Skip directly to search Skip directly to A to Z list Skip directly to site content




Community health concerns were obtained from local citizens during a November 23, 1993 meeting at the Metz Community Center. These concerns are summarized below.

  1. Are rates of cancer, respiratory illness, or other health problems higher in the vicinity of the site?

  2. Are PCBs released from the site causing adverse health effects?

  3. Are PCBs in the water?

  4. Do the electromagnetic fields measured in the neighborhoods and on the playground by the plant represent a health hazard?

  5. Is it safe for children to play on the playground and on the basketball courts near the plant?

  6. Does the noise produced by the plant have a negative affect on the community?

  7. Can the respirable particulates produced by the plant result in adverse health effects?

  8. Are potential fires at the plant a health hazard?

  9. People still fish in Town Lake regularly. Is it dangerous to eat fish from Town Lake?


Contaminants of Concern

This section contains site-specific information about potential health hazards associated with the site; however, inclusion in this section does not imply that a particular hazard represents a threat to public health. Based on citizen input and known site activities, we have identified six main areas of concern: 1) air pollution; 2) noise; 3) electric and magnetic fields; 4) PCB contamination in Town Lake; 5) groundwater contamination; and 6) physical and other hazards. Subsequent sections of this report will evaluate these potential health hazards to determine if they have public health significance.

Air Pollution

Under normal operating conditions, the Holly Street Power Plant burns natural gas. Under some conditions, such as when natural gas supplies become scarce or when special burns are required, the plant burns fuel oil. Table 1 lists the amount of fuel used by the plant over a 5-year period (1989-1993) and estimates of the number of hours that fuel oil was burned. We estimate that between 1989 and 1993, fuel oil was used less than 1% of the plant's total operating time.

Table 1.

Holly Street Power Plant Fuel Oil Use 1989-1993
Year Month Gallons of
Fuel Oil Used
Estimated Number of Hours in Operation Estimated Percent of
Yearly Operation
1989 December
60 hours
34 hours
1 hour
1990-1991 Not Applicable 0 0 hours 0.0
1992 August
63 hours
12 hours
60 hours
Not Applicable
0 0 hours 0.0
Based on design fuel oil firing rates at maximum continuous steam flow of 7,140, 7,140, 12,600, and 13,650 gallons per hour for Units 1, 2, 3, and 4 respectively.
Assumed operation at 50% of maximum rate.

It is well known that the burning of fossil fuels can result in the generation of air pollutants such as particulate matter (PM10), total suspended particles (TSP), sulfur dioxide (SO2), carbon monoxide (CO), and nitrogen oxides (NOx). Although air monitoring data were not available for the area immediately around the Holly Street Power Plant, the TNRCC used an air dispersion model to estimate emissions from the plant during natural gas curtailment conditions. The pollutants that they modeled included benzene, formaldehyde, CO, PM10, TSP, SO2, NO2, and polycyclic aromatic hydrocarbons (PAHs). The model showed that predicted maximum short-term and average annual impacts for all contaminants, except SO2 would be below corresponding TNRCC Regulations, National Ambient Air Quality Standards (NAAQS), and TNRCC Effects Screening Levels (ESLs). Although the predicted 3-hour and annual SO2 levels were below the NAAQS annual standard, the 1-hour and 24-hour predicted concentrations exceeded both the TNRCC and NAAQS standards (Table 2). The highest concentrations of SO2 were predicted to occur either along the shoreline or over Town Lake; lesser concentrations were predicted to occur in the adjacent community.

Table 2.

Summary of TNRCC Modeled and Background Ambient Air Concentrations for Pollutants Emitted from Holly Street Power Plant During Simultaneous burning of Natural Gas and Fuel Oil
Pollutants Averaging Timea Maximum Modeled Concentration (g/m3)b Background Concentrationd (g/m3) Total Ambient Concentration (g/m3) TNRCC Regulation (g/m3) NAAQS (g/m3)
Carbon Monoxide (CO) 1 hour 441 6057 6498 -- 40,000
8 hour 171 3667 3838 -- 10,000
Benzene 1 hour 0.05 -- -- 30 --
Annual 0.0009 -- -- 3 --
PAHs Annual Negligible - - 0.003-1.0e --
Nitrogen Dioxide (NO2) Annual 23 32 55 -- 100
Total Suspended Particulate Matter (TSP) 1 hour 124 -- -- 400 --
3 hour 80 -- -- 200 --
Particulate Matter (PM10) 24 hour 17 53 70 -- 150
Annual 0.88 19 19.88 -- 50
Sulfur Dioxide
1 hour 2352 (1787)c -- -- 1021 --
3 hour 1148 (872)c 26 1174 (898)c -- 1300
24 hour 540 (410)c 26 566 (436)c -- 365
Annual 9.64 (7.90)c 5.33 14.97 (13.23)c -- 80

a Short-term and long-term concentrations based on short-term and long-term emission rates respectively.
b Micrograms per cubic meter of air.
c Concentrations inside and outside parenthesis related to emission rates corresponding to 0.227 and 0.3 % sulfur content of fuel oil
d Background concentrations are the highest monitored numbers in Austin available at time of modeling (1993 data were used for CO, NO2,
   and PM10. 1991 data were used for SO2).
e Varies with different PAHs

Polychlorinated Biphenyl Contamination

In 1987, the City of Austin EUD discovered polychlorinated biphenyl (PCB) contamination at the Holly Street Power Plant. The PCBs appeared to have originated from oil leakage and overspray from rotating screens in the air screen system. The contamination extended to the ventilation system, sewer line sediments from Units 2, 3, and 4, and a storm water skimmer pit. No PCBs were found in unfiltered effluent water samples from the skimmer or in water samples taken from Town Lake (Table 3). In 1992, the drainage and air ventilation systems were cleaned to below EPA and TNRCC standards and the plant set up ongoing monthly analysis of plant discharges during storm events to detect PCBs. While PCBs were not found in the sediment and water samples from Town Lake, they were detected at trace levels in the effluent of the oil/water separator. The PCBs were detected at levels above the minimum analytical limit of 1.0 g/L (micrograms per Liter) in eight of 30 quarterly sampling events (years 1992 through 1998). Currently, the City of Austin EUD is working to increase the capacity of the main oil/water separator to further minimize PCB levels.

Table 3.

Polychlorinated Biphenyl Sample Results for Holly Street Power Plant (Maximum Detected Values)
Operating Unit Sample Type Maximum Detected Value Description of Sample Location Source
Oil Soil
Oil Screen Reservoir Air Screen Sludge
City of Austin
City of Austin
2 Wipe Sample
6.6 g/cm 2
0.65 ppm
0.05 ppm
0.13 ppm
Wall wipe Tunnel at level 2A
Top 1 inch
2-3 inches
Wipe Sample
469.4 g/cm2
2.5 ppm
4.5 ppm
Wall, 6" above floor
Northern Shaft of Tunnel
Beneath Air Duct
Wipe Sample
705 g/cm2
0.32 ppm
0.085 ppm
Oil stain location
Oil stain location @ 2"
Level 4A Tunnel
Sewer Lines
Wastewater Sediment
Effluent Wastewater
44.64 ppm
7.8 ppm
<1.0 ppb
Sewer Sediment
Effluent Side of Oil Skimmer
Unfiltered Oil Skimmer Effluent
City of Austin
Town Lake
Several Locations in Town Lake
Several Location in Town Lake

MDL = Minimum detection limit.
g/cm2 = micrograms per square centimeter.

Groundwater Contamination

Reports of quarterly groundwater monitoring data (1992-1993) collected from 11 on-site monitor wells were provided by the City of Austin [1, 2]. See Figure 2 for well locations. During all sampling events available in these reports, BTEX compounds (benzene, toluene, ethylbenzene, and xylene) were reported below minimum detection limits (MDLs). The MDL for benzene, toluene, and ethylbenzene was 1.0 g/L; the MDL for xylene was 3 g/L. These MDLs are significantly below the drinking water standards for these contaminants.

Total petroleum hydrocarbons (TPH) also are generally reported below MDLs; however, the detection limits varied from 200 g/L up to 800 g/L. TPHs were detected at levels ranging from 300 g/L to 600 g/L in instances when the MDL was set at 200 g/L. The highest reported TPH concentrations were 2,200 g/L (MW-6; September 23, 1993) and 4,100 g/L (MW-North; June 17, 1993).

