COSTS AND BENEFITS OF USING THE AIR FORCE OVER-THE-HORIZON RADAR SYSTEM FOR ENVIRONMENTAL RESEARCH AND SERVICES
T. M. Georges
NOAA Technical Memorandum ERL ETL-254
July 1995
EXECUTIVE SUMMARY
NOAA has been testing oceanographic and other remote sensing applications of the U.S. Air Force over-the-horizon air defense radar system known as OTH-B, which has now been deactivated. The results of these tests show sufficient promise that an assessment of the costs and benefits of operating the system for environmental research and services is now warranted. The incremental cost to NOAA of such operations is believed to be reasonable, only because the Air Force is required to maintain the OTH-B radars in an operable state called "warm storage."
An informal cost study was carried out in cooperation with the Air Force OTH-B program office and Lockheed Martin, the present contract radar operator. The study focused on finding the most cost-effective way to operate the three East Coast OTH-B radars for a nominal ten hours a week for NOAA tasks. As more radar functions were identified as unnecessary for NOAA operations, and as more functions were recognized as necessary for the Air Force to maintain system recoverability during warm storage, the incremental cost to NOAA began to decrease substantially from early estimates of $5-6M. The estimated add-on cost to perform NOAA ocean-monitoring tasks in all three Atlantic sectors is now between $1M and $1.5M per annum. Two cost drivers were identified: added radar crew and electronic equipment repair. Even lower operating costs could be achieved if labor and administrative costs were tightly controlled, and if radar staff capable of performing multiple functions were employed. NOAA could reimburse the Air Force for the cost increment.
Assessments of the tangible benefits of limited operation were solicited from the scientific and service communities, including NOAA operational units. Fifteen distinct environmental monitoring applications have been identified, including marine weather research and forecasts, calibration and validation of satellite sensors, solar-radar research, air-traffic control, ionospheric research, ocean circulation, search and rescue, and climate change. Marine-weather and air-traffic-control products could be delivered immediately, while the rest would require further development.
1. BACKGROUND
Since 1989, the U.S. Air Force and the U.S. Navy have permitted NOAA to conduct ocean-monitoring tests with their over-the-horizon (OTH, or skywave) air-defense radar systems (Georges and Thome, 1990; Georges and Harlan, 1994a, 1994b; Georges et al., 1993; Harlan and Georges, 1994; Georges and Harlan, 1995; Georges et al., 1995). These tests have demonstrated the feasibility of using OTH radar to map surface winds, waves and currents over very large ocean areas in near real time. Such a mapping capability could fill large gaps in basic meteorological data over the ocean and thus advance our understanding of marine meteorology and ocean circulation--an important part of NOAA's mission.
The Air Force OTH-B East Coast Radar System (ECRS) in Maine is of particular interest to NOAA because its coverage includes most of the North Atlantic Ocean and the Caribbean Sea, ocean regions that affect the weather and climate of North America and Europe. For example, its ability to map ocean-surface winds with mesoscale resolution, on demand and through clouds and rain, over synoptic-scale ocean areas cannot be matched by existing or planned sensors. With the end of the Cold War threat, the Air Force has deactivated the ECRS, but it intends to maintain the system in an operable state called "warm storage," in case it is needed again for air-defense purposes. In such a state, it is believed that the ECRS could be operated for environmental research and services at modest incremental cost. The possibility of redirecting this unique radar system to serve NOAA's ocean-monitoring mission has prompted an examination of the costs, benefits, and technical issues involved.
The OTH-B radar system was originally deployed to provide round-the-clock early warning of Soviet bomber and cruise missile attacks on North America. Three radars were deployed on the U.S. West Coast, and three were deployed in Maine to cover the approaches to the U.S. East Coast (Figs. 1 and 2). Both have been operated under contract with the USAF Air Combat Command (ACC) by the Lockheed Martin Company (formerly Martin Marietta Services), and were staffed by military personnel from the United States and Canada. Following the end of the Cold War in 1991, the West Coast Radar System (WCRS) was shut down, and the East Coast Radar System (ECRS) was redirected to perform counter-narcotics surveillance for 40 hours per week. This scenario became known as a limited operations and permitted NOAA to piggyback its environmental monitoring tests. In 1995, the transition to full contractor operations was completed.
Limited operation of OTH-B has been a point of contention between the Defense Department and the Congress since the end of FY 93. The Air Force terminated ECRS operations at the end of FY 93 and again at the end of FY 94, but limited operations were subsequently restored by Congress, partly to permit NOAA to continue its assessment of weather-related applications. In the FY-95 Defense Appropriation, the Conference Committee directed the Air Force to continue OTH-B operations and stated: "The Committee emphasizes that FY-96 funding would be contingent upon (1) receipt and positive review of the cost-benefit analysis directed by Secretary Aspin; (2) receipt and positive review of a report by NOAA regarding the use of OTH-B to support its mission; and (3) determination by the President, in consultation with the Secretaries of Defense and Commerce, as to the future of this program."
At a Pentagon meeting on October 4, 1994, Deputy Assistant Secretary of Defense (C3I) Deborah Castleman agreed with NOAA Chief Scientist Kathryn Sullivan to cooperate in assessing the costs and benefits of using OTH-B in support of NOAA's mission. In the words of the meeting minutes: "To facilitate continued weather research using over-the-horizon radars, the parties agreed to follow up on a two-track plan. On the first track, the OTH-B Research Consortium, made up of members of the scientific community, would approach the drug enforcement community to pursue research efforts on the [Navy's] relocatable radars. On the second track, NOAA and Air Force contracting representatives would develop cost estimates for a minimal bare-bones capability in FY-95 that supports scientific research at the Bangor, ME site, taking into account Air Force warm storage considerations." This report is a direct result of that agreement.
Section 2 of this report summarizes the joint efforts of NOAA, the contract radar operator, and Air Force contract personnel to devise a cost-effective way for NOAA and others to use the OTH-B as an environmental observatory, while satisfying the Air Force requirement to warm store the facility for future recall. The add-on cost to NOAA of such operations is estimated.
Section 3 of this report contains specific recommendations from the scientific community about civilian research and services to which OTH-B might be applied. The social, economic, and scientific benefits of these applications are quantified where possible. The Appendix contains more detailed supporting documentation from the scientific and user communities.
Recent NOAA tests using the two U.S. Navy ROTHR radars have successfully mapped ocean surface current vectors for the first time. These results will be reported elsewhere.
Fig. 1. One of three OTH-B transmitting antennas near Moscow, Maine. Each array is 1.1 km long and radiates one-million watts of power in the 5-28-MHz frequency band. The 7.5 transmit beam is steerable over 60 in azimuth. The total ocean area covered is approximately 45-million km2.
Fig. 2. The nominal coverage of the OTH-B East Coast Radar System includes most of the North Atlantic Ocean and the Caribbean Sea.
2. COSTS
Many barriers and pitfalls obstruct any attempt to estimate the cost of using a Defense Department asset for civilian purposes. Not the least of these are the barriers to open communication imposed by federal contracting regulations and by the competitive environment of the defense procurement system. It is no secret that the final cost of a defense system can escalate to many times the cost of an equivalent civilian system, and that programs that seem small by Defense Department yardsticks can be very large by civilian standards. Our initial attempts [Georges and Harlan, 1994a] to estimate the cost of using OTH-B to acquire environmental data were based on the costs of performing similar functions under Defense Department contract. These estimates produced greatly inflated numbers ($5-6M per year) well beyond NOAA's reach. At first, it appeared that OTH-B would be so expensive to operate and maintain that no civilian agency could afford to do so.
To obtain more realistic numbers quickly but legally, it was necessary to bypass the rigor, formality, and expense of a full-blown solicitation of bids. Instead, we decided to establish an informal atmosphere in which the present contract operator of the OTH-B could sit down with representatives from the Air Force OTH-B Program Office and from NOAA to openly discuss common-sense ways to arrive at an OTH-B operating scenario that NOAA could afford. The resulting cost estimates are therefore little more than informed guesses and must not be considered in any way binding on the parties involved, or as excluding any other parties from consideration. Nevertheless, almost all technical and logistical aspects of operating OTH-B for NOAA's purposes were discussed in depth.
