National Academies Press: OpenBook
« Previous: 3 Physical Mechanisms for RF Effects on Biological Systems
Suggested Citation:"4 Exposure Levels." National Research Council. 2005. An Assessment of Potential Health Effects from Exposure to PAVE PAWS Low-Level Phased-Array Radiofrequency Energy. Washington, DC: The National Academies Press. doi: 10.17226/11205.
×

4
Exposure Levels

OVERVIEW

This chapter summarizes the current data on exposure levels produced by PAVE PAWS. It begins with an overview of the operating characteristics of the radar and then discusses the existing exposure data measured in terms of peak and average power density (in µW/cm2). Recent measurements of PAVE PAWS power density recorded by census tract obtained as part of the current initiative to re-examine the potential health effects of exposure to low-level phased-array RF energy are included. Discussion also considers exposure levels of the general population to other broadcast radiation sources of comparable spectral content.

PAVE PAWS OPERATION

The environmental impact and safety questions related to human health for radar installations (and other broadcast sources) have traditionally been characterized in terms of the power density produced as a function of time and location within the surrounding population (often neglecting details of the local terrain). Power density remains an important metric for safe operation of a radar in a populated area. As such, the primary physical characteristics of PAVE PAWS that determine its power-density distribution (e.g., its beam attributes) are briefly reviewed (detailed descriptions of the full operational behavior of PAVE PAWS have appeared in prior reports [NRC 1979a, b]). Historically, power levels below the limits known to cause thermal effects in tissue have been deemed acceptable. Current safety standards, which have been in place in this country for a number of years, continue to evolve and are periodically revised, but are essentially based

Suggested Citation:"4 Exposure Levels." National Research Council. 2005. An Assessment of Potential Health Effects from Exposure to PAVE PAWS Low-Level Phased-Array Radiofrequency Energy. Washington, DC: The National Academies Press. doi: 10.17226/11205.
×

on this threshold (e.g., IEEE Standard C95.1-1999). However, concerns have recently been raised regarding the safety of exposure to phased-array radiofrequency energy. Specifically, two questions have been articulated. The first is whether or not the human body responds differently (and detrimentally) to a radar beam formed from a large number of individual antenna elements radiating waveforms that are slightly shifted in time (to form a beam whose direction can be electronically scanned by controlling the phasing between elements) relative to a single source generating an equivalent beam (in terms of radar function) that is mechanically redirected. The second is whether or not the transient characteristics of the waveforms generated by PAVE PAWS are sufficient to produce so-called precursors—spectral components of the composite signal created through interactions with the human body that have been hypothesized to propagate to distances into tissue extending beyond that expected from the primary (signal) frequency comprising the PAVE PAWS emission. As a result, the summary of PAVE PAWS operation presented here will focus on its beam-forming capabilities. The characteristics of the time-domain signal measured from PAVE PAWS and the conceptual and experimental basis for the existence/nonexistence of precursors generated in tissue by PAVE PAWS exposure will be developed more fully in Chapter 5.

THE BEAM CHARACTERISTICS

PAVE PAWS consists of two planar arrays of active antenna elements (the centers of each array face are pointed 20° above horizontal and 120° apart in azimuth) that are phased to radiate a directed narrow (main) beam (2° in width, broadside) in short, high-energy bursts (pulses) containing approximately 90% of the total transmitted power at the intended RF frequencies (any one of 24 discrete frequencies between 420 to 450 MHz). The remaining energy is distributed in sidelobes to the main beam which form clusters of surrounding auxiliary beams that transmit significantly reduced intensities pointed in directions fanning out from the orientation of the main beam with respect to the horizontal. For example, the first and second sidelobes radiate powers that are no more than 1% and 0.1% of the main beam, respectively are directed approximately 4° (1st sidelobe) and 6° (2nd sidelobe) off the central axis of the main beam (horizontally). Given the local topography of the Cape Cod PAVE PAWS installation (which slopes away from the radar at 1° or more below horizontal), it is primarily the secondary sidelobes that intersect the ground when the main beam is pointed at its lowest (3° above horizontal) elevation.

Transmitted Power

Each face of the PAVE PAWS radar consists of a regular grid of 2677 antennas of which 885 are inactive, leaving 1792 active that are directly connected to

Suggested Citation:"4 Exposure Levels." National Research Council. 2005. An Assessment of Potential Health Effects from Exposure to PAVE PAWS Low-Level Phased-Array Radiofrequency Energy. Washington, DC: The National Academies Press. doi: 10.17226/11205.
×

individual solid-state transmitters radiating about 320 watts at peak power. The peak transmitted power of PAVE PAWS is rated at 580 kW (320 W × 1792), since typically only one of the two antenna faces radiates at a time. As described in more detail below, the sequence of pulses radiated depends on the function that the radar is performing at the time. Controls built into the system limit the rate at which pulses are transmitted so that the fraction of total time during which power is radiated does not exceed 25%, making the average transmitted power less than 145 kW (580 kW × 0.25). Correspondingly, the average radiated power in the first and second sidelobes is approximately 1.45 kW and 145 W, respectively.

Waveform Generation

PAVE PAWS operates by transmitting pulses of radiated power interspersed with periods of time dedicated to signal reception, the details of which depend on the radar’s immediate functional task (e.g., searching, tracking), making the type of pulse or pulse burst that is emitted, and the timing between bursts, complex and dynamic. Further, the variability in the pulse pattern is coupled to directional changes and small pulse-to-pulse frequency shifts in the main-beam transmissions that create sidelobe radiation patterns that are constantly changing at any given location on the ground. It is, therefore, difficult to quantify precisely exposure conditions over time because the patterns of pulses emitted by PAVE PAWS depend on the scanning directions being searched by the main beam along with the number and location of targets being tracked at a particular instant in time.

The operation of PAVE PAWS is synchronized to a 54 msec cycle time in which one or more pulses may be transmitted in each of several directions at slightly different frequencies. Seventeen (17) consecutive 54 msec cycles are devoted to radar function with an 18th used for self-testing. Within a cycle, pulse widths ranging from 0.25 to 16 msec may be generated. Within some pulses, the main transmission frequency is varied (modulated or “chirped”) by 2 MHz or less depending on radar function. Hence, the maximum possible duty cycle (fraction of time that power is transmitted) during a single pulse repetition interval is 30% (16/54 msec) and 28% over 18 consecutive cycles (17/18 × 30%), although on average the maximum duty cycle does not exceed 25%.

