This paper aims to study how alternative trade policies will help mitigate the effects of climate change in agriculture in South Asia. We use a modified version of MIRAGE CGE for long term projections and allowing modeling of climate change effects (impact on yield) at a subregional level (163 geographical units at the world level) to simulate the effects of 13 SRES scenarios in 8 different trade policy landscapes. Based on these results, we discuss the ranking of trade policy options based on expected values but also in terms of variance using the theory of decision in uncertainty. Choices between unilateral and regional strategies for the countries of the sub regions are compared. Our results confirm that South Asia will be one of the most adversely affected regions in terms of the impacts of climate change on agricultural yield. Both the overall level of economic activity and trade flows will react to this change (-0.5 percent of real income for the region in average, up to -4 percent for Pakistan). Beyond national real income, we also look at the distributional effects of climate change. Unskilled worker real wages, proxy for poor people income, are largely and generally negatively impacted by climate change. We show that trade policies weakly affect the overall economic impact of climate change but leads to more significant changes for the poor.