Effect of agricultural practices on water quality in Ndakaini Dam, gatanga, Muranga, County, Kenya

Abstract

Change in Land Use and Land Cover (LULC) is inevitable and accuracy and time are important to detect the change in different seasons to monitor trends and status of water quality. Understanding LULC and Water Quality Parameters (WQP) is necessary for efficient water management and reduction of operation water treatment cost. Therefore, the effect of agricultural land uses on water quality in Ndakaini, Gatanga, Muranga County, the LULC and WQP were analyzed using Remote Sensing (RS), Geographical Information System (GIS) and R Studio. Landsat 8 Ol.I was used to classify different LULC using. band 4, 3, 2. Semi- Automatic Classification Algorithm (SCP) was used to create a Region Of Interest (ROI) and Spectral Angle Mapping Algorithm (SAM) threshold of 90 and pixel size of 1500 was used to create thematic maps. Classification accuracy was 97.5 % overall accuracy and 0.90 kappa coefficiency in dry season, 97.8 % overall accuracy and 0.94 kappa coefficiency in a wet season. Digital Elevation Model (DEM) was also used to delineate channel network. Kiama (14), Kimakia (13), Chania (6) rivers were the longest and largest, and the slopes ranged between 2.78° and 58.3° The land surface factor values were 1.45 mis, 3.5 m/s, 5.54 m/s, 7.59 m/s and 9.63 m/s. Topographical Wetness Index (TWI) had -1.28 T.U,-0.314 T.U ,0.654 T.U,1.62 T.U and 2.59 T.U and flow direction -1.0 m/s to 6.99 m/s. Water samples were collected and referenced in dry and wet seasons from 8 sampling rivers. Eight WQP including pH, turbidity, conductivity, color, N03, N02, COD, and alkaline analyzed in the laboratory correlated in a different season with WHO standards where p= 0.001. The results showed that LULC in the wet and dry season significantly influenced WQP.