![Figure: Case study of August 11, 2018. Convective/stratiform split of the raining system observed by GPM-core satellite (orbit: 025293). From left to right: (a) PMW-retrieved (GPROF) – a current operational benchmark; (b) Dual-frequency Precipitation Radar-derived product – the truth; (c) Bayesian model prediction ResNetV2; (d) Entropy for the Bayesian model prediction – uncertainty map.](https://essic.umd.edu/wp-content/uploads/2022/02/bayesian-model-predictions.png)
Using Bayesian Deep Learning to Improve Precipitation Retrievals
ESSIC/CISESS Scientist Veljko Petković co-authored a study on the application of new and emerging field of BDL concepts to mitigate problems associated with the accuracy of precipitation retrievals from satellite-borne passive microwave (PMW) radiometers, which was published in IEEE Geoscience and Remote Sensing Letters.