Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

A model-based approach investigating killer whale (Orcinus orca) exposure to marine vessel engine exhaust Lachmuth, Cara Leah 2008

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
24-ubc_2009_spring_lachmuth_cara_leah.pdf [ 3.45MB ]
Metadata
JSON: 24-1.0067192.json
JSON-LD: 24-1.0067192-ld.json
RDF/XML (Pretty): 24-1.0067192-rdf.xml
RDF/JSON: 24-1.0067192-rdf.json
Turtle: 24-1.0067192-turtle.txt
N-Triples: 24-1.0067192-rdf-ntriples.txt
Original Record: 24-1.0067192-source.json
Full Text
24-1.0067192-fulltext.txt
Citation
24-1.0067192.ris

Full Text

A MODEL-BASED APPROACH INVESTIGATNG KILLER WHALE (ORCINUS OR CA) EXPOSURE TO MARINE VESSEL ENGINE EXHAUST  by Cara Leah Lacbmuth B.Sc., The University of Calgary, 2000  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE  in  THE FACULTY OF GRADUATE STUDIES (Zoology)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  December 2008  © Cam Leah Lachmuth, 2008  ABSTRACT The summer habitat of the southern resident population of killer whales (Orcinus orca) in British Columbia and Washington experiences heavy traffic by vessels involved in whale-watching, sport fishing, other recreational activities, and shipping. Behavioural changes caused by vessel proximity and the impacts of vessel noise have been previously documented, but this is the first study to assess direct impacts of air pollutant emissions from vessel traffic. The concentration and composition of air pollutants from whale-watching vessels that southern resident killer whales are exposed to during the peak tourist season were estimated, as were the health impacts of the exposure. Specifically, the study a) estimated the output of airborne pollutants from the whale-watching fleet based on emissions data from regulatory agencies, b) estimated the vertical dispersion of such pollutants based on air stability data collected in the field and from climatological sources, c) used a dispersion model incorporating data on whale, vessel, and atmospheric behaviour to estimate exposure, and d) examined the likely physiological consequences of this exposure based on allometric extrapolation of data from other mammalian species. The results of these exercises indicate that the current whale-watching guidelines are usually effective in limiting pollutant exposure to levels just at or below those at which adverse health effects would be expected in killer whales. However, under ‘worst-case’ conditions and even under certain ‘average-case’ conditions the pollutant levels are much higher than those predicted to cause adverse health effects. With this information, recommendations are made for further studies that would fill in missing information, and increase confidence in the models, and the predicted impact on the southern resident killer whales. Recommendations for limiting killer whale exposure to air pollutants are also provided.  11  TABLE OF CONTENTS Abstract  .  ii  Table of Contents  iii  List of Tables  vi  List of Figures  vii ix  List of Abbreviations  xiii  Acknowledgements 1  INTRODUCTION  1  1.1 INTRODUCTION  1  1.1.1  Frequency and Duration of Whale-Watching  3  1.1.2  Whale-Watching Guidelines  6  1.1.3  Study Site  7 9  1.2 METHODS  11  1.3 REFERENCES 2  EXHAUST EMISSION DISPERSION IN THE MARINE ATMOSPHERIC 14  BOUNDARY LAYER 2.1.1  14  INTRODUCTION  2.1.1.1 Airshed Description and Air Pollution Sources  14  2.1.1.2 Air Quality Objectives and Standards  21  2.1.1.3 Ambient Air Quality  23  2.1.1.4 Atmospheric Mixing and Boundary Layer Stability  27  2.1.2  MARJNE ATMOSPHERIC BOUNDARY LAYER MEASUREMENTS.. .31  2.1.3  MARINE ATMOSPHERIC BOUNDARY LAYER DATA  .  