부산대학교 도서관

The Mekong Delta is the most important food production area in Vietnam, but salinity intrusion during the dry season poses a serious threat to agricultural production and livelihoods. A seasonal forecast of salinity intrusion is required in order to mitigate the negative effects. This communication presents a statistical seasonal forecast model based on logistic regression using either the ENSO34 index or streamflow as a predictor. The model is able to reliably predict the salinity intrusion up to 9 months ahead (receiver operating characteristic (ROC) scores: >0.8). The model can thus be used operationally as a basis for timely adaptation and mitigation planning.