Univ. of Arizona: "Mesoscale modeling for Arizona"

Final Report

SECTION I: OBJECTIVES AND ACCOMPLISHMENTS

A. COMET FUNDED WORK

The objectives outlined in the proposal were met through the following activities

• Maintained an operational system of real-time, numerical forecasts for the Southwestern United States and Northern Mexico. The computational grids have been nested, with the highest resolution grid over the state of Arizona. In this region, the resolutions have ranged from 2- to 7-km. The period of time covered in the forecasts was 24-48 hours and the initialization at 00 UTC.
• Provided the output from real-time forecasts to communities at NWS and UA. In addition, this output was converted into GEMPAK format for display with visualization programs. A graphical version of the forecasts was provided to the general public via the world-wide-web (http://nimbo.wrh.noaa.gov/tucson/mm5/mm5tus.html and www.atmo.arizona.edu/mm5/mm5_az.html).
• Conducted a workshop in Tucson to discuss forecast issues of the Southwest United States and Mexico. This workshop, entitled the Southwest Weather Symposium, was attended by over 70 people including representatives from the local media, military, southwest NWS, University of Arizona and the Mexican weather service. We received permission from Vickie Johnson to use unspent COMET funds to partially support the symposium. The first symposium was very well received and we sponsored a second symposium in 2000 through a separate COMET grant. A third symposium is being planned for 2000.
• Performed an analysis of validation and verification of the model forecasts. The validation considered the cases in which convective activity developed over the area of interest. In general, the model has some skill in reproducing the diurnal cycle of convection in the summer and to simulate the propagation of baroclinic systems in winter. The verification was designed for selected stations and compared fields from the forecasts and observations. This analysis has been used to identify the strengths and deficiencies of the model simulations.
• Performed sensitivity simulations to evaluate specific aspects of the physical parameterizations in the MM5 model. These runs included the schemes for boundary layer, convective parameterization and explicit moisture.

Work conducted by the University that was a result of the Outreach Program collaboration but was not directly funded by it.

• The Center for Computing and Information Technology (CCIT) at the University of Arizona provided computer time on the 32-processor SGI Origin 2000 machine. The CCIT allocated a three hour block of time in the late evening during which we had exclusive use of the computer. The output from the forecasts were archived at the Department of Atmospheric Sciences, which provided the resources for the data storage.
• The output from the real-time forecasts were made available to state and federal agencies, such as the Arizona Department of Environmental Quality and the Pacific Northwest National Laboratory. Another group that was interested in our products is the Tucson Soaring Club, which received specific displays for soaring purposes in one of our web-sites.
• Initiated collaboration with the Environmental Fluid Dynamics group in the Department of Mechanical & Aerospace Engineering at Arizona State University. This collaboration is focused in the study of the orographic flows over the mountains in Phoenix AZ.
• A case study of the landfall of Hurricane Nora (1997) was performed. This east Pacific hurricane made landfall on the coast of Baja California and the remnants passed into western Arizona and caused localized heavy flooding. MM5 simulations revealed details of the structural changes and evolution of heavy precipitation.

Work conducted by the NWS which was a result of collaboration with the university but was not directly funded by it

• The Tucson NWS used their local travel budget to pay for David Stensrud (NSSL) to attend the development meeting in Tucson. Western Region Headquarters also participated in the meeting and sent two meteorologists from Scientific Services Division.
• The NWS paid for one third of the cost of a connection to the University of Arizona campus network to facilitate the transfer of data between the NWS office and the Department of Atmospheric Sciences.
• David Bright invested a considerable amount of time into pre- and post-processing development work and also into the development of the real-time system. Several members of the NWS Tucson staff have helped to evaluate the MM5 output.
• David Bright attended the NCAR MM5 Users Workshop and presented a poster describing the real-time system in Arizona. This trip was paid for through NWS funds.
• MM5 ensemble forecasting for Arizona during the monsoon (Bright D.R. and S.M. Mullen, 2001: Short range mesoscale ensemble forecasts of precipitation over the southwest United States during the summer convective season. AMS Precipitation Extremes Conference. Albuquerque, NM, January 2001.)
• Using the MM5 to study Gulf of CA moisture surges (McCollum D.M. and DR Bright, 1998: The Gulf of California: A significant source of moisture during Arizona's summer thunderstorm season? Sensitivity experiments using the MM5. AMS 19th Conference on Severe Local Storms. Minneapolis, MN, September, 1998.)