Physical and Other Hazards

Electric and Magnetic Fields

The City of Austin EUD has taken magnetic field readings throughout the neighborhood surrounding the Holly Street Power Plant. Readings were taken with milligauss meters. Figure 1 presents the readings taken by the city. Readings ranged from 0.2 to 32.0 milligauss. The highest readings were measured adjacent to power lines along Holly (15.8 to 32.0 milligauss), Garden (0.8 to 29.1 milligauss), Chicon (12.0 to 27.5 milligauss), and Perdernales (6.2 to 13.9) streets. Non-power line affected readings taken around the perimeter of the plant were generally low and ranged from 0.3 to 2.3 milligauss.

At the request of TDH, the City of Austin EUD calculated peak magnetic field strengths along power line corridors for various locations throughout Austin, Travis County. Table 4 lists the calculated maximum magnetic field strengths produced along various power line corridors.

Table 4.

Calculated Maximum Magnetic-Field Strengths Produced Along City of Austin Electric Utility Department Transmission Corridors
Circuit No. Location Distance From Pole/Tower
0 feet
50 feet
100 feet
Magnetic-Field Strength (Milligauss)
Maximum Calculated Magnetic Field Strengths for Holly Street Neighborhood
Pedernales Street
Garden Street
River Crossing; S. Lakeshore
Holly Street
Between Holly and Garden Streets
Inside Holly Street Power Plant
Maximum Calculated Magnetic Field Strengths for other Areas in Travis County
N. Mopac; Research Blvd.
N. Lamar; N. Park Estates
S. Austin; N of Ben White Blvd.
N. Mopac
Gracy Woods; Braker Ln.
Colony Meadows
Braker Ln. to N, Mopac
E. & S. of Motorola (Ed Bluestein)
N. Mopac
E. Travis Co. to Bastrop Co.
Indian Hills; Creek Bend
University Hills; Springdale
Braker Ln. at Mopac
S.E. Travis Co.
First Street; Lake Austin Blvd.
Westlake Hills
Westlake Hills
Westlake Hills; Mt. Bonnell Shores
Allendale; Koenig Lane
Govalle; Del Valle; Hwy 183 @ Berg
Hyde Park
Bluff Springs Estates; Beacon Ridge
Slaughter Ln; Brodie Ln; Oak Hill
Convict Hill
Jester Estates
Jollyville Rd.; Woodcrest
Hwy 183 S; Near Bergstrom
Hwy 183 S. Near Bergstrom
Oak Hill Motorola
Great Hills

Field strengths were calculated at a position one meter above the ground at zero, 50, and 100 feet from the pole. At zero feet (a position directly beneath the wires), calculated maximum magnetic field strengths range from 1 to 81 milligauss. The field strengths listed in this table are estimates based on system peak levels for the year 1990. Since magnetic field strengths are dependent upon many factors, actual field strengths may be different than those listed.


On June 15, 1993, Radian Corporation [3] provided the City of Austin EUD with a "Screening Study of Noise Impacts from the Holly Street Power Plant." The purpose of the study was (1) to evaluate noise levels experienced by residents living closest to the plant, (2) to compare the levels experienced by these residents to background neighborhood noise levels, (3) to compare the background levels and the levels experienced by the residents living closest to the plant with noise experienced in a suburban Austin neighborhood, and (4) to correlate noise from the plant with changes in plant operation.

Radian [3] determined average day/night noise (Ldn) values for three sites (A, B, & C) positioned near the plant at nearby residential street corners. Neighborhood background measurements were taken at a fourth site (D). Noise measurements (taken in October 1991) from a previous study were used for the suburban comparison (site E). Locations for the various sites are presented in Table 5.

Radian [3] compared noise levels to the U.S. Department of Housing and Urban Development (HUD) Noise Assessment Guidelines for determination of acceptability in residential areas. According to HUD, an Ldn of 75 or greater is "unacceptable" for residential areas and an Ldn between 65 and 75 is "normally unacceptable."

Since the measurements presented in Table 5 were taken while only one unit (Unit 4) was operating, additional measurements were taken while a second unit (Unit 3) was brought on line. Table 6 compares Ldn impacts for Unit 4 only, Unit 4 and Unit 3 startup, and Units 3 & 4 together (site locations as described in Table 5). Measurements were not taken during the simultaneous operation of all four units.

Between 1994 and mid-1996, the City of Austin EUD invested approximately $2 million on a variety of sound abatement measures at the Holly Street Power Plant to reduce sound levels generated by plant equipment. Sound abatement measures included enclosure of the forced draft fans on top of each of the units, enclosure and muffling of approximately 40 other sources, and the construction of a 100-foot by 62-foot sound absorbing wall on the west side of unit 4. Measurements taken before and after the abatement program indicate that in the1-block area immediately surrounding the facility, the abatement measures reduced sound levels by 12 to 14 dBA (plant operating at 220 MW). In areas two to three blocks from the facility, sound levels have been reduced by 5 to 10 dBA. Reductions of 10 dBA are generally perceived by the human ear as a reduction in loudness of one-half. After improvements, the number of acres of land subjected to normally unacceptable noise levels (65 Ldn) was reduced from 193 acres to 54 acres (plant operating at 400 MW). The number of homes and businesses subjected to normally unacceptable noise levels was reduced from approximately 700 to 107 (plant operating at 400 MW). Thus, there continues to be a small area immediately surrounding the facility where, when the plant is operating at a 400+MW output, noise from the facility could reach levels that are considered normally unacceptable.

Table 5.

Comparison of Time-Weighted Noise Levels (Ldn) for Five Locations*
Site Location Noise Level

Qualitative Interpretation of
HUD Guidelines for
Residential Noise (Ldn)

A In parking lot of athletic field west-southwest of the plant. 72.1 "normally unacceptable"
B On the northeast corner of Mildred and Holly Streets northwest of plant boundary.


C Northeast of the plant just south of a home on the northwest side of Metz Park. 66.8 "normally unacceptable"
D Three-quarters of a mile northwest of the plant on the northeast corner of Navasota and Canterbury Streets. 55.2 "acceptable"
E In Shady Hollow neighborhood in southwest Austin 50 feet from Brodie Lane.



*Adapted from Screening Study of Noise Impacts from the Holly Street Power Plant Report, June 15, 1993, Radian Corporation [3].

Table 6.

Projected Time-Weighted (Ldn) Pre-abatement Noise Impacts for Two-Unit Operations*
Site Condition Noise Level (Ldn) Qualitative Interpretation of HUD Guidelines for
Residential Noise (Ldn)
A Unit 4 Only 72.1 Normally Unacceptable
  Unit 4 with Unit 3 in Start-up 77.0 Unacceptable
  Units 3 & 4 in Full Operation 77.0 Unacceptable
B Unit 4 Only 75.9 Unacceptable
  Unit 4 with Unit 3 in Start-up 79.1 Unacceptable
  Units 3 & 4 in Full Operation 78.6 Unacceptable
C Unit 4 Only 66.8 Normally Unacceptable
  Unit 4 with Unit 3 in Start-up 76.5 Unacceptable
  Units 3 & 4 in Full Operation 78.5 Unacceptable

*Adapted from Screening Study of Noise Impacts from the Holly Street Power Plant, June 15, 1993, Radian Corporation [3]. Locations as described in Table 5.

The EUD also funded a program to soundproof and weatherize homes surrounding Holly. The City of Austin soundproofed approximately 200 homes. This includes two-thirds of the 107 homes still exposed to sound levels above 65 Ldn. Areas with unacceptable sound levels (75 Ldn or greater) are now confined to the plants' boundaries.

After abatement, there were noticeable reductions (from 6 to 11 dBA) in sound levels during start-up. With Unit 4 active (55 MW) and Unit 3 inactive, sound levels were reduced from 68.9 dBA to 57.7 dBA. With Unit 4 active (55 MW) and activation of the Unit 3 forced draft fan, sound levels were reduced from 69.0 dBA pre-abatement to 62.6 dBA post-abatement. With both units operating (55 MW), sound levels went from 68.5 dBA pre-abatement to 60.4 dBA post-abatement.

