Talking points for MonitoringGulf Moisture Return with GOES Imagery

  1. Title
  2. Objectives
  3. Response curves for GOES 3.9, 6.7 and 10.7 um imagery. The 6.7 um (water vapor) channel peaks around 400-600 mb, too far above the surface to adequately detect low-level moisture. The 3.9 and 10.7 um channels peak at low-levels, allowing them to detect low-level moisture. Note, there is a slight difference between the 10.7 um and 3.9 um weighting functions but they are similar.
  4. 10.7 um imagery at night, clear skies
  5. GOES-6 11.2 um imagery from 00:00 – 10:00 UTC 13 May 1985. Although the IR channel at this time was 11.2 um, it is similar to the 10.7 um channel on current GOES satellites. The loop shows a warmer (darker in this color table) air mass advecting northwestward across Texas. Note the surface observations show this warmer air mass to be associated with higher dewpoints than the cooler (lighter colored) air mass to the west where dewpoints are much lower. Keep in mind, this is at night, therefore the dry air mass will cool more quickly than the moist air mass. At night, the dry air mass will appear cooler (lighter) and the moist air mass will appear warmer (darker).
  6. Soundings taken from Midland, TX (KMAF on previous slide) taken at 06:00 UTC (frame 1) and 12:00 UTC (frame 2), 13 May 1985. The 06:00 UTC sounding is clearly still in the dry air – note in the 10.7 imagery at this time that KMAF is in the cooler (lighter) region until just after the sounding release time. At 12:00 UTC KMAF is in the moist air mass (darker region) and the moisture depth is just under 100 mb. Convection developed around 09:00 UTC along the leading edge of the moisture return, so that by 12:00 UTC the sounding may at least partially be showing outflow from that convection. The satellite imagery with surface observations show the location of the low-level moist air mass, while the soundings show the depth of the moisture.
  7. 10.7 um imagery at night, cloudy skies.
  8. 10.7 um imagery at night, 06:45 – 11:01 UTC 22 March 2005. Low-level moisture return is very difficult to see in the 10.7 um loop (insufficient thermal contrast). Overlay the surface observations to show the moisture returning in southern Alabama. The key is to use the fog product which will show the low-level clouds associated with the moisture return (shown in next slide).
  9. Fog product at night, 06:45 – 11:01 UTC 22 March 2005. Now, the low-level moisture return associated with the low-level clouds is easy to see. The key is to verify that the low-level clouds are associated with the warm sector, the METARs show this is true.
  10. Daytime tracking of low-level moisture. The visible imagery with its higher resolution should be the primary choice (especially with low-level clouds that identify the moist air mass), however the 10.7 imagery will show the moist air mass if the skies are clear / mostly clear and there is sufficient thermal contrast.
  11. 22:15 UTC 22 May 2004 visible (frame 1) and 10.7 um (frame 2) images. The visible image shows a dryline from the eastern Texas panhandle extending northward into Kansas. Note there are clear skies on either side of the dryline. The 10.7 um image shows the moist air mass east of the dryline to be cooler (lighter in this linear color table) and the dry air mass west of the dryline to be warmer (darker in this color table). Note that this signature is opposite of what we see at night. During the daytime, the dry air mass will warm up faster than the moist air mass and make this signature detectable if there is sufficient thermal contrast.
  12. Visible imagery with METARs from 16:45 UTC 21 April – 00:15 UTC 22 April 2001. In this case, the low-level moist air mass is characterized by cloud streets that are parallel to the low-level flow. The northwestward moving warm front is easy to observe across Kansas. The visible imagery combined with the METARs give you the precise location of the low-level moist air mass. In this case, a tornadic supercell erupted at the western end of the warm front (at the intersection with the dryline).
  13. 10.7 um imagery from 16:45 UTC 21 April – 02:15 UTC 22 April 2001. The low-level clouds in the warm sector are still detectable in this loop, but not as easy to discern as the higher resolution visible imagery.
  14. 23:31 UTC 21 April 2001 visible image with MSLP, METARs and Eta 12 hr forecast surface dewpoint overlay options. The optimum way to view the low-level moisture air mass is to overlay the METARs with the imgery, but one can also overlay the forecast dewpoint field to verify a model forecast
  1. Visible loop of moisture return – Morning of May 6th 2003 between 11 and 17 UTC (early AM to Noon). Important cities: watch Ponca City, OK and Wichita, KS for increasing dew points through the morning as LL moisture moves north into E Cntrl KS. Ongoing elevated storms moving towards ern KS. Note interaction of LL moisture boundary with ern most storm just to the west of Emporia KS…with subsequent tornado produced at around 1548 UTC, one mile N of Emporia.
  1. Visible image also from May 6th 2003…taken about 40 min after end of the last loop (in 15 above…1740 UTC). Both METARS and 12Z initialized ETA dew point contours (isodrosotherms) 6 hr forecast for 18Z overlayed. Toggle/Fade in hand drawn 65 dew pt contour…show difference between where actual moisture is and how it’s displaced from the model forecast (east/west/north south). Model trend is pretty good, but differences in specific placement of the moisture, especially for the purpose of convective initiation and or intensification, is very important particularly in the NOWCAST and in making upcoming warning decisions.
  1. 10.7 um IR loop that runs for 24 hrs (late spring case form May 22 2004 running from around 03Z on the 22nd to 03Z on the 23rd). Hide the time when you come into this page. Let it loop a couple of times. Then stop loop and move through frames to some of the more important times/frames that illustrate some of the problems and principles introduced through the previous portions of the session. Ask a few questions regarding the time of day and what would be expected to see visually in the IR…and where the moist/dry, warm/cool boundaries are. Frame 44 – early AM on May 22nd…good example of insufficient thermal contrast…why? Note how dryline ID washes out at this time. Side note – lighter color area over CNTRLTX where there is enough thermal contrast … with hill country showing as cool (lighter color) region. Frame 81 (1910 UTC)…early afternoon…find dryline. What other LL moisture cloud features can be identified? Frame 98 (2140 UTC)…late afternoon…again find dryline and note differences on either side. Frame 118 (evening )….what is going on here? (another good example of insufficient thermal contrast)…why? What other LL boundaries can be seen clearly? (look north to the outflow from big storms in NE and IA). Frame 133 (last - 0245 UTC on the 23rd)…note and describe time of day as well as the appearance of the areas on both sides of the dryline…why does it look different from the afternoon images?
  1. Future Satellite Products. Particularly stress higher resolution (especially vertically)coming with GOES R in 6 to 7 years.
  1. Conclusions. Self explanatory.
  1. Conclusions cont. Self explanatory. Stress importance of using Satellite together with SFC Obs, Uppr Air soundings, and model data to locate returning moisture boundaries…in order to better nowcast and help in the decision making process leading to possible warning scenarios.