STREAM‐Sat: A Novel Near‐Realtime Quasi‐Global Satellite‐Only Ensemble Precipitation Dataset

Abstract Satellite‐based precipitation observations can provide near‐global coverage with high spatiotemporal resolution in near‐realtime. Their utility, however, is hindered by oftentimes large uncertainties that vary substantially in space and time. This problem is particularly pronounced in regio...

Full description

Saved in:
Bibliographic Details
Main Authors: Kaidi Peng, Daniel B. Wright, Yagmur Derin, Samantha H. Hartke, Zhe Li, Jackson Tan
Format: Article
Language:English
Published: Wiley 2025-03-01
Series:Water Resources Research
Subjects:
Online Access:https://doi.org/10.1029/2023WR036756
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849422713283674112
author Kaidi Peng
Daniel B. Wright
Yagmur Derin
Samantha H. Hartke
Zhe Li
Jackson Tan
author_facet Kaidi Peng
Daniel B. Wright
Yagmur Derin
Samantha H. Hartke
Zhe Li
Jackson Tan
author_sort Kaidi Peng
collection DOAJ
description Abstract Satellite‐based precipitation observations can provide near‐global coverage with high spatiotemporal resolution in near‐realtime. Their utility, however, is hindered by oftentimes large uncertainties that vary substantially in space and time. This problem is particularly pronounced in regions which lack dense ground‐based measurements to quantify or reduce such uncertainty. Since this uncertainty is, by definition, a random process, probabilistic representations are needed to advance their operational application. Ensemble methods, in which uncertainty is depicted via multiple realizations of precipitation fields, have been widely used in numerical weather and climate prediction, but rarely in satellite contexts. Creating such an ensemble dataset is challenging due to the complexity of observational uncertainties and the scarcity of “ground truth” to characterize them. In this study, we attempt to resolve these two challenges and propose the first quasi‐global (covering all continental land masses within 50°N‐50°S) satellite‐only ensemble precipitation dataset (STREAM‐Sat), derived entirely from NASA's Integrated Multi‐SatellitE Retrievals for Global Precipitation Measurement (IMERG) and GPM's radar‐radiometer combined precipitation product (2B‐CMB). No ground‐based measurements are used to generate STREAM‐Sat, and it is suitable for near‐realtime use without extending the 4‐hr latency and 0.1°, 30‐min spatiotemporal resolution of IMERG Early. We compare STREAM‐Sat against several precipitation datasets, including global satellite‐based, rain gage‐based, atmospheric reanalysis, and merged products. While our proposed approach faces some limitations and is not universally superior to the comparison datasets in all respects, it does hold relative advantages due to its unique combination of accuracy, resolution, rainfall spatiotemporal structure, latency, and utility in hydrologic and hazard applications.
format Article
id doaj-art-cf19e1d8a3f14420b1761629fa8ad9ba
institution Kabale University
issn 0043-1397
1944-7973
language English
publishDate 2025-03-01
publisher Wiley
record_format Article
series Water Resources Research
spelling doaj-art-cf19e1d8a3f14420b1761629fa8ad9ba2025-08-20T03:30:57ZengWileyWater Resources Research0043-13971944-79732025-03-01613n/an/a10.1029/2023WR036756STREAM‐Sat: A Novel Near‐Realtime Quasi‐Global Satellite‐Only Ensemble Precipitation DatasetKaidi Peng0Daniel B. Wright1Yagmur Derin2Samantha H. Hartke3Zhe Li4Jackson Tan5Department of Civil and Environmental Engineering University of Wisconsin‐Madison Madison WI USADepartment of Civil and Environmental Engineering University of Wisconsin‐Madison Madison WI USADepartment of Civil and Environmental Engineering University of Wisconsin‐Madison Madison WI USANational Center for Atmospheric Research Boulder CO USADepartment of Electrical and Computer Engineering Colorado State University Fort Collins CO USANASA Goddard Space Flight Center Greenbelt MD USAAbstract Satellite‐based precipitation observations can provide near‐global coverage with high spatiotemporal resolution in near‐realtime. Their utility, however, is hindered by oftentimes large uncertainties that vary substantially in space and time. This problem is particularly pronounced in regions which lack dense ground‐based measurements to quantify or reduce such uncertainty. Since this uncertainty is, by definition, a random process, probabilistic representations are needed to advance their operational application. Ensemble methods, in which uncertainty is depicted via multiple realizations of precipitation fields, have been widely used in numerical weather and climate prediction, but rarely in satellite contexts. Creating such an ensemble dataset is challenging due to the complexity of observational uncertainties and the scarcity of “ground truth” to characterize them. In this study, we attempt to resolve these two challenges and propose the first quasi‐global (covering all continental land masses within 50°N‐50°S) satellite‐only ensemble precipitation dataset (STREAM‐Sat), derived entirely from NASA's Integrated Multi‐SatellitE Retrievals for Global Precipitation Measurement (IMERG) and GPM's radar‐radiometer combined precipitation product (2B‐CMB). No ground‐based measurements are used to generate STREAM‐Sat, and it is suitable for near‐realtime use without extending the 4‐hr latency and 0.1°, 30‐min spatiotemporal resolution of IMERG Early. We compare STREAM‐Sat against several precipitation datasets, including global satellite‐based, rain gage‐based, atmospheric reanalysis, and merged products. While our proposed approach faces some limitations and is not universally superior to the comparison datasets in all respects, it does hold relative advantages due to its unique combination of accuracy, resolution, rainfall spatiotemporal structure, latency, and utility in hydrologic and hazard applications.https://doi.org/10.1029/2023WR036756precipitationuncertaintysatelliteensembleIMERGGPM
spellingShingle Kaidi Peng
Daniel B. Wright
Yagmur Derin
Samantha H. Hartke
Zhe Li
Jackson Tan
STREAM‐Sat: A Novel Near‐Realtime Quasi‐Global Satellite‐Only Ensemble Precipitation Dataset
Water Resources Research
precipitation
uncertainty
satellite
ensemble
IMERG
GPM
title STREAM‐Sat: A Novel Near‐Realtime Quasi‐Global Satellite‐Only Ensemble Precipitation Dataset
title_full STREAM‐Sat: A Novel Near‐Realtime Quasi‐Global Satellite‐Only Ensemble Precipitation Dataset
title_fullStr STREAM‐Sat: A Novel Near‐Realtime Quasi‐Global Satellite‐Only Ensemble Precipitation Dataset
title_full_unstemmed STREAM‐Sat: A Novel Near‐Realtime Quasi‐Global Satellite‐Only Ensemble Precipitation Dataset
title_short STREAM‐Sat: A Novel Near‐Realtime Quasi‐Global Satellite‐Only Ensemble Precipitation Dataset
title_sort stream sat a novel near realtime quasi global satellite only ensemble precipitation dataset
topic precipitation
uncertainty
satellite
ensemble
IMERG
GPM
url https://doi.org/10.1029/2023WR036756
work_keys_str_mv AT kaidipeng streamsatanovelnearrealtimequasiglobalsatelliteonlyensembleprecipitationdataset
AT danielbwright streamsatanovelnearrealtimequasiglobalsatelliteonlyensembleprecipitationdataset
AT yagmurderin streamsatanovelnearrealtimequasiglobalsatelliteonlyensembleprecipitationdataset
AT samanthahhartke streamsatanovelnearrealtimequasiglobalsatelliteonlyensembleprecipitationdataset
AT zheli streamsatanovelnearrealtimequasiglobalsatelliteonlyensembleprecipitationdataset
AT jacksontan streamsatanovelnearrealtimequasiglobalsatelliteonlyensembleprecipitationdataset