CREST is a fully distributed hydrologic model, with water & energy balances coupled module and a snow process


RAPID is a fully automated inundation mapping system for non-obstructed flood driven by Synthetic Aperture Radar (SAR) data and rely extensively on the use of multi-source remotely sensed and surveyed datasets.


The first global-scale distributed basin characteristics (GDBC) of 30 arcsec spatial resolution


A comprehensive flood-event database (FEDB) automatically retrieved from USGS flow records and Stage IV precipitation field

Released Flood Events


The inundation maps in GeoTiff format are generated from 10 m resolution Sentinel-1 images using the automatic RAdar Produced Inundation Diary (RAPID) algorithm. Inundation (pixels with label 1) refers to non-obstructed flooding only. The maps can only be used for non-commercial research, educational, and humanity purposes only. Please cite our publications when using/redistributing the maps.

Dr. Shen's research interests include Humans Disasters and the Built Environment (HDBE), flood Inundation Numerical Modelling, and remote sensing retrieval, machine learning, ecohydrology, and microwave remote sensing. You can find his Curriculum Vitae here.

Contact Us

Phone: (860) 486-8913
Address: Room 212B
Innovation Partnership Building
University of Connecticut
159 Discovery Dr
Storrs, CT 06269