Using Sensorweb Technologies to Monitor Flooding in Thailand

By Chien, et all , posted on December 2nd, 2013 in ESTO Showcase 2013

Steve Chien1, David Mclaren1, Joshua Doubleday1, Daniel Tran1,,
Veerachai Tanpipat2,Royol Chitradon2,Surajate Boonya-aroonnet2, Porranee Thanapakpawin2, Daniel Mandl3
1Jet Propulsion Laboratory, California Institute of Technology, 4800 Oak Grove Drive, Pasadena, CA, 91109-8099, USA Email:{firstname.lastname}
2Hydro Agro Informatics Institute, Bangkok, Thailand,
3 Goddard Space Flight Center, Greenbelt, MD, USA



Flooding has a tremendous impact on humanity and is worldwide in scale. Thailand (as well as greater Southeast Asia) is particularly prone to flooding, as observed during the 2010-2011 and 2011-2012 flood seasons in Thailand. The Thailand flood of October-November 2010 was responsible for more than 200 deaths, caused more than $1.67 billion (U.S.) in damages, and affected more than 7 million people. The 2011-2012 flood season was even more severe (See Figure 1), accounting for more than  600 deaths and $45.7 billion (U.S.) in damage.

In a collaboration  of NASA’s Jet Propulsion Laboratory and the Hydro and Agro Informatics Institute of Thailand, sensorweb technologies are being used to enhance monitoring of flooding in Thailand. This Thailand Flood Sensorweb utilizes automation techniques to detect flooded areas, alert interested parties, retask remote sensing assets to acquire high resolution imagery, derive analysis products from this imagery, and deliver this imagery. This system has been in operations during the last three flooding seasons and has rapidly delivered more than 100 analysis products.

Image Credit: JPL/Caltech.

Image Credit: Color Imagery: EO-1/NASA GSFC, Analysis Products: JPL/Caltech

The sensorweb system can automatically detect flooded areas using MODIS imagery and can also accept manual notifications from Thai authorities. The automatic MODIS-based flood detection retrieves MODIS band 7-2-1 subset data and uses green and infrared bands to distinguish flooded areas from land.  Flooded area pixel counts are then compared to dry season baseline measurements to detect flooding. These detections/alerts then trigger an automated response system using the Earth Observing-1 (EO-1) spacecraft but also including semi‐manual and manual alerts to a range of space assets. The EO-1 spacecraft has an automated tasking system that accepts urgent, “last minute” electronic observation requests and will fit them into the operational schedule based on pre-determined mission priorities. In partnership with other missions, we also generate electronic requests and notifications of these flooded area targets that are then forwarded to appropriate disaster response points of contact. Once the data is acquired, a combination of automated and semi-automated processes are used to access the data. Once the data is retrieved, workflows automatically determine surface water extent using both support vector machine-learned classifiers and ratios of green to infrared spectral bands. Surface water extent maps are then combined with digital elevation map information to produce water depth maps. Both of these products are then automatically electronically pushed to HAII.

Depending on timing and the sensor, this entire process from detection to product delivery can take anywhere from one to several days. In the past, three flood seasons products have been developed and delivered using the Thai Flooding Sensorweb from a range of space-based satellites and sensors including: EO-1 (ALI, Hyperion), Worldview-2, Ikonos, GeoEye‐1, Landsat 7, and Radarsat‐2. Samples of the input imagery and products for EO‐1/ALI, Landsat 7, and Worldview-2 are shown in Figures 2-4.

Future plans include integrating in-situ data sources and hydrological models as well as additional sensors such as ASTER.

Portions of this work were performed by the Jet Propulsion Laboratory, California Institute of Technology under contract with NASA. Copyright 2013. All rights reserved.


Image Credit: JPL/Caltech.

Image Credit: Color Imagery: USGS, Analysis Products: JPL/Caltech











Image Credit: JPL/Caltech.

Image Credit: Color Imagery: Digital Globe, Analysis Products: JPL/Caltech



facebooktwittergoogle_plusredditpinterestlinkedinmailby feather