United States Office of Radiation Programs ORP/EAD 78-5
Environmental Protection Las Vegas Facility June 1978
Agency PO. Box 15027
Las Vefeas NV89114
Radiation
*xEPA Population Exposure to
VHP and UHF Broadcast
Radiation in the
United States
-------
Office of Radiation Programs Technical Publications
Nonionizing Radiation
Publications of the Office of Radiation Programs are available
from the National Technical Information Service (NTIS), Springfield,
VA 22161. Current prices should be obtained directly from NTIS
using the indicated NTIS Order number. Single copies of some of
the publications listed below may also be available without
charge from the Office of Radiation Programs (AW-461), 401 M St.,
SW Washington, DC 20460.
EPA ORP/SID 72-3
EPA/ORP 73-2
EPA-520/2-73-001
/
EPA-520/1-74-005
EPA-520/2-74-008
ORP/EAD 75-1
ORP/EAD-76-1
ORP/EAD-76-2
EPA-520/2-76-008
ORP/EAD-77-2
ORP/EAD-77-3
Reference Data for Radiofrequency Emission
Hazard Analysis (NTIS Order No. PB 220 471)
Environmental Exposure to Nonionizing
Radiation, (Available NTIS only, Order
No. PB 220 851)
Nonionizing Measurement Capabilities: State
and Federal Agencies (Available NTIS only,
Order No. PB 226 778/AS)
RF Pulse Spectral Measurements in the
Vicinity of Several ATC Radars (NTIS Order
No. PB 235 733)
An Evaluation of Satellite Communication
Systems as Sources of Environmental Micro-
wave Radiation (NTIS Order No. PB 257 138/AS)
An Analysis of Broadcast Radiation Levels
in Hawaii (NTIS Order No. PB 261 316/AS)
Radiation Characteristics of Traffic Radar
Systems (NTIS Order No. PB 257 077/AS)
A Measurement of RF Field Intensities in
the Immediate Vicinity of an FM Broadcast
Station Antenna (NTIS Order No. PB 257 698/AS)
An Examination of Electric Fields Under EHV
Overhead Power Transmission Lines (NTIS
Order No. PB 270 613/AS)
An Investigation of Broadcast Radiation
Intensities at Mt. Wilson, California
(NTIS Order No. PB 275 040/AS)
An Analysis of Radar Exposure in the San
Francisco Area (NTIS Order No. PB 273 188/AS)
-------
POPULATION EXPOSURE TO VHF AND UHF BROADCAST RADIATION IN
THE UNITED STATES
Richard A. Tell
Edwin D. Mantiply
June 1978
U.S. Environmental Protection Agency
Office of Radiation Programs
Electromagnetic Radiation Analysis Branch
P.O. Box 15027
Las Vegas, Nevada 89114
USA
-------
DISCLAIMER
This report has been reviewed by the Office of Radiation
Programs, U.S. Environmental Protection Agency, and approved for
publication. Mention of trade names or commercial products does
not constitute endorsement or recommendation for their use.
11
-------
PREFACE
The Office of Radiation Programs of the U.S. Environmental
Protection Agency carries out a National program designed to
evaluate population exposure to ionizing and nonionizing radiation,
and to promote development of controls necessary to protect the
public health and safety. This report presents the latest estimates
of population exposure to radiofrequency radiation determined by
this agency. Readers of this report are encouraged to inform the
Office of Radiation Programs of any omissions or errors. Comments
or requests for further information are also invited.
Floyd L. Galpin, Director
Environmental Analysis Division
Office of Radiation Programs
111
-------
ABSTRACT
The U.S. Environmental Protection Agency has been collecting
broadcast signal field intensity data for over two years to
estimate population exposure to this form of nonionizing radiation.
Measurement data have been obtained at 373 locations distributed
throughout 12 large cities and collectively represent approximately
11,000 measurements of VHF and UHF signal field intensities. The
VHF and UHF broadcast service is the main source of ambient
radiofrequency exposure in the United States. A computer algorithm
has been developed which uses these measurement data to estimate
the broadcast exposure at some 39,000 census enumeration districts
within the metropolitan boundaries of these 12 cities. The
results of the computations provide information on the fraction
of the population that is potentially exposed to various intensities
of radiofrequency radiation. Special emphasis has been placed on
determining the uncertainty inherent to the exposure estimation
procedure and details are provided on these techniques. A median
exposure level (that level to which half of the population is
exposed greater than) of 0.005 yW/cm2 time averaged power density
has been determined for the population of the 12 cities studied,
the cumulative population of which represents 18 percent of the
total United States population. The data also suggest that
approximately 1 percent of the population studied, or about
380,000, are potentially exposed to levels greater than 1 yW/cm*,
the suggested safety guide for the population in the USSR.
Alternative techniques of using the measurement data to estimate
population exposure are examined and future extensions of this
work are discussed.
IV
-------
TABLE OF CONTENTS
Page
ABSTRACT . iv
LIST OF FIGURES vi
LIST OF TABLES viii
BACKGROUND 1
METHOD OF MEASUREMENTS 3
APPROACH USED TO DETERMINE POPULATION EXPOSURE 5
MODELING METHOD 9
POPULATION EXPOSURE RESULTS 12
DIRECT ESTIMATION METHOD 23
CONCLUSIONS 25
FUTURE WORK 27
REFERENCES 28
-------
LIST OF FIGURES
Page
Figure 1.
