Project to Intercompare Regional Climate Simulations- Workshop II

5 - 6 June 1997 Jointly Sponsored by the
Electric Power Research Institute
and the
International Institute of Theoretical and Applied Physics

INTRODUCTION

The Project to Intercompare Regional Climate Simulations (PIRCS) held its second modelers workshop 5 - 6 June 1997 at Iowa State University, Ames, Iowa, USA. Twenty-five scientists representing sixteen modeling groups attended. The overall goal of PIRCS is to advance the status of regional climate simulation using atmospheric mesoscale models. In accordance with this goal, active participants convened several months after the release of the boundary conditions for the first PIRCS simulation experiment (Takle, 1995) to discuss several topics:

Technical problems with performing the simulation

Initial results from some models

Strengths and limitations of the PIRCS boundary conditions

Output archive requested from participating models

Observational data needs

An underlying theme of these discussions was the challenge presented to the mesoscale climate modeling community by the Working Group on Numerical Experimentation (WGNE) to evaluate thoroughly the feasibility of using mesoscale models for climate simulation.

PIRCS Experiment 1 was established during the first PIRCS workshop in November 1994 (Takle, 1995). Participants at that workshop determined that the initial experiment should focus on the central U.S. drought of 1988 (Exp. 1a) and floods of 1993 (Exp. 1b), two periods offering strong signals of climate variability for the models to reproduce. These choices were motivated also by the substantial observational record available for evaluating simulations and the interest of the GEWEX Continental-Scale International Project (GCIP) in these episodes. A prominent mesoscale feature of this region is a nocturnal, low-level jet that plays a central role in the region's hydrologic cycle. Participants in the November 1994 workshop and follow-on discussions developed a standard output archive for participating models for diagnosing this feature and its interaction with regional energy and water cycles.

Workshop II was supported by the International Institute of Theoretical and Applied Physics (IITAP) at Iowa State and the Electric Power Research Institute (EPRI). Early in the workshop, James Vary, Acting Director of IITAP, reviewed IITAP's support of PIRCS in the context part of IITAP's goal as a UNESCO-sponsored institution to promote forefront research by scientists in developing countries. He also surveyed participants later in the workshop for their collaboration with developing country scientists, stating that IITAP may be able to sponsor such interaction. Chuck Hakkarinen, of the EPRI Environment Group, noted that EPRI has supported substantially efforts to understand climate change on the regional scale, partly to aid the Intergovernmental Panel on Climate Change (IPCC) in its periodic review of the science of climate change and its impacts on human society. Comments by Vary and Hakkarinen provided a broad context for the ensuing discussions.

INITIAL AND BOUNDARY CONDITIONS

The initial and boundary conditions for Experiment 1, particularly soil moisture, generated substantial discussion. The relatively short lengths of Exps. 1a and 1b (60 days) prevent models from spinning up their own soil moisture field, so their simulations may be sensitive to the initial condition, which is poorly observed. (Informal discussion at the workshop suggested that 90 days or more of simulation is necessary to spin-up soil moisture.) The experiment's initial soil moisture and atmospheric boundary conditions come from the reanalysis (Kalnay et al., 1997) produced by the U.S. National Centers for Environmental Prediction (NCEP) and National Center for Atmospheric Research (NCAR). John Roads showed that the NCEP/NCAR reanalysis soil moisture is damped toward a specified climatology and so may not capture fully the climatological extremes of Exp. 1. Anji Seth relayed a suggestion from Filippo Giorgi that PIRCS provide multiple sets of initial soil moisture for a suite of simulations that would reveal sensitivity to this specification. No definitive conclusion was reached at the workshop, but the PIRCS organizers will continue to discuss this suggestion by electronic mail with participants and possibly prepare alternative initial soil-moisture conditions. William Gutowski noted that one of the intents of PIRCS is to provide a common framework for sensitivity studies, such as the one suggested here, so that such simulation would be encouraged even if not formally part of Exp. 1. Finally, John McGregor pointed out that deep-soil temperature also needs to be included in the initial conditions.