2.1 Cost Assessment Meeting
On 14 and 15 December 1994, a meeting was held at the OTH-B Operations Center in Bangor, Maine. This location was chosen so that radar engineering and support personnel could be called upon to answer questions requiring their expertise. The principals were:
Dr. Tom Georges, NOAA Environmental Technology Laboratory
Mr. Jordan Matejceck, NOAA Systems Program Office
Mr. Marvin A. Franz, USAF Air Combat Command (ACC) Contract Programs Squadron (CPS) Program Management Office (PMO)
Lt. Col. Vincent Azzarelli, Commander, Det. 1, USAF, Northeast Air Defense Sector
Mr. Ray Healy, Lockheed Martin Project Manager
Mr. Larry Meyer, Lockheed Martin, OTH-B Operations Manager
Capt. Henrick Smith, USAF/ACC, Ground-Air Control Systems Operations Manager
Capt. Zena Culp, USAF/ACC, Air Defense Systems Management
Capt. Frank Faupel, USAF/ACC, OTH-B Contracting Officer
Mr. Milton Cates, USAF/ACC, Logistics Manager
The principals quickly agreed that their task would be to examine and estimate the cost to NOAA of operating the east-coast OTH-B radars for a nominal NOAA mission. The cost to NOAA would be the total cost of the labor and materials required for such operations, over and above those required to warm store the system. Although we hoped that the costs of the proposed operation would be closer to the cost of warm storage than to the cost of limited military operations, this scenario came to be known as modified limited operations. As more radar functions were identified as unnecessary for NOAA operations, and as more functions were recognized as necessary for system recoverability during warm storage, the incremental cost to NOAA began to decrease substantially from the early estimates of $5-6M.
2.2 A Nominal NOAA Task
For the purposes of this exercise, a nominal "straw man" NOAA task was posed. This task would involve collecting oceanographic data for about one hour at the beginning and at the end of each eight-hour shift, for five days a week (10 hours per week), in all three east-coast radar sectors. Low transmitter power would normally be used, meaning 10% of maximum. Some surge efforts would be required, for example, during hurricane season, when seven-day operations would be required for approximately eight weeks. Other brief tests would occasionally be required for exploratory research, not to exceed four hours of low-power radar operation per day for 20 days. During the remainder of the shifts, the radar crew would be available to perform service, maintenance, and tests required for warm storage.
The costs for more or fewer hours of NOAA operations do not scale linearly, mainly because most of the radar crew would normally perform multiple tasks over 8-hour/5-day weeks. Doubling the number of NOAA hours, for example, would increase operator workload but would not appreciably increase costs, because warm-storage tasks could be interleaved with NOAA tasks. Weekend operations, on the other hand, would increase the number of operator hours, unless more flexible work schedules could be arranged. Such scheduling details would depend on the flexibility of the Air Force and the warm-storage crew and will not be precisely known until the warm-storage phase is under way.
2.3 What is Warm Storage?
The precise meaning of warm storage is crucial for any estimate of NOAA's incremental costs, because warm-storage costs are paid by the Air Force, and the increment to operate the radars would be paid by NOAA. The maintenance functions that fall under the warm-storage specification would be required to operate the radar, but NOAA would not be responsible for them.
In general terms, warm storage is a state that preserves the physical and electrical integrity of the radar system, so that it could be recalled to perform air defense tasks at some time in the future, should the threat arise. It is often assumed that warm storage includes "mothballing" the OTH-B hardware or moving it to some other location. This is not the case. In principle, the warm-stored system could be powered up at any time, given personnel qualified to do so. (In practice, equipment failures are expected after extended power-down.) Therefore, in the modified limited operations scenario, no cost savings are expected from warm storage tasks not performed. Maintaining the radar in operable condition requires, for example, maintaining fire and safety systems, keeping transmission lines pressurized with dry nitrogen, and keeping de-ionized water circulating through the transmitting tubes. Continuous security protection is required, as well as periodic maintenance and service approaching levels required for full operation, except that the radar and associated subsystems (e.g., computers) would not be turned on. To maintain a recall capability, it is believed that the Air Force expects to pay approximately $4.3M for warm storage of the East Coast Radar System (ECRS), with an expected staffing requirement of 29 to 33, including 17 security guards.
A more precise definition of warm storage (i.e., a baseline) must be specified before the incremental cost of NOAA operations can be determined with confidence. Such a precise specification does not yet exist and is being negotiated between the Air Force and Lockheed Martin. Given the USAF requirement that the radar system be recoverable in 24 months, Lockheed Martin maintains that it needs to preserve a certain level of resident system expertise, because of the intricacy and idiosyncrasies of the radar system. The USAF maintains that the system could be recovered in less than 24 months, including training a new operations and support staff, using only the existing system documentation. The resolution of this difference affects the cost to NOAA of limited radar operations, since NOAA would pay for the amount of resident engineering expertise over and above that required for warm storage. For the purposes of this exercise, we accepted the USAF definition of warm storage, which would result in higher costs to NOAA. If the Air Force warm-storage contract specifies true recoverability and retains the expertise, the incremental cost of NOAA's tasks would be much less.
2.4 Cost Drivers
Two high-cost items (cost drivers) appear to dominate the expenses of operating the radar in support of NOAA's tasks (that is, modified limited operations). These are the salaries of additional radar crew and the cost of repairing electronic equipment, mainly the 260 elemental radar receivers. The sum of the remaining costs is believed to be below the level of uncertainty in estimating these two cost drivers. The following estimates of these major costs use existing Air Force requirements as a starting point and could most likely be reduced by relaxing (for NOAA's purposes) military specifications for redundancy, reliability, and system performance.
2.5 Radar Manpower
Additional radar crew labor would be required to operate the radar for NOAA. Lockheed Martin recommended an addition of 11.5 man-years to what is presumed to be the warm-storage requirement, at a cost of approximately $633K, assuming an average rate of $55K (Lockheed Martin figure). Of the 11.5 additional man-years, 3 would be required at the transmit site, 5 at the receive site, and 3.5 at the operations center. This was called a "worst case" estimate, and the possibility was briefly discussed of reducing this requirement substantially by carefully selecting personnel who could perform both NOAA's tasks and those required in the warm-storage specification. The magnitude of any such reductions was not quantified, but it was agreed that there was room for reductions, particularly since NOAA's tasks would not occupy the radar engineers full time. The willingness of some of the crew to work part time was also discussed and is believed feasible. It is worth noting that during weekend operations for NOAA in the fall of 1994, only one person at each site (in addition to security guards) was needed to perform NOAA's 1-2 hour data collection. In round numbers, then, it seems reasonable to assume that $500K would suffice for the incremental cost of radar crew labor.
2.6 Equipment Repair
The second cost driver is the repair of electronic equipment, mainly the 260 elemental receivers. In warm storage, the receivers would be turned off, so no repairs would normally be required. (More precisely, repairs would be delayed.) In military operations, these receivers are said to fail at a rate of about one per week, although recent reports put the failure rate much lower. (It may sound like these receivers are unreliable, but since there are about 260 of them, a once-per-week failure rate would give a mean time to failure of about 5 years.) Some "failures" are in fact increases in the receiver noise level above military specifications, but which would cause no noticeable degradation of signals of interest to NOAA. The cost to repair them at an Air Force depot in Sacramento has averaged about $10K each, or $520K per year.
Under NOAA's task, some repairs would still be required, but the cost of doing so is expected to be somewhat less than under USAF operations. First, the contract with the Air Force depot will terminate with the implementation of warm storage, so Lockheed Martin would have to develop an equivalent test and repair capability, at a one-time cost of about $150K. With the closure of the depot, the receiver test equipment may be transferred to Bangor, reducing this cost somewhat. Second, receiver repair should be required less frequently because of less demanding NOAA applications. Third, when only minor receiver adjustments or tuning is required, Lockheed Martin staff could perform them at substantial saving over depot repair costs. The savings that would result from turning receiver repair over to Lockheed Martin could not be quantified, but discussions with the engineers indicated that a 50% reduction could be achieved. To be conservative, then, we assume that all necessary repairs could be performed by Lockheed Martin for $400K per annum.
2.7 Electric Power
It is perhaps surprising that the incremental electric power consumed by the transmitters is not a significant cost for NOAA operations. NOAA has tested the radar for ocean mapping using 10% of the maximum radiated power and found no noticeable degradation in data quality. On January 30, 1995, power consumption was measured at the transmit site during 10% power transmissions, and the rate of consumption was approximately 1300 kW. On the average, electricity costs about $0.10 per kWh at the transmitter, so the transmit facility could be run continuously at 10% power for 8769 hours (slightly more than a year) before reaching the $1.14M annual fixed minimum charged under the contract with the Central Maine Power Company. Therefore, NOAA's transmissions (even using all three radar segments) should not increase the transmit site electric bill.
Electricity for the ECRS Operations Center and Receive Site is provided by the Bangor Hydro-Electric Company at a rate of about $0.08 per kWh. Recent records indicate that the average consumption during normal USAF operations was about 530 kW at the Operations Center and about 357 kW at the Receiver Site, for a total annual cost of about $622K. The incremental power that would be consumed by NOAA operations is unknown but could be measured, once a warm-storage baseline is established. For reference, the warm-stored Receive Site at the West Coast Radar System uses only 44 kW. If only the equipment essential to NOAA operations were turned on, it should be possible to keep the electricity costs closer to those of warm storage than to those of Air Force operations. For example, at the Operations Center, NOAA would require electric power to only about 1/3 of the computers (whose heat is required to heat the building). Only two or three radar consoles are required. All of the flight-data support and backup systems can remain off. No additional administrative or support staff is required, so most of the lights can remain off. At the Receive Site, only the 260 receivers and their supporting systems will require power above the warm-storage level. Electronic systems generally have lower failure rates if they are left on all the time, than if they are turned on and off as needed. However, the final tradeoff between the electricity to keep them on all the time and the increased receiver mortality resulting from turning them on and off as needed is not known and would have to be learned by experience.
