Hoofdmenu

Over Forces...
Thema's
Posters
Artikelen
Analyses
Media
Vliegen
Links
Forces NL Forum
Commentaren
Steun Forces
Archief
Horror Stories

 Navigatie

Hoofdpagina
Zoekpagina
Inhoudsopgave
Wist U...?
All Time Favorites

 Internationaal

Forces Psychiatry

Canada
Manitoba (email)
Italy
New Zealand
UK (email)
Russia

 VS afdelingen

California
Connecticut
Delaware
Duluth
Georgia
Indiana
Maine
Massachusetts
Minnesota
Rochester
USA
Virginia

 Affiliates

Smokers' Club
NYC C.L.A.S.H.
Smoking Paradise
MA Citizens for Freedom
Real Texas Freedom
Ontario Smoking

	The Evidence Archive
What you think you know about tobacco may surprise you

Forces Comité van Aanbeveling

Forces Nederland
E-Mail:
info@forces-nl.org

5. SYNTHESIS AND DISCUSSION

Selected results from the previous section are synthesized and discussed in Section 5.1. In Section 5.2 ETS contaminants and pollutants are further analyzed and discussed in terms of the consistency of results and factors related to variations in measured concentrations.

5.1 SYNTHESIS OF MONITORING RESULTS

5.1.1 ETS Contaminants

Average values for various measurement parameters related to ETS contaminants are summarized by monitoring location for both smoking and nonsmoking flights in Table 5-1. The results are segregated by particle-phase versus gas-phase measurements. Both Gravimetric and optical particle-phase measurements are given in the table. As noted in Section 4.2, there was greater uncertainty for the Gravimetric measurements due to relatively short monitoring durations for a number of flights. Further, as shown later in this section, the optical results were more strongly correlated with observed smoking rates than were the Gravimetric results. At the same time, however, the Gravimetric method is a well-established technique that has been successfully used for measuring average RSP levels in many other environments, whereas the optical method has had more limited use.

Peak RSP levels measured with optical sensors (Table 5-1) indicated even more pronounced differences between the boundary region and other no-smoking locations on smoking flights. The peak-to-average ratios for RSP were nearly identical in the smoking and boundary sections, and the ratios in these sections were higher than for the other no-smoking locations on smoking flights, The ratios in these other locations, however, were still higher than those for nonsmoking flights. Thus, tobacco smoking impacted all other sections of the aircraft in terms of peak RSP levels that were measured optically, and the effects were most pronounced in the boundary section (in addition to the distinct effects in the smoking section itself.

Effects of tobacco smoking, based on gas-phase measurements, were more discernible for nicotine than CO (Table 5-1). Beyond the marked increase in nicotine in the smoking section, the boundary region was most affected. Differences between nicotine levels for the remaining no- smoking locations and levels on nonsmoking flights were within the range of measurement uncertainty, but nicotine levels were more often above detection limits in the no-smoking locations than on nonsmoking flights. Further, cases where nicotine was detected on nonsmoking flights may reflect residual contamination from prior smoking flights. The only discernible effect for CO was in the smoking section itself. The lack of any other measurable effect may be due to the relatively low levels that prevailed, thereby increasing measurement uncertainty, coupled with background levels due in part to intrusion of ground-level emissions.

Results of statistical tests to contrast levels of ETS contaminants on smoking versus nonsmoking flights are given in Table 5-3. Comparisons were made for the monitoring locations common to both types of flights (i.e., smoking/rear and middle locations) using both parametric and non-parametric tests (the non-parametric tests do not require assumptions of normality or homogeneity of variances), For the smoking/rear location, levels of all six ETS measurement parameters were significantly higher (p C 0.05) on smoking than nonsmoking flights. For the middle location, levels were significantly higher for continuously monitored parameters (optical RSP and CO) but not for integrated-sample parameters (Gravimetric RSP and nicotine). The only discrepancy between the two types of statistical tests was for average optical RSP at the middle location, for which the parametric test was significant at the 0.05 level but the significance level for the non-parametric test was 0.09.

Results of statistical tests to contrast different sections within smoking flights are given in Table 5-4. Comparisons were made of the smoking versus boundary locations and the boundary versus middle locations, again using both parametric and non-parametric tests. Levels of all six ETS measurement parameters were significantly higher (p < 0.05) in the smoking than the boundary location. The boundary location was significantly higher than the middle location for all ETS tracers except C0. The only discrepancy between the two types of statistical tests was for nicotine at the boundary versus middle locations, for which the non-parametric test was significant at the 0.05 level whereas the parametric test had a significance level of 0.08. Thus, these tests indicate a clear difference between ETS levels in the smoking versus boundary sections and, to a lesser extent, between the boundary and middle sections (particularly for particle-phase constituents).

