Approximately 1.57 million accidents occur each year in poor weather on the nation's roads and an estimated 713,000 injuries and 7,400 deaths per year (based on an 8-year average) occur, creating an annual economic toll of approximately $42B. Weather plays a role in about 28% of the total crashes and 19% of the total fatalities. Weather also reduces capacity and significantly impacts efficiency, triggering congestion, particularly on roads operating near capacity. Approximately 15% of all congestion is due to poor weather and related road conditions. Since 1998 RAL has had a leadership role in the Road Weather Research and Development Program within the USDOT’s, Federal Highway Administration, bringing the surface transportation and weather communities together to improve surface transportation safety and mobility. RAL continues to provide national leadership in surface transportation weather by organizing and participating in national and international surface transportation weather workshops, conferences, training programs, and committees.
Maintenance Decision Support System (MDSS):
Since 1999, RAL has led a team of national laboratories in the development of the prototype winter Maintenance Decision Support System (MDSS), a unique decision support system that provides real-time snow and ice control guidance (e.g., treatment times, chemical choices, rates, and locations) for user-defined roadway segments. In FY07 the modeling, pavement (bridge and roadway) heat balance models, data fusion system, and rules of practice components of the MDSS were enhanced to improve the overall performance of the system. Other highlights included working with Environment Canada to improve and integrate their pavement heat balance model called METRo into the MDSS system; disseminating MDSS Version-4.0 software to over 60 road weather organizations (public and private); and successfully supporting the FHWA’s annual MDSS stakeholder meeting, which included more than 80 participants from the surface transportation community including 30 State DOTs.
The prototype MDSS was utilized in real-time by the City and County of Denver and the E-470 Public Highway Authority during the particularly severe winter of 2006-2007. A commercial version of the MDSS is being developed and demonstrated as part of a DOT Pooled Fund Research Program. The Pooled Fund MDSS was evaluated by 11 DOTs last winter and another demonstration is planned for the winter of 2007-2008.
Vehicle Infrastructure Integration (VII):
The USDOT has embarked on a relatively new program with the automotive and consumer electronics industries to develop a national capability to support wireless communications for vehicles. The VII Program is working toward deployment of advanced vehicle to vehicle and vehicle to infrastructure communications that could keep vehicles from leaving the road and enhance their safe movement through intersections. RAL is working closely with the FHWA Road Weather Management Program to develop concepts and assess the feasibility of utilizing vehicle data to enhance the diagnosis and prediction of weather and road conditions along our nation’s roadway network. In 2007, RAL prepared and delivered to the FHWA a major vision document titled Weather Applications and Products Enabled Through Vehicle Infrastructure Integration (VII). This document discussed how probe data from millions of vehicles could be used in the future to support the diagnosis and short term prediction of weather.
MDSS: The MDSS will continue to be developed and validated in 2008 using Colorado as a test bed. Prototype MDSS products will be provided to the E-470 Public Road Authority and the City and County of Denver. We anticipate that Denver International Airport may also want to participate in the MDSS demonstration to evaluate the system’s capabilities for supporting snow and ice control operations for ramp, runways, and Pena Boulevard. RAL will also work with the Utah DOT to configure and transfer the prototype MDSS to Utah DOT as part of a evaluation project to assess the portability of the prototype MDSS to state DOTs.
Clarus Initiative: RAL will continue to support the Aurora Program in the development of new weather and road condition hazard applications for travelers and will provide input and feedback to the FHWA on the development of the Nationwide Surface Transportation Weather Detection and Forecast System Initiative known as the Clarus Initiative
Vehicle Infrastructure Integration (VII): In FY2008, RAL will participate in the VII testbed near Detroit, Michigan. RAL will obtain data from test vehicles and begin to analyze their characteristics including quality, density, and geographical and temporal distribution. RAL will also begin the design and development of a Weather Data Translator (WDT) that will be used to parse, ingest, process, quality control, and generate advanced weather and road condition analyses utilizing vehicle probe data. This will be a significant scientific and engineering activity that will eventually facilitate the collection and processing of millions of new surface weather observations.back to top
Graphic showing the general size of the new Weather Research and Forecasting (WRF) Model domains that became operational in the AOAWS in October 2007. The WRF model replaced MM5 as the Civil Aeronautics Administration’s (CAA) aviation weather prediction system. The WRF model is now being run on the Central Weather Bureau’s (CWB) IBM supercomputer. The WRF runs in a nested configuration with a model grid spacing of 45, 15, and 5 km.
Since 1998 RAL and MMM have collaborated in the development of an Advanced Operational Aviation Weather System (AOAWS) for the Civil Aeronautics Administration (CAA) of Taiwan. The initial AOAWS project (AOAWS-I) was completed and deployed for operational use in June 2002 and provides the CAA, the airlines, and the flying public with state-of-the-art aviation weather technology to detect and forecast hazardous weather phenomena that affect aviation operations, airspace efficiency, and capacity and safety at Taiwan’s major hub airports.
In January 2005, the AOAWS Project entered a second five-year phase. This new phase focuses on the replacement of the MM5 model with the WRF model, advancing the data assimilation system (WRF-VAR) to incorporate new data types (e.g., COSMIC GPS, and other satellite data), upgrading the icing and turbulence products to incorporate the latest advancements developed as part of the FAA Aviation Weather Research Program, and developing and implementing new JAVA based display systems.
The AOAWS Program had several major accomplishments this year including the development and introduction of the WRF model as the primary model used by both the CWB and CAA to support general weather prediction in Taiwan and aviation weather forecasting, respectively. The transition from the MM5 to the WRF was no small task as it required the CWB to adopt this new model as its primary mesoscale model, and it required the CAA to agree to upgrade and reconfigure the entire AOAWS to accept the new model data. NCAR successfully completed the transition to the WRF-based system in September 2007 and the system was declared ready for operational use by the CAA in mid-October.
The upgraded AOAWS system includes the WRF model and the WRF-VAR data assimilation system, upgraded MDS, model display, and SMD, and the new JMDS. System enhancements also include refinements to the in-flight icing and turbulence products. The Thunderstorm Identification, Tracking, and Nowcasting (TITAN) was developed and implemented to process data from the CWB radar mosaic and the TTY Doppler weather radar.
RAL, MMM, and our Taiwan technical partner, the Institute for Information Industry, will continue to collaborate on the following research and development activities in 2008:
- Develop and evaluate enhancements to WRF-VAR
- Evaluate new data types and their contribution to WRF performance
- Evaluate the performance of the in-flight icing algorithm using the WRF output
- Evaluate the performance of the jet stream turbulence algorithm using the WRF output
- Continue to refine the Java-based Multi-dimensional Display System (JMDS)
- Begin redesign of the web-based Multi-dimensional Display System (WMDS)
- Continue to provide training of CAA personnel
The Aviation Digital Data Service (ADDS) is jointly developed by NCAR-RAL, NOAA-Global Systems Division, and the National Weather Service's Aviation Weather Center (AWC) with funding from the FAA Aviation Weather Research Program. ADDS disseminates weather products to the aviation community via the web. The AWC provides 24 hour-a-day support for Operational ADDS, while an experimental version of ADDS resides at NCAR-RAL and provides next-generation products and services.
1) Sample screen capture of the new, version2 Flight Path Tool. Composite Level-2 NEXRAD radar data shown for 2045 UTC 08 Oct 2007.