Figure 2.
Figure 3.
Figure 4.
Figure 5.
Figure 6.
Figure 7.
Figure 8.
Figure 9.
Figure 10.
Figure 11.
Figure 12.
Figure 13.
Figure 14.
Measured FM radio broadcast field
intensity spectrum in Portland, Oregon
Accumulative fraction of population in
Boston exposed £ log S(yW/cm2)
Accumulative fraction of population in
Atlanta exposed <_ log S(yW/cm2)
Accumulative fraction of population in
Miami exposed <_ log S(yW/cm2)
Accumulative fraction of population in
Philadelphia exposed £ log S(yW/cm2)
Accumulative fraction of population in
New York exposed £ log S(yW/cm2)
Accumulative fraction of population in
Chicago exposed <, log S(yW/cm2)
Accumulative fraction of population in
Washington exposed <. log S(yW/cm2)
Accumulative fraction of population in
Las Vegas exposed <. log S(yW/cm2)
Accumulative fraction of population in
San Diego exposed <. log S(yW/cm2)
Accumulative fraction of population in
Portland exposed ^ log S(yW/cm2)
Accumulative fraction of population in
Houston exposed <. log S(yW/cm2)
Accumulative fraction of population in
Los Angeles exposed <. log S(yW/cm2)
Accumulative fraction of population in
12 cities exposed <. log S(yW/cm2)
15
15
16
16
17
17
18
18
19
19
20
20
21
VI
-------
Figure 15.
Figure 16.
LIST OF FIGURES (Continued)
Distribution of uncertainties in exposure
calculations
Site exposure and population exposure in
Los Angeles
Paqe
21
24
vii
-------
Table 1.
Table 2.
Table 3.
Table 4.
LIST OF TABLES
Measurement system uncertainties in
VHF and UHF broadcast bands
Summary of information relevant to
environmental RF and MW field studies
Population exposure results in 12
cities
Summary of exposure test program
results
22
22
viii
-------
BACKGROUND
The United States (US) Environmental Protection Agency (EPA)
is presently gathering information pertinent to the development
of guidance to Federal agencies within the US concerning limitations
on radiofrequency (RF) and microwave (MW) exposure of the general
population. This information consists of both detailed descriptions
of the biological effects of RF and MW energy in experimental
test animals and man, and normally encountered environmental
exposure levels throughout the country. This report provides
detailed information on the results of our environmental measure-
ments program and presents our most current estimates of population
exposure based on these measurement data. It is pertinent to
describe the general approach used by the USEPA in collecting
these data; in the first instance, numerous and widely distributed
measurement points, generally selected on the basis of population
distributions, located throughout many US high density metropolitan
areas have been used to determine ambient exposure levels of RF
and MW energy. These measurement data are then used in conjunction
with a computer automated algorithm which contains census data to
provide estimates of the fraction of the studied population
exposed to various intensities of RF and MW radiation. Via this
method, good estimates of exposure of most of the population are
obtainable. In the second instance, many field intensity measure-
ments are conducted without regard to population distributions
but rather from the viewpoint of determining the maximum or
highest intensities of exposure that are possible to be found in
the environment. The principle purpose of this report is to
provide the results of our efforts in the first instance, but to
the extent that the secondly described measurement approach
provides relevant exposure data, we will discuss these "specific
source" types of measurements.
1
-------
Previous discussions of USEPA activities in this area are
available (Janes, et al., 1977a; Janes, et al., 1977b; Athey, et
al., 1978). This report contains new and more extensive data and
results for US cities and uses an improved propagation modeling
technique for generating estimates of population exposure.
Additionally, a technique is discussed which provides insight to
the consideration of the accuracy with which exposure estimates
are obtained.
-------
METHOD OF MEASUREMENTS
Detailed discussions of the development of a specially
instrumented mobile electromagnetic radiation analysis van used
in the collection of the environmental exposure data are available
elsewhere (Tell, et al., 1976a). The instrumentation approach
involves spectrum analysis techniques coupled with on-line
computer assisted data acquisition for purposes of recording,
correcting, and processing of the acquired spectral intensity
data. A series of calibrated antenna systems appropriate to the
frequency bands of primary consideration are used to provide
signal input to the spectrum analyzer. Appropriate account is
taken for the polarization of the impinging waves in certain
bands by the use of orthogonal dipolar antenna systems. The
mini-computer system provides various features including signal
averaging whereby fluctuating signal amplitudes are processed to
obtain time-averaged values of field intensity, and the capability
to retain instantaneous peak signal intensity excursions during
the overall observation period. Extensive efforts resulted in
our ability to specify the measurement system uncertainties as
outlined in Table 1. It is noted that the mobile measurement
system has been designed to principally operate in the bands
assigned to domestic broadcasting within the US; this was done
because of the generally higher environmental levels of RF and MW
energy being the result of the broadcast service. Several
changes in the mobile measurement system are currently underway
which include a new super-broadband antenna system capable of a
flat response over the 50-900 MHz region and a spectrum analysis
system which will result in an enhanced capability for measurement
of pulsed, radar field intensities.
-------
Use of more portable instrumentation has been made in different
studies of unique exposure situations, such as the main beam
illumination of tall buildings and other locations not generally
accessible by the mobile van system. Some of this instrumentation,
the applicable studies involving its use, and discussions of
accuracy limitations have been described in previous reports
(Tell and Nelson, 1974a, 1974b; Tell and O'Brien, 1977; Tell,
1976; Tell, 1978) .
