Abstract

The operational network of WSR-88D systems is in place. These radars provide a large increase in performance and coverage over the radars they replaced—and in some locations are the first operational weather radars. The early years of experience using these radars have shown that they have contributed to increased weather warning and forecast performance. These systems must be regularly upgraded, however, to maintain a viable network, meet new requirements, and take advantage of new science and technology. In addition to the NEXRAD program’s standard modification and retrofit process, the NEXRAD agencies (Department of Commerce/National Weather Service, Department of Defense, and Department of Transportation/Federal Aviation Administration) have established a NEXRAD product improvement program to initiate major hardware upgrades to the WSR-88D. This paper describes the status of the NEXRAD program, improvements to the WSR-88D that have taken place, improvements that are planned, and a discussion of possible future changes to the WSR-88D and application of radar data.

1. Introduction

The Next Generation Weather Radar (NEXRAD) program has completed the development and installation of the network of Weather Surveillance Radar-1988 Doppler (WSR-88D) systems. The radars provide coverage for nearly the entire contiguous United States, Alaska, Hawaii, and at selected overseas locations. The radars provide information on storm structure and other atmospheric phenomena never before available to operational forecasters. The nation has already begun to reap the benefits of the WSR-88D network and the new capabilities provided by the radar.

The WSR-88D system must evolve, however, to meet changing mission requirements of the NEXRAD agencies to improve the efficiency and reliability of these radars, and to take advantage of advances in computer and radar engineering, and in hydrometeorological science and algorithm development. As a result, the NEXRAD agencies [Department of Commerce/National Weather Service (DoC/NWS), Department of Defense (DoD), and Department of Transportation/Federal Aviation Administration (DoT/FAA)] have committed to a long-term NEXRAD Product Improvement (NPI) program.

The management of the WSR-88D systems remains a triagency team effort, just as was the development, acquisition, and deployment. The NEXRAD Program Management Committee (PMC) provides broad operational management of the NEXRAD program. The PMC responsibilities cover all operational aspects during the life cycle of the radars to ensure that both common and agency-unique requirements are met. The PMC also has a high-level configuration management responsibility.

This paper provides a status update on the NEXRAD program, a summary of NPI activities and likely near-term projects, and a visionary look ahead to how the WSR-88D will support forecast and warning operations in the future.

2. NEXRAD program status

Crum and Alberty (1993) documented major aspects of the NEXRAD program history, an overview of the WSR-88D system, and deployment plans. The major milestones projected in that publication have been met. This section provides highlights of those milestones and of operational experiences from the early years of operating the WSR-88D systems.

a. WSR-88D network

The last of the 166 WSR-88D systems (158 operational radars) and 407 Principal User Processors (PUPs) were installed in 1997. Due to changing requirements of the NEXRAD agencies, some WSR-88D sites were added and others deleted from the original deployment plan. Figure 1 depicts the location of the operational WSR-88D sites in the contiguous United States. The DoC/NWS deployed 125 systems in the contiguous United States (5 are used for logistics support, training, and operational support). The DoD deployed 29 systems in the contiguous United States (3 are used for training), Guam, the Azores, Korea, and Japan. The DoT/FAA deployed 12 systems in Alaska, Hawaii, and Puerto Rico. The NEXRAD agencies are operationally using their WSR-88D systems and have decommissioned most of their earlier generations of weather radars.

Fig. 1.

Location of operational WSR-88D RDA units in the contiguous United States as of December 1997.

Fig. 1.

Location of operational WSR-88D RDA units in the contiguous United States as of December 1997.

b. WSR-88D role in improved forecasts and warnings

The WSR-88D has played an important role in the improvement of short-range forecasts and warnings for severe thunderstorms, tornadoes, and flash floods. Polger et al. (1994) were among the first to document that NWS forecasters using the WSR-88D had an improved severe weather warning performance, issued warnings for severe local storms and tornadoes with a longer lead time (on average), and had a greater probability of detecting severe local storms and tornadoes. Bieringer and Ray (1996) found similar results at selected sites in a separate study. In addition, a report by the National Research Council (1995) concluded there was a distinct improvement in storm warning statistics associated with the installation of WSR-88D systems. Figures 2 and 3 depict the upward trend of NWS severe local storms and tornado warning performance since the deployment of WSR-88D systems, beginning in 1991. While many factors can influence warning performance statistics [e.g., a strong NWS modernization emphasis on professional training, new Geostationary Operational Environmental Satellite (GOES) data, profiler data, reliance on spotter reports, year-to-year variability of the verification results], we suggest that the availability of WSR-88D data in forecast offices has had a major role in the upward trend in warning performance.

