Enhanced precipitation prediction through the integration of gauge observations with satellite-based precipitation prediction models utilizing the Bayesian model averaging (BMA) technique in Kelantan, Malaysia

Husniyah Binti Mahmud; Takahiro Osawa

doi:10.1117/12.3038111

3 January 2025 Enhanced precipitation prediction through the integration of gauge observations with satellite-based precipitation prediction models utilizing the Bayesian model averaging (BMA) technique in Kelantan, Malaysia

Husniyah Binti Mahmud, Takahiro Osawa

Author Affiliations +

Proceedings Volume 13262, Remote Sensing of the Atmosphere, Clouds, and Precipitation VIII; 1326203 (2025) https://doi.org/10.1117/12.3038111
Event: Asia-Pacific Remote Sensing, 2024, Kaohsiung, Taiwan

Abstract

This study aimed to merge monthly precipitation from remote sensing products and rain gauge observations to provide high-quality long-term historical precipitation data to support climate change studies. In this study, two satellite precipitation products, the TRMM 3B43V7 and GPM-3IMERGM Final Level 3 precipitation analysis, were used as input datasets. This study used Autoregressive Integrated Moving Average (ARIMA) and Long Short-Term Memory (LSTM) prediction models to train a monthly satellite precipitation dataset from 1998 to 2018 and predict precipitation from 2019 to 2023. The Bayesian Model Averaging (BMA) method was employed to combine the modeled predicted precipitation dataset with 28 rain gauge data in Kelantan, Malaysia to consider each input impact and enhance the precipitation prediction. The BMA weights were assigned to three models: model_0 with the input variable ARIMA and rain gauge, model_1 with LSTM and rain gauge input, and model_2 with ARIMA, LSTM, and rain gauge as input variables. After assigning weights to each model, new ensemble datasets were calculated using the weight sum method and averaged to obtain the ensemble average model. The performance of the merged precipitation products was evaluated using RMSE, R², and NSE. The ensemble using the weighted average of all precipitation prediction models demonstrated an increase in R² to 18.70% and a reduction in RMSE of 3.56%. Another significant finding is that increasing the input variable does not enhance the accuracy of rainfall prediction, meanwhile by using ensemble averaging the gap between merged precipitation products and rain gauge data able to be reduced.

(2025) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Husniyah Binti Mahmud and Takahiro Osawa "Enhanced precipitation prediction through the integration of gauge observations with satellite-based precipitation prediction models utilizing the Bayesian model averaging (BMA) technique in Kelantan, Malaysia", Proc. SPIE 13262, Remote Sensing of the Atmosphere, Clouds, and Precipitation VIII, 1326203 (3 January 2025); https://doi.org/10.1117/12.3038111

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

;

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
16 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Satellites

Earth observing sensors

Rain

Bias correction

Data modeling

Climate change

Correlation coefficients

Show All Keywords

Keywords/Phrases

Search In:

Publication Years