Who Are We?

The RMRS Rocky Mountain Center (RMC) is part of the Wildland Fire Management RD&A Program. The RMC Team is engaged in development and deployment of science-based computer applications for real-time delivery of high-resolution fire-weather intelligence and smoke dispersion forecasts over the Continental USA. Our vision is to provide comprehensive weather support to wildland fire operations, prescribed burns, and air resource management.

Our Mission

The strategic goal of RMC is to:
• Help accomplish agency prescribed burning targets;
• Help burn safer;
• Help burn cleaner and avoid air quality problems;
• Help combat wildfire with relevant weather intelligence;
• Facilitate rapid implementation of new weather information products;
Our long-term vision is to solve problems of fire and smoke management by providing regional simulations of weather and weather-dependent phenomena including fire danger, fire behavior, and smoke dispersion.

Operational Products

We deliver real-time spatially explicit hourly information about current and future values of weather elements such as air temperature, relative humidity, precipitation, cloud cover, wind speed and wind direction, as well as several fire indices (i.e. Ventilation, Haines, Fosberg, and Ketch-Byrum). Weather forecasts and analysis are performed over the Conterminous USA at 8-km and 15-km resolution. Forecasts extend up to 240 hours into the future. Web products are produced for the Main Domains, and numerous sub-domains (windows) covering territorial units managed by the GACC offices as well as many individual states.

In addition, we provide real-time smoke dispersion forecasts for the entire Conterminous USA using the BlueSky Modeling System. We have recently expanded our list of operational products to include maps of crucial indices from the National Fire Danger Rating System (NFDRS). Our information products are delivered to consortium members and the general public via the World Wide Web in the form of interactive 2-D maps, point observations & point forecasts, and actual 3-D data fields.


Conterminous USA

  • Model Version:

  WRF and MFF Models

   MM5 Version 3.6.3

  • Horizontal Grid:

   448x448 12-Km grid

  180x100 18-Km

  • Vertical levels:

  41 levels
  41 levels

  • Lowest model level:


  •Initialization Time:

  6Z & 18Z;
  NOAA FSL LAPS Analysis
(6-Km Grid) 
  00Z & 12Z;
  Eta Analysis
(40-Km CONUS Grid)

  • Δ T:

  18 seconds
  54 seconds
  • Model Physics: 
  Reisner mixed phase including
graupel (Reisner2) microphysics

  Reisner mixed phase including
graupel (Reisner2) microphysics

  • PBL Scheme:

  • Radiation:

  RRTM scheme (longwave),
cloud-radiation scheme (shortwave)

  RRTM scheme (longwave),
  cloud-radiation scheme (shortwave) 

  • Surface Scheme:

  NOAH Land-Surface Model

  NOAH Land-Surface Model

  • Boundary Conditions:
  NCEP Eta forecast (6hr interval)

  NCEP Eta forecast (6hr interval)

  • Cu-Parameterization:


  Grell scheme

Method and Tools

Analysis of observed (past) weather are performed using the Local Analysis and Prediction System (LAPS) supported by the NOAA Forecast Systems Laboratory in Boulder, CO. Weather forecasts are generated by the latest version of the MM5 Community Model jointly developed by the Pennsylvania State University (PSU) and the National Center for Atmospheric Research (NCAR), and the new MFF Model (a coupling between MM5 and the FORFLUX surface biophysical ecosystem model). MM5 is a limited-area, non hydrostatic, terrain-following sigma-coordinate model designed to simulate 3-D mesoscale atmospheric circulation.

LAPS assimilates thousands of meteorological observations every hour from land surface and the upper air (including data from satellite and Doppler radar stations) as well as large-scale model output, and interpolates them to a continuous 3-D grid of certain resolution. Data interpolation employs statistical methods as well as algorithms based on laws of atmospheric physics. The analysis produces a consistent mass-balanced 3-D data fields of cloud cover, temperature, water vapor, wind speed and wind direction. These fields are passed to the MM5 and MFF models that perform temporal integration and produce the weather forecasts. We use MM5 version 3.6.3. with Reisner mixed phase microphysics (Reisner2) including graupel, the MRF PBL scheme.

WRF and MFF models are run twice per day for the two main domains. Model initialization times are 11:00 and 23:00 MST for the Continental US domain, and 8:00 and 20:00 for the Western US domain. The fields of observed weather generated by LAPS provide initial conditions to the forecast models. Boundary conditions are extracted from the NCEP NAM model. The LAPS analysis offer a unique model initialization for all domains by specifying mass-balanced 3-D fields of actual cloud cover and precipitable water content during the first hour of simulation. Known as a Hot Start, this method eliminates the need for model spin-up to achieve diabatic initialization, and allows the production of a reliable forecast from the commence of model simulation.

WRF and MFF run on a 176 CPU Core Parallel Super-computer running Linux CentOS. Forecasts of smoke dispersion are produced by the BlueSky Modeling System through a collaboration with the Northwest Regional Modeling Consortium. BlueSky requires forecast meteorological fields for input that are provided by WRF and MFF model runs.

Webmaster: Ned Nikolov (nnikolov02@fs.fed.us)