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

1. ECPC Experimental Forecast System

The ECPC previously used the reanalysis I version (Kalnay et al. 1996) of the National Centers for Environmental Prediction’s (NCEP’s) global spectral model (GSM; Roads et al. 2001) along with the corresponding NCEP Regional Spectral Model (RSM; Juang et al. 1997; Roads 2003, 2004) to make routine experimental global to regional forecasts (daily out to 7days and weekly out to 16-weeks) starting from the NCEP operational 00UTC global analysis and using persisted SST anomalies (+climatology) as boundary conditions. These forecasts are routinely posted at http://ecpc.ucsd.edu/m2s/m2s_ECPC_forecasts.html. The GSM is now being replaced by an updated version of NCEP's seasonal forecast model (SFM; Kanamitsu et al. 2002a), which is based on updated physics from the NCEP/DOE reanalysis II (Kanamitsu et al. 2002b). The SFM has a nominal (a reduced grid technique is used near the poles) horizontal resolution of T62 (about 2^o). There are 28 levels in the vertical sigma coordinate system. A number of theoretical predictability experiments have been performed with this later generation GSM (Reichler and Roads 2003, 2004). The new RSM will have the same number of vertical levels but increased resolution for the US (50 kms) and CA/NV (25kms).

The SFM also now provides an augmented ensemble set of ECPC seasonal forecasts. Starting from slightly perturbed initial conditions, and forced with observed SST anomalies, 10 simulations are made up to present. Then, an ensemble of forecast SSTs are used to generate a 7-month forecast ensemble. The forecast SSTs come from simplified models (3) for the tropical Pacific and are produced by the International Research Institute (IRI); four 7-month forecasts are produced for each IRI SST prediction. This new SFM is now being coupled to the MIT ocean model and sometime in the future we hope to demonstrate that our coupled system will be as useful as current persisted or forecast SSTs as well as our current uncoupled experimental ocean forecasts, which use forecast GSM anomalies, and may eventually contribute to a community coupled model forecast. A major advantage of the SFM over the GSM is that the computer code of the SFM was completely rewritten to run on massively parallel processor (MPP) machines using Message Passing Interface (MPI) routines. Normally the SFM runs on 64 LINUX processors and takes 2 hours to make a 7-month forecast. Depending upon the number of ensemble members, a normal 7-month forecast takes between 1-2 days. During the rest of the month background runs are being made to augment the growing ensemble climatology. In addition, as changes are made in the model new climatologies have to be developed. It should also be noted that there are a few physical parameterization differences between the ECPC SFM and the NCEP Coupled Forecast System (CFS). In particular, the ECPC SFM has an updated land surface model (Noah).

2. NCEP CFS/RSM forecasts

In addition to using and applying forecasts from the experimental ECPC forecasts, we are also downloading NCEP CFS/RSM 7 month forecasts for applications to fire danger forecasting. At NCEP the operational CFS currently starts from the operational atmospheric and ocean analysis and uses a coupled model, Global Forecast System (GFS) for the atmosphere and Modular Ocean Model (MOM3) for ocean, for the coupled seasonal forecast. Corresponding climatology runs start from the NCEP/DOE atmospheric reanalysis II (Kanamitsu et al. 2002) and MOM3 ocean analyses.

NCEP is now nesting the US RSM for the regional seasonal forecast. The RSM was originally developed to emulate the global model but operate at regional scales and is now in operational use for the 10 km RSM daily weather forecast for Hawaii and as a contributor to the 48 km RSM short-range ensemble forecast for the CONUS. A newer version of the RSM has since been developed at NCEP to more closely emulate the new Climate Forecast System physics at NCEP and is similar to the ECPC RSM. The NCEP regional model US domain covers the CONUS and its vicinity, from 130W to 65W and 20N to 55N with 50 km resolution. The CFS/RSM currently outputs binary restart files as well as GRIB files on pressure surfaces using the model grid every 6 hours. The standard output has been enhanced in order to develop the needed input for danger forecasts. All output, including binary restart files and GRIB files, are grouped together by file type and stored in the IBM HPSS mass storage site. A rotating archive is accessed by ECPC to drive the fire danger code, described below.

The new 7-month NCEP/ECPC CFS/RSM forecast archive began in Oct. 2004. NCEP actually re-runs the global and regional forecasts at the same time since this provides the best flexibility for future implementation and application of this experimental product. Each month, three hindcasts per year are made from 1982 to 2004, which provides a total of 23x3 members of hindcasts to develop the model climatology. Ten forecasts starting on 5 different days at 0000 and 1200 UTC are also made as part of the 10-member ensemble forecast. More hindcasts may be added later in order to construct more-stable model climatology. The number of ensemble members is dependent upon available computer time. In addition, a continuous 1-day forecast run from Jan. 1982-present has now been developed. This run, along with observed precipitation, was required to initialize the fire danger code. This initialized fire danger code also serves as the validation (see Roads et al. 2005; Reinbold et al. 2005).

3. National Fire Danger Rating System

The NFDRS (Burgan 1988) indices describe characteristics of fire danger, given the conditions of fuel, topography, and weather. The basic inputs to the NFDRS include precipitation, temperature, relative humidity, cloud cover and wind speed as well as fuels and slope. The standard weather input to the NFDRS comes from weather station data, which is assumed to apply to a large (~10³ hectares) area surrounding each weather station; vegetation (fuel) types and slope are also defined for each weather station and assumed to apply to the same surrounding area. The Wildland Fire Assessment System (http://www.wfas.us/content/view/16/31/) constructs fire danger maps based on either observed or forecast data for a weather station. The major difference in NFDRS calculations at ECPC is that here gridded fuels, weather forecasts and topography data are used. The fuels and orography (slope) data were initially defined at 1km spatial resolution and then the nearest 1km grid point was used for the NFDRS 25km grid. The observed precipitation (25 kms) and forecast model output (50 kms) are similarly interpolated to the NFDRS grid and output is subsequently interpolated back to the RSM grid, which is used for all evaluations.

