Skip to Navigation Skip to content

Regional and Mesoscale Meteorology Branch

Search the RAMMB website

A Year in a Week – VIIRS Captures Colorado Flooding

Posted On: October 2, 2013 - By: Curtis Seaman


A year’s worth of precipitation fell on parts of Colorado in one week’s time (9 September to 17 September 2013). As Colorado State Climatologist Nolan Doesken said, “Whenever you get your annual precip in a few days time, you’re in trouble.” So it is that this blog returns to flooding once again. Flooding that hit real close to home.

If you have an hour and a half available, you might want to watch this video with preliminary results and discussion about what happened given by scientists from the Colorado State University (CSU) Department of Atmospheric Science and CIRA (including Nolan Doesken and fellow JPSS Imagery Team member Dan Lindsey). If you don’t have an hour and a half, here’s an article with a good background on the events as they happened in Boulder (although if you’re a slow reader, it may not save you much time since it’s pretty comprehensive). A less comprehensive, 4-page summary of the event was put together by the University of Colorado-Boulder, the Colorado Climate Center (at CSU) and NOAA’s Earth System Research Laboratory (ESRL) which may be found here (PDF document).

The Colorado Climate Center and the Department of Atmospheric Science at CSU have put together this website to document the flood event. If you haven’t seen enough pictures of the flooding on the news or elsewhere on the internet, these two pages here and here give a good idea of the damage that resulted. By the end of September, 8 people were confirmed dead in Colorado as a result of the flooding.

Just to make sure that all of you have seen this, here are the precipitation totals (in inches) from various National Weather Service (NWS) Cooperative Observers, trained weather spotters, automated rain guages and CoCoRaHS members for the 7-day period ending on the 16 September 2013, put together by the Denver/Boulder NWS Forecast Office:

Preliminary rainfall totals over Northern Colorado, 9-16 September 2013

Preliminary rainfall totals (in inches) over Northern Colorado, 9-16 September 2013. Image courtesy NWS.

Remember to multiply those numbers by 25.4 if you’re used to using millimeters as the standard measure of rain. Also, keep in mind that this part of the world averages somewhere between 12 and 20 inches of precipitation per year.

From a satellite perspective, there really isn’t much (that isn’t classified) that can beat Digital Globe, a private company that specializes in high-resolution satellite imagery. Here’s what you can see with 0.5 m resolution. (Oh, how meteorologists would love to have data and forecast models on that kind of resolution – even if we’d all be drowning in yottabytes of data!)

In contrast, the high resolution imagery channels on VIIRS have ~350 m resolution, which is not enough to see each individual puddle, but it is enough to capture the flooding that occurred on the South Platte River subsequent to the 5-18 inches of rain that fell along the Front Range mountains.

Here’s what the “Natural Color” RGB composite of channels I-01 (0.64 µm, blue), I-02 (0.87 µm, green) and I-03 (1.61 µm, red) looked like before the flooding occurred:

VIIRS false-color RGB composite of channels I-01, I-02 and I-03, taken 19:49 UTC 7 September 2013

VIIRS false-color RGB composite of channels I-01, I-02 and I-03, taken 19:49 UTC 7 September 2013

Click on the image, then on the “1172×866” link below the banner to see the full resolution version. Note that you can’t actually see the South Platte River before the flooding occurred, but you can see the dark olive color of the river valley (caused by the mixture of trees, other ground vegetation and rich soils along the river) and the swath of light green irrigated farmland on either side of the river.

The week that the flooding occurred, it was very cloudy (duh!), so VIIRS wasn’t able to see much. But, on the 14th (which people around here refer to as “that Saturday” because each day that week brought specific memories to those that lived through it) the clouds briefly broke enough for VIIRS to see that the South Platte River valley had begun to flood:

VIIRS false-color RGB composite of channels I-01, I-02 and I-03, taken 19:17 UTC 14 September 2013

VIIRS false-color RGB composite of channels I-01, I-02 and I-03, taken 19:17 UTC 14 September 2013. The yellow arrow indicates the furthest east extent of flooding along the South Platte River.

Look for the dark, bluish-greenish color (scientific term) extending as far east as the yellow arrow. That arrow is pointing to the leading edge of the flood water, which was near the town of Weldona at this time. Places upriver from there all the way to the north side of Denver were experiencing significant (even record breaking) flooding.

Three days later (17 September 2013, about one week after the flooding began) it was a really clear day over Colorado, which made it easy to see that the flooding made it past Fort Morgan and Sterling out to little Sedgwick:

VIIRS false-color RGB composite of channels I-01, I-02 and I-03, taken 20:01 UTC 17 September 2013

VIIRS false-color RGB composite of channels I-01, I-02 and I-03, taken 20:01 UTC 17 September 2013. The yellow arrow indicates the furthest east extent of flooding along the South Platte River at this time.

Two weeks after the flood began, flood waters made the South Platte River visible all the way to (and past) North Platte, Nebraska, another site of record flooding roughly 250 miles away from where the heavy rains occurred!

VIIRS false-color RGB composite of channels I-01, I-02 and I-03, taken 19:30 UTC 24 September 2013

VIIRS false-color RGB composite of channels I-01, I-02 and I-03, taken 19:30 UTC 24 September 2013. Note that the South Platte River is visible from Denver, CO to North Platte, NE as a result of the flooding.

Here’s a short animation of this sequence of images:

Animation of VIIRS false-color composites of channels I-01, I-02 and I-03 from 7-24 September 2013

Animation of VIIRS false-color composites of channels I-01, I-02 and I-03 from 7-24 September 2013

You have to click on the image, then on the “1172×866” link to see the images loop.

It should also be said that this event didn’t just affect Colorado. Parts of New Mexico reported over 12 inches of rain and at least 1 death. Cheyenne, Wyoming just recorded the second wettest month on record (dating back to the late 1800s). And, as mentioned above, the flooding made it down the Platte River all the way to central Nebraska. And, as a piece of good news, this flood water is being used to refill the Ogallala Aquifer, which has been low due to long-term, drier-than-normal conditions.

Events like this generally bring more questions than answers: Was it a “100-year flood” or a “1000-year flood”? Could the forecasts have been better? If the forecasts were better, would anyone have believed them? How do you prepare for unprecedented events?


Leave a Reply

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