GC23C-1160
Attributing the Risk of Late Onset of the Rainy Season in Southern Africa to Climate Change

Tuesday, 15 December 2015
Poster Hall (Moscone South)
Piotr Wolski, University of Cape Town, Cape Town, South Africa
Abstract:
Rainfed subsistence agriculture in Sub-Saharan Africa accounts for approximately 96% of all cropland. This, combined with strong intra-seasonal and interannual variability of rains makes food production sensitive to climate variations, and increases the potential and frequent occurrence of climate-triggered famines. Farmers often identify the timing of the onset of the growing season (in many areas dependent predominantly on rainfall) as a key climate characteristic which influences crop yields, influences planting activities, and can be used to adapt to changing seasonal conditions without requiring additional resources. It is thus important to understand factors affecting the timing of the onset of rains, particularly how anthropogenic climate change may increase the risk of later onsets.

Here, we present a study designed to assess the level of influence of anthropogenic CO2 emissions on the risk of late onset in southern Africa. Considering numerous definitions of rainy season onset, we use one that describes onset related to the growing of maize, as it is the most wide-ranging and consumed staple in the region. Attribution is done using a risk-based event attribution methodology in which we use dedicated simulations by AGCMs (HadAM3p-N96 and CAM5.1-1degree) performed in the framework of the Climate of the 20th Century Plus (C20C+) Detection and Attribution Project. These simulations enable comparison of event (in our case the timing of the onset of a particular rainy season) probabilities under real world climates with those under a climate that might have been had human activities not emitted greenhouse gases. The fraction of risk of later onsets, attributable to an increase in greenhouse gases, is done in a spatially-explicit way for onset events derived from observed and reanalysis data for the 2007-2014 period.