In view of these conservation measures, the meeting participants agreed that the average electric power required by NOAA's limited operations, over and above that used for warm storage, should be no more than 20% of that required by Air Force limited operations. This would amount to no more than 106 kW at the Operations Center and 84 kW at the Receive Site, for an annual cost of $124K per annum.
2.8 Security Issues
No additional security costs are expected for NOAA's operations, since a full-time security force (approximately 17 guards) is required for ECRS warm storage.
Virtually all of the radar specifications and capabilities that apply to NOAA's remote sensing requirements are unclassified. Security has been an issue because NOAA's data must be recorded in the classified radar environment. At present, any data recorded for NOAA must be cleared before release to NOAA. This clearance has become routine and now presents no problems for NOAA's data collection, even when near-real-time products are required.
Since the termination of counter-narcotics surveillance, secure communication lines to NORAD have been disconnected, eliminating the main barrier to direct unclassified connections to the radar data stream. It is likely that the security classification of the system will be downgraded as soon as certain classified hardware and software (not needed for NOAA's tasks) can be secured. This may permit NOAA to connect to the radar data stream and to perform certain reversible system modifications that would enhance the utility of the radar data to NOAA. At the least, there would be no further need to declassify radar data tapes before release to NOAA.
2.9 Maintenance Contracts
The only major subsystem requiring an additional maintenance contract under NOAA's limited operations is the VAX computers. The computers would be turned off in warm storage, whereas about 1/3 of them would be required for NOAA operations. NOAA would therefore have to negotiate a new service contract. As with the other radar subsystems NOAA would use, computer maintenance would be on a "fix what we break" basis. A 72-hour response time from a computer failure was tentatively suggested as adequate for NOAA's needs. An estimate by the present service contractor put the annual cost of such a contract at $144K, which would include a full-time computer maintenance person on-site. It seems likely, however, that a contract for a 72-hour response by an off-site contractor could be negotiated for much less. We estimate that a satisfactory contract could be negotiated by NOAA for $50K.
2.10 Contractual Issues
NOAA expressed the desire to set up the simplest possible arrangement with the Air Force that would permit "piggybacking" limited radar operations on the warm-storage contract between the Air Force and the contractor, with NOAA reimbursing the Air Force for the incremental expenses incurred. Until NOAA identifies sources of funds to pay these expenses, however, any such arrangements must be considered hypothetical and cannot be formally pursued. In the words of the ACC contracting representative, "Anything can be accomplished contractually, if the funds are available to make it happen." At this time, however, no provision for limited operations has yet been written into the ECRS warm-storage statement of work. Alternatively, NOAA could set up a support agreement with the Air Force and award a separate contract to operate the system as needed.
Since government and contractor personnel will continue in place to assure technical and security compliance during warm storage, NOAA should not incur any additional expenses for such administrative costs.
2.11 Uses of OTH-B for Other Research
It was agreed that OTH-B would be available for other research efforts on a non-interference, pay-as-you-go basis, provided the integrity of the radar system is not compromised. NOAA could coordinate scheduling and cost-sharing of other tests proposed by the scientific community and would request their approval from the OTH-B System Program Office.
2.12 Cost Summary
Here is a breakdown of the major costs to NOAA of operating the OTH-B East Coast Radar System for the nominal 10-hour-per-week mission described in Sec. 2.2:
NOAA Costs ($K per annum)
Extra crew 500
Equipment repair 400 + 150 first year
Electricity 124
Computer service 50
TOTAL 1,074 + 150 first year
The uncertainty of these estimates is large: approximately ±30%. This means that the cost of operating the OTH-B for the nominal NOAA mission described in Sec. 2.2 could be as much as $1.5M in the first year. On the other hand, if costs are carefully controlled, and if the radar crew can perform multiple tasks, first-year costs could be less than $1M. This estimate assumes that the full cost of warm-storing the facility will be borne by the Air Force. In keeping with the usual practice at national facilities, these costs cover radar operation only and do not include research and development costs associated with processing and interpreting the radar products or developing new applications. At present, these tasks are performed by NOAA/ETL staff. No permanent NOAA staff is required at the radar, however, since data collection can usually be remotely scheduled and processed.
3. BENEFITS
The benefits of operating an over-the-horizon radar for environmental research and services were considered more than 20 years ago (Barrick, 1973; Rhodes, 1975). Both authors envisioned relatively inexpensive ground-based HF pulse-Doppler radars that could illuminate millions of square kilometers of open ocean. These radars would map weather conditions at the ocean surface that would otherwise require an impossible number of widely dispersed in-situ instruments. The economic benefits they envisioned included warning oil tankers of high seas, thus saving fuel and averting disasters, supporting ship routing and tracking, aiding search and rescue missions, and providing input to global ocean and atmosphere numerical weather prediction systems. The dollar value of the benefits to the U.S. of operating OTH radars on its east and west coasts was estimated at that time to be more than $100M.
In February 1994, more than 60 scientists from academia, government, and the private sector formed an OTH-B Research Consortium to pool research ideas and prevent the loss to science of this national asset. The scientific community sees many potential environmental research and monitoring applications for over-the-horizon radar. Most of these applications address environmental hazards that threaten the populace with physical harm and economic loss.
Quantifying the benefits of these applications is as difficult as quantifying the value of the radar system for national defense. Where possible, we roughly estimate the economic impact of various environmental hazards that improved monitoring would address, but attempts to construct specific mitigation scenarios for any new technology tend to produce vague claims and fictitious valuations. No one claims that OTH radar would completely mitigate these hazards; only that it could help us better understand and monitor them--the first step toward forecasting and avoidance. Following is a summary of the research and service applications that have been contemplated. The Appendix contains more detailed discussions of specific scientific issues by specialists in each field.
3.1 Hurricane Forecasting
The NOAA National Hurricane Center has estimated that our Nation could save $90 million for every 300-nautical-mile improvement in hurricane landfall forecast accuracy. This figure is derived mainly from lost productivity and avoidable property damage, and does not include unavoidable destruction. As an example of an avoidable loss, the U.S. Navy spent $690K evacuating its ships from Norfolk prior to Hurricane Emily, which missed the port. Most hurricane research and modelling attempts to improve methods of forecasting landfalls and storm intensity, including high winds and storm surge. However, these models depend on weather observations over the notoriously data-sparse open ocean.
The present hurricane observing system consists mainly of geostationary-satellite imagery and hurricane-hunting aircraft that penetrate a storm and drop pressure and temperature sensors near its center. Other than ships of opportunity and the inhabitants of land that happens to lie in its path, there is no system for measuring weather conditions at the storm's periphery, for example the contour of gale-force winds, used by law-enforcement agencies to guide evacuation decisions. Satellite-based sensors that measure sea state cover the ocean in narrow strips a few hundred kilometers wide and so interrogate a given point on the sea only once every three days. Some, like the SSM/I microwave radiometer, cannot see through clouds and rain. Improving hurricane landfall and intensity forecasts will require an observation system that continuously covers large ocean areas and can move with the storm as it approaches the coastline of the United States.
OTH radar offers a unique capability for continuously mapping sea-surface winds and waves over very large fixed ocean areas. In particular, its ability to map sea-surface wind direction permits an accurate assessment of the location, shape, and growth of the tropical waves that intensify and often develop into tropical storms and hurricanes. It also maps synoptic and mesoscale meteorological features that determine whether tropical waves will grow or die. No present or planned observing system offers this capability. It fills a gap in the peripheral surface-wind information needed to predict the genesis, intensification, and landfall of hurricanes and tropical storms. For details of how experimental OTH-B surface wind data have already been used by the National Hurricane Center, see letters 3, 4, and 6 in the Appendix.
If OTH radar observations improve the mapping of peripheral winds in one Atlantic hurricane per year, to the extent of improving the landfall forecast accuracy by 30 nautical miles, then the savings to the public, by the above estimate, would be $9M. Additional savings would result from more efficient deployment of aircraft missions into developing tropical waves and storms.
