Close Window

Univ. of Utah: "Mesoscale weather prediction over Utah"

Final Report

Section 1. Project Accomplishments

Scientific exchange and cooperation between the University of Utah and the Salt Lake City NWSFO has continued to grow during the past two years. A Memorandum of Understanding between the University of Utah and the Department of Commerce has led to the establishment of the NOAA Cooperative Institute for Regional Prediction at the University of Utah during 1996. The Institute moved into its new facility in the Intermountain and Network Scientific Computation Center recently. A dedicated T-1 line provides fast communications between the NWS and researchers at the Institute.

The establishment of the Cooperative Institute would not have been possible without the track record of research and scientific interaction between the Department of Meteorology, the Salt Lake NWSFO, and Western Region of the National Weather Service. That interaction was facilitated principally with support over the past 6 years from COMET. Further information on the Cooperative Institute is available at

The COMET Cooperative project has had several specific goals during the past two years:

(1) develop and apply the Utah Local Area Model (LAM) to mesoscale forecasting problems in the intermountain region.

(2) develop and apply the Utah Mesonet as a means to develop and validate conceptual models of the interactions between mesoscale weather features and the complex terrain of the intermountain region.

(3) host the annual Intermountain Weather Prediction Workshop as a means to facilitate exchange of information between researchers and operational forecasters.

(4) evaluate techniques to improve precipitation estimates over regions of high terrain through the incorporation of Utah Mesonet observations and output from WSR-88D radars.

(5) begin planning for the Winter Olympics to be held in Salt Lake City in February 2002.

The Utah LAM no longer represents a state-of-the-art model and its use and significance at the University and forecast office have diminished correspondingly.What was novel five years ago about the application of the Utah LAM in the forecast office is now commonplace. The Eta Model at 32 km is viewed by us as being a better model for operational forecasting in the intermountain region. Intercomparison of the ULAM, NGM, and Eta models is described by White et al. (1998).

Of greater significance to the first goal of this project has been the increased application of real-time MM5 forecasts. Jim Steenburgh supervises the twice-per-day forecasts of the MM5 with spatial resolution comparable to that of the Mesoscale Eta Model. This model is now run on either an in-house Sun workstation or a SGI Origin 2000 computer managed by the Center for High Performance Computing. This work has received considerable visibility locally with demonstrations for high performance computer users and SGI representatives.

Occasionally, as conditions warrant, Jim performs forecasts in real-time at enhanced resolution (on the order of 6 km) (Onton and Steenburgh 1998). At the present time, forecasters can access the output from the MM5 via the Department’s Web server rather than raw files transmitted to the forecast office. Work is underway to transfer the model output into the AWIPS system. The latest graphics from the MM5 can be obtained at

The Utah Mesonet has been a major cornerstone of the COMET Cooperative Project. It relies on frequent coordination between the Salt Lake NWSFO, Colorado River Basin Forecast Center, and other participating agencies with the University of Utah. Data is collected at the University from the Salt Lake NWSFO and other agencies every 15 minutes. The data is processed, archived, converted into graphical displays, and graphics and text files are transmitted to the NWSFO every 15 minutes. The graphical products from the Utah Mesonet have been used extensively for real-time evaluation of frontal passages, lake-effect snow storms, and many other weather events in the intermountain region. The Utah Mesonet graphics can be obtained at the University via David Zaff and others at the NWSFO have developed additional software to display the model output for operational purposes. This interface is available at http://

The Utah Mesonet component of the COMET project continues to expand to include other agencies. PI’s John Horel and Larry Dunn have spent innumerable hours attending meetings with personnel associated with the Utah Department of Air Quality and Utah Department of Transportation and other local transportation agencies. Additional networks have been added in Nevada (ARL SORD) and Idaho (Agrimet). Efforts are now being placed on reaching agreements to access commercial networks, such as Kennecott and Utah Power.

Some of the most significant improvements to the Utah Mesonet have been behind the scenes. Dan Judd, Judd Communications, has been hired with other funds from the NWS and the University to troubleshoot communications problems. This has improved access to weather data at remote stations in the Wasatch Mountains and near the Great Salt Lake. Quality control measures have been implemented to flag data that is in error or suspicious (Splitt and Horel 1998).

Jason Stiff completed the requirements for the M.S. Degree using information from the Utah Mesonet (Stiff 1997). Although supported by a NWS-COMET Graduate Fellowship, Jason is now working in the broadcasting industry as a result of the NWS hiring freeze during 1997. Jonathan Slemmer is nearing completion of the requirement for his M.S. using radar and Mesonet data (Slemmer 1998). He will be joining the Oxnard NWSO during May 1998. After delays resulting from his full-time work commitments, Alan Haynes completed the requirements for his M.S. (Haynes 1998). Alan is a lead forecaster at the Pueblo,Colorado NWSO.