Although the waveform modulation scheme is complicated by the multi-task operation of PAVE PAWS, the modulation of transmitted power experienced at a fixed location on the ground has been estimated based on the assumption that the radar is functioning in an enhanced search mode (where the main beam targets 120 different azimuthal locations at 3° above the horizon every 2.5 seconds) (NRC 1979b). In this analysis, the signal (assumed to be a nominal 435 MHz carrier) was considered to be modulated both by the pulse pattern of the burst transmissions and by the scanning of the sidelobes as the main beam is repositioned. The power spectrum of the envelope of the enhanced search mode pulse train with a repetitive period of 54 msec contains a small amount of energy at 18.5 Hz (1/54

Suggested Citation:"4 Exposure Levels." National Research Council. 2005. An Assessment of Potential Health Effects from Exposure to PAVE PAWS Low-Level Phased-Array Radiofrequency Energy. Washington, DC: The National Academies Press. doi: 10.17226/11205.
×

msec) and its harmonics. The periodic interruptions (once every 18 cycles for antenna self-testing) in transmissions create small sidebands about these frequencies spaced approximately every 1 Hz (18 × 54 msec = 0.972 sec; 1/0.972 = 1.03 Hz). The 1979 NRC report (NRC 1979b) incorrectly states that 30% of the total power is at zero frequency. This finding was based on the envelope of the pulses. In reality, each pulse is filled with oscillations at the radar frequency (420-450 MHz); each pulse has a very small DC component. Because of this, the 18 msec maintenance sequence used every 972 msec (one of 18) also has very small DC components.

As noted in NRC 1979b, continuous operation in enhanced search mode is not probable in practice, meaning that the 18.5 Hz peak would be reduced with more power distributed around its overtones (37 Hz, 74 Hz, 148 Hz) resulting from the factors of two reduction in pulse widths that occur with the variation in transmission bursts. Additional sidebands would be present as well due to the quasi-periodic recurrence of the shorter interval pulses occurring within the 54 msec repetition rate. Further spectral spreading results from the pulse-to-pulse sampling of the antenna pattern sidelobes experienced at the fixed location on the ground as the direction of the main beam is changed. This net effect was estimated to create a uniform loss (~6.6 dB) across the band with further spectral spreading into the sidebands of the pulse-train envelop spectrum leading to a distribution dominated by sidelobe power in the 0-3 Hz band with less than 1% of the total in the 15-20 Hz range (NRC 1979b). The report describes a sample strip-chart recording of power measurements in the field ,which showed a 2 Hz fluctuation with a peak amplitude of approximately 0.4 µW/cm2 on which spikes in amplitude of about 1.4 µW/cm2 were superposed with a regular period of 0.4 Hz. These findings tend to confirm the estimate that the modulation of the sidelobe power is a few hertz or less on the ground.

PAVE PAWS POWER-DENSITY ESTIMATES AND MEASUREMENTS

Exposure levels on the ground from PAVE PAWS have been estimated and measured a number of times at various locations under several different assumptions and conditions. Six sources of such data are reviewed and summarized here: (1) the 1979 NRC (National Research Council) Engineering Panel report on Radiation Intensity of the PAVE PAWS Radar System (NRC 1979b); (2) the 1979 NRC report on Exposure Levels and Potential Biological Effects of the PAVE PAWS Radar System (NRC 1979a); (3) the 1986 Engineering Report (#86-33) issued by the 1839 Engineering Installation Group (EIG) from the Engineering Division at Keesler Air Force Base entitled “Radio Frequency Radiation Survey” for the AN/FPS-115 PAVE PAWS Radar (EIG 1986); (4) the 2000 report by Mitre Corporation on “RF Power Density Exposure at Ground Level for the PAVE PAWS Radar at Cape Cod—Questions and Answers” (MITRE 2000); (5) the Broadcast Signal Laboratory (BSL) final test report “A Survey of Radio Fre-

Suggested Citation:"4 Exposure Levels." National Research Council. 2005. An Assessment of Potential Health Effects from Exposure to PAVE PAWS Low-Level Phased-Array Radiofrequency Energy. Washington, DC: The National Academies Press. doi: 10.17226/11205.
×

quency Energy Field Emissions from the Cape Cod Air Force Station PAVE PAWS Radar Facility” issued in April of 2004 (BSL 2004b); and (6) the power density measurement data analysis from the Coalition to Operate PAVE PAWS Safely (COPPS) provided in May 2004.

1979 NRC Engineering Panel Report

The 1979 National Research Council Engineering Panel reviewed measurement data obtained from June 1978 through August 1978 as well as other recordings presented to that committee in September 1978 (NRC 1979b). These measurements were obtained with the radar operating in the enhanced search mode at a fixed carrier frequency with 20% duty cycle and were corrected for equivalency to measurements at 1 km distance from the radar with a 25% duty cycle (for comparison purposes). Data from 4 station locations ranging from 1600 to 3900 feet from the radar (at 0° to 63° in azimuth from boresight) resulted in average power densities of 0.38 to 3.26 mW/cm2, which when converted to equivalent 1 km, 25% duty cycle data, ranged from 0.64 to 0.98 mW/cm2 with an average (from the 4 locations) of 0.82 mW/cm2.

The Panel concluded that the measurement protocol and instrumentation used followed good engineering practice and that the variability in the measurement data (nearly 2 dB) was acceptable given the calibration procedures employed and differences in line of sight to the various measurement locations. Interestingly, in this same report (NRC 1979b), the panel made its own estimation of nominal and worst-case estimates of average power density on the main axis of a secondary sidelobe taking into account characteristics of the radar, and reported numbers of 8.2 µW/cm2 (nominal) and 11 mW/cm2 (worst-case) at 1 km, which are approximately an order of magnitude larger than the actual measurements but reassuring as conservative estimates. When factoring in more of the pulse-train specifics associated with enhanced search mode, the panel’s estimate reduced to 1.8 µW/cm2 at a distance of 1 km, which is only about 6 dB greater than the measurement data.

