Estimation of net ecosystem exchange at the Skukuza flux site, Kruger National Park, South Africa

Abstract

Annual estimates of gross primary production (GPP) and ecosystem respiration (Reco) were obtained for the Skukuza flux site, Kruger Park, South Africa, based on the eddy covariance flux data. A new method of extrapolating night-time respiration to the entire day and filling gaps in eddy-covariance data in semi-arid systems was developed. The purpose for developing this method was to better account for the manner in which net ecosystem exchange (NEE) in drylan systems occurs as pulses driven by rainfall events, compared to current standard interpolation procedures developed primarily for temperate flux sites. The standard techniques furthermore do not take into account the decrease in respiration at very high soil temperatures. An artificial neural network (ANN) model was used to model GPP and Reco by incorporating fraction absorbed photosynthetically active radiation (fAPAR), the timing and magnitude of rainfall events, and temperature. The ANN predicted measured fluxes accurately (MAE 0.42 gC/m2/day), and was able to represent the seasonal patterns of photosynthesis and respiration at the site. The annual integral of the filled NEE data was found to range from -138 to +155 g C/m2/y over the five years eddy covariance measurement period. A full explanation of the methods and a full analysis of the data can be found in Archibald et al. (2009).