Close Window

Saint Louis Univ.: "Research on Midwestern severe convective storms"

Final Report

UNIVERSITY RESEARCHERS PREPARING REPORT: James T. Moore, Yeong-jer Lin. Graduate Students: Scott Rochette, Patrick Market, Sean Nolan and Stephen Considine

NWS RESEARCHERS PREPARING REPORT: Ted Funk, Ron Przybylinski, Fred Glass, Dan Ferry, Gary Schmocker, Kevin Darmofal, Don Kirkpatrick, Van DeWald, Mike Shields, Todd Shea, and Greg Lewis


There are two components to this COMET cooperative project. The first component deals with understanding the development of mesoscale convective systems (MCSs) in the mid-Mississippi Valley resulting in heavy precipitation. Research in this part of the project involved collecting numerous case studies of heavy convective precipitation, defined as rainfall greater than or equal to 4 inches within a 24-hour period. The case studies utilized conventional surface and upper-air data, WSR-88D radar data, profiler data and satellite imagery to develop conceptual models of the development and evolution of MCSs that result in heavy rain and possible flash flooding. The major goals noted in the original proposal included:

1. Beginning the development of the climatology of MCSs associated with heavy rainfall in the Mid-Mississippi River Valley. The climatology would include the major synoptic scale forcing mechanisms noted during the heavy rain/flash flood episodes, including the role of the low-level jet (LLJ), vertical wind shear, precipitable water and instability.

2. Using WSR-88D data to study the propagation patterns associated with these heavy rain episodes. This includes categorizing the characteristics of the MCSs and attempting to distinguish between synoptic environments that favor backward and forward propagation. The radar imagery would be compared to the satellite imagery to see the advantages and disadvantages of both remote sensors in studying MCSs.

3. Testing some of the concepts of MCS behavior advanced by Weisman (1992) in terms of how low-level vertical wind shear interacts with the thunderstorm-generated cold pool to affect precipitation type distribution (convective vs. stratiform) within the MCS and longevity of the MCS.

4. Comparing model generated precipitation fields to observed precipitation to estimate the effectiveness of large-scale models (NGM and Eta) in heavy rain situations.

Major Accomplishments of the MCS-Heavy Convective Rainfall Team

1. We installed the SLUBREW software on the NWS St. Louis HP workstation. This software enables the NWS participants to use our datasets to construct surface, upper-air and isentropic charts as well as calculate numerous stability indices and kinematic parameters useful for both parts of the COMET work.

2. WATADS software from NSSL is now on a SLU SPARC workstation, enabling the SLU participants to look at WSR-88D data for the heavy rain or severe storm part of the project.

3. We presented numerous papers at professional conferences, and have written numerous preprints and 1-2 journal articles (see section 6).

4. We developed software to estimate propagation vectors, compute frontogenetical vectors, objectively analyze Missouri cooperative rainfall data, compute parameters related to precipitation efficiency, and compute moisture transport vectors on pressure and isentropic surfaces.

5. We developed a conceptual model for cool sector (elevated) thunderstorms associated with an east-west quasi-stationary boundary and a low-level jet (LLJ) acting normal to the boundary. This model, developed by Fred Glass and Dan Ferry, is described in a preprint paper and is being worked on for journal submission. The Glass et al. (1995) conceptual model is appended to this report (see Figs. A and B).

6. We have established guidelines for discriminating the pre-convective synoptic environments favorable for forward versus backward propagation based on the numerous case studies. These guidelines have been shown at numerous QPF workshops, COMET residence courses and at our COMET workshops in St. Louis.

7. We wrote software to compute composites of surface and upper-air parameters associated with cool sector and warm sector thunderstorms associated with heavy convective rainfall. The list of cases used for these composites is expanding and will also appear in the journal article noted in (5).

8. We wrote software to compute CAPE using virtual temperatures instead of the dry-bulb temperatures from a sounding. Also, we developed an 'elevated CAPE" based on the maximum equivalent potential temperature in the lowest 300 mb of the sounding, as described in an article by Rasmussen and Doswell (1994) in Weather and Forecasting. We can also objectively analyze these CAPE values over a U.S. grid. A conceptual diagram illustrating the difference between the standard CAPE calculation and the elevated or "max qe" CAPE is appended to this report (see Fig. C).

9. We collected numerous case studies over the three years of the grant to use in the composite study, and provided examples of heavy rain cases for seminars and workshops for educational use and for training at local workshops and at COMET courses.
These case studies are also being used in the current COMET cooperative to further investigate the relationship of low-level wind shear to the development of the MCS.

10. In the first year of the project we did comparisons between the NGM QPFs and actual cooperative network reports (see section 6). During the last year and one-half of the project research on this aspect of the QPF problem was delayed in lieu of doing compositing work. In the new COMET cooperative project more work will be done in this area.

11. We created a web home page of COMET activities within the St. Louis University, Department of Earth and Atmospheric Sciences home page. The web page is available at "" (the department home page) under the box labeled COMET. Our preprint and journal articles are listed there, in addition to descriptions of various conceptual models useful to forecasters.

12. We have begun a study of precipitation efficiency and are currently testing two potential indices to estimate the precipitation efficiency of the pre-storm environment. This testing will continue into the new COMET cooperative period.

13. Fred Glass, Jim Moore and Scott Rochette have completed their first review of the Flash Flood chapter written by Mr. Robert Davis of NWS Pittsburgh for the Severe Local Storms Monograph, which is currently being edited. The monograph is being written under the auspices of the American Meteorological Society.

The second component of this cooperative research effort with the National Weather Service centered on MCSs associated with straight-line wind damage, notably bow echoes/derechoes. The "bow echo" team set the following goals:

1. Diagnose and document the pre-storm sounding characteristics (thermodynamic and dynamic) of those MCSs (derecho type) in cold and warm seasons resulting in strong winds, hail and tornadoes.

2. Document the evolution of the storm reflectivity structures of bowing convective storms using WSR-88D volumetric reflectivity and velocity data.

3. Test the numerical simulations of bow echoes discussed by Weisman (1992,1993). They especially wanted to examine the weak-moderate shear and strong shear environments which created bow echoes to compare and contrast the synoptic characteristics and resultant severe weather.

4. Use vertical motions derived from single Doppler radar data (assuming a 2-D bow echo which is normal to the radar beam) to test its effectiveness as a short-term forecasting tool in forecasts of severe straight-line winds, hail and tornado production.

