Overview

Statistical postprocessing is a method of improving on numerical weather models' ability to forecast by relating model outputs to observational or additional model data using statistical methods. MDL currently uses a variety of techniques, most notably Model Output Statistics (MOS). These methods can be used to produce products from hourly to climatological time scales, but are most commonly produced either hourly (LAMP) or multiple times a day(Station-based MOS, Gridded MOS).

MDL Statistical Postprocessing Products

MOS (Model Output Statistics)

Station-based MOS

MOS forecast guidance products are developed by relating NWP model data to observations using statistical methods. Station-based MOS provides general guidance based at meteorological observation stations.

There are currently MOS products available based on two NWP models:

GO TO: GFS MOS GO TO: NAM MOS

Gridded MOS (GMOS)

GMOS forecast guidance products are developed by relating NWP model data to observations using statistical methods. GMOS provides guidance according to data on a geographic grid.                                 

GO TO: GRIDDED MOS

Ensemble Kernel Density MOS (EKDMOS)

Ensemble Kernel Density MOS (EKDMOS) uses output from the North American Ensemble Forecast System (NAEFS) to produce reliable and accurate probability distributions of sensible weather elements. Grids are generated twice daily, based on the 0000 and 1200 UTC runs of the NAEFS. To express the probability distribution, grids are produced for a number of percentiles as well as for the mean forecast.

GO TO: EKDMOS

NBM (National Blend of Models)

The National Blend of Models (NBM) is a nationally consistent and skillful suite of calibrated forecast guidance based on a blend of both NWS and non-NWS numerical weather prediction model data and post-processed model guidance.

GO TO: NBM Home Page
 

LAMP (Localized Aviation MOS Program)

Station-based LAMP

LAMP produces forecast guidance products developed by relating Model Output Statistics (MOS) forecast guidance to observations using statistical methods. LAMP provides hourly aviation station-based forecast products that cover a 25-hour period.

GO TO: Station-based LAMP

Gridded LAMP (GLMP)

Gridded LAMP (GLMP) provides gridded analyses of observations and operational LAMP forecasts for aviation forecasting on the 2.5-km CONUS NDFD grid.  The GLMP grids are produced hourly and cover a 25-hour period.                                                                                  

GO TO: Gridded LAMP