A joint project between Western Region, the NWSFO, NSSL, and the University developed to evaluate winter precipitation and high wind algorithms for the WSR-88D. Weather sensors were deployed at a number of locations and incorporated into the Mesonet. Success has been mixed, as the goal to define in real-time a Z-S relationship has been difficult.

The 2002 Winter Olympics that are to be held in Salt Lake City has led to the deployment of weather sensors at many venue sites. Most of the equipment was contributed by the NWS, with some equipment purchased by the Salt Lake Olympic Committee or the University. The data are incorporated into the Utah Mesonet and available to forecasters, researchers, and Olympic planners.

Brett McDonald is nearing completion of his Ph.D. studies. His thesis is on validation of precipitation forecasts from operational and research models. He has made several presentations at national conferences (McDonald and Horel 1997, 1998a,b).

2. Summary of University/ NWS Exchanges

Interaction between John Horel and Jim Steenburgh at the University and Larry Dunn at the NWSFO has continued on a frequent basis. Most of this exchange focuses on making sure that the Utah Mesonet is operating properly, discussing operational and research model forecasts, or the forecast problem of the day.

A seminar series sponsored by the NOAA Institute began during Spring 1997. Several speakers have been invited with attendance by many NWSFO and Western Region personnel. Interaction continues through other means such as informal meetings and the local chapter of the American Meteorological Society.

The Third Annual Intermountain Weather Prediction Workshop was very successful during 1996. Over 100 individuals from the National Weather Service, other federal and local agencies, and universities attended this event. The scheduling of the Workshop in conjunction with related workshops for Science and Operations Officers and Warning Coordination Meteorologists facilitated enhanced participation from around the Western Region.

The Fourth Annual Intermountain Weather Prediction Workshop received highly favorable reviews by participants. The Workshop was held during September 12, 1997 and was organized by Jim Steenburgh. 85 individuals attended, a slight drop from the previous year since no NWS meetings preceded it. The workshop schedule and attendee list can be obtained from

3. Presentations and Publications

M. S. Theses

Haynes, A., 1998: A Comparison of Techniques to Evaluate Quasi-Geostrophic Vertical Motion. M. S. Thesis. University of Utah. 107 pp.

Stiff, J., 1997: The Utah Mesonet. M.S. Thesis. University of Utah. 110 pp.

Slemmer, J., 1998: Characteristics of Winter Snowstorms near Salt Lake City as Deduced from Surface and Radar Observations. M.S. Thesis. University of Utah.142 pp.

Conference Proceedings and Refereed Articles

Horel, J. D., and J. Stiff, 1997: The evolution of winter storms over Utah as determined from the Utah Mesonet. Fifteenth Conference on Weather Analysis and Forecasting. Norfolk, VA. 52-53.

Lazarus, S., C. Ciliberti, and J. Horel, 1998: Application of a Local Analysis System in Highly Variable Terrain. Sixteenth Conference on Weather Analysis and Forecasting. Phoenix, AZ. 262-265.

McDonald, B., and J. Horel, 1997: Precipitation forecast skill over California of NCEP models.

Fifteenth Conference on Weather Analysis and Forecasting. Norfolk, VA.

____ and ____, 1998: Evaluation of Precipitation Forecasts from the NCEP’s 10 km Mesoscale Eta Model. Twelfth Conference on Numerical Weather Prediction. Phoenix, AZ. J27-30.

______, _____, C. J. Stiff, W. J. Steenburgh, 1998: Observations and Simulations of three downslope wind events over the northern Wasatch Mountains. Sixteenth Conference on Weather Analysis and Forecasting. Phoenix, AZ. 62-64.

Onton, D. J., and W. J. Steenburgh, 1998: Simulations of post-frontal convective snowbands to the lee of the Great Salt Lake. Sixteenth Conference on Weather Analysis and Forecasting. Phoenix, AZ. 353-354.

Splitt, M., and J. Horel, 1998: Use of Multivariate Linear Regression for Meteorological Data Analysis and Quality Assessment in complex Terrain. Tenth Symposium on Meteorological Observations and Instrumentation. Phoenix, AZ. 349-350.

White, B. G., W. J. Steenburgh, J. Paegle, J. D. Horel, R. T. Swanson, J. Miles, 1998: Short-term forecast validation of six models for winter 1996. Twelfth Conference on Numerical Weather Prediction. Phoenix, AZ. 135-136.

White, B. G., J. Paegle, W. J. Steenburgh, J. D. Horel, R. T. Swanson, 1998: Short-term validation of six models for winter 1996. Accepted. Weather and Forecasting.

4. Summary of Benefits and Problems Encountered

University Perspective
Over the 6 years that COMET has supported research activities at the University of Utah and Salt Lake NWSFO, there has been a steady maturation of the joint efforts. Initial efforts focussed on access to real-time forecasts of the ULAM. It quickly became apparent that the lack of access to real-time data in many regions of the intermountain region affected the quality of the initial analyses and evaluation of the model forecasts. Thus, the shift in effort to develop the Utah Mesonet.

