Statistical Bias Correction for Predictions of Indian Ocean Dipole Index with Quantile Mapping Approach

Authors: S Nurdiati, A Sopaheluwakan, MK Najib. 

Abstract: The Indian Ocean Dipole (IOD) is a phenomenon of ocean-atmosphere interaction that affects climate conditions in Indonesia. The IOD index shows the difference between the western and eastern Indian Ocean sea surface temperature. The impact of the IOD can increase the risk of forest fires, floods and crop failure. Thus, an IOD index prediction model is needed to anticipate the impact of the IOD. One of prediction models of sea surface temperature is the ECMWF prediction model. However, this prediction model has systematic errors that can be corrected using a quantile mapping approach. This method corrects the systematic error of the ECMWF model by connecting the distribution between the ECMWF model and OISST in a transfer function, such as different of quantile and polynomial function. Based on the results, the linear function has the highest chance to improve the accuracy of the model. Moreover, the result shows that statistical bias correction is a good method to improve the accuracy of the ECMWF model especially in Januari-April and September-December. 

Keywords: statistical bias correction, quantile mapping, indian ocean dipole.

link: disini

Dipublikasikan pada "the 1st International Conference on Science and Mathematics (IMC-SciMath) 2019"