Technology

  • Modeling
  • Auto-Nowcasting
  • Tools
  • Hardware

For numerical weather prediction (NWP), the Army ranges use the Four-Dimensional Weather System (4DWX), the product of extended collaboration between NCAR’s Research Applications Laboratory (RAL) and the Army Test and Evaluation Command (ATEC). 4DWX the WRF Model as its predictive core, and assimilates observations via the RT-FDDA scheme.

RT-FDDA: Real-Time Four-Dimensional Data Assimilation

The RT-FDDA scheme assimilates into 4DWX a multitude of observations from commonly available and specialized sources.  RT-FDDA involves modifications to an NWP model’s predictive equations, which permit the model to be gently adjusted, or nudged, toward observed conditions during the analysis stage.  The scheme is computationally efficient and preserves the precise timing of observations, which gives 4DWX a more accurate depiction of the weather at any instant.  New analyses and multi-day forecasts are computed every three hours.

3DVAR: Three-Dimensional Variational Data Assimilation

An alternative method of data assimilation, 3DVAR, is currently being tested and refined at the mesoscale.  This method incorporates a number of non-standard observations that cannot be included in RT-FDDA's observation-nudging scheme, such as satellite radiance, GPS, and radar.  3DVAR is currently being integrated into 4DWX, producing model analyses that will take advantage of a wealth of remotely sensed observations.

WRF Model: Weather Research and Forecasting Model

4DWX’s primary model core is the WRF Model, the current state of the art in mesoscale modeling for both operations and research.  It is a fully compressible, Eulerian, nonhydrostatic NWP model with a large suite of options for numerical schemes and parameterizations of physical processes.  As a community model, it benefits from countless contributions by the world’s leaders in weather modeling.  RAL takes the public version of the code and further modifies it for the specific needs of the Army ranges.  Although the MM5 is still used, the WRF Model is the current model core of choice for 4DWX.

Coupled Applications

NWP models provide forecasters with weather predictions, but substantial value can be added to such predictions by diagnosing how the weather affects other processes and conditions, such as sound propagation and the trajectories of missiles.  To achieve this added value, RAL couples analyses and forecasts to secondary models, sometimes called coupled applications.

Noise Assessment and Prediction Capability (NAPS)
Developed by the Army Research Laboratory (ARL), NAPS is used to predict sound levels generated during the testing of explosive materiel at the Aberdeen Test Center, MD.

Second order Closure Integrated Puff (SCIPUFF)
Developed by Titan System Corporation is part of the Defense Threat Reduction Agency's (DTRA) Hazard Prediction and Assessment Capability (HPAC), SCIPUFF is fed real-time model data in the MEDOC format, and predicts plume concentration and dosage based on the characteristics of a material and its release, as specified by the user.

Lewis Model
Developed by White Sands Missile Range (WSMR) personnel, the Lewis model is a simple missile trajectory model with empirical fits to rocket-specific performance data which allows rapid computation of missile downrange and crossrange impact locations given input measured or forecasted winds.

Auto-Nowcasting

The primary role of the nowcast environment is to collect weather data; execute algorithms for producing a combined thunderstorm forecast; and provide a graphical display tool for viewing the various datasets.

Data incorporated into the nowcast environment includes WSR-88D (NEXRAD) radar, Goes satellite, sounding, and surface data.

The software applications in the nowcast environment include algorithms for identifying and tracking thunderstorm movement, identifying boundaries, wind retrieval from radar, as well as a fuzzy-logic engine which allows the user to combine the weighted outputs from the various algorithms to produce a single, combined forcast. Subsequent verification of generated forecasts are available both visually and statistically.

In realtime operations the nowcast environment initiates and maintains process control through an auto-restart mechanism. The auto-restarter provides the user with an automated, hands-free forecasting environment.

Many components of the nowcast environment also can be operated in archive mode. Archive mode allows the user to re-run algorithms with newly specified parameters and to view historical datasets.

The key hardware components:

  • A collection of Linux workstations on a local area network
  • A tape drive for archiving data
  • The key software components:

  • Data ingest and distribution
  • control mechanism
  • Analysis algorithms
  • Graphical data displays
  • The key data components:

  • Doppler radar data
  • Sounding data
  • Mesonet data
  • Goes satellite data
  • The "main engines" of the 4DWX system are the modeling and auto-nowcasting capabilities, but a wide range of tools accompany a typical 4DWX system implementation. These provide the user with clear interfaces to the underlying data, the configuration of the various models and their output products, and visualization of observations and model results.

    Web Portal

    A Web Portal is a web site comprised of a collection of "portlets". Each portlet in the new 4DWX Web Portal offers different tools and content related to 4DWX.

    Weather Warning Tool (WWT)

    The Weather Warning Tool (WWT) provides a GUI interface for the creation and dissemination of textual and graphical weather warnings.

    Met Alert Tool (MAT)

    The Met Alert Tool monitors current weather observations at specific range sites and issues alerts when observed weather conditions fall outside of a certain threshold.

    Multiple Radar Regional Display (MRRD)

    The Multiple Regional Radar Display (MRRD) tool allows users to display radar, surface, and lightning data from several radars in the vicinity of the installation.

    MIR Database

    The MIR database is a redesign of the previous WXDB database that works in tandem with ATEC's ARMADA database for storing observational data.

    UGUI: Model Image View

    UGUI is a convenient and customizable web-based tool for viewing real-time model output.

    NCAR has partnered with Dell to provide hardware systems for running the models and other 4DWX subsystems:

    The Data Application System

    The Data Application System. The Data Application System (DAS) is a data-collection, data-repository platform, which also includes tools for data visualization and analysis of output from the 4DWX System.

    The Model Application Cluster

    The Model Application Cluster (MAC) runs the WRF numerical weather prediction model to create RTFDDA forecasts. The MAC sends resultant imagery to the range DAS systems.

    The High Performance Cluster

    The High Performance Cluster (HPC) is a system purchased by the US Army HPCMO for Dugway that has been used in conjunction with NCAR to create RTFDDA ensemble forecasts.