Analysis of the Exposure Levels and Potential Biologic Effects of the PAVE PAWS Radar System (1979)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

CHAPTER 1 CHARACTERIZATION OF PAVE PAWS EXPOSURE CONDITIONS OPERATING CHARACTERISTICS OF PAVE PAWS RADAR SYSTEM AS RELATED TO HUMAN EXPOSURE "PAVE PAWS" is the name used by the U.S. Air Force for a fixed-base solid-state radar system that includes two radars of identical design— one at Otis Air Force Base on Cape Cod, Massachusetts, scheduled to go into operation in April 1979, and another at Beale Air Force Base in California, scheduled to go into operation a year later. The radar at Otis Air Force Base is the subject of this report. This section describes the operating characteristics of the PAVE PAWS radar system, not in the context of engineering and electronic detail, but with reference to how the characteristics are related to human exposure to the radar. In several instances, to provide a familiar reference source, there are comparisons with frequency-modulated (FM) radio broadcast-station operating conditions. The primary function of PAVE PAWS is the early detection of the approach of ballistic missiles launched from submarines or ships. During normal operation, the radar scans near the horizon across the sector of potential approach. It is designed with the power and sensitivity to detect a launch booster as it appears above the horizon. Once detected, the launched object is continuously tracked and its trajectory estimated. Any object that separates from a booster, as a missile would, is tracked as it approaches. A secondary function of PAVE PAWS is to track satellites as part of the National Space Track net. Information about objects that are classified as threatening on the basis of their trajectories and expected points of impact is transmitted to the North American Air Defense complex in Cheyenne Mountain, Colorado, to the Strategic Air Command in Omaha, Nebraska, and to the National Military Command Center and Alternate National Military Command Center. PAVE PAWS is a phased-array radar, whose radiation is emitted from a mechanically fixed antenna, rather than from an antenna that scans or is pointed by being moved. The antenna consists of an array of many small radiating elements. The radiation beam is focused and pointed in a desired direction by controlling the manner in which the individual elements radiate. -7-

If the beam is to be pointed to the left of straight ahead (or "boresight"), the signals from the elements on the left side of the array are delayed relative to those emitted from the elements on the right, with the delay increasing progressively across the array from right to left. The radar, housed in a triangular building, has two antenna faces, which face in directions 120 apart, as shown in Figure 1. At the Otis site, the two boresight directions are 47° true north and 167° true north. The radar scans the sector from 347° (13° west of north) to 227° (47° west of south). Each antenna face is tilted back by 20°, so that the boresight line tilts 20° above the horizontal. Only one antenna face transmits at a time. In the normal search operation, the beam scans in a stepwise manner in a somewhat regular sequence across the sector of view, at 3° above the horizon. Free time during a search operation is used to direct the beam to any targets that are under track. Targets under track can be anywhere in the hori- zontal sector of search and from 3° to 85° above the horizon. The actual pattern of beam positions followed during any one scanning and tracking cycle depends on the particular search mode that has been selected and on the positions of the targets being tracked. Each face of the antenna is a metal plane from which a regular array of 5,353 antenna elements protrude, as shown in Figure 2. Of these 5,353 elements, 2,676 around the periphery are totally inactive; they would make it possible to increase the size of the antenna and the power of the radar. The 2,677 active elements in the central region of each array constitute the present antenna proper—each antenna face is approximately 72 ft (22 m) in diameter. Of the elements in each face, 1,792 are active transmitt- ing elements; each is connected to a solid-state transmitter (and receiver) module. The remaining 885 electrically active elements in each face are not connected to power sources and serve only to improve control of the shape of the beam. In normal operation, the PAVE PAWS beam is not directed below an elevation of 3° above the horizon. At Otis Air Force Base, the radar is on high ground, and the main beam of the antenna is always at least 100 ft (30 m) above ground level at the nearest points of public access. In an analysis of public exposure to PAVE PAWS radiation, then, the questions that bear on the radar's performance are: o In normal operation, what are the nature and intensity of the radiation that "spills over" to the ground from the axis of the beam? o Can abnormal conditions, such as rain, unfavorably affect the snape of the beam or the intensity at ground level? -8-

FIGURE 1. PAVE PAWS radar, showing triangular structure and the two antenna faces. -9-

FIGURE 2. Regular array of antenna elements protruding from one face of PAVE PAWS radar. -10-

o What safety features can detect abnormal operation or prevent in- advertent focusing of the beam below the normal 3° minimum? The shape of the antenna beam in normal operation has been determined by theoretical and computational analysis, and predictions of exposure levels derived from this analysis can be verified by measurements in situ. A panel of the National Research Council Assembly of Engineering (AE) has reviewed the antenna design (which is related to the first two questions), the safety features (which are related to the third question), and the Air Force's measurement program. The present report describes the general features of PAVE PAWS operation and the radiation fields that may occur in nearby public areas. FREQUENCY The radar operates in ultrahigh-frequency (UHF) band. There are 24 frequencies, from 420 to 450 megahertz (MHz), at which transmission may take place. No significant transmitted energy falls outside the band from 420 to 450 MHz. These frequencies correspond to wavelengths of 60-70 cm (23.6-27.6 In.). POLARIZATION The transmitted wave is right-circularly polarized. POWER A long-range radar like PAVE PAWS operates by transmitting brief pulses of energy or short trains of pulses, followed by longer periods during which the transmitter is off and the receiver is sensitive to re- turning echoes. The radiated power is described in terms of "peak power" and "average power." Peak power is the rate at which energy is emitted during the period of one pulse; that is, peak power is the energy of one pulse divided by the duration (in seconds) of that pulse. Average power is the time-averaged rate of energy emission over a period that is long compared with the pulse duration. The duty cycle is the fraction of time, on the average, that the transmitter is transmitting. The average power is determined by multiplying the peak power by the duty cycle. In assessing the effects of radiation, both peak power and average power must be considered. Peak power governs the instantaneous inten- sity of the electric and magnetic fields induced in or near a body on which radiation impinges. Average power governs the rate at which energy impinges on a body or medium in the field and therefore governs the rate at which heat may be generated or other effects may be produced in an absorbing body or medium. -11-

TRANSMITTED POWER Each of the 1,792 transmitting elements in each of the two PAVE PAWS antenna arrays is connected to its own transmitter; the combination of element and transmitter is called a "transmitting module." The pulsed output of each transmitter is such that the element radiates at approximately 320 W at peak power. The peak transmitted power of the radar is thus about 580 kW (320 W x 1,792). The time sequence of pulses transmitted by PAVE PAWS during a given interval depends on the functions being performed and the number of tar- gets being tracked. However, controls limit the rate at which pulses are transmitted, and the duty cycle never exceeds 0.25. Therefore, the average transmitted power never exceeds 145 kW (0.25 x 580 kW). In round numbers, this is about 3 times the average power transmitted by a typical large TV station and somewhat more than the average power transmitted by a typical high-power-output FM broadcast station (but see the discussion of antenna patterns below). The most powerful FM and TV stations radiate more power than PAVE PAWS. ANTENNA PATTERN Th3 key to the operation of the phased-array antenna is that, at dis- tances beyond a few hundred feet from the antenna and in all directions that are separated by even a few degrees in angle from the direction in which the beam is pointed, the electromagnetic fields of the individual transmitting elements cancel each other, whereas in directions along the beam the fields add to each other. The detailed structure of the result- ing radiation pattern is complex, in that it varies with the precise position in which the beam is pointed and with the transmission frequency. However, the general features of the pattern can be conceptualized and measured. If one thinks of the antennas as a searchlight, an observer would see the main beam and then numerous secondary beams, called "sidelobes." The sidelobes would be of much lower intensity and would point in directions divergent from that of the main beam. If an observer were to move as little as 1° off the axis of the main beam (i.e., about 90 ft off the axis at 1 mile away, or 17 m at 1 km away), he would observe that the apparent intensity of the searchlight was decreased to half that on the main-beam axis. The main beam would appear to extinguish by the time he moved 2° from the axis. A typical sidelobe has a shape resembling that of the main beam, but possibly broader. The so-called first sidelobes form a cluster whose direction is close to that of the main beam. Each, on its own axis, has an intensity not more than 1% of that of the main beam. If the observer were 4° off the axis of the main beam, even the first sidelobes would all be extinguished. Beyond the cluster of the first -12-