Fig. 2.

NWS probability of detection (POD) and false alarm ratio (FAR) for all severe local storms for 1986–96.

Fig. 2.

NWS probability of detection (POD) and false alarm ratio (FAR) for all severe local storms for 1986–96.

Fig. 3.

NWS lead time for all severe local storms and for tornadoes only for 1986–96.

Fig. 3.

NWS lead time for all severe local storms and for tornadoes only for 1986–96.

The WSR-88D has also proven to be an excellent tool in detecting and forecasting conditions other than those directly associated with severe storms. The increased sensitivity and resolution of the system have helped forecasters detect and forecast the time of passage of atmospheric boundary layer convergence lines (e.g., drylines, thunderstorm outflow boundaries, land-/sea-breeze fronts), even in the optically clear air. These convergence lines were not generally detectable with previous generations of weather radars. This has resulted in improved timing for the onset and duration of events such as wind shifts, convective storms, and heavy rain/flood events. The display capabilities (e.g., time lapse, improved background maps) and scientific algorithm output of the WSR-88D (e.g., precipitation accumulation, storm cell identification and forecast) have contributed to forecasts with improved timing and location of events. These have enabled forecast offices to issue more reliable forecasts, with more spatial and temporal specificity.

c. WSR-88D data dissemination

The WSR-88D has provided exciting new opportunities to study atmospheric structure and phenomena. The high quality and resolution of the data, combined with the ability to transmit and record the data digitally, have facilitated the widespread use of these data in research laboratories, universities, and the private sector.

The archived WSR-88D data best suited for research are termed “level II”. These are the high-resolution base data (reflectivity, mean radial velocity, and spectrum width) produced by the signal processor at the full spatial and temporal resolution of the WSR-88D system (Crum et al. 1993). The NEXRAD program has implemented network-wide, continuous recording of WSR-88D level II data on a “noninterfering” basis, when equipment and maintenance resources at the operational site permit. Over 60 000 level II, 4.7-GB, 8-mm tapes have been recorded and are available at the National Climatic Data Center (Crum 1995a). Copies of over 7000 data tapes have been produced for a wide range of uses (e.g., support to WSR-88D algorithm development, WSR-88D training, better understanding of atmospheric processes, bird and insect migration studies, cloud modification experiments, and meteorological graphical display development). Beginning with data recorded in late-1998, the level II data tapes will also include notch width information, clutter maps, adaptation data values, and system performance information. These data will be extremely useful to researchers who need to know the calibration and clutter information to better interpret development results (e.g., rainfall estimation from reflectivity data).

The demand for real-time access to WSR-88D data has also grown since the inception of the NEXRAD program. To support many users, but primarily the broadcast industry, the NEXRAD Information Dissemination Service has provided real-time access to a predetermined set of WSR-88D products from operational systems (Baer 1991). Recently, the demand for real-time WSR-88D data access has expanded to include the high-resolution base data. These data are needed to support the real-time needs of the NEXRAD agencies, government research laboratories, and universities working collaboratively with the NEXRAD agencies. Initial connections have been made available on a limited and experimental basis through the use of a stand-alone workstation, which serves as the interface between the WSR-88D and the external user (Crum and Kelleher 1997). This interface is the Radar Interface and Data Dissemination System developed by the National Severe Storms Laboratory (NSSL) (Rhue and Jain 1995). The need for external access to real-time WSR-88D base data is expected to continue to grow. Future configurations of the WSR-88D, as discussed below, will have the capability to provide external users real-time access to base data within the baseline architecture of the WSR-88D.

d. WSR-88D technical needs

Sophisticated algorithms and data quality control techniques to support warning and forecast operations in a robust, reliable manner are vital components of the weather forecast office tools required to meet the ever-changing NWS mission (Elvander et al. 1997). Based on operational experience, the NEXRAD agencies have identified technical areas where work is needed to improve or regionalize the initial suite of WSR-88D meteorological algorithms, add new algorithms to meet unfulfilled or new operational requirements, add other new capabilities, and improve data quality. The processes for improving the performance of WSR-88D algorithms, products, and data to meet the WSR-88D technical requirements have been implemented (Crum 1995b; Crum and Alberty 1993). Annually, the NEXRAD Technical Advisory Committee (TAC), identifies and prioritizes WSR-88D technical needs based upon their operational impact, scientific or technical feasibility, and cost effectiveness. The NEXRAD agencies then use the PMC-approved prioritized list of technical needs to allocate resources for development of technical improvements and their transition to the WSR-88D baseline. Below is the 1997 priority list of WSR-88D technical needs identified by the TAC:

  • evolution of WSR-88D hardware and software to implement advances in technology and science;

  • data quality improvements;

  • system performance assessment;

  • WSR-88D level II data archive of storm phenomena;

  • data acquisition rate needs and strategies;

  • severe weather detection and forecasting;

  • precipitation analysis techniques;

  • feature detection, tracking, and forecasting techniques;

  • wind analysis techniques;

  • turbulence analysis techniques;

  • tropical cyclone analysis techniques;

  • icing analysis techniques;

  • data compaction and transmission techniques;

  • interpretative techniques/human interface techniques.

e. Changes to WSR-88D system baseline

Changes to the WSR-88D system baseline are controlled by a NEXRAD triagency configuration management process. Changes are made to meet new requirements, correct deficiencies, and enhance the performance of the system. Below is a discussion of the major changes that have been made or are in development.

1) Major software changes

The WSR-88D Operational Support Facility (OSF) (Crum and Alberty 1993) is the focal point for developing, testing, and releasing new WSR-88D software. The OSF has released several versions of software to the network of WSR-88D systems. The objectives of the releases have increasingly matured from being primarily corrective in nature to adding significant new capabilities, products, product displays, and meteorological algorithms. For example, in 1996 among other changes made, a new storm cell identification and tracking algorithm, and a new hail detection algorithm, were implemented. In 1998, the WSR-88D software will be modified to support an increase in the narrowband communication rates from 9.6 to 14.4 kbps. Additionally, an anomalous propagation mitigation algorithm, a new tornado detection algorithm, and a remote maintenance monitoring capability will be added to the software baseline. Further, human factors have received considerable attention in the NEXRAD program. New display techniques such as trends of radar-measured quantities and calculation of rotational-velocity shear have been implemented. Even more significant improvements to the user interface will be implemented in the future.

2) Major hardware changes

As with the software baseline, significant hardware changes have been made. During the deployment phase, the WSR-88D system design was improved to include changes to the pedestal; the central processing units of the Radar Data Acquisition (RDA), Radar Product Generation (RPG), and Principal User Processor (PUP) computers; and the narrowband communication protocol. Lightning/grounding protection systems have been upgraded. A transition power system is being added to improve operational availability and reduce maintenance required due to power fluctuations. Finally, a capability for operators and maintainers to remotely troubleshoot system problems and issue corrective commands is being added. Hardware changes will continue to occur through the operational life of the system.

f. Lessons learned and research/development under way

Some WSR-88D systems are located where no meteorological radar observations have been made before. As a result, forecasters at these offices have required time to adjust their interpretation of meteorological algorithm output and adaptable parameters to their specific local conditions. Research using archived data is under way to develop new values for adaptation parameters (e.g., number of pattern vectors required for mesocyclone indication, shear threshold for tornado indication, and ZR relationships for precipitation accumulations) to be used in existing WSR-88D meteorological algorithms and to develop improved and new algorithms to better address the wide range of WSR-88D operating environments.

As more experience has been gained in operating the WSR-88D, the need for improved data quality control has become apparent. For example, manual intervention to optimize clutter suppression parameters is laborious at best and, in most cases, even with the automated techniques, does not result in fully suppressing all ground clutter. More effort to mitigate the effects of clutter and to improve the human–computer interface for modifying clutter rejection tools is under way (Pratte et al. 1997). Maintaining the network of radars to an absolute calibration has also been a formidable task. New techniques have been provided to field technicians to improve the reliability and network consistency of the calibration of the WSR-88D systems. Automated detection and mitigation of anomalous propagation continues to be an area requiring improved capability, especially to improve precipitation estimates.

The WSR-88D has provided a new capability for forecasters to track radar estimates of precipitation accumulation. Even in cases where radar estimates of the absolute amount of precipitation are incorrect, the location and gradients of precipitation accumulations are usually accurately depicted and very useful to forecasters. The current capability has provided a large step forward in providing the public improved flood and hydrologic information. However, much work remains to obtain consistent and reliable accumulation estimates (Hunter 1996). Work is in progress to improve precipitation estimates, especially for heavy precipitation accumulations. The focus of this work is on integration of real-time rain gauge information to help determine any radar estimation bias and through improved understanding of how to adjust the algorithm adaptation parameters for different types of precipitation processes.