3.2 Numerical Weather Prediction
Although NOAA claims that weather affects as much as one-sixth of our Gross Domestic Product, neither the magnitude of the effects nor the dollar value of accurate forecasts has been credibly quantified. Yet NOAA and other environmental agencies around the world spend millions of dollars each year improving the numerical weather prediction models upon which these forecasts are based. Although the models themselves continue to improve, the amount of weather data the models ingest is actually decreasing. Surface observations over land are decreasing as weather stations are closed. A weak link in the global weather observing system, upon which numerical weather prediction models depend, is the sparsity of weather reports over the three-quarters of the Earth occupied by the world's oceans. Both the number of weather reporting ships and the number of data buoys are decreasing. An example of a major surface analysis problem over the ocean is locating surface lows and fronts.
OTH radar provides high resolution surface wind data over the oceans, which can be used to validate and initialize these models. In particular, a streamline analysis of radar-derived surface wind directions gives a realistic snapshot of synoptic and mesoscale surface meteorology features that can be used to correct numerical weather prediction models. See letter 5 in the Appendix.
3.3 Space Weather Forecasts
A geomagnetic storm on 13 March 1989 knocked out Quebec's power grid for nine hours. Canadian power companies have since spent over $1 billion hardening their systems. Studies have estimated that such a storm could cause damage and losses to the U.S. electric power grid costing $6 billion. Indirect costs include the effects of blackouts on crime, disruption of critical services, etc. These storms also damage sensitive electronics in spacecraft costing hundreds of millions of dollars. They also pose radiation hazards to manned space missions, as well as to the passengers and crews of high-flying aircraft.
It is not widely known that VHF-radar echoes from the Sun were first observed in the 1960s. The Sun's radar cross section increases by orders of magnitude during events known as coronal mass ejections (CMEs). CMEs are known precursors of geomagnetic storms. An OTH radar experiment is underway in which the radar looks directly at the Sun shortly after sunrise. If the experiment is successful (as preliminary results suggest), the capability of measuring the intensity of CMEs 3 to 4 days before their high-energy particles reach the Earth would greatly extend the time for preventive action. Such "space weather forecasts" are part of the mission of the proposed National Space Weather Service. See letters 7 and 8 in the Appendix. Because OTH radars look only at low elevation angles, OTH-B is not seen as an operational solar radar, but instead as a unique research tool for probing the structure and dynamics of the solar corona and as a test bed for developing HF solar radar technology.
3.4 Marine Weather Forecasts
The world ocean freight bill is roughly $40 billion per year. Marine storms and high seas take their toll on shipping by increasing fuel costs, routing time, ship wear, and the danger of breakup and pollution. Fishing fleets are even more vulnerable to storms at sea. Data over the oceans are so sparse that conventional forecasts of storm strength and location are often inaccurate. OTH radar can track ocean storms in real time by mapping surface wind directions (streamlines). Storm precursors, such as tropical waves and polar lows, can be identified by OTH radar by their surface-streamline signatures. High-resolution surface streamline maps could be provided immediately to commercial ship-routing services. See letter 1 in the Appendix.
3.5 Pollution Control
More than $50 million is spent in the average year to clean up toxic spills at sea. To make better use of expensive cleanup resources, models that predict spill trajectories need better and up-to-date current measurements. OTH radar can map ocean surface currents with high resolution, through clouds and rain, over very large ocean areas. More information about surface currents is precisely what is needed to predict the trajectories of surface-borne pollutants. OTH radar can also monitor bursts of strong currents or sudden changes in current direction that can damage offshore oil platforms and cause spills. The main limitation is imposed by the particular ocean areas accessible to existing OTH radars (see Fig. 1). If 10% of the average cleanup cost were mitigated by OTH radar surface current observations only once in 5 years, the savings would pay for radar operations.
3.6 Search and Rescue
The U.S. Coast Guard spends approximately $15 million annually on ocean search-and-rescue missions. Of this, approximately $4 million is spent determining ocean surface currents, for the purpose of estimating the track of a target whose position was known at some time in the past. The Coast Guard routinely measures ocean currents by deploying aircraft that drop satellite-tracked drifter buoys ($170K per mission for an average of 24 missions). OTH radar could support search and rescue missions by mapping surface currents with high resolution over very large ocean areas. The main limitations are imposed by the particular ocean areas accessible to existing OTH radars (see Fig. 1), and by the fact the one radar measures only the radial component of surface current. However, vector current fields can often be inferred from one component, using knowledge of the surface temperature field, which is routinely available from satellite infrared imagery. If one-quarter of the current measurements now made by aircraft and drifter buoys were made instead by OTH radar, the savings would pay for radar operations.
3.7 Calibration and Validation of Satellite Ocean Sensors
Spaceborne ocean sensors, such as those planned for NASA's Mission to Planet Earth, must be empirically calibrated and validated by comparison with simultaneous in-situ sensors under a wide variety of sea conditions. No other ground truth so closely matches satellite sensors' coverage and spatial averaging as OTH radar. For example, OTH radar-derived surface currents can be used to validate and fill in dynamic height maps from TOPEX and ERS-1 altimetry, and OTH radar-derived surface winds can fill the gaps and remove ambiguities in surface winds measured by ERS scatterometers and SSM/I radiometers. The cost of deploying an equivalent number of sensors over large ocean areas could exceed the cost of a typical satellite, which is roughly $150 million. Operationally, the capabilities of OTH radar would complement those of oceanographic satellites, whose coverage is global but limited to infrequent space-time samples by their orbital dynamics. See letters 4 and 9 in the Appendix.
3.8 Air Traffic Control
Since OTH-B was designed to track aircraft, this capability already exists and could be adapted immediately to air-traffic control. Every day, dozens of commercial airliners, costing up to $155 million each, not including the human lives, cross the North Atlantic Ocean with no positive radar tracking. The East Coast OTH-B radar, with no modifications, could provide routine tracking and control of commercial air traffic in the crowded North Atlantic corridor. The capability for correlating flight plans with radar tracks is already in place at OTH-B. The warm-stored West Coast OTH-B, if operated continuously, could provide similar control over the northern Pacific, where the Korean Air Lines disaster could have been averted with appropriate radar tracking. Air-traffic control is the responsibility of the Federal Aviation Administration (FAA), not NOAA. Various FAA officials have expressed interest in OTH-B's capabilities, and a demonstration was given, but fiscal constraints have prevented further pursuit of this interest. Using the OTH-B for this purpose would require continuous operation at high power and would cost much more than NOAA's tasks. It is believed that the FAA desires instead to focus its limited resources on developing a GPS-based navigation and tracking system.
3.9 Fisheries Oceanography
The annual income to the world's fisheries is roughly $20 billion. Ocean-surface winds affect upwelling of nutrients, and warm and cold currents and eddies define the boundaries of fisheries and transport eggs and fry. Fisheries management involves, in part, understanding these variable (and largely unknown) oceanographic influences on fish populations. OTH radar can map both ocean surface winds and surface currents over large, fixed areas, through cloud cover and precipitation. The main limitation is in the particular ocean areas accessible to the existing radars.
3.10 Coastal Wave Forecasting
In the average year, approximately $1 billion in property losses are attributable to high surf and storm surges generated by marine storms. With better warnings, fixed property can be secured, transportable property moved, and false alarms avoided. Specific beneficiaries include coastal residents and industries (fisheries and offshore platform operators, for example) that must take protective measures. The role of OTH radar in improving observations of storms at sea is discussed in Sec. 3.1. With further development of the algorithms for analyzing the second-order sea-echo spectrum, OTH radar can also map parameters of the ocean wave directional spectrum in the open ocean, which could be used to initialize coastal wave models.
3.11 Ionospheric Research
Understanding the electrodynamics of the upper atmosphere, particularly in polar regions, has implications for electric power generation, mitigation of geomagnetic storm effects, and estimating satellite drag. It may someday be possible to tap some of the immense electric power flowing in ionospheric currents during geomagnetic storms. Ionospheric heating using high power (such as OTH) radar is a convenient way to study the upper atmosphere's response to known energy input. Researchers have also proposed studying auroral fine structure and the dynamics of artifical barium clouds with the OTH-B radar. See letters 10 and 11 in the Appendix.
3.12 Marine Weather Research
Accurate global and synoptic analysis requires more weather information over the oceans than ships of opportunity alone can provide. Because weather is mostly convective in the tropics and subtropics, mesoanalysis is important. For example, it is known that polar lows and other mesoscale features make important contributions to midlatitude and subpolar weather. Little is known about the mesoscale surface circulation in extratropical storms, for example in the North Atlantic. Under some poorly understood conditions, and without warning, such storms suddenly intensify into "bombs." In 1978, one such bomb caused $50K damage to the ocean liner, Queen Elizabeth II and the loss of an American trawler with its crew. Application of streamline analysis to high resolution OTH radar maps of surface winds far at sea can help meteorologists better understand synoptic and mesoscale circulation conditions that favor the intensification of extratropical cyclones. Radar-derived real-time surface wind data could also improve warnings of such events. For more details, see letter 1 in the Appendix.
