Grass fires in Kansas, Oklahoma and Texas

March 6th, 2017 |

Widespread large grass fires began to burn across parts of northwestern Oklahoma, southwestern Kansas, and the Texas Panhandle on 06 March 2017. The fires grew very quickly during the late morning and early afternoon hours, due to strong southwesterly winds (with gusts as high as 67 mph in Oklahoma) behind a dryline (surface analyses); a cold front then moved southward across the region during the late afternoon and evening hours, bringing strong northerly/northwesterly winds.

GOES-16 (left) and GOES-13 (right) 3.9 µm Shortwave Infrared images [click to play MP4 animation]

GOES-16 (left) and GOES-13 (right) Shortwave Infrared (3.9 µm) images [click to play MP4 animation]

*  GOES-16 data posted on this page are preliminary, non-operational and are undergoing testing *

In the 2-panel comparison shown above (also available as a 204 Mbyte animated GIF), Shortwave Infrared (3.9 µm) images — 1-minute interval (Mesoscale Sector) 2-km resolution GOES-16 vs. 5-7 minute interval (Rapid Scan Operations) 4-km resolution GOES-13 — fire “hot spots” (dark black to yellow to red pixels) from the large Starbuck Fire can be seen making a very fast northeastward run from the eastern Oklahoma Panhandle into southwestern Kansas, behind the dryline; later, after the passage of the cold front, the leading edge of that fire and another large Kansas fire turned southward and moved back into Oklahoma. Another large fire in the Texas Panhandle (the Perryton Fire) moved rapidly eastward and crossed the border into Oklahoma (moving a distance of about 45 miles), before also turning abruptly southward in the wake of the aforementioned cold frontal passage. A total of 7 deaths resulted from these fires (CNN).

===== 07 March Update =====

On the following day, the large size of the grass fire burn scars could be seen in comparisons of true-color and false-color Red/Green/Blue (RGB) images from Terra MODIS (1732 UTC), Suomi NPP VIIRS (1857 UTC) and Aqua MODIS (1912 UTC) images viewed using RealEarth (below). The burn scars appeared as dark areas in the true-color images, and shades of tan to darker brown in the false-color images.

Terra MODIS true-color and false-color images [click to enlarge]

Terra MODIS true-color and false-color images at 1732 UTC [click to enlarge]

Suomi NPP VIIRS true-color and false-color images [click to enlarge]

Suomi NPP VIIRS true-color and false-color images at 1857 UTC [click to enlarge]

Aqua MODIS true-color and false-color images [click to enlarge]

Aqua MODIS true-color and false-color images at 1912 UTC [click to enlarge]

The creation of true-color and false-color images such as these will be possible using the ABI spectral bands available on GOES-16 and the GOES-R series of satellites. A separate blog post highlighting other multi-spectral GOES-16 views of these fire burn scars on 07 March  is available here.

Fog Detection using GOES-16 Channel Differences

March 6th, 2017 |

GOES-R IFR Probability Fields at 1230 UTC on 6 March 2017 (Click to enlarge)

Note: GOES-16 data shown on this page are preliminary, non-operational data and are undergoing on-orbit testing.

Here is what this blog post will show: It is vital to tweak the supplied default AWIPS Enhancements so that important atmospheric information is better highlighted.

GOES-R IFR Probability fields (Click here for a website that shows many examples), shown above, use present GOES Data and Rapid Refresh Data to forecast the probability that IFR conditions exist. (There are also Low IFR Probability fields and Marginal VFR Probability fields as well, data from this site). The inclusion of surface information via the Rapid Refresh Model output (that details low-level saturation) is vital to screen out false fog detection (regions where mid-level stratus does not extend to the surface) and to highlight IFR conditions that exist under cirrus that block the satellite detection of low clouds.

GOES-16 data in AWIPS includes pre-defined channel differences judged to have utility in Decision Support Services. One of these is Fog detection (the infrared Brightness Temperature Difference between 3.9 µm and 11.2 µm) that extracts information at night based on emissivity differences from water-based clouds at those two wavelengths. This is a product that can detect stratus clouds at night, if cirrus clouds do not block the satellite’s view. If those stratus clouds extend to the surface, then fog is a result. A GOES-16 Channel Difference field, shown below with the default AWIPS enhancement, contains information about the fog/low clouds that are present over North Dakota, and over Texas (click here for a graphic from the Aviation Weather Center that highlights regions of IFR conditions — Dense Fog Advisories were issued on 6 March over North Dakota).

The Fog signal in the Brightness Temperature Difference field at night occurs when the value is negative; the default color enhancement, below, contains a lot of color gradations that grab the eye in regions where the Brightness Temperature Difference is positive; for Fog Detection, those extra colors in regions of positive difference are needless visual clutter.

Brightness Temperature Difference fields (3.9 µm – 11.2 µm) over the United States, 1227 UTC on 6 March 2017 (Click to enlarge)

To get useful information from this field, alter the Brightness Temperature Difference enhancement to highlight negative values. That has been done in the toggle below with the IFR Probability field. Fog regions over North Dakota and Texas are apparent.  (Note that the scale for the Brightness Temperature Difference field here has also been flipped — click here to toggle between the two Brightness Temperature Difference field enhancements).

GOES-R IFR Probability fields and GOES-16 Brightness Temperature Difference fields, ~1230 UTC on 6 March 2017 (Click to enlarge)

GOES-R IFR Probability fields and GOES-16 Brightness Temperature Difference fields, ~1230 UTC on 6 March 2017 (Click to enlarge)

It is vital to tweak the supplied default AWIPS Enhancements so that important atmospheric information is better highlighted.