The NEXRAD agencies have several development projects under way to improve and add new algorithm capabilities to the WSR-88D baseline in response to the technical needs. Foremost are the efforts to better address the “Doppler dilemma,” the inherent characteristic of Doppler radars whereby increasing the maximum unambiguous velocity decreases the maximum unambiguous range and vice versa (Rinehart 1991). Mitigating the effects of the Doppler dilemma has been and continues to be a major concern of the NEXRAD program. This problem can obscure severe storm signatures and clear air phenomena. The NEXRAD agencies have invested in a long-term project to provide potential hardware and/or software solutions to this problem. Methods being examined include multiple-pulse-repetition frequency, spectral decomposition, intermittent pulse staggers, and random pulses. Many of the possible solutions will require increased processing capacity for the RPGs and RDAs.

Other algorithm efforts are under way. In response to the users’ needs for detection and estimation of snowfall, the NEXRAD program is funding the development of a snow accumulation algorithm that capitalizes on the increased sensitivity of the WSR-88D. The algorithm will be adaptable to various climatic and geographical regimes. This algorithm is expected to serve as a tool for forecasters to warn the public of snowfall accumulations and provide liquid water equivalent information to assist in runoff estimation for river basins. Projects are under way to improve the detection and classification of mesocyclones, develop a predictive and diagnostic damaging downburst (convective) algorithm, implement an algorithm to detect and track gust fronts, and improve rainfall accumulation estimates.

The FAA, NWS, and U.S. Weather Research Program are sponsoring research activities to develop very short period forecasting techniques for thunderstorms, rainfall, and snowfall that use the WSR-88D. This has led to the development of WSR-88D algorithms to detect convergence lines, retrieve single Doppler wind fields, extrapolate echoes, and develop storm growth and decay algorithms that use a variety of operational datasets including the WSR-88D and GOES.

3. NEXRAD product improvement program

The WSR-88D system is an outstanding success and has contributed to better forecasts and warnings for severe weather. The system hardware and software, however, generally represent technology and scientific knowledge as of the mid-1980s, the period when the system design was finalized. In the decade since, computer technology has advanced on two fronts: basic processing capacity and widespread use of common standards, that is, open systems. There has also been growth in weather radar engineering (e.g., polarization) and in the understanding of relationships between radar-observed data, data obtained from other remote sensors (e.g., satellite, lightning detection networks), and corresponding weather features (e.g., tornadoes and downbursts). Much of the new science and technology require greater signal and data processing capacities than the existing WSR-88D platform can provide or be economically expanded to support. From its inception, the NEXRAD program has recognized that allowances must be made for system growth. In the 1981 NEXRAD Joint Operational Requirements (JOR) document, this need was expressed in terms of 1) long-term service as the nation’s primary weather radar network for at least 20 years, 2) necessity for software additions and refinements to achieve the highest level of automated warning assistance, 3) need to incorporate new meteorological technology, and 4) need to address increased communications loading as more information is derived from WSR-88D systems.

To meet the need for orderly system evolution under existing NEXRAD change management procedures, the NEXRAD agencies have established the NPI program as a long-term subactivity of the overall NEXRAD program (Saffle and Johnson 1997). The NPI program management is led by the NWS Office of Systems Development, but individual NPI projects enlist the participation and support of all of the NEXRAD agencies. The NEXRAD agencies have developed an update of the JOR to reflect changes in mission responsibilities and other weather systems modernization programs. This document, the Tri-Agency Requirements for Operational Use of Weather Radar Data (NEXRAD 1997a), will guide the selection and execution of NPI projects.

a. Open systems project

The first NPI project is under way, focusing on minimizing the long-term maintenance costs and improving the processing capacity of the WSR-88D RPG. The development and maturity of the open systems concept for computer hardware and software provide the basis for achieving expanded capability in a manner that 1) provides the processing and communications capacities necessary for full utilization of WSR-88D data with other modernized weather systems; 2) provides the greater capability at much less cost than purchasing additional components for the current system; 3) provides capacity to implement future, highly complex data and signal processing algorithms; 4) enables a more robust development-to-operations partnership between the NEXRAD agencies and development groups; and 5) puts the government on a path to postpone indefinitely the expense of a major systems acquisition for an entirely new national weather radar system. The individual functional areas of the WSR-88D are being addressed as separate tasks, but with overall systems engineering oversight by the OSF to ensure architectural and design consistency among the developments. The project schedules for the open systems tasks are presented in Fig. 4.

Fig. 4.

Major milestones for the NEXRAD NPI.

Fig. 4.

Major milestones for the NEXRAD NPI.