Louisville, KY Office

The following squall line and bow echo events have been researched at Louisville as part of the COMET project (a complete version of Ted Funk's report is appended to this report):

1. 15 April 1994: Detailed research of this event across Kentucky and southern Indiana is near completion. The squall line produced widespread wind damage, numerous mesocyclones, and several tornadoes associated with 7 bowing line segments. We evaluated the pre-storm and "near-storm" environments to assess stability, vertical wind shear and forcing fields. KLVX WSR-88D data also were analyzed countless times to study the detailed evolution of rear inflow jets (RIJs); multiple circulations/mesocyclones, some of which produced tornadoes; mesoscale airflow structure; bookend vortices aloft; mid-altitude radial convergence (MARC) signatures; and embedded high precipitation (HP) supercell structure. Numerous time-height cross-sections were constructed to study the typical evolution of bow echo mesocyclones and their relationship to damaging straight-line winds and tornadoes. Based on this event, NWS Louisville wrote a conference paper that Ted Funk presented at the 18th Conference on Severe Local Storms in San Francisco, CA in February 1996 (see section 6). We then greatly expanded the conference paper and composed a detailed journal article for Weather and Forecasting based on this case. The manuscript will be submitted to the journal for formal peer review in Spring 1997.

2. 14 May 1995: This event over Kentucky is currently being researched. We have assessed the structure of the line and calculated rotational velocities associated with a tornadic bow echo mesocyclone. This case will result in a future publication as a conference paper and/or journal article.

3. 18 May 1995: This case over south central Kentucky produced multiple mesocyclones, several tornadoes, and much wind damage. Mesocyclone tracks and the evolution of the squall line have been analyzed. Van DeWald and Ted Funk developed a detailed workshop from this event for use at the COMET Cooperative Project Workshop at SLU in November 1996. Our workshop stressed the identification of severe weather signatures and their evolution, radar sampling issues with respect to mesocyclone evolution, and application of this knowledge to the severe weather warning process. This event will result in a future publication as a conference paper and/or journal article.

4. 20 April 1996: NWS Louisville also has begun investigating this case over Kentucky and southern Indiana. This line produced a few tornadoes and embedded HP supercell structures. This event also will likely appear in a conference preprint and/or journal article.

Based on the bow echo events studied at NWS Louisville, we have developed a preliminary conceptual model for bowing line segments within squall lines (see Fig. D). Key aspects of this model include:

1. Highest reflectivity values within organized bowing segments often exhibit a meso-/convective-scale low and frontal structure. The mesolow exists within a distinct or subtle comma head with a boundary extending poleward (north or east) from the low and a bulging cold (gust) front south of the low (a line echo wave pattern (LEWP)-type structure; see attached Fig. D). Straight-line wind damage occurs along the bulging cold front, which is coincident with a rear inflow jet (RIJ) in velocity data and often a weak echo channel (WEC) in reflectivity data. Cyclonic circulations and tornadoes occur immediately north of the most bulged-out portion of the cold front, near or just south of the mesolow (i.e., southeast quadrant of the comma head). Minimal wind damage usually occurs poleward of the mesolow.

2. Multiple cyclonic circulations often develop within long-lasting bowing segments, with new circulations located just south of older ones, which typically results in one or more transient tornadoes.

3. Cyclonic circulations associated with well-defined line segments/LEWPs typically develop upward well into the middle levels of storms from their low-level origins. Some of these circulations apparently meet supercell mesocyclone rotational velocity and depth criteria.

4. Rapid deepening and strengthening of mesocyclones typically occurs just prior to and during tornadogenesis.

5. Embedded HP supercell structures have been noted at times within organized bowing segments, apparently due to a convective-scale boundary interaction that produces a deep-layered mesocyclone and redistribution of the precipitation field.

In July 1996, Van DeWald placed information about COMET and collaborative research efforts among SLU, NWS St. Louis, and NWS Louisville on the NWS Louisville Internet homepage.

The following face-to-face meetings were held between project participants at NWS Louisville and those at SLU and NWS St. Louis during the 3-year COMET project to set goals, discuss interim and final research results and composed publications, discuss future efforts, etc.:

1. July 1994: NWS Louisville traveled to SLU and NWS St. Louis for a 2-day meeting.

2. August 1994: NWS St. Louis traveled to NWS Louisville for a 2-day meeting.

3. March 1995: SLU and NWS St. Louis traveled to NWS Louisville for a 2-day meeting.

4. Summer 1995: NWS Louisville traveled to NWS St. Louis for a 3-day meeting.

5. June 1996: NWS Louisville traveled to SLU and NWS St. Louis for a 2-day meeting.

A great deal of effective information exchange and science sharing also occurred among NWS Louisville, NWS St. Louis, and SLU via numerous telephone conversations, e-mail and regular mail throughout the project.

St. Louis, MO Office

During the three-year period of the study, 12 convective cases were analyzed at NWS St. Louis. Cases used in the analyses included the following:

Cool Season/Strong Dynamics-Low Instability: 15 April 1994 and 21 November 1994.

Warm Season/Weak-Moderate Dynamics-High Instability: 2 July 1992, 10 August 1992, 4 June 1993, 8 June 1993, 22 September 1993, 8 June 1995, 8 July 1995, 5 May 1996, 25 May 1996, and 27 May 1996.

Storm surveys performed by staff members were completed for 5 warm season and 2 cool season MCS events. Tables 1 and 2, appended to this report, give a summary of the environmental and storm characteristics associated with each convective case.

The first objective focused on investigating the kinematics, dynamics, and antecedent conditions (pre-convective environments) associated with linear MCSs, i.e., bowing convective line structures, which produced damaging winds.

1. Objective #1 - Pre-Convective Environmental Analyses - Findings and Accomplishments
a. Comparisons Between Johns et al. (1993) and Our Results:

Detailed analyses of the pre-convective environment for all of the cases highlighted above were completed. Composite charts were constructed for each case and compared to those completed by Bob Johns of the Storm Prediction Center (SPC), which satisfied the derecho criteria. For the late spring-summer cases, we found minor differences between our surface, upper-level and instability composites compared to Johns' results. Key features included: surface dewpoint pooling, low-level convergence, presence of an east-west frontal boundary, 850 mb warm air advection and dew point pooling were all nearly similar to the study by Johns et al. (1993). We found that the presence of an east-west frontal boundary, where convection focused and intensified, was a critical feature in all of our cases. At upper levels (500 mb and 300 mb) we found that our wind fields for the cases examined were 5-10 m s-1 (on average) stronger compared to Johns' findings.