3.13 Shortwave Radio Forecasting
A resurgence in military and commercial use of the crowded shortwave radio spectrum has prompted a renewed interest in HF channel characterization and better ionospheric models for HF radio frequency management. These "climatological" ionospheric models are derived from global soundings and adjusted for solar and geomagnetic influences. At least $5 million is spent annually collecting ionospheric diagnostic data globally and assimilating it into shortwave prediction models and warnings. These models lack the precision, however, to be useful for optimizing point-to-point radio communication.
Existing OTH radars can map ionospheric propagation conditions over large parts of the earth that are inaccessible to conventional ionosondes. They can also map the occurrence of transient phenomena in the ionosphere, such as polar and equatorial disturbances, as well as sporadic-E ionization. OTH radars also measure the spectral broadening of ionospherically propagated ground clutter as a function of range, azimuth, frequency, and time, which can be used as an index of the information capacity of an ionospheric path.
3.14 Climate Change
Changes in global and regional climate have inestimable potential socio-economic impact. NOAA has requested $89M for FY-96 research to understand climate and global change. Air-sea interaction is a critical unknown, both in terms of global observations, and for understanding how the climate machine works. OTH radar can monitor surface winds, waves, and currents over large, fixed ocean areas that influence the weather and climate of North America and Europe. It can characterize sea state, a critical unknown in assessing the effects of air-sea interaction on global climate change. Sea state affects ocean albedo and thus the absorption of solar radiation. Sea-surface roughness affects the uptake of greenhouse gases. Surface wind stress, determined by wind speed and surface roughness, affects ocean circulation and the global heat budget. The recently demonstrated capability of OTH radar for mapping surface currents could contribute directly to monitoring ocean circulation and heat fluxes. Long-term archiving of data on sea-surface processes over large ocean areas is essential to understanding their role in global environmental change. See letter 2 in the Appendix.
3.15 Iceberg Tracking
The U.S. Coast Guard, as part of the International Ice Patrol (IIP), is responsible for monitoring shipping lanes for dangerous icebergs. Although OTH radar cannot track individual icebergs (they are transparent to HF radar waves), it can track large, threatening icebergs on which transponders have been dropped from aircraft. The ECRS covers the part of the North Atlantic Ocean of interest to the IIP. A similar scheme using transmitters tracked by the ARGOS satellite system is already in use by the IIP, however.
3. CONCLUSIONS
An amendment to the NOAA Authorization Bill (H.R. 1815) for FY 96 requires NOAA to "expand its efforts to develop interagency agreements to further the use of defense-related technologies to support its oceanic missions." An opportunity to do so presents itself in the recently deactivated Air Force OTH-B air-defense radar system. In a series of tests over the last four years, the NOAA Environmental Technology Laboratory has demonstrated OTH-B's capabilities to map ocean surface winds, waves and surface currents over very large data-sparse ocean areas. For less than 1% of the $1.5 billion that taxpayers have already paid to deploy them, all the OTH-B radars could be redirected to ocean monitoring and other environmental research and services. The research capability and development expertise already exist within NOAA to exploit these capabilities. Complementary capabilities have been demonstrated for the Navy's ROTHR radar system.
We estimate that the cost to NOAA of operating the OTH-B East Coast Radar system for ten hours a week, in concert with Air Force warm-storage tasks, would be between $1M and $1.5M per annum. Benefits far in excess of this cost are realizable now. Oceanic products could be immediately provided to marine weather researchers, modelers and forecasters. Other products would require further development and investments by potential customers to reach their full potential.
Many of the remote-sensing capabilities of OTH-B are unique and cannot be duplicated by other sensors. Polar-orbiting satellites, for example, cover the earth in narrow strips and interrogate a given spot on the ocean once every two to three days. This is adequate for climate studies, but not for synoptic observations. Many satellite-based ocean sensors do not function through clouds and rain, whereas OTH radar can. Microwave scatterometers on satellites map ocean wave height and surface wind speeds reasonably well but wind directions are unreliable and not continuous enough to give a synoptic picture. OTH radar offers a complementary capability of measuring wind directions reliably and speeds poorly, and its coverage of large fixed areas permits construction of surface streamlines on a synoptic scale.
This report has examined only the operation of the OTH-B East Coast Radar System, or ECRS. A virtually identical (and virtually unused) three-radar system called the West Coast Radar System (WCRS) lies in warm storage near the California-Oregon border. Like the ECRS, the WCRS could be re-activated and used for environmental monitoring. The ocean area illuminated by the WCRS contains the weather systems that a few days later pass across the United States. For about the same cost as operating the ECRS, the WCRS could be operated remotely from the ECRS Operations Center in Bangor, Maine, using the existing WCRS satellite telemetry system.
REFERENCES
Barrick, D. E. (1973): The use of skywave radar for remote sensing of sea states. Mar. Technol. Soc. J., 7(1), 29-33.
Georges, T. M. (1993): Defense radars try environmental monitoring. The World & I vol. 8 no. 2, 220-227.
Georges, T. M. and J. A. Harlan (1994a): New horizons for over-the-horizon radar? IEEE Antennas and Propagation Magazine 36(4), 14-24.
Georges, T. M. and J. A. Harlan (1994b): Military over-the-horizon radars turn to ocean monitoring, Marine Technol. Soc. Jour. 27(4), 31-38.
Georges, T. M. and J. A. Harlan (1995): Ocean surface wind directions measured by the Air Force over-the-horizon radar during the 1994 hurricane season, NOAA Tech. Memo. ERL ETL-246, 93pp.
Georges, T. M., J. A. Harlan, L. R. Meyer, and R. G. Peer (1993): Tracking hurricane Claudette with the U. S. Air Force over-the-horizon radar. J. Atmos. Oceanic Technol. 10(4), 441-451.
Georges, T. M., J. A. Harlan, R. R. Leben, and R. A. Lematta (1995): Ocean surface currents mapped with an over-the-horizon radar, submitted to IEEE J. Geophys. Remote Sens.
Georges, T. M. and G. D. Thome (1990): An opportunity for long-distance oceanographic and meteorological monitoring using over-the-horizon defense radars. Bull AMS, vol. 71, 1739-1745.
Harlan, J. A. and T. M. Georges (1994): An empirical relation between ocean-surface wind direction and the Bragg line ratio of HF radar sea echo spectra, JGR Oceans 99(C4), 7971-7978.
Rhodes, R. R. (1975): A preview of benefits from skywave measurements of sea state. Mar. Technol. Soc. J., 9(2), 29-33.
APPENDIX
STATEMENTS FROM THE SCIENTIFIC AND USER COMMUNITIES REGARDING THE SCIENTIFIC AND PUBLIC BENEFITS OF OTH-B APPLICATIONS
1. Applications of OTH Radar to Marine Meteorology --
by Associate Professors George S. Young and Gregory S. Forbes,
Department of Meteorology, Pennsylvania State University
20 May 1995
As active researchers in marine meteorology, we feel that over-the-horizon (OTH) radars offer an outstanding opportunity for advancing the state of the science in the otherwise data-sparse regions of the Northern Hemisphere oceans. Questions as fundamental as the climatology of atmospheric response to ocean fronts will remain unanswered unless wind observations with mesoscale resolution are available for the open ocean. OTH radar provides just such data and at ranges far exceeding that of the Weather Service's coastal zone monitoring radars (WSR 88Ds). It is easy to list a variety of research and operational meteorology applications of the OTH radar. Just a few are mentioned here and described more fully below.
Research Applications:
1. Improved knowledge of the climatology of coastal front formation
2. Improved knowledge of oceanic front formation associated with the Gulf Stream
3. Improved knowledge of the role of the coastal front and its initial circulations on cyclogenesis
4. Improved knowledge of the mechanisms of formation of oceanic squall lines and Mesoscale Convective Systems
5. Improved knowledge of the conditions under which explosive Gulf Stream convection develops
Operational Applications:
1. Detection of the formation, location, and movement of coastal fronts
2. Detection of the formation and location of coastal cyclones and oceanic cyclones
3. Detection of coastal fronts near the polar ice sheet and the development of incipient polar lows along them
4. Detection of polar lows at sea
5. Prediction of Gulf Stream convective development
6. Detection and short-range forecasting of anomalous coastal winds
7. Improved oceanic surface analysis
8. Improved initial analysis of boundary-layer features in operational numerical models
9. Verification of the accuracy of numerical models, leading to improvements in their accuracy
Coastal fronts, their interaction with the Gulf Stream, and the severe weather they can trigger have been the focus of several major field experiments including GALE (Genesis of Atlantic Lows Experiment) and ERICA (Experiment on Rapidly Intensifying Cyclones over the Atlantic). Results from these experiments demonstrated the importance of coastal fronts in determining the location of both rapid cyclogenesis and severe precipitation events. Once the coastal front is well developed, temperatures across the front can vary by 10C or more, and winds can shift by 90 degrees or more. Two of the key forecasting questions pertaining to forecasting for coastal sites in the winter, therefore, are whether a coastal front will form and march inland, and when. Part of the information needed to make accurate predictions, then, is information on whether or not a coastal front has formed offshore, where, and its movement.
