Freshwater availability in sufficient quantity and quality is necessary for both people and nature. Environmental flow data is useful in the management and allocation of water resources. This study aimed at quantifying stream flows and their trends in the Malewa Basin rivers in central rift valley, Kenya. Daily stream flow data in four gauges (2GB01, 2GB05, 2GB0708 and 2GC04) were subjected to exploratory data analysis, fixed interval method of baseflow separation and Mann Kendall trend test. The results shows that on average, the Malewa river at Gauge 2GB01 discharge (excluding abstractions) about 191.2 million cubic metres of water annually, equivalent to a discharge of 6.06 m 3 /s. While discharges had not experienced a step change, huge annual fluctuations were noted suggesting periodicity with changes in climatic conditions. No trend was noted in annual stream data for the four gauges assessed. However, extreme low and high flows, median flows and baseflows for daily data showed either positive or negative trends. The baseflow index for daily flows showed trends: 2GB01 (Z = 4.519), 2GB05 (Z = -6.861), 2GB0708 (Z = -16.326) and 2GC04 (Z = 5.593). The findings suggest that Malewa rivers are likely experiencing effects of extreme climatic conditions and land cover changes. Land cover degradation seems to create conditions of increased flow, although the intensity varies from sub-catchment to another. The data also seems to suggest that stream discharge is much dependent on baseflows. There is need to regulate water use, improve soil cover and manage or adapt to the adverse effects of climate change.

Detection of land cover change helps in the understanding of how humans modify the natural environment. Modification is attributed to both restoration and degradation processes. Such information guides decisions on mitigating landscape degradation and advancing restoration. This study sets to determine land cover changes from 1973 to 2013 in the Malewa River Basin (1,760 km 2) in central rift valley, Kenya. Satellite imageries from Landsat (Landsat Multispectral Scanner, 1973; Landsat TM (Thematic Mapper), 1986; ETM+ (Enhanced Thematic Mapper Plus), 2000; and SPOT, 2013) were analyzed using various imaging techniques available in ArcGIS 10.1 and ERDAS Imagine software. The results showed a cumulative growth of 25,617.0 ha (28.8%) in area under cropland, an increase of 4,310.1 ha (11.3%) in forestland and 688.0 ha (490.7%) of wetland. There was a net decrease of 28,953.8 ha (72.2%) in the area under shrubland and 1,747.4 ha (19.2%) under grassland. The findings suggest that increased demand for arable land is mainly driven by food and income needs of the human population. This exerted enormous pressure particularly on shrublands and grassland. Increased forest cover suggests an improvement in forest restoration efforts during the last ten years. There is need to manage expansion into new arable areas by improving land productivity and tackling the drivers of land cover change.