ADDS developers at RAL worked closely with the helicopter emergency medical services (HEMS) user community to tailor the ADDS Flight Path Tool to meet their needs for weather information in a very narrow range of altitudes and short distances. Version 1 of a new HEMS tool was made available on 1 Nov. 2006, and throughout 2007 its usage grew as word spread and a number of testimonials from users indicated that it has improved their situational awareness of weather and the safety of their operations.
During 2007, NCAR-RAL integrated some of the interfaces and features of the newly created HEMS software application into a second version of the Flight Path Tool. Key modifications include the ESRI ArcIMS map solution, simpler interface to weather products via menus, and new weather products including NEXRAD Level-2 radar data. This new version of the application is currently being tested and prepared for release at the end of the year.
Besides the new Version 2 Flight Path Tool, a new visualization of observed and forecast weather data at airports was created. In a single graphic, the observed weather data at airports (in METAR format) is seamlessly merged with terminal forecasts (TAFs) of the same weather conditions into the future. In one quick glance, users can now see past, present, and future weather conditions at an airport and the immediate impact to their aviation flight category.
2) Sample "meteogram" time-series showing seamless combination of observed and forecast weather data at Little Rock, AR. Past 6 hours observed data is combined with the "terminal forecast" for future 18 hours.
In the next 12-18 months, ADDS will focus on digital delivery of all weather products in industry-standard formats. For example, aviation weather text products including observations, forecasts, and advisories are currently available in XML and CSV formats. In the near future, these will be served using a Web Feature Service, a standard created by the Open Geospatial Consortium (OGC), allowing users to integrate these ADDS data into a variety of GIS applications. Likewise, ADDS developers are researching the capabilities of an OGC Web Coverage Service to deliver its gridded weather forecasts of icing, turbulence, and more. In coordination with the FAA's System Wide Information Management Program and the Joint Planning and Development Office, work will continue to make ADDS data available on a machine-to-machine basis. And finally, further enhancements will be made to the HEMS tool: radar and satellite data will be incorporated; user-specific data (e.g., location of hospitals, helipads/bases, etc.); street-level maps will be made available; and a search interface created.back to top
NCV real-time ceiling height analysis as it appears on the Flight Path Tool on ADDS.
Adverse ceiling and visibility (C&V) conditions create serious flight safety hazards for general aviation, as well as costly reductions in traffic flow efficiency for commercial operations. On a yearly basis, C&V hazards result in 30-40 fixed-wing general aviation accidents, ~65-75 deaths and losses of ~$150-160M. Reduced C&V conditions are second only to convective weather as a cause of U.S. flight delays. RAL research funded by the FAA Aviation Weather Research Program (AWRP) directly addresses both safety and efficiency concerns through the development of improved C&V weather information and decision support systems.
The National Weather Service (NWS), the Joint Planning and Development Office, and the National Ceiling and Visibility (NCV) Research Team at NCAR have determined that NCV’s automated diagnosis and forecast systems are viable technologies for the operational production of data grids that will flow into the NWS forecast process. They are included in NextGen planning as a critical part of the planned “four-dimensional data cube.”
A new NCV analysis system was delivered to the Aviation Weather Center (AWC) in Kansas City in summer 2006 and during FY2007 support was provided to the AWC and the FAA for ongoing operational system testing and evaluation. The NCV real-time analysis grids for ceiling, visibility and flight category have been made available as experimental products used by the National Weather Service’s Eastern Region aviation forecast modernization process through the year. They are shown to be accurate for local conditions and have good reliability in representing the most probable conditions between reporting sites. These results encourage further development and use of NCV gridded ceiling and visibility products.
The operational NCV forecast integration system was re-implemented to update the weights of the various forecast components (including RUC, LAMP MOS, persistence and data mining) on an hourly basis. A playback system based on the operational system was implemented for internal research and development. The forecast integration system now supports both agile selection as well as a weighted majority vote algorithm. Both schemes are actively being tuned and evaluated by the science staff for performance. The NCV forecast system was adapted to incorporate the new GFS LAMP and its more frequent updates and sites leading to improved forecast performance.
Testing and evaluation of the NCV analysis product will continue in FY2008, and we anticipate that the analysis system will be approved for operational use. In addition, the NCV forecast integration system will be evaluated by NOAA/GSD with the goal of moving it to the Experimental ADDS website and transferring it to the Aviation Weather Center.back to top
For more than twenty years the FAA has funded research and development efforts aimed at improving short-term forecasting of storm hazards affecting aviation. This work has led to the creation of more than 15 separate forecasting tools, algorithms, and systems. In FY07 the FAA changed its strategy and created a new research team tasked with development of a single forecast system. This new “Consolidated Storm Prediction for Aviation” (CoSPA) program brings together researchers from NCAR, MIT/Lincoln Laboratory, and NOAA’s Global Systems Division to create a 0-8 hour forecast for both summer and winter storms. Forecast products from this system will be designed to satisfy the current needs of Air Traffic Management (ATM) as well as the future demands of the Next Generation Air Transportation System (NextGen) in which much of the strategic air traffic decision making will be made by automated decision support tools.
The first steps in the development of CoSPA have included: creation of an inventory of existing technologies in the operational nowcasting community; evaluation of existing technologies; identification of gaps in existing technologies; development of an appropriate system architecture; and creation of a research and development agenda to address technological gaps.
Highlights in system design and software engineering thus far include:
Establishing methods for data exchange and agreement on a common data format. Coordination with the FAA’s System-Wide Information Management (SWIM) team will allow CoSPA to take advantage of the proposed network-centric approach to data sharing between the FAA and other government agencies.
Creating a system architecture with a modular design to foster ease of plug-and-play experimentation with new technologies.
Developing a new Graphical User Interface to provide a menu-driven selection of datasets, preprocessing techniques, and blending methods, as well as produce standard statistical skill scores for graphical display or output to files for further processing.
A scientific R&D agenda has been established and work is underway in the following areas:
Extrapolation – New techniques for improving the skill of extrapolation forecasts at longer lead times are needed. To meet this need, the variational echo tracking technique is being compared with a Lagrangian TITAN-based technique; results to date indicate that both techniques have associated strengths and weaknesses.
Nowcasting (0-2 hr) Storm Initiation –. The ability of current technologies to predict storm initiation is being assessed using the Random Forest statistical technique. This technique will also be used for to analyze different predictor fields’ ability to forecast storm growth and decay.
Data Assimilation - Team members have been testing and evaluating new techniques for assimilating radar data into rapidly-updating NWP models. Current work focuses on improved assimilation of the 3D radar reflectivity through latent heat nudging and variational analyses of the 3D wind field.
Blending – Work is underway to use new blending techniques to combine the relative strengths of observation-based and model-based forecasts into a seamless forecast can be easily understood by Air Traffic Managers. A testbed has been created for the evaluation of extrapolation, model and blending forecasts, and intercomparisons are being made. Results indicate that critical components of the blending will be the climatological processing of the extrapolation forecasts, phase correction of the model forecasts (see figure below) and statistical weighting of the two forecasts.
Precipitation Calibration – Scientists are working to better calibrate automated precipitation measurements. Accurate measurements of snowfall rates, as well as accirate detection of the phase of precipitation, are essential for implementing correct de-icing procedures and hold-over times for planes during winter storms.