33  111  2.1.4  MARINE ATMOSPHERIC BOUNDARY LAYER CONCLUSIONS  2.2.1  MODELING DISPERSION IN THE MARINE ATMOSPHERIC  40  42  BOUNDARY LAYER  3  2.2.1.1 Emission Dispersion Models  44  2.2.1.2 Marine Engines, Fuel, and Emissions  46  2.2.1.3 Wet and Dry Exhaust Systems  48  2.2.1.4 The Whale-Watching Fleet  50  2.2.1.5 Engine Emission Factors  51  2.2.2  NETLOGO DISPERSION MODEL  54  2.2.3  NETLOGO DISPERSION MODEL RESULTS  64  2.2.3.1 Results of the Sensitivity Analysis  65  2.2.3.2 Results of the Average-Case Trials  69  2.2.3.3 Results of the Worst-Case Trials  72  2.2.4  DISPERSION MODEL CONCLUSIONS  73  2.2.5  REFERENCES  77  PHYSIOLOGICAL EFFECTS ASSOCIATED WITH EXPOSURE TO AIR POLLUTION  84  3.1 INTRODUCTION  84  3.1.1  Compounds in Diesel and Gasoline Exhaust  86  3.1.2  Health Effects from Exposure to Diesel and Gasoline Exhaust  88  3.1.3  Retention and Clearance of Air Pollutants in the Lungs  90  3.1.4  Killer Whale Respiratory Anatomy and Physiology  95  3.1.5  The Effects of Diving and Breath Holding  99  iv  3.1.6  Physiological Models Used to Estimate Internal Pollutant Dose and 102  Health Effects 3.1.7  Allometric Scaling to Estimate Internal Pollutant Dose and Health 104  Effects  4  3.2 METHODS  107  3.3RESULTS  110  3.4 CONCLUSIONS  113  3.5 REFERENCES  116  CONCLUSIONS AND FURTHER STUDIES  125  4.1 GENERAL CONCLUSIONS  125  4.2 UNCERTAINTY AND ASSUMPTIONS IN THE MODELS  130  4.3 FUTURE RESEARCH  132  4.4 REFERENCES  135 138  APPENDICES Appendix A: Be Whale Wise Guidelines  138  Appendix B: Air Pollutant Emissions  139  Appendix C: Air Quality Standards  141  Appendix D: Alternative Fuels and Fuel Additives  142  Appendix E: Programming Code for the NetLogo Dispersion Model  144  Appendix F: Sensitivity Analysis Results from the Dispersion Model  149  Appendix 0: Classes of Compounds in Diesel Exhaust  156  Appendix H: Health Effects from Exposure to Air Pollutants in Exhaust  157  Appendix I: University of British Columbia Animal Care Certificate  168  V  LIST OF TABLES 25 and 03.22 Table 2.1 The Canada-Wide Standards for PM 22  Table 2.2 The Metro Vancouver Air Quality Objectives  Table 2.3 Monthly average ambient air pollutant concentrations (maximums in parentheses) 24  at the Christopher Point, BC air quality monitoring station from 2005-2007 Table 2.4 Times of the year when ambient air pollutants reach their maximum and  25  minimum concentrations in the Georgia Basin Airshed Table 2.5 USEPA non-road model emission factors for pre and post-2006 model  recreational marine diesel engines with power ratings less than 175 to 300 hp. .52 . .  Table 2.6 USEPA non-road model emission factors for pre and post-2006 model 52  recreational marine gasoline engines  Table 2.7 Air pollutant multiplication factors for different marine engine configurations.. .57 61  Table 2.8 Variables in the dispersion model with their default and range of values 2 per kg body mass that male and female killer whales Table 3.1 The dose of CO and NO and humans receive during average-case, worst-case, 1 -hour, and 8-hour  110  exposures 2 (1-hour) toxicity values (Am) for male and Table 3.2 CO (1-hour and 8-hour) and NO  111  female killer whales 2 per kg body mass for male and female killer Table 3.3 Toxicity doses of CO and NO  111  whales and humans, using toxicity values (A) 2 per kg body mass that male and female killer Table 3.4 Total dose of CO and NO  whales are estimated to receive under average-case or worst-case whale-watching 112  conditions  vi  LIST OF FIGURES Figure 1.1 The number of vessels accompanying southern resident killer whale pods per 5  month as measured from the Soundwatch vessel 1998-2005  Figure 1.2 Map of the summer habitat of the southern resident killer whales, and inset of the 8  regional location map  16  Figure 2.1 The Georgia Basin and Lower Fraser Valley Airsheds  Figure 2.2 Percentage of air pollutants attributed to all marine vessels in the LFV Airshed in 17  the year 2000 emission inventory  Figure 2.3 Air pollutant contributions by vessel category in BC outside of Metro Vancouver 19  and FVRD in the year 2000 Figure 2.4 Graphs of temperature change with height showing (a) an unstable atmosphere  30  and (b) a stable atmosphere Figure 2.5 Profile view of the thermocouple setup on the zodiac  32  Figure 2.6 Map of southeast Vancouver