Global Climate Indices and Its Effect on Extreme Climate Events in Indonesia

Abstract

Many facts show that El-Nino Southern Oscillation (ENSO) phenomenon and Dipole Mode are closely related to rainfall event in Indonesia, but the magnitude of its impact varies with site. Therefore, it is needed to determine the most singnificant global climate indices that has closely related to Indonesian’s rainfall. The significant indices are able to be used in predicting rainfall in Indonesia.The objective of the present study is to detemine global climate indicators that have the significant effect to rainfall, climate/season anomaly, the occurence flood and drought in Indonesia. The study has been done through the following steps ; 1) regression analysis of rainfall with global climate indices of Sea Surface Temperatur Anomaly/SST in Nino 3.4 zone, Dipole mode Index (DMI), Southern Oscillation Index (SOI), interacton of SSTA with DMI, and interaction of SOI with DMI), 2) Plotting of rainfall anomaly and global climate indices for determining rainfall deviates with deviant of global climate forcing indices, 3) analysis of probability of exceedence for determining onset and lenght of wet season on climate extreme event, and 4) analysis impact of climate extreme event on flood and drougt occurences and damage areas of ricefield. The result showed that the closest relationship between global climate forcing indices and rainfall in Indonesia is SST in Nino 3.4 zone and only have significant relationship to rainfall in transisional season (August-November). Negative correlation between SST and rainfall indicates that the increase of SST anomaly causes the decrease of rainfall on August-November period. Probability of exceedance analysis showed that if the SST on September decrease below -0.5 -0.5 oC (indicate La-Nina event), the wet season will start earlier with longer period. In contrast, if SST increases above 0.5 oC (indicate El-Nino event) the wet season will delay with shorter period. Impact of drought on damage of ricefield is more significant than the flood occurence.