In a recent article, Fosberg and Fujioka (1987) described the use of national

ECPC's U.S. Forecasts

J. Roads, S. -C. Chen, J. Ritchie

Experimental Climate Prediction Center Scripps Institution of Oceanography UCSD, 0224 La Jolla, CA 92093

1. ECPC's Global to Regional Atmospheric Forecast System

The Scripps Experimental Climate Prediction Center's (ECPC's) atmospheric forecast system was previously described by Roads et al. (1998, 1999b). At the largest space (global 200 km resolution) and time (monthly to seasonal) scales, we use the National Centers for Environmental Prediction's (NCEP's) medium range forecast (MRF) model or global spectral model (GSM; Kalnay et al. 1996; Roads et al. 1999a) and start these forecasts from the NCEP operational global analysis. The GSM then forces a regional spectral model (RSM; Juang and Kanamitsu, 1994; Juang et al. 1997; Chen et al. 1999, Anderson et al. 1999) in order to gain increased spatial resolution (50-25 km resolution) at shorter time scales (4xdaily to 4-weekly) for several selected regions (US, CA, SW). At even smaller space (2-km resolution) and time (8xdaily to daily) scales either the GSM or the RSM can force a corresponding nonhydrostatic mesoscale spectral model (MSM; Juang 1999; Roads et al. 1999c) for the Hawaiian Islands (Roads et al. 1999c). All atmospheric models are based upon the same physics used in the GSM and can be easily updated as the GSM is updated. Output products from the atmospheric models include a fire weather index (FWI, see Roads et al. 1997) and associated variables such as 2m-temperature, relative humidity and 10m windspeed as well as precipitation and soil moisture for the globe and for several selected regions (e.g. US). Since the global and regional atmospheric models are now forcing ocean and land surface models, we are also beginning to display surface water and energy forcings in conjunction with ocean and land forecasts. Forecasts of these forcings and corresponding output from the ocean and land models will be presented in later ELLFB issues.

2. GSM Forecast Skill Evaluations

The current background climatology used to calculate anomalies for the analysis and forecasts comes from the NCEP reanalysis (Kalnay et al. 1996) and does not take into account the significant biases that these models produce. Separating out the systematic biases as well as evaluating the true forecast skill from the limited numbers of forecasts is frustrating and we have had to be patient in developing comprehensive evaluations. Nonetheless, as discussed by Roads et al. (1998), still limited forecast samples suggest that the GSM provides more skillful forecasts of temperature, precipitation, soil moisture, and fire weather index than persistence, even at long forecast ranges. Although the greatest skill occurs initially and then decays toward zero, daily, weekly or monthly forecast skill does not ever reach zero and forecasts averaged into monthly and seasonal averages demonstrate small but significant skill, which may be comparable to other long-range forecast methodologies.

3. May Forecasts, May Analysis, June Forecasts

Fig. 1 compares the GSM monthly forecast for May (top panel) of the FWI with the May FWI calculated from 4xdaily analysis (middle panel). Note that the northern US and Canadian fire weather index anomaly was successfully forecast. However, the higher FWI over the Southwest and Mexico that emerged during the month was not. There are a few systematic biases like this that we are still trying to correct in the model. This Mexican FWI anomaly is forecast to continue (bottom panel) in June along with the Canadian anomaly. The western states, especially California are showing reduced danger, due in part to higher humidities associated with the increased precipitation being forecast.

Fig. 2 compares the GSM monthly forecast for May (top panel) of precipitation with the May precipitation calculated from 4xdaily analysis (middle panel). There were some regions such as the northeast, South Atlantic coast, and northwest that were forecast well. Note that the dry patterns in northeast Mexico and Texas-Arkansas area are forecast to continue. The biggest forecast problem was the western slope of the Rocky Mountains and northern Mexico, where the precipitation was opposite of what occurred. Since we are forecasting a continuation of this pattern (bottom panel), we should be more wary of forecasts in these regions. Again, we are trying to correct systmatic biases like this.