TABLE 1. MEASUREMENT SYSTEM UNCERTAINTIES IN VHP AND UHF
BROADCAST BANDS
Band Frequency Range (MHz) RMS System Error(dB)
Low VHP TV 54- 88 2.5
FM Radio 88-108 2.1
High VHP TV 174-216 2.3
UHF TV 470-806 2.0
-------
APPROACH USED TO DETERMINE POPULATION EXPOSURE
The method used for our assessment of population exposure
incorporates (a) identification of sites representative of the
population distribution in a given metropolitan area, (b) measure-
ment of the ambient field intensities existing at these representative
sites, and (c) subsequent use of a model, to estimate the exposure
that would have been measured at many other locations throughout
the city. The results of this modeling phase are then analyzed
to determine the fraction of the population potentially exposed
to different intensities of RF and MW radiation.
An important, underlying factor in our approach is the
availability of detailed census data for the entire US suitable
for machine processing. These census data, based on the 1970
census of the US, represent the number of persons residing in
specific geographical cells called Census Enumeration Districts
(CEDs) and the geographical coordinates of the centroid of each
CED. A CED is a relatively small geographic area, consisting of,
for example, a few city blocks within densely populated areas
such as cities, but may be larger in rural regions wherein the
population is more sparsely distributed. The entire US population
is contained within some 257,000 such CEDs.
We have developed a method for selecting environmental
measurement sites which are representative of the population
within a city. First, general boundaries are defined for a city
which include essentially all of the metropolitan area population
and all corresponding CEDs within these boundaries are then
selected for subsequent processing from the overall census data
base. In effect, each of these CEDs is assigned a weighting
factor, according to the population within each CED. We then use
5
-------
a random process to select any desired number of these CEDs to
use as measurement sites. Thus, we use a technique which incor-
porates an equal likelihood of choosing any particular CED,
except that those CEDs having a greater population are weighted
in such a way as to increase their chance of being selected as a
measurement site. Out of this process, we obtain those sites
which are deemed to be most representative of the total city
population. Field measurements are then accomplished at each of
the selected sites, usually between 30 and 40, from which subsequent
propagation models are generated. In addition to these sites,
selected irrespective of RF and MW source locations, a few
measurement sites are also included very near to selected trans-
mitters to ensure a comprehensive approach to defining the full
range of environmental levels.
Field measurements are then performed at each selected site
using the aforementioned mobile measurement van. This field
activity is normally accomplished during an intensive two-week
period of time. The actual measurement process is performed by
situating the measurement van at a specific stationary location.
No attempt is routinely made to evaluate standing wave phenomena
in the vicinity of each measurement site and thus seek out either
maximum or minimum field intensities which are characteristically
present in such measurements. The extent to which such immediate
location variability affects the resulting measurements is
reflected in the scatter of the final data and is inherent in the
variance with which we subsequently predict field intensities via
a model.
The results presented in this report are the product of
USEPA field measurements conducted in 12 US cities which include
in the order that they were studied Boston, Atlanta, Miami,
Philadelphia, New York, Chicago, Washington, Las Vegas, San
Diego, Portland, Houston, and Los Angeles. The total population
studied in these 12 cities is 38,144,845 and includes 38,548
-------
CEDs yielding a mean population per CED of 990 persons. From
these field studies, approximately 11,000 individual signal field
intensities were determined from a total of 373 measurement
sites. Figure 1 illustrates the type of field intensity data
collected; in this case the spectral data show one of the measure-
ments of the FM broadcast band obtained in Portland. Here each
spectral peak observed is a single FM radio station signal. In
this particular case the measurement site was very near to a
multiple broadcast transmission center and the measured power
density was 14 yW/cm2. Table 2 summarizes the relevant informa-
tion pertaining to each city investigated.
FM AVERAGE FIELD STRENGTH
... TOTAL rOUER OEMS!TV (FOft IMTCMftTIOM
! T MMC 0* M 0»> 14. 212« UM/CN/CN
7»
»» IN
MCQUKNCV <«WZ>
25 MANS
SM OHTM POJMTS
1189 MOWS
4 HIM IT
317 MVS 1»7(
M - ! KMZ
fITC CO»E- «
OPMftTOft COW- 7
1M
Figure 1. Measured FM radio broadcast field intensity
spectrum in Portland, Oregon
-------
TABLE 2. SUMMARY OF INFORMATION RELEVANT TO ENVIRONMENTAL RF AND MW FIELD STUDIES
oo
Boston
Atlanta
Miami
Philadelphia
New York
Chicago
Washington
Las Vegas
San Diego
Portland
Houston
Los Angeles
#CEDs
2003
1249
1897
3606
11470
4646
2291
356
1113
1194
1127
7596
Population
1953665
1221431
1661012
3407059
12269374
4743905
2516917
264501
1071887
818040
1265933
6951121
Number of Stations
# of Field Low High
Strength Values FM VHF VHF UHF
# of
Total Sites
252
396
448
941
1426
1378
1107
632
956
816
810
1801
14
11
13
17
23
20
17
6
17
12
14
29
3
2
3
2
3
2
2
2
1
3
1
3
1
2
2
2
4
3
2
3
2
3
3
4
3
3
2
3
3
3
3
0
2
0
2
7
21
18
20
24
33
28
24
11
22
17
20
43
9
16
16
31
36
39
37
42
38
38
33
38
TOTAL
38548
38144845
10963
193
26 31
31
281
373
-------
MODELING METHOD
Athey et al., (1978) described a method whereby the actual
measurement data were used to modify a presumptive propagation
model for calculation at all CED sites throughout a city. Athey"s
report made use of a propagation model form which was obtained by
analyzing measured field intensity data obtained in Miami which
suggested a classically recognized decrease in electric field
intensity with increases in distance between FM broadcast stations
and measurement sites. This form for the model was then applied
to data obtained in all VHP and UHF broadcast bands to determine
exposure. In the present case, we have developed an enhanced
method for predicting exposure at the various CEDs by taking into
consideration the fact that each city and individual stations
possess their own distinctive propagation characteristics.