Concerning strong dynamic-low instability environments, we tested Johns' findings for our two cases and documented similar results. Additional cases will be needed to test parameters associated with linear MCSs, which evolve in environments of this type.

b. Evaluating Weisman's Method on the Operational Utility of 0 to 3-4 km Vertical Wind Shear

We tested Weisman's method of evaluating the utility of the 0 to 3-4 km vertical wind shear as one of the important short-term forecast parameters to predict the growth of linear MCSs with bowing structures that produce damaging straight-line winds. Data from the WSR-88D vertical wind profiler (VWP) at NWS St. Louis and nearby wind profilers (Bloomfield, MO; Winchester, IL; and Conway, MO) were used in the analyses. Weisman's (1992,1993) numerical simulations showed that long-lived convective squall lines can evolve and be sustained in environments having moderate (13-19 m s-1) to strong (³20 m s-1) vertical wind shear, provided sufficient instability and other forcing mechanisms were present. We have found very good agreement between his magnitudes of the 0 to 3-4 km vertical wind shear and our results for the linear MCS cases we examined. Note: see Tables 1 and 2 for additional details.

2. Objective #2 - Storm Reflectivity and Doppler Velocity Analyses - Findings and Accomplishments

This objective focused on the storm reflectivity and Doppler velocity analyses including the evolution of tornadic and non-tornadic vortices. We focused our energies on discovering new methods and techniques for improving lead times in forecasting the initial onset of damaging winds and non-supercell tornado activity. We shared a number of ideas with the NWS Louisville team. Several sub-components were derived from this objective. These components included the following:

a. A survey of the evolution of storm reflectivity structures during the entire linear MCS's life cycle.

b. Assessing the evolution of mesoscale airflow and other velocity signatures associated with linear MCSs.

c. Analyzing and understanding vortex evolution during each phase of the MCS's (bowing structures) life cycle.

a). Reflectivity pattern evolution

A component of this part of the cooperative project allowed us to compare our recent radar analyses to previous studies completed by Houze et al. (1990), Smull et al. (1990), Johns and Hirt (1987), and Przybylinski and DeCaire (1985).

1. Early stages: In seven of our ten warm season cases results from our studies showed that convection initially took on the form of either a linear multicellular cluster or a circular multicellular cluster. In the other three warm season cases and the two cold season cases, convection was well into the mature stage when it moved into central and eastern Missouri.

2. Mature stages: Of the ten warm season MCS cases surveyed, we classified the overall reflectivity pattern into the following groups: a. 3 MCSs - symmetrical, b. 4 MCSs - asymmetric, c. 3 MCSs - serial. Of the 2 cold season MCSs, one MCS took on an asymmetrical reflectivity pattern, while the second evolved from a classic LEWP to a serial pattern, as described by Johns and Hirt (1987).

Based on the cases examined, results show that clusters' damaging winds frequently occurred from just south of the apex of the bow northward along the cyclonic shear side of the bow.

b) Mid-Altitude Radial Convergence (MARC) Velocity Signature

One of the critical objectives and goals of this component of the cooperative project was to identify reflectivity and velocity precursors within developing linear MCSs to forecast the initial onset of damaging winds and rapid growth of non-supercell tornadoes. Much of the earlier research efforts focused on detecting damaging microburst winds from isolated pulse convective storms. However, very limited research was completed on detecting the initial onset of damaging winds from organized linear MCSs.

The MARC velocity signature (derived from work completed by Eilts (1990) with pulse microburst storms) is shown as a narrow zone of enhanced radial convergence along the leading edge of the MCS. This signature was tested on 8 of the 10 warm season linear MCSs. Gary Schmocker and Ron Przybylinski constructed 15 time-height cross-sections to show the evolution of magnitudes of MARC, VIL and the initial onset of damaging winds and non-supercell tornadoes. Preliminary results show that for nearly 80% of the cases sampled, the MARC velocity signature has proven to be a strong indicator for forecasting the initial onset of damaging winds, and frequently provided a 5 to 15 minute lead time in forecasting the initial occurrence of non-supercell tornadoes. Magnitudes of the MARC velocity differential of 25-30 m s-1 or greater proved to be a reliable precursor of severe wind gusts of 25 m s-1 or greater. A preliminary model of the MARC is shown in Fig. E appended to this report. See Przybylinski et al. (1995) and Schmocker et al. (1996) in section 6 for details.

We also examined the evolution of the mesoscale rear inflow jet (RIJ). Specifically, the local velocity maxima embedded within the current were examined to see how these features possibly affected accelerations within the bow line segments. Our data sample here is limited and more research will be needed to complete this task from both NWS Louisville and St. Louis sites.

c) Vortex Evolution

Another critical objective and goal was to survey and understand the evolution of tornadic and non-tornadic vortices which form along the leading edge of rapidly moving bowing line segments and near the northern end of linear convective line segments. Very limited research was completed (Wakimoto and Wilson, 1989; Burgess and Smull, 1990; Przybylinski, 1988; Przybylinski et al. 1993) prior to the studies performed at NWS St. Louis and Louisville.

Numerous time-height cross-sections of rotational velocities (Vr) were constructed to show the intensity and trend of tornadic and non-tornadic circulations and their relationships to damaging winds and tornadoes. Ted Funk's group (NWS Louisville), Jim O'Sullivan (SLU grad student) and our group (NWS St. Louis) conducted a thorough study of numerous tornadic and non-tornadic vortices associated with the 15 April 1994 squall line. Similar results were noted at both sites. After completing this initial study, both groups, along with Dr. Lin and the graduate students, learned much about the origins, mature and dissipating stages of vortices and the time of tornado and wind damage occurrence in relation to vortex evolution.

Tornadic and non-tornadic circulations evolution was examined for 8 warm season and 2 cool season linear bowing MCS events. 26 time-height cross-sections of Vr were completed for vortices associated with 8 warm season MCSs across Missouri and Illinois. 12 time-height cross-sections of Vr were completed for vortices associated with 2 cool season MCSs across Missouri and Illinois.

Members of the damaging winds team (NWS St. Louis and NWS Louisville) are currently developing conceptual models for vortex evolution for both warm and cool seasons and comparing our findings to those of Burgess et al. (1982), concerning vortices associated with supercells. Based on results from both groups, tornadic and non-tornadic circulations initially form near the apex of bowing line segments, then deepen and intensify as they traverse along the cyclonic shear side of the bow. Tornadoes occur just north of the bulge of the line segment, while damaging winds are noted within the region of the apex of the bowing structure.

