Predicting areas suitable for wheat and maize cultivation under future climate change scenarios in Pakistan

Abstract

Agriculture is greatly impacted by climate change, which can lead to situations of food security or insecurity at both regional and global levels. Pakistan is predicted to experience an area reduction and geographical shifting of major crops in the near future. In the present study, we assessed the potential future distribution of wheat and maize in Pakistan. Based on current locations of these crops, we ran a Maxent species distribution model to predict future changes in crop distributions. We used 58 presence records for wheat and 48 presence records for maize. The model simulated current and future climate change scenarios (RCP 4.5 and RCP 8.5) based on the CMIP5 model, MPI-ESM-LR. Results from our model showed a decline in production area, with a 30-35% reduction in wheat and a 23-36% reduction in maize in the year 2070, depending upon which climate change scenario was modelled (i.e. RCP 4.5 or RCP 8.5). The model predictions were highly accurate, with test AUC values of 0.88 for wheat and 0.89 for maize. A jackknife test for variable importance indicated that irrigation, precipitation seasonality and precipitation of the warmest quarter are the most important environmental variables determining the potential geographic distribution of the crops. Due to the varying severity and nature of climate impacts, adaptation strategies are needed. This study can aid policy makers in devising policies that can help reduce the threat of future food insecurity in the region.