The method we have used includes the following features. For
each station under consideration, the field intensity obtained
for the station at each measurement site is used to obtain a
linear, least squares fit of the data. This provides a functional
form describing the way by which the electric field strength
varies as a function of distance from the station. Since this
model is generated from actual measurement data for each station,
note that no specification of transmitter power or antenna height
is necessary. If, by chance, because of poor data, i.e., high
variability in measured values of field strength, the resulting
computed slope of the least squares fit is positive, the slope is
changed arbitrarily to be equal to zero. This in general is not
a common problem, occurring in only 12 instances for the entire
set of measurements reported. Next, the straight line model is
used to calculate the field intensity which would be expected at
each CED within the cities' bounds. From extensive tests we
9
-------
determined that maximum accuracy was usually obtained in the
modeling procedure by using the predetermined slope of the line
model but shifting this line model vertically to form a least
squares fit with the measurement data obtained in the neighbor-
hood of the calculational point Ca CED location). We observed
that this shifting process was effective in reducing the uncer-
tainty whenever the particular station was closer than 5 km to
the CED. Thus we incorporated this feature of appropriately
shifting the line model to best fit the measurement data obtained
at the two nearest measurement sites. Tests revealed a non-
significant reduction in uncertainty by shifting the model to
best fit more than the two nearest sites. The effect of this
process is to lend weight to the local measurement data in
improving estimates primarily of high intensity exposures. It
was found that the shifting technique produced little, if any,
apparent improvement in other than the higher exposure levels.
If it occured in the calculational process that a CED was identi-
fied as being closer than 100 meters from a nearby station, then
the actual distance was arbitrarily changed to correspond to
100 m. This was accomplished to protect against the erroneous
computation of very high exposure levels when the CED - station
distance was very short.
An important feature in the development of our work was the
construction of a test program which could be used to estimate
the uncertainty associated with the modeling method. In lieu of
performing additional measurements to examine the accuracy of the
method, we elected to make use of the metropolitan area measure-
ments themselves in a special way. The process consists of
starting at one specific measurement site where data has been
obtained and then creating the least squares line model for each
station based on the measurements obtained at all other measure-
ment sites, but not including the site under test. The exact
calculational process described above is then used, always
rejecting any data obtained at the test site, to arrive at the
10
-------
estimated field strength for each station. Then, a direct
comparison is made between the predicted field and the field
strength actually measured at the site. This is accomplished for
each station involved and in addition to individual signal field
strength differences, a comparison is made between the predicted
total power density of exposure and that actually measured and
being the result of exposure from all signals present at the
site. The process is then repeated at each other measurement
site to obtain an indication of the goodness of the modeling
procedure. Once the process has been completed for all measure-
ment sites in a city, the results are assessed statistically by
determining the mean deviation between actual and predicted field
strengths and the mean deviation between actual and predicted
total power densities of all signals. These results are then
used as an indicator of the quality of the more comprehensive
calculations performed at all CEDs within a city. Undoubtedly,
the variances of the deviations apparent in this process are
partly due to the immediate location variability discussed
previously. Longely (1976) has discussed this subject in detail.
Repeated application of the test program, using different
criteria for shifting, provided the insight by which the final
modeling criteria were determined. Extensive computer time was
spent before arriving at the optimum criteria.
11
-------
POPULATION EXPOSURE RESULTS
The aforementioned modeling method was applied to the
measurement data obtained in each of the 12 cities. Exposure
levels were computed at each CED location and the resulting
exposure was assumed to apply to all of the population associated
with each CED. After calculation of the exposures the number of
persons associated with various ranges of intensities were
determined; in particular, approximately one-third decadic power
density ranges were used to classify exposure, i.e., 0.001,
0.002, 0.005, 0.010, 0.020, 0.050, 0.100 yW/cm2, etc. The final
results of the analysis are presented in terms of the accumulative
fraction of the population which are potentially exposed equal to
or less than these different one-third decadic power density
intervals. Results for each of the cities under study are presented
in Figures 2-13 wherein the exposure level is plotted logarith
mically and the population fraction follows a near normal distri-
bution. Figure 14 provides the results for all cities taken
together.