sidelobes, there are other sidelobes of lower intensity, none having more than 0.1% of the intensity of the main beam. RADIATION AT GROUND LEVEL Where the main beam of PAVE PAWS is pointed to 3° or more above the horizon, the maximum of the radiation field at or near ground level is governed by the sidelobes of the antenna pattern other than the first sidelobes. This field is therefore limited in intensity to 0.1% or less of the intensity of the main beam. At 0.1%, the radiation field corre- sponds to that which would be observed from a 1,000-kW transmitter radiating uniformly in all directions. An observer on the ground, situated exactly in the direction of peak intensity of one of the secondary sidelobes, would be exposed to radiation at no more than 0.1% of the intensity of the main beam. Peaks of this intensity are necessarily distributed very sparsely over all angles, because most of the 145 kW of average transmitted power is designed to radiate along the direction of the main beam. Only a small fraction is left to radiate in other directions. Thus, over all angular positions remote from the main beam, the antenna output is similar, on the average, to a source whose average power is only a minute fraction of the maximal intensity of a secondary sidelobe, as estimated above. In normal operation, the main beam of the antenna is directed se- quentially to many different directions even during a fraction of a second, and transmissions occur at different frequencies. This has the effect, of altering from pulse to pulse the sidelobes to which an observer at ground level may be exposed. The average exposure of this observer therefore tends to be comparable with that encountered at an "average" angular position remote from the main beam. Both calculations and measurements on the PAVE PAWS antenna indicate that the average power at ground level near PAVE PAWS should be no greater than one-fifth to one-fourth of the maximum at the peak of a secondary sidelobe. With respect to average radiation intensity at nearby points on the ground, the PAVE PAWS radar can therefore be regarded as a source radia- ting isotropically at an average power not greater than 250 kW (0.25 x 1,000 kW), which is comparable with the power of the most powerful FM and TV broadcast stations. Furthermore, these broadcast stations use antennas that concentrate radiation near the horizon, to serve the intended listening areas better. Radiation intensity in a listening area is therefore greater than that received from an isotropic source of the same power. The PAVE PAWS radar operates at a wavelength about one-fifth that of an FM station or of TV stations on channels 1-6, about the same as that of UHF TV stations, and about twice that of TV stations on channels 7-13. -13-

PAVE PAWS and other types of radar radiate energy in short pulses of high peak power, creating a signal that is very different from that of an FM or TV station. The nature of radar signals is examined further below. MODULATION The pattern of pulses emitted by the PAVE PAWS radar can be quite complex, because it depends on the number of targets being tracked, the distances of the targets from the radar, and the region being searched for targets. There are, however, some regularities. The operations of the radar are timed to a basic cycle of 0.054, or 54 milliseconds (ms). During one such cycle, several pulses may be transmitted in different directions and at different frequencies. The pattern of one cycle need not, in general, be repeated exactly in the next cycle or in any cycle occurring soon after. There are several basic kinds of cycles, called "templates." In one template, all the pulses are transmitted during the first 16 of the 54 ms. The other templates correspond to division of the 54-ms cycle into two, four, or eight.sub- cycles, scaling down the transmitting period in each subcycle to 8, 4, or 2 ms. Every eighteenth cycle of operation is dedicated to testing and calibrating the radar. There tends to be some regularity in the manner in which the templates alternate or repeat from cycle to cycle. The resulting signal, which is thus rather random, has elements of a repetitive character at periods of 54, 27, 13.5, and 6.75 ms. These repetition periods correspond to modulating tones at frequencies of about 18.5, 37, 74, and 148 Hz. In addition, sidelobe energy received at any point on the ground is subject to further modulation, because the successive pulses are typically transmitted in different directions and at different frequencies. This results in low-frequency components (2 Hz and below) in the modulation and in the spreading of energy about the concentrations at 18.5 Hz and its harmonics. RADAR CONTROL The detailed pattern of operation of PAVE PAWS radar is controlled by computers that are integral parts of the system. Signals received by the radar are processed and interpreted by a central computer. The results are displayed to operators and are made available for further processing and for transmission to users of the data. Under general conditions of operation that are selected or set by entering instructions into the central computer through an operator's keyboard, the central computer and a radar-control computer generate beam-steering orders and transmitting orders for each 54-ms cycle of -14-

transmission and reception. The beam-steering orders appear as electric signals unique for each of the 1,792 transmitter-receiver modules. Each module delays the transmitting or receiving of signals as directed by the instructions from the computer. In the transmitting mode, the transmitter-receiver module acts simply as an amplifier (with an appro- priate delay) for whatever pulse it has been directed to transmit. In this control process, there are, at least in principle, a number of ways in which the cycle of normal operation could be in error. Errors that could affect exposure estimates are the following kinds: o A well-formed beam might be steered to an angle below the 3° limit, and reflection from the ground might increase the power density. o A poorly formed beam might have sidelobes more intense than those described. o Transmission might occur at a duty cycle greater than 0.25. In the estimates previously referred to, it was noted that in areas of public access, under normal radar operating conditions, an observer could be exposed to the sidelobes of a rapidly scanning antenna, and that, on the average, the radiation intensity would be much less than that esti- mated for the peak of a secondary sidelobe. There is, therefore, a fourth possible effect: o A well-formed beam, properly steered, might remain stationary, rather than operating in a continuously scanning mode. Thus, there could be, at some points in the public area, steady illumination at an average intensity greater than that which would be received from a powerful FM station at the position of the radar. The safeguards against these malfunctions that are incorporated into the design and testing procedures of the PAVE PAWS radar are the subject of a detailed review, cited earlier, by a panel of the AE. The next section summarizes the relevant portion of that review. SAFEGUARDS The four kinds of malfunction noted above might arise from the Issu- ance of improper orders by the computers to the signal-generating and beam- steering circuits of the radar or from improper execution of valid orders by the radar. Safeguards for both computer and radar operation have been built into the system. First, internal checks in the computer programs are -15-

