Aircraft and Airport-Related Hazardous Air Pollutants: Research Needs and Analysis (2008)

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

37 This section reviews the measurements of ambient HAP concentrations on or near airports and is divided this into three main categories: (1) inside terminal buildings, (2) on airport grounds, and (3) in nearby communities. Instead of presenting an exhaustive review of all known studies, we pre- sent only key results from various studies and refer to two excellent documents that present a more in-depth review of past measurement campaigns—the 2005 ORD final environ- mental impact statement (FAA 2005), and the 2003 FAA review (URS 2003). 6.1 Hazardous Air Pollutants Concentrations Inside the Airport There are few measurements of HAPs inside airport build- ings. Measurements of aldehydes in an airport in Strasbourg, France found a negligible difference between indoor and out- door concentrations at the airport, though the outdoor concentrations were among the highest recorded at various outdoor locations. Average formaldehyde and acetaldehyde concentrations were 11 μg/m3 and 4 μg/m3, respectively (Marchand, Buillot et al. 2006). PAH measurements inside an Italian airport found benzo[b+j+k] fluoranthene and benzo [a]-pyrene concentrations to be of concern (Iavicoli, Carelli and Bergamaschi 2006). Other PAHs measured included naphthalene, 2-methylnaphthalene, 1-methylnaphthalene, and biphenyl. 6.2 Hazardous Air Pollutants Concentrations on Airport Grounds Among the only peer-reviewed papers are measurements of CO, NOx, and numerous VOCs from Zurich airport (Schurmann, Schafer et al. 2007). VOC measurements were limited to several “snapshot” measurements near the taxiway. Due to the limited number of measurements, it is hard to infer any information regarding average concentrations. For example, the benzene concentrations reported varied from 0.15 to 13.1 ppb. During the recent APEX campaigns, most instruments used for the study sampled airport air continuously between dedicated engine tests and thus collected several days worth of data at OAK and Cleveland Hopkins International Airport (CLE). At the OAK site, the measurement team was set up at a ground run-up enclosure that was downwind of an active taxiway. As such, the site was exposed to advected exhaust plumes 10 to 60 sec in duration from idling aircraft. Over long time periods (more than 3 hr), the average concentra- tions were minimally affected by this intermittent compo- nent and were very close to the “background” concentrations. At times of heavy air traffic, however, the average concentra- tions could be significantly higher depending on the how often idling aircraft passed upwind of the site. For example, during one busy 45-min time period, the average concentra- tion of formaldehyde was 2.4 ppb, compared to background concentrations of 0.6 ppb. These measurements show that the air at OAK is not well mixed, and that accurately quanti- fying the air quality at an airport would require measure- ments at numerous locations. In contrast, the measurement site at APEX3 was situated on the outskirts of the airport property and no discernable aircraft exhaust plumes were observed. There has been little analysis of diurnal data taken at this site. 6.3 Hazardous Air Pollutants Concentrations in Adjacent Neighborhoods Much more effort has been devoted to the measurement of HAPs in adjacent neighborhoods and at airport boundaries, with mixed results. Some measurements have shown that concentrations are no higher than in other urban settings (McGulley, Frick and Gilman, Inc. 1995), whereas others S E C T I O N 6 Ambient Hazardous Air Pollutant Measurements