1) Open systems RPG

The NPI open systems RPG (ORPG) task includes rehosting existing RPG functionality to a standards-based, commercial-off-the-shelf (COTS) computer, and the development of a software architecture employing distributed processing, fault tolerance, and dynamic load allocation. The ORPG partnership includes the NSSL for software development; the NWS Office of Systems Development for communications infrastructure development; the OSF for systems engineering, hardware specification, test and evaluation, technical documentation, and field deployment, and other components of the NWS Office of Systems Operations for procurement support. The NEXRAD technical requirements document has been updated to be the WSR-88D system specification (NEXRAD 1997b) to better support the documentation of system specification changes related to the Open Systems upgrade and other NPI projects.

Major features of the initial version of the ORPG will include 1) a graphical user interface to facilitate use of WSR-88D functionality, 2) sufficient processing capacity to meet immediate NEXRAD agency needs and projected needs for at least five years after the system is implemented, 3) support for distribution of base data to multiple external users, and 4) local area network interface support to enable future, high-speed connections to external systems. The ORPG is scheduled to be deployed in the 1999–2001 time frame. More detailed information on the ORPG development can be found in a collection of 12 papers in the preprints of the 13th International Conference on Interactive Information and Processing Systems (Session 8).

2) Open systems RDA

The NPI open systems RDA (ORDA) task includes the development of a standards-based COTS platform similar to the ORPG for the host computer part of the RDA and replacing the current hard-wired signal processor/programmable signal processor combination with a modern digital signal processing module. The NSSL will develop the application software, the technical specifications for the signal processors, and a preproduction version of the ORDA. The OSF and Office of Systems Operations roles will be similar to those in the ORPG. The ORDA implementation will support more complex signal processing approaches to mitigate impacts of range and velocity folding, although the selection of a particular approach has not yet been made. Improvements in data quality, residual clutter removal, and anomalous propagation removal are expected benefits of the ORDA. The ORDA will also support the development of new capabilities to provide a given NEXRAD agency with a “tailored” feed of base data. For example, the FAA would like to have its products generated from base data that have been processed with nationally consistent, aggressive clutter filtering parameters, while the NWS requires flexibility in these settings to prevent any degradation of rainfall estimation or weather detection through overly aggressive clutter filtering.

In addition to such benefits as those discussed above, the ORDA is a necessary step to the possible future modification of adding a polarization capability to the WSR-88D. The expanded signal and data processing capacities of the ORDA are a necessary foundation for polarization diversity.

3) Open systems PUP

The open systems PUP (OPUP) will be developed and implemented as part of the NPI program. The NSSL will develop the application software and the OSF will perform the system engineering, hardware specification, test and evaluation, technical documentation, and field deployment in a manner similar to that for the ORPG. At this time, the only sponsor for the OPUP is the DoD. The OPUP will not only host the functionality of the existing PUP, but will also provide the DoD with a path to better interoperability with, and eventual incorporation of OPUP within, their future weather system modernization platforms. The NWS and the FAA plan to incorporate WSR-88D product displays and requests within the Advanced Weather Interactive Processing System (AWIPS) and the Weather and Radar Processing (WARP) System, respectively.

b. Transmitter and receiver evolution

Implementation of open systems upgrades will provide the WSR-88D with processing platforms capable of supporting advances in the basic radar transmitter, receiver, and antenna technology. At this time, the NEXRAD agencies believe the most likely candidate for NPI in these areas is the addition of polarization capability.

Other radar engineering technology advances being monitored for potential NPI projects include solid-state transmitters, digital receivers, and phased array antennas. The current work in these areas offers the NEXRAD community exciting prospects for future operational benefits.

4. Future

a. Radar enhancements

Present research radars offer new capabilities that may become part of the WSR-88D within 5–10 years. Most prominent among these is polarization diversity. In fact, the WSR-88D operated by NSSL is being modified for this purpose. Zrnić (1996) has discussed some of the operational applications that may be forthcoming with polarimetric radar. Two of the most promising variables are the differential reflectivity between the horizontal and vertical pulses (ZDR) and differential phase change between the horizontal and vertical polarizations (ϕDP). The differential phase has been shown to be particularly effective in estimating high rainfall rates (Ryzhkov and Zrnić 1996). Differential reflectivity combined with horizontal reflectivity can be used to identify hail and estimate hail size (Aydin et al. 1986). Polarimetric variables have also been shown to be useful for identifying precipitation particle type, ground clutter, and biological targets.

