Repurposing climate reconstructions for drought prediction in Southeast Asia

Abstract

Climate information may help smallholder farmers in Southeast Asia takeinformed risks and reduce their climate vulnerability. Model-based seasonalforecasts are limited in the skill they offer with longer lead times, and intheir accessibility to farmers in remote regions. Climate reconstructions derivedfrom tree rings can complement model-based forecasts, providing probabilisticestimates of future climate at an annual resolution. This study demonstratesthe skill of PDSI reconstructions derived from the Monsoon Asia DroughtAtlas (MADA) to forecast a climate signal several years into the future, andillustrates the advantage of such information vis-a-vis the simple strategy ofassuming climate to persist from year to year. This work shows the potentialfor such climate reconstructions to improve inter-annual decisions on SoutheastAsian farms, and highlights the social research necessary to transmute thispotential into a valuable climate decision aid.In much of Southeast Asia the capacity of smallholder farmers to predict andprotect against drought is limited. Farmers commonly employ conservative, risk-averse strategies to farm planning and operation that reduce losses to climate-relateddamages, but that also reduce surpluses in years of favorable climate (Meinke andStone 2005). Farmers operating close to thresholds of sustenance and self-sufficiencywho fail to take advantage of these rents can fall into ‘poverty traps’ (Holling 2001;Sachs 2005), unable to lift themselves from degraded land conditions and a lackof capital. To help smallholders become informed risk takers and capitalize on the