designed to sense malfunctions of the computer of any kind and specifi- cally to test steering orders for violations of the 3° limit, as well as other limits. Second, there are instruments that monitor, essentially independently, many features of the radar performance and many elements of the radar. Some of these instruments make periodic tests called for by the computer; others measure such quantities as temperature and voltage, independently of the computer. Displays at the operating consoles summarize the status of the system at all times and warn of minor malfunctions. Serious malfunctions, which result in the sounding of alarms, can shut down the power supply of the transmitters. The system reacts to a major malfunction within a few seconds. The more important safeguards that are parts of the computer pro- gram are discussed in more detail in the report of the AE panel. The following observations are related to the four general kinds of possible malfunctions listed above. o Beam steered below 3°: Two simultaneous failures would have to occur for a beam to be steered below 3°—one a failure in a program in the central computer, the other a failure in the operation of an entirely separate computer. There are independent guards and monitors against each kind of failure. In the opinion of the AE panel, there appears to be no practical possibility of the simultaneous failures that would result in a beam direction of less than 3° above the horizon. o Poorly formed beam (severe sidelobe); Generation of a poorly formed beam requires that a number of antenna elements radiate essenti- ally in a cooperative manner, like a separate small antenna that has been steered in an improper direction. For example, doubling the inten- sity of one of the secondary sidelobes of the normal antenna requires that about 50 individual radiating elements, properly spaced in the whole array, exhibit improper signal delays in a systematic way. In the opinion of the AE panel, the manner in which steering orders are generated and distributed to the individual transmitting modules makes such a malfunction extremely unlikely. In any case, monitors of the radar's operation would detect such a malfunction within 30 s. Because there are as many as 1,792 transmitting elements, a few random failures of individual elements to transmit properly or to execute steering (delay) orders properly would result in only a small distortion of the antenna- beam pattern. At every eighteenth radar cycle, essentially once each second, the computer interrupts the operation of the radar and puts groups ("subarrays") of 32 transmitting modules each through tests of response to steering orders. Performance is observed by a test monitor mounted on a pole approximately 100 ft (30 m) in front of the antenna -16-

face. The existence of a malfunction of any kind is displayed on a main- tenance console, and the malfunctioning subarray is shut down auto- matically. The entire cycle of antenna operations is tested every 30 s. Antenna elements that are electrically damaged—for example, by ice or snow—would be detected by these same antenna tests. Heating elements in the antenna face prevent icing and the accumulation of snow. o Excessive transmitting power: Quite apart from checks within the computer software, there are monitors of temperature and of other indicators of power-supply overload that are designed to shut the system down under overload conditions. Even if it were possible for transmission at a duty cycle of 0.30 s to be called for, the system would shut down in a matter of seconds. o Beam not scanning; When the beam is properly pointed, at more than 3° above the horizon, failure of the beam to scan would result in exposures no greater than the maximum estimated earlier. As a safeguard, when the beam is under control of the central computer, an alarm sounds if more than 16 consecutive pulses are emitted in the same direction. When the beam is under manual control, transmission is prevented by the radar-control computer if the angle of elevation is less than 6°. WEATHER EFFECTS Rain and fog can scatter electromagnetic radiation. At UHF wave- lengths, the effect of scattering is not significant. Traveling through many miles of rain or fog, the beam loses a few percent of its total energy (much of the loss by absorption, rather than by reradiation or scattering). Therefore, in the neighborhood of the radar, the energy scattered to the ground by rain, fog, or snow is a minute fraction of that already accounted for in the sidelobe calculations above. An anomalous distribution of temperature in the atmosphere can refract (bend) the radar beam. Even under worst possible atmospheric conditions, the amount of bending is so small that it does not affect the radiation environment described here. MEASURED AND CALCULATED POWER DENSITIES OF THE PAVE PAWS RADAR This discussion is limited to a review of three studies that are directly applicable to the prediction of electric fields and power densities produced by PAVE PAWS at locations outside the boundaries of Otis Air Force Base. The first gives results of measurements in August 1978 on Otis Air Force Base in front of the south face of the radar with only the south face operating. The second gives results of measurements in October 1978 outside the base boundaries with both the north and south faces of the radar operating. The third is an -17-

analytic study. Other measurements have been made, but most of these were for testing the operation of the radar, not for measuring the environmental fields. MEASUREMENT METHOD The system used by the Air Force for measuring peak electric- field strength and average power density is shown in Figure 3. Peak power density was calculated from the measured peak electric field strength. According to the Air Force, this system has been reviewed by the National Bureau of Standards (NBS).1Z1 The calibration of the system is traceable to NBS, and the overall uncertainty is determined by dividing and multiplying the measured values of field strength and power density by 1.6. At each measurement location, a dipole antenna on a tripod was raised to a height of 2 m. The antenna was moved horizontally to lo- cate a maximal value, and the received signal was measured in three orthogonal planes. A Singer NM-37/57 field-intensity meter was used to measure the peak electric-field strength, and the data were processed by a small computer and recorded on magnetic tape at a rate of 50 samples per second in August and 100 samples per second in October. Average power density was measured with a Hewlett-Packard 8484A power sensor and a 436A power meter sampled by the computer at a rate of 167 samples per second in August and 100 samples per second in October. RESULTS The peak electric-field strength and the average power density were measured at four locations in front of the south face of the radar on August 26, 1978. The south face was operated at a frequency of 435 MHz, a duty factor of 0.2, and normal power (nominal average, 146 kW). At three of the four sites, data were collected for beam eleva- tions of 3°, 6°, and 10C; only 3° and 6° data were collected at the fourth site. At each elevation, the beam scanned back and fourth. The re- sults, taken from PAVE PAWS System Program Office and arranged in order of increasing distance are given in Table 1. Although only the south face was operating, all four sites were within 3,900 ft (1,190 m) and 63° of boresight for the south face, and the measured values should be close enough to those obtained when both faces are operating to permit valid comparisons between measured and predicted values. Similar measurements were made at 21 sites in Bourne, Sandwich, Mashpee, and Falmouth, Massachusetts, on October 20 and 21, 1978. The operating conditions were as described above, except that both the north and south faces were operating with a duty factor of 0.18. Measurements were made for 3°, 6°, and 10° of radar-beam elevation, -18-

CO oo vO oo *^ oo CM No ^^ h 0 ? 4J l *H CD 9 CM 0. CN x- >-i a* oo U iH 4J 00 o ) CO 0 vO l 4J •<; rH X! 4J CM c 0) t-> o *n O r^ Z So o. o CM Z CU rH 00 -- « CO 4-l OH 00 -^ 5C 0* O> cn 35 H cn Ed ON cn H 0) o 4J ^^^ (-l •< > I H fH o o 3 0) CJ vO ** ^ « 01 Z O DC U ^cn < I/ ' f_ N m cu m l/^ oo 4J CM o CM 0) *O fli O CM < tH CX rH 2 oo 0 -? 0 r-~ o < < 00 < X 0 CM ., O -H >> m vO l ft l •H CO CO V. 0) fa COO TJ Z* ^ r^ Ui CO lO O) -* CM rj oo m O rH CM rH <H t* CO CO -* 9 M NO Ui iH CM O > OI O »-! rH O O rH 4J -H st 00 0) 35 P-i 0) PU CO J^ 0 Z C u Z Z 4-> Z C Z c •-. M cn C K S^n PL. 0i ^~- IH O cn cn / V S K cn cn 4-) Ol s ^ • c T3 o i •H a h iH 1- 0 Ol 0) U •H ^ 3 co cn o — • 0) 1 4-l 4J o b 9 O I co o- U NO « tO (C •H (D 3 CM 4J •H w cd > M c m c eu w dj I Ol O 9) O H 4J CO E oi CO <S fcj-, cd <J *j 3 rH l M 0) •< Z I-i eu o m O) rl iH "-» 4J K exo 3 •H o> cxcn co •rH -H CM D. -o H1 C 0. 0 i SOU t-i S S S cn W Q |t4 01 U5 4J >v ^ 4J ^H 0) O ^ -, *§ W CO CO ^8 1 ! cd O 1* V S x TJ CO — ' £ 60 Z M •H — cn cn U. -19-