Comparisons of warm and cool season circulations were performed. Preliminary results from the dataset suggest two interesting results:

a) The origins of circulations, which evolve in the cool season form at a lower height, compared to their warm season counterparts; cool season vortices form at 1-3 km, while warm season vortices form at 3-5 km.

b) For cool season cases we found a good correlation between the rapid intensification and deepening of the mesocyclone to tornado and damaging wind occurrence. Rapid vortex deepening and intensification often preceded tornadogenesis by as much as 15 minutes.

c) For warm season cases we discovered that the first circulation formed near the northern end of the linear convective line prior to bowing. This vortex often became the primary "bookend vortex." Subsequent vortices frequently evolved from near the apex poleward along the cyclonic shear side of the bow. [Note: See Figs. F and G appended to this report for conceptual models; and publications by Przybylinski et al. (1993), Funk et al. (1996) and Przybylinski et al. (1996) for more details.]

Bookend Vortex Overview: 13 warm season bookend vortices were examined. Our survey over Missouri and Illinois showed that the bookend vortex is a low-mid level feature with the strongest shears detected between 2-6 km. Several cases revealed that the strongest wind damage was confined to the cyclonic shear side of the bow echo, south-southeast of the bookend vortex. (Note: See Tables 1 and 2 appended to this report for additional details).

During the winter and spring seasons of 1997, members of the NWS St. Louis team put forth considerable effort developing a journal article for Weather and Forecasting. This paper is an expanded version of the conference paper published in the 18th Conference on Severe Local Storms (see section 6). The NWS Louisville team is working on a companion journal article.

COMET Cooperative Workshops

We held two COMET cooperative research workshops during the period of the grant. The first workshop was held from 30 November - 1 December 1995 at Saint Louis University in the Busch Memorial Center. About 35 people attended, including students and faculty from SLU, and forecasters from Scott AFB and the NWS. We covered two central themes: 1) the heavy rainfall prediction problem, and 2) forecasting convective high wind events. The workshop involved seminars from the university participants and from the NWS research participants. Case studies of heavy rainfall and severe straight-line winds were also used during the afternoon workshops.

The second COMET cooperative research project workshop was also held at Saint Louis University on 12-14 November 1996. Daily attendance averaged 40-55 people, which included staff members from numerous NWS forecast offices, private sector meteorologists, faculty and students from the University of Kansas, Northern Illinois University and Saint Louis University, and forecasters from Scott Air Force Base. Appended to this report is a memorandum describing the workshop and an agenda for the workshop that notes the speakers and their seminar topics, as well as the workshops given on each day. As noted in the agenda all of the COMET project participants presented a seminar and/or workshop. This workshop set the standard for future workshops of its kind.


Jim Moore has presented workshops/seminars at the following NWS offices during the COMET cooperative grant period: Des Moines, IA; Pleasant Hill, MO; Wilmington, OH; Tulsa, OK; Wakefield, VA; Morehead City, NC; Wilmington, NC; Davenport, IA (with Fred Glass); Detroit, MI; Jackson, MS; Upton, NY; Louisville, KY; and St. Charles, MO. Topics included isentropic analysis, jet streak circulations, conditional symmetric instability, QPF techniques, frontogenesis, elevated thunderstorms, Q-G theory and Q vectors, and applications of PC-GRIDDS. Recently, Jim presented a three-hour seminar/workshop at Weatherdata, a private meteorological firm located in Wichita, KS.

Dr. Frank Lin presented a two-hour seminar to the forecasters of NWS Louisville entitled, "Applications of Single Doppler Radar Analysis to Local Storm Research" during the bow echo team's visit to Louisville on 13-15 March 1995. The COMET cooperative work was also discussed.

Jim Moore has also conducted numerous seminars and workshops at COMET residence courses (COMAP, Managers Course, WMO Course, Hydrometeorology Course and the Mesoscale Course) over the last three years. Notes prepared for these courses at COMET have been given to both Ted Funk (SOO; Louisville, KY) and Ron Przybylinski (SOO; St. Louis, MO) for use in training the staff at both offices. Notes include the following topics: Q-G theory and Q vectors, conditional symmetric instability and other mesoscale instabilities, frontogenesis, jet streak circulations, isentropic analysis, latent heating effects, and cyclogenesis.

SLU graduate student Scott Rochette presented a seminar on composite surface and upper-air analyses of elevated thunderstorm environments to the COMAP class on 30 April 1996.

Two SLU graduate students, Patrick Market and Sean Nolan, presented seminars at the NWS St. Louis. Titles of the presentations were "Structure of an Occluded Cyclone Determined from MAPS Initializations" and "The Role of Frontogenesis in the December 18-19, 1995 Heavy Snowstorm in Missouri," respectively.


Ron Przybylinski has created several packets of "handouts" on topics of severe weather to the University for use in undergraduate and graduate courses. These include packets on "Characteristics of Classic, HP and LP supercells," "Mesocyclone Characteristics" and "Evolution of MCS-Bowing Line Structures." These handouts, which include color WSR-88D imagery, have proven to be extremely useful in the classroom and for the training of graduate students in the project. Four sets of briefing slides and notes covering the topics of bow echo-mesocyclone evolution were developed by Ron Przybylinski, Gary Schmocker and Dr. Frank Lin as a result of the knowledge gained through the COMET cooperative. Copies of these briefing slides and notes have also been presented to COMET SOO Residence and Mesoscale Courses, and to many NWS forecast offices over the central United States. Ron has given seminars at the NWS Louisville, NWS St. Louis and SLU to acquaint the new research members, students and faculty with the COMET cooperative work concerning bow echoes and key features noted on the WSR-88D radar.

Each spring (1994, 1995, and 1996) Ron Przybylinski gave two seminars to staff members at NWS St. Louis on squall line-bow echo evolution, which evolve in cool and warm seasons, respectively. Discussions included the mesocyclones that evolve along the leading edge and cyclonic shear side of bow echoes. Much of the information presented was based on knowledge gained through the COMET cooperative and publications derived from this project.

Gary Schmocker presented a paper on the MARC velocity signature at the AMS 15th Conference on Weather Analysis and Forecasting (see section 6) and the COMET cooperative workshop in November 1996 (see appended material). Ron Przybylinski gave presentations entitled "Comparisons of Vortex Evolution in Warm Season MCSs" and "Overview of the 15 April 1994 Squall Line over Missouri and Illinois".

Development of a Central Region CBL Module focusing on the 2 July 1992 bow echo/derecho event is currently underway.