1979 NRC Exposure Level Report

This report (NRC 1979a) reviewed similar data measured by the Air Force in 1978 (August and October) at various points within, and up to 5 miles beyond, the restricted area, which is 1000 feet from the radar. It also considered analytical results for PAVE PAWS produced by the Environmental Protection Agency (EPA) in 1977. The August 1978 data were the same as reviewed in the 1979 NRC Engineering Panel report, although much more information was included in NRC 1979b. Specifically, at each of the 4 measurement sites, data from main-beam elevations of 3°, 6°, and 10° were evaluated along with both peak and average power-density recordings. The results showed a 344 µW/cm2 peak power

Suggested Citation:"4 Exposure Levels." National Research Council. 2005. An Assessment of Potential Health Effects from Exposure to PAVE PAWS Low-Level Phased-Array Radiofrequency Energy. Washington, DC: The National Academies Press. doi: 10.17226/11205.
×

density at 3° elevation at the closest distance (1600 feet) dropping to a peak power density of 179 µW/cm2 at 10° beam elevation with average power densities corresponding to 3.26 to 2.90 µW/cm2, respectively, for these two beam elevations. At the largest distance (3900 feet), the 3° peak power density reduced to 48 µW/cm2 with an average of 0.38 mW/cm2, while at 6° beam elevation these numbers were 17 µW/cm2 peak and 0.20 µW/cm2 on average (the 10° beam elevation was not reported).

Measurements were also summarized for 21 sites in surrounding locations (Bourne, Sandwich, Mashpee, and Falmouth, MA ranging in distance from 1.0 to 22.2 km from the radar) on 2 days in October 1978 using both faces of the radar with an 18% duty cycle and 3° beam elevation. The committee compared these data to calculated values of electric field strength and average power density predicted to be emitted by the radar. It found that, in general, (1) exposures decreased with distance from the radar (but not uniformly), (2) measured values decreased more rapidly than calculated quantities farther from the source (likely due to attenuation not modeled from terrain, atmosphere, etc.), and (3) the ratio of peak to average power density varied significantly depending on beam elevation, azimuthal angle, and distance from the radar. More specifically, the committee concluded that average power density outside the Air Force base was not likely to exceed 1 mW/cm2, the average power density at the exclusion fence surrounding PAVE PAWS was approximately 5 µW/cm2, while the average measured intensity at locations where the public would most likely be exposed was 0.06 mW/cm2 (Route 6, 1.0 km from the radar).

1839 EIG Engineering Report (#86-33)

In the fall (September 18-30) of 1986, time-averaged power-density measurements were recorded to document the RF exposures in the lighting and security camera areas within and around the security fence surrounding the PAVE PAWS radar facility on Cape Cod (EIG 1986). These data were requested because installation contracts for new lighting and camera systems at all four PAVE PAWS sites were being issued and construction workers would need access to locations directly in front of the radar. At the same time, the State of Massachusetts Department of Public Health was investigating possible causes of larger than normal cancer clusters on Cape Cod and the opportunity arose to also record RF power density levels at locations within the local community around the radar.

On site, measurements were reported for 17 lighting pole locations and 7 security camera positions immediately inside the security fence (approximately 120 feet from the radar face). Data were recorded at heights of 6 and 40 feet (above ground level) for the light poles and 6 and 20 feet (above ground level) for the security cameras. Additionally, ground-level (6 feet above) measurements were recorded at 10 other locations: 4 of these were outside the security fence but

Suggested Citation:"4 Exposure Levels." National Research Council. 2005. An Assessment of Potential Health Effects from Exposure to PAVE PAWS Low-Level Phased-Array Radiofrequency Energy. Washington, DC: The National Academies Press. doi: 10.17226/11205.
×

inside a secondary fence (between 120 and 200 feet from the radar face); 3 were outside the second fence (but within 250 feet of the radar); 2 were inside the security fence (one at the access gate lateral to the north face of the radar and the other directly in front of the north face 50-60 feet away); and 1 was not labeled on the map of measurement sites contained in the report but was listed in the data table of power-density recordings (presumably, it was associated with an 18th lighting pole not explicitly shown in the diagram, although data were only recorded at the 6 foot height). Power-density data were acquired at the 20 and 40 foot elevations (above the ground) using a bucket-loader truck that raised the test antenna and an operator to a height typically expected for maintenance personnel servicing either the security cameras (20 feet) or the lighting system (40 feet). A surveying transit was used to estimate the height of each measurement location relative to the bottom edge of the radar face. Because of the sloping terrain around the radar, these distances varied depending on the antenna test position. For the 6-foot (above ground) measurements, the heights ranged from even with to 14 feet below the bottom edge of the radar (at the greatest distance away, approximately 250 feet). The 20-foot elevations (at the security cameras) ranged from 5 to 14 feet above the bottom edge of the radar, while the 40-foot elevations (at the lighting poles) ranged from 31 to 45 feet above the radar’s bottom edge.

The equipment used during the recording sessions was calibrated and certified by the Keesler Air Force Base Precision Measurement Equipment Laboratory based on specifications issued by the National Bureau of Standards. Results showed that the time-averaged power-density recordings had an overall accuracy of ±2.0 dB when measuring a pulsed RF signal. When measurements were acquired, the PAVE PAWS radar was operating in its normal mode. The results showed that the highest time-averaged power-density at 6 feet above ground level was 1850 µW/cm2 with a minimum of 38 µW/cm2. The average was 529 µW/cm2 at the 33 locations comprising the lighting poles, security cameras, and selected positions in and around the security fence. Data at the elevation of the 7 security camera locations (20 feet above ground) had a maximum power density of 745 µW/cm2, a minimum of 100 µW/cm2, and a mean of 273 µW/cm2. Data at the elevation of the 17 lighting poles (40 feet above ground) had a maximum power density of 3716 µW/cm2, a minimum of 127 µW/cm2, and a mean of 1190 µW/cm2.