Prior to the abatement, sound levels inside the portable classrooms were slightly above EPA's 45 dBA indoor guideline. After abatement, with the plant operating at a 94 MW power output, there was a 5.8 to 8.6 dBA reduction in noise levels inside the portable classrooms to levels below EPA's guideline.

In response to recommendations made in previous drafts of this document, the EUD took measurements inside the portable classroom 4 at Metz Elementary school with Holly operating at a power output of approximately 490 MW (Unit 1, 73 MW; Unit 2, 90 MW; Unit 3, 193 MW; and Unit 4, 140 MW). With the room air conditioning unit turned off, the measurements taken inside the classroom averaged 40.7 dB, below EPA's indoor guideline.

Fire Hazards

The Holly Street Power Plant stores fuel in aboveground storage tanks. As outlined in the background section of this report, since 1974 there have been several documented accidental releases of fuel oil. In 1993, the release of 200 gallons of #5 fuel oil resulted in a fire that burned for approximately 30 minutes.

Quality Assurance and Quality Control

In preparing this public health assessment, TDH relied on information provided by other agencies, contractors, and individuals. In most instances, quality assurance/quality control (QA/QC) data were not available; we assumed that adequate QA/QC measures were followed with regard to chain-of-custody, laboratory procedures, and data reporting. The analysis and conclusions reached in this public health assessment are valid only if the referenced information is complete and valid.


To determine whether nearby residents are exposed to environmental contaminants originating from the site, we evaluated the environmental and human components that lead to human exposure. This evaluation or pathway analysis consists of looking at the five elements of an exposure pathway: a source of exposure, transport through an environmental medium, a point of exposure, a route of human exposure, and an exposed population.

Exposure pathways are categorized as completed, potential, or eliminated. For a person to be exposed to a contaminant, the exposure pathway must be completed. An exposure pathway is considered completed when all five of the pathway elements are present and exposure has occurred, is occurring, or will occur in the future. A pathway is considered to be a potential pathway when at least one of the five elements is missing and may or may not be present in the future. Table 7 identifies the exposure pathway(s) that we have identified for this site. Inclusion of a pathway in the table does not imply that the pathway presents a risk to public health. An evaluation of the pathways for public health significance will be presented in the following section. An exposure pathway is eliminated when one or more of the elements is missing and will never be present to complete the pathway. PCB exposure and groundwater contamination were eliminated as possible exposure pathways.

Table 7.

Potential Exposure Pathways
Source Environmental
Point of
Route of Exposure Exposed
Air Holly Street Power Plant Air Neighborhood
Holly Street Power Plant
Inhalation Residents
Plant Workers

Potential Exposure Pathways


We were not able to classify this pathway as completed or eliminated because ambient air monitoring data were not available. However, air modeling data suggest that in the past, under some meteorological conditions, ground level concentrations of SO2 could have exceeded the 1-hour and the 24-hour standards for this contaminant.


The following sections discuss the possible adverse health effects that have been associated with exposure to specific contaminants associated with the potential exposure pathway that was identified in the previous section. In addition, to address community concerns, we have included a discussion of electric and magnetic fields and noise. This section also evaluates state and local databases, and addresses community health concerns.

Toxicological Evaluation

Air: Sulfur Dioxide (SO2)

Sulfur dioxide is a highly irritating, colorless gas with a pungent taste and odor that results from activities associated with the burning of fossil fuels (coal, oil) such as at power plants or from copper smelting. In nature it can be released to the air from volcanic eruptions.

Sulfur dioxide is extremely irritating to the eyes and upper respiratory tract. Exposure to concentrations above 6.0 ppm (15,600 g/m3) produces instantaneous mucous membrane irritation. Symptoms such as eye irritation, tearing, rhinorrhea, cough, shortness of breath, chest tightness or discomfort, and choking sensation are common.

In humans, numerous controlled exposure studies have shown that acute exposure to SO2 can result in lung function changes indicative of bronchoconstriction. In healthy, non-asthmatic individuals, exposure to SO2 at concentrations up to 1.0 ppm (2,600 g/m3) for up to 40 minutes was associated with a slight increase in subjective mild upper-respiratory symptoms such as sore throat [4]. Decreased tidal volume and increased respiratory flow rate were observed in healthy people at exposures of 1 to 8 ppm (2,600 to 20,800 g/m3) SO2 for 10 minutes [5] and specific airway resistance was increased in healthy individuals at exposures of 0.6 to 0.8 ppm (1,500 to 2,080 g/m3) SO2 for five minutes [6].

Controlled laboratory studies in humans have established that people with asthma are particularly sensitive to the respiratory effects of SO2. Increases in airway resistance were observed in asthmatics exposed to 1, 3, and 5 ppm (260, 780, and 1,300 g/m3) SO2 for 10 minutes by mouthpiece while at rest [7]. Exercise has been found to exacerbate the effects of SO2 on asthmatics; significant increases in airway resistance were observed in moderately exercising asthmatics exposed to 0.4 to 1.0 ppm (1,040 to 2,600 g/m3) SO2 for 3 to 10 minutes [4, 8, 9, 10, 11]. Significant changes in airway resistance were observed in young adult mild asthmatics exposed through a mouthpiece to SO2 concentrations as low as 0.25 ppm (650 g/m3) while exercising [12]. In this study, the two most sensitive individuals exhibited some degree of bronchoconstriction (slight increase in specific airway resistance) following inhalation of 0.1 ppm (260 g/m3) SO2 through a mouthpiece for 10 minutes. Others have observed increases in specific airway resistance in exercising asthmatics exposed to 0.25 ppm (650 g/m3)SO2 for 5 minutes [8]. During some controlled acute studies, asthmatics exposed to SO2 concentrations as low as 0.5 ppm (1,300 g/m3) exhibited pulmonary responses severe enough to warrant termination of the experiment and/or medical attention [13, 14]. Severe asthmatics may be more responsive to the effects of sulfur dioxide than the mild asthmatics examined in the controlled human experiments because of their lower lung reserve capacity.

Other factors such as exposure to cold dry air can exacerbate the respiratory effects of sulfur dioxide. Recent evidence suggests a potential interaction with other pollutants commonly found in urban environments; preexposure to a low concentration of ozone (0.12 ppm) for 45 minutes potentiates the bronchoconstrictive effects of an otherwise subthreshold dose of SO2 in adolescent asthmatics [15].

Acute studies in animals have supported the human data on the pulmonary effects of sulfur dioxide. Exposure to high concentrations of sulfur dioxide has been shown to result in increased airway resistance and decrease compliance in guinea pigs (2.6 ppm [6,760 g/m3]), degenerative changes in the olfactory epithelium in mice (20 ppm [52,000 g/m3]), cellular necrosis in the trachea and bronchus of rats (800 ppm [2,080,000 g/m3]), and bronchitic lesions in hamsters (650 ppm [1,690,000 g/m3]).

In summary, the available information indicate that 0.25 ppm (650 g/m3) may be close to the threshold for significant changes in lung function in most healthy asthmatics; however, there is evidence that some sensitive individuals may develop slight bronchoconstriction after inhaling 0.1 ppm (260 g/m3).

The predicted 1-hour (1,787 g/m3) and 24-hour (410 g/m3) concentrations of SO2 for the Holly Street Power Plant using 0.227% sulfur content fuel oil, which are of sufficient magnitude to potentially endanger sensitive individuals, were predicted to occur southeast of the site over Town Lake. In the surrounding community, the maximum predicted 1-hour and 24-hour concentrations were predicted to be approximately 1,212 g/m3 and 337 g/m3, respectively. Although the average SO2 concentrations to which local area residents would be exposed would be much lower, mixed fuel combustion using 0.227% or 0.3 % sulfur content fuel oil could pose a potential public health hazard.

Assuming a reduction in SO2 emissions proportional to the reduction in the SO2 content of the fuel, we estimate that using fuel containing 0.05% sulfur by weight would result in predicted 1-hour and 24-hour SO2 concentrations in the community of approximately 200 g/m3 and 56 g/m3, respectively. These levels are below their respective regulatory limits and below the levels normally associated with adverse health effects in humans. Based on available information, the levels of sulfur dioxide produced by the Holly Street Power Plant currently poses no apparent public health hazard

Air: Nitrogen Dioxide (NO2)

Nitrogen dioxide is one of many poisonous, highly reactive gases called nitrogen oxides. These gases are produced when fuel is burned at high temperatures. The principal source of these gases include motor vehicle exhaust and stationary sources such as electric utility and industrial boilers. Nitrogen dioxide is a suffocating, brownish gas that reacts in the atmosphere to form nitric acid. It also is a major participant in the photochemical generation of ground-level ozone, the primary constituent of smog.