60 « >• 4-J 4) -H -5 0 OVDCM r^oi~» -***ox \o m I U-l CM**vO *^vOCO CMmcO CMOOl o e 0 rH rH iH rH o p *H 4J M CO SI IH ;x 0) o O O w PC Pu a. B 4J v-4 O) a 9 M i~H C 0) OOCMC7* t-.CMiH OOCO*J COCMl IP M 4-I f-t 00 T-l CO -< CMCOCM rHrHrH OOO OO ^3 over Ox iH t* > U eous o a Be- »-. vO M * c CM 4i <J" O OX rH CM rH 00 f^- vO OO r^- l cd 4= •g I cd -a-ini*- oo r- <r omr-i -* rH l JU) G 4-l O £ CO rH rH rH C * i B a 13 4J 0) fch M B 4-I J 0 w -H <tj iH B a « o 0 •o 0) J^ •H •r4 •8 M oo in in in c\i vo oo ~^ rH 1 cd 0) O,. 4J *T~J 2 X 0 W vOCOvO t*»vOCM O*^l*» COOO COCMCM rHrHrH CMrH rH M cd tu iH 0) „ O VM •o 0) rH •H ^ o .^ O •H rH CO P- W 0) 2 4-l rH U oi 4J fc| 0L4 CM > 1 •H 0) b M A ^3 4J rH H ^4 0 C O 0) O ijf 0) 4-i 0* rees t U B 0) H cd i COvOO COvOO COvOO COvOO C H M 0) rH o s fit W cd O) <u iH rH rH r-t r H 0) U 1 iH N $ 0 ^ ^ i IH Q O « V 43 *J to CO *** IH >. B IH rH 01 0 0) o • 0 4J 0 CM 0 -H •a 4) O O CO § 4-) 3 -H rH 0) a B rH w 4) to •H 0) CO X 9 ^ 0) 0) 60 ^ a 1 B .£> "So M cy» co CM o O 4J CO I 00 ^ •o P- 60 •H $ C j 0) 0 "9 Ta ^3* O fl CO r-l 0 'M 9 S 0) « C I-- o •O 4-l M j> O 1^ O a Ox O -* vO CB O *£ aco B C 0 ^^ i**1* oo i*** S-t ^% 4J ca ---. 4-) O CO m M aics 1 e o -H CO O s tf to 'e o -••. •H cd 3 O1 0) 4J > =S MH 4J V > "T^ cd 0) 00 O -* O 0) B SC^T o rH 4-l in co r^ m 0) m o H IH CM CM CM CM pj rH] cQ T^ o w r3 ^ 60 9 rH to >~~l <J^ c7 O'* 4r IH 4J • O o in o PH CO •H i C c/x in *^ CJN l IIH •a 0) -^ m CTI rH g rH c 1 a 0) M O 0) cd U 4JCM 0) O f^ I-i ,G 0) •H CO S 0 CO IH 4-) f H *9 -^ g O O O O 3 B 4J o o o o cd e 4J o 0 3 cd (/j vO 00 rH Ox 4-i •H O 0) rH 3i 0) «4 w cd N o rH X o rH rH CO CO Q U 00 t J2 u T3 -20-

but only 3° data were recorded. The results, reproduced from PAVE PAWS System Program Office, are given in Table 2. Table 3 shows the same data arranged ln order of distance and includes the ratios of peak power to average power. Hankln has calculated the electric-field strength and average power density expected to be produced by the radar. Hankln1s data are repro- duced here as Table 4. Locations 4 through 22 lie outside the boundaries of Otis Air Force Base. Of these 19 sites, four correspond roughly to sites where the Air Force took measurements; the results for these four sites are compared in Table 5. As examination of the data in Tables 1-5 leads to two conclusions: the average power density produced by the radar at locations outside the boundaries of Otis Air Force Base is not likely to exceed 1 yW/cm near the base boundaries and will be even less at more distant locations, and the ratio of peak to average power density is between 37 and 320. The point nearest to the radar that lies outside the base boundaries is along U.S. Route 6. The calculated and measured values (Table 5) of the peak electric-field strength are in good agreement for this location and in reasonably good agreement with the data in Table 1 for distances of 3,100 and 3,900 ft (945 and 1,190 m). The average power densities do not agree as well, but they are not inconsistent. In computing aver- ages, Hankln assumed that the sidelobe exposure was never less than the maximal value of a secondary sidelobe, i.e., less by a factor of 1,000 than the main-beam exposure at an equivalent distance. However, this, in effect, assumes a stationary beam. If one accounts for the motion of the beam, the average should be lower by about a factor of 6.3 than the stationary average value, or lower by a factor of 6,300 than the main- beam exposure at an equivalent distance. If Hankin's calculated value of 4.7 yW/cm is reduced by a factor of 6.3, the result, 0.75 yW/cm , is about the same as the measured values at distances of 3,100 and 3,900 ft (see Table 1)—between 0.2 and 0.87 for various ratios of peak to average power. These values are still about an order of magnitude greater than the average power density measured at the Route 6 site (see Table 2).* The data in Tables 3 and 4 show that, in general, exposure decreases with distance from the site, but not uniformly; that, far from the source, measured values decrease more rapidly than calculated values, presumably because calculations do not take into account attenuation by the atmos- phere, trees, etc.; and that the ratio of peak to average power density varies widely and, from the limited data in Tables 1 and 3, appears to depend on beam elevation, azimuthal angle, and distance from the source. The following paragraphs show that the observed range of the ratio of peak to average power density is consistent with assumptions generally made. However, the data in Tables 1 and 3 are too limited to uncover any particular trends or patterns. *At approximately equivalent distances from the radar source, measurements may differ because of ground topology and other factors. -21-