The same equipment and procedures were used to measure power densities within surrounding Cape Cod communities at a height of 6 feet above the ground for all except two locations: (a) the Sandwich Fire Tower, which was 86 feet above the ground and in direct line of sight of the radar, and (b) the Otis Central Control Tower which was 106 feet above ground. Locations were selected by the Massachusetts Department of Public Health to represent areas of dense population and to augment data obtained from earlier measurement efforts (NRC 1979a, b). Once a location was selected, the test antenna was placed in direct line of sight of the radar whenever possible or in an open area to record maxi-

Suggested Citation:"4 Exposure Levels." National Research Council. 2005. An Assessment of Potential Health Effects from Exposure to PAVE PAWS Low-Level Phased-Array Radiofrequency Energy. Washington, DC: The National Academies Press. doi: 10.17226/11205.
×

mum signal intensities. Data were measured at 15 locations in the surrounding communities ranging in distance from 1.2 to 8.8 miles from the radar. At 10 of the sites, power-density levels were below the measurable threshold of the equipment (determined to be 0.001 µW/cm2). At locations where levels above this threshold were received, the largest value, 0.139 µW/cm2, was documented at the Sandwich Fire Tower (3.2 miles away) at the 86-foot elevation, which was in direct line of site of the radar. The minimum measurable value, 0.003 µW/cm2, was recorded at the Otis Central Tower (5.9 miles away elevated 106 feet above ground). Of the measurable levels at 5 sites, the average power density was 0.041 µW/cm2. These locations averaged 3.2 miles away from the radar (ranging from 1.2 to 5.0 miles away).

2000 MITRE Report

In that report (MITRE 2000), two-dimensional peak and average power-density maps on the ground were presented for Cape Cod based on simulation studies of PAVE PAWS. Areas with line of sight to the radar were considered based on digital terrain elevation data for the Cape over a distance of 50 nautical miles. At these locations, the power density was calculated by considering the distance from the radar, the main beam elevation angle, and the time-averaging factors associated with the pulse-transmission scheme. For the peak-power calculation, a peak radiated power (543 kW), antenna gain (38.4 dB), and antenna pattern (beam at 3° elevation), along with a scan loss for off-broadside angles, were used to construct the maps. Generally, the peak-power densities in the 1-5 mile range from the radar cluster in the 1-10 µW/cm2 level but fall off to 0.3-1.0 µW/cm2 at distances 5-10 miles from the radar, although some variation with azimuthal angle is evident.

To estimate an average power-density map, the peak-power calculation was modified by the duty cycle, the scan revisiting time, and the azimuthal weighting factor. A duty cycle of 25% was used, a scan revisit factor (fraction of time the main beam is pointed in a particular direction given its beam width) of 0.066 was assumed, and an azimuthal-weighting factor that increased the average power 4 times (far from broadside) was applied. The map shows that beyond the first couple of miles from the radar, the average power density is within 0.01-0.1 mW/ cm2 and generally less than 1 mW/cm2 within the first couple of miles (except very close to the radar).

To simplify the analysis, peak and average intensity maps that assume the terrain is flat at a height of either 0 or 100 feet above sea level (with the radar at a height of 270 feet above sea level) are also included in this report. These results show peak-power densities (assuming sea-level terrain) of 1.0-10.0 mW/cm2 up to 5 miles or more away from the radar and a similar distribution of peak intensity at 100 feet above sea level with several isolated “hot spots” of 10-100 µW/cm2 at distances up to 2 nautical miles from the radar. The average power-density maps

Suggested Citation:"4 Exposure Levels." National Research Council. 2005. An Assessment of Potential Health Effects from Exposure to PAVE PAWS Low-Level Phased-Array Radiofrequency Energy. Washington, DC: The National Academies Press. doi: 10.17226/11205.
×

at sea level yielded values between 0.01 and 0.1 µW/cm2, with pockets having exposures between 0.1 and 0.3 µW/cm2 approximately 2 miles from the radar. The 100-foot elevation map is similar but with more uniform exposure between 0.1 and 0.3 µW/cm2 in a zone extending up to 4 miles from the radar with levels between 0.01 and 0.1 µW/cm2 existing beyond this zone.

The Broadcast Signal Lab Report

During the first quarter of 2004, the Broadcast Signal Lab, Medfield, MA (BSL 2004a, b), executed a survey of RF emissions from the PAVE PAWS radar facility located at the Massachusetts Military Reservation on Cape Cod. The final test plan was approved by the PAVE PAWS Public Health Steering Group (PPPHSG) based on its solicitation for competitive proposals to complete the work. It consisted of three distinct tasks: (1) measurement of radar emissions during its normal operation at 50 open, publicly accessible locations throughout Cape Cod; (2) measurement of ambient emissions from all other sources in the VHF and UHF radiofrequency spectrum (30 MHz to 3 GHz) at 10 locations on Cape Cod (these data are discussed in the subsection on other environmental exposures); and (3) estimation of the radar exposure on Cape Cod based on calculations from a mathematical description of the PAVE PAWS antenna used in conjunction with propagation-modeling software provided by the MITRE Corporation.

All measurement methods and procedures were consistent with guidelines and consensus standards for performing such work, and the 50 field locations for assessing radar exposures were selected in consultation with the PPPHSG and the International Epidemiology Institute (IEI). A number of factors including (1) distance, (2) elevation, (3) population density, and (4) beam sweep coverage were considered in the site selection process. A detailed discussion of these parameters and the 50 field measurement locations is available elsewhere (BSL 2004a). Table 4-1, reproduced from the Final Test Plan (BSL 2004a), is a useful summary of the number of measurements fulfilling certain site criteria regarding height, intervening terrain, distance, etc. Maps of the specifics of locations by town, county, and approximate coordinates can be found in the test plan (BSL 2004a) and final report (BSL 2004b). The actual measurement site locations were modified slightly at the time of data acquisition to ensure safety, accommodate access issues, and improve measurement quality as determined by the field crew. Measurements were recorded for 90-minute periods (as six 15-minute routines where the detector was moved 3 feet after each acquisition sequence) in order to acquire realistic estimates of the average and peak radar emissions at each location.