Nitrogen dioxide can irritate the lungs and lower resistence to respiratory infections such as influenza. The principle site of acute toxicity is the lower respiratory tract with the severity of symptoms depending more on the concentration of nitrogen oxides inhaled than on the duration of exposure. The effects of short-term exposure are not clear; however, animal studies suggest that exposure to NO2 (at concentrations usually greater than 5ppm) could increase susceptibility to infection. Some epidemiologic studies have suggested that NO2 may increase the susceptibility of children and adults to respiratory illnesses.

Mild exposure may result in transient nonspecific symptoms such as dyspnea, cough, headache, fatigue, nausea, vertigo, and somnolence which disappear in hours to days but could persist up to two weeks without clinically detectable pulmonary findings. Moderate exposure may result in a multi phasic course of disease that could be fatal if left untreated. Symptoms could include dyspnea, cough, wheeze, chest pain, palpitations, weakness, diaphoresis, nausea, vomiting, and headache. Extreme exposure may result in sudden death from laryngospasm, bronchospasm, or asphyxiation.

Several studies investigating the effects of gas cooking on the lung function of children found a weak association between indoor levels of nitrogen dioxide and respiratory illness. A loss in forced vital capacity was observed in adult male volunteers exposed to 1.0 ppm (1,887 g/m3) NO2 for two hours with intermittent exercise. Asthmatics also may be more sensitive to the effects of NO2 exposure. Exposure to 0.30 ppm (530 g/m3) during 60 minutes of intermittent exercise caused decreased forced vital capacity in asthmatic subjects and short-term exposure to 500 g/m3 NO2 was found to have effects on airway resistence and bronchial responsiveness in subjects with mild asthma [16].

The highest concentrations of NO2 predicted to be emitted by the Holly Street Power Plant are significantly lower than the levels that have been associated with adverse human health effects. Based on available information, the levels of nitrogen dioxide produced by the Holly Street Power Plant poses no apparent public health hazard.

Air: Particulate Air Pollution

Exposure to air pollution has been implicated as a causative factor for increased morbidity and mortality. Elevated levels of air pollution have been associated with declines in pulmonary function [17, 18, 19], increased respiratory symptoms [20, 21, 22], asthma attacks [23], increased respiratory hospital admissions [24,25, 26], and mortality [27, 28, 29, 30, 31].

Although there are many components of air pollution, inhalable particulate matter (particulate matter smaller than 10 microns or PM10) has been implicated as a factor causally related to adverse health effects. However, since monitoring of PM10 was not adopted as the reference method in the United States until 1987, few existing studies have actually measured PM10. In most instances, alternative measures of air pollution have been used including total suspended particulates (TSP), fine particulates (< 2.5 microns), British smoke (BS), sulfates, coefficient of haze (COH), and KM (a measure of light transmittance sensitive to smaller particles). These different units of measure make comparisons with current monitored particulate levels difficult.

One of the earliest reports of an association between particulate matter and mortality was based on a study of the 1958-1959 "London Fog" episode [27]. A statistically significant relationship was reported between daily levels (24-hour average) of British Smoke and daily mortality. The association was reported primarily among those with preexisting cardiovascular or respiratory illness. In 1990, the relation between air pollution and mortality in London was examined for the winters from 1958 through 1972 [28]. Mean levels of particulate matter (BS method) ranged from a low of 59 g/m3 to a high of 536 g/m3. The mean level of BS for the first four winters was 336 g/m3, while the mean level for the last four winters was 68 g/m3 (comparable to levels commonly found in the United States). The authors reported a highly significant relation between daily mortality and levels of particulate matter after controlling for other factors. The relationship was non-linear with no apparent threshold and the exposure-response curve was steeper at lower air pollution levels.

In 1992, Schwartz and Dockery [32] examined the association between daily measurements of TSP and daily mortality in Stubenville, Ohio for the eleven-year period from 1974 to 1984. They reported a mean TSP concentration of 111 g/m3 for the eleven year period. After controlling for effects of weather and seasonality they found a statistically significant relationship between daily levels of TSP and daily mortality (Relative Risk = 1.04, 95% CI = 1.02-1.06). An increase in particulates of 100 g/m3 was associated with a 4% increase in mortality on the succeeding day. The relationship appeared to continue at levels well below the current National Ambient Air Quality Standard for TSP. In a similar study using eight years of data from Philadelphia, Pennsylvania, TSP (mean concentration = 77 g/m3) was significantly associated with all-cause mortality [32]. The association was strongest for people over age 65. Mortality from chronic obstructive pulmonary disease, pneumonia, and cardiovascular disease were each separately associated with TSP.

Several studies that used actual PM10 measurements have recently been published. Using data from Birmingham, Alabama for August 1985 through 1988 and controlling for various factors, a significant association was found between inhalable particles and daily mortality [30]. A relative risk of 1.11 (95% CI 1.02-1.20) was estimated for a 100 g/m3 increase in inhalable particles. There was no evidence of a threshold down to concentrations as low as 20 g/m3. A recent prospective cohort study, using data from a 14-to-16-year mortality follow-up of 8,111 adults in six U.S. cities, estimated the effects of air pollution on mortality [31]. A statistically significant association between air pollution and death from lung cancer and cardiopulmonary disease was observed after adjusting for smoking and other risk factors. Mortality was associated with fine particles (rate ratio, 1.26; CI 1.08-1.47), inhalable particles (rate ratio, 1.27; CI 1.08-1.48), and sulfate particles (rate ratio, 1.26; CI 1.08-1.47). The mean air pollution levels for these variables in the least- and most-polluted cities during the study period were: 11.0 and 29.8 g/m3 for fine particles; 18.2 and 46.5 g/m3 for inhalable particles; and 4.8 and 12.8 g/m3 for sulfate particles.

Using general estimates of the relationships between alternative measurements of particulate matter, Ostro [21] converted the results from several studies into a common metric (PM10). He found a striking consistency in the results. The estimated mean effect based on several studies [28, 29, 33, 34, 35, 36, 37] was a 1.0% increase in mortality for each 10 g/m3 increase in PM10.

Daily fluctuations in respirable particulates consistently have been associated with increased morbidity and mortality. The consistency of association between particulate air pollution mortality has been observed across a broad range of pollution levels. Some investigators observed the associations at extremely high concentrations of pollution and others have reported associations at concentrations commonly experienced in metropolitan areas throughout the United States. Data also suggest a dose-response relationship across the entire range of particle concentrations studied with no evidence of a threshold.

The biological mechanism for the particulate/mortality association is not known. Potential biological mechanisms are difficult to evaluate because particulate matter consists of a heterogenous mix of chemicals. Some evidence suggests that small particles may penetrate deeply into the lung resulting in bronchoconstriction and possible alteration in respiratory mechanics, causing pulmonary and cardiorespiratory disease. In addition, some constituents of particulate matter may irritate the upper airway and deep lung, reduce bronchial clearance, and modify the lung's resistance to infection. In this manner, exposure to particulates may exacerbate the conditions of individuals already compromised by respiratory infections.

The TNRCC modeling results suggest that particulate matter emissions from the Holly Street Power Plant when both natural gas and fuel oil are burned should be low. Predicted maximum 24-hour and annual average concentrations were 17.0 g/m3 and 0.88 g/m3, respectively. The maximum total ambient air 24-hour and annual average concentrations (background plus contribution from Holly) were estimated to be 70 g/m3 and 19.88 g/m3, respectively. These maximum predicted levels are significantly below the current ambient air quality standards and below levels normally associated with measurable health outcomes. Additionally, although there are uncertainties associated with these maximum predicted values, the actual concentrations to which people would regularly be exposed are likely much lower. Based on available information, we conclude that, although the Holly Street Power Plant may contribute to PM10 in the surrounding community, total PM10 levels, including maximum predicted emissions from Holly Street Power Plant, pose no apparent public health hazard.

