Using ENSO conditions to optimize rice yield for Nepal's Terai

The direct application of forecasts from seasonal prediction systems (SPSs) in agriculture is limited by their skill, and SPSs are more skilled at El Nino-Southern Oscillation (ENSO) prediction than precipitation prediction. An alternative to the direct application of forecasts from SPSs could be to link the forecast of ENSO conditions with dynamic crop models to evaluate alternate crop management options prior to the start of the actual planting. Although potential benefits of this approach have been tested in many areas of the world, so far limited evidence exists regarding its application in Nepal's Terai region. The overall goal of this study was to determine the potential relationship between ENSO and summer monsoon precipitation over Nepal's Terai and ascertain SPSs' skill in predicting ENSO. This analysis included disentangling the relative contribution of precipitation to interannual variability in rice yield from other factors using a cropping system model, namely, the Crop Environment Resource Synthesis-Rice (CSM-CERES-Rice). The crop model was also employed to explore options for increasing rice yield and minimizing risk by adjusting crop management. This study found that precipitation was the main variable affecting interannual variability in rice yield, that SPSs are good at predicting ENSO, and that the ENSO signal can be used to predict seasonal precipitation anomalies in the study area in all years except ENSO neutral years. Prior knowledge of seasonal precipitation anomalies can then be used to optimize rice yield using a crop model, and ultimately to assist farmers with decision making.