Performance Comparison of Bias-Corrected Satellite Precipitation Products by Various Deep Learning Schemes

Precipitation observations from a ground-based gauge provide a reliable data source for hydrological and climatological studies. However, these data are sparse in many regions of the world, particularly the Mekong River Basin (MRB). Satellite-based precipitation products (SPPs) are the sole data source available with worldwide coverage. Despite this, there is a mismatch between SPPs and gauge-based observations, and the correct procedures should be utilized to minimize systematic bias in SPPs. This study aimed to benchmark the efficacy of four state-of-the-art bias-correcting deep learning models (DLMs) for the tropical rainfall measuring mission-based precipitation product named TRMM_3B42 (hereafter TRMM) over the entire MRB. These models were designed mainly based on convolutional neural network (CNN) and encoder-decoder (ENDE) architectures, including ConvENDE, ConvUNET, ConvINCE, and ConvLSTM. The bias-corrected dataset by DLMs was then confirmed against the gauge-based dataset (Asian precipitation-highly resolved observational data integration toward evaluation of water resources, APHRODITE). From the results obtained, all four DLMs effectively minimized the bias of the TRMM product. Among them, ConvENDE and ConvUNET had a higher consistency and performance level compared to ConvINCE and ConvLSTM. Additionally, the complexity of DLMs did not enhance their efficiency, as is the case with ConvINCE and ConvLSTM, despite using many computing resources. Given the observed data shortage for the MRB since 2016, the application of DLMs, such as ConvENDE and ConvUNET, can serve to improve the reliability of existing rainfall datasets and provide valuable input for various research purposes in the MRB.

Automated summary

This study evaluated the efficacy of four state-of-the-art deep learning models in reducing bias in precipitation data in the Mekong River Basin (MRB). The results showed that ConvENDE and ConvUNET models had a higher level of consistency and performance compared to ConvINCE and ConvLSTM models.