Each figure provides the population exposure determined for
each band separately and for all measured bands together. The
results suggest that the exposure levels are approximately
normally distributed and reveal the interesting finding that of
the exposure contributed by the various VHF and UHF broadcast
bands, the FM radio broadcast band is clearly discernable as
being most responsible for overall exposure, particularly at the
highest exposure levels. This finding supports the earlier
proposition offered by Tell and Janes (19751 implicating FM radio
broadcast transmissions as generally dominant in creating the
highest ground levels of RF fields. Despite the lower effective
radiated powers authorized for FM broadcasting compared to other
VHF and UHF television emissions, a combination of relatively low
12
-------
tower heights and broad vertical antenna radiation patterns for
FM transmission conspire to produce these relatively high fields.
It is also interesting to note the relatively low contribution
provided by the UHF TV band in as much that UHF television
stations in the US carry the maximum power authorizations.
In our experience we have found it informative to discuss
these results using two different indices. The first is the
median exposure level, i.e., that power density at which 50
percent of the population are exposed less than and 50 percent
are exposed greater than. The second is the measure of the
fraction of the population potentially exposed above 1 yW/cm2.
The data for total band exposure presented in Figures 2-14 have
been summarized from the point of view of these two indices in
Table 3. The most significant results are for the accumulative
population of all the cities in which a median exposure of
0.005 yW/cm2 was determined while something less than 1 percent ,
of the population are apparently exposed at intensities greater
than 1 yW/cm2. It is worthy to reemphasize that these data apply
only to the domestic broadcast service in the US and cannot
account for population mobility. Though the population data base
itself is dated, we feel that the results are probably representa-
tive for the actual present distribution of population.
The results of the test program designed to estimate the
uncertainty associated with exposure calculations are presented
in summary form for the 12 cities in Table 4. The tabulated data
refer to the average of all individual field strength deviations
and power density deviations at all measurement sites within each
city. The observed high deviation in power density calculations
in Boston undoubtedly reflects the few measurement sites used in
that study.
13
-------
In order to assess the uncertainty in our overall estimates
of population exposure for all cities studied to date, Figure 15
was prepared which provides the frequency of occurrence of
deviations between measured and calculated values of exposure at
all 373 sites visited. Figure 15 shows that the distribution of
these uncertainties is approximately chi-squared in nature
suggesting that the population of power densities from which
these determinations were obtained is normally distributed, this
being in consort with the general appearance of Figure 14. The
most significant point of Figure 15 is that the most likely
uncertainty appears to be about 3dB while 70 percent of all our
exposure calculations are within 8dB.
14
-------
flCCUMULflTIVE FRACTION OF POPULBTION
EXPOSED flS fl FUNCTION OF POMER OCNSITV
99
93
. 9-
. 8--
7-
3 -
2-
.1--
91
BOSTON
h-
-3
~^Z 37
TOTAL
FH WHO
LOM VHP TV
o HIQN VMF TV
* OMf TV
-f-
f
-5 -4
LOQRR1THM OF POMER DENSITV IN UM/CH'
Figure 2. Accumulative fraction of population in
Boston exposed <. log S(yW/cm2)
ACCUNULBTIVE FRACTION OF POPULATION
EXPOSED flS R FUNCTION OF POWER DENSITV
1 * ".
99^- ? »
99
.9
. 6
. 7-
:|:
. 4-
. J-
. a-
. i-
89
81
ATLANTA * .
*
*
1 . ' -
*
r ° . *
» »
.. "
.
*
" ' A
. Torm.
LOU VHP TV
o HIOH VHP TV
* UMF TV
1 1 1 1 , 1 1
-3 -4 -2-2-1 8 1 2
LOGARITHM OF POWER DENSITV IN UH/CM2
Figure 3. Accumulative fraction of population in
Altanta exposed <. log S(yW/cm2)
15
-------
ACCUMULATIVE FRACTION OF POPULATION
EXPOSED- AS A FUNCTION OF POWER DENS I TV
99
ei
MlftlK
mm.
- m «noio
LOU VMF TV
° HIOH VMF TV
* UHF TV
-5 -4 -3 -2
-1
0
LOGARITHM OF POWER DENSITV IN UWVCH*
Figure 4. Accumulative fraction of population in
Miami exposed <. log S(yW/cm )
ACCUMULATIVE FRACTION OF POPULATION
EXPOSED AS A FUNCTION OF POUER DENS1TV
99
95
.9-
. «--
7-
3-
.2-
. 1-
85
01
*
TOTW.
* rn RADIO
LOW VH» TV
MIOH VHP TV
* IMf TV
1 1 1 1 \ 1
-3 -4 -3-2-1 8 1 2
LOGARITHM OF POWER DENSITY IN UW/CM2
Figure 5. Accumulative fraction of population in
Philadelphia exposed ^ log S(yW/cm2)
16
-------
ACCUMULATIVE FRACTION OF POPULATION
EXPOSED AS fl FUNCTION OF POWER DENSITV
.99
. 93
. 9
. 9-
I
. 3
. 2-
. 1-
95
NEW VO«K
TOTBL
rn «noiO
LOU VHP TV
o NIQM VHP TV
* UHF TV
-3 -4-3-2-1 8 12
LOGARITHM OF POWER DENSI TV IN UH/CM2
Figure 6. Accumulative fraction of population in
New York exposed £ log S(yW/cm2)
ftCCUMULRTIVE FRftCTIOH OF POPULATION
EXPOSED AS A FUNCTION OF POWER DENSITV
99
95
o
o »
74-
es
torn
FH WPIO
LOU VHT TV
° HtOH VMF TV
UMf TV
-3 -4 -3-2-1 0 i 2
LOGARITHM OF POWER DENSITV IN UW/CH2
Figure 7. Accumulative fraction of population in
Chicago exposed £ log S (yW/cm2)
17
-------
ACCUMULATIVE FRACTION OF POPULATION
EXPOSED AS A FUNCTION OF POWER DENSITY
99
95
. 9-
7--
3
. 2-
. 1-
05
01
MftfHlffOTOM
8
TOTAL
FH RADIO
LOW VHF TV
o HIQH VHF TV
» u«r TV
-5 -4 -3 -2 -1
0
LOQARXTHH OF POWER DENSITY IN UM/CIT
Figure 8. Accumulative fraction of population in
Washington exposed ^. log S(yW/cm2)
ACCUHULflTIVE FRACTION OF POPULATION
EXPOSED AS A FUNCTION OF POWER DENSITY
99
95
. 9--
.8
7-
3
2-
. !