Summary tables of power-density measurements at the 50 sites within the communities of Cape Cod show that the average values of the recordings at each location are extremely small. The largest average value was 0.035 µW/cm2, which was recorded in Shawme Crowell State Park (site #15) at a location close to the

Suggested Citation:"4 Exposure Levels." National Research Council. 2005. An Assessment of Potential Health Effects from Exposure to PAVE PAWS Low-Level Phased-Array Radiofrequency Energy. Washington, DC: The National Academies Press. doi: 10.17226/11205.
×

TABLE 4-1

Measurement Site Characteristic

No. of Sites

% of Sites

Measured with 30 ft-high antenna

15

30

Radio line of sight to radar

36

72

Terrain obstruction to radar

13

26

Man-made obstruction to radar

1

2

Beneath beam sweep

40

80

Beyond beam sweep (includes rear)

10

20

Less than 3 miles from radar

16

32

3 -10 miles away from radar

19

38

More than 10 miles away from radar

15

30

Barnstable County

44

88

Plymouth County

6

12

radar (less than 1 mile) that was relatively high (167 feet above sea level). The next highest sets of recording levels were nearly an order of magnitude lower (~0.002-0.005 µW/cm2) and occurred at 4 other locations (site # 9, 21, 23, and 40) that generally tended to be closer to the radar (within 3 miles) although at variable heights (9-167 feet above sea level). One interesting exception was the Scargo Hill site whose average power density was 0.0038 µW/cm2 at some 18.5 miles from the radar (65 foot elevation). Direct line from the radar to this location does cut across the coast line of the Cape, which may explain its relatively high (compared to radar distance) values, although the direct paths to sites 1-5 and 9 cut across mostly water (at distances ranging from 24 to 31 miles from the radar) and do not appear elevated compared to other locations. The 5 sites (# 9, 15, 21, 23, and 40) with the highest average values, not surprisingly, also shared the highest peak-power recordings in the 14-15 µW/cm2 range. These peaks were, again, nearly an order of magnitude greater than peaks measured at the other 45 locations. The next cluster of peak values hovered in the range of 1.4-1.5 µW/cm2 and were found at 10 other sites (# 14, 16, 18-20, 22, 25-27, 47), which were all within 6 miles of the radar and averaged a distance of 2.6 miles away.

Of the 5 sites with the highest average and peak power-density measurements, 4 were located in the sidelobe overlap zone (the 5th was in the northern face sweep zone). The majority of the 10 sites with the next highest levels of peak power density (all but one, #14) appeared in either the northern sweep (5 sites) or sidelobe overlap (4 sites) zones. Although there is an overall decrease in detected signal levels with distance, sites close to the radar (e.g., Route 6E Canal Over-look, #29, 1.9 miles away) can have low levels (0.135 µW/cm2 peak), while locations at much further distances (e.g., Rock Harbor Parking, #5, 27.5 miles away) can exhibit relatively higher levels (0.153 µW/cm2 peak), reflecting the importance of the intervening and local terrain in addition to the direct-line distance.

Comparisons between the current measurements and those recorded within

Suggested Citation:"4 Exposure Levels." National Research Council. 2005. An Assessment of Potential Health Effects from Exposure to PAVE PAWS Low-Level Phased-Array Radiofrequency Energy. Washington, DC: The National Academies Press. doi: 10.17226/11205.
×

the community in either 1978 (NRC 1979a,b) or 1986 (EIG 1986) show that the present equipment was significantly more sensitive and able to achieve recordings of weaker signals than was possible during the earlier efforts. At sites within 7 miles of the radar (previous studies did not measure responses at greater distances), the 2004 data are consistently lower than the prior recordings. Specifically, the average power-density readings for sites within 7 miles were below 0.01 µW/cm2 in the 2004 data with many points below 0.0001 µW/cm2. The 1978 and 1986 results showed average power densities between 0.1 and 0.01 µW/cm2 for most sites within 3 miles of the radar with the largest value being above 0.1 µW/cm2 at 3-plus miles from the radar in the 1986 data set. About one-half (4 locations) of the 1978 readings were between 0.01 and 0.001 µW/ cm2 and a third (2 sites) of the 1986 results were in this range (levels below 0.001 µW/cm2 were not considered to be measurable with the instrumentation used in the earlier power-density surveys).

The modeling effort produced results from the propagation model combined with an antenna pattern to generate a matrix of public-exposure estimates to the PAVE PAWS emissions. The propagation model was evaluated by comparing its predictions to measurements obtained in 3849 latitude-longitude cells acquired during a 250-mile drive test. The comparison between computed and measured values indicates that the model is accurate to within the uncertainty of the measurements. Specifically, the average variance between the propagation model prediction and the drive-test recordings was −1.6 dB. A MITRE antenna model was used to create an antenna pattern for the propagation software. This led to the creation of the exposure matrix for Cape Cod, indexed by latitude and longitude, containing values for each position in terms of dBµW/cm2. To validate the results when combining the antenna and propagation models, measurements at 35 sites were compared with the corresponding calculations. Departure between the model estimates and the measured data was 6.5 dB on average over all sites with a standard deviation of 8.5 dB. Excluding 4 sites behind the radar reduced the average departure to 5.9 dB. Overall, the model estimates slightly over-predicted the levels measured in the field.

COPPS Data

The Coalition to Operate PAVE PAWS Safely (COPPS) performed power density measurements on Cape Cod at 26 sites (23 firehouses, 2 commuter parking lots, and 1 lighthouse) spanning 311 days beginning in May of 2003. Analysis of existing data as of April 2004 was made available to the committee through website postings (see COOPS 2004). Specifically, a town-by-town map of measurement sites, a color-coded rank ordering by measured (http://www.pavepaws.com/measured_sites.pdf) peak-power density (in µW/cm2) of the towns where data were recorded (http://www.pavepaws.com/peak_measurements.pdf), and plots of peak-power versus various parameters of interest (e.g. elevation, dis-

Suggested Citation:"4 Exposure Levels." National Research Council. 2005. An Assessment of Potential Health Effects from Exposure to PAVE PAWS Low-Level Phased-Array Radiofrequency Energy. Washington, DC: The National Academies Press. doi: 10.17226/11205.
×

tance, obstruction, bearing, weather, season) (http://www.pavepaws.com/additional_findings.pdf) were provided. The results are generally consistent with the Broadcast Signal Lab (BSL) report (BSL 2004b) in a number of important respects, although the details of specific measurements may be different (and are not easily compared given information presently available). The COPPS data show peak-power densities of less than 1 µW/cm2 regardless of location, with more than 75% of the 40 peak values reported (http://www.pavepaws.com/additional_findings.pdf) being near or below 0.1 µW/cm2. These levels are lower than the highest peak values recorded by BSL but consistent with the second and third tiers of peak values into which the majority of the BSL measurement sites also fall. Both the BSL and the COPPS data sets reveal that the locations with higher peak power-density recordings are preferentially distributed in the northern sector of the radar. Both data sets also indicate that some of the highest peak power-densities occur at locations having the greater distances from the radar, especially when the straight-line distance from the radar to the measurement site cuts across the coastal waters of Cape Cod Bay. Other findings in the COPPS data are consistent with expectations (and the BSL results) that the degree of topological obstruction between the measurement location and the radar facility is a critical determinant of the measured power-density exposure. Positive correlations between power-density levels and clear weather and the fall/winter seasons were reported in the COPPS analysis. Again, these results are reasonably anticipated from the perspective that inclement weather, increased foliage, and other factors are expected forms of obstruction to radiowave propagation at PAVE PAWS frequencies.

