by  Brian Motta, Kevin Schrab, Tony Mostek
with contributions from Dan Bikos, Scott Bachmeier, and Jim Ladue.

Introduction

Goal: To demonstrate the use of integrated displays in AWIPS at different scales.

This lab utilizes an FSL-archived AWIPS data set from 02 September 1998. This day contained many interesting weather systems, forecast challenges, and weather warnings. Most groups will be simulating WFO operations as the Oxnard/Los Angeles office with AWIPS.

In keeping with the "forecast funnel" concept, large-scale data products will be displayed to provide an overview and demonstrate how they can be used with the other AWIPS data sets. The approach will be to move to smaller scales and eventually toward warning operations for a brief time. One group will be selected as the Storm Prediction Center. Some "real-life" surprises may occur which require fast and resourceful decisions to be made. Most of these were experienced in real-time by WFO-LOX (Oxnard, CA).

This lab will be synchronized by section in order make sure that we progress through the material in a timely fashion. ===============================================================

Section 1: "The Big Picture"(35 mins)

A. Under "Options", please set the time/date to 15:30 UTC on 09/02/1998.

Load the largest scale (N. Hemispheric) IR Window imagery (3.9um, IR, WV, and 12um) available from the "every image" part of the satellite product menu.

1) What is the best satellite enhancement to use at this scale for each channel ?

Look for one that shows the features at the warm and cold ends of the temperature spectrum.  Pay close attention to the appearance of cloud systems in the tropical, oceanic, northern, and convective regions.

2) What other alternate but appropriate color curves would you suggest to your fellow meteorologists in your office/lab group ?     Why ?

Now load the largest scale (N. Hemispheric) WV data from the top satellite menu.

3) At what time frequency are these GOES-East and GOES-West "composites" available ?
    Why ?

4) What is the difference between the top part of the satellite menu and the "every image" part of the satellite menu ?

• Notice the difference in temporal frequency of the imagery as you change from the N. Hemispheric scale to the North American scale.

• Now try overlaying the 400 mb winds from the profilers on a North American water vapor image (menu:Upper Air, Profiler, Plots).While it would be outstanding to have the GOES-generated winds sets available, do your best in determining if the profiler winds are good using the GOES imagery and other data sets. Try adding levels below and above 400mb to the display.

5) Where do the flow patterns match the profiler wind observations ? (you may want to  zoom to the central US) List sites which appear to agree well and those that do not.

What other levels or analyses would you use with water vapor imagery ?
 
 

6) How can you tell what the quality of the profiler-generated winds are ?
 

7) What are the major weather systems on the NH scale at 12 UTC ?
 

Stop to answer previous questions (Brief Q&A)

B. At the N. American scale, use volume browser to analyze tropical systems by overlaying AVN, ETA, NGM, and RUC model data to compare with the satellite for the following storms:

Danielle (Atlantic) - groups 1 & 2
Earl (Gulf) - groups 3, 4, & 5
Isis (Pacific) - groups 6 & 7
The SPC perspective provided by group 8
LOX perspective provided by group 9

What are the best combinations of satellite and model fields to look at ?
 

Which model run shows the largest difference when compared to reality (ie. satellite data) ?

Note a satellite image type/product and an appropriate color table for your weather system of interest at this scale/resolution.

Other than the tropical cyclones, what other areas are you concerned about ?
 

Review large scale analyses with groups.
Discuss SPC Guidance for the rest of the afternoon (group 8) and LOX outlook (group 9).

Section 2: "Morning Nowcast Update" (20 mins)
 

• Now, be sure your display has three windows by clicking on the view menu and the three-pane-layout choice. Still using the same timeframe as above, we will now change scales. Load the CONUS satellite imagery for the IR, water vapor, and visible.
• How does the AWIPS CONUS scale satellite imagery differ from the larger-scales ?
• Why are there differences in the appearance of the visible imagery ?
• Do you see these kinds of differences in the IR data as well ?
• What can be said about the convection in southern California at this time ?

• Use the Regional scale to load the visible data.
• How does this imagery compare with what you just saw at the CONUS scale ?
• Using the standard IR and water vapor imagery at the regional scale,
• what features are apparent in the water vapor imagery over southern California ?
• Now, select the enhancement called "GOES WV (CIRA)".
• What can be said about the upper-air pattern over southern California and the current weather ?
• List several features that you can see in the visible imagery clearly in the regional domain (that were not visible at the larger CONUS scale). You may need to change the color table to low-light visible.
• Are there any features in the imagery that would raise concern for the LOX nowcast or short-term forecast ?

SHORT BREAK

Section 3: "Local Scale Displays"(30 mins)

A. Clear all three panels. Change the time to 1998 09/02/1998 at 20:30 UTC.

One of the valuable additions to the databases in AWIPS is the availability of high resolution topographic data. Load the topo data at the STATE scale. Notice the resolution is lower at scales above the state scale (even if you zoom in). Discuss some uses of terrain data within your group.