0) 9 rH f-Xl rHXi o xi xi . 0 rH NO rH CN CO X> X) X) X> 0. X> h NO CM CN O m rH o o o ^ S O O O O ooooo •H <; do do ooooo CfjCM II ^ ^ u"l r~-- s£> NO ^l^ fNJ QN 00 lO 00 CM CM rH. 0 X 4-t *4 3. ^^ ^J ^^ Csj ^5 C5 co r^ ^^ oo *^ 0 0 0 0 -H CO 0) «,< ^^ ^x ^^ vjD ^^ t^^ O 0 O CM O CM rH CO 0 0 0 0 4J o cu ^^ C^ ^^ CO t^_p t^^ 0 0 O in CM O O O o o O O cu Clj {Xi CL. rH A (_. C cu •H Jj 4-l <»N a o CO CO <. At ^4 i gy ^ r^ oo NO o CN in rH ONfO. ON *^ ON ON *^ *^ r*x oo ON vO. O Xi CU •H in rH st 1^ st rH m o O st f*» OO 00 rH rH O 0 0 O- rf} c o W 4-) TJ oo co o co o o O 0 O -* CN O O rH o o 0 0 CO cr^ Jl! y rH CO T— < cd cu cu X cu 0) rH -H 4-i y A PL- W fa •H CO rH co PL. CM c • cu x; Q 4-) 4-J g CN 2 J O -* -3- ON rH ** r^ r>. oo CM NO CM r*. in NO co *a- CN CO in o M O (_. *•• •* rH c*> en f> r^ r^ OO rH st CM CM CO CM ^ rH CM rH -* ~* <r ON •O CO Xi 41 S rH rH CM rH rH rH rH rH tH •o O M B 0) c 4-l O (0 co CM •H a y p 4-l O) NO rH rH st ** NO sr CO CN oo vo o r-* oo o •* rH oo ON O 00 Q {A ._^ M l x: _ a -H -H O CM CN CM ** ** m r~ ON CO rH CN rH CN rH r-i r^ OO 00 ON rH rJ y 4-i *• < 0 S rH t—l • PQ t-i t-i ft ^•v ^J (4 O f"» CM H M e 4-l C 3 .w c _j it o f ^^ M .M 0 ai ^ 0 rH m o 0) O y *| | •H rH O •rt rH O X 4-) •9 cu CO 1 A rH •H XI ^^ ^5 <4H O • C •H 1 8 cu 0 • 3 rH Q CO PC ^3 M rl y B o o 9 C Ol (4 M (0 co C fi 4J J^j § CO 0) Q rH CO (2 rH 0) CU rH -H O C CO CO 4J to 03 TJ g co e 00 X! (4 (0 CO 3 -H 3 4J rH O -H •rj C O 4-> 3 CO CO 06 cn O rH 4J rH 33 5C O 4J O NO O O O TJ O K O CO X! rH 4J y O C C/3 -H 0) rH PQ -d cQ o C O..C CO 4-i C S o CU CO CO cu h x! c x: o cu cu y cu o cn M •H 0) rH X! g rH XJ ^^^ < 0) 0) U M O O 3 e iX CO S §. 4-l ^ to cn S rH 0) 4J h O ^ 4-i 3 CO TJ O) o 01 cu U 00 r) CO QEi o r* CO o ta OX 00 C J3 ^ 4J M Co 0 iO 60 CQ 0 y rH TJ "»H TJ TJ CO 4-) 0) rH • >s 0) CO 3 VH -H a. c x; > rH cu o. 3 M CO P* •TJ 2 S X S CO O iH TJ O rH O > rl IS co "co rH (0 c (0 -ex cu cd . x x ^ o PU S Q. O 0) § O Q oi x: 0) O O 0 rH c 0) T) rH S co x: CO 0) o o (00) C -H C "oj 01 d. CO 4-) CO p» t ^j pu cu •H .4-i 0) 0) MOC O rH O y cu 0) O rH K 4J a >> in s cd > rH O, O. 3 S i-l CU PU 4J C £ 0) co B X x: o W J3 CO CO M C X TJ rH 0) 4-t C (0 7 4—* Ua 6 ^> Cy •< W y O CO CO TJ 0) O CQ x: c h rH o CU (u CO O rH "rH CO CO CJ 5C CO 5C •H 4J CO 0) < PQ o cd (^ rl SS CO 32 Q W W u' cn x u- B o. O o cu U t-l * 4J C , . CO O 4-) rH CO -H CO OI 0) O rH CM CO *^ in NO r^ oo ON CO rH CM CO *^ in vO r** 00 ON O rH a eo H cm rH rH rH rH rH rH rH rH rH rH CM CM 0 X) -22-

TABLE 3 Ratio of Peak to Average Power Density (Based on Table 2) Test Point Approximate Distance Power Density yW/cm2 Ratio of Peak to Average Power Density Miles km Peak Average 0.061 1 0.6 1.0 19.5 320 15 1 1.6 0.345 0.003 115 16 1.4 2.3 0.008 a — 11 1.6 2.6 5.23 0.051 103 13 1.7 2.7 0.209 0.001 209 12 2.0 3.2 2.07 0.016 129 2 2.1 3.4 2.7 0.027 100 3 2.1 3.4 0.055 a — • 4 2.4 3.9 3.6 0.02 180 14 2.8 4.5 0.188 0.002 94 5 4.4 7.1 0.047 0.001 47 6 4.6 7.4 0.006 a ~ 7 5.4 8.7 0.026 a — 17 7.1 11.4 0.002 a — 8 7.3 11.7 0.002 a -- 18 8.8 14.2 0.002 a — 19 8.9 14.3 0.001 a — 20 9.0 14.5 a a -- 9 9.2 14.8 a a — 21 11.8 19.0 a a — 10 13.8 22.2 0.002 a ~ aBelow reportable value (less than 0.001 vW/cm2). -23-

I 4J •Ss tJS 0) C ' •O ^ rH cd oi OI -H OH fa S 0i I 01 •h CO IdCM C OJ 0 V P| V ° «£ S3* i 53 H I •H M 4J y 0) iH H (0 •o 0I < •O S I OH M At O M OOOiHi-lrHiHiHiHi-HCMCMi-HCMcM<J'instl-.a<COiH oooooooooomooooocoooooo C8 Cd •H C iH C T5 C ed C cd 0) iH OI rH M iH CM CM CM CO c cd a <d ti cd • l~ CM O O 60 4-) CO W O O) 00 0) £ co cd 3 Ol O M 4J 00 60 00 « « 00 (X) OI i—t TJ C M O « •H *O O H CO -H X 3 »-i y o jr. es 0I OI CM QJ -r4 -H -H -H -H U) O O « i-ioi)cco4J4JV4acgcoo.cooiBS*OM 0) O H can cd « 10 CM CO - OI ja 4-t •H O O Oi I ca C/3 0) o I 00 $ r-i ,c a. s_i y y cd o 3 3 y p p > 5-j CO 4J 4J OH y K W 03 00 CD tH 00 J3 cd 0) oi co en 4-i co C C rH -H -H l_ -r-i c5S^JSS^S in I E O 4J C B -H rHrHiHiHiHiHi-lrHrHiHCMCMCM -24-

ID • i « ^ 10 cn o o CO i-t iH o <N iH 9 9 ^ 9 B 9 o o O O y y •a 3. u m r- >> 1 • 4J "8 •H O 0O 4J in -* W 0 CO fx oo O iH • • • co Cl 9 *^ O o 0 M-! y Vl tH ^ Of 4 4J ^ CJ 9 o •o * c 00 A) •H w*t 9 i-j 9 iH O S '( • *-> h vO m CO 1 o o y c 0) O O Cti • • .^tj 3 O o O co o) Q •8 V Wi V £ M ^ 3 U-l nj 3 •o t-c O 4J •a 4J 0) CO ^•y pf") i-4 0 P x iFH • 9 • •H 9 oo o> co CN O •o y y c 3 •H "3 4J 5 M w a w S 4J •o y -»» OI 0) > Q 4J •g « w • cd iH T3 3 Jd iH i Ox 00 O m 0) y <0 0) vO -3- 00 o ji i»! 0) -H [*t y • • • cd ^ oo <t O o in m 0) ijy M CM IM CO >M 00 00 4-l O i"H C 0) 4-) JB H ^ 0» iH ^ O O cd r" CM • iH rH a oi rH CO •H •H y x ^ h x a ^B 43 vO CO O ^0 O O o O a I-t 4-l O • ^3 *o U-l s (X CO ^> m ^-f m cd 0) 0) O o CX -H ^j M A A A 4J fH f—( < 0 PL. en 00 Ox vO cd '— o. U CO •a -H •H O 4J 4-l 4J rH iH C 3 3 0) B S >« 5 ^ S <• .* CO H ^ h cu ^ b 3 0) ^> h 0) rH CO CO iH i-t f CO oo •o §4-) c Cd JT- J3 •a o) cd -H J3 0) a H H 0) 3 J_^ y B H o PC cfl 8 O 0) a S 3 0 ^3 v*O iH O y o O •o CD Cl) W •« M 0) O •H 4J O* (0 4J o o iu c/b 17 fa cd .4-l h 4J 0) 3 S -H c CO CO CO CO O H x co o u w 4-l 4_) y O (£ oo •rt 4J •H CO CO CO o d CO V) w S- 4J CA O v-' CO"' ^ Q Q y -25-