Saint Louis University graduate students Jim O'Sullivan and Merl Heinlein visited with Ron Przybylinski and Gary Schmocker about twice a month from June to December 1996. Excellent exchanges of ideas concerning storm conceptual models, vortex evolution and WSR-88D case studies were the highlights of these visits. Both the WSR-88D PUP and SOO SAC workstation were used to display WSR-88D cases. Jim O'Sullivan and Ron Przybylinski have examined 5 vortices associated with the 15 April 1994 bow echo-squall line case.

Table 3 presents a summary of presentations conducted at many NWS forecast offices by Ron Przybylinski. A number of staff members from the U.S. Air Force and local media also attended and benefited from these presentations. Topics at these presentations included:

*CAPE/shear issues related to bow echo development

*Bow echo morphology in cool and warm season environments

*Role of mesoscale airflow structures within the MCS

*Mesocyclone evolution related to tornadic and non-tornadic vortices

*Role of bookend vortices in the warm season and their relationship to the production of damaging winds

*Utility of the Mid-Altitude Radial Convergence (MARC) velocity signature for forecasting the onset of damaging winds and vortex development

The following cases were used in the presentations: 2 July 1992, 22 September 1993, 15 April 1994, 8 July 1995, 25 May 1996 and 27 May 1996.

Fred Glass has presented workshops/seminars at the following locations during the COMET cooperative grant period: Lincoln, IL (NWSO ILX), Pleasant Hill, MO (NWSO EAX/Missouri River Basin Forecast Center), Omaha, NE (First NWS Central Region QPF Workshop), Davenport, IA (Second NWS Central Region QPF Workshop), St. Charles, MO (St. Louis FAA Flight Service Station), and Weldon Spring, MO (NWSFO LSX - four different talks over the three-year period). Topics included MCS initiation/propagation, environmental conditions and conceptual models of heavy rainfall events, elevated thunderstorms, synoptic climatology of recent heavy rainfall events, MCS forecast techniques and the application of gridded datasets, and QPF techniques. During the talks/workshops at other NWS offices, Fred has freely distributed PC-GRIDDS command files used to apply MCS, heavy rainfall, and severe weather forecast techniques.

Fred Glass will be presenting a seminar on elevated thunderstorms and MCS forecasting techniques to the NWSO Paducah, KY staff in May 1997. He will also be presenting a talk on the 29 November 1991 F4 Nixa-Springfield, MO tornado to the "Diagnosis and Prediction of Severe Local Storms" class on 30 April 1997. This is a graduate-level course taught by Jim Moore.

At the Fourth National Heavy Precipitation Workshop in Scottsdale, AZ on 12-16 September 1994, Fred Glass, Dan Ferry, Jim Moore and Scott Rochette conducted two workshops on "Forecasting the Initiation and Propagation of Mesoscale Convective Systems."

At the First COMET Cooperative Project Workshop at SLU on 30 November-1 December 1995, Fred Glass presented a seminar entitled "Composite Analyses and Environmental Conditions Associated with Heavy Convective Rainfall Events Across the Mississippi Valley Region."

At the Second COMET Cooperative Project Workshop at SLU on 12-14 November 1996, Fred Glass presented a seminar entitled "Composite Analysis and Characteristics of Recent Heavy Rainfall Events."

Dr. Frank Lin's Severe Storms course (7 graduate students) visited the NWS St. Louis and attended at two hour informal presentation on the structure and evolution of the 15 April 1994 severe convective squall-line and tornadic and non-tornadic vortices. This session was held on 3 May 1996. On 18 November 1994 a similar exchange had taken place with the students in Dr. Lin's course during that term.

In 1995, Ted Funk developed a set of notes for forecaster-training purposes called "Bow Echoes and Squall Lines." These notes were partly based on research results from the COMET project. The notes were utilized in some of the severe weather seminars, and were distributed to neighboring NWS offices including NWS Indianapolis, NWS White Lake (Detroit), NWS Paducah, NWS Jackson, and NWS Wichita.

Based on completed and ongoing COMET research results, Ted Funk wrote a small document in late 1996 that discussed a preliminary conceptual model of the location of straight-line wind damage, tornadoes, mesolow and frontal structures with respect to bowing segments within squall lines. The document was given to all Louisville forecasters, and was distributed at the 19 March 1997 and 27 March 1997 severe weather seminars.

The dates shown below indicate training seminar days conducted at NWS Louisville or in Lexington on which NWS Louisville project members presented or attended throughout the period of the COMET project. Many of the seminars discussed squall line, bow echo, and tornado structure and evolution based on events studied at Louisville. Much of the information presented was based directly or indirectly on research and knowledge gained from the COMET project.

25-26 August 1994 18 April 1995 28 March 1996 16 May 1996

14 March 1995 28 June 1995 24 April 1996

28 March 1995 27 March 1996 1 May 1996

30 March 1995

Through their attendance at one or more of the seminars, staff members from the following agencies have benefited from the knowledge of detailed convective storm structure gained during the COMET research project: NWS Louisville; NWS St. Louis; NWS Jackson; NWS Paducah; NWS Indianapolis; NWS Wilmington; NWS Wichita; various TV meteorologists/weathercasters in Louisville, Lexington, Bowling Green, KY, and Evansville, IN; United Parcel Service; and the University of Louisville. In addition, training materials based on the events studied also were shared with NWS Nashville, NWS Detroit, NWS Des Moines, and NWS Central Region Headquarters.

At the first COMET Cooperative Project Workshop at SLU on 30 November-1 December 1995, NWS Louisville's Don Kirkpatrick attended and presented a seminar entitled, "An Overview of the April 15, 1994 Squall Line over Kentucky and Southern Indiana."

At the second COMET Cooperative Project Workshop at SLU on 12-14 November 1996 the NWS Louisville participation included: Ted Funk discussed severe weather observations from the 10 April 1996 squall line over Kentucky and southern Indiana. Don Kirkpatrick and Kevin Darmofal (NWS Wichita) presented research results from the 15 April 1994 squall line. Ted Funk and Van DeWald conducted a successful workshop on the 18 May 1995 squall line over south central Kentucky.


There have several areas where Saint Louis University has benefited from this COMET cooperative research. They include:

1. Exposure of our graduate students to the NWS environment of real-time forecasting. Students visiting the local NWS St. Louis have been impressed with the technology and have seen first hand the need to study Doppler imagery analysis and storm structure, in addition obtaining a better understanding of the GOES 8/9 imagery for forecasting mesoscale convective systems.

2. Informal discussions of the faculty (Jim Moore and Frank Lin) have led to the addition of new course material in the syllabi of "Diagnosis and Prediction of Severe Local Storms," "Synoptic Meteorology I/II," "Meteorology of Severe Storms" and "Radar Meteorology."