Evaluation of Physical and Other Hazards

Electric and Magnetic Fields

Electric and magnetic fields are associated with power lines, building wiring, lighting, and all electric appliances. An electric field represents the electric force that an object is capable of exerting on other charged particles; magnetic fields are produced when charges are in motion. The combination of electric and magnetic effects is known as an electromagnetic field (EMF). The 60 Hertz (Hz) fields associated with the 60 cycle alternating currents (AC) used in power distribution systems in the U.S. are often described as extremely low frequency fields (ELF). Initially, when power companies began utilizing extra high voltage (EHV) transmission lines, a variety of issues were raised: 1) aesthetics, 2) right-of-way, 3) TV/radio interference, and 4) noise. More recently, as a result of numerous laboratory and epidemiologic studies, public concern has focused on potential adverse health effects from EMF exposures.

Laboratory Studies

Although some laboratory studies have reported potentially significant biological effects, the results must be evaluated cautiously. In many experimental studies using isolated test systems, the overwhelming majority of the effects were associated with exposure to field strengths many times higher than those seen in ambient residential settings. Calcium release from chick brain tissue [38, 39, 40, 41, 42, 43], inhibition of nocturnal melatonin synthesis by the pineal gland [44, 45, 46, 47, 48, 49, 50, 51], and the blocking of melatonin's oncostatic action on the proliferation of cultured human breast cancer cells [52] were the only effects observed under conditions analogous to ambient residential settings. Additionally, extrapolation from isolated test systems or animals to human exposure may be fraught with methodological difficulties since coupling of ELF fields to the whole body is complicated by factors such as body orientation, field polarization, grounding conditions, contact with other conductors, and body geometry. For instance, the internal fields and current densities in a rat and a human exposed to the same external field can differ significantly.

In general, the laboratory data are quite diverse and rather confusing. Few observations provide links directly to health endpoints. Biologic plausibility is difficult to establish because there is no significant transfer of energy from electric and magnetic fields at extremely low frequencies (60 Hz and less) to biologic systems. One hypothesis suggests that pineal function and melatonin synthesis may have hormonal consequences directly relevant to cancer. This hypothesis may be plausible since suppression of melatonin production has been linked experimentally with increased susceptibility to cancer in rodents [53] and with the blocking of melatonin's oncostatic action on human breast cancer cell proliferation in tissue culture [52].

Human Epidemiologic Studies

The results from human epidemiologic studies often are expressed in terms of single relative risk (RR) or odds ratio (OR) figure with a 95% confidence interval (95% CI). A relative risk greater than 1.0 represents a positive relationship between the exposure and the disease outcome, and a relative risk figure that is less than 1.0 represents a negative relationship between the exposure and the disease outcome. The 95% CI indicates the range in which we would expect the RR or OR to fall 95% of the time. A study result is considered statistically significant only if the confidence interval does not include the value 1.0. If both values of the confidence interval are greater than 1.0 the results is interpreted as a statistically significant positive relationship. For example, if the RR for a particular exposure and disease outcome is 2.3 and the 95% CI is (1.45-3.32), we interpret this to mean that the best estimate of the risk for the disease is 2.3 times higher than expected, and that we are 95% confident that the true RR is no less than 1.45 and no greater than 3.32.

Much of the current EMF controversy dates back to 1979, when Nancy Wertheimer and Ed Leeper [54] published a study which examined the electrical wiring configurations near the former homes of 344 Denver children who had died of cancer between 1950 and 1973. They reported a higher proportion of presumably "high current wiring configurations" near the homes of cases when compared to the homes of controls. Specific cancer sites which seemed to show higher relative risks included central nervous system (CNS) and/or brain cancer, leukemia, and lymphoma. In a second study published in 1982, the same authors reported a slightly increased total cancer mortality among adults who resided at several locations in Colorado; this increase also appeared to be associated with high current wiring configurations. These two studies were the first to suggest possible human cancer risks associated with residential exposure to power-frequency EMFs.

Subsequently, over 100 epidemiologic studies have been published which investigate the potential adverse health effects of residential and occupational EMF exposures. A number of these epidemiologic studies have reported positive associations of varying degrees between childhood or adult cancers such as leukemia and/or CNS/brain cancer and surrogate measures of residential or occupational exposure to power-frequency electric and magnetic fields [54, 55, 56, 57, 58, 59]. However, a number of other studies, utilizing similar methodologies, have reported no increased rates for these same cancers [60, 61, 62, 63, 64]. There is considerable inconsistency and disagreement, even among those studies that have reported "positive" results. Furthermore, both the positive and negative studies have a number of methodological flaws that make the results difficult to interpret. For example, in the initial Wertheimer and Leeper [54] study, EMF exposures were presumed rather than measured, multivariate analysis was not performed, and in many cases, birth and death addresses were summed, resulting in double counting [65]. The authors also knew the status of the household (case or control) when they evaluated and ranked the power line configurations near the home. Few, if any, studies have examined other potential confounding factors such as heredity, parental occupation, hobbies, dietary factors, smoking habits, and other environmental exposures.

Like the Wertheimer [55] study, many of the subsequent studies failed to make actual environmental measurements to confirm that homes situated near "high current configurations" actually had higher electric or magnetic field exposure than those situated near "low current configuration." The few that did perform actual environmental measurements often reported unexpected results. Savitz [57] analyzed 356 cases of childhood cancer in the Denver area from 1976 to 1983. When the wiring configuration for power distribution lines was used as a surrogate measure of exposure, there was a weak but significant association with CNS/brain cancer (odds ratio, 2.04; 95% confidence interval, 1.11-3.76). However, when the data were analyzed with respect to measured magnetic fields, the apparent correlation with CNS/brain cancer disappeared (odds ratio, 1.04; 95% confidence interval, 0.22-4.82). The odds ratios for leukemia were slightly increased, but the results were not statistically significant for either measured magnetic fields or for wiring configurations. While the Savitz [57] study was a well-executed case-control study, interpreting the results is complicated by a number of confounding variables. When these data were reanalyzed for other common factors, there was a weak, but statistically significant, association between leukemia and traffic density near the residences of cases [66]. Since gasoline contains benzene (which is known to cause leukemia in humans), it is at least equally plausible to hypothesize that the slight excess in leukemia may have resulted from long-term exposure to motor vehicle emissions. Socioeconomic status may also be an important potential confounding variable. Childhood cancer rates, in particular leukemia rates, may be associated with socioeconomic status. Thus, modestly elevated odds ratios could be explained by this source of study bias.

In 1991, London [67] conducted a case-control study based on 232 cases of childhood leukemia diagnosed from 1980 to 1987 in Los Angeles County. Controls and cases were matched for geographic area, gender, age, and race. Wire code configuration assessments were made according to the system of Wertheimer and Leeper [55] and actual magnetic fields were measured for 24 hours or longer in the child's sleeping area. The authors found no association between measured field exposure and leukemia risk with a reported odds ratio of 1.50 (95% CI 0.66-3.29) for the highest exposure group (> 2.68 milligauss). However, when wiring configuration was used as a surrogate measure of exposure, they reported an odds ratio of 2.15 (95% CI 1.08-4.26).

Recently, Feychting and Ahlbom [59], used a nested case-control study to test the hypothesis that exposure to magnetic fields from high voltage power lines (transmission lines) increased cancer incidence in children. The study base consisted of everyone under age 16 who lived on a property at least partially located within 300 meters of any of the 220 and 400 KV power lines in Sweden during the period from 1960 to 1985. Exposure was assessed by spot measurements and by computer estimates of the historic magnetic fields generated by the power lines, taking distance, line configuration, and load into account. When historical calculations were used as a surrogate measure of exposure, there was a weak but significant association with leukemia for 2 milligauss or more (relative risk, 2.7; 95% confidence interval 1.0-6.3). For 3 milligauss or higher, the relative risk was estimated at 3.8 (95% confidence interval 1.4-9.3). When adjustments were made for demographic variables, socioeconomic status, and levels of nitrogen dioxide, the results were virtually unchanged. These associations were based on seven observed cases of leukemia. For all cancers combined and for CNS system tumors or lymphoma, there was no relation with historical estimates of magnetic fields. When the data were analyzed with respect to measured magnetic fields (i.e., spot measurements), there was no excess risk for total cancer or for leukemia. For CNS tumors, the relative risk was estimated at 2.5 (95% confidence interval 0.9-6.6) for the intermediate exposure level, but at 1.5 (95% confidence interval 0.4-4.9) for the highest level.