CALCULATION OF RATIO OF PEAK TO AVERAGE POWER DENSITY Consider the power density, <S>, produced at a point in space by a scanning radar. We assume that the point is illuminated each time the source is pulsed. The pulse amplitude will depend on the sidelobe structure of the antenna and the position of the beam. In Figure 4, this is depicted as a series of pulses with monotonically decreasing ampli- tudes. For simplicity, we assume that all pulses have the same width, <T , and period, T , and that the scan is periodic with period J\t where fl is the number of pulses per scan. The time-averaged power density, t, is given by: f t - IT <S>t = (1/TU) , Sdt (1) I t = o = (1/TJt) / a- Si i=1 Si], a = <T/T. (2) 1=1 Without loss of generality, we assume S^>Sj+j for all i. We further assume that there is no exposure to the main beam. Hence, let S-, be the power density associated with primary sidelobe, Sn that associated with the secondary sidelobe, etc. We then take the following two quantities as given: S2/(1/T) I S± = 6.3, i.e., 8dB (ref /^ ); (3) S/S = 10 (ref -7JT). (4) If there is no exposure to the primary sidelobe, then ST = 0, and the ratio of the peak power density, S , to the time average Is given by (Sp/<Sj.t) = (S2/<S>t) = S2/a[(l/-n) Z S±] (5) = 6.3/a = 35 for duty factor (a) of 0.18. If the first sidelobe is present, then from Equation 4: (Sp/<S>t) = S1<S>t = 10 S2/<S>t (6) = 350 (a = 0.18). Therefore, the ratio of peak to average power density lies in the range 35 £ [Sp/<S>t] £ 350, for a duty factor of 0.18. -26-

w JT time FIGURE 4. Schematic representation of pulses from a scanning radar -27-

POPULATION EXPOSED TO PAVE PAWS RADIATION The estimated number of persons living in the vicinity of the PAVE PAWS antenna is shown in Table 6. It should be understood that Cape Cod's transient summer population is greater than indicated by the numbers in the table. It is evident that there is little or no population within 1 mile (1.6 km) of the antenna site. TABLE 6 Distribution of Population around PAVE PAWS Radar Site, Otis Air Force Base, Massachusetts3 No. Census Radius Enumeration No. Housing Miles km Districts Units Population 1 1.6 0 0 0 5 8.0 2 916 1,239 10 16.1 19 8,059 11,235 20 32.2 88 28,195 63,289 aThe population data base from which this table was prepared is an edited and compressed version of the 1970 U.S. Bureau of the Census Master Enumeration District List with Coordinates. The computer program and data base were adapted from those developed primarily for marketing purposes by the U.S. Department of Commerce. ' The population data base contains the housing and population counts for each census enumeration district (CED) and the geographic coordinates of the popula- tion centroid for the district. In the Standard Metropolitan Statistical Areas, the CED is a "block group," which usually consists of a city block. In other areas, the district is called an "enumeration district," and it may cover several square miles in areas of low population density, as in the case of the PAVE PAWS site. There are approximately 250,000 CEDs in the United States, and the average population is about 800. The posi- tions of the population centroids for each CED were marked on the district maps by the individual census officials responsible for the districts and are based only on their judgment from inspection of the population distribution on the maps. The resolution of the data base as applied to the PAVE PAWS site allows confident estimation of population for distances somewhere between 5 and 10 miles (8 and 16.1 km) from the site. Population figures for shorter distances should be viewed only as estimates. bPAVE PAWS coordinates, 70°32'18" West, 41°45'11" North. -28-

ENVIRONMENTAL LEVELS OF RADIOFREQUENCY RADIATION The entire population of the United States is exposed to radiowaves, including microwaves, from a variety of communication, medical, industrial, and consumer-product sources—e.g., radio and television broadcast systems, radars, radiotelephones, citizen's band radio, microwave relay links, medical diathermy units, radiofrequency heat sealers, and microwave ovens for industrial and home use. The discussion here is limited to sources that produce readily detectable radiofrequency radiation in locations that are accessible to the general population and does not treat occupational or medical exposure. It is convenient to define two kinds of environmental radiofrequency exposure. One occurs at distances far from individual sources and is due to the superposition of the fields from many sources operating at different frequencies; in the discussion that follows, we call this the "general radiofrequency environment." In a relative sense, whether exposure in the general environment is high or low may depend on the locations and types of sources that contribute to the exposure. The other kind of exposure, ac- tually a special case of the first, occurs so close to a particular source (or sources) that the radiowave environment is dominated by the source(s) at that location; we call this the "specific-source radiofrequency environment.' The quantity most commonly used for specifying exposure to radiowaves is power density. Power density is the rate at which energy crosses a unit area; in the International System of Units, it is given in watts per square meter (W/m ). In temperate latitudes on a cloudless day. the rate at which sunlight falls on the earth's surface Is about 1,000 W/m2. For historical reasons, radiofrequency exposure is often expressed in milliwatts (mW) or microwatts (uW) per square centimeter (cm ). The interrelationships among these units are shown in Table 7. TABLE 7 Units of Radiowave Power Density 2 W/m2 mW/cm2 yW/cm 0.01 0.001 1 0.1 0.01 10 1 0.1 100 10 1 1,000 100 10 10,000 -29-