The original PIRCS Exp. 1 boundary conditions specified monthly mean sea-surface temperature (SST). Michael Botzet showed how regional simulations could be sensitive to temporal evolution of SST, suggesting that SST should be updated from observations more frequently than monthly. Song-You Hong stated that the NCEP/NCAR reanalysis updates SST to daily intervals by interpolating weekly observational values. Participants generally agreed that while the simulation domain for Exp. 1 does not contain substantial ocean, PIRCS should nonetheless specify SST more frequently than monthly, in part to maintain consistency with the NCEP/NCAR atmospheric reanalysis output used for boundary conditions.

Further discussion of boundary conditions focused on specified surface properties. There was substantial disagreement over whether or not standards should be enforced for these and other specified features such as surface roughness and albedo. One group felt that adherence to such standards would reduce uncertainty over the causes of inter-model differences. Another felt that model surface specifications are determined in part for compatibility among all model parameterizations present and that the greater value of PIRCS simulations lies in comparing model performance with observations. Participants agreed, however, that PIRCS organizers should give maps of standard vegetation distributions and topography on the PIRCS Web site, both as references and to establish map contouring and coloring standards.

ARCHIVE OF MODEL OUTPUT AND OBSERVATIONS

Another area of substantial discussion was the standard output archive that PIRCS organizers will collect from the modelers and maintain. The workshop gave the modelers an opportunity to review this archive after some had run initial simulations. Special attention focused on output for comparison with observations and output for inter-model diagnosis of mesoscale behavior.

Participants recommended that the standard archive continue to emphasize diurnal cycles of energy and moisture fluxes at the surface. However, several consensus modifications emerged:

Surface and top-of-the-atmosphere fluxes should be archived as 3-h accumulated values (eight values per day) at all points on each model's PIRCS domain, to accommodate output storage restrictions faced by some.

Surface wind-stress should be similarly accumulated and saved.

Surface pressure, sea-level pressure, precipitable water, air temperature and skin temperature should be sampledevery three hours.

Convective and stable precipitation should be saved separately.

Soil (surface) heat flux is defined as the net heat flux from the atmosphere into the model domain below the atmosphere.

Runoff, accumulated over 24 h from 0 UTC, should be archived.

Archived soil moisture is the total amount of water in a gridbox's soil, sampled daily at 12 UTC.

Each model should give temperature, horizontal wind, specific humidity, geopotential height and pressure at all levels on its grid at 00, 06, 12 and 18 UTC. (This will allow diagnosis of the low-level jet without having to estimate a priori the location of its maximum strength.)

Each modeling group should provide the latitude and longitude of their model's gridpoints so that PIRCS can interpolate all ouput to a common 0.5 ° lat-lon grid.

Modeling groups are encouraged to submit a supplemental data set:

Hourly precipitation
Hourly atmospheric mixed-layer depth
Vertical "sounding" profiles at selected locations

All participants agreed that having a variety of observations available would be of great benefit to individual as well as collective efforts to understand model behavior. Comparison with observations will also be necessary for addressing the WGNE challenge described above. These observations will be collected by PIRCS. Summary plots of observations will be displayed on the PIRCS Web site (http://www.pircs.iastate.edu) to allow modelers an initial check of their output with observations and also to establish map contouring and coloring standards for comparative study. Observations will emphasize diurnal variations, especially those like precipitation that are closely linked to mesoscale behavior such as the nocturnal, low-level jet. Some specific observational fields include hourly precipitation from first-order stations, four-times daily NCEP and ECMWF analyses, daily maximum/minimum temperatures, and satellite observations of radiative fluxes.