It is difficult to come to firm conclusions regarding the potential health risks associated with EMF exposure from available studies. The numerous epidemiologic studies of EMF exposure and cancer have shown inconsistent and even contradictory results. Most non-occupational studies of EMF exposure and cancer have considered the wiring configuration at the person's place of residence at the time of death as a surrogate measure of EMF exposure, but most studies have not considered the duration of residence. Of all the conditions studied, leukemia and EMF exposure remains the most suggestive of a possible association. Although there is a persistent unexplained association between surrogate measures of exposure and leukemia risk, the risk does not correlate with actual measured electric and magnetic fields. Associations with CNS/brain cancer also remain suggestive but inconclusive.

In 1995, TDH, as part of the State of Texas Environmental Priorities Project (STEPP) review of the EMF issue, plotted power consumption trends (per capita U.S. electric power consumption) together with cancer mortality trends. The age-adjusted leukemia (Figure 3) and CNS/brain cancer (Figure 4) mortality rates among children have shown nearly a 3-fold decrease since approximately 1955, while the U.S. per-capita power consumption has continued a nearly exponential growth rate which began around the turn of the century (producing nearly a 4-fold increase in power consumption over the same period [68]). If EMF exposure were contributing significantly to either of these diseases, we might expect a strong correlation between those mortality rates and power consumption trends in the U.S. While this type of analysis cannot disprove an association between EMF exposure and cancer, it would tend to indicate that if such an association does exist, the effect must be relatively minor. This inconsistency, between the hypothesis and the observations, would suggest that the epidemiologic evidence should be considerably more substantial before a causal association can reasonably be assumed.

For most studies examining adverse health conditions and EMF exposure, the reported relative risks are generally in the range of 1.2 to 3.8. In most cases, the higher relative risks are produced by a very few cases occurring when one or less than one case would be expected. Thus, the lower 95% confidence limits are usually only slightly greater than 1.0. EMF studies also have rarely shown a significant dose-response effect. For most contaminants a dose-response relationship often is considered a prerequisite for inferring a cause and effect relationship.

The energy produced by 60 Hz electric and/or magnetic fields from power lines generally is considered far too low to cause any breakage of molecular bonds or other disruption of DNA molecules. Presently, evidence for other plausible biologic mechanisms to explain how EMF exposure might result in the development of cancer is weak and needs further study. The most plausible hypothesis involves the suppression of melatonin production. The results of experimental animal studies have led some scientists to conclude that EMFs do not act as a cancer initiator. Although some argue that EMF may be a cancer promoter, the experimental data is lacking to confirm this hypothesis.

Although available information suggests that further studies of EMF exposure are warranted, the evidence for adverse health effects is currently inconclusive. Because of numerous inconsistencies, methodological deficiencies, and contradictory findings, current evidence is insufficient for establishing a cause-and-effect relationship between EMF exposure and adverse health effects from which to quantify the risk (if any) which may result from exposure to EMF.

The National Institute of Environmental Health Sciences (NIEHS) recent report to Congress, "Health Effects from Exposure to Power-Line Frequency Electric and Magnetic Fields [69], concluded:

"The scientific evidence suggesting that [power-frequency electromagnetic field] exposures pose any health risk is weak. The strongest evidence for health effects comes from associations observed in human populations with two forms of cancer: childhood leukemia and chronic lymphocytic leukemia in occupationally exposed adults. While the support from individual studies is weak, the epidemiological studies demonstrate, for some methods of measuring exposure, a fairly consistent pattern of a small, increased risk with increasing exposure that is somewhat weaker for chronic lymphocytic leukemia than for childhood leukemia. In contrast, the mechanistic studies and animal toxicology literature fail to demonstrate any consistent pattern across studies although sporadic findings of biological effects (including increased cancers in animals) have been reported. No indication of increased leukemias in experimental animals has been observed.

Epidemiological studies have serious limitations in their ability to demonstrate a cause and effect relationship whereas laboratory studies, by design, can clearly show that cause and effect are possible. Virtually all of the laboratory evidence in animals and humans and most of the mechanistic work done in cells fail to support a causal relationship between exposure to [power-frequency electromagnetic fields] at environmental levels and changes in biological function or disease status. The lack of consistent, positive findings in animal or mechanistic studies weakens the belief that this [epidemiological] association is actually due to [power-frequency electromagnetic fields], but it cannot completely discount the epidemiological findings...

The National Toxicology Program routinely examines environmental exposures to determine the degree to which they constitute a human cancer risk and produces the "Report on Carcinogens" listing agents that are "known human carcinogens" or "reasonably anticipated to be human carcinogens." It is our opinion that based on evidence to date, [power-frequency electromagnetic field] exposure would not be listed in the "Report on Carcinogens" as an agent "reasonably anticipated to be a human carcinogen." This is based on the limited epidemiological evidence and the findings from the EMF-RAPID Program that did not indicate an effect of [power-frequency electromagnetic field] exposure in experimental animals or a mechanistic basis for carcinogenicity."

Regardless of the controversies surrounding the potential adverse health effects associated with exposure to electric and magnetic fields, the Holly Street Power Plant is not the source of the electric and magnetic fields observed in the community. Electric and magnetic fields are associated with power lines, building wiring, lighting, and all electric appliances; these sources would not be eliminated with the cessation of operations at Holly Street Power Plant. The EMF measurements taken in the neighborhood near the facility were associated primarily with the transmission and distribution lines in the neighborhood. Since the time that the measurements reported in this document were taken, the city has moved one of the major power lines from Holly Street Power Plant to the Town Lake shoreline. Although the data were not available, moving this power line should have reduced the EMF readings along the south side of Holly Street Power Plant.


Exposure to noise

can result in both auditory and non-auditory adverse health outcomes with the types of outcomes dependent upon the frequency, intensity and duration of the exposure. The role of noise in causing sensorineural hearing loss has long been recognized, and research has progressed to the point where the relationship between the level of noise, the duration of exposure, and degree of expected damage can be determined. There is much less agreement on the relationship between noise and non-auditory adverse health outcomes.

Auditory Effects

Exposure to intense levels of noise (140 to 160 dB) can cause acute acoustic trauma (AAT) or permanent damage to the middle and inner ear. The most common symptoms of AAT are hearing loss and tinnitus [70]. Exposure to noise can result in temporary or permanent hearing losses [71]. Losses generally occur in the higher frequencies (4,000 to 6,000 Hz) and have been described as a 4,000 Hz shift. Temporary losses of hearing or "temporary threshold shifts" or "auditory fatigue" (at these higher frequencies) may be caused by loud industrial noises typically at sound pressure levels > 80 dB. Temporary threshold shifts occur within the first two hours of exposure. However, full recovery may take two to 48 hours after leaving the noisy environment. Permanent hearing loss or a permanent 4,000 Hz shift can occur from long-term (generally 10 years or more) exposure to industrial noise. With continued exposure, the loss extends to frequencies above and below 4,000 Hz. Since noise-induced hearing losses occur at frequencies outside normal conversational frequencies, the losses generally occur without the person knowing it.

Most of the information concerning the auditory effects of noise have come from studies conducted on adults. Conclusions pertaining to children are often based on extrapolations from adult data. Although the data are limited and often contradictory, it has been suggested that children may be more susceptible to the effects of noise [72].

Non-Auditory Effects

Exposure to noise can produce a number of acute physiological responses in both animals and humans. The basis for these acute effects appears to be grounded in the stress response governed by sympathetic activation of the autonomic nervous system. The best-documented non-auditory physiological effects of noise have been demonstrated in several studies which have found an association between high noise levels and elevated blood pressure [73, 74, 75, 76]. Studies of human volunteers have supported such a relationship in both normotensive and hypertensive individuals [73]. Wang [77] reported an increased prevalence of hypertension in a group of retired textile workers exposed to noise above 100 dBA compared with age-matched controls from the same population exposed to low noise levels. A recent study of noise and hypertension in 1,101 female textile workers was the first to provide evidence for an association and a dose-response relationship between exposure to noise and hypertension [76]. The workers were grouped according to exposure level (75-80 dBA, 86-90 dBA, 96 dBA, and 104 dBA) and the study controlled for independent risk factors such as salt intake, family history, and age. The study estimates that a 30 dBA increase in sound pressure level roughly doubles the risk of hypertension, independent of other risk factors. The authors reported that the prevalence of hypertension predicted using a logistic model fitted to the data at 70 dBA were comparable with the general population.