3. Installing our "SLUBREW" software on the HP UNIX workstation at the NWS St. Louis office has resulted in good feedback on our software package, which has helped us to modify and enhance the package. Other sites (notably Louisville, KY; Paducah, KY; and possibly the NWS Training Center and SSD in CRH) have expressed interest in this software package, as well.

4. Both the NWS Louisville and the NWS St. Louis offices have provided WSR-88D imagery of examples of bow echo structures and MCSs associated with heavy convective rainfall, which have been useful as illustrations in the classroom. Also, both offices have provided numerous handouts and schematics, which have been included in class notes.

5. Please list the benefits to the NWS office resulting from the collaboration (promising new forecasting techniques, heightened interest in research in the office, better understanding of new observing systems, potential new hires, use of university personnel as resource, etc.). Please be as specific as possible, particularly in regard to any improvements in forecasting operations resulting from the COMET project.

St. Louis, MO Office

Forecasters at NWS St. Louis have significantly improved their diagnostic and prognostic skills in forecasting heavy rain-producing MCSs due to the collaborative research effort. Collective composite analyses and examination of the individual heavy rainfall cases has resulted in a strong signal of the environmental conditions associated with heavy rain-producing MCSs and the physical processes which produce these conditions. A conceptual model was first developed in Spring 1994 based on preliminary research findings, then refined and introduced in 1995 (Glass et al. 1995). This model has been one of the highlights of the heavy rain component of this research effort. When still in its developmental stage, the conceptual model was used in the forecast process to issue flash flood watches with good lead time for the 10-11 April 1994 and 27-28 April 1994 heavy rain/flash flood events. Recently the conceptual model and COMET findings were applied to forecast with significant skill the 27-28 April 1996 and 14 May 1996 heavy rain/flash flood events. Application of our findings and techniques has also been successful in defining cases where there was a limited heavy rainfall threat, or where the main threat area would lie outside the zone of forecast responsibility.

The COMET cooperative research has significantly expanded our understanding and knowledge on forecasting damaging winds and tornadoes associated with linear and bowing convective systems. Participants in the project, including forecasters at NWS St. Louis, NWS Louisville, and many other offices across the central United States, have improved their knowledge base concerning bow echo/vortex evolution. Today, they are beginning to put into practice new conceptual models and techniques that have been developed over the past three years in weather warning situations. Observations and results from the past three years of our study have allowed all team members to build upon earlier conceptual models and foundations and gain a deeper understanding of the roles of mesoscale airflow structures, vortex evolution, and other internal processes associated with bowing MCSs.

Forecasters at NWS St. Louis have benefited from the COMET cooperative in at least four ways.

1. Acquired a greater awareness and understanding of the role of the pre-environmental conditions (e.g., CAPE, vertical wind shear) associated with MCSs which produce damaging winds and tornadoes. Forecasters are viewing nearby profiler data and WSR-88D VWP data much more today compared to the convective seasons of 1993 and 1994.

2. Acquired a greater awareness and understanding of the storm reflectivity patterns and characteristics associated with bowing MCSs.

3. Acquired a much greater understanding and appreciation in the area of circulation evolution (e.g., vortices that spawn tornadoes and damaging winds). The information gained from this subcomponent of the project provided new information and insights to all of our forecasters.

4. Became more aware of the usefulness of the MARC signature as a very important warning tool for issuing severe thunderstorm and tornado warnings.

Examples of the benefits of the COMET cooperative research on severe weather associated with bow echoes include the following cases:
Case #1: 8 July 1995: A multicell cluster of storms evolved into a bowing structure over central through southeast Missouri. Prior to the convective system bowing out, forecasters were quickly able to use the MARC technique to determine the potential for damaging straight-line winds and non-supercell tornadoes. The MARC signature showed magnitudes of 30-m s-1 velocity differentials along the leading edge of the multicell cluster, 30 to 35 miles west of Columbia and Boone County. Severe thunderstorm warnings were issued 15-25 minutes prior to the first report of damaging winds over western Boone and Coles Counties in central Missouri. Forecasters understood the potential for non-supercell tornado development to occur within the updraft/downdraft interface in the region of the MARC. A vortex rapidly formed near the apex of the bow over western Boone County and intensified/deepened along the northern flank of an evolving bowing structure. The severe thunderstorm warning was upgraded to a tornado warning. About 10 minutes later a brief tornado touched down 3 miles north of Columbia, MO in Boone County.

Case #2: 25 May 1996: Between 2030-2100 UTC 25 May 1996 a multicell cluster formed again west of Columbia in central Missouri. The MARC signature showed magnitudes of 25-30 m s-1 velocity differential over northern Boone and western Audrain Counties in central Missouri. Based upon the MARC velocity differentials, severe thunderstorm warnings were issued for eastern Boone County, and parts of Audrain and Callaway Counties. In this case the convective complex gradually evolved into a nearly solid, linear reflectivity pattern; no bowing structure was observed. However, forecasters at NWS St. Louis were again cognizant that non-tornadic and tornadic vortices could easily form within the region of the MARC. Between 2130-2145 UTC, 3 circulations formed within the region of the MARC. A tornado warning was issued for southern Pike County in Missouri based on the location and rapid growth of one of the three vortices. The strongest of the three vortices spawned a F1 intensity tornado, while the other two vortices produced damaging winds.

In both cases, the MARC velocity signature proved to be a reliable warning tool for anticipating severe thunderstorm winds and the growth of non-supercell tornadoes. Severe thunderstorm warnings with lead times of 15-20 minutes were issued prior to the first occurrence of wind damage. Research conducted from the COMET cooperative project played a critical role in the warning process and directly aided in the issuance of timely and accurate warnings for each event. As of 31 December 1996 the MARC velocity signature is being tested on past cases at 8 NWS forecast offices across the Central Region.

Staff members of the NWS St. Louis have significantly benefited from both the heavy rainfall and damaging wind components of the COMET cooperative. Forecasters at NWS St. Louis have acquired a higher level of understanding from results achieved in both components of the project. Synoptic, mesoscale and storm-scale conceptual models and findings derived from observations collected during the course of the last three years have given NWS forecasters (not only in St. Louis and Louisville, but across the Midwest) much keener insights and understandings of key mesoscale and storm-scale features and processes associated with MCSs which produce heavy rainfall, damaging winds and non-supercell tornadoes. The number of vortices sampled at NWS St. Louis in both warm and cool seasons have given forecasters a greater degree of knowledge concerning vortex evolution and the occurrence of tornado activity. The roles of storm reflectivity and mesoscale airflow structure with respect to both the occurrence of damaging winds and heavy convective rainfall have been better documented to help forecasters issue more timely and accurate watches and warnings. The 12 damaging wind MCS cases surveyed by the St. Louis team is only a small sample of the cases needed to produce a high quality, complete and accurate study. Through educational and training efforts derived from the COMET cooperative, NWS forecasters at St. Louis, Louisville and other nearby sites have increased their awareness and understanding of MCS evolution in relation to severe weather and flash flooding. Much progress has been made over the past three years with the COMET cooperative study, however much remains undone.