5.1.2 Carbon Dioxide and Pollutants

Average values for various measurement parameters related to pollutants are summarized by monitoring location for smoking and nonsmoking flights in Table 5-5. Most noteworthy are the relatively high CO2 concentrations, which exceeded 1,000 ppm (the ASHRAE level associated with satisfaction of comfort criteria) on 87 percent of the monitored flights. Further discussion of the C02 measurement results is given in Section 5.2.

Ozone levels were relatively low, averaging nearly an order of magnitude below the FAA 3-hour standard of 0.1 ppm and never exceeding the standard on monitored flights. Fungi levels were also very low, indicating little problem with sources attributable to the aircraft themselves. Monitoring of fungi levels earlier in the flight might have better reflected the extent of intrusion from ground-level outdoor sources, but this strategy was avoided to remain unobtrusive throughout most of the flight. Bacteria levels were slightly higher in the smoking sections; the measured bacteria levels need to be contrasted with measurements from other environments to obtain further insights concerning their relative significance.

5.2 FURTHER ANALYSIS OF MONITORING RESULTS

Additional analyses described and discussed in this section focus on (1) comparisons between two measurement methods for RSP, (2) RSP-to-nicotine ratios that were measured in this study, (3) factors related to variations in measured levels of ETS contaminants, (4) comparisons between measured and modeled C02 levels, and (5) factors related to variations in measured levels of pollutants.

5.2.1 Comparison of RSP Measurement Methods

As previously summarized in Table 5-1, the optical RSP results were similar to the Gravimetric results for the smoking section on smoking flights, whereas the Gravimetric results were higher at all other monitoring locations, both for smoking and nonsmoking flights. One possible explanation is that the optical method is less sensitive to RSP from sources other than ETS. As indicated by Ingebrethsen et al. (1988), the mass density of ETS particulate matter is lower than that of standard test aerosols such as Arizona Road Dust. Consequently, the MINIRAM optical sensors that were calibrated in an ETS-dominated chamber environment may have under-reported RSP concentrations when the prevailing average mass density was higher, as may have been the case on nonsmoking flights.

Further insights were obtained by modeling average RSP concentrations for the entire cabin as a single chamber. A dynamic model for cabin air quality can be stated as follows:

where

Under steady-state conditions (i.e., dCin/dt=0), the above equation reduces to:

Modeling was performed using nominal fresh-air intake rates and recirculation rates given in Section 4.0, smoking rates estimated from technician observations, and an emission rate of 26,000 ug per cigarette (National Research Council 1986). An outdoor concentration of zero and a filter efficiency of 90 percent were assumed. Measured cabin-wide RSP concentrations were determined by weighting the monitoring results from each of the four measurement locations in proportion to the number of rows associated with each. Modeling was restricted to domestic smoking flights due to uncertainties concerning smoking rates in the business-class section of international flights.

Predicted and measured RSP concentrations for the two different methods are shown in Figure 5-3, together with the line of best fit for each. Predicted RSP values were 50 to 100 percent higher than measured values (a similar outcome was obtained in modeling results from the chamber tests used for calibration. The over-prediction may be due in part to the fact that a term for particle deposition was not included in the model due to uncertainty concerning an appropriate value for this parameter.

The correspondence between predicted and measured values was better for optical measurements (correlation coefficient of 0.65) than for Gravimetric measurements (correlation coefficient of 0.31). In addition, the average difference between predicted and measured values was lower for optical (55 percent) than Gravimetric (64 percent) measurements. The y-intercepts for regression of measured against predicted values indicate measurement results that can be expected in the absence of smoking. The larger intercept for Gravimetric results (40.2 ug/m3) than for the optical results (18.7 ug/m3) may reflect a higher sensitivity of the Gravimetric method to non-ETS sources of RSP. The intercept for the optical measurements is consistent with the optical results that were obtained during periods prior to smoking, which averaged near 18 ug/m3.

Standard errors for the intercept and regression coefficients are given in parentheses in the above equation. For Gravimetric measurements, there was only one significant predictor (smoking rate), which explained 11 percent of the variance; the following regression equation was obtained:

A final comparison was made between the two methods based on five Northwest Airlines nonsmoking flights that were monitored during the study. These flights were of relatively longer duration and should have had little or no residual ETS levels due to Northwest's no-smoking policy for all flights within the continental United States. Both the Gravimetric and optical results (Table 5-6) for this subset of flights were somewhat lower, based on the average of the two monitored locations, than for all nonsmoking flights as a whole. The Gravimetric results, however, were quite different at the two locations and had relatively high standard deviations, reflecting measurement uncertainty.

