Presentation | H13C: Advances in Machine Learning for Earth Science: Observation, Modeling, and Applications II Oral
Oral
H13C-06: Leveraging Passive Microwave Data (MiRS) With U-Net Architecture for Satellite Rainfall Estimation in High-Latitude Regions: A Case Study over Alaska
Author(s): Mohammad bolboli Zadeh, University of California Irvine (First Author, Presenting Author) Vu Dao, University of California Irvine Kuo-lin Hsu, University of California, Irvine Phu Nguyen, University of California, Irvine Soroosh Sorooshian, University of California, Irvine
This study focuses on the MiRS dataset over the Alaska region. We use a U-Net architecture to: 1) bias-correct the MiRS rain rate over Alaska, resulting in a 24% reduction in mean squared error (MSE) compared to the original MiRS estimates; and 2) estimate rainfall using other MiRS variables besides rain rate, achieving a 25% reduction in MSE compared to the original MiRS estimations.