GENERAL RADIOFREQUENCY ENVIRONMENTS Broadcast Sources The urban general radiofrequency environment is dominated by radio and television broadcast transmissions.8'79'80'152'15^ On the basis of measurements made at 373 locations (about 30 in each of 12 cities) with a total 1970 population of over 38 million, it is estimated that the median continuous exposure in urban areas of the United States is 0.005 yW/cm , i.e., 50% of the population is exposed to higher power densities and 50% to lower. The results of these studies, shown in Tables 8 and 9, indicate that 95% of the population is exposed to less than 0.1 yW/cm and that less than 1% may be exposed to greater than 1 yW/cm . These estimates do not include contributions from AM radio transmission; the absorption of energy at AM broadcast frequencies (0.535-1.605 MHz) by humans is less than the absorption of energy at FM and TV frequencies (54-890 MHz) by several orders of magnitude. Nor do these estimates include such refinements as accounting for population mobility, for exposures at heights greater than 6 m (20 ft), for attenuation by typical buildings, or for periods when sources are not transmitting. The estimates are based simply on the population that resides in areas more than several hundred feet from FM and TV broadcast antennas where an unobstructed measurement 6 m above the ground would result in the indicated values. Nonbroadcast Sources For a number of reasons, the nonbroadcast sources do not appear to contribute significantly to the general radiofrequency environment, although the contribution of the higher-powered devices to specific-source environments may be large. Examples of low-power devices include micro- wave relay links, personal radios (such as radiotelephones and citizen's band radios), and the traffic radars used by law-enforcement agencies for measuring the speed of vehicles. Examples of higher-power nonbroadcast sources include satellite communication systems and radars (including military acquisition and tracking radars), civilian air-traffic control and air-route surveillance radars, and weather radars. Because all these high-power systems use highly directive antennas, they form beams with small cross sections; thus, only a small volume of the available space is irradiated at any given time. For many of the systems, the beam is high above the ground or the antenna is angled 2° to 3° above the horizon, so the possibility of exposure to the main beam is severely limited. Most radar systems rotate, and that further reduces the average expo|ure. Table 10 summarizes the measurements made in one large urban area. The highest observed average power density was 0.001 yW/cm . -30-

TABLE 8 Population Exposure in 12 U.S. Cities, 54-900 MHza ~~ Median Exposure, Percent of Population City iiW/cm Exposed at <1 yW/cm Boston 0.018 98.50 Atlanta 0.016 99.20 Miami 0.007 98.20 Philadelphia 0.007 99.87 New York 0.002 99.60 Chicago 0.002 99.60 Washington 0.009 97.20 Las Vegas 0.012 99.10 San Diego 0.010 99.85 Portland 0.020 99.70 Houston 0.011 99.99 Los Angeles 0.005 99.90 All cities 0.005 99.41 aData from Tell and Mantiply.152 -31-

TABLE 9 Cumulative Population Exposure at 54-900 MHz0 Power Density, Cumulative Percent of yW/cm Population Exposed 0.002 17 0.005 49 0.01 69 0.02 83 0.05 92 0.1 95 0.2 97.5 0.5 99 1.0 99.5 aData from Tell and Mantiply.152 h For example, 17% are exposed continuously at less than 0.002 69% at less than 0.01 yW/cm , etc. -32-

TABLE 10 Typical Urban Radar Environments in San Francisco, California* No. Radars Detected Average Power Density, yW/cm Location Mt. Diablo 8 0.000026 Palo Alto 10 0.00027 Bernal Heights 10 0.0011 'Data from Tell.146 -33-

SPECIFIC-SOURCE RADIOFREQUENCY ENVIRONMENTS Broadcast Sources The antennas used for very-high-frequency (VHF) and UHF TV broadcasting are highly directive, as shown in Figure 5. With such antenna patterns, one would expect power densities at high elevations close to the source to be considerably higher than those found at ground level. Power densities have been measured in some tall buildings that either support broadcast antennas or are within a city block or so of another tall building that supports a broadcast antenna. ' Table 11 summarizes the results of most of these measurements. The values range from less than 1 yW/cm2 to 97 yW/cm2 for a location inside a building or 230 yW/cm2 at an unshielded location on the roof of a building. Note that two cir- cumstances are required to obtain these higher values: high elevation and proximity to a high-power antenna. The upper floors of tall buildings far from broadcast antennas are not exposed to power densities that differ substantially (factors of 10) from those near the ground at equivalent distances. Some FM broadcast antennas have antenna patterns that can produce relatively high power densities at ground level near the antenna tower. In addition to the main antenna lobe (Figure 5), they produce a para- sitic or grating lobe that is coaxial with the antenna tower. Power densities near the bases of FM towers typically are 1-10 yW/cm (Figure 9 in Athey ^it al. ). However, some types of FM antennas .can produce fields of 100-350 yWcm at the tower base in areas that are accessible to trans- ient foot traffic (R. A. Tell, unpublished data). These power densities decrease rapidly with distance from the antenna tower, but densities near a few residences may range from 50 to 100 yW/cm . In a single unusual, if not unique, case, measured fields near the base of an FM antenna tower were between 1,000 and 7,000 yW/cm ; exposures in open areas--i.e., not close to conducting structures—did not exceed 2,000 yW/cm . Nonbroadcast Sources Because of their number and variety, there is less information on nonbroadcast than on broadcast specific-source environments. For pur- poses of discussion, it is convenient to distinguish between low- and high-power sources. The distinction is somewhat arbitrary. The conven- tion chosen here is the definition of a high-power source as one that can produce a main-beam power density of 100 yW/cm at a distance of 100 m from the source's antenna. This strategem distinguishes these sources from low-power sources that may produce equivalent power densi- ties very close to a source, but not elsewhere. Low-Power Sources. The three types of low-power sources of interest here are microwave relay links, low-power radar, and mobile communication equipment—radiotelephones, citizen's band radios, hand-held walkie- talkies, etc. -34-

OCGRILS roou HORIZONTAL PL»HE FIGURE 5. Vertical radiation pattern of a UHF TV transmitting antenna. -35-

TABLE 11 Radlofrequency Power Densities in Tall Buildings Near FM and TV Antennas* Location Empire State Building, New York 86th-floor observatory 102nd-floor observatory Near window Near elevator Power Density, yW/cm FM ~" TV 15.2 30.7 1.35 1.79 World Trade Center, New York 107th-floor observatory Roof observatory Pan Am Building, New York 54th floor 0.10 0.15 3.76 1.10 7.18 6.52 One Biscayne Tower, Miami 26th floor 30th floor 34th floor 38th floor Roof (shielded location) Roof 7 5 62 97 134 148 Sears Building, Chicago 50th floor Roof 32 201 34 29 Federal Building, Chicago 39th floor 5.7 0.73 Home Tower, San Diego 10th floor 17th floor Roof Roof 18 0.2 119 180 Milam Building, Houston 47th floor 35.8 31.6 aData from Tell and Hankin.150 -36-

Microwave relay links used for long-distance communication are lower- power devices with transmitter powers usually less than 5 W. The maximal power density in the near-field of the antenna is calculated to be about 700 yW/cm ; except for service personnel, this is not an accessible lo- cation. Maximal values at ground level are calculated to be less than 1 yW/cm2 (N. N. Hankin, unpublished data). The radars used for measuring the speed of vehicles have transmitter powers of about 100 mW. These devices are either hand-held or vehicle- mounted. They are continuous-wave, rather than pulse-modulated, devices and determine speed from the Doppler frequency shift of the returned signal. The maximal calculated near-field power density for typical devices ranges from 170 to 400 yW/cm at the face of the device. Power densities at 3 and 30 m are calculated to be less than 24 and 0.2 yW/cm , respectively. Two other low-power radars that are in common use are weather radars in aircraft and navigational radars used on small, pleasure boats. Under normal operating procedures, aircraft weather radars are not operated when the aircraft are on the ground. When they have been operated on the ground, measured power densities for a number of radar-aircraft combinations were less than 10 mW/cm , except on the surface of the radome housing of one system, and less than 1 mW/cm at distances greater than 3.5 m (11.5 ft). For marine radars, the computed average power density for any of the six units that were studied was less than 50 yW/cm at the antenna's turning- circle radius; ^ one of the units has an option for sector scanning, and the maximal power density was about 250 yW/cm when it was operated in this mode. Most of the information on specific-source environments produced by personal radio devices is based on systems mounted on vehicles or on hand- held walkie-talkies. Interpretation of such data is difficult, because most of the measurements are made in the near-field and the fields are not uniform over volumes comparable with the size of humans; i.e., the reported values are not equivalent to whole-body exposures. Furthermore, the absorption patterns for these complex near-fields may differ appre- ciably from those produced by far-field whole-body exposures: the absorp- tion may be higher or lower, and the sites of maximal absorption may differ from those in the case of far-field whole-body exposure. Some measured values for fields in and around vehicles equipped with radios are presented in Table 12 in units of volts per meter (V/m). The values range from a few volts per meter to 475 V/m (Bronaugh et al.; J. W. Adams, M. Kanda, J. Shafer, and Y. Wu, unpublished data; and D. L. Lambdin ). Only electric fields have been measured. To convert to power density properly, the magnetic fields would also have to be mea- sured, because the impedance in these complex fields is not, in general, 377 ohms (the free-space impedance for a plane wave). Some authors have defined an "equivalent" free-field power density by assuming the imped- ence value for free space and calculating the power density according to the equation S(W/m2) = E2(V/m)2/377(n) (7) -37-