85
91
LHS
TOTHL
LQN VHF TV
VMf TV
TV
-f-
-3 -4-3-2-1 0 1 2
LOQARITHH OF POWER DENSITY IN UW/CK2
Figure 9. Accumulative fraction of population in
Las Vegas exposed <. log S(yW/cm2)
18
-------
ACCUMULATIVE FRACTION OF POPULATION
EXPOSED AS A FUNCTION OF POWER DENSITV
99
95
. 9
.8--
7--
3
. 2--
. 1
8s
81
sun orcoo
F» RADIO
LOW VMf TV
o Ml OH VHF TV
» WKF TV
-3 -4 -3-2-1 8 1 2
LOGARITHM OF POWER OENSITV IN UW/CH2
Figure 10. Accumulative fraction of population in
San Diego exposed £. log S(yW/cm2)
ACCUNULflTIVE FRACTION OF POPULATION
EXPOSED AS A FUNCTION OF POWER DENSITV
. 99
.95
. 9
.8--
. 7--
. 3-
.2-
. 1-
85
81
PMTLANt
TOTW.
' rn MOID
LOU VHT TV
o MIO« VMf TV
UHF TV
4-
-3 -4 -3-2-1 8 1 2
LOGARITHM OF POWER DENSITY IN UU/CH*
Figure 11. Accumulative fraction of population in
Portland exposed ± log S(yW/cm2)
19
-------
ACCUMULATIVE FRACTION OF POPULATION
EXPOSED AS A FUNCTION OF POWER DENSITV
99
93
. 9 +
. 8
it
. 4-
.3-
.2-
.!-
09
81
HOUSTON
TOTflt
* PM MO 10
LOW VMF TV
° MIOH VMF TV
UMT TV
-5 -4 -2-2-1 6 1 2
LOGARITHM OF POWER DENS!TV IN IW/CM2
Figure 12. Accumulative fraction of population in
Houston exposed <. log S(yW/cm2)
ACCUMULATIVE FRACTION OF POPULATION
EXPOSED AS ft FUNCTION OF POWER DENSITY
99
95
n
. 3-
.2--
. 1
es
ei
LOS
* *
TOT«L
FH
LOH V«F TV
oHIOH VMP TV
*UHP TV
-f-
-3 -4 -3-2-1 8 1 2
LOGARITHM OF POWER DENSITY IN UU/CH*
Figure 13.
Accumulative fraction of population in
Los Angeles exposed <. log S(yW/cm2)
20
-------
flCCUIIULRTIVE FRRCTION OF POPULflTION
EXPOSED flS R FUNCTION OF POWER DEMSITV
. 99
,95
. 9
. 8
. 7
. 4-
. 3
.2-
. i
05
01
MSTOM
N1ANI
fHILADCLPHtn
NCM VORK
CMICAOO
WASHINGTON
LAS VC0AS
SAN OICOO
.PMUAN9
HOUSTON
LOS flNKUS
S
« *
o
A
TOTAL
* FH flADIO
LOU VHF TV
oHIOH VHT TV
*UM» TV
~5 ~4 ~i ~2 ""d. 0 i 2
LOQARITHH OF POWER OEHSITV IN UM/CH*
Figure 14. Accumulative fraction of population in
12 cities exposed £ log S(yW/cm2)
so..
40..
(A
111
30
cc
UJ
m
20..
10..
DISTRIBUTION OF UNCERTAINTIES IN EXPOSURE CALCULATIONS
373 SITES
POWER DENSITY ERROR FOR POINTS NOT PLOTTED
33.2 33.7 34.4 38.5 42.7 49.1 66.5
.
HIIIIIIIh
HIh
Hh^HA
01 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
UNCERTAINTY (dB)
Figure 15.