Island, BC  33  Figure 2.7 Results from the first ten 1 5-minute trials at Oak Bay, BC, demonstrating that the 34  average temperature increased with height above the water Figure 2.8 Plot of the nine trials at Oak Bay, BC, that had deviations from temperature  34  increasing with height  Figure 2.9 Air-sea surface temperature difference (TairTsea) for all 411 5-mm trials, by the 35  time of day the trial started Figure 2.10 Average monthly air-sea surface temperature difference (Tair-Tsea) at Race  36  Rocks, BC in 2007, with standard error of the mean bars Figure 2.11 Average monthly air-sea surface temperature difference (TairTsea) at Race  vii  37  Rocks from 2002-2006, with standard error of the mean bars Figure 2.12 Average monthly air-sea surface temperature difference (Taji-Tsea) at Halibut  39  Bank Buoy from 2002-2005, with standard error of the mean bars  Figure 2.13 Average monthly air-sea surface temperature difference (TairTsea) at Hem Bank 40  Buoy from 2004-2007, with standard error of the mean bars  56  Figure 2.14 Image of the NetLogo interface 2 concentration as a function of wind speed and angle under Figure 2.15 CO and NO  70  average-case whale-watching conditions  2 concentration as a function of the vertical mixing height and wind Figure 2.16 CO and NO 71  angle under average-case whale-watching conditions 2 concentration as a function of the wind speed and angle under Figure 2.17 CO and NO  73  worst-case whale-watching conditions  viii  LIST OF ABBREVIATIONS ACSM  American College of Sports Medicine  AQO  Air Quality Objectives Toxicity value for a test animal  ATSDR  Agency for Toxic Substances and Disease Registry  A  Toxicity value for a wildlife species  b  Allometric scaling factor  BC  British Columbia, Canada  BCLA  British Columbia Lung Association  BCME  British Columbia Ministry of Environment  BCPHO  British Columbia Provincial Health Officer  1 bm  Breaths per minute  CAPMoN  Canadian Air and Precipitation Monitoring Network  °C  Degrees Celsius  °C m 1  Degrees Celsius per meter  CO  Carbon monoxide  2 CO  Carbon dioxide  CO  Carbon oxides  CWS  Canada-Wide Standards  DALR  Dry Adiabatic Lapse Rate  DFO  Department of Fisheries and Oceans Canada  DNA  Deoxyribonucleic acid  EC  Environment Canada  ix  ELR  Environmental Lapse Rate  FR  Breathing frequency  FVRD  Fraser Valley Regional District  1 g hp’ hf  Grams per horsepower hour  GVRD  Greater Vancouver Regional District  HC  Hydrocarbon  HEI  Health Effects Institute  3 HNO  Nitric Acid  hp  Horsepower  IPCS  International Programme on Chemical Safety  kg  Kilogram  km  Kilometer  LFV  Lower Fraser Valley  L b’  Litre per breath  L m’  Litre per minute  L s’  Litre per second  m  Meter  Mb  Body mass  mg  Milligram  3 mg m  Milligram per meter cubed  mg s 1  Milligram per second  m  Meter per second  s  MTBE  Methyl Tertiary Butyl Ether  x  MV  Metro Vancouver  NESCAUM  Northeast States for Coordinated Air Use Management  4 NH  Ammonium  nrn  Nanometer  NOAA  National Ocean and Atmospheric Association  NO  Nitrogen oxide  2 NO  Nitrogen dioxide  3 NO  Nitrate  NO  Nitrogen oxides  02  Oxygen  03  Ozone  PAH  Polycyclic Aromatic Hydrocarbon  PBPK  Physiologically-Based Pharmacokinetic model  PCB  Polychlorinated biphenyl  PM  Particulate matter  10 PM  Particulate matter smaller than 10 microns in diameter  25 PM  Particulate matter smaller than 2.5 microns in diameter  POP  Persistent Organic Pollutant  Ri\4R  Resting Metabolic Rate  RPM  Revolutions Per Minute  RRER  Race Rocks Ecological Reserve  SARA  Species At Risk Act  SEM  Standard Error of the Mean  xi  2 SO  Sulfur dioxide  2 4 S0  Sulfate ion  SO,  Sulfur oxides  SRKW  Southern Resident Killer Whale  SST  Sea Surface Temperature  TAF  Transient Adjustment Factor  TLC  Total Lung Capacity  UF  Uncertainty Factor  ig m 3  Micrograms per meter cubed  pm  Micrometer  USEPA  United States Environmental Protection Agency  VA  Alveolar volume  Vc  Vital capacity Dead space volume  Veff  Effective ventilation