Several models performed Exp. 1a before the workshop and participants informally compared their preliminary output, with several noteworthy features emerging. Models generally showed agreement with the observed time-average, large-scale circulation, though there were differences in details that were rather large at some points. Low-level jet output showed a diurnal variation with maximum strength at night. The degree of agreement in simulated precipitation between models and with observations varied greatly. Additional simulations and further diagnosis are clearly needed.

WRAP UP

Concluding discussions focused on the time table for Exp. 1a. Workshop participants agreed to the goal of completing the updated Exp. 1a and producing its output archive by the end of 1997. This archive will be the basis for one or two initial scientific papers. The end-of-1997 goal does not preclude later participation by other models. Participants also agreed that PIRCS should explore hosting a PIRCS participants review session in San Francisco that coincides with the December 1997 American Geophysical Union meeting.

Final discussions considered potential follow-on PIRCS experiments. Jens Hesselbjerg Christensen advocated a suite of simulations with relevant observations and model output that any person with a new model might run to assess model performance. Potential sites discussed included South-east Asia, southern Africa and South America. In this light, Clemente Tanajura presented some results from mesoscale simulation of the South American low-level jet.

REFERENCES

Kalnay, E., M. Kanamitsu, R. Kistler, W. Collins, D. Deaven, L. Gandin, M. Iredell, S. Saha, G. White, J. Woollen, Y. Zhu, M. Chelliah, W. Ebisuzaki, W. Higgins, J. Janowiak, K. C. Mo, C. Ropelewski, J. Wang, A. Leetma, R. Reynolds, R. Jenne and D. Joseph, 1996: The NCEP/NCAR 40-year reanalysis project. Bull. Amer. Meteor. Soc. 77, 437-471.

Takle, E. S., 1995: Project to Intercompare Regional Climate Simulations (PIRCS), Preliminary Workshop, 17-18 November 1994. Bull. Amer. Meteor. Soc. 76, 1625-1626.