SECTION II: BENEFITS TO THE UNIVERSITY

The University of Arizona has obtained the following benefits:

• The interaction of UA researchers and NWS forecasters has led to an improvement in our understanding of model performance. The MM5 model is used in a wide variety of applications and there are a number of options for physical parameterizations, most of which have been designed for specific applications.
• Collaboration with NWS personnel allowed us to focus on problems that were of particular local interest and support the operational forecast mission of NWS. In addition, we received immediate feedback on the accuracy and utility of the MM5 output from the NWS forecasters.
• The UA was allowed access to additional observational data for model initialization and verification. GOES 9 sounder information and additional surface observations were used to augment the ETA analysis and forecast grids used in the real time MM5 runs.
• The UA/NWS collaboration gave our students experience with problems related to regional weather forecasting. The MM5 output was displayed on the department web page and hard copy was posted to the map board. This allowed our students to compare MM5 output with that from the other NCEP models and gain first hand experience with regional model verification and evaluation.

SECTION III: BENEFITS TO NWS

• The NWS has benefited from the project in several ways. All of the forecasters attended at least part of the MM5 workshop and benefited from many of the presentations. Most of the forecasters have improved their understanding of mesoscale models as a result of the workshop and the real-time MM5 forecasts.
• The MM5 model has been incorporated into the real-time forecast process at NWSO Tucson. One advantage of the locally run MM5 model is the availability of hourly output. Qualitative evaluation suggests the model forecast is of high quality; however, a quantitative evaluation has not yet been started. Most forecasters are very interested in the MM5 convective precipitation during the first 12 hours of the forecast and the transport of low level moisture during the second 12 hours of the forecast.
• Our proof of concept simulations showed that low level moisture initialization was sometimes poor over the Southwest. In order to improve this initialization we acquired an hourly grid of special soundings from the GOES-9 Sounder over the Southwest U.S. and Northern Mexico. These data were incorporated into the MM5 pre-processing system and are also available to the forecasters every hour for real-time analysis. All of the forecasters have learned something about the Sounder data and have also come to rely of these data for real-time analysis.
• David Bright (Tucson SOO) shared the real-time software developed specifically for this project, and experience gained in real-time MM5 modeling, with several NWS offices, including: WFO St. Louis (to assist with a COMET Fellowship Project), WFO Reno, WRH Salt Lake City, WFO Spokane, WFO Phoenix, WFO Flagstaff, and the First COMET Symposium on NWP.

SECTION IV: OUTREACH PROGRAM-RELATED REFERENCES (publications, presentations, theses, student papers)

Bright, David: Real-time MM5 forecasts for Arizona during the monsoon. (Presentation) MM5 Users Conference, Boulder, CO, 1997.

Luis M Farfan, David R Bright and Joseph A Zehnder: Numerical Simulation of Mesoscale Weather Systems in Arizona. AMS 12th Conference on Numerical Weather Prediction. Phoenix Arizona, January 11-16 1998.

Luis M Farfan: Real-time weather forecasts for Arizona. Institute of Atmospheric Physics Seminar. Tucson Arizona, January 30 1998.

Farfan, L.M., J.A. Zehnder and D.R. Bright, 2000: Verification of mesoscale model simulations for Arizona. Southwest Weather Symposium, post-print volume.

Farfan, L.M. and J.A. Zehnder, 2001: An analysis of the landfall of hurricane Nora (1997). Mon.Wea.Rev.