A major milestone for Summer 2008 will be the first real-time demonstration of the experimental CoSPA system. It will produce 0-6 hour high resolution (3.3 km) forecasts using heuristic-based extrapolation forecasts and output from a high-resolution mesoscale model with real-time radar data assimilation. The products will be rapidly updating so that the latest model and observational data are incorporated. Work will also continue on all of the R&D tasks outlined above.back to top
The Aviation Weather Group of the National Weather Service has funded NCAR during the past three years to install and run the Thunderstorm Nowcasting System (Auto-nowcaster or ANC) at the NWS Dallas/Ft.Worth (DFW) Forecast Office (FO). Two primary objectives of the Forecaster-Over-the-Loop demonstration are to assess 1) the role of the NWS forecaster in providing value-added enhancements to gridded, automated, nowcast products produced by the Auto-Nowcaster, and 2) the usefulness of these products as guidance for the forecaster in producing short-term forecasts, area warning updates and CWSU outlooks for terminal/enroute aviation traffic. The overarching goal is to improve the consistency, reliability, and accuracy of 0-2 hour convective forecast products for automated aviation weather digital products (4-D grids) for the National Aerospace System’s Next Generation Air Transportation System (NGATS).
Previous demonstrations of the ANC system for FAA-funded activities have shown that forecaster input into the ANC process, particularly in producing storm initiation nowcasts, added consistency, reliability and accuracy to the 0-1 hr short term, time and location specific thunderstorm nowcasts. The primary objective this year has been the evaluation of performance of the Auto-nowcaster thunderstorm initiation nowcasts with and without forecaster input into the process. Using standard statistical calculations of probability of detection (POD), false alarm rate (FAR) and critical skill index (CSI) over the whole NWS County Warning Area (CWA) domain, it is difficult to separate out the CSI skill associated with the storm initiation nowcasts, as nowcasts of extrapolated storms generally dominate the statistics over the very large domain where many storms are occurring simultaneously. For this reason, a new approach was taken by Rita Roberts, Eric Nelson and Tom Saxen to perform the statistical evaluation of storm initiation nowcasts. The approach was simply to compute the statistics over smaller domain sizes that are more relevant to the scale on which the many different discrete areas of convection are occurring (Figure 1). We have found that this approach provides us a more detailed and informative look at nowcast performance and a better scientific understanding of the factors that increase or decrease the accuracy of the nowcast.
Overlaid onto Fig. 1 are the CSI skill scores of ANC performance with (blue) and without (magenta) forecaster-input during the complete event for five of the subset boxes. The CSI scores in the top two plots (Boxes 1,2 and 4,3) show minor differences in skill with and without forecaster input. Within these boxed regions, very little initiation is occurring and most nowcasts are based on the extrapolation of existing storms. The benefit of forecaster involvement is much clearer in the statistics for the other three subset boxes (Boxes 1,1; 2,2; and 3,2). Forecaster-entered convergence boundaries clearly had an impact in the increased accuracy of the nowcasts. Significant new convection initiated in these boxed regions during the 10 hr period spanning this event and forecaster-entered boundaries aided in the timely nowcasts of new convection triggered by these boundaries. Increases of CSI scores ranged from 0.3 – 0.5, a substantial increase in accuracy.Figure 1. Comparison of CSI performance of Auto-nowcaster thunderstorm nowcasts with (blue curves) and without (magenta curves) forecaster-input into the system computed for the boxed (green) regions shown. The white polygon is the 60 min nowcast of new storm initiation produced by the Auto-nowcaster. See text for more details.
Expanding upon these results an additional step was taken to combine the statistics from the individual sub-grid boxes over the duration of the event to evaluate the overall performance in storm initiation nowcasting when forecaster-entered boundaries are used in determining storm initiation. Results indicate that the use of forecaster-entered boundaries in the forecast process leads to increased POD of storm initiation and increase in overall accuracy of CSI.
Encouraged by these results, the NWS Meteorological Development Laboratory (MDL) has been collaborating with Rita Roberts,Dave Albo and Dan Megenhardt to transfer the components of the NCAR ANC forecaster interactive tools to MDL for inclusion in a prototype AWIPS system running at MDL. Surface convergence boundaries will be entered by the forecasters on AWIPS and the boundary information will be sent to the ANC machine for inclusion in the ANC processing. This is anticipated to 1) increase the ease of forecasters for entering boundaries because of their familiarity with the AWIPS displays and 2) to facilitate incorporating this task into the everyday routine of the NWS Short Term Forecaster. Transfer of the other software components of the ANC into the MDL prototype AWIPS will occur during FY2008.back to top
Since late FY04, RAL has been sponsored by the Department of Defense's Defense Threat Reduction Agency (DTRA) to develop tailored meteorological decision-support applications for the military and domestic emergency response communities. In particular, these applications are used to enhance DoD's Chemical, Biological, Radiological, Nuclear, and Explosive (CBRNE) hazard prediction toolsets such as the Hazard Prediction and Assessment Capability (HPAC) and more recently the Joint Effects Model (JEM). RAL’s work has two primary objectives: development of an operational algorithm which can both estimate an unknown CBRNE source and predict a refined downwind hazard from that source, using available CBRNE and meteorological sensor observations; and integration of this algorithm into the HPAC/JEM hazard-prediction toolsets. To support testing and evaluation of this product, RAL is developing a virtual testing and evaluation environment (VTHREAT) which will enable simulation of a realistic CBRNE release scenario, placement of CBRNE and meteorological sensors, and extraction of the resulting synthetic sensor readings. These synthetic observations can then be used by the evolving algorithms to evaluate their ability to recreate the CBRNE event.
In FY07 RAL developed and demonstrated of a prototype VTHREAT system which provides the capability to emulate a CBRN attack scenario. During the development of the VTHREAT prototype, DTRA requested that RAL use the VTHREAT concept to provide field test design guidance for DTRA’s FUSION Field Test 2007 (FFT-07). The question posed to NCAR was “What is the optimal release and collection configuration to ensure a high probability of detection, while also avoiding saturation of the front line detectors AND supporting the distinction of multiple simultaneous releases”. Subsequently, VTHREAT was used to generate three typical atmospheric conditions for the field test area, Dugway Proving Ground, Utah. Using synthetic atmospheric datasets, myriad material releases were simulated with three different proposed sensor configurations. The synthetic sensor readings were extracted and analyzed to determine the probability of detection under each of these configurations. These results were then presented to the FFT-07 Science Team Board in July 2007. After reviewing NCAR’s recommended sensor configuration, the board approved utilization of this layout for the operational tests in September. The FFT-07 tests were recently completed and the field test has been deemed a success by the DTRA sponsors. The VTHREAT system also includes a gui application which allows users to interactively create synthetic sensor and meteorological data sets. This VTHREAT prototype application was demonstrated in September 2007 at the FFT07 VIP at Dugway Proving Ground, Utah.Variational SDF Prototype Phase I Sample Results.
FY07 also marked the development of the Variational Sensor Data Fusion (SDF) prototype. Phase I of this effort used the existing L3-Titan SCIPUFF code (and adjoint) to calculate a source estimate and then used that source estimate to run a single forward SCIPUFF simulation. This phase of the prototype was recently completed and demonstrated in September 2007 using synthetic chemical and meteorological sensor data generated with the VTHREAT prototype. (Figure 1)
The VTHREAT GUI application will continue to be refined during FY08. Current plans call for the development of additional types of meteorological sensors (towers, rawinsonde, and LIDAR) and the ability to dynamically place grids of chemical sensors. The release of the first official version of the VTHREAT system is scheduled early FY09.