Distribution of uncertainties in exposure
calculations
21
-------
TABLE 3. POPULATION EXPOSURE RESULTS IN 12 CITIES
Boston
Atlanta
Miami
Philadelphia
New York
Chicago
Washington
Las Vegas
San Diego
Portland
Houston
Los Angeles
All cities
Median Exposure (yW/cm2)
0.018
0.016
0.0070
0.0070
0.0022
0.0020
0.009
0.012
0.010
0.020
0.011
0.0048
0.0053
Percent of Population
Exposed <1 juW/cm2
98.50
99.20
98.20
99.87
99.60
99.60
97.20
99.10
99.85
99.70
99.99
99.90
99.99
TABLE 4. SUMMARY OF EXPOSURE TEST PROGRAM RESULTS
No. Mean Field Mean Power Density
Sites Error (dB) Error (dB)
Boston
Atlanta
Miami
Philadelphia
New York
Chicago
Washington
Las Vegas
San Diego
Portland
Houston
Los Angeles
9
16
16
31
36
39
37
42
38
38
33
38
11.9
5.8
6.5
7.3
7.2
6.9
6.1
7.2
8.4
9.7
7.3
5.8
16.8
4.4
7.6
6.9
6.2
7.6
5.5
5.2
10.5
5.2
5.6
6.6
22
-------
DIRECT ESTIMATION METHOD
Our choice of the population weighted random method for
selection of CEDs as measurement sites was prompted by a desire
to establish a consistent approach from city to city. In the
beginning phases of the metropolitan area studies, measurement
sites were not chosen on this basis but were decided upon by
common sense and the apparent distribution of population as
infered from city maps. An interesting observation from applica-
tion of the computer selection method, however, is that if
measurements are conducted at locations which are truly random in
the population space, then a simple inspection of the measurement
data according to sites should provide a direct assessment of
population exposure in the general area. To illustrate this
process, measurement sites corresponding to CEDs (most do) are
sorted according to increasing power density and the accumulative
fraction of sites are plotted against the logarithm of power
densities on probability paper. Figure 16 provides an example of
this method applied to data obtained in Los Angeles. From the
data, which is seen to be almost perfectly log-normally distributed,
one obtains a median exposure value of about 0.006 yW/cm2 which
compares favorably with the most comprehensive method which
necessitates many calculations at all CEDs in the area. Note
that this method, after the initial site selection is completed,
requires no further information on population. We have observed
a generally good agreement between the two approaches in deter-
mining population exposure, particularily near the median exposure
values, and often utilize the direct method, in favor of its
simplicity, to obtain preliminary estimates of results.
23
-------
SITE EXPOSURE AND POPULATION EXPOSURE IN LOS ANGELES
99.9.
99.
98.
DC
0
« W QS
UJO 9&
to
o
UJ m
sl80-
is
UJ Ul
UJ t
UJ Z
-2
_J J
ACCUMI
POPL
W 0
2-
1
0.5-
H
*
> H *
J
* .
^^
/ CALCULATED
POPULATION
I EXPOSURE
. MEASURED CED SITE
EXPOSURE
> *
_
1 1 I 1 I 1
99.9
.99
.98
.95
.90
.80
50
.10
.5
2
1
0.5
-4 -3 -2-10 1
LOGARITHM OF POWER DENSITY IN uW/cm2
Figure 16.
Site exposure and population exposure
in Los Angeles
24
-------
CONCLUSIONS
Results of the methods outlined here suggests that, of the
population group studied, representing 18 percent of the total US
population, a median exposure value of about 0.005 yW/cm2 time
averaged power density exists and perhaps, more interestingly,
less than 1 percent of the population are potentially exposed at
levels above 1 yW/cm2. It is observed that the FM radio broadcast
service is responsible for most of the continuous illumination of
the general population. Indeed, that fraction of the population
exposed beyond 1 yW/cm2 needs more careful definition and the
absolute maximum intensities observed demand precise determina-
tion, but it is interesting to note from our results that, even
at this time, at least 99 percent of the population studied are
not exposed to levels above the suggested level of safety estab-
lished in the USSR of 1 yW/cm2 (Gordon, 1974). Additional data
obtained by the USEPA, in special areas wherein mainbeam illumina-
tion of tall buildings occur nearby various high power broadcast
installations, has shown that it is difficult to find areas where
intensities exceed 100 yW/cm2 (Tell, 1978).
These data must be viewed from the standpoint of long term
exposure and certainly, it is true that, on occasion, localized
exposures may greatly exceed 1 yW/cm2. The authors recognize the
case of limited time exposure of some individuals to microwave
oven leakage, portable or mobile communication equipments, and
various other sources of RF and MW exposure including pulsed
sources, however, we feel that at this time, there do not exist
adequate quantitative techniques for evaluating these more
extreme exposure regimes in terms of their impact on our popula-
tion exposure estimates provided in this report. It is our
observation that these higher intensity situations must be
25
-------
addressed on the basis of the length of time spent in the field
and will require an accentuated emphasis upon field measurements
conducted from the viewpoint of determining absolute maximum
exposure values that may be encountered such as inside building
measurements.
26
-------
FUTURE WORK
The evidence provided by the rather extensive environmental
measurements program conducted by the USEPA within the US seems
to overwhelmingly support the contention that most of the general
population is not chronically exposed to high intensity
(i.e., >100 yW/cm2) RF and MW radiation. Accordingly, future
field measurement efforts will include to a greater extent
examination of those unique kinds of exposure circumstances
wherein relatively high intensity exposures are possible or
expected. A more detailed investigation of environmental levels
of pulsed RF and MW fields is currently being developed. Addition-
ally, we are examining our data from the viewpoint of developing
deterministic propagation models, provided transmitter effective .
radiated power and antenna height, for different classes of
transmitting stations. Our particular interest is in being able
to more accurately model close-in exposure conditions, and in
this connection we will be comparing our data and resulting
propagation models with other existing models (Kalagian, 1976).
27
-------
REFERENCES
Athey, T.W., R.A. Tell, N.N. Hankin, D.L. Lambdin, E.D. Mantiply,
and D.E. Janes (1978): "Nonionizing Radiation Levels and Popula-
tion Exposure in Urban Areas of the Eastern United States."