NFDRS gridded fuels and slopes are also used for these calculations. Slope is important in assessing fire danger because fire generally burns faster spreading upslope than on flat ground. Vegetation type, quantity and structure are also important for describing fire danger. Sixteen of the twenty NFDRS fuel models are being used to represent the vegetation types across the U.S., (Burgan 1988) defining fuel characteristics such as depth, load by live and dead classes, heat content, fuel particle size, etc. Because the variability of fuel bed characteristics is infinite, each fuel model in the fire danger rating system must necessarily represent a rather broad range of vegetation types. The basic vegetation data source was the 1km resolution land cover map released in 1991 by the EROS Data Center. The LC map was converted to an NFDR fuel model map through a combination of 2546 ground sample plots scattered across the U.S., and consultation with fire managers from across the country.

Roads et al. (2005) and Reinbold et al. (2005) provide a summary description of the NFDRS fire danger variables being examined here.

4. ECPC SFM Seasonal Forecasts

Above normal temperatures (Fig. 1) are being forecast by the ECPC SFM for most of the US and Mexico, especially over the Northeast during the spring to early summer. There is forecast to be some tapering toward climatology in the later summer months and even below normal anomalies are being forecast over the mountain states in the late summer. This is in contrast to the initially below normal temperatures being forecast for Northwest Canada and Alaska, which also forecast to become more moderate during the summer, except for central Canada, where the cool weather is forecast to continue.

Except for the Great Lakes region, precipitation (Fig. 2) is forecast to be mostly below normal across the US initially, although the Monsoon is being forecast to provide above normal rainfall this summer in the Southwest, especially during the late summer.

The 500 mb height forecast anomalies associated with these forecast precipitation and temperature anomalies (Fig. 3) indicates that there are high pressure centers being forecast over the East Coast and Western Pacific Ocean, along with a low pressure over Northwest Canada. The intensity of these centers decreases during the summer.

5. Fire Danger Forecasts

The ECPC/NCEP fire danger forecasts, which are initialized from the continuous validating Fire Danger Code, consist of 6 indices: (1) Fosberg Fire Weather Index, FWI (not a part of the NFDRS); (2) Burning Index, BI; (3) Ignition Component, IC; (4) Energy Release, ER; (5) Spread Component, SC; and (6) the Keetch/Byram, KB, drought index. Further details about these indices can be found in Roads et al. (2005) and references therein.

Related to the decreased precipitation being forecast by the ECPC global model ensemble during the summer monsoon, almost all of the firedanger indices are indicating above normal fire danger in the Southwest and Southeast, especially Florida. Above normal ER (Fig. 7) indices are being forecast on the Northwest Coast and Eastern slopes of the Rocky Mountains, which is somewhat consistent with the FWI (Fig. 4), Burning Index (Fig. 5), and ignition component (Fig. 6). The SC (Fig. 8) and KB drought index (Fig. 9) are less consistent over the northwest coast and northeast; however, all of the indices (Figs. 4-9) do indicate that the greatest seasonal forecast danger is appearing over the Southwest.

References:

Burgan, Robert E. 1988. 1988 Revisions to the 1978 National Fire-Danger Rating System. Res. Pap. SE-273. Asheville, NC: U.S. Department of Agriculture, Forest Service, Southeastern Forest Experiment Station. 39 pp.

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.

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

Kanamitsu, M., A. Kumar, H.-M. H. Juang, W. Wang, F. Yang, J. Schemm, S.-Y. Hong, P. Peng, W. Chen and M. Ji, 2002a: NCEP Dynamical Seasonal Forecast System 2000. Bull. Amer. Met. Soc., 83, 1019-1037.

Kanamitsu, M., W. Ebisuzaki, J. Woolen, J. Potter and M. Fiorino, 2002b: NCEP/DOE AMIP-II Reanalysis (R-2). Bull. Amer. Met. Soc. 83, 1631-1643.

Reichler, T. J. and J. O. Roads, 2003: The Role of Boundary and Initial Conditions for Dynamical Seasonal Predictability. Nonlinear Processes in Geophysics, 10 (3) 1-22.

Reichler, T. and J. Roads, 2004: Time-space distribution of long-range atmospheric predictability. J. Atmos. Sci., 61 (3), 249-263

Reinbold, H, J. O. Roads, T. Brown, 2005: Evaluation of the Experimental Climate Prediction Center’s fire danger forecasts with remote automated weather station observations. International Journal of Wildland Fire, 14, 19-36.

Roads, J.O., S-C. Chen and F. Fujioka, 2001: ECPC’s Weekly to Seasonal Global Forecasts. Bull. Amer. Meteor. Soc., 82, 639-658.

Roads, J.O., S. -C. Chen, S. Cocke, L. Druyan, M. Fulakeza, T. LaRow, P. Lonergan, J. Qian and S. Zebiak, 2003: The IRI/ARCs Regional Model Intercomparison Over South America. J. Geophys. Res., 108 (D14), 4425, doi:10.1029/2002JD003201.

Roads, J. 2004: Experimental Weekly to Seasonal U.S. Forecasts with the Regional Spectral Model. Bulletin of the American Meteorological Society 85(12) Dec 2004.

Roads, J., J. Ritchie, F. Fujioka, R. Burgan, 2005: Seasonal Fire Danger Forecasts for the USA. International Journal of Wildland Fire, Special Issue: Fire and Forest Meteorology, 14, 1-18.