POPULATION EXPOSURE TO OTHER RF SOURCES

It is informative to consider general and specific population exposures to other RF sources that have some similarities to PAVE PAWS for gaining a perspective on the power-density levels described in this chapter. Interestingly, the 1979 engineering report (NRC 1979b) compared the transmitter characteristics of PAVE PAWS with two other radar installations that radiate much higher peak powers (megawatts) and higher (for one) and lower (for the other) average powers but no power-density measurements on the ground were reported. In NRC (1979a), data obtained by the EPA in the 1970s (EPA 1978a) from population exposures in 12 large U.S. cities over the FM and TV broadcast bands (54-900 MHz) were summarized. These studies estimated that the median exposure in urban areas was 0.005 µW/cm2 and that 95% of the urban population is exposed to less than 0.1 mW/cm2. These numbers are generally consistent with the PAVE PAWS data in terms of power-density exposure. The EPA has also reported power-density measurements in tall buildings that either house, or are near, broadcast antennas (EPA 1978b). These studies show that, depending on the location within the building and/or the degree of shielding from the antenna site, values

Suggested Citation:"4 Exposure Levels." National Research Council. 2005. An Assessment of Potential Health Effects from Exposure to PAVE PAWS Low-Level Phased-Array Radiofrequency Energy. Washington, DC: The National Academies Press. doi: 10.17226/11205.
×

can range from less than 1 µW/cm2 to 97 µW/cm2 on the inside to as much as 230 µW/cm2 outside (and near) the source.

The Broadcast Signal Laboratory study (BSL 2004b) selected 10 sites for assessing the ambient RF exposure that were chosen in consultation with the PPPHSG and IEI, and a detailed discussion of the rationale for their selection appears in BSL 2004a. A bracketing strategy was adopted because, as with the PAVE PAWS radar, itself, the levels of ambient VHF and UHF radiation measured at a given location depend on distance from the source, the intervening terrain, and local topological environment. At least a dozen FM radio stations (a number of which operate with radiated powers of 45 kW or more) and a couple of TV stations (including one UHF station licensed to operate at 1150 kW) are present in the area. Numerous (more than 5000) land-mobile facilities, typically licensed for 100 to 1000 watts per channel emissions (e.g. police, fire, business systems) were also identified. The final 10 sites selected attempted to represent high, low, and middle-level exposure conditions to these sources as a means of providing a representative sampling of the ambient RF exposures on Cape Cod. The propagation model was used to define the power levels and associated locations that would be expected to bracket exposure conditions given the relevant sources of radiation identified. Descriptions of the final site selections are reported (BSL 2004b). Four pre-measurement bracketing level estimates spanning 10−1-10−5 (i.e., 10−1, 10−2, 10−4, 10−5) µW/cm2 were used to categorize the 10 sites.

Measurements at the 10 sites selected to bracket the levels of ambient RF exposure across the VHF and UHF signal bands were reported in terms of the maximum permissible exposure (MPE) for low (30-320 MHz), mid (320-950 MHz), and high (950-3050 MHz) bands that were combined into a final figure of merit that could be compared to the MPE for PAVE PAWS. These findings were tabulated as dB below MPE. For example, the MPE for PAVE PAWS (determined by applying the IEEE weighting for the 420-450 MHz band) is 290 µW/cm2 thus the maximum and minimum measured power densities from PAVE PAWS (at the 50 measurement sites) convert to −48 dB (below MPE) and −103 dB, respectively, whereas the equivalent values for the ambient signals from the 10 sites resulted in maximum and minimum measured values of −19 dB and −50 dB (relative to MPE). These results indicate that the ambient exposure level at the most energetic site (near the FM tower at the Exit 6 commuter parking lot on Route 6) was 100 times less that the MPE for the general population. The other sites were an additional 10 to 1000 times less exposed to VHF and UHF emissions. Overall, the highest average PAVE PAWS exposure at any of the 50 measurement sites was comparable to the lowest ambient levels observed at the 10 locations sampled. Further, the variation between ambient sites was no more than a factor of 1000 (30 dB), while the lowest observed PAVE PAWS signal was about a million (60 dB) times weaker than the highest.

Cell-phone exposure has generated considerable interest, and there have been

Suggested Citation:"4 Exposure Levels." National Research Council. 2005. An Assessment of Potential Health Effects from Exposure to PAVE PAWS Low-Level Phased-Array Radiofrequency Energy. Washington, DC: The National Academies Press. doi: 10.17226/11205.
×

a number of efforts to quantify exposure. Two issues are relevant: first, the exposure level to the cell-phone user due to the radiated power during a call, and second, the exposure of the general population to cell-phone relay stations. The former issue has been studied in some depth. The power radiated for a typical cell phone is generally less that 0.6 W and it has been shown that approximately half of this power is absorbed in the hand and head of the user, leading to specific absorption rates on the order of 1 W/kg (Jensen and Rahmat-Samii 1995).

The committee is not aware of extensive studies of power-density levels near relay stations (base stations) in the United States, but there are 3 informative evaluations from Canada, the United Kingdom, and Australia.1 In Vancouver, Canada, Thansandote and others (1999) measured RF levels in 3 schools with base stations on or near their campuses and 2 schools with no antennas nearby that served as controls. Maximum measured exposures in the 3 schools with adjacency ranged from 0.16 µW/cm2 to 2.6 µW/cm2 depending on the location of the station (2.6 mW/cm2 resulted from a base station on the roof of a school). The 2 schools without base stations in their vicinity had a maximum power density of 0.01 µW/cm2.