Environmental Protection Agency Technical Report ORP/EAD-77-008,
May.
Gordan, Z.V. (ed.): Biological Effects of Radiofrequency
Electromagnetic Fields, translated from Moscow 0 Biologicheskom
Deystrun Elektromagnitnykh Poley Radiochastot in Russian No. 4,
1973. Available Through NTIS as JPRS Document 63321, October 30,
1974.
Janes, D.E., R.A. Tell, T.W. Athey, and N.N. Hankin (1977a):
"Radio-frequency Radiation Levels in Urban Areas," Special
Supplement in biology to Radio Science, editors A.W. Guy and D.R.
Justesen, SS-1 (in press) 1977.
Janes, D.E., R.A. Tell, T.W. Athey, and N.N. Hankin (1977b):
"Nonionizing Radiation Exposure in Urban Areas of the United
States," Communication 304, accepted for presentation in Session
No. S.07, Nonionizing Radiation, IVth International Congress of
the International Radiation Protection Association, 1977.
Kalagian, G.S. (1976): "Field Strength Calculation for TV and FM
Broadcasting (Computer Program TVFMFS)." Federal Communications
Commission Technical Report FCC/OCE RS 76-01, Washington, DC,
January.
Longley, A.G. (1976): Location Variability of Transmission Loss-
Land Mobile and Broadcast Systems." Department of Commerce,
Office of Telecommunications Report OT 76-87, May.
Tell, R.A. (1976): "A Measurement of RF Field Intensities in the
Immediate Vicinity of an FM Broadcast Station Antenna." Technical
Note, ORP/EAD-76-2, U.S. Environmental Protection Agency, Silver
Spring, MD, January (NTIS Order No. PB 257 698/AS)*.
Tell, R.A. (1978): "Measurements of Radiofrequency Field Intensities
in Buildings with Close Proximity to Broadcast Stations," Environ-
mental Protection Agency Technical Note, April.
Tell, R.A., and D.E. Janes (1975): "Broadcast Radiation: A
Second Look." In Biological Effects of Electromagnetic Waves,
ed. by C.C. Johnson and M.L.Shore,TSelected papers of the USNC-
URSI 1975 annual meeting, Boulder, CO, October (.2 Volumes) ,
USDHEW Publication (FDA) 77-8011.
28
-------
Tell, R.A. and J.C. Nelson (1974a): "RF Pulse Spectral Measure-
ments in the Vicinity of Several Air Traffic Control Radars."
EPA Technical Report EPA-520/1-74-005, 45 pages, May.
Tell, R.A., and J.C. Nelson (1974b): "Microwave Hazard Measure-
ments Near Various Aircraft Radars." Radiation Data and Reports,
Vol. 15, No. 4, pp. 161-179, April.
Tell, R.A., N.N. Hankin, J.C. Nelson, T.W. Athey, and D.E. Janes
(1976a): "An Automated Measurement System for Determining Environ-
mental Radiofrequency Field Intensities II." In Proceedings of
NBS symposium on Measurements for the Safe Use of Radiation.
March 1976, NBS publication NBS SP456, (ed. S.P. Fivozinsky), pp.
203-213. Also presented at 1974 Meeting at USNC/URSI, October 14-
17, 1974, Boulder, CO.
29
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
2.
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
POPULATION EXPOSURE TO VHF AND UHF BROADCAST RADIA-
TION IN THE UNITED STATES
5. REPORT DATE
June 1978
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Richard A. Tell and Edwin D. Mantiply
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING PRGANIZATION NAME AND ADDRESS
U.S. Environmental Protection Agency
Office of Radiation Programs
Electromagnetic Radiation Analysis Branch
P.O. Box 15027
Las Vegas, Nevada 89114
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
13. TYPE OF REPORT AND PERIOD COVERED
Technical Note
Same as above
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
The U.S. Environmental Protection Agency has been collecting broadcast signal
kCJFi'L-tmtjj-Ly data £UL uvtji. twu yecLLtj tu est-uikitfc; population exposure to tills £01.111
16.
of nonionizing radiation. Measurement data have been obtained at 373 locations
distributed throughout 12 large cities and collectively represent approximately
11,000 measurements of VHF and UHF signal field intensities. The VHF and UHF
broadcast service is the main source of ambient radiofrequency exposure in the
United States. A computer algorithm has been developed which uses these measurement
data to estimate the broadcast exposure at some 39,000 census enumeration districts
within the metropolitan boundaries of these 12 cities. The results of the computa-
tions provide information on the fraction of the population that is potentially
exposed to various intensities of radiofrequency radiation. Special emphasis has
been placed on determining the uncertainty inherent to the exposure estimation
procedure and details are provided on these techniques. . A median exposure level
(that level to which half of the population is exposed greater than) of 0.005 yW/on2
time averaged power density has been determined for the population of the 12 cities
studied, the cumulative population of which represents 18 percent of the total
United States population. The data also suggest that approximately 1 percent of the
population studied, or about 380,000, are potentially exposed to levels greater than
1 yW/on , the suggested safety guide for the population in the USSR. Alternative
techniques of using the measurement data to estimate population exposure are
examined and future extensions of this work are discussed.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
18. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (Rev. 4-77) PREVIOUS EDITION is OBSOLETE
------- |