The above analysis and discussion indicate that the RSP results obtained by optical methods are more internally consistent and predictable than the results obtained by Gravimetric methods. Thus, there are indications that optical measurements may be more sensitive to ETS than Gravimetric measurements and the level of uncertainty associated with the Gravimetric measurements may be high for cases of low airborne RSP concentrations and short sampling durations, However, as stated previously, the average of the RSP measurement results from the two methods was used for risk assessment purposes.

5.2.2 Ratios Between RSP and Nicotine

Based on a subset of 57 smoking flights with complete results for nicotine and RSP by both measurements methods, the average nicotine concentration in the smoking section was 13.0 ug/m3. Average RSP concentrations in this section were 181.7 ug/m3 by the optical method and 182.6 Ug/m3 by the Gravimetric method. These aggregate results imply an RSP-to-nicotine ratio near 14 for the smoking section. Netting out RSP levels not due to ETS (i.e., 19 ug/m3 for optical results and 40 ug/m3 for Gravimetric results) would result in a ratio between 11.0 and 12.5. This range of ratios is consistent, for example, with the 11:1 ratio assumed by Repace and Lowrey (1988) in developing an indoor concentration model for nicotine.

RSP-to-nicotine ratios calculated for each flight, and then averaged across flights, would be misleading because very large ratios would be obtained for flights with low nicotine levels. Instead, the nicotine results for the smoking section on each flight were regressed on RSP results for the same monitoring location. The following equations were obtained:

Nicotine = -2.38 + 0.084 * Optical RSP (R2 = 4.36)

Nicotine = 0.12 + 0.070 * Gravimetric RSP (R2 = 0.24)

The inverse of the regression coefficients imply an RSP-to-nicotine ratio between 11.9 and 14.3, consistent with the ratios based on aggregate data. The equations also imply that no nicotine would be detectable until the optical measurement reaches near 30 ug/m3, whereas some nicotine would be detectable for Gravimetric results near zero. As indicated by the R2 values shown above and the scatter about the regression lines shown in Figure 5-4, the nicotine measurements were more strongly correlated with optical than with Gravimetric measurements.

The RSP-to-nicotine ratios for the boundary section, calculated from aggregate data presented earlier in Table 5-1, were much higher (150 to 260). These much higher ratios for the boundary section indicate that nicotine is being preferentially removed (relative to RSP) before or as ETS leaves the smoking section. RSP is subject to some removal through deposition, whereas nicotine can react with various types of materials including clothing, seats, and carpeting on the cabin floor. Netting out RSP levels not due to ETS would result in RSP-to-nicotine ratios between 80 and 105 for the boundary section.

RSP-to-nicotine ratios higher than those observed in the smoking section have been measured by some researchers. Nicotine levels measured in this study were generally lower than those measured in the boundary section as part of a smaller field study reported by Mattson et al. (1989). However, in that study the higher nicotine values were obtained on a wide-body flight for passengers seated in aisle seats adjacent to the smoking section. Because the middle and side sections of wide-body aircraft are offset by about half the width of a seat, passengers in the boundary section sitting in outer seats could easily be exposed to ETS levels rivaling those in the smoking section. Thus, the RSP-to-nicotine ratios measured in the boundary section during this study, although relatively high, are not implausible.

5.2.3 Factors Related to Variations in ETS Concentrations

Nicotine measurement results for each monitoring location on smoking flights are summarized in Table 5-7 in relation to four factors -- type of aircraft, air recirculation, air exchange rate, and cigarette smoking rate. Compared to aircraft without recirculation, aircraft with recirculation had lower levels in the smoking section coupled with somewhat higher levels in the no-smoking section. Levels in all sections were lower on narrow-body than wide-body aircraft. Levels in the smoking section were strongly related to smoking rates. Air exchange rates appear to have had little impact.

RSP measurement results are summarized in relation to the same factors in Table 5-8 (for Gravimetric measurements) and in Table 5-9 (for optical measurements). The smoking rate had the greatest impact, in this case influencing levels in the boundary section in addition to those in the smoking section. The effects of aircraft type, air recirculation, and air exchange rate were less consistent, but levels in the smoking section were lower on narrow-body aircraft and on flights with air recirculation. More rapid removal of ETS contaminants from the smoking section, and some redistribution to other sections, could be occurring due to recirculation.

CO measurement results are summarized in relation to the same factors in Table 5-10. The only discernable pattern for CO was that of higher levels in the smoking section when smoking rates were higher, particularly at the upper extreme (i.e., 30 or more cigarettes per hour).

Measurement results for nicotine, RSP, and CO in the boundary section are summarized in Table 5-11 in relation to the technician's proximity to the smoking section. There was no discernable pattern for gas-phase tracers (nicotine and CO), but both average and peak RSP levels were highest when the technician was located in the row immediately bordering on the smoking section.

