DFG (Germany): Land use impact on hydro-biogeochemical fluxes and processes in tropical catchments of Kenya

Long term field measurements are required to investigate the effects of land use change on large scales and over extended periods of time. To achieve this, we installed four fully automatic measurement systems with UV hyperspectral sensors in the South West Mau, an area of tropical montane forests in the highlands of Kenya. They provide long-term data to quantify and compare hydro-biogeochemical dynamics (discharge, nitrate, turbidity, total and dissolved organic carbon) in catchments with natural forest, smallholder agriculture and commercial tea plantations. Measurements at the outlet of each subcatchment give a good indication of the effect land use on stream flow patterns and hydro-biogeochemical dynamics during dry and wet seasons.

Within the DFG project, three different approaches were used to gain more knowledge on the influence of land use on hydro-biogeochemical processes in the South West Mau:

• Intensive spatial sampling campaigns, each lasting 48 hours, were carried out during the dry seasons of 2015 and 2016 to investigate spatial patterns in dissolved organic carbon and dissolved nitrogen. The results showed that land use specifically influenced concentrations of total dissolved nitrogen and nitrate. This is most likely due to fertilizer inputs on agricultural land.

• Data on water level and nitrate concentrations recorded at 10-minute intervals by the automatic measurement systems were analysed to investigate nitrate dynamics. Strong seasonal differences were observed in the catchments dominated by smallholder agriculture and commercial tea plantations, with highest concentrations during the rainy season due to inflow of groundwater with high nitrate concentrations. This nitrate probably originates from leached fertilizer. Again, nitrate concentrations were lowest in the natural forest catchment. The analysis of rainfall events suggested that surface runoff occurred in both agricultural catchments, but not in the forest.

• Weekly samples of rainfall, soil water, and stream water were collected over a period of eighteen months and analysed for stable isotopes of water. These were used to estimate the mean transit time - how long it takes for rainfall to become streamflow - and assess the contribution of 'young' water (less than 2 to 3 months old) to streamflow. Additional samples were analysed for trace elements and used to investigate water flow paths in the three land use types. All catchments had a mean transit time around 4 years and a young water fraction of less than 15%, indicating that groundwater is the most important source of stream flow in all land use types. Unlike the natural forest catchment, the smallholder agriculture and commercial tea plantation catchment showed an increased input of groundwater during the rainy season.

• Further information

  ---