| dc.description.abstract | Sand harvesting has potential consequences on the river ecosystem. This study sought if sand harvesting impacts the riverine ecosystem and consequently, riparian conservation on two rivers in Kakamega County. Specifically, the study sought to map the distribution of sand harvesting on two rivers in the County; determine the effects of sand harvesting on river water quality and soil physico-chemical properties of the riparian areas; quantify the impacts of sand harvesting on floral and soil macro-fauna diversity within the riverine ecosystem; and assess the mitigative strategies and challenges of Water Resources Users Associations (WRUAs) and river system conservation. The study combined quasi experimental and qualitative research design. There were five quasi experimental sites and five control sites that were used, in three replicates in the months of June, September and December 2020, to undertake the study. Distribution of sand harvesting activities was assessed by a reconnaissance survey on the rivers. Areas identified were mapped using a GPS. Turbidity was determined in-situ by a turbidimeter. Riverbank erosion was assessed by satellite imageries. 500ml water samples were collected and taken for analysis of total suspended sediments (TSS) in the Water Resources Authority (WRA) Regional laboratory in Kakamega. Soil moisture content, nitrogen (N), phosphorus (P), pH, organic carbon (OC) and textural class were determined by collecting a 500g composite soil sample and taken to Kenya Agricultural and Livestock Research Organization (KALRO) Kakamega Laboratory for analysis. A total of thirty 10mx10m quadrats were established randomly on either side of the riverbank in which trees, saplings, herbs and soil macro-fauna were sampled. 105 questionnaires, interviews and Focused Group Discussions (FGDs) were used to assess mitigative strategies and challenges of conservation by WRUAs. Questionnaires were tested for validity and reliability by pre-testing which gave a Cronbach alpha index of 0.717. Turbidity, TSS, soil moisture, N, P, OC and pH were analysed using IBM SPSS Statistics Version 22. Shannon-Wiener diversity index was used to determine floral and soil macro-fauna diversity. Data from the questionnaire was entered into SPSS Statistics and analysed. A spatial map of 18 sand harvesting hotspots identified was developed using ArcGIS. Sand harvesting significantly affected changes in river width (p = 0.008), TSS in rivers (p = 0.001), turbidity in rivers (p = 0.006), total nitrogen content in the riparian area (p = 0.036), soil organic carbon content (p = 0.003) and soil pH (p = 0.001), as there was significant difference between the control and the sand harvesting sites for these parameters. However, it did not significantly affect riparian soil phosphorus content (p = 0.810), soil percent moisture content (p = 0.309) and soil textural classes (p = 1.000). Also, sand harvesting significantly affected tree abundance in the riparian area (p = 0.048), but there was no correlation between the tree species diversity index and sand harvesting activities (r=.125); significantly affected sapling abundance (p = 0.036) leading to invasive species like Psidium being most abundant, but had no correlation with the sapling species diversity index (r=-.284); significantly affected herbaceous species abundance (p = 0.042) as well as had negative correlation with herbaceous species diversity index (r= -.608); and significantly affected soil macrofauna species abundance (p = 0.049) and had no correlation with soil macrofauna species diversity index (r= -.317). Most applied river system conservation strategy by WRUAs was planting of trees (83%, n=87). About 54% (n=57) WRUAs understood their role and training was proposed as one of the measures to improve their confidence in riverine ecosystem conservation concepts, as there lacked a recognized institution for consulting on sand harvesting issues. Hence, it is evident river Shiatsala and Lusumu are facing significant pressure from sand harvesting. | en_US |
