Advancing Satellite Precipitation Retrievals With Data Driven Approaches: Is Black Box Model Explainable?
Open Access
- 23 February 2021
- journal article
- research article
- Published by American Geophysical Union (AGU) in Earth and Space Science
- Vol. 8 (2)
Abstract
Satellite‐based precipitation retrieval is an essential and long‐standing scientific problem. With an increase of observational satellite data, the advances of data‐driven approaches such as machine learning (ML)/deep learning (DL) are favored to deal with large data sets and potentially improve the accuracy of precipitation estimates. In this study, we took advantage of new technologies by wrapping up a ML/DL‐based model pipeline (LinkNet segmentation + tree ensemble). This approach is applied to the Advanced Microwave Sounding Unit (AMSU) on National Oceanic and Atmospheric Administration 18 and 19 flight, and compared with the MultiRadar MultiSensor. Four simulations were configured to examine the performance gain by incorporating three components: (1) precipitation identification, (2) nonlocal features, and (3) precipitation classification. More importantly, we examined the interpretability of the “black box” model to get a better understanding of the underlying physical connections. First, the results by this model pipeline suggest the advantages of the ML model by reducing the systematic error and instantaneous error to a factor of two. Second, identifying precipitation pixels helps to reduce the systematic error by 130%, and predicting precipitation classification benefits improved correlations by 32%. Last, channels at higher frequencies (beyond 150 GHz) are favored to identify precipitation regions, and also channels at 89 and 150 GHz are ranked as the two most important features to precipitation retrieval. This study explores the potentials of AMSU precipitation estimates with ML algorithms and provides means of interpreting the models to facilitate the better prediction of precipitation.Keywords
This publication has 33 references indexed in Scilit:
- Global Precipitation Estimation and Applications2012
- Detection and Measurement of Snowfall from SpaceRemote Sensing, 2011
- A Study of the NOAA Near-Nadir AMSU-A Brightness Temperatures over AntarcticaJournal of Atmospheric and Oceanic Technology, 2010
- PERSIANN-MSA: A Precipitation Estimation Method from Satellite-Based Multispectral AnalysisJournal of Hydrometeorology, 2009
- Precipitation Estimation from Remotely Sensed Imagery Using an Artificial Neural Network Cloud Classification SystemJournal of Applied Meteorology and Climatology, 2004
- Advanced microwave sounding unit cloud and precipitation algorithmsRadio Science, 2003
- Precipitation characteristics over land from the NOAA‐15 AMSU sensorGeophysical Research Letters, 2000
- Estimation of physical variables from multichannel remotely sensed imagery using a neural network: Application to rainfall estimationWater Resources Research, 1999
- Water vapour profile retrievals from SSM/T-2 data constrained by infrared-based cloud parametersInternational Journal of Remote Sensing, 1997
- A simplified scheme for obtaining precipitation and vertical hydrometeor profiles from passive microwave sensorsIEEE Transactions on Geoscience and Remote Sensing, 1996