TABLE 12 Electric Field Strength In and Around Radio-Equipped Vehicles Field Frequency, MHz Transmitter Power, W Vehicle Type Strength , V/m 27.075 27.610 5 Sedan Sedan 21-25^ 80s 40.27 40.27 110 110 Sedan Sedan 10-190 75-368b 40.27 41.31 41.31 110 100 100 Tractor-trailer Compact Pickup truck 5-475 5-106d 7-165d 162.475 110 Sedan 8-201 164.45 60 Sedan 5-52 164.45 164.45 60 60 Station wagon Van 5-64d 5-95d Reference 24 e c c c 92 92 c 92 92 92 aLegal power is 5 W, assuming 80Z efficiency (4 W); illegal power used with special authorization of the Interagency Radio Advisory Committee. Vehicle was placed on an electrically grounded plane. CJ. W. Admas, M. Kanda, J. Shafer, and Y. Vu, unpublished data. in j ^_"/_3\ _ n nn/. Si?2/^2/_2' u Calculated from the reported electric energy density (Ug). given i the original report from the equation UF(nJ/m ) • 0.0043E2(V2/m2), where E is the electric field. -38-

When used in this manner, however, "equivalent" does not necessarily mean equivalent heating power density. If one ignores this compli- cation and performs the calculation, then the range of 2-475 V/m corresponds to an "equivalent" power-density range of 1 yW/cm to 60 mW/cm2. Little information is available on the fields produced by hand- held walkie-talkies. In one study, electric fields 12 cm from a 2.5-W hand-held unit operating at 27.12 MHz were measured to be as much as 205 V/m, for an "equivalent" power density of 11 mW/cm . 31 The measured magnetic field was 0.9 A/m, which from Equation 8 is "equiva- lent" to 31 mW/cm2.131 S(W/m2) = 377( Q )H2(A/m2) (8) n Maximal fields of 200 nanojoules per cubic meter (nJ/m )—212 V/m or 11.9 mW/cm "equivalent"—have been measured for a 1.8-W hand-held unit operating at 164.45 MHz; the measured exposure rapidly diminishes by a factor of 10 within 1 or 2 in. (2.5 or 5 cm) of the maximal-exposure site. It was noted earlier that the complex near-fields were not well characterized by power density and that predictions of thermal impact could not be directly extrapolated from far-field whole-body exposures. One study that used dielectric phantom models of human heads predicted temperature increases of less than 0.1°C in the region of the eye for a 6-W, 150-MHz hand-held unit whose antenna was positioned 0.5 cm from the eye. Such techniques have been extended to higher frequencies (Balzano et^ jLL. an<* Balzano, 0. Gavay, and F. R. Steel, unpublished data. High-Power Sources. The principal nonbroadcast high-power sources are satellite communication systems and the larger radars. The satellite communication systems are continuous-wave sources and the radars are usually pulse-modulated. A study by the Electromagnetic Compatibility Analysis Center in 1972 for the U.S. Environmental Protection Agency identified 223 continuous-wave sources with effective radiated power greater than 1 MW and 375 pulsed sources with peak effective radiated power of 10 gigawatts* (GW) or greater. The main-beam power densities generated by these systems can be greater than 10 mW/cm . ' However, the environmental significance of specific systems depends on the power densities in locations that are accessible to people. High-power sources use directional antennas to achieve high effective radiated power. The probability of being irra- diated by the primary beam of one of these sources is quite small, for three reasons: with the exception of height-finder radars, the primary beam is usually 3° or more above the horizon; many of these sources are remote and are surrounded by an exclusion area that limits access to *1 GW * 109 W. -39-

the primary beam; and, some sources are mechanically or electrically equipped to limit the pointing direction of antennas or to reduce or shut off power when occupied areas are scanned. Thus, the directional radiation distribution pattern of most high-power sources reduces the probability of human exposure to high-power densities. Persons who live or work near airports or military bases are exposed to sldelobe radiation from systems with stationary or slowly moving antennas, as well as many types of radar with rapidly moving antennas. Calculated exposures are 10-100 yW/cm at distances of up to 0.5 mile (0.8 km) from some of these systems (N. N. Hankln, unpub- lished data). The motion of the antennas of acquisition radars re- duces the time-averaged power density from these systems. For high- power radars, a combination of mitigating factors, such as beam motion and antenna elevation angle, makes it unlikely that power densities will exceed 50 yW/cm at distances of over 0.5 mile, at least in loca- tions that are accessible to people (N. N. Hank in, unpublished data). The fields produced by these high-power sources have not been evaluated generically. Until further work is done, each high-power source must be evaluated individually. SUMMARY Low-power systems—e.g., microwave relay links, low-power radars, and mobile communication systems—contribute the least to environ- mental radiofrequency radiation. The general radiofrequency environ- ment to which most persons in the United States are exposed is dominated by broadcast transmitters, i.e., radio and television. Approximately 99Z of the people who live in metropol at power densities below 1 yW/cm ; litan areas are exposed to radiation the median exposure is 0.005, yW/cn Of the low-power sources, only mobile communication equipment pro- duces densities that, at least on cursory examination, appear high. These "equivalent" power densities, calculated on the basis of electric field, do not provide a sufficient basis for evaluating their impact. Other complicating factors include partial-body exposure and the inter- mittent mode of operation of these devices. More work needs to be done, but the exploratory efforts that have been made indicate that tissue temperature increases should not exceed 0.1°C under conditions of normal operation. The specific-source radiofrequency environment for most broadcast sources is well below 100 yW/cm . The only fields in excess of this occur ln the immediate vicinity of a few FM antenna towers and on the roofs of tall buildings within a block or so of FM and TV broadcast antennas. With respect to other high-power sources, such as radars and satellite communication systems, the potential for human exposure to obviously high power densities is small, owing to system character- istics that include colllmated beams and, in many cases, a combination of such other factors as remoteness, operation that precludes primary- beam exposure, and motion of the primary beam. -40-

These factors result in estimated time-averaged power densities for most systems that are below 50 yW/cm at distances greater than 0.5 mile from the source. Higher power densities may occur closer to the source, especially if exposure to the primary beam is possible. -41-