Bistatic radars (Wurman et al. 1993), which are low-cost radar receivers, can be distributed around a WSR-88D to obtain dual-Doppler wind fields. To be practical, this cannot be done until the WSR-88D has been made a polarimetric radar since the area of dual Doppler winds is quite restricted for horizontally polarized pulses, but extensive for vertical pulses.

In the 10–20-yr time frame, much higher time and space resolution may be possible using new technologies now emerging in the research community such as wide bandwidths, pulse compression, and electronically scanned phased array antennas.

b. Directed research

The national loss of property and lives caused by weather disasters such as hurricanes, floods, snowstorms, and severe thunderstorms/tornadoes is increasing at a rapid rate. Policy makers have stated a need to reduce the damage from natural disasters (National Mitigation Strategy 1995). The research and operational community is responding to this need. Work is particularly evident in the U.S. Weather Research Program where National Oceanic and Atmospheric Administration, university, and National Center for Atmospheric Research scientists are beginning to work directly with operational forecasters to solve pressing forecast problems such as quantitative precipitation forecasting and hurricane landfall (Emanuel et al. 1995).

Major advances are taking place in the ability of computers to economically manipulate, integrate, display, and communicate datasets. These include initializing numerical models with radar and satellite data. The modernization of the national network of observation facilities is beginning to pay off in improved forecasts and warnings of hazardous weather. Continued improvements are expected as researchers and forecasters gain experience with the modernized data platforms.

c. Data integration

We envision an ever-growing trend toward integrating WSR-88D data with data from other sources in efforts to improve the detection, warning and very short period forecasting activities. These other data sources include satellite, surface stations, rawinsonde, profilers, lightning detection networks, and numerical models. For this purpose, interactive displays are required so that a forecaster can easily overlay and time lapse data on a common grid. Data displays for this purpose have already been developed and utilized by the research community (Corbet et al. 1994) and tested operationally at several weather forecast offices (WFOs) (Johnson et al. 1995; Roberts et al. 1997).

The NWS has begun to implement such displays at a few experimental sites and plans to deploy the capability within AWIPS at all WFOs. The ORPG will aid the end-to-end forecast process by contributing to the integration of multiple data types into the AWIPS workstation. The FAA has two programs to integrate WSR-88D data: WARP, and the Integrated Terminal Weather System (ITWS) (Evans and Ducot 1994). The WARP will mosaic the WSR-88D data within each of the FAA regions and the ITWS will combine WSR-88D data with other sensor data in the terminal area.

d. Automated forecast systems

The very short period forecasting of thunderstorms and precipitation is extremely human intensive and it is unreasonable to expect humans to perform these functions routinely without automated assistance (Wilson and Mueller 1993; Rothfusz et al. 1997). As Wilson and Carbone (1984) hypothesized, the ideal nowcasting system is one in which the computers conduct the highly repetitive and computational intense functions. This allows the forecaster more time to use his/her physical reasoning and pattern recognition capabilities to assess data quality, evaluate automated forecast material, and apply broad meteorological reasoning to the forecasts.

Under FAA sponsorship, completely automated techniques for very short-period forecast of thunderstorms and snowfall are under development that will integrate the data from the various collection platforms and numerical models (Wolfson et al. 1997; Mueller et al. 1997; Lindholm et al. 1997). Similar techniques are under development for testing in the NWS, however, with the philosophy that the automated output is a tool for the forecaster and the human would have the ability to easily edit the input data, forecast rules, and machine forecast.

e. Programmatic efforts

The NEXRAD agencies must provide programmatic infrastructure to ensure that new data and scientific techniques are validated and implemented into forecast operations in a timely fashion. The National Oceanic and Atmospheric Administration/NWS management is strongly supporting efforts to solve forecast and warning problems with an integrated rather than a single-sensor approach. Infrastructure to enhance such technology transfer processes must include common development and testing tools, as well as organizational cultures that encourage the use of these tools and participation in multiagency collaborative efforts. Two NWS programs that address this area are the Collaborative Science, Technology and Applied Research and the System for Convection Analysis and Nowcasting (SCAN) (Smith et al. 1998). SCAN involves efforts of the NWS, NSSL, and National Center for Atmospheric Research scientists to work together to develop an AWIPS functionality ecompassing existing algorithms and products from these groups.

