<![CDATA[The South Manokwari climatology station provides monthly rainfall forecast information in the form of deterministic and probabilistic rainfall distribution forecast information maps. Assessing the quality of forecast is verified the rainfall forecast information service can meet the standards and needs for the community. This research aims to compare the results of monthly rainfall forecast between ECMWF model and ARIMA statistical model with rainfall observation results to determine the performance of forecast results closer to observation. The data used in this research are observation rainfall data from 15 rain posts and 5 BMKG stations in West Papua and ECMWF reanalysis data. Observation data used as baseline is monthly rainfall data in 2019. The method used is ARIMA method as forecast method, then contingency table to calculate forecast suitability, PC to see accuracy and HSS for excellence. Furthermore, the results used rainfall criteria/classes consisting of 9 classes called quantitative rainfall forecasts and criteria consisting of 4 classes called qualitative rainfall forecasts. The results of monthly rainfall forecast in West Papua show that ECMWF data tends to overestimate the observation compared to ARIMA. PCH ARIMA quantitatively and qualitatively, the frequency of accuracy of forecasts to the appropriate observations is more than PCH ECMWF. PC value from PCH ARIMA is qualitatively more accurate and HSS value has advantages over PCH ECMWF. The performance of ECMWF and ARIMA in each rain post qualitatively has better accuracy, especially in southern part of West Papua. Meanwhile, quantitatively shows worse results almost all rain posts.]]>
Copyrights © 2021