Ray Arritt Department of Agronomy Iowa State University 3010 Agronomy Ames, IA 50011-1010 USA tel. (515) 294-9870 fax (515) 294-3163 email: rwarritt@iastate.edu	Sebastien Biner Département des sciences de la terre Université du Québec à Montréal C.P. 8888, Succursale "Centre-Ville" Montréal (Québec) H3C 3P8 CANADA tel. (514) 987-3000 #6813 fax (514) 987-7749 email: biner@phy.uqam.ca
Michael Botzet Max Planck Institute for Meteorology Bundesstrasse 55 D-20146 Hamburg GERMANY tel. +49-40-41173-374 fax +49-40-41173-298 email: botzet@dkrz.de	Daniel Caya Département des sciences de la terre Université du Québec à Montréal C.P. 8888, Succursale "Centre-Ville" Montréal (Québec) H3C 3P8 CANADA tel. (514) 987-3000 #3126 fax (514) 987-7749 email: caya@phy.uqam.ca
Jens Hesselbjerg Christensen Danish Meteorological Institute Lyngbyvej 100 DK-2100 Copenhagen DENMARK tel. +45-39157428 fax +45-39157460 email: jhc@dmi.min.dk	Dave Flory Department of Geol. & Atmos. Sciences Iowa State University 3010 Agronomy Ames, IA 50011-1010 USA tel. (515) 294-4758 fax (515) 294-3163 email: flory@iastate.edu
Bill Gutowski Department of Geol. & Atmos. Sciences Iowa State University 3021 Agronomy Ames, IA 50011-1010 USA tel. (515) 294-5632 fax (515) 294-3163 email: gutowski@iastate.edu	Chuck Hakkarinen Electric Power Research Inst. 3412 Hillview Avenue Palo Alto, CA 94304-1395 USA tel. (415) 855-2592 fax (415) 855-1069 email: chakk@eprinet.epri.com
Song-You Hong ECM/NCEP/NOAA 5200 Auth Road, Room 207 Camp Springs, MD 20746 USA tel. (301) 763-8000, ext. 7014 fax (301) 763-8545 email: songyou.hong@noaa.gov	Yimin Ji COLA 4322 Rowalt Drive #102 College Park, MD 20740 USA tel. (301) 902-1273 fax (301) 595-9793 email: ji@cola.iges.org
Jinwon Kim Lawrence Livermore Nat. Lab L-256, PO Box 808 Livermore, CA 94551 USA tel. (510) 422-1848 fax (510) 422-6388 email: kim1@LLNL.gov	Bill Lapenta Global Hydrology and Climate Center NASA Marshall Space Flight Center 977 Explorer Blvd. Huntsville, AL 35806 USA tel. (205) 922-5834 fax (205) 922-5723 email: bill.lapenta@msfc.nasa.gov
Glen Liston Department of Atmospheric Science Colorado State University Fort Collins, Colorado 80523-1371 USA tel. (970) 491-7473 fax (970) 491-8449 email: liston@iceberg.atmos.colostate.edu	John McGregor CSIRO, Div. of Atmospheric Research PB1 Aspendale Victoria 3195 AUSTRALIA tel.+61-3-9239-4520 fax +61-3-9239-4444 email: jlm@dar.csiro.au
Zekai Otles Department of Geol. & Atmos. Sciences Iowa State University 3010 Agronomy Ames, IA 50011-1010 USA tel. (515) 294-0264 fax (515) 294-3163 email: zekai@iastate.edu	Jan Paegle 819 Browning Building Meteorology Department University of Utah Salt Lake City, UT 84112 USA tel. (801) 581-7180 fax (801) 585-3681 email: jpaegle@atmos.met.utah.edu
Zaitao Pan Department of Geol. & Atmos. Sciences Iowa State University 3010 Agronomy Ames, IA 50011-1010 USA tel. (515) 294-0264 fax (515) 294-3163 email: panz@iastate.edu	Renato Ramos Da Silva Department of Geol. & Atmos. Sciences Iowa State University 3010 Agronomy Ames, IA 50011-1010 USA tel. (515) 294-2551 fax (515) 294-3163 email: renato@iastate.edu
John Roads Scripps Institution of Oceanography UCSD 0224 8605 La Jolla Shores Dr., NH 441 La Jolla, CA 92093-0224 USA tel. (619) 534-2099 fax (619) 534-8561 email: jroads@ucsd.edu	Moti Segal Department of Agronomy Iowa State University 3027 Agronomy Ames, IA 50011-1010 USA tel. (515) 294-2804 fax (515) 294-3163 email: segal@iastate.edu
Anji Seth Institute for Atmospheric Physics PAS Bldg 81 University of Arizona Tucson, AZ 85721 USA tel. (520) 621-6840 fax (520) 621-6833 msg. (520) 621-8836 email: seth@halo.atmo.arizona.edu	Andrew Staniforth Recherche en prévision numérique Environnement Canada 5e étage 2121 route Transcanadienne Dorval, P.Q. H9P 1J3 CANADA tel. (514) 421-4748 fax (514) 421-2106 email: Andrew.Staniforth@ec.gc.ca
Gene Takle Department of Geol. & Atmos. Sciences Iowa State University 3013 Agronomy Ames, IA 50011-1010 USA tel. (515) 294-9871 fax (515) 294-3163 email: gstakle@iastate.edu	Clemente A. S. Tanajura CPTEC/INPE Rodovia Presidente Dutra km 40 Cachoeira Paulista SP 12630-000 BRAZIL tel. +55-12-560-8479 fax +55-12-561-2835 email: cast@cptec.inpe.br
Rafael Terra UCLA, Department of Atmospheric Sciences Box 951565 Los Angeles, CA 90095-1565 USA tel. (310) 825-7407 fax (310) 206-5219 email: rterra@atmos.ucla.edu	Dennis Todey Department of Agronomy Iowa State University 3010 Agronomy Ames, IA 50011-1010 USA tel. (515) 294-4758 fax (515) 294-3163 email: dptodey@iastate.edu