Browsing by Author "WATURU MARGARET"

Filter results by typing the first few letters
Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    IMPACTS OF HUMAN ACTIVITIES ON WATER QUALITY IN THE UPPER ATHI RIVER CATCHMENT, KENYA
    (TUK, 2024-05) WATURU MARGARET
    V ABSTRACT In the past few decades, the impacts of human activities on water quality (WQ) have been demonstrated to have a strong negative effect on downstream users. In spite of this linkage, little is known about the specific effects on the Upper Athi River Catchment (UARC). The catchment has experienced rapid industrialization coupled with an increased human population that has led to the encroachment of the watershed. This study aimed at: (1) determining the levels/concentrations of physicochemical parameters (pH, Electrical Conductivity (EC), temperature, Total Suspended Solids (TSS), Total Nitrogen (TN), Total Phosphorus (TP), Biological Oxygen Demand (BOD), Chemical Oxygen Demand (COD), Chlorophyll a and heavy metals (Iron, Zinc, Lead, Copper, Manganese and Chromium) at different sampling sites in different seasons, (2) assessing Land Use Land Cover Changes (LULCC) in relation to water quality in the UARC using Geographical Information System (GIS) and Remote Sensing (RS) technologies, and (3) generating scenarios to predict future water quality trends by utilizing one-dimensional river and stream water quality (QUAL2Kw) and forecasting models. WQ samples were collected every month from February 2017 to December 2018, and analyses for nutrients and heavy metals were done in the laboratory following standard methods for water and wastewater analysis. WQ data was analysed using principal component analysis (PCA) and further subjected to the Kruskal-Wallis test. Dry and wet seasons data were analysed using ANOVA. Historical Landsat satellite data was acquired for the periods 1990, 2004, 2010, 2014 and 2018 to prepare LULCC maps of the study area. A hybrid classification technique was used to classify LULCC into six categories. The physicochemical parameters showed spatial-temporal variations, with a significant increase in TSS, Turbidity, and BOD recorded during the wet seasons. Nutrient concentrations were higher in stations adjacent to industrial, agricultural, and domestic effluents as compared to those in upstream areas. TP values were higher in two tributaries, Nairobi River at Njiru (0.17 mg/l) and Mathare (0.25 mg/l) than in Kikuyu springs (0.009 mg/l). Multiple regression analysis showed that EC, DO, Zn and Pb were significantly associated with urban areas (p < 0.009, p < 0.042, p < 0.031, and p < 0.02, respectively). Correlation results presented strong R2 values between the observed and predicted: temperature: 0.82, electrical conductivity: 0.99, total dissolved solids: 0.94, biochemical oxygen demand: 0.66, chlorophyll a: 0.89, total nitrogen: 0.75, and total phosphorus: 0.94. These results depict the model's reliability in predicting water quality parameters in rivers and streams, even in watersheds with little data availability. There was a strong correlation between the urban area and water quality parameters, with Cr (r = 0.56) and Pb (r = 0.80) as the most significant parameters. Shrubland was the most dominant land use type (56.5%) in 1990 but declined considerably to (49.16%) in 2018, mainly due to conversion into urban and agricultural land uses, whose increase tripled (from 0.93% to 3.11% and 5.66% to 18.17%, respectively). From the models, it was shown that DO will continue to decline towards the year 2030, indicating increased pollution. In contrast, EC, TSS, BOD, and iron will increase significantly in tandem with the present trends in LULCC. The implication is that pollutants will increase to unsafe levels in the UARC if appropriate watershed management actions are not taken in good time. Significant land cover degradation is expected to occur if no mitigation measures are instituted, creating a threat to biodiversity conservation and the survival of local communities.