The primary activity of the SDF program during FY08 will be the continued development of the variational sensor data fusion algorithm. Phase II of this algorithm is designed to improve upon the initial source estimate, and associated forward prediction demonstrated in FY07 utilizing a recently developed Eulerian plume model (and adjoint), that iteratively refines the source estimates with variational data assimilation techniques. Phase II development is currently underway and is on schedule for an initial operational demonstration in November 2007. Continued refinement of the Phase II algorithm will be conducted throughout FY08. Current plans call for the development of the capability to include observations taken at multiple times and the ability to identify the source locations from multiple release scenarios.back to top NexGen MDS and API Conceptual Overview.
Since late FY04, RAL has been sponsored by the Department of Defense's Defense Threat Reduction Agency (DTRA) to develop tailored meteorological decision support applications for the military and domestic emergency-response communities. In particular, these applications are used to enhance DoD's Chemical, Biological, Radiological, Nuclear, and Explosive (CBRNE) hazard prediction toolsets such as the Hazard Prediction and Assessment Capability (HPAC) and more recently the Joint Effects Model (JEM) (see figure at right).
In FY07, RAL continued to refine the NexGeneration Meteorological Data Server (MDS) and its associated Application Programming Interface (API). New MDS versions were released in February and April 2007 to DTRA facilities in Alexandria, VA and Albuquerque, NM. The next version is scheduled for operational deployment in October 2007.
The effort to refine and improve the HPAC weather interface in support of the HPAC 5.0 release also continued with major improvements made in new, more modular component architecture for the software system including published Application Programming Interfaces (APIs); replacement of the legacy MDS Client interface with the new NexGen MDS API; and general GUI Enhancements.
Developmental support for HPAC 5.0 Service Pack 1, scheduled for delivery January 2008, is underway. Weather interface related improvements are focused on the NexGen MDS GUI. These GUI updates include: notification to user of estimated download file size, if it exceeds a prescribed maximum allowable file size; estimated file size and download status; identification of data source (useful for those using the “AUTO” download option); and estimate of the time it takes to generate NWP analyses.
4DWX is accredited for operational use at seven test ranges.
Over the past decade, the U.S. Army Test and Evaluation Command (ATEC) has sponsored the research and development of the Four-Dimensional Weather (4DWX) system, a cutting-edge weather modeling system based on MM5 and the WRF Model. 4DWX provides high-resolution mesoscale modeling capability, short-term thunderstorm prediction, multi-dimensional integrated displays, and fine-scale climatological analysis tools, enabling the Army to test military hardware under precise conditions across the full spectrum of arctic, tropical, desert, and other natural and controlled environments. 4DWX is accredited for operational use at seven test ranges:
Real-Time Four-Dimensional Data Assimilation (RTFDDA):
This unique scheme assimilates observations from a variety of data feeds, preserving the data’s temporal dimension during assimilation. New analyses and forecasts are made every one to three hours, depending on the range, providing the operational forecaster with very timely information. A recent addition to RTFDDA is the Model Manager, which is a user interface that provides more modular, automated control of non-operational RTFDDA simulations, and also provides the resource management necessary to efficiently use the Test Center’s computing resources.
3-Dimensional Variational Data Assimilation (3DVAR):
3DVAR techniques at the mesoscale are being evaluated, and a number of non-standard observations that cannot be included in RTFDDA's observation-nudging scheme, such as satellite radiance, GPS, and radar, are being incorporated. 3DVAR is currently being integrated into the RTFDDA system, yielding a model-based solution that will account for all available observations.
Late in FY07 RAL successfully fielded at Dugway Proving Ground (DPG) a test version of an operational ensemble of 4DWX forecasts in support of FFT07, the FUsing Sensor Information from Observing Networks (FUSION) Field Trial. The ensemble system (called E-RTFDDA) extends the pseudo-deterministic, single RTFDDA realizations by running a suite of RTFDDA forecasts, all valid at the same place and time. The ensemble comprises 30 members whose differences are induced by varying initial conditions, boundary conditions, model physics, and model cores. The system cycles every 6 h and produces four 36-h forecasts per day. Currently forecasts are uncalibrated, but research into calibration (including first and second moments of ensemble distributions) continues. As a critical step in this work, we are generating a 3-year archive of retrospective forecasts that will provide statistics necessary for an initial attempt at calibration and will allow us to optimize the size of the ensemble.
Coupled Applications: The direct NWP output from 4DWX is the essence of the forecast guidance used by staff at the ATEC ranges, but a great amount of added value is achieved by coupling this direct output to secondary models, also known as coupled applications. These include:
- Noise Assessment and Prediction System (NAPS)
- Second-order Closure Integrated Puff (SCIPUFF) model
- General Electric Missile and Satellite Simulation (GEMASS) Program
- Open Burn / Open Detonation Model (OBODM)
Model Verification: In FY07, RAL continued development of approaches to model verification that go beyond traditional metrics such as root-mean-squared errors calculated at points. Among the newer approaches to verification are those that treat predicted and observed weather as one or more “events,” or “objects”—a change in wind direction, or a coherent region of rainfall, for instance. Object-based approaches do not inherently handicap more resolved models and are often more consistent with the mental images that end-users have of forecasts and observations.
Web Portal: Development continued in FY07 on a dramatically new look and feel to the 4DWX range Web pages using portal technology, which permits administrative and user customization. The portal’s flexibility, accessibility, modularity, and extensibility make it ideal for serving as the new foundation for most 4DWX user interfaces.
Climate FDDA (C-FDDA: RAL has begun generating a test archive of mesoscale climatographies for use by two ATEC ranges for long-range scheduling of the optimal time, day, season, and location for materiel testing under specific weather conditions.
Plans For FY08:
The first version of MetVault, a central data repository for observations, model output, and sundry other electronic information, will be released. It includes a sophisticated search engine, a suite of services that will enable users to extract, interpolate, combine subsets of stored data, and write out the data in a variety of formats, and a quality control system for observations ingested into the 4DWX database.
Work will also continue to refine and extend the Global Meteorology on Demand (GMOD) tool, which enables non-scientists to launch highly resolved, rapidly executed NWP simulations anywhere on the globe with a simple, Web-based user interface. And RAL will continue to explore methods for assimilating radar data into 4DWX. In FY08 we anticipate significant advances from an approach that is based on the coupled cycling of RTFDDA and the Variational Doppler Radar Assimilation System (VDRAS).
The Future Combat System (FCS) is the U.S. Army's program to modernize American war-fighting capabilities and achieve the ability to rapidly deploy a dominant ground force anywhere in the world within days. Because most weapons testing and military operations rely heavily on accurate weather information, the Army relies on tools such as the 4-Dimensional Weather Real-Time Four Dimensional Data Assimilation (4DWX-RTFDDA) system developed at RAL in collaboration with the Army Test and Evaluation Command (ATEC). This modeling system provides high-resolution 4-D synthetic weather analyses and forecasts by continuously merging all available observations with full physics models. It has become a very valuable tool for providing weather modeling support for FCS Modeling and Simulation (FCS/MS). Since 2004, NCAR RAL has been continuously adapting and refining this modeling system and proving realistic simulated battle-field-scale weather events for the FCS/MS applications.
