports and existing literature were reviewed to supplement the primary data. A questionnaire was the main tool used for data collection in this study. Data obtained were analyzed through: descriptive and inferential statistics; binary logistic regression; linear regression; and stochastic frontier analysis. The logit binary model showed that age of household, gender, education, income, and size of the land were important factors that influenced farmers’ awareness of the devolution of agricultural extension services. Further, this study established that most of the sampled respondents reported insufficient performance in extension service provision by the county government due to challenges such as inadequate transport, salaries not paid on time, lack of proper staff promotion, lack of clear terms of service without duplication, unconducive work environment, and low facilitation for extension activities. There is minimal interaction between agricultural extension functions run by county and national governments due to the minimal involvement of county extension staff in the development and implementation of the work plans as well as monitoring and supervision at the national level. For example, the sampled smallholder maize farmers who had access to agricultural extension services had their yield productivity increase by 16.4%. The devolution of agricultural extension services resulted in a significant improvement in agricultural productivity and farmer’s income by 27.2% and 13.8%, respectively. This study recommends that more campaigns with focus on women's groups and elderly farmers should be held in the vast Kitui County to create awareness about the devolution of agricultural extension services. Greater involvement of extension staff in development and implementation of work plan at the national level as well as monitoring/supervision should be enhanced in order to contribute to better interactions between national government and county governments. Also, there is a need to provide incentives to extension officers through adequate facilitation, remuneration, and promotion. Therefore, adequate funds should be allocated to the devolved agricultural extension services, for example, a specified percentage of the agriculture sector budget as a way of enhancing overall agricultural productivity and households’ incomes. Description: Doctor of Philosophy in Agricultural Economics, 2022 2023-10-19T00:00:00Z