Exposure to noise has resulted in reduced fertility and enlargement of the ovaries in laboratory rats [78, 79]. However, data on the effects of prenatal noise exposure in humans are limited. Available studies lack information on individual exposure levels, have inadequate sample populations, and do not have appropriate control populations [80]. One case-control study of 1,200 women found no relationship between occupational exposure to noise (greater than 80 dB) and the risk of premature birth or low birth weight babies [81].

Children may be more susceptible to certain non-auditory effects of noise because they have less precise speech, limited vocabulary, and less developed familiarity with language rules [82]. Chronic exposure to high levels of noise during the periods in which children are acquiring speech, language, and listening skills may contribute to poor performance in reading and other academic areas [82]. One study involving children in grades 3 and 4 reported that noise-exposed students (school located under the flight path of the Los Angeles International Airport) more often failed to solve or gave up on a puzzle-solving task. The performance on a distraction task was initially better but became worse as the duration of the exposure increased [83]. A similar study in New York City reported that the percentage of children reading below grade level increased as noise increased [84]. A study of freeway noise in 100 classrooms in 15 schools in California also found that reading scores decreased as noise levels increased [85]. The authors noted that when the community was quiet, relatively more noise could be tolerated in the classroom without negatively affecting reading achievement. Although the specific mechanisms have not been clearly identified, the above studies suggest that noise may interfere with cognitive development and academic performance.

From analysis of reading achievement results, Lucas et al. derived a tentative criterion for noise inside classrooms equivalent to about 46 dBA, a level similar to the indoor level of avoidance of interference and annoyance established by the EPA of 45 dBA. A noise limit approximately equivalent to a day/night average sound (Ldn) of 60 dBA was recommended for the community around the school because of the interactive nature of school and community noise levels found in the study. This level is lower than the 65 dBA threshold that HUD has set for "normally unacceptable" residential noise.

The day/night average noise levels measured near the Holly Street Power Plant in the preliminary assessment by Radian [3] exceed the levels normally considered acceptable by HUD. Although these levels are well below those reported to result in hearing loss or other physiologic changes, the levels could have been considered to be a nuisance by nearby residents. These levels also were well within the range of levels reported to interfere with cognitive development in children. Pre-abatement measurements made by Radian [86] suggested that Metz Elementary School was within the 65 to 70 Ldn contours and the indoor sound levels in the portable buildings were slightly higher than EPA's 45 dBA indoor guideline. Outdoor levels were above the 60 Ldn recommended for communities around schools and above the 65 dBA threshold that HUD has set for "normally unacceptable" residential noise.

Between 1994 and mid-1996, the EUD took significant measures to abate the effects of the noise from the facility on the surrounding community. These measures resulted in reductions in noise levels equivalent to a reduction in loudness of one-half. Except for an area immediately adjacent to the facility, these measures resulted in reductions in sound levels to those normally deemed to be acceptable. The EUD also has provided soundproofing to approximately 200 homes in the Holly Street neighborhood. This includes approximately two-thirds of the 107 homes in an area with sound level measurements above 65 Ldn. The potential sound levels in the area immediately adjacent to the facility do not pose a public health hazard.

The post-abatement measurements taken inside the portable classrooms showed reductions in indoor sound levels well below EPA's 45 dBA indoor guideline; however, all measurements at Metz were made with the plant operating at a 94 MW power output. In Austin, children begin school during the month of August, one of the peak demand months for electricity. Higher power output levels at Holly Street Power Plant (such as those that may occur during peak demand months) could result in higher noise levels at Metz. Since we could not predict with any degree of certainty the magnitude of this response we asked the EUD to take measurements in the classroom during a peak demand month. On August 9, 1999, the EUD took measurements inside one of the portable classrooms at the school with the plant operating at a power output of approximately 496 MW. The measurements averaged 40.7 dB, well below EPA's indoor guideline. Based on these measurements we would not expect noise from Holly to pose a public health hazard to children in the school.

Health Outcome Data Evaluation

To address the community concern with air pollution and the possible carcinogenic effects of EMFs, we conducted an evaluation of cancer incidence for the zip code area in which the Holly Street Power Plant is located. We limited the evaluation to those types of cancers that are either known or suspected to have associations with the contaminants of concern--lung cancer, brain cancer, and leukemia.

To determine if there is an excess in these specific cancers in this area, we compared the number of cancer cases in the neighborhood for each selected type of cancer to the expected number of cancer cases for the area (based on the age, sex, and race-specific cancer incidence of a control population). Due to incomplete reporting in the State of Texas, the number of expected cases for African Americans and Whites is based on U.S. Surveillance, Epidemiology, End Results (SEER) Program cancer incidence rates for the period of 1986 through 1990. For Hispanics, the number of expected cases is based on cancer incidence rates for California during the period of 1988 through 1990. The ratio of observed to expected cancer cases is called a standardized incidence ratio (SIR). When the SIR for a selected type of cancer is equal to 1.00, then the number of observed cases is equal to the number of expected cases. Confidence intervals were also calculated for each SIR. The 95% confidence interval indicates the range within which we would expect the SIR to fall 95% of the time. The confidence interval is a statistical reflection of the precision of the risk estimate. If the confidence interval includes 1.00, no statistically significant excess of cancer is indicated.

It should be noted that limitations and uncertainties are associated with the incidence rates presented in this report. While the incidence of cancer was reported by zip code, reliable population data were available only by census tract. The Holly Street zip code includes parts of ten census tracts, but the boundaries do not match exactly. Therefore, calculated incidence rates are imprecise. In addition, the area encompassed by the Holly Street zip code includes people who live up to 1 1/2 miles from the site and are unlikely to be affected by most contaminants that could be associated with the site.

Table 8 summarizes the cancer incidence data for three selected cancer sites (lung, brain, and leukemia) based on available data for the 78702 zip code. This table lists the number of observed cancers, the number of expected cancers, the SIRs, and the 95% confidence intervals. Data are presented by sex. For the years 1989 to 1992, the lung cancer experience for males living in this zip code area is significantly greater than the expected rate. This excess lung cancer experience can be attributed primarily to a significant excess lung cancer for African American males (observed number = 39, expected number = 21.97, SIR = 1.78, CI = 1.26 - 2.43). The lung cancer experience for White and Hispanic males is not significantly different than the expected rate. However, the lung cancer experience for White females is significantly different than would be expected (observed number = 8, expected number = 3.25, SIR = 2.46, CI = 1.06 - 4.85). This excess lung cancer cannot necessarily be attributed to the Holly Street Power Plant for a number of reasons: 1) if elevated cancer rates were associated with an environmental exposure, we would expect elevated levels across both sexes and all races within the exposed population; 2) although actual air monitoring data have not been collected, available information does not indicate that the Holly Street Power Plant contributes significantly to air pollution within the zip code area; and 3) the major risk factors for lung cancer such as cigarette smoking and occupational exposures have not been evaluated for this population.

The cancer incidence data show that the cancer experience of these residents, for the cancers most often associated with exposure to EMFs (brain and leukemia), is not significantly different than would be expected. Since EMF exposure has been associated most often with childhood cancer, an age distribution for the selected cancers is presented in Table 9. There were only two cases of leukemia and one case of brain cancer in individuals under the age of 65. Of the 79 cases of lung cancer observed, 78 were reported for people 40 years old or older.

Table 8.

Number of Observed and Expected Cancer Cases and Race-Adjusted Standardized Incidence Ratios, Selected Sites, Austin, Texas Zip Code 78702 For the Years 1989-1992
LUNG 56 32.78 1.71* 1.29 - 2.22
BRAIN 1 2.07 .48 .01 - 2.69
LEUKEMIA 3 4.48 .67 .14 - 1.96
LUNG 23 18.69 1.23 .78 - 1.85
BRAIN 2 1.88 1.06 .13 - 3.84
LEUKEMIA 3 4.24 .71 .15 - 2.07

Note: The SIR (Standardized Incidence Ratio) is defined as the number of observed cases divided by the number of expected cases. The number of expected cases is based on White/Black cancer incidence rates, U.S. SEER during the period 1986-1990, and Hispanic cancer incidence rates, California during the period 1988-1990.
* Significantly higher (at the 5% level) than expected.

