Climate variability and change and its potential impact on maize yield in Northeastern Zimbabwe
Masanganise, Joseph N
MetadataShow full item record
The main aim of this study was to investigate climate variability and change and its potential impact on maize yield in Northeastern Zimbabwe. Understanding the impact of climate change on crop production may help optimize schemes to increase or stabilize yields. The research used historic climatic data (precipitation and mean air temperature) as well as downscaled model predictions of the same parameters. The models used in the study were the same as those used by the Intergovernmental Panel on Climate Change (IPCC) in formulating the IPCC Special Report on Emissions Scenarios (SRES). Historic (1971-2000) climate trends for five climatological stations in Northeastern Zimbabwe were derived from observational data provided by the Zimbabwe Meteorological Services Department (ZMSD). Time series plots for temperature showed that the mean temperatures recorded at all the stations have increased by about 0.5 - 0.8 ºC during the period 1971 to 2000. Rainfall at Mt Darwin, Wedza and Rusape showed slight decreases, while no major changes were observed at Karoi and Agricultural Research Trust (ART) Farm. At Mutoko, rainfall showed an increase during the period 1971 to 2000. Five coupled global climate models were evaluated for simulating monthly precipitation and monthly minimum and maximum temperature. Model performance was assessed statistically using the following quantitative statistical indicators: coefficient of determination (R2), root mean square error (RMSE) and model efficiency (ME). For rainfall, in addition to R2 and RMSE, a frequency analysis test was applied at 50 % probability of exceedance. A t-test was then performed at 5 % level of significance to assess the correlation between observed and simulated data. Across all the three variables, four models performed relatively well but one of the models deviated substantially from the other four. All the models however did not perform well in predicting precipitation. The five models were ranked according to their performances and the highest performing model was applied in the study. Using the Climate Change Explorer tool, the highest performing model was applied to investigate the past and future climates at each station. In the case of temperature, downscaled model simulations from the best model consistently predicted a warming of between 1 and 2 ºC at most of the stations in the 2046-2065 regime above the baseline (1971-2000) period. In the case of rainfall, all models showed a wide variation in prediction. A crop growth simulation model, AquaCrop, was validated for simulating maize yields. Three quantitative statistical measures were used to assess its performance. These were R2, RMSE and the slope. A t-test was performed at 5 % level of significance to assess the usefulness of the model. Statistical analysis showed that AquaCrop is a usable and reliable tool in predicting maize yields. The potential impact of climate change on maize yield by 2046-2065 was investigated. Downscaled model simulations predicted that climate change will shift planting dates towards delayed planting in the 2046-2065 period. AquaCrop was used to simulate the impact on yield. Results of the simulations showed that if current planting dates are maintained in the period 2046-2065, maize yield will decrease but delayed planting will result in increased yields. AquaCrop also demonstrated that if planting is delayed, yields will be highest for short season varieties as compared to medium and long season ones.