The U.K. National Radiological Protection Board measured RF power-density levels at 118 sites around 17 base stations in 2000 (NRPB 2000). Typical levels were less than 0.1 µW/cm2, while the highest recorded power density at any location was 0.83 µW/cm2. The measured upper bound on power density as a function of distance from the base of a building or tower with an antenna showed an increase (from just below 0.01 µW/cm2) for approximately the first 60 meters (to a level approaching 0.1 µW/cm2) followed by a more gradual decrease (back to 0.01 µW/cm2) up to 200 meters away. The power-density auditing program in the U.K. has continued with data available from recording in 2001-2004. Results are summarized in terms of a percentage of the ICNIRP (International Commission for Non-ionizing Radiation Protection) power-density guidelines for a band exposure equivalent that factors in the changing safety standard as a function of frequency from 400 MHz to 2 GHz, which for the PAVE PAWS emissions would represent an upper limit of 200 mW/cm2 (at 400 MHz). The highest values reported at any location for each year would represent PAVE PAWS-like exposures (i.e., band equivalent exposure extrapolated to 400 MHz) of 0.72 µW/cm2, 0.27 µW/cm2, 0.32 µW/cm2, and 0.26 µW/cm2 for the years 2001, 2002, 2003, and 2004, respectively.

The survey of GSM base stations by the Australian Radiation Protection and Nuclear Safety Agency measured 13 sites that yielded an average power density

Suggested Citation:"4 Exposure Levels." National Research Council. 2005. An Assessment of Potential Health Effects from Exposure to PAVE PAWS Low-Level Phased-Array Radiofrequency Energy. Washington, DC: The National Academies Press. doi: 10.17226/11205.
×

of less than 0.1 µW/cm2; the highest power density measured was less than 0.2 cm2.

SUMMARY OF PAVE PAWS EXPOSURE DATA

While not extensive per se, the power-density measurements that do exist for the PAVE PAWS radar on Cape Cod, which have been recorded by different groups on a number of separate occasions over the radar’s operational life, are generally consistent with expectations. The most extensive efforts in terms of the number and distribution of measurement sites are the recently issued 2004 reports (BSL 2004a, b; COPPS 2004), which have focused on assessments of exposure levels within the surrounding communities on Cape Cod. These studies reveal heterogeneity in the measured levels that demonstrate the strong influence of the intervening terrain and local topological environment on the recorded power densities and consequently the public exposure. While there is a general decrease in the power-density levels measured with distance from the radar that is evident in the various datasets that have been assembled, it is also clear that locations much farther (10-plus miles) from the radar can be exposed to higher levels of radiation than sites nearer (within 3 miles) the facility depending on topographical features. Certainly, both the BSL and the COPPS data (BSL 2004a, b; COPPS 2004) demonstrate this characteristic at selected locations. Nonetheless, the average power densities have been consistently below 0.1 mW/cm2 and generally in the 0.001-0.01 mW/cm2 range at locations where the public would be expected to be exposed. The simulation studies that have been reported estimate higher values overall, but for the most part still suggest that the exposure of the public is less than 1 µW/cm2 in terms of average power density. Measured peak levels are generally less than 1 µW/cm2, although values as high as 15 µW/cm2 have been found at a few elevated locations near the radar. These levels are well below the widely accepted thresholds for the power absorption that would be required to generate harmful thermal effects in tissue. The degree to which they may cause other biological effects in cells and animals is discussed in Chapters 6 and 7, respectively, and the implications for human health are developed in Chapters 8 and 9. The degree to which emissions from PAVE PAWS may contain special characteristics pertaining to its time-domain waveform is described in the following chapter (Chapter 5).

REFERENCES

ARPANSA (Australian Radiation Protection and Nuclear Safety Agency). 2000. Line, P., W.A. Cornelius, M.J. Bangay and M. Grollo. Levels of Radiofrequency Radiation from GSM Mobile Telephone Base Stations (Tech Rep 129). Available from http://www.arpansa.gov.au/pubs/eme_comitee/rfrep129.pdf as of 10/25/04.

Suggested Citation:"4 Exposure Levels." National Research Council. 2005. An Assessment of Potential Health Effects from Exposure to PAVE PAWS Low-Level Phased-Array Radiofrequency Energy. Washington, DC: The National Academies Press. doi: 10.17226/11205.
×

BSL (Broadcast Signal Lab). 2004a. Survey of RF Energy Field Emissions from the PAVE PAWS Radar Located at Cape Cod Air Force Station, Massachusetts: Final Test Plan, prepared in consultation with International Epidemiology Institute, Rockville, MD, by Broadcast Signal Lab, Medfield MA.

BSL. 2004b. A Survey of Radio Frequency Energy Field Emissions from the Cape Cod Air Force Station PAVE PAWS Radar Facility: Final Test Report, prepared for PAVE PAWS Public Health Steering Group by Broadcast Signal Lab, Medfield MA.


COPPS (Coalition to Operate PAVE PAWS Safely). 2004. Available from http://www.pavepaws.com/measured_sites.pdf; http://www.pavepaws.com/peak_measurements.pdf; http://www.pavepaws.com/additional_findings.pdf.


EIG (Engineering Installation Group). 1986. Radio Frequency Radiation Survey for the AN/FPS-115 PAVE PAWS Radar Cape Cod MA, 18-30 Sep 86, Report #86-33, 1839 Engineering Installation Group, Keesler Air Force Base, Biloxi, MS.

EPA (Environmental Protection Agency). 1978a. Population Exposure to VHF and UHF Broadcast Radiation in the United States, R.A. Tell and E.D. Mantiply, EPA Technical Report ORP/EAD 78-5.

EPA. 1978b. Measurements of Radiofrequency Field Intensities in Buildings with Close Proximity to Broadcast Stations, R.A. Tell and N.N. Hankin, EPA Technical Report ORP/EAD 78-3, March 1978.


Jensen, M. and Y. Rahmat-Samii. 1995. Interaction of handset antennas and a human in personal communications. P IEEE 83:7-17.


MITRE. 2000. RF Power Density Exposure at Ground Level for the PAVE PAWS Radar at Cape Cod—Questions and Answers. MITRE Technical Report, MITRE Corporation, Bedford, MA.


NRC (National Research Council). 1979a. Analysis of the Exposure Levels and Potential Biologic Effects of the PAVE PAWS Radar System. Washington, DC: National Academy Press.

NRC. 1979b. Radiation Intensity of the PAVE PAWS Radar System, Engineering Panel on the PAVE PAWS Radar System, National Research Council, Final Report. Washington, DC: National Academy Press.

NRPB (National Radiological Protection Board). June 2000. Mann, S.M., T.G. Cooper, S.G. Allen, R.P. Blackwell, and A.J. Lowe. Exposure to Radio Waves near Mobile Phone Base Stations. Chilton, U.K.: National Radiological Protection Board.