5. Summary

The NEXRAD Program has completed the installation of the United States’ first true digital network of weather radars. These radars have been providing forecasters with the tools to improve forecasts and warnings, providing the public with nationwide radar coverage, and providing improved real-time and archive datasets to aid researchers in better understanding the atmosphere. The NEXRAD agencies have recognized the need to refine the capabilities of the WSR-88D systems and to add new capabilities to meet new requirements. The planned improvement of WSR-88D software and hardware has already begun. The NEXRAD agencies have also committed to a longer-term program that will significantly expand and improve the computational capability of the WSR-88D via the NEXRAD Product Improvement Program. These projects (the open systems RPG, open systems PUP, and open systems RDA) will help ensure that the quality of WSR-88D data will continue to improve and provide a platform for the next generation of algorithms, which will combine the data streams from multiple remote sensors. Even more dramatic changes are likely to occur in the future. The possible implementation of polarization diversity into operations and the merging of data from diverse sensors are likely to provide a large benefit to operational forecasting.

Acknowledgments

The support of the NEXRAD agencies to the OSF and NPI program is gratefully acknowledged. The authors thank Paul Polger for providing the NWS warning performance results. Art Thomas provided Fig. 1. Dwayne Johnson provided the graphics support for Figs. 2–4. The review comments of Jim Belville, Rich Vogt, Don Burgess, Art Levy, and the anonymous reviewers were very helpful and appreciated.