Louisville, KY Office

Several people at NWS Louisville have used the WSR-88D imagery from various squall lines and knowledge learned from our COMET research at many Spotter Training Classes across central Kentucky during 1996-1997.

The majority of Kentucky's severe weather results from squall lines and bow echoes. COMET collaborative research performed thus far on several squall lines across Kentucky and Indiana, and review of related literature has greatly increased the knowledge of convective storm structure of most NWS Louisville forecasters. A more detailed knowledge and understanding of RIJs, mesocyclones and tornado development, mesoscale airflow and frontal structures, and squall line evolutionary processes in general, have resulted. The project also has had a tremendous positive impact on the ability of the NWS Louisville forecasters to issue accurate and timely severe weather warnings. In several cases in 1995, 1996, and early 1997, forecasters have been able to issue severe thunderstorm and tornado warnings for squall lines and bow echoes with more precision and timeliness than before the project. Research results may have at least partially helped warning operations at neighboring offices as well.

An example of research benefits at NWS Louisville occurred on 20 April 1996, when a squall line moved across southern Indiana and Kentucky. Forecasters were able to differentiate those portions of the squall line where wind damage would be greatest, and highlight warnings accordingly. In addition, severe thunderstorm warnings were upgraded to tornado warnings for two counties based on WSR-88D reflectivity and velocity patterns, and detailed knowledge gained from our COMET research. In both counties, tornadoes developed and did damage, particularly in Madison County where the tornado warning was issued with 9 minutes lead time, despite a rapidly moving and evolving line. In Ted Funk's opinion, the COMET research clearly and directly helped with the issuance of accurate and timely tornado warnings for the two counties.

Squall line events on 7 November 1996, 1 March 1997 and 28 March 1997 across Kentucky and southern Indiana were also handled well by the NWS Louisville forecasters. In these cases, tornadoes were associated with rather subtle radar signatures; however, ingrained knowledge of bow echo mesocyclone structure derived from our COMET research permitted accurate tornado warnings to be issued for several counties. In addition, a Louisville TV weathercaster used information obtained at the 19 March 1997 seminar at NWS Louisville to show the public likely areas of wind damage and tornadoes with respect to a bowing line segment on 28 March 1997. His knowledge apparently was based on our COMET research.


Moore, J. T., and F. H. Glass, 1992: Mesoscale Convective Systems: Initiation and Propagation. Preprints, 4th Atmospheric Environment Service, Canadian Met. And Ocean. Soc. Workshop on Operational Meteorology, Whistler, Canada, 215-224.

Moore, J. T., C. H. Pappas, and F. H. Glass, 1993: Propagation Characteristics of Mesoscale Convective Systems. Preprints, 17th Conference on Severe Local Storms, St. Louis, MO, Amer. Meteor. Soc., 538-542.

Moore, J. T., and F. H. Glass, 1993: The Great Flood: The St. Louis Perspective. Invited talk, 17th Conference on Severe Local Storms, St. Louis, MO, Amer. Meteor. Soc.

Moore, J. T., S. M. Rochette, D. L. Ferry, and F. H. Glass, 1993: Mesoscale Convective Systems Associated with Heavy Rainfall in Missouri. Presentation at the National Weather Association Annual Meeting, Raleigh, NC.

Zarauz, J. V., 1994: Intercomparison of Areal Rainfall Estimates Obtained through WSR-88D Radar, a NMC Model, a Research Model and Raingage Measurements. M.S. Thesis, Saint Louis University, 73 pp.

Rochette, S. M., 1994: Evolution and Propagation of Heavy Rainfall-Producing Mesoscale Convective Systems. M.S. Thesis, Saint Louis University, 208 pp.

Moore, J. T., S. M. Rochette, F. H. Glass, and D. L. Ferry, 1994: Environmental Conditions Favorable for the Production of Heavy Convective Rainfall During the Warm Season. Presentation at the Fourth National Heavy Precipitation Workshop, Scottsdale, AZ.

Glass, F. H., D. L. Ferry, J. T. Moore, and S. M. Rochette, 1994: The Use of Gridded Model Data in Forecasting Mesoscale Convective Systems Which Produce Heavy Rain. Presentation at the Fourth National Heavy Precipitation Workshop, Scottsdale, AZ.

Moore, J. T., S. M. Rochette, F. H. Glass, D. L. Ferry, and S. M. Nolan, 1994: Characteristics of Elevated Thunderstorms Associated with Heavy Rainfall. Presentation at the National Weather Association Annual Meeting, Salt Lake City, UT.

Rao, G. V. and J. V. Zarauz, 1994: An Intercomparison of the Areal Rainfall Estimates Obtained through WSR-88D radar, a NMC model, a Cloud Model, and Raingages in Missouri. Preprints, 10th Interactive Information and Processing Systems Conference, Nashville, TN, Amer. Meteor. Soc., 247-248.

Rochette, S. M., J. T. Moore, P. S. Market, F. H. Glass, and D. L. Ferry, 1995: Heavy Rain Events Associated with Elevated Thunderstorms in the Midwest. Postprints, 4th National Winter Weather Workshop, Kansas City, MO, NOAA Tech. Memo. NWS CR-112, 25:1- 12.

Glass, F. H., D. L. Ferry, J. T. Moore, and S. M. Nolan, 1995: Characteristics of Heavy Convective Rainfall Events across the Mid-Mississippi Valley during the Warm Season: Meteorological Conditions and a Conceptual Model. Preprints, 14th Conf. On Weather Analysis and Forecasting, Dallas, TX, Amer. Meteor. Soc., 34-41.

Market, P. S., and J. T. Moore, 1995: An Isentropic Perspective on Elevated Thunderstorms Associated with Heavy Rainfall. Postprints, Conf. on Isentropic Analysis and Forecasting, Millersville Univ., Lancaster, PA, 37-39.

Moore, J. T., P. S. Market, S. M. Rochette, and S. M. Nolan, 1995: An Evaluation of Precipitation Efficiency Factors for Forecasting Heavy Rainfall. Presentation at the Missouri Academy of Science Annual Meeting, Maryville, MO.