Thansandote, A., G.B. Gajda, and D.W. Lecuyer. 1999. Radiofrequency radiation in five Vancouver schools: exposure standards not exceeded. Can Med Assoc J 160:1311-1312.

Suggested Citation:"4 Exposure Levels." National Research Council. 2005. An Assessment of Potential Health Effects from Exposure to PAVE PAWS Low-Level Phased-Array Radiofrequency Energy. Washington, DC: The National Academies Press. doi: 10.17226/11205.
×
Page 37
Suggested Citation:"4 Exposure Levels." National Research Council. 2005. An Assessment of Potential Health Effects from Exposure to PAVE PAWS Low-Level Phased-Array Radiofrequency Energy. Washington, DC: The National Academies Press. doi: 10.17226/11205.
×
Page 38
Suggested Citation:"4 Exposure Levels." National Research Council. 2005. An Assessment of Potential Health Effects from Exposure to PAVE PAWS Low-Level Phased-Array Radiofrequency Energy. Washington, DC: The National Academies Press. doi: 10.17226/11205.
×
Page 39
Suggested Citation:"4 Exposure Levels." National Research Council. 2005. An Assessment of Potential Health Effects from Exposure to PAVE PAWS Low-Level Phased-Array Radiofrequency Energy. Washington, DC: The National Academies Press. doi: 10.17226/11205.
×
Page 40
Suggested Citation:"4 Exposure Levels." National Research Council. 2005. An Assessment of Potential Health Effects from Exposure to PAVE PAWS Low-Level Phased-Array Radiofrequency Energy. Washington, DC: The National Academies Press. doi: 10.17226/11205.
×
Page 41
Suggested Citation:"4 Exposure Levels." National Research Council. 2005. An Assessment of Potential Health Effects from Exposure to PAVE PAWS Low-Level Phased-Array Radiofrequency Energy. Washington, DC: The National Academies Press. doi: 10.17226/11205.
×
Page 42
Suggested Citation:"4 Exposure Levels." National Research Council. 2005. An Assessment of Potential Health Effects from Exposure to PAVE PAWS Low-Level Phased-Array Radiofrequency Energy. Washington, DC: The National Academies Press. doi: 10.17226/11205.
×
Page 43
Suggested Citation:"4 Exposure Levels." National Research Council. 2005. An Assessment of Potential Health Effects from Exposure to PAVE PAWS Low-Level Phased-Array Radiofrequency Energy. Washington, DC: The National Academies Press. doi: 10.17226/11205.
×
Page 44
Suggested Citation:"4 Exposure Levels." National Research Council. 2005. An Assessment of Potential Health Effects from Exposure to PAVE PAWS Low-Level Phased-Array Radiofrequency Energy. Washington, DC: The National Academies Press. doi: 10.17226/11205.
×
Page 45
Suggested Citation:"4 Exposure Levels." National Research Council. 2005. An Assessment of Potential Health Effects from Exposure to PAVE PAWS Low-Level Phased-Array Radiofrequency Energy. Washington, DC: The National Academies Press. doi: 10.17226/11205.
×
Page 46
Suggested Citation:"4 Exposure Levels." National Research Council. 2005. An Assessment of Potential Health Effects from Exposure to PAVE PAWS Low-Level Phased-Array Radiofrequency Energy. Washington, DC: The National Academies Press. doi: 10.17226/11205.
×
Page 47
Suggested Citation:"4 Exposure Levels." National Research Council. 2005. An Assessment of Potential Health Effects from Exposure to PAVE PAWS Low-Level Phased-Array Radiofrequency Energy. Washington, DC: The National Academies Press. doi: 10.17226/11205.
×
Page 48
Suggested Citation:"4 Exposure Levels." National Research Council. 2005. An Assessment of Potential Health Effects from Exposure to PAVE PAWS Low-Level Phased-Array Radiofrequency Energy. Washington, DC: The National Academies Press. doi: 10.17226/11205.
×
Page 49
Suggested Citation:"4 Exposure Levels." National Research Council. 2005. An Assessment of Potential Health Effects from Exposure to PAVE PAWS Low-Level Phased-Array Radiofrequency Energy. Washington, DC: The National Academies Press. doi: 10.17226/11205.
×
Page 50
Suggested Citation:"4 Exposure Levels." National Research Council. 2005. An Assessment of Potential Health Effects from Exposure to PAVE PAWS Low-Level Phased-Array Radiofrequency Energy. Washington, DC: The National Academies Press. doi: 10.17226/11205.
×
Page 51
Suggested Citation:"4 Exposure Levels." National Research Council. 2005. An Assessment of Potential Health Effects from Exposure to PAVE PAWS Low-Level Phased-Array Radiofrequency Energy. Washington, DC: The National Academies Press. doi: 10.17226/11205.
×
Page 52
Next: 5 PAVE PAWS Exposure Conditions »
An Assessment of Potential Health Effects from Exposure to PAVE PAWS Low-Level Phased-Array Radiofrequency Energy Get This Book
×
Buy Paperback | $64.00 Buy Ebook | $49.99
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

PAVE PAWS is a phased-array warning system designed to detect and track sea-launched and intercontinental ballistic missiles operated on Cape Cod since 1979 by the U.S. Air Force Space Command. In 1979, the National Research Council issued two reports to address concerns from Cape Cod residents about the safety and possible health effects of the radiofrequency energy from the radar. Following up on the1979 report, the new report finds no evidence of adverse health effects to Cape Cod residents from long-term exposure to the PAVE PAWS radar. The report specifically investigated whether the PAVE PAWS radar might be responsible in part for the reported higher rates of certain cancers in the area, but concludes there is no increase in the total number of cancers or in specific cancers of the prostate, breast, lung, or colon due to radiation exposure from PAVE PAWS. The report did find in the scientific literature a few biological responses to radiofrequency exposures that were statistically significant. Such responses do not necessarily result in adverse health effects, but the report recommends additional studies to better discern the significance, if any, of those findings.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    Switch between the Original Pages, where you can read the report as it appeared in print, and Text Pages for the web version, where you can highlight and search the text.

    « Back Next »
  6. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  7. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  8. ×

    View our suggested citation for this chapter.

    « Back Next »
  9. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!