REFERENCES

REFERENCES
Aydin, K., T. A. Seliga, and V. Balah, 1986: Remote sensing of hail with a dual linear polarization radar. J. Appl. Meteor., 25, 1475–1484.
Baer, V. E., 1991: The transition from the present radar dissemination system to the NEXRAD Information Dissemination Service (NIDS). Bull. Amer. Meteor. Soc., 72, 29–33.
Bieringer, P., and P. S. Ray, 1996: A comparison of tornado warning lead times with and without NEXRAD Doppler radar. Wea. Forecasting, 11, 47–52.
Corbet, J., C. Mueller, C. Burghart, K. Gould, and G. Granger, 1994:Zeb: Software for integration, display, and management of diverse environmental datasets. Bull. Amer. Meteor. Soc., 75, 783–792.
Crum, T. D., 1995a: An update on WSR-88D level II data availability. Bull. Amer. Meteor. Soc., 76, 2485–2487.
——, 1995b: Enhancing and regionalizing the WSR-88D meteorological algorithms. Preprints, 27th Conf. on Radar Meteorology, Vail, CO, Amer. Meteor. Soc., 167–169.
——, and R. L. Alberty, 1993: The WSR-88D and the WSR-88D Operational Support Facility. Bull. Amer. Meteor. Soc., 74, 1669–1687.
——, and K. K. Kelleher, 1997: Real-time distribution of WSR-88D base data, present and future. Preprints, 28th Conf. on Radar Meteorology, Austin, TX, Amer. Meteor. Soc., 282–283.
——, R. L. Alberty, and D. W. Burgess, 1993: Recording, archiving, and using WSR-88D data. Bull. Amer. Meteor. Soc., 74, 645–653.
Elvander, R. C., L. D. Johnson, and S. M. Holt, 1997: The NEXRAD algorithm survey and analysis for the rehosted radar product generator (RPG) subsystem. Preprints, 13th Int. Conf. on Interactive Information and Processing Systems for Meteorology, Oceanography, and Hydrology, Long Beach, CA, Amer. Meteor. Soc., 228–233.
Emanuel, K., and Coauthors, 1995: Report of the First Prospectus Development Team of the U.S. Weather Research Program to NOAA and the NSF. Bull. Amer. Meteor. Soc., 76, 1194–1208.
Evans, J. D., and E. R. Ducot, 1994: The Integrated Terminal Weather System (ITWS). Lincoln Laboratory Journal, Vol. 7, No. 2, MIT Lincoln Laboratory, Lexington, MA, 449–474.
Hunter, S. M., 1996: WSR-88D rainfall estimation: Capabilities, limitations, and potential improvements. Natl. Wea. Dig., 20 (4), 26–38.
Johnson, J. T., and Coauthors, 1995: Operational testing of enhanced WSR-88D algorithms and display concepts in National Weather Service offices. Preprints, 27th Conf. on Radar Meteorology, Vail, CO, Amer. Meteor. Soc., 170–172.
Lindholm, T. A., F. Hage, C. Wade, and R. M. Rasmussen, 1997: Weather support to ground deicing decision making system (WSDDM) real-time snowfall rate display, operational utility, and benefits. Preprints, Seventh Conf. on Aviation, Range, and Aerospace Meteorology, Long Beach, CA, Amer. Meteor. Soc., 152–157.
Mueller, C. K., and Coauthors, 1997: The Memphis ITWS convective forecasting collaborative demonstration. Preprints, Seventh Conf. on Aviation, Range, and Aerospace Meteorology, Long Beach, CA, Amer. Meteor Soc., 249–257.
National Mitigation Strategy, 1995: Partnerships for building safer communities. Federal Emergency Management Agency, Washington DC, 32 pp. [Available from National Weather Service, Silver Spring, MD 20910.]
.
National Research Council, 1995: Assessment of NEXRAD Coverage and Associated Weather Services. National Academy Press, 104 pp.
NEXRAD, 1997a: Tri-agency requirements for operational use of weather radar data. National Weather Service, 13 pp. [Available from WSR-88D Operational Support Facility, Norman, OK 73072.]
.
——, 1997b: WSR-88D system specification (draft), November 1997. WSR-88D Operational Support Facility, 187 pp. [Available from WSR-88D Operational Support Facility, Norman, OK 73072.]
.
——, 1981: NEXRAD joint operational requirements (JOR), January 1981. NEXRAD Joint System Program Office, 47 pp. [Available from WSR-88D Operational Support Facility, Norman, OK 73072.]
.
Polger, P. D., B. S. Goldsmith, R. C. Przywarty, and J. R. Bocchieri, 1994: National Weather Service warning performance based on the WSR-88D. Bull. Amer. Meteor. Soc., 75, 203–214.
Pratte, F., D. Ecoff, J. VanAndel, and R. J. Keeler, 1997: AP clutter mitigation in the WSR-88D. Preprints, 28th Conf. on Radar Meteorology, Austin, TX, Amer. Meteor. Soc., 504–505.
Rhue, D. T., and M. H. Jain, 1995: A RISC-based WSR-88D interface and workstation. Preprints, 11th Int. Conf. on Interactive Information and Processing Systems for Meteorology, Oceanography, and Hydrology, Dallas, TX, Amer. Meteor. Soc., 235–239.
Rinehart, R. E., 1991: Radar for Meteorologists. Knight Printing, 334 pp.
Roberts, W. F., P. C. Kucera, C. M. Lusk, and D. C. Walker, 1997: Operational use of WFO-Advanced at the Denver WSFO. Preprints, 13th Int. Conf. on Interactive Information and Processing Systems for Meteorology, Oceanography, and Hydrology, Long Beach, CA, Amer. Meteor. Soc., 340–343.
Rothfusz, L. P., S. K. Rinard, J. T. Johnson, and M. Valverde, 1997:Lessons learned from the 1996 Olympics weather support: The operational impact of a comprehensive collection of cutting-edge forecast, warning and observational tools. Preprints, 13th Int. Conf. on Interactive Information and Processing Systems for Meteorology, Oceanography, and Hydrology, Long Beach, CA, Amer. Meteor. Soc., 39–42.
Ryzhkov, A., and D. Zrnić, 1996: Assessment of rainfall measurement that uses specific differential phase. J. Appl. Meteor., 35, 2080–2090.
Saffle, R. E., and L. D. Johnson, 1997: NEXRAD product improvement overview. Preprints, 13th Int. Conf. on Interactive Information and Processing Systems for Meteorology, Oceanography, and Hydrology, Long Beach, CA, Amer. Meteor. Soc., 223–227.
Smith, S. B., and Coauthors, 1998: The System for Convection Analysis and Nowcasting (SCAN). Preprints, 16th Conf. on Weather Analysis and Forecasting, Phoenix, AZ, Amer. Meteor. Soc., J22–J24.
Wilson, J. W., and R. Carbone, 1984: Nowcasting with Doppler radar:The forecaster–computer relationship. Nowcasting II, K. Browning, Ed., European Space Agency, 177–186.
——, and C. K. Mueller, 1993: Nowcasts of thunderstorm initiation and evolution. Wea. Forecasting, 8, 113–131.
Wolfson, M. M., C. K. Mueller, and M. D. Eilts, 1997: Convective weather forecasting for FAA applications. Preprints, Seventh Conf. on Aviation, Range, and Aerospace Meteorology, Long Beach, CA, Amer. Meteor. Soc., 238–243.
Wurman, J., S. Heckman, and D. Boccippio, 1993: A bistatic multiple-Doppler radar network. J. Appl. Meteor., 32, 1802–1814.
Zrnić, D. S., 1996: Weather radar polarimetry—Trends toward operational applications. Bull. Amer. Meteor. Soc., 77, 1529–1534.

Footnotes

Corresponding author address: Dr. Tim Crum, WSR-88D Operational Support Facility, 1200 Westheimer Drive, Norman, OK 73069.