Performance evaluation and error decomposition of CMORPH satellite precipitation estimation for Klang Valley, Malaysia
Satellite Precipitation Estimations (SPEs) have gained traction as a viable substitute for estimating urban rainfall. However, their performance assessment in Malaysia continues to be constrained. As an effort to enhance urban rainfall estimations, this study examines the performance of Climate Pred...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Proceedings |
| Language: | en |
| Published: |
Researchgate
2025
|
| Subjects: | |
| Online Access: | https://eprints.ums.edu.my/id/eprint/44112/1/FULL%20TEXT.pdf https://eprints.ums.edu.my/id/eprint/44112/ https://www.researchgate.net/publication/389307814_Performance_evaluation_and_error_decomposition_of_CMORPH_satellite_precipitation_estimation_for_Klang_Valley_Malaysia http://dx.doi.org/10.1088/1755-1315/1453/1/012050 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Satellite Precipitation Estimations (SPEs) have gained traction as a viable substitute for estimating urban rainfall. However, their performance assessment in Malaysia continues to be constrained. As an effort to enhance urban rainfall estimations, this study examines the performance of Climate Prediction Center Morphing technique (CMORPH) over urbanized areas of the Klang Valley from 2011 to 2020, through continuous and categorical statistical metrics. The continuous statistical metrics consists Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Percentage Bias (PBIAS) and Pearson’s Correlation Coefficient (CC). Meanwhile, the categorical statistical metrics include False Alarm Ratio (FAR), Probability of Detection (POD), Critical Success Index (CSI) and Accuracy (ACC). The performance assessment was conducted on seasonal scale, comprising of Northeast Monsoon (NEM), Southwest Monsoon (SWM), Intermonsoon 1 (IM1) and Intermonsoon 2 (IM2). CMORPH exhibits strongest correlation with rain gauge during IM2 with maximum correlation coefficient (CC) recorded at 0.52. However, it exhibited significant inaccuracies in PBIAS, MAE and RMSE with values of 2514.04%, 111.41 mm/day, and 210.94 mm/day, respectively. Besides, characteristics of errors in CMORPH were identified using the error decomposition technique, which contains overhit bias, underhit bias, false bias and miss bias. The analysis shows that overhit bias is the main error component that contribute to the total bias in CMORPH at seasonal scale. High false bias and substantial total bias, primarily due to overestimation and false detection, underscore the need for further refinement of the CMORPH algorithm. |
|---|