Rochette, S. M., J. T. Moore, P. S. Market, and S. M. Nolan, 1995: Synoptic Scale Characteristics of Cold Season Heavy Rain Events in Missouri. Presentation at the Missouri Academy of Science Annual Meeting, Maryville, MO.

Rochette, S. M., 1995: A Case Study of a Heavy Rainfall Event Associated with Elevated Convection. Abstracts, 20th Annual Northeastern Storm Conference, Saratoga Springs, NY, 18.

Moore, J. T., S. M. Nolan, and S. M. Rochette, 1995: Techniques for Predicting Cool Sector versus Warm Sector Heavy Convective Rainfall Events in the Midwest. Presentation at the National Weather Association Annual Meeting, Houston, TX.

Moore, J. T., S. M. Nolan, F. H. Glass, D. L. Ferry and S. M. Rochette, 1995: Flash Flood Producing High-Precipitation Supercells in Missouri. Preprints, 14th Conf. on Weather Analysis and Forecasting, Dallas, TX, Amer. Meteor. Soc., (J4)7 (J4)12.

Przybylinski, R. W., Y-J. Lin, G. K. Schmocker, and T. J. Shea, 1995: The Use of Real-Time WSR-88D, Profilers, and Conventional Data Sets in Forecasting a Northeastward Moving Derecho over Eastern Missouri and Central Illinois. Preprints, 14th Conf. on Weather Analysis and Forecasting, Dallas, TX, Amer. Meteor. Soc., 335-342.

Moore, J. T., S. M. Rochette, F. H. Glass, D. L. Ferry, and P. S. Market, 1996: Elevated Thunderstorms Associated with Heavy Rainfall in the Midwest. Preprints, 18th Conference on Severe Local Storms, San Francisco, CA, Amer. Meteor. Soc., 772-776.

Funk, T. W., B. F. Smull, and J. D. Ammerman, 1996: Structure and Evolution of an Intense Bow Echo Embedded within a Heavy Rain Producing MCS over Missouri. Preprints, 18th Conference on Severe Local Storms, San Francisco, CA, Amer. Meteor. Soc., 521- 526.

Przybylinski, R. W., Y-J. Lin, C. A. Doswell, G. K. Schmocker, T. J. Shea, T. W. Funk, J. D. Kirkpatrick, K. E. Darmofal, and M. T. Shields, 1996: Storm Reflectivity and Mesocyclone Evolution Associated with the 15 April 1994 Derecho. Part I: Storm Evolution over Missouri and Illinois. Preprints, 18th Conference on Severe Local Storms, San Francisco, CA, Amer. Meteor. Soc., 509-515.

Funk, T. W., K. E. Darmofal, J. D. Kirkpatrick, M. T. Shields, R. W. Przybylinski, Y-J. Lin, G. K. Schmocker, and T. J. Shea, 1996: Storm Reflectivity and Mesocyclone Evolution Associated with the 15 April 1994 Derecho. Part II: Storm Structure and Evolution over Kentucky and Southern Indiana. Preprints, 18th Conference on Severe Local Storms, San Francisco, CA, Amer. Meteor. Soc., 516-520.

Schmocker, G. K., R. W. Przybylinski, and Y-J. Lin, 1996: Forecasting the Onset of Damaging Downburst Winds Associated with a Mesoscale Convective System (MCS) using the Mid- Altitude Radial Convergence (MARC) Signature. Preprints, 15th Conf. On Weather Analysis and Forecasting, Norfolk, VA, Amer. Meteor. Soc., 306-311.

Rochette, S. M., and J. T. Moore, 1996: Initiation of an Elevated Mesoscale Convective System Associated with Heavy Rainfall. Wea. Forecasting, 11, 443-457.

Rochette, S. M., J. T. Moore, and P. S. Market, 1996: Surface Composites of Elevated Heavy Rainfall Episodes in the Midwest. Presentation at the Annual Meeting of the Missouri Academy of Science, Springfield, MO.

Rochette, S. M., and J. T. Moore, 1996: Composite Analyses of Heavy Rainfall-Producing Elevated Thunderstorms in the Midwest. Abstracts, 21st Annual Northeastern Storm Conference, Saratoga Springs, NY, 24.

Rochette, S. M., P. S. Market, and J. T. Moore, 1997: Conventional and isentropic analyses of a heavy rainfall episode associated with elevated convection. Presentation at the Third Annual University of Wisconsin Winter Weather Workshop, Madison, WI.

Rochette, S. M., J. T. Moore, F. H. Glass, and D. L. Ferry, 1997: The Importance of Parcel Choice in Elevated CAPE Computations. Submitted to Weather and Forecasting.

Przybylinski, R. W., Y.-J. Lin, G. K. Schmocker, T. W. Funk, K. E. Darmofal, J. D. Kirkpatrick, and V. L. DeWald, 1997: Storm Reflectivity and Mesocyclone Evolution Associated With the 15 April 1994 Squall Line, Part I: Storm Structure and Evolution Over Missouri and Illinois. To be submitted to Weather and Forecasting.

Funk, T. W., K. E. Darmofal, J. D. Kirkpatrick, V. L. DeWald, R. W. Przybylinski, G. K. Schmocker, and Y-J. Lin, 1997: Storm Reflectivity and Mesocyclone Evolution Associated with the 15 April 1994 Squall Line, Part II: Storm Structure and Evolution Over Kentucky and Southern Indiana. To be submitted to Weather and Forecasting.


Probably the greatest problem at the moment is finding time to meet with our NWS counterparts to discuss research problems and swap ideas. Although we have been able to do a great deal through e-mail, fax, phone, etc., it doesn't make up for face-to-face meetings. Staffing problems, fewer supernumerary shifts, and increased workloads on NWS personnel make doing COMET-related work quite difficult at times. We are always impressed by the amount of effort that the NWS members put into the COMET cooperative activity and their commitment to the project.


The only major problem encountered during the COMET Cooperative project has been finding adequate time for project participants to work on relevant cases due to shift work, operational work responsibilities, and focal point duties. However, sufficient time has been allocated to permit significant research results.

WSR-88D Level II data was unavailable for research of bow echo events during late 1994 and 1995. However, Level II data and the WATADS software are available for study of 1996 and 1997 events.

Two project participants have left NWS Louisville during the project (Mike Shields and Kevin Darmofal) which impacted the continuity of the project research. However, excellent replacements for these forecasters have greatly minimized any negative impact.