While WSR-88D radars will typically give adequate coverage of the coastal front east of the United States, dramatic coastal fronts can form offshore in other Northern Hemisphere locations, notably in the vicinity of the ice cap. There, small but intense polar lows can develop on the coastal front and then drift with the wind farther out to sea. Accompanying wind circulations can impact transportation and offshore drilling operations, and perhaps even influence the drift of icebergs. While most polar lows are accompanied by satellite-detectable cloud clusters in their developmental stages, there are many more such cloud clusters than polar lows. It is not always possible with satellites to determine whether or not the circulation is reaching the ground, and conventional surface observations can be very sparse relative to the small scale of the polar low. Thus, the OTH radar could prove valuable in the early detection of polar lows.
Unfortunately, the Gulf Stream plays such a role in the dynamics of the coastal front, coastal cyclogenesis and oceanic cyclogenesis that many of the critical events occur far enough offshore that their boundary-layer winds are below the horizon of the coastal WSR 88D Doppler radars. The location of most rapid deepening of the strongest oceanic cyclones typically falls into this category. Moreover, ship of opportunity and moored buoy data are generally too sparsely distributed to support the required mesoscale analyses. High clouds often obscure the low-level cloud street patterns that can reveal the coastal front and other mesoscale circulations. OTH radar overcomes these problems with its ability to resolve the mesoscale field of surface flow in offshore regions. The most dangerous of the oceanic storms tend to intensify very rapidly shortly after their inception, often reaching nearly full strength in six hours. Numerical models systematically predict these developments too slowly, and may not exactly predict the location of formation. Moreover, the ERICA field program revealed that in the initial stage of some cyclogeneses there appear are multiple circulation centers. How important are these in general? By monitoring the development of the mesoscale surface flow with OTH radar, it will be possible to determine the precursors of the coastal front and incipient oceanic cyclones and ascertain their association with the Gulf Stream, and to detect and monitor the coastal front and polar lows in more northern latitudes. During storm development, the transition from frontogenesis to cyclogenesis can be related to larger scale phenomena with much less ambiguity when OTH radar is used to locate the key events accurately enough in time and space.
The offshore environment is also extremely active on the convective end of the meteorological spectrum. For both the tropics and the Gulf Stream region, a hotly debated question is the origin and triggering of mesoscale convective systems. Lack of horizontal resolution in the current surface observing systems has hindered study of the links between convective initiation and preexisting mesoscale flow features. Thus, such fundamental questions as whether most tropical maritime convective systems arise from the aggregation of random convective elements or are triggered by a preexisting mesoscale convergence line remain open. Careful comparison of OTH radar observations with GOES imagery will provide the combination of mesoscale flow analysis and convective verification needed to address this question.
A similar approach using OTH radar surface streamline analyses with WSR 88D mid-level reflectivity analyses can be applied to determining the origins of the strong convective systems observed over the Gulf Stream. There are many instances in which convection near the Gulf Stream has developed to intensities rivalling that of Midwestern storms, and with equal rapidity. Yet, unlike their Midwestern counterparts, in Gulf Stream explosive convective developments there is often little information to reveal the cause or timing of the storms. A working hypothesis is that a sudden shift of boundary-layer wind causes unstable tropical air that had been stagnant over or east of the Gulf Stream to head shoreward, both bringing it under cooler air aloft and lifting it over the cooler air at the surface near shore.
OTH radar provides the only comprehensive means of verifying the surface layer performance of large scale numerical weather prediction forecasts. The operational NWP model forecasts are currently verified against analyses which are dominated by first guess fields derived from the model's own forecasts. This incestuous relationship between forecasts and verifying analyses has severely limited our ability to constructively critique the NWP models' performance over the open ocean. The advantages of OTH radar data for model verification will continue to increase as the horizontal resolution of the operational models improves to match that of the OTH radar. The case for using OTH radar data for this task is all the more compelling because these scales are much too fine to be resolved by the current ship of opportunity observations. OTH radar can also provide an invaluable source of data for preparation of the operational surface analyses for maritime regions. Despite the availability of satellite imagery, the analyzed location of major weather systems such as cyclone centers and fronts has often been called into question. This problem stems both from the extremely low resolution of the available ship of opportunity data and from the tendency of upper level clouds to occur in conjunction with surface weather systems thereby obscuring them in satellite images. OTH radar data goes a long way towards solving this problem of system placement by providing exact locations of the confluence zones characteristic of surface fronts and the cyclonic circulation centers.
Diagnosis of cyclogenesis, not only for coastal lows and polar lows discussed above, but also for tropical storms and all types of oceanic cyclones, is another thorny operational problem for which OTH radar data would provide significant help. Again, the presence of high clouds generally obscures the satellite signatures of low-level cyclogenesis in both the tropics and mid latitudes. In both regimes, the transition from a relatively benign wave phenomenon to rapid and intense cyclogenesis has proved all-too-easy to miss. In contrast, streamline analyses prepared from OTH radar are not affected by satellite's limitations and have the horizontal resolution required for diagnosing the transition from an open circulation easterly wave to a closed circulation tropical depression. Similarly, researchers in the ERICA experiment noted that satellite images frequently gave little indication of exactly where surface cyclogenesis was focused in cases of rapid and intense cyclogenesis. OTH radar has the horizontal resolution needed to sort out the confusing multiple centers and keep track of which ones are growing to dominate the system.
In short, OTH radar represents a unique resource both for advancing our scientific understanding of meso/synoptic scale marine meteorological phenomena and for improving our ability to forecast such phenomena in an operational setting. Saving and utilizing this irreplaceable resource will be well worth the cost.
2. Letter from Commerce Secretary Ron Brown
to Defense Secretary Les Aspin, September 16, 1993:
It is my understanding that the Air Force is considering the termination of the Over-the-Horizon Backscatter (OTH-B) radar system which supports the Nation's air defense system. Although still in a research stage, these radars have also demonstrated promise for monitoring wind and sea conditions over very large ocean areas.
At present, the lack of such weather observations over the world's oceans is a weak link in the National Oceanic and Atmospheric Administration's (NOAA) ability to make more accurate weather forecasts and warnings, and to understand global climate change. Continued operation of these radars with civilian access to the environmental data streams will fill the gap of ocean surface information that is not currently available by any other means. The enclosed article [Georges, 1993] describes some of the tests recently performed with these radars and the potential they hold for environmental monitoring.
I am told that the incremental cost of operating these radars in a dual-use mode appears to be low. I also understand that the Air Force Rome Laboratory, which has cooperated with NOAA in its experiments, has the technical expertise to manage OTH-B as a national facility. Given the vast potential benefits to the Nation that these radars have to offer, I urge that we jointly convene a task force to study the range of options for the ultimate disposition of these radars. NOAA, along with other interested Federal agencies, would work with the Office of Management and Budget and Congress to develop an equitable approach to pay for the operation of these radars, if the task force determines that the benefits justify the costs.
I think the country has the potential to reap some real "peace dividends" through the productive dual-use of these radars. I look forward to working with you on this issue.
3. Memorandum to Dr. Kathryn D. Sullivan, NOAA Chief Scientist,
from Dr. Robert C. Sheets, Director, NOAA National Hurricane Center and
Dr. Peter G. Black, NOAA Hurricane Research Division
October 3, 1994
Through the efforts of Tom Georges and Jack Harlan of the ERL Environmental Technology Laboratory, Over-the-Horizon (OTH) radar wind direction maps were made available once per day to the National Hurricane Center (NHC), AOML-HRD, and others in real time via Mosaic and the World Wide Web for the period from August 19 to October 1. Through the efforts of Paul Chang of the NESDIS Oceanic Sciences Branch, the OTH data were merged with both the European Space Agency ERS-1 satellite scatterometer real-time surface wind vector data and the Air Force/Navy DMSP satellite microwave radiometer (SSM/I) surface wind speed data. These merged data sets were also made available in real time via Mosaic and the World Wide Web. This proved useful for operations and research work in two important ways:
The OTH data, when available, provided an accurate location of the trough axis for tropical waves that traversed the Atlantic and Caribbean during the hurricane season. NHC carefully monitors these features for possible associated adverse weather and signs of tropical cyclone development.
The OTH data provided a spatial continuity to help connect the many separate ERS-1 and SSM/I satellite passes over the tropics. When satellite wind speed estimates were merged with OTH wind direction estimates, a more coherent image of the wind field within the tropical waves, and other tropical weather features, was made possible.
