Abstract:
In Botswana crop production has been the most vulnerable component of the agricultural sector
due to its heavy reliance on rainfed agriculture. At the same time, the availability of rainfall is
uncertain due to natural climatic variability which is worsened by climate change. Already the
local farmers have knowledge of the climatic conditions around them, and from time to time, they
have been adjusting their farming practices in line with the prevailing climatic conditions. Climate
change is already imposing threats to most social- economic livelihoods including agriculture.
Therefore this study focuses on the impact of climate variability and climate change on dry land
crop productivity within Barolong Farms located in the Goodhope Sub District of Botswana. In
order to understand and quantify changes in the future, knowledge of the historical (baseline)
climate is important as such the initial component of the study involved historical analyses of
precipitation and temperature for the period 1979-2014. A relatively low variability (3.8%) in the
mean annual temperature is observed, but with a warming of 0.03 oC/year and 0.05 oC/year
respectively in the average maximum and minimum temperatures. Contrary to temperature,
historical rainfall exhibited high variability ranging from 59% to 91% for the wet seasons.
Projected sorghum yield for the period 2030-2060 under the Representative Concentration
Pathway (RCP) scenario of RCP 4.5 was simulated using the AquaCrop model which was
calibrated and validated based on observed sorghum yields of the Goodhope Agricultural
Experimental Station and Barolong Farms respectively. For calibration, the model performed well
for the study area as indicated by the coefficient of determination (R2) of 0.77, a root mean square
error (RMSE) of 0.41 and a Nash Sutcliffe Coefficient of Efficiency (CE) of 0.78. Using five
selected GCMs the calibrated model was then used to simulate sorghum yields under the projected
climate of 2030-2060. The mean average temperature for the cropping season is projected to
increase through a range of +5% to +20%, while the rainfall may change by +5% to -25% across
all the GCMs under the climate scenario of RCP 4.5. The future yield based on the projected
temperature and rainfall is expected to decrease by -5% to -15%. This study concludes that there
is a need to explore adaptation strategies under different climate scenarios that are climate smart
and tailor made for the local situation.
