Colorado Water Resources – NASA DEVELOP Fall 2013 @ NCCSC


[NASA Intro] [Sound of helicopter]>>Ryan Anderson: On September 12, 2013, tropical
moisture drawn north from Mexico by a low-pressure system descended upon the Rocky Mountains
and began to release its contents over the Colorado Front Range. Ensuing rains resulted
in the loss of nine lives and thousands of others displaced or stranded as damage to
200 miles of roadways cut off access to numerous residential areas. The Colorado Water Resources
team worked to observe and assess the damage caused by these historic flood waters.>>Steve Chignell: Hi, I’m Steve Chignell.>>Ryan Anderson: I’m Ryan Anderson.>>Amber Weimer: I’m Amber Weimer.>>Sky Skach: I’m Sky Skach.>>All: We are the Colorado Water Resources
Team.>>Steve Chignell: This project had two main
objectives. The first was to map the flood extent and calculate total inundated area
across the Front Range. The second was to compare these results with the region’s
pre-existing 100 and 500 year flood maps developed by FEMA.>>Melinda Laituri: This really provides the
basis for a number of different projects that are going to come out of this. I see this
as the opportunity to compare across the different drainages that experience the flood. We see
this as the basis for understanding and demonstrating levels of impact across the extent of where
the flood occurred. We also see this as the opportunity to partner with other entities
that are doing research on the flood and provide again this base layer to build a comprehensive
regional approach to understanding this flood event.>>Amber Weimer: The study area includes counties
affected by flooding in north-central Colorado excluding areas above 1800 meters. Landsat
8 imagery from August 16th and September 17th, 2013 were used to represent pre flooding and
during flooding conditions. The National Hydrography Dataset layer was added to both images in
order to capture all water pixels on the landscape. Before and during flood imagery was used to
calculate the Modified Normalized Difference Water Index and determine water and non-water
pixels for each scene. In the pre image, water was assigned a value of 1, while in the post
image, water was assigned a value 2.>>Sky Skach: When these layers were added
together the resulting values of 0 represent non-water areas, values of 1 represent water
to non-water, a rare occurrence which may indicate land-use change, and values of 2
represent flooded areas where non-water became water. Finally, values of 3 represent areas
of water in both images, such as pre-existing rivers and lakes. Calculations conducted on
pixels with a value of 2, the flooded areas, revealed that approximately 16,000 acres of
land were inundated 2 days after peak flow, at the time that the Landsat image was captured.
The comparison between total inundated area and existing FEMA flood hazard maps for both
100 and 500 year floodplains largely agreed, but revealed areas where flood waters were
not predicted. Roughly 3,100 acres were flooded that had not predicted by the 500 year flood
map and an additional 400 acres not predicted by the 100 year flood map.>>Melinda Laituri: By providing this layer
that is comprehensive across the entire region I think that that’s a really valuable perspective
of the flood. Granted its not going to replace the data that the cities and counties need
but its going to be a step towards understanding this event at a regional scale which more
and more we need to do this with these extreme events. [Credits Play]

Leave a Reply

Your email address will not be published. Required fields are marked *