The NHC prepares twice-daily analyses of surface wind and pressure for the tropics, which are transmitted via fax to other users worldwide. A key element of this analysis is the location of the axis of tropical waves, which generally move in a steady progression from the African west coast across the tropical Atlantic, the Caribbean Sea and into the eastern Pacific. It was found that the surface wind directions provided by OTH clearly marked the axis of the tropical waves. These wave-axis locations agreed quite well with approximations determined from GOES and METEOSAT cloud imagery, scattered ship reports and island upper-air soundings. Spot comparisons of OTH directions agreed well with those from NOAA Data Buoy Center (NDBC) platforms in the Bahamas and Caribbean, scattered ships and an automatic weather station on the eastern tip of Barbados, operated by the University of Miami. The OTH and merged OTH-ERS1-SSMI products were made useful to operational forecasters by their availability on Mosaic, on an NHC computer workstation.
The incorporation of ERS-1 and SSM/I microwave data provided quantitative speeds along their respective satellite swaths. The OTH direction data provided much needed spatial continuity of features in the normally data void regions between satellite passes, which, in the tropics, can be quite large. The merged data thus provided much greater utility that either data source used separately. If this program continues, we suggest routine imaging of these data sets. Perhaps these merged OTH-satellite data sets hold clues to distinguishing waves that develop from those that do not. Waves with sharply peaked crests, i.e., greater curvature in the wind field, may have potential for development, when located in regions of favorable upper atmosphere conditions, than those with flatter, broadly curved surface wind fields. The latter may be indicative of a stronger-than-usual African easterly jet following the waves, which is sometimes associated with African dust outbreaks seen on GOES and METEOSAT visible images. Availability of these data sets on a routine basis would permit investigation of these features.
The merged OTH-satellite data were also useful for planning HRD research aircraft missions into prospective tropical cyclones. These data were evaluated along with other data sources during mission planning for deployments to tropical storms Debby and Chris. The data can also serve as useful comparisons with airborne scatterometer and stepped-frequency microwave radiometer surface wind estimates in a research mode.
Initial impressions, gleaned from the merged OTH/satellite data products, suggest potential for obtaining quantitative information in data-void areas. This season, the data were generally used as confirmatory information. If these data sets are available next year, we would suggest that the data be made available in concert with operational deadlines for twice-per-day analyses and entered directly into the data stream that gets plotted and analyzed together with conventional data sources. This proposed test may also yield new insights into the process by which the surface wind field evolves as tropical waves develop into tropical storms, especially in regions of the Atlantic beyond the range of reconnaissance aircraft.
4. Letter to T. M. Georges from Dr. Ed Rappaport, Hurricane Specialist
National Hurricane Center
2 June 1995
As a follow-up to the 10/94 memo from Bob Sheets and Peter Black to Kathryn Sullivan, I have prepared a few notes summarizing National Hurricane Center (NHC) specialists' comments on OTH radars:
Given the paucity of surface wind observations over the Atlantic, OTH radar appears to provide a unique and potentially important data base for the analysis of tropical and subtropical weather systems in the Atlantic. This conclusion was reached based on our qualitative and subjective assessments of OTH radar wind direction/streamline information made available to the NHC during the past few hurricane seasons. Specifically, those data helped us to identify with confidence some otherwise undetectable characteristics of tropical waves, as well as other features of the surface wind field.
So far, we have recognized two limitations that, if overcome or ameliorated, might make the OTH data indispensable: (1) temporal resolution -- data have been available only once per day, and (2) absence of wind speed data.
Ultimately, the OTH data would be most useful to the NHC (hurricane forecasters and TAFB unit meteorologists) if it (and other data bases such as SSM/I and ERS) were routinely incorporated into the operational surface analyses used by the NHC, and if the data were found to significantly improve the NMC model initializations/predictions upon which the NHC is so dependent.
Regarding wind speed information, you indicated that one or more of the existing radar systems could be modified (at some expense) to provide quantitative information about surface wind speed. Such information could be invaluable to us in our analyses of the intensity, size, and potential impact of tropical cyclones. It could also be used by us to better advise emergency managers and decision makers charged with issuing public evacuation orders (e.g., the radii of 34 kt winds are important parameters that are not now well observed). Hence, I would again like to propose that a "proof-of-concept" field program to evaluate the utility of quantitative OTH radar surface wind speed and direction data be designed and executed. In particular, you might consider a two-to-four-week program in which OTH surface wind data would be collected at either 6 or 12 hour intervals (discuss with NMC). The program should probably be run in September or October, when there is likely to be an opportunity to observe both significant tropical and extratropical cyclones. The impact of the data on the NMC models could then be assessed in a study evaluating the accuracy of the models, with and without that data.
On behalf of the hurricane specialists, I thank you for your efforts in making the OTH data available to us. The ability of the OTH radars to observe such a relatively large surface wind field is likely to remain unsurpassed for some time, and it is easy to foresee potentially important applications of this data at the NHC.
5. Letter to T. M. Georges from Dr. Eugenia Kalnay, Chief, Development Division,
NOAA National Meteorological Center, April 14, 1994
Thank you very much for your and Jack Harlan's stimulating seminar on the possible use of Over-the-Horizon Radar to determine surface winds.
We were impressed by the ability of the system to provide high resolution surface wind direction information, in large oceanic regions, even in the presence of clouds. The possibility of developing methods to infer wind speed is very interesting. I appreciate that you sent us also comparisons between the inferred OTH surface winds and the estimated wind directions from the NMC global analyses. The fact that they differ substantially from each other can be considered both good news (if the OTH directions are correct there is great potential for impacts on the NMC weather analyses and forecasts) or bad news (the OTH wind directions may be wrong in a significant number of areas), and further research and comparisons with other data is required to clarify which of these two scenarios is dominant.
We are now in a position to assimilate information such as wind direction, or directly measured reflectivities, rather than retrieved physical parameters, such as winds, because of our use of a variational data assimilation system (Spectral Statistical Interpolation, or SSI). This was impossible a few years ago. So this would support our interest in further studying the possible use of your data. On the other hand, past experience suggests that surface data alone does not tend to contribute too much to the global analysis. We are somewhat concerned with the assumption that the wind and wave directions for 10-m waves line up, which may not be a reliable assumption for wind speeds less than 8 m/sec (Tolman, pers. comm.). Perhaps you could plot the same comparisons with the NMC analysis deleting those winds in areas where the NMC analysis estimates the wind speeds to be less than 8 m/sec. It seemed that there are several areas of disagreement between the two wind direction estimates that were small scale, and which could therefore be real features missed by the global analysis, but on the other hand, some of these small features are anticyclonic, which is somewhat surprising.
We hope you will keep us abreast of your progress and look forward to further interaction with you and your group.
6. Letter to T. M. Georges from Julian M. Wright, Federal Coordinator,
Office of the Federal Coordinator for Meteorological Services and Supporting Research
May 18, 1994
We were pleased to host your presentation on the OTH-B Radar to the Interdepartmental Committee for Meteorological Services and Supporting Research (ICMSSR), which includes membership from 14 Federal agencies with a vested interest in improving the quality of weather services to our Nation. On behalf of ICMSSR, for which I serve as chairman, we are convinced that the OTH-B radar is truly a unique scientific tool that has tremendous potential for the meteorological community.
Specifically, the modeling of hurricane tracking, intensification, and genesis are all in need of improvement. A particularly difficult aspect of hurricane prediction is the paucity of observations over the open oceans. Accurate prediction of hurricane motion and intensity changes will require observations of the large scale flow in the storm environment as well as detailed observations near the storm center. Your ability, through the use of the OTH-B radar, to provide near-real-time information about ocean surface wind and wave fields, on demand, through cloud cover, and over very large ocean areas could significantly impact our ability to improve the accuracy of coastal wind and wave forecasts and severe coastal event warnings for landfalling hurricanes and tropical storms. It has been estimated that our Nation could save $300,000 for every mile improvement in the forecast warning accuracy of a landfalling hurricane. As a result, it is not difficult to conclude that an investment in the OTH-B radar for ocean environmental monitoring has the potential to pay for itself.
Surely the formation of the OTH-B radar research consortium and your efforts to realize a significant "peace dividend" from this billion-dollar defense program are worthy of consideration by all members of ICMSSR. Please feel free to contact me if the OFCM can be of further assistance.
7. Letter to T. M. Georges from Antony C. Fraser-Smith, Professor of Electrical Engineering and of Geophysics and Von Eshleman, Professor of Electrical Engineering (Emeritus),
Stanford University
February 28, 1994
Following up our earlier conversations, I want to point out the interest of Professor Von Eshleman and myself, at Stanford University, and Drs. James R. Barnum and Robert L. Showen at SRI International, in conducting solar radar experiments using the Air Force's Over-the-Horizon Backscatter (OTH-B) radar facility.
