Abstract:
Streamflow variability poses challenges to water security in Nairobi City, which
relies heavily on the Thika River watershed upstream of RGS 4CB05 in the Upper
Tana catchment. Over the past few decades, the watershed has faced marked
streamflow declines that have intensified water shortages and entrenched rotational
rationing. This was evident particularly during severe droughts such as the
2016/2017 event when Thika Dam dropped to critically low levels. At present,
Nairobi’s installed production capacity of approximately 525–560 million liters/day
remains far below the city’s demand, which exceeds 850 million liters per day.
Therefore, there is urgent need to disentangle the impacts of climate and land use
changes on Thika River watershed hydrology, as they are the main drivers of
streamflow variability. The objective of this study was to assess and quantify these
impacts on streamflow dynamics in the Thika River watershed. Mann-Kendall and
sequential Mann-Kendall tests were applied to detect trends and breakpoints, while
the Wang-Tang Budyko model partitioned the relative contributions of climate
variability and land use change. Drought patterns were analyzed using the
Standardized Precipitation-Evapotranspiration Index (SPEI) and Standardized
Runoff Index (SRI). Future streamflow was simulated using the SWAT (Soil and
Water Assessment Tool) model under the IPCC RCP 4.5 scenario for 2050–2065.
Findings revealed a significant decline in streamflow with a breakpoint in 2003.
Climate variability accounted for 46% of the reduction, mainly through decreased
precipitation and rising evapotranspiration, while land use changes dominated by
deforestation and agricultural expansion contributed 54%. Post-2003, hydrological
droughts became more severe and recurrent, with projections indicating worsening
conditions mid-century due to temperature rise and rainfall variability. The study
recommends strengthening hydro-climatic monitoring and drought preparedness
systems, mainstreaming climate and land use considerations into watershed planning,
and enforcing land use regulations alongside community-based reforestation,
afforestation, agroforestry, and soil-water conservation. Further research should
incorporate a wider range of climate (RCP 2.6, 6.0, 8.5) and land use scenarios to
better capture future uncertainties. By integrating the Budyko framework with
drought indices, this study provides a robust, transferable approach for managing
water resources in the Thika River watershed and similar climate-sensitive
catchments globally.
