Description

Abstract: The Livestock Early Warning System Project (LEWS) has created a technology suite of products to predict forage conditions on the ground in pastoralist regions of East Africa. The system monitors the impact of emerging weather events on forage supply for livestock. The PHYGROW model is the foundation of the LEWS toolkit. Primary inputs for the model include: soil parameters, plant community characteristics, and livestock management decision rules. These inputs are driven by satellite-based gridded weather data for a particular location, simulating daily forage available for livestock and wildlife. Regular verifications are conducted to ensure that PHYGROW’s simulation output of available forage agrees with observations in the field. LEWS scientists sampled 81 of over 300 monitoring sites scattered across Ethiopia, Kenya, Tanzania, and Uganda throughout the entire vegetation production cycles of 2001 and 2002. The zonal teams estimated the total forage standing crop (kg/ha) for 50 plots in each of the 81 grids, based on pre-established reference quadrats. The field data collected by the zonal teams was highly correlated with the simulated model output (R2= 0.96 and SEP = 161 kg/ha). The methodology was judged effective in providing large-scale estimates of forage dynamics on the ground, offering a low-cost mechanism for translating weather data into forage over large regions. Satellite-based weather data, coupled with a robust biophysical model like PHYGROW, is a viable base to support early warning efforts in developing countries.