Articles | Volume 15, issue 17
Research article
02 Sep 2022
Research article |  | 02 Sep 2022

GPROF-NN: a neural-network-based implementation of the Goddard Profiling Algorithm

Simon Pfreundschuh, Paula J. Brown, Christian D. Kummerow, Patrick Eriksson, and Teodor Norrestad​​​​​​​

Data sets

GPM GMI Common Calibrated Brightness Temperatures Collocated L1C 1.5 hours 13 km V07 W. Berg

GPM MHS on NOAA-18 Common Calibrated Brightness Temperature L1C 1.5 hours 17 km V07 W. Berg

GPM DPR and GMI Combined Precipitation L2B 1.5 hours 5 km V07 W. Olson

MRMS QPE Multi-RADAR Multi-Sensor Archiving

Model code and software

GPROF-NN: A neural network based implementation of the Goddard Profiling Algorithm S. Pfreundschuh

Short summary
The Global Precipitation Measurement mission is an international satellite mission providing regular global rain measurements. We present two newly developed machine-learning-based implementations of one of the algorithms responsible for turning the satellite observations into rain measurements. We show that replacing the current algorithm with a neural network improves the accuracy of the measurements. A neural network that also makes use of spatial information unlocks further improvements.