have inferred an airport influence based on comparison of upwind and downwind measurements (Rhode Island Department of Environmental Management 2007). Most studies have relied on measurements with low temporal res- olution and have struggled to identify the cause of the enhanced concentrations (aircraft versus nearby roads, etc). Most of the compounds emitted by aircraft exhaust (NOx, CO, black carbon, VOCs, SO2) are also emitted by other urban sources, particularly on-road vehicles. There are few if any unique tracer compounds that have been used to connect ambient measurements to airport emissions (which include both aircraft and vehicle emissions). Furthermore, a conclu- sion that demonstrates that pollutant concentrations near an airport are no higher than in other urban locations does not imply that the airport had no influence. 6.3.1 Discerning an Airport Influence on Elevated Concentrations Results from the TF Green Air Monitoring Study at Prov- idence Airport (PVD) (Rhode Island 2007) showed that no VOC concentrations exceeded the acute or chronic non- cancer health benchmarks at five sites near the airport in Warwick. Average concentrations, however, of benzene, 1,3- butadiene, formaldehyde, acetaldehyde, acetone, chloroform, carbon tetrachloride, and perchloroethylene exceeded the cancer benchmark levels. At all measurement sites, the con- centrations of these HAPs were lower than at an urban meas- urement site in Providence. Since the chlorinated com- pounds are mainly caused by use of chlorinated solvents (as opposed to combustion sources such as aircraft or diesel ex- haust), the Rhode Island Department of Environmental Management is inspecting sources of such solvents in the area. Inference of an airport influence based solely on the measured VOC concentrations and wind direction was in- conclusive, mainly because the averaging time was 24 hr for most data (with some 3-hr samples) and thus correlations with wind speed were not very meaningful. However, 1-min averaged measurements of black carbon, which is emitted by both aircraft and diesel engines, showed a clear increase at the four measurement sites closest to the airport when the wind was blowing from the airport. Furthermore, the temporal profile of the elevation in black carbon measurements coin- cided with the increase in jet activity at the airport in the early morning. Further measurement studies and investigation of lung cancer data are planned. In contrast, a suite of 24-hr average VOC measurements conducted for one month at various locations near Ted Stevens Anchorage International Airport was unable to dis- cern a difference between upwind and downwind sites, and concluded that “Sites in and around airport tended to have lower BETX (benzene ethylbenzene toluene xylene) concen- trations presumably because they were affected less by auto emissions.” Only eight compounds were above the instru- mental detection limit (Municipality of Anchorage 2003, 7). Ambient pollutant concentrations were measured at Toronto’s Lester B. Pearson International Airport (LBPIA) from May 2005 through April 2006 (Tradewind Scientific 2006). The purpose of the study was to assess the air pollution resulting from airport activities and aircraft operations. The study employed a mobile laboratory that was moved between five carefully chosen locations on the airport grounds. Stationary monitoring was conducted at each location for extended time periods. Mixing ratios of several pollutants were reported as hourly averages. The study continuously measured levels of CO, NO2, NO, NOx, O3 and PM2.5 as well as wind speed, wind direction, and ambient temperature. In addition, total suspended particulates (TSP) and speciated hydrocarbon compounds were determined using samples acquired during 24-hr periods every sixth day. The measure- ment data were analyzed using a variety of methods includ- ing interspecies correlation, correlation with wind speed, and comparison with simultaneous measurements made at a nearby air quality monitoring station. This study did not focus on HAPs measurements though there is undoubtedly some information about HAPs emissions in the less frequent speciated hydrocarbon measurements. The major conclusions of the study were that the meas- ured pollutant concentrations at the airport are strongly affected by the background concentrations surrounding the airport and that this data set could be used for further studies of air pollution impacts, pollutant distribution, dis- persion processes and emission chemistry. Also, the com- plete data set could be used to test dispersion models of the airport’s pollutant emissions. One specific finding was that CO levels were elevated at the measurement site that was located on the apron near one of the terminals. Local sources at that location would probably include aircraft service vehicles, idling aircraft and the terminal itself. As the study concluded, further analysis of the measurement data is needed to draw stronger conclusions about source strengths and locations. 6.3.2 Source Apportionment The 1999 study by KM Chng attempted to use “advanced chemical fingerprinting” to ascertain the influence of aircraft exhaust on soot deposition near ORD. The study concluded that samples collected near ORD more closely resembled on- road vehicle exhaust than they did aircraft exhaust (KM Chng 1999). Although this study focused on particulate matter, its findings on VOC emissions and concentrations were similar. This study was subsequently criticized by the Park Ridge studies (ENVIRON 2000). 38

6.3.3 San Leandro Measurements At the conclusion of the JETS-APEX2 study, the Aerodyne Mobile Laboratory spent two days at the San Leandro Marina (Figure 14), which is ~2 km downwind of the OAK runway. Winds were consistently from the northwest. Although lim- ited in duration, these measurements are unique in that they enabled observations of diluted airport emissions with no interferences from non-airport sources, since there is no land in between the emissions and the measurement site. A 6-hr time-series of HCHO, CO, NOx, CO2, and PM number concentration is shown in Figure 15. Individual air- craft exhaust plumes resulting from idle, take-off, and landing activity could be resolved. The average HCHO concentration in the time series shown is 1.3 ppb, while the interpolated background value is approximately 0.8 (simi- lar to the background value observed on the airport grounds). More importantly, a dilution factor of ~5000 can be inferred for most of the observed plumes based on com- parison of the observed CO2 to known CO2 concentrations at the exit of a high-bypass turbine engine. Such a simplified source-receptor scheme would be an ideal scenario for test- ing dispersion models. 39 Figure 14. San Leandro Marina and Oakland International Airport. Figure 15. Time series of measurements recorded 2 km downwind of OAK.