RAL produced model simulations of events for FCS with 4DWX-RTFDDA final analyses for two events - one at the US Army Dugway Proving Ground (DPG), Utah, involving nocturnal boundary layer flows; and the other at the Army Cold Region Test Center (CRTC), Alaska, involving an extreme cold-weather event. Both weather processes present unique modeling challenges due to complex terrain and the scarcity of measurements. The complete model output for the simulated events includes the meteorological products and celestial-object information (moon phase, solar and lunar illumination).
A second accomplishment involved the study of weather impacts on soil “trafficability”, a term used to describe the mobility of a broad variety of military ground vehicles over specific local terrain, land use, soil type, composition and moisture. Soil moisture is, of course, heavily dependent on recent and current weather conditions, especially precipitation. The 4DWX-RTFDDA system is being used to generate a summer trafficability scenario at DPG and surrounding region in Utah. A case study demonstrates the great potential of applying 4DWX-RTFDDA modeling technology to provide an estimate of real-time trafficability parameters (e.g. surface soil strength, stickiness, slipperiness, and critical-layer depth) and short-term (0 – 48h) forecasts for battle field planning.
Development of new capabilities to support the FCS/MS missions will be further explored. Specifically, to better simulate trafficability, RAL will work with the Army DPG modelers to couple the 4DWX-RTFDDA model system with Army trafficability models such as the NATO Reference Mobility Model. Meanwhile, RAL will continue to improve the 4DWX-RTFDDA modeling capability, which provides real-time support for the upcoming FCS Distributed Test Events (DTEs). Special effort will be committed to exploring the weather-sensitive variables that are commonly used by military planners, such as Cn2 for laser targeting applications and vertical wind shear over complex terrain for unmanned aerial vehicle (UAV) operations.back to top
Hydrological impacts of significant weather and climate events often translate into staggering human, economic and environmental costs. Despite decades of research, significant gaps remain in how weather and climate information are used for reliable hydrological prediction. Additionally, it is now well-recognized that terrestrial hydrologic processes are not only influenced by weather and climate but also exert a significant feedback to the coupled Earth system. The nature of the relationship between terrestrial hydrologic processes and the atmospheric processes that simultaneously drive and respond to them are the focus of RAL’s Hydrometeorological Processes efforts.
Romania: During FY07 RAL continued to provide hydrological modeling support for a World Bank-funded project for the country of Romania. Working in collaboration with scientists from Baron Advanced Meteorological Services (BAMS) we have completed a suite of enhancements to the Noah-distributed hydrological model and implemented and tested the new modeling system in several new river basins in both the U.S. and in Romania. Work also focused on fully parallelizing all components of the Noah-distributed hydrological modeling system for the execution of parallel computations on NCAR supercomputers.
Colorado Front Range: A new project aimed at improving predictions of short term (hours to 1 day) flash flooding events in the Colorado Front Range was initiated during FY07. This project deploys, in an unprecedented manner, the newly developed Noah-distributed hydrological modeling system over a large region in north-central Colorado. During FY07, the model domain was defined and attributed and case study simulations were executed which focused on simulating the 1997 Ft. Collins flash flood (Figure 1). Additionally, the Noah-distributed model has been coupled to the Advanced Weather Research and Forecasting model for fully-coupled simulations of high-impact hydrometeorological events. Results from an initial round of sensitivity studies have been completed and were presented at the annual WRF User’s workshop in late June.back to top
National Transportation Safety Board records indicate that in-flight icing causes more than 25 accidents annually, with over half of these resulting in fatalities and damaged aircraft. The cost of injuries, fatalities and aircraft damage is estimated to be $100M annually. The RAL In-Flight Icing Program addresses this problem by producing improved operationally -available, high-resolution, accurate diagnoses and forecasts of aircraft icing conditions. This work is funded by the FAA's Aviation Weather Research Program.
An upgraded version of the Current Icing Product (CIP) depicting icing severity has been thoroughly evaluated and cleared for unrestricted operational use by all aviation decision makers. This designation opens the door for vendors to include it in their product suites for in-cockpit use by pilots. CIP is the first product developed by the FAA’s Aviation Weather Research Program to receive this designation.
Progress was made in improving the microphysical parameterizations within the MM5, Rapid Update Cycle (RUC), and Weather Research and Forecast (WRF) models to provide better forecasts of potentially hazardous icing conditions. In particular, a parameterization of snow density varying with crystal size has resulted in significant improvements to our ability to model precipitation on the ground and hydrometeor content aloft.
Continued evaluation of NASA’s advanced satellite products to improve the severity product is very encouraging, pinpointing areas of hazardous icing. Additionally, freezing drizzle detection using NEXRAD data looks feasible, and RAL scientists are now engaged in determining how and where to implement the new detection algorithm.
Work will continue to move the forecast version of the icing algorithm (FIP) to operational status for the National Weather Service and FAA. Work is underway to make the new WRF Rapid Refresh model the basis for our icing diagnosis and forecasting algorithms. New data sets (e.g., NEXRAD 3D mosaic and advanced satellite products) will be tested for use in the Current Icing Product. Further development of microphysical parameterizations using data sets from the ICE-L field experiment will be conducted, and a confidence field will be created for our icing algorithms to help them become part of automated decision-making systems. Finally, we will begin planning for North American and global versions of our icing algorithms.
back to top
Geographic display of Juneau Aviation Warning System. Hazard warning areas are outlined as boxes.
Sheep Mountain Anemometer Site in February 2007.
The Juneau, Alaska, environment is characterized by rugged terrain and adverse weather that combine on occasion to produce moderate to severe terrain-induced turbulence and wind shear for flights into and out of the Juneau International Airport. To address this problem, the FAA tasked RAL in FY97 with the development and implementation of a new wind hazard warning system for the airport. Over the last ten years, RAL engineers and scientists have built a prototype warning system that relies on statistical correlations between wind-related parameters observable by the system (i.e., speed, direction, shear, variance, etc.) and the location and severity of turbulence. Since terrain is fixed and there are distinct strong-wind scenarios, correlations are expected between winds and hazards. These correlations were established using the warning system’s input measurements and hazards measured by research aircraft during three intensive field projects. The graphical display was designed for a variety of users including Automated Flight Service Station specialists, airline dispatchers, and pilots. It contains information that depicts current alerts and conditions as well as conditions during the past hour. The prototype warning system is used every day by airlines and pilots flying in and out of Juneau to assess the current turbulence hazards around the airport.
During FY2007, the FAA evaluated options for the final configuration of the terrain induced turbulence warning system in Juneau. RAL supported this effort by providing information to the FAA about the capabilities and operation of the prototype system. RAL also operated and maintained the prototype warning system in Juneau to allow airlines and pilots to assess the important and unique weather information it provides.
RAL will continue to operate and maintain the warning system to continue service for the Juneau airlines and pilots. The FAA will complete its evaluation and decide on the final warning system configuration in FY2008. RAL will then complete the technology transfer of the prototype system to the FAA in FY2008 and FY2009.back to top VDRAS vertical velocity field (color) and wind vectors at 0714 UTC, Aug. 1, 2006. The black contours are the observed 30 dBZ reflectivity at 0846 UTC. The vertical velocity field shows a clear updraft maximum near the Beijing C-band radar (marked as BJRC) 1 hour and 32 min before the 30 dBZ convective cell appears.
