Preliminary study on the validation of ECMWF with the ground-based data for the Indonesia Fire Danger Rating System (Ina-FDRS)

Abstract

The validation process of the European Centre for Medium-Range Weather Forecasts (ECMWF) and ground- based rainfall data was carried out during the period 2014 to 2017 at South Sumatra. Data validation was carried out by comparing 4 (four) weather parameters such as total precipitation, relative humidity, temperature, and wind speed, obtained from Meteorological, Climatological, and Geophysical Agency (BMKG) with ECMWF data at the same coordinate locations.Rainfall data from an AWS at Palembang Station is in accordance with ECMWF rainfall data, where if there is no rainfall on AWS then there will be no rainfall in ECMWF, and vice versa. The highest correlation value of average rainfall at Banyu Asin is 0.1509. The lowest correlation value of 0.0721 is at the Ogan Komering Ilir (OKI). Based on the correlation value, the average rainfall has a higher correlation than the total rainfall from ECMWF. For the relative humidity, the correlation results of the Palembang Station are 0.51 and OKI Station has a value of 0.49. The temperature from 2014 – 2017 shows the same composition and pattern of events but there are several events of ECMWF data that have a greater value than the Automatic Weather Station (AWS) data. The correlation is 0.494, this is quite large, and it is possible that the correlation value is due to the data in 2015 which was slightly overestimated. The validation of wind speed data has different patterns and values between surface and ECMWF data and the correlation results are 0.233.