There is no need to point out to you the importance of the Sun to our life and activities on Earth. Strangely, however, considering the importance of the Sun, there have been few attempts to conduct active experiments to investigate its outer atmosphere, even though this outer atmosphere (i.e., the corona) is not well understood, and it is known that active experiments are feasible. In fact, scientists at our Laboratory were the first to obtain radar echoes back from the Sun [Eshleman, 1960], and later experiments by scientists at MIT's Lincoln Laboratory using an array in Texas clearly demonstrated both the feasibility of routine soundings of the Sun's corona and the variability of this outer region of the Sun's atmosphere.
Preliminary calculations we have carried out at Stanford, as well as calculations done by Dr. Francis J. Kelly of the E. O . Hulbert Center for Space Research at the Naval Research Laboratory, show that the OTH-B radar can probably detect changes in the coronal plasma some of the time -- primarily when there are coronal mass ejections (CMEs). It is primarily these CMEs that produce magnetic storms several days later when enhanced solar wind plasma reaches the earth. Further preliminary calculations we have carried out indicate that even greater sensitivity and something approaching routine soundings of the Sun's corona would be possible by combining transmissions from the OTH-B radar with reception by SRI International's Wide Aperture Research Facility (WARF) located in central California.
The WARF facility was originally constructed as part of a pioneering over-the-horizon radar (OTHR) project here in our laboratory, and it has now been greatly improved and expanded following the move of the project to SRI International in the early 1970s. The WARF OTHR utilizes a 2.3-km-long receiving antenna array, which greatly increases the sensitivity of the radar, reduces clutter, and improves the accuracy with which echoes can be located. Because of its pioneering role in the development of OTHR, the WARF facility uses essentially the same instrumentation as the Air Force's OTH-B radars. This latter point is important because there have been remarkable developments in the technology used to send and receive OTHR signals, which could greatly improve our capability of obtaining echoes from the Sun, as compared with the situation in the 1960s, when echoes were first obtained. We have discussed use of this combination of an OTH-B radar transmitter with the WARF receiver with Dr. James Barnum of SRI International, and he has indicated his great interest, and that of a colleague, Dr. Robert Showen, in a project of that nature.
We cannot help pointing out the great advances in knowledge about the Earth's ionosphere that resulted from development of sounding techniques. Even today, there are many ionosondes in operation all over the world, measuring ionospheric parameters that are of crucial importance to many different activities, including worldwide short-wave radio communications, in particular. We believe routine "soundings" of the Sun could similarly greatly increase our knowledge of the properties of its outer atmosphere, as well as providing information about changes that could impact upon our life on Earth (the CMEs, for example).
... With its past experience obtaining radar echoes from the Sun, in the development of over-the-horizon radar, and close relations with colleagues at SRI International (and with Dr. Kelly, if he is available), the STAR Laboratory is well qualified to help organize a study of the possibility of "sounding" the Sun by using a combination of the OTH-B and WARF radars. Both Professor Eshleman and I are enthusiastic about the possibility of such a study, and we believe it would provide a particularly appropriate application of the U.S.'s OTH-B radar technology to dual use.
8. Portion of Email To Dr. Kathryn D. Sullivan from Dr. Ernie Hildner, Director,
NOAA Space Environment Laboratory
April 14, 1995
I thought that perhaps you might not be aware that one of the experiments caught short by the OTH-B radar's abrupt closure is an exciting solar radar experiment. This is a speculative experiment, but potentially valuable. It might show us the way to measure the properties of the solar wind leaving the Sun, including the occasional coronal mass ejections (CMEs) that cause geomagnetic storms.
Physics Background -- In principle, a radio wave directed at the sun will be reflected from where the coronal plasma frequency equals the radio frequency of the radio wave; the higher the frequency, the deeper the wave will penetrate before reflection. Thus, if we vary the transmitted frequency, the times of flight (ranges) of the returned signals tell us the profile of density in the corona as a function of height. Similarly, the Doppler shifts for various frequencies allow us to determine the outflow velocity profile as a function of height. More subtle, but still possible in principle, would be to infer the inhomogeneity of the corona at each particular plasma density (height) from the scatter in the return signal at the appropriate frequency and (wild speculation here!) to infer the temperature or microturbulence in the outflowing plasma from the spectral broadening.
Radar Background -- There is no other facility in the world which has the transmitting power and receiving sensitivity to perform this experiment at the frequencies which would probe the region of acceleration of the solar wind and would see coronal mass ejections headed Earthward to make geomagnetic storms. We intended the experiment -- receiver tests and software development have already been carried out -- to be a proof of concept, carried out daily at sunrise for half a solar rotation or so to sample a variety of solar structures at the sub-Earth point. A major unknown is the degree to which the transmitted and returning signals can penetrate the ionosphere; at the slant angles and frequencies incorporated in the radar's design, only some fraction of the power will pass through.
Future Hopes -- If a return signal was received, and if the signal/noise was sufficient to analyze to form the height profiles of the parameters, we planned to start a campaign to get a (non-NOAA) research radar funded, using the performance data gathered with the OTH-B radar for the proposal.
9. Letter to T. M. Georges from Gary S. Brown, Professor and Director,
Electromagnetics Interaction Laboratory,
Virginia Polytechnic Institute
April 8, 1994
The purpose of this letter is to provide you with some background information on what I feel to be an important potential use of the OTH-B radar. As you are no doubt aware, the OTH-B system is capable of providing tracking information on hard targets and basic data on the state of the sea where its antenna footprint intersects the ocean. Information such as current speed and direction, waveheight directional power spectral density, and surface wind speed can be extracted from the basic radar data. These data can be provided over a relatively large area with antenna beam scanning.
The OTH-B radar as a remote sensor of oceanic characteristics is more mature than spaceborne remote sensors simply because it has been operated longer than any recent, active, spaceborne sensor. The confidence we have in its data, along with its relatively large coverage area make it an ideal device to provide oceanic data for "truthing" data from spaceborne sensors. That is, what it measures and the area over which it measures are significantly more relevant to checking the data from spaceborne sensors than any techniques we currently use. For example, we presently check spaceborne sensors by either carrying out extensive and very costly underflights by aircraft or by waiting for the satellite to complete enough overflights of buoys to build up a statistical data base for comparison. Neither of these "surface truthing" techniques is really adequate, and they give us only a relatively crude check on the spaceborne sensors and algorithms.
With the OTH-B radar, we could illuminate a large area on the ocean surface and compare the satellite sensor's data (when we have an overflight of the area) with that of the OTH-B radar. The two data sets would be very comparable in terms of both spatial and temporal averaging; this is something that we have never achieved in the past, as buoys provide a point measurement. In fact, the OTH-B radar is the only way I see of ever obtaining data that is compatible with spaceborne sensors. Thus, the OTH-B radar provides an invaluable and in fact, singular opportunity to really do surface truthing the way it should be done. In short, there is no other way to check spaceborne oceanographic remote sensors in a true "apples and apples" comparison.
If you have any questions on what I have proposed herein, please do not hesitate to contact me. I honestly feel that the OTH-B radar system represents an extremely valuable asset for the remote sensing community, and I would hate to see it not used in this application.
10. Letter to T. M. Georges from Professor Syun-Ichi Akasofu, Director,
Geophysical Institute, University of Alaska
March 2, 1994
We would like to express our interest in promoting the Over-the-Horizon Radar Research Consortium as a vehicle for facilitating use of the USAF OTH-B East and West backscatter sounder radars. We would be particularly interested in using Beam 1 of the OTH-B East and Beam 3 of the OTH-B West radars to investigate "weak-scatter backscatter" from the auroral ionosphere during moderately disturbed and magnetic storm periods. The gain and resolution of the OTH-B radars should give us some information on the fine structure of the auroral F region heretofore not obtainable. Dr. Robert Hunsucker would probably be the most involved in this effort.
As you have already demonstrated, the OTH-B West radar beam 3 should also give excellent surface wind direction in the Gulf of Alaska. Of course, these efforts are contingent upon adequate funding to carry out the experiments.
11. Letter to T. M. Georges from Dr. Min-Chang Lee, Head
Ionospheric Plasma Research Group, Plasma Fusion Center, Massachusetts Institute of Technology
1 April 1994
I have prepared the following brief statement to support the continued operation of the OTH-B for ionospheric research:
Wave injection experiments using the ground-based VLF and HF transmitters have been actively conducted in the radio science and space plasma research communities for diagnosing the ionosphere and magnetosphere, and for investigating artificial communication paths to extend the capability of radio communications. There are two HF heating facilities in the United States, one in Puerto Rico and the other one in Alaska, dedicated to scientific research. High power HF waves are primarily transmitted vertically into the upper atmosphere from these two facilities. However, it is expected that the OTH radars, transmitting HF waves obliquely will illuminate much larger ionospheric regions and complement the aforementioned radio and space plasma research. Unfortunately, there are no OTH radars dedicated to scientific research in the United States. It is therefore desirable to propose that the OTH-B in Maine be operated for the scientific community to promote atmospheric research and radio science.