The research and development efforts of RAL's Convective Weather Group are aimed at improving short-term (0 – 6 hour) thunderstorm forecasting and bridging the gap in skill between observation-driven expert systems and numerical weather prediction. One of the programs conducted by the Convective Weather Group is a collaborative effort with the Institute of Urban Meteorology (IUM) of the Beijing Meteorological Bureau (BMB) to transfer the NCAR AutoNowcaster (ANC) to Beijing in support of China's role as host of the 2008 Summer Olympics. As part of this multi-year project, scientists from NCAR and the BMB are studying the local characteristics of thunderstorm initiation and evolution to modify and tune the ANC algorithms for optimal performance in the Beijing area. Efforts are also underway to train the IUM staff on thunderstorm nowcasting techniques and the use of the ANC.
The last component of ANC, the software for satellite applications, has been installed in Beijing and the Chinese satellite data were ingested to ANC. Training of the IUM staff took place on the use of satellite applications and ANC in general. The ANC system that is running at BMB was named BJ-ANC. It includes the Variational Doppler Radar Assimilation System (VDRAS) which retrieves and assimilates boundary-layer winds and thermodynamics from Doppler radar, surface stations, and sounding data. The BJ-ANC has been demonstrated during the summer 2007 along with a number of other nowcasting systems including NCAR’s Niwot (a system that blends nowcasting and numerical weather prediction techniques) to prepare for the WMO-sponsored forecasting demonstration during the Beijing 2008 Summer Olympics.
Several severe storm cases that were documented during the past summers were studied to examine Beijing convective weather characteristic and forecast challenges. It was found that the storms had various initiation mechanisms and terrain forcing played a critical role. VDRAS analyses of these cases revealed convergence and humidity features that existed a few hours before storm initiation in the regions of metropolitan Beijing districts. These features will be further examined and analyzed to investigate their connections to storm initiation forecasting. Fig. 1 shows an example of how the analyzed updraft correlates to the location of storm initiation. The cold pool and gust front characteristics and their contribution to storm development are also studied.
NCAR scientists will participate in the summer 2008 forecasting demonstration project before and during the Olympics in Beijing. Further training of the BMB forecasters on nowcasting techniques will be conducted. The results from the case studies will be written up as journal papers.back to top
The cloud top height output is shown for 12 September 2007 at 2045 UTC for the Gulf of Mexico domain. Overlaid onto the cloud top height in a) is output from the convective diagnosis oceanic algorithm (shown in magenta) and in both a) and b) the 1-hour nowcasts of convection location are indicated by the red polygons with storm motion vectors.
Convection is an aviation hazard that can produce turbulence, icing, hail and lightning. Over the continental United States, the aviation community is well-served by the national Weather Surveillance Radar 1988 Doppler (WSR-88D) network that gives accurate and timely detections of the presence of hazardous convection. Over remote oceanic regions, however, the aviation community is hampered by the scarcity of available data that can be used to create products that detect and warn pilots, dispatchers and air traffic controllers of the current and future locations of hazardous convection. Unexpected convective development while en route can lead to costly re-routing and delays. Transmitting this critical information into the flight deck is difficult and further compounds the hazards faced by oceanic flights. The long duration of oceanic flights requires convective forecasts with longer lead times than those required for the continental United States. Under NASA Earth Observing System sponsorship, the Oceanic Convection Program at the RAL is working to overcome these limitations through the use of geostationary and polar-orbiting satellite observations and global numerical model results within an intelligent system that generates 0-2 hour nowcasts of convection location. Collaborators in this research include the National Center for Atmospheric Research, the Naval Research Laboratory-Monterey (NRL), and the Massachusetts Institute of Technology Lincoln Laboratory (MIT LL).
This R&D effort is currently focused on the Gulf of Mexico. Figure 1 shows an example of the cloud top height product overlaid with the convection diagnosis oceanic (CDO) product. The CDO has been devised using a scaled combination of three satellite-based algorithms that define where convection is present. The CDO is then tracked in time using the thunderstorm initiation, tracking and nowcasting (TITAN) algorithm that provides the 1-hour and 2-hour nowcasts of storm location, indicated by the red polygons in Figures 1a and 1b. Accomplishments in FY07 include software development of ingestors for polar-orbiting satellite data sets of the QuikSCAT near-surface winds, the sea surface temperature and soundings. The storm tracking algorithm was refined to improve results and an updated version of the NRL cloud classification algorithm was implemented. Investigation has commenced into characterizing the environment of regions where convection initiation has occurred.
Verification of oceanic products has been hindered by the limited number of independent data sources. However, two such data sources do exist and are beginning to be exploited: the TRMM Precipitation Radar and the CloudSat Cloud Profiling Radar. Lincoln Lab is taking the lead in verifying the oceanic convection products using these radar data. In addition, the oceanic convection products are being generated over the continental United States to enable comparison of the oceanic convection products to the WSR-88D reflectivity field.
Development of the Oceanic nowcasting system will continue. Characterization of the environment near initiating convection will be used to devise prediction methodologies. Likewise, characterizing the environment near dissipating convection will be undertaken. The total precipitable water global satellite field will be ingested to better define regions where convection can be sustained. The merger of global model fields with observations will be undertaken such that the maximum amount of mesoscale detail will be retained. Expansion into the Pacific domain is planned.back to top A schematic description of E-RTFDDA system
RAL modelers have continued their effort to implement the “observation-nudging”-based FDDA scheme in the Weather and Research Forecast (WRF) model. A reliable research version of the algorithm has been included in WRF-ARW version 2.2, which was released for public use in Dec. 2006, along with utility programs for observation-data preparation and a brief user guide. In FY07, RAL modelers focused on improving the “observation-nudging”-based WRF RTFDDA modeling system for real-time operations at the Army test ranges. A robust data quality-control (QC) algorithm was built using the WRF-VAR framework and RTFDDA real-time cycling model outputs.
A variety of data-assimilation approaches have been developed for WRF. Each data-assimilation approach has its pros and cons with respect to assimilation of different kinds of observations for weather of different scales. Therefore it is potentially advantageous to join the schemes in a reasonable way so that the advantages of the different approaches can be combined. In FY07, a hybrid data-assimilation approach that is based on the RAL “observation-nudging”-FDDA, the PSU “grid-nudging”-FDDA, and the ESSL WRF-3DVAR data assimilation schemes was developed. The idea is to enhance the 4DWX-RTFDDA data-assimilation capability by using “grid-nudging” with intermittent 3D analyses produced by WRF-3DVAR. With this algorithm, the WRF-3DVAR capability that assimilates remotely-sensed observations such as satellite radiance and radar radial winds and reflectivity can be merged with the “observation-nudging” capability for meso- and small-scales that assimilates temperature, moisture and wind measurement from diverse observation platforms. Results from initial tests are encouraging.
