IN034-06
An automated river ice mapping system over northern watersheds in the USA using VIIRS imagery

Tuesday, 15 December 2020: 04:20
Virtual
Marouane Temimi, Stevens Institute of Technology, Union City, NJ, United States, Naira Chaouch, NOAA CREST Institute/City University of New York, New York, NY, United States, Paul Alabi, NOAA-CREST/City College, CUNY, New York, United States, Peter Romanov, NOAA-CREST/City College, CUNY, New York, NY, United States, William C. Straka, University of Wisconsin Madison, Cooperative Institute for Meteorological Satellite Studies, Madison, WI, United States, Jay Hoffman, University of Wisconsin-Madiso, Madison, WI, United States, Mitch Goldberg, NOAA NESDIS Senior Scientist, Silver Spring, MD, United States and Bill Sjoberg, GST contractor for NOAA/NESDIS JPSS Program Office, Lanham, MD, United States
Recorded Presentation
Abstract:
Ice on rivers in northern US watersheds plays a significant role in the control of flow routing. Ice jams and breakup could lead to hazardous conditions and local flooding events especially when they develop and evolve rapidly. It is critical to develop a system to monitor river and lake ice conditions across all northern watersheds to predict ice break up and the potential flooding that it may lead to.

Here, we introduce an automated system for the mapping of ice on rivers and lakes within five northern river forecast centers (RFCs) domains part of NOAA National Weather Service. The system uses images from the Visible Infrared Imaging Radiometer Suite (VIIRS) onboard NOAA 20 and NPP satellites which provide the necessary spatial and temporal coverage to continuously map and monitor changes in ice conditions in river and lake ice. The system is a candidate for implementation as a NOAA official operational product.

Over high latitude watersheds, like in Alaska, VIIRS images could be collected from frequent overpasses which allow for continuous monitoring of ice conditions. The sensors' I-band images at a spatial resolution of 370 m are suitable for the monitor of ice condition on large rivers. In the case of narrow rivers, unmixing techniques were applied to separate the land and river signal and process the river inferred reflectance. Maps of ice extent and concentrations are developed operationally. The system uses the NOAA CLAVRX cloud probability product to develop the cloud mask over waterbodies. Moreover, the day and night band (DNB) images have shown great potential in the mapping of ice conditions especially in Alaska were light pollution is minimal. VIIRS images were processed to develop daily and running-cloud free composites. Aerial photography, radar images, namely Sentinel 1, and in situ observation were used for the assessment of the performance of the system.

<< Previous Abstract | Next Abstract >>