For instance, select the "high range enhanced" color table. The Salt Lake City, UT (SLC) WFO has successfully utilized the terrain data in combination with the radar data and the basin outline overlays to help them issue flash flood warnings for Zion National Park. Try this now by doing the following:

- under options menu, select image combinations (or use icon)
- select the reflectivity from Radar/Nowrad  or build your own radar composite from single-site imagery. Which is better ?   Why ?
- now select the drainage basins from the maps menu
- select image properties and select color table (OSF16 level) for radar and CIRA (IR default) for the topography
- Add County Names

Which county has the most significant radar echoes over the highest terrain? (Use the image fade and sampling functions to help you)

B. Clear all panels. Change time to 0000 UTC on  09/ 03/ 1998
 

Compare the cloud-top-temperatures to the 17 UTC KEDW equilibrium temperature. Use the interactive skewt to get the equilibrium temperature (EL). You can visualize the EL on an IR loop by highlighting the brightness temperatures for a degree on either side of the EL using the enhancement editor. Use an IR loop on a state or WFO scale to make these comparisons because we'll need that scale for radar comparisons next. At 2130 UTC, is the cell near KEDW utilizing the full CAPE available to it assuming the KEDW sounding is still representative? How about the storm near the KSOX radar for the same time?

Let's Build a 4-panel of GOES-IR4, VIS, low-level reflectivity from KSOX, and VIL from KSOX. Do this by right clicking on the panel with the IR loop and select 4 panel display. The IR loop should redisplay on the 1st panel with enhancements retained. Next, display the visible loop in panel 2, the KSOX 0.5° reflectivity in panel 3 and the KSOX VIL in panel 4.

Pay special attention to the IR cloud-top temperatures and anvil expansions. Do these satellite indications of intense storms match the radar observations ?   Focus on the time between 2100 and 2130 UTC for the storm nearest to KSOX.  Compare the trends in the KSOX VIL to that of the storm top brightness temps. Which trend would you believe? Would satellite imagery influence your warning decision process?

Sporadic reports of flooding start coming in at 22 UTC as thunderstorms continue to develop. Evaluate the situation as best you can with the satellite, radar, lightning, and model data that is available. You might try loading the terrain data at the WFO scale and combining the image with the visible loop. Select the the image properties window. To better show the terrain/visible imagery, try sliding the fader bar between the two image enhancement selectors.

    1) Is there an enhancement problem with the visible data?

    2) Is there any convection being enhanced by "upslope flow" or is some other process going on?

Now, load the KVTX precipitation products. The first data on precipitation is now being received…integrate the radar products with the satellite data for an assessment of the flash flood potential.

Two of the DoD WSR-88Ds in the CWA have been struck by lightning and taken off line. The remaining NWS WSR-88D is now down too.

    3) What do you do ?  Cause of outage and expected time of restoration is unknown.  The SOO
        looks on the "bright side" of the situation and says "we still have lightning data and
        satellite imagery!"

    4) Do you expect flooding ?

    5) Someone asks "What about GOES Rapid Scan Operations (RSO)?"

A report of a tornado is now heard on the NAWAS communications network. While a specific location was not given, there were a few distinct storms which are obvious candidates. Examine the available radar and satellite examination. Try to identify the storm.

    6) A tornado warning is immediately issued...do you have enough information to determine what county/area?

Technicians call back and say expect the DoD radars up soon. The radars come up but the narrowband links to the PUP and AWIPS are down- problem unknown. No radar data still. 88D radar data is missing from 22:30 through 00:06 UTC. Fortunately, we have NOWRAD data through the period. However, remember that the three closest radars that actually cover the area are down. Use this mosaiced data with caution. Flash flooding now reported in LA.

Section 4 - Using GOES RSO and Radar (30 mins)

Clear all panels. Change time to 2:00 UTC on  09/ 03/ 1998.
Use either the State or WFO scale!
The satellite focal point initiated a RSO call. The RSO data arrived at LOX an hour later on their RAMSDIS. Let’s examine the RSO data now. RSO began after the 00 UTC full-disk scan. The images start at 00:41 UTC for VIS and 00:30 for IR. Radar data are now available again. Identify the most active areas of convection using both the satellite and radar data (as of 2:00 UTC).
Display the radar echo tops product as an image combination.

• Are there any threats left?  Examine Storm Total Precipitation mosaic and coastal areas.
• What can you say about the data near the radar's "cone of silence" ?
• How useful is the NOWRAD data in this vicinity?
• Which satellite imagery provides additional information about the convection over the radar site?

• What other map overlay would be particularly useful if monitoring an area for flash flooding?

You may want to explore image combinations and the differences in single-site radar imagery versus the NOWRAD composite imagery that is available for this case. By combining the imagery, you can clearly see which reflectivity areas are within the LOX radar vicinity.

Display a visible satellite loop at the WFO scale. Now, overlay the radar-generated storm tracks.

• How well do they agree ?
• What are the differences in these depictions ?
• Discuss the pros and cons of using these data sets together.

Epilogue (15 mins)

As you may have noticed, despite the severe weather, there was very little in the way of lightning data displayed. For this case, there appears to have been a problem with archiving the lightning data. The real-time display of lightning data was quite plentiful on the displays at LOX when the event was unfolding.

The media was quite active in covering the public-reported tornadoes, flash floods, wildfires, and resulting impact in LA and vicinity. Links to CBS2 NEWS summarize the events of the day.
Sep 02, 1998
Sep 02, 1998
Severe Reports for the 09/02/98 Scenario

• What would you do differently in your office based on this case ?