The last, but the most noteworthy, accomplishment is the development of a 4DWX-RTFDDA based mesoscale ensemble analysis and forecast system (E-RTFDDA). E-RTFDDA extends the RTFDDA deterministic forecasting capability to probabilistic weather prediction by executing an ensemble of RTFDDA models which sample the uncertainties in the model components. In August 2007, a 30-member E-RTFDDA ensemble analysis and forecast system was run on an Army High Performance Computer Modernization Program (HPCMP) HPC machine at the Army Dugway Proving Ground (DPG) to support the FFT07, a major sensor data fusion field campaign which was also at DPG. The ensemble system is composed of fifteen MM5 and fifteen WRF members which sample the uncertainties in model physics, boundary conditions, measurements, data assimilation and analysis algorithms, and model dynamics. The system is summarized in Fig. 1. The system is considered to be one of the most advanced operational systems of its kind in the world, with its continuous data assimilation and forecast-cycling capability, multiple scales with a 3.3 km grid fine mesh, and inclusion of multiple ensemble perturbation schemes. It runs 4 cycles a day and produces 6-h analyses and 36 h forecasts for each cycle. The ensemble system successfully ran through the field campaign, and continued to run afterward for longer-term system verification. The web-based probabilistic-forecast graphics products are produced and can be viewed at http://dpg-ingest.4dwx.org/images/ens/index.html).
For hybrid data assimilation development, the focus will be on exploration of assimilation of satellite radiance and radar radial winds using the hybrid scheme. Currently, studies have been undertaken with the IHOP-2002 cases to tune up the data assimilation parameters, and the implementation of the hybrid scheme for RTFDDA operations is planned during late FY08. Aside from the “nudging-3DVAR” hybrid development, a “VDRAS” (Variational Doppler Radar Analysis System)-“observation-nudging” hybrid approach will also be studied. This hybrid scheme will make use of VDRAS to work as a bridge for incorporating Doppler radar radial winds and reflectivity measurements into high-resolution WRF-RTFDDA through the “observation-nudging” mechanism.
Research will be undertaken to 1) develop and evaluate existing and new ensemble perturbation schemes, including the ensemble-transform Kalman Filter approach and the ensemble adjustment Kalman Filter approach; 2) develop more experience and conduct statistical verification relative to operational ensemble forecasting; 3) generate re-forecasts for the last three years and develop ensemble calibration algorithms; and 4) start to investigate a 4-D EnKF scheme which makes use of “Kalman-Gain” to define the spatial weighting factors of “observation-nudging” data assimilation.back to top
RAL scientists and engineers are continuing work on a number of projects that involve the acquisition and application of meteorological satellite data to enhance understanding of atmospheric processes and applications. Much of this work is being supported by NASA, through the Advanced Satellite Aviation Weather Products (ASAP) initiative and through a system of competitive grants awarded by NASA’s Applied Sciences Program. The goal of ASAP is to enhance the transition of new and existing satellite information and products into operational products by collaborating with the FAA’s Aviation Weather Research Program (AWRP) Research Teams (RT), many of which are based at NCAR. This effort is specifically addressing hazards such as in-flight icing, convective weather, turbulence (clear-air and cloud-induced), and the monitoring of weather in the data-sparse areas over oceans. RAL’s role in ASAP is to coordinate the contributions from the participating universities and laboratories with the needs and requirements of the FAA’s aviation weather Research Teams and to evaluate the accuracy and usefulness of these advanced satellite products. This program is described in an article team members published in the October 2007 issue of the Bulletin of the American Meteorological Society.
During 2007 the In-Flight Icing RT continued testing satellite-based cloud microphysical products produced by NASA Langley for possible incorporation into operational icing products such as the Current Icing Product (CIP). RAL scientists worked with scientists from the University of Alabama Huntsville (UAH) and the University of Wisconsin’s Cooperative Institute for Meteorological Satellite Studies (UW/CIMSS) to test the utility of satellite-based, high-resolution, early cloud imagery for identifying favorable areas for convective development capable of growing into hazardous storms and to identify cloudy and clear-air features that may help identify areas of strong turbulence.
Three new, NASA-funded projects got underway during FY07. Each is aimed at enhancing the use of NASA Earth Science data sets in decision support systems targeting areas of national need. These projects, include studies of oceanic aviation weather hazards (in collaboration with scientists at the Navy Research Laboratory Monterey and MIT/Lincoln Laboratory); aircraft avoidance of convectively-induced turbulence due to thunderstorms (in collaboration with scientists at UAH and UW/CIMSS); and improved monitoring and forecasting of soil moisture and temperature for agriculture (in collaboration with DTN/Meteorlogix).
As a spin-off of this effort to make better use of satellite observations for practical applications, RAL Project Scientist David Johnson has proposed a method for enhancing satellite imagery for enhanced resolution near the edges of the image area. During FY07 UCAR submitted a patent application for this hardware-based enhancement device.
The RAL NASA ASAP effort will continue during 2008 with a gradual transition of the new satellite products into operational use through collaborations with the FAA aviation weather Research Teams.
Encounters with turbulence pose significant safety, efficiency and workload issues for commercial and general aviation. The number of pilot-reported encounters with turbulence is substantial, with moderate-or-greater turbulence pilot reports averaging about 65,000/year and severe-or-greater reports averaging about 5,500/year. More often than not, pilots will try to avoid or exit turbulent air, so turbulence significantly impacts national airspace (NAS) efficiency and air traffic controller workload. Fortunately, not every significant encounter with turbulence results in an injury; nevertheless, according to NTSB numbers, each year turbulence accounts for approximately 71% of all weather-related accidents and incidents. The cost to U.S. airlines due to injuries (medical attention and liability suits), cabin and aircraft damage, flight delays, and time lost to inspection and maintenance is substantial, with estimates in the $150-$500 million/year range. In order to help reduce the number and severity of turbulence encounters and the impact of turbulence on the NAS, RAL scientists are working on improving the detection and forecasting of turbulence and providing operationally useful products directly to users and will supply input for the automated decision support systems planned for the Next Generation Air Transportation System.
Aviation turbulence R&D at RAL is funded primarily through the FAA’s Aviation Weather Research Program and is augmented by NASA’s Advanced Satellite Aviation-weather Products (ASAP) program and the Boeing Corporation.
- A new turbulence detection and on-board quality control algorithm was developed. The software has been provided to Southwest and Delta Airlines for deployment by the end of CY07.
- The NEXRAD Turbulence Detection Algorithm (NTDA) was implemented and tested in collaboration with the FAA’s Advanced Weather Radar Techniques team, and the software was delivered to the NWS for deployment on all NEXRADs.
- A real-time demonstration of the NTDA was expanded to use data from 83 NEXRADs east of the Rockies. A 3-D graphical product was made available to meteorologists and dispatchers via a web-based Java display, and a customized text-based graphic of in-cloud turbulence ahead was uplinked to select United Airlines aircraft.
- The Graphical Turbulence Guidance-2 (GTG2) algorithm continues to progress toward Aviation Weather Technology Transfer (AWTT) Board approval for operational status, which is expected in early 2008. Experimental product displays are available on the Experimental ADDS web site.
- A prototype GTG-Nowcasting system was developed and implemented in RAL.
- A comprehensive study regarding the use of airborne GPS receivers to measure turbulence was performed and delivered to the Boeing Corporation.
Work will be concentrated in the following areas:
In situ measurements: Verification of the edr measurements against pireps will continue, and ICAO documentation on the algorithm will be completed. Fleet-wide implementation of the winds-based edr algorithm on approximately 280 Southwest Airlines planes will be completed. Further, implementation of the algorithm on 140 Northwest Airlines planes will begin. We are also in discussion with other airlines, both national and international, about implementation on their respective fleets.