5.2.4 Modeling of CO2 Concentrations

A single-chamber steady-state model similar to that described previously for RSP was used to model average C02 concentrations for all study flights. Because the filters in aircraft with recirculation are not currently designed to remove C02, the equation previously used can be simplified to the following:

Cin = Cout + S/F

where Cin and Cout refer to indoor and outdoor C02 concentrations, S indicates the emission rate, and F indicates the fresh-air intake rate. Nominal air exchange rates were used for the model together with an assumed outdoor concentration of 330 ppm and an emission rate of 0.3 1/min (16,000 ml/h) per passenger (ASHRAE 1989). As illustrated in Figure 5-5, a reasonable association between predicted and measured values was obtained (r = 0.55). However, measured values (averaging 1,609 ppm) were nearly a factor of two higher than those predicted by the model (average of 841 ppm). The modeled values shown in the figure do not include emissions from the flight and cabin crew members, but adding emissions from 10 additional persons to account for the crew would increase the modeled values only to 888 ppm.

There are four possible explanations for the discrepancy between measured and modeled values: (1) the measurements may have a positive bias, due to proximity to the breathing zone or the measurement device used, (2) there may be short-circuiting between the supply and exhaust points within the aircraft, resulting in poor ventilation efficiency, (3) the nominal air exchange rates used for modeling may be higher than prevailing rates during the monitored flights, or (4) C02 emission rates may be higher than those used in the model. One study (Balvantz et al. 1982)has suggested that C02 exhalation rates in airliner cabins could be as high as 0.5 1/min per passenger due to factors such as environmental stress and food/alcohol consumption. With this higher emission rate, average measurement values still exceeded average modeled values (1,180 ppm) by a third. Further measurements at different heights in the aircraft, with more sophisticated monitoring devices, are needed to fully resolve the issue. However, even if the monitoring results were biased high by a factor of two, there would still be a substantial number of monitored flights (about 24 percent) exceeding 1,000 ppm C02.

5.2.5 Factors Related to Variations in C02 and Pollutant Concentrations

Average C02 levels measured at smoking and middle seats on all smoking flights (domestic plus international) are summarized in Table 5-12 in relation to type of aircraft, air recirculation, air exchange rate, and load factor (i.e., percent of seating capacity filled by passengers). Higher C02 levels were associated with narrow-body aircraft, aircraft with recirculation, lower air exchange rates, and higher load factors, with load factor having the strongest association. The relationships with most of these factors were in opposite directions for bacteria versus fungi (Tables 5-13 and 5-14); bacteria levels were somewhat higher on wide-body aircraft, aircraft with recirculation, and flights with lower nominal air exchange rates, whereas fungi levels were somewhat lower in each of these cases. Bacteria and fungi levels both were generally higher in the presence of higher load factors"

5.3 REFERENCES

ASHRAE. 1989. Ventilation for Acceptable Indoor Air Quality . ASHRAE Standard 62-198 . American Society of Heating, Refrigerating, and Air Conditioning Engineers, Atlanta, GA

Balvarz, Jr., S. Bowman, A. Fobelets, T. Lee, K. Papamichael, and R. Yoder. 1982. Examination of the Cabin Environment of Commercial Aircraft. Iowa State University, Ames, A

Ingebrethsen, B.J., 0. Heavner, A. Angel, J. Conner, T. Steichen and C. Green. 1988. "A Comparative Study of Environmental Tobacco Smoke Particulate Mass Measurements in an Environmental Chamber," JAPCA 38(4):413-417

Mattson, M., G. Boyd, D. Byar, C. Brown, J. Callahan, D. Corle, J. Cullen, J. Greenblatt, N. Haley, K. Hammond, J. Lewtas and W. Reeves. 1989. "Passive Smoking on Commercial Airline Flights," J. Amer. Med. Assoc., 261(6):867-872

National Research Council. 1986. Environmental Tobacco Smoke: Measuring Exposures and Assessing Health Effects. National Academy Press. Washington, D.C.

Repace, J.L. 1987. "Indoor Concentrations of Environmental Tobacco Smoke: Field Surveys," in Environmental Carcinogens Methods of Analysis and Exposure Measurement: Volume 9--Passive Smoking. Ed. I.K. O'Neil, K. Brunneman, B. Dodet, and D. Hoffman, International Agency for Research on Cancer, lyon, France.

Repace, J.L., and A. Lowrey. 1988. "Indoor Air Modeling of the Aerosol from Environmental Tobacco Smoke." Presented at the 1988 Annual Meeting of the American Association for Aerosol Research, Chapel Hill, NC

DOT English | Summary | Chapter 2 | Chapter 3 | Chapter 5 | Chapter 6 | Chapter 7 | Chapter 8 | Chapter 9 | Chapter 10