It can not be stressed enough that the Utah Mesonet is a joint effort. The NWS bears the brunt of routine costs, such as large phone bills, has contributed significant resources (computers and mesonet stations), and has the personnel to maintain the system 24 hours per day. The University role has been to develop the software to access, archive, and display the data. The activities required to maintain the Utah Mesonet will continue indefinitely. However, the type of effort has shifted from being experimental to operational. Thus, support for the Utah Mesonet will continue as part of the base activities of the Cooperative Institute. Efforts are planned for Summer 1998 to improve the communications to remote sites, which remains one of the most time-intensive aspects of the Mesonet.

The most significant indication of the maturation of these joint activities has been the establishment of the NOAA Cooperative Institute for Regional Prediction. This would not have been possible without the track record of research and scientific interaction between the Department of Meteorology, the Salt Lake NWSFO, and Western Region of the National Weather Service. That interaction was facilitated initially with the support from COMET.

It should be recognized, however, that some aspects of the interaction between the NWS and the University of Utah will remain experimental in nature, applicable to or involving other NWSFO’s, and worthy of support from COMET. Such activities will not be automatically supported by base funding for the Cooperative Institute.

NWS Perspective
The level of interaction between the SLC forecast office and the University of Utah Meteorology Department continues to grow. We share data, we share model output, we share research results, and we share ideas. The results of this interaction have had a positive impact on virtually all aspects of our operation.

Without question, the biggest forecast problem faced by the SLC office is how to deal with mesoscale phenomena. In the past, there has been almost no research into mesoscale meteorology in the Great Basin. The result is that there is little understanding of the phenomena observed, and the accompanying forecast skill has been low. The interaction with the University Meteorology Department has changed the situation on a number of fronts.

Observations are collected, decoded, and displayed in real-time. This is an important first step towards understanding the extremely complex mesoscale processes that take place in this region. The Utah Mesonet project is likely a leader nationwide in showing how data from many different agencies can be brought together and shared by all in real-time. The Utah Mesonet has been the basis for bringing cooperation among the weather-sensitive agencies and the NWS to a new level, and it would not have happened without the University.

We are seeing and documenting phenomena in the mesonet that have not been observed before. Forecasters routinely have mesonet displays on their workstations throughout their shift. The view of surface conditions in the region is unprecedented for us. In the coming year the mesonet will be brought into AWIPS and integrated with radar, satellite and lightning data. One of the reasons the NWS is testing the Local Data and Dissemination System portion of AWIPS at SLC is because of the extensive real-time mesonet that exists here. The mesonet continues to grow and become more useful.

The observational data is also the basis for an automated analysis scheme that will eventually be used to initialize local models. At this time output from MM5 runs at the University are made available in real-time to the forecasters. The resolution is similar to the meso-Eta, but the ouput is at hourly resolution, and the different physics and coordinate system offers the forecaster an alternative view of the situation at hand.

The modeling activity has also been important in a post-analysis mode. We have looked to the University to rerun certain cases at very high resolution and with various physical processes turned ‘on’ or ‘off’ to try and gain a better understanding of the atmospheric phenomenon, and to assess likely sources of error in the NCEP models. This includes post-frontal precipitation banding cases and other situations in which the NCEP models erroneously forecast major snow events in strong southwest flow. This latter problem is persistent in certain synoptic regimes.

The analysis and local modeling activity will become a cornerstone for the high resolution forecasts that will be required to support the 2002 Winter Olympics. The goal of accurate forecasts and warnings with high temporal and spatial resolution is at the core of the NWS mission. The 2002 Winter Olympics offer a very tangible target to aim at in accomplishing this goal, and the interaction with the University Meteorology Department on research, modeling, and real-time issues is critical for this effort.

Finally, the office training and professional development program has moved to a new level as a result of the interaction with the University. On-station training and local research results are increasingly web-based. It is possible to include high resolution imagery, model output, color graphics, and animation within self-paced training material and even station drills. This is being done at a national level in the NWS, but also at the local level. Much of the material produced at the SLC office would not be possible without the extensive archive maintained at the University. Our requests for all sorts of data are filled quickly. We could not maintain anywhere near this sort of archive ourselves.

A study of windstorms in southwest Utah produced a forecast decision tree that is used extensively. This project could never have produced such robust results without access to the University archive. We have been maintaining a list of all lake-effect snow events since the installation of the KMTX WSR-88D. The university archive allows us to ‘rebuild’ these cases to look at them in more depth. An on-line downslope windstorm station drill, a severe convection drill, and a heavy snow training package are all highly interactive with imagery and animation due to the University archive.

In summary, the interaction between the University of Utah Meteorology Department and the SLC forecast office is an example of how a relatively small amount of funding from COMET has had a huge impact on NWS operations. It is producing exactly the type of results envisioned when the Outreach program was conceived.