Remote sensing observations: NTDA-1 will be implemented on all NWS NEXRADs in summer 2008. Work on the next generation algorithm, NTDA-2, will commence to accommodate upcoming scheduled NEXRAD upgrades, e.g. phase coding, dual-polarization, super-resolution, new volume coverage patterns, and a possible new spectrum width estimator. This will provide improved coverage, particularly at lower altitudes. NTDA uplink demonstrations are expected to continue. R&D on a CIT diagnosis model will continue, using empirical relationships and conceptual models based on NWP model, satellite, lightning, NTDA EDR, etc., which will eventually become part of GTG-N.
Nowcasting/forecasting: GTG2 should become "operational" and available on the ADDS web site in early CY08. This product produces forecasts of clear-air turbulence sources out to 12 hours, updated hourly. Preliminary testing of the next version, GTG3, will commence next year. This version will provide probabilistic forecasts, will use all available in situ reports, and will include explicit mountain wave turbulence diagnostics. Work will continue on the nowcast product (GTG-N) which combines all observations of turbulence (including NTDA-2, in situ, PIREPs, and satellite-based diagnostics) with GTG3 analyses to produce a probabilistic nowcast, updated at 15 min. intervals.
Weather, especially convective storms, continues to exert a disruptive influence on aviation, both in the terminal area and en-route air traffic flow. Aviation users need 0 – 6 hour forecasts that provide not only details about the likely weather outcome, but also information about storm structure, intensity, and organization, and associated forecast uncertainty. This emphasizes the need for short-range (0 – 2 days), high-resolution (<10 km spatial resolution) ensemble weather forecasting systems. Optimization of air traffic management, especially under future scenarios of anticipated much increased demand, requires automated decision support tools that integrate probabilistic weather information to estimate airspace capacity and provide guidance for managing air traffic flows under consideration of the associated prediction uncertainties. Under NASA sponsorship, RAL is defining and refining new concepts of how probabilistic weather forecasts can be tailored for aviation needs and integrated with automated decision support tools.
A detailed report has been compiled that reviews short-term weather forecasting techniques, from the perspective of both observation-driven expert nowcasting and numerical weather prediction systems. Emphasis has been placed on numerical ensemble forecasting procedures that will become highly valuable for aviation users. Moreover, a novel approach of how ensemble weather forecasts in the not-too-distant future may be analyzed from an aviation point of view and packaged for integration with automated air traffic management decision support tools has been presented (see figure below). This new approach draws upon recent experience gained with probabilistic convective scenario forecasts. The focus of the report has been on convective storms primarily because of their disruptive influence on air traffic flows. However, the concepts developed there may be applicable to other en-route weather hazards, such as turbulence and icing, as well.
The new concepts developed during the FY07 will be further refined and feasibility analyses conducted. In particular, use will be made of high-resolution ensemble model simulations to create probabilistic weather forecasts with a specific tailoring from an aviation perspective.back to top
Winter weather research and development at RAL has focused on developing a real-time nowcasting system called the Weather Support to Deicing Decision Making (WSDDM) system. Recent additions to the system included the Check Time system, and a high resolution winter weather modeling system using radar data assimilation. “Check Time” is a UCAR patented technology for ground deicing that provides users with an aircraft independent wall clock time that indicates when an applied aircraft deicing fluid is close to failure based on the current minute by minute snowfall rate and temperature. A snow Check Time system was developed in previous years and tested at Denver International Airport and shown to provide value to users. An All-Weather Check Time system was developed last year and adds three new sensors: Vaisala PWD-11 precipitation type sensor, Rosemount Freezing Rain sensor, and a GEONOR snowgauge. The addition of these three sensors allows the system to provide Check Times for all precipitation types (including ice pellets, snow pellets, freezing rain and drizzle).
The key accomplishment for FY07 was the development of a Liquid Water Equivalent (LWE) real-time system consisting of a GEONOR snowgauge, a hotplate snowgauge, a WXT weather, and a precipitation type sensor. The motivation for this development is the need for real-time liquid equivalent snowfall rates for use in holdover time determination. Data were collected at the Marshall test site for this purpose and provided to the FAA for approval. The evaluation and approval process is ongoing. We also successfully demonstrated the ability of the RAL-developed snowmachine to perform deicing fluid testing at temperature close to 0 ºC for a variety of deicing fluids, a temperature range in which testing has proven difficult in the past.
The main focus for FY08 will be the demonstration of the LWE system at four sites: Pittsburgh International, Chicago O’Hare, Denver International, and Minneapolis/St. Paul airports. Data from the LWE system will be provided to airlines, pilots and deicing users via a web site and radio broadcast. The FAA will evaluate the results of the demonstration and provide final approval to use the system operationally. The data from the Marshall site will also be used in this evaluation. Vaisala will develop a commercial version of the system that will be made available to users after FAA approval. A diagram of the LWE system is given below.back to top Comparison of two 3 hour forecasts, one without reflectivity nudging (middle panel) and the other with reflectivity nudging (right panel), of a storm occurred near Chicago on July 8, 2006. The Stage IV precipitation analysis is shown on the left for verification.
High-resolution data assimilation and numerical modeling are critical elements in the effort to improve short-term (0-6 h) forecasts of convective storms. Radar data assimilation using advanced techniques allows a model to start up with initial conditions that account for the presence of existing storms and can result in significant improvement in the timing and location of forecasting convective storms. At RAL’s Convective Weather Group our modeling efforts emphasize the assimilation of radar reflectivity and radial velocity level II data from operational network into high-resolution cloud-resolving models. Two data assimilation systems have been developed and continue to be improved. One is reflectivity nudging using Real-Time Four-Dimensional Data Assimilation (RTFDDA), and the other is a 4D-Var based Doppler radar retrieval system known as VDRAS. These two systems are used both for research and real-time demonstrations.
Several cases documented during the FAA-supported summer 2006 forecasting demonstration period in the Indiana/Illinois domain have been rerun to evaluate the impact of reflectivity nudging on forecast improvement of convective events. Fig. 1 compares 3-hour forecasts with and without the reflectivity nudging. Recently, we have started to migrate the reflectivity nudging scheme from a MM5 modeling system to the WRF model, version 2.2. The mosaic reflectivity data ingest module has been adapted to WRF already. The reflectivity and latent heating nudging algorithms are now made compatible with the WRF modeling infrastructure.
In an effort to improve the computational efficiency of VDRAS, the code has been modified to enable a MPI parallelization capability. Tests on SCD’s Linux Cluster computer showed that the parallelization was successful. A scheme for radial velocity dealiasing that uses VDRAS analysis as reference wind was developed and showed improvement over using radiosonde observations. Additionally, exploratory analyses reveal that it might be beneficial to couple the RTFDDA and VDRAS data assimilation systems to improve temperature and wind fields around convective storms. This would be accomplished through an iterative process in which the initial RTFDDA analysis is used as background field for the VDRAS wind and thermodynamic retrieval, which then are ingested into the RTFDDA assimilation cycle.
Plans call for running the 3km WRF reflectivity nudging in real-time next summer in a domain covering the northeastern United States. Tests will be conducted with the 3km WRF domain initialization using different model analyses, including the RUC 13 km analysis in collaboration with NOAA/GSD scientists. Work will continue on code development and testing of the WRF-based reflectivity nudging. Also, the assimilation of VDRAS wind into the RTFDDA will be further evaluated.