The effect of land surface characteristics on runoff generation and nitrate fluxes from a Kenyan tea plantation

Date:

Aaron Neill (1), Suzanne Jacobs (2), Lutz Breuer (3), and Sim Reaney (4)

1 Institute of Hazard, Risk and Resilience, Durham University, Durham, United Kingdom (aaron.james.neill@gmail.com) 2 Centre for International Development and Environmental Research (ZEU), Justus Liebig University, Giessen, Germany. 3 Institute for Landscape Ecology and Resources Management (ILR), Justus Liebig University, Giessen, Germany. 4 Department of Geography & Institute of Hazard, Risk and Resilience, Durham University, Durham, United Kingdom.

The conversion of tropical montane forests to commercial plantation agriculture affects both the generation of runoff and nutrient water quality, degrading ecosystem service delivery and impacting downstream freshwater environments. Previous empirical studies have inferred that changes to nitrate dynamics at the catchment scale likely reflect the interplay of local topography and climate, crop characteristics (e.g., type and density), fertiliser usage and soil management practices. However, the relative importance of such factors is not well understood. Through the 20th century, the Mau Forest Complex in Kenya experienced dynamic and rapid land use change, including the conversion of pristine forest to commercial tea and tree plantations. Utilising a unique long-term (2015-2021), high-resolution (10-minute) discharge and nitrate dataset collected for such a plantation (33.3 km2), we developed a simple, semi-distributed conceptual model to disentangle the drivers of runoff generation and nitrate fluxes. Insights from weekly stable water isotope data helped to further constrain simulated flow paths. The model represented the main land surfaces of the plantation (tea, eucalyptus, compacted tracks, and impervious and built surfaces) as a set of conceptual stores. Rainfall inputs were weighted by the proportional area of each surface and, where relevant, fertiliser inputs were estimated based on application rates reported in the literature. Lateral water and nitrate fluxes from each conceptual store to the river were delayed and transformed as a function of the mean distance to the river and slope gradient. The available long-term data combined with the structure of the model allowed the relative contribution of each land surface to runoff and nitrate fluxes to be successfully simulated under a range of local hydroclimatic conditions. These insights provide a valuable knowledge base for optimising fertiliser use and implementing mitigation measures to sustain water quality and ecosystem service delivery under conditions of expanding plantation agriculture.

How to cite: Neill, A., Jacobs, S., Breuer, L., and Reaney, S. 2024: The effect of land surface characteristics on runoff generation and nitrate fluxes from a Kenyan tea plantation, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-16580, https://doi.org/10.5194/egusphere-egu24-16580.