Validation of Satellite and Ground Station-based High Resolution Rainfall Datasets over the Philippines
Abstract
Characterizing the rainfall variability over the Philippines remains to be challenging due to the archipelago's complex shorelines and high terrain gradients, compounded by its location at the confluence of the synoptic- and global-scale climate systems of the Pacific Ocean and the warm active seas of the Maritime Continent. Few studies using ground observation data have been published partly due to the sparse spatial coverage of weather stations in the country. High resolution gridded rainfall products can supplement the need for observed rainfall data in the mountainous regions and remote islands, but require careful validation with existing ground-based measurements. This study assesses the performance of four gridded datasets, APHRODITEv1101, CHIRPSv2, TRMM 3B42v7 and PERSIANN-CDR, all at 0.25 spatial resolution, over the daily, monthly and seasonal scales. These datasets were compared to ground observation rainfall data from 49 stations of the national weather agency PAGASA located all over the Philippines from 1998-2005. Performance metrics such as bias, correlation coefficient, root-mean-square error, rainfall event distribution and partitioning, False Alarm Ratio, and True Skill Score were calculated. Results show that APHRODITEv1101 performs best in terms of skill score for rainfall amounts of up to 40mm/day. TRMM 3B42v7, on the other hand, has best skill at rainfall values that are greater than 40mm/day. APHRODITEv1101 has the highest daily correlation coefficients with station data (R2=0.633) , followed by TRMM 3B42v7 (R2=0.436) and CHIRPSv2(R2=0.292) and lastly PERSIANN-CDR(R2=0.290). Among the datasets, TRMM 3B42v7 has the least overall average daily bias per month ranging from -2.54 to +2.89 mm/day and most closely resembles the rainfall event distribution and partitioning seen in the observation station data.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2018
- Bibcode:
- 2018AGUFM.H43F2474P
- Keywords:
-
- 3354 Precipitation;
- ATMOSPHERIC PROCESSESDE: 3360 Remote sensing;
- ATMOSPHERIC PROCESSESDE: 1817 Extreme events;
- HYDROLOGYDE: 1847 Modeling;
- HYDROLOGY