Fig. 3 compares the GSM monthly forecast for May (top panel) of surface water (snow plus total soil moisture) with the May surface water calculated from 4xdaily analysis (middle panel). The pattern was forecast well because surface water is very persistent. This persistence is best illustrated by the similarity of the forecast pattern from May to June, which shows dry conditions in the northern US and Canada and wet conditions in the Southern US and Mexico. Note the wet surface water conditions in Mexico are drying, consistent with the decreased precipitation being forecast in these regions.

Fig. 4 compares the GSM monthly forecast for May (top panel) temperature with the May temperature calculated from 4xdaily analysis (middle panel). The general pattern of positive anomalies in the northern US and Canada with negative temperature anomalies in the southern US and Mexico was forecast although there is a noticeable positive bias in the Canadian temperatures. The largest changes for June occur over the West Coast where the temperature is forecast to be below normal. This temperature forecast is consistent with the increased precipitation, surface water and decreased FWI anomaly being forecast for June.

References

Anderson, B.T., J. O. Roads, S. -C. Chen, and H. -M. Huang, 1999: Regional Modeling of the Low-level Monsoon Winds Over the Gulf of California and Southwest United States: Simulation and Validation, (submitted).

Chen, S. -C., J. O. Roads, H. H. -M. Juang, and M. Kanamitsu, 1999 Global to Regional Simulations of California Wintertime Precipitation. J. Geophys. Res. (in press, special precipitation issue)

Juang, H. -M. H., and M. Kanamitsu, 1994: The NMC nested regional spectral model. Mon. Wea. Rev., 122, 3-26.

Juang, H.-M. H., S.-Y. Hong and M. Kanamitsu, 1997: The NCEP regional spectral model: an update. Bulletin Amer. Meteor. Soc., 78, 2125-2143.

Juang, H. -M. H., 1999: The EMC/NCEP mesoscale spectral model: A revised version of the nonhydrostatic regional spectral model. Mon. Wea. Rev., submitted.

Kalnay, E. et al., 1996: The NMC/NCAR reanalysis project, Bull. Am. Meteor. Soc., 77, 437- 471, 1996.

Roads, J.O., S. -C. Chen, F.M. Fujioka, H. Juang, and M. Kanamitsu. 1997. Global to Regional Fire Weather Forecasts. Int. Forest Fire News, 33-37.

Roads, J., Chen, S. -C., Ritchie, J., 1998: Evaluation of the Experimental Climate Prediction Center's global to regional and daily to seasonal prediction system. Proceedings of the 23^rd Annual Climate Diagnostics Meeting. Miami, Florida

Roads, J. O., S. -C. Chen, M. Kanamitsu, H. Juang, 1999a: Surface Water Characteristics in NCEP's Reanalysis and Global Spectral Model. J. Geophys. Res.-Atmos. (in press, special GCIP Issue)

Roads, J., S. -C. Chen, J. Ritchie, 1999b: ECPC's Weekly to Seasonal U.S. Forecasts of FWI, Soil Moisture, and Precipitation. ELLFB bulletin, Mar. 1999

Roads, J., S. Chen, D. Stevens, C. McCord, H. Juang, F. Fujioka, 1999c: Weather and climate analyses and forecasts at MHPCC. Seventh International Conference on High Performance Computing and Networking. Amsterdam, The Netherlands, April 12-14, 1999.

Fig. 1 FWI (dimensionless). Upper panel, 1 month forecast for May 1999. Middle panel, analysis for May 1999. Lower panel, 1 month forecast for June 1999.

Fig. 2 Precipitation (mm/day) and moisture flux anomalies. Upper panel, 1 month forecast for May 1999. Middle panel, analysis for May 1999. Lower panel, 1 month forecast for June 1999.

Fig. 3 Soil Moisture anomaly (mm). Upper panel, 1 month forecast for May 1999. Middle panel, analysis for May 1999. Lower panel, 1 month forecast for June 1999.

Fig. 4 Temperature anomaly (K). Upper panel, 1 month forecast for May 1999. Middle panel, analysis for May 1999. Lower panel, 1 month forecast for June 1999.