Adapting Nyando Smallholder Farming Systems To Climate Change And Variability Through Modelling

Abstract

This study was carried out in Nyando, Kisumu County to model maize production under different climate scenarios and project the yields for the years 2030 and 2050. A crop model, Decision Support System for Agrotechnology Transfer (DSSAT) was used under rain fed conditions to simulate the effects of climate change on maize production and project the future yields. Three maize varieties were used; Katumani Comp B as early maturing variety, Hybrid 511 as a medium maturing variety and Hybrid 614 as a late maturing variety. Three global coupled models (GCMs) CSIRO-MK3-6-0, HadGEM2-ES and MIROC-ESM under representative concentration pathways (RCP) 4.5 and 8.5 were used to downscale Nyando’s climate data for the years 2030 and 2050. This data together with past 50 year’s climate data was entered into Weatherman and ran. Minimum annual temperatures were getting warmer by 0.0050C while maximum annual temperatures were increasing by 0.0070C. Trends in annual rainfall showed reduction in coefficient of variation from 39 % in the period 1981 to 1990 to 24% from the year 2001 up to 2015. The projected maize yields showed that the yields will reduce in the years 2030 and 2050. This could be due to the negative effects of projected increase in temperatures in the three GCMs. However, projections showed that Katumani Comp B maize variety will have better yields compared to H511 and H614 because it requires less rain and also hardy in hot climate. The yield under RCP 4.5 for the year 2030 for Katumani Comp B was 2369 kg ha-1 under HadGEM while H511 had lowest projected yields of 1661 kg ha-1 under MIROC. Projection under RCP 8.5 for the year 2030 showed Katumani Comp B and H511 will yield 3319 and 3003 kg ha-1, respectively under MIROC. The lowest simulated yields were 1867 kg ha-1 for H614 under CSIRO. xv The maize yield projections for the year 2050 under RCP 4.5 showed that Katumani Comp B will give better yields by 3142 kg ha-1 under MIROC with H511 yielding lowest by 1643 kg ha-1 under CSIRO. The same trend was observed under RCP 8.5 with simulated yields of Katumani Comp B of 2819 kg ha-1 under MIROC. H614 projected lower yields of 1534 kg ha-1 under HadGEM. Lack of fertilizer application showed yield reduction of up to 40.8% in Katumani Comp B, 38.3% loss in H511 and 37.7% loss in H614. In conclusion, the study found out that Katumani Comp B maize variety responded well to climate change compared to H511 and H614 maize varieties therefore well adapted in Nyando. Also, the use of DSSAT crop model was good enough to project ideal maize yields in Nyando under present and projected future climatic conditions. Key words: Climate change, DSSAT, Global Coupled Models, Maize yield.