Table 9.

Age Distribution for Selected Cancers in Austin, Texas Zip Code 78702 For the Years 1989-1992
00-04                     1
20-24     1                 
30-34             1        
35-39                     1
40-44 1                    
45-49 3         1        
50-54 5         1        
55-59 8         2        
60-64 5         2      
65-69 6     1 3        
70-74 10         4 1 1
75+ 18     2 9 1    

Community Health Concerns Evaluation

We have addressed each of the community concerns about health as follows:

  1. Are rates of cancer, respiratory illness, or other health problems higher in the vicinity of the site?
  2. For the years 1989 to 1992 the lung cancer experience for African American males and White females living in this zip code area is significantly greater than would be expected. The observed rates for brain cancer and leukemia, the cancers most often associated with EMF exposure, were not different than expected rates for this population. Smoking and occupational exposures, by far the greatest contributor to excess lung cancer, have not been evaluated for this population.

    Data are not available to determine if rates of respiratory illnesses or other health problems are higher in the neighborhood around the Holly Street Power Plant. According to available data, the only time that the plant would be likely to produce contaminants associated with respiratory illness would be on the occasions when it burned high sulfur content fuel oil. Based on the air modeling results, it is possible that when high sulfur content fuel was burned in the past, sulfur dioxide levels could have been high enough to temporarily affect sensitive individuals. The voluntary conversion to a low sulfur content fuel oil should ensure that sulfur dioxide emissions remain below the levels potentially able to cause adverse reactions in sensitive individuals. Under current operating conditions, it is not likely that the plant would be responsible for increased rate of respiratory illnesses or other health problems.

  3. Are PCBs released from the site causing adverse health effects?
  4. We were able to identify four spills in which PCBs were implicated. However, except for one event where plant personnel detected PCBs in water discharged from the plant, all other events were confined to the site. Subsequent testing of water and sediment from Town Lake by an independent consulting firm has shown no detectable levels of PCBs. We were not able to find any evidence that people have come into contact with PCBs from the Holly Street Power Plant. Without contact adverse health effects would not occur.

  5. Are there PCBs in water from Town Lake?
  6. Available sampling data does not indicate that water from Town Lake has been contaminated with PCBs from the Holly Street Power Plant.

  7. Do the electric and magnetic fields from the plant measured in the neighborhoods and on the playground represent a health hazard?
  8. Although the Holly Street Power Plant is a source of electricity, the plant itself is not the source of the electric and magnetic fields observed in the neighborhood. Electric and magnetic fields are associated with power lines, building wiring, lighting, and all electric appliances. The EMF measurements reported by the City were associated primarily with transmission and distribution lines in the neighborhood. Sources of EMFs in the home include, microwave ovens, hair dryers, electric shavers, electric dial clocks, and electric blankets.

    Because of numerous inconsistencies, methodological deficiencies, and contradictory findings, evidence of a cause-and-effect relationship between EMF exposure and adverse health effects is insufficient for deriving a health-based standard for EMF exposure. Although further studies of EMF exposure are warranted, the current evidence for adverse health effects is inconclusive.

    Since the main sources of EMFs are power lines, building wiring, and electric appliances, the fields measured in the surrounding neighborhood should not be different from those measured in other neighborhoods with power lines. Magnetic-field strengths calculated near other power lines support this conclusion. Although we cannot state definitively whether EMF exposure does or does not represent a hazard, we can state with reasonable certainty that removing or closing the plant would not reduce the EMFs in the neighborhood unless the power lines were de-energized or removed and home sources also were removed.

    Childhood leukemia and brain cancer, the primary adverse health outcomes that are thought to be associated with exposure to EMFs, were not elevated in this neighborhood.

  9. Is it safe for children to play on the playground and on the basketball courts near the plant?
  10. With respect to EMFs and noise levels, we believe that playing on the playground or basketball courts near the plant does not pose a significant health risk. In the past, sulfur dioxide emissions from the plant could have affected the most sensitive individuals on the rare occasions when fuel oil was burned. The plant has agreed to use fuel oil only in emergency situations. The EUD also has agreed only to use low sulfur fuel oil. Air pollution resulting from the burning of the low sulfur fuel oil should be minimal and should not be associated with any significant adverse health effects. Although the probability of a catastrophic event occurring is low, there have been non-catastrophic events (such as fire) at facilities in Austin. We have recommended proactive actions that the City of Austin EUD can take to further minimize risk to the surrounding community.

  11. Does the noise produced by the plant have a negative effect on the community?
  12. The day/night average noise levels (Ldn) measured near the Holly Street plant, taken during Radian's preliminary assessment, exceeded levels normally deemed acceptable by HUD. Although these levels were below those normally reported to result in physiologic changes or hearing loss, they were well within the range of values that have been reported to interfere with cognitive development. Abatement efforts undertaken by the City of Austin have, for the most part, reduced noise levels in the community to levels normally deemed acceptable. There continues to be a small area close to the facility where noise levels during high power output intervals could be marginally unacceptable. While these sound levels would not pose a health hazard, sensitive individuals could consider them annoying.

    Children may be more susceptible to certain non-auditory effects of noise since they have less precise speech, limited vocabulary, and less developed familiarity with language rules. Chronic exposure to high levels of noise during the periods in which children are acquiring speech, language, and listening skills may contribute to poor performance in reading and other academic areas. Pre-abatement sound measurements suggested that sound levels at some areas of Metz Elementary School were slightly higher than recommended guidelines. Post-abatement measurements taken at Metz showed reductions in indoor sound levels well below the recommended indoor guideline. Although all measurements at Metz were made with the plant only operating at a 94 MW power output, the measurements were taken with Unit 4 operating with both forced draft fans operational. These forced draft fans were the largest contributor to the elevated noise levels at Metz School prior to the initiation of the abatement project. At our request, the City of Austin EUD took additional measurements inside the Metz portable classrooms with Holly operating at a high power output (496 MW). The measurements taken inside the classroom were lower than EPA's recommended guidelines.

  13. Can the respirable particulates produced by the plant result in adverse health effects?
  14. The Holly Street Power Plant operates with clean burning natural gas greater than 99 percent of the time. Under some conditions, such as when natural gas supplies become scarce or when special burns are required, the plant burns fuel oil. Air modeling data obtained from the TNRCC indicate that respirable particulates emitted by Holly Street Power Plant do not pose a threat to public health.

  15. Are potential fires at the plant a health hazard?
  16. Any fire can present a health hazard either from the fire itself or from the smoke and other contaminants generated by the fire. On March 13, 1993, 200 gallons of #5 fuel oil were released, ignited, and continued to burn until it was extinguished by the City of Austin Fire Department. The fire was put out in less than 30 minutes. However, a great quantity of smoke was produced according to witnesses. Increased PM10 and other pollutants produced during a fire could present an increased risk to sensitive individuals. On June 14, 1999, an accident caused a fire requiring the evacuation of approximately 60 employees. The City of Austin EUD has taken proactive measures to further reduce the risk of such events.

  17. People still fish in Town Lake regularly. Is it dangerous to eat fish from Town Lake?
  18. Since the consumption advisory is based on the presence of chlordane in the fish tissue, this issue is not related to the Holly Street Power Plant. People are allowed to fish in Town Lake, but are advised to limit or refrain from eating the fish that they catch. TDH is in the process of reevaluating the fish consumption advisory for Town Lake.

Next Section     Table of Contents The U.S. Government's Official Web PortalDepartment of Health and Human Services
Agency for Toxic Substances and Disease Registry, 4770 Buford Hwy NE, Atlanta, GA 30341
Contact CDC: 800-232-4636 / TTY: 888-232-6348

A-Z Index

  1. A
  2. B
  3. C
  4. D
  5. E
  6. F
  7. G
  8. H
  9. I
  10. J
  11. K
  12. L
  13. M
  14. N
  15. O
  16. P
  17. Q
  18. R
  19. S
  20. T
  21. U
  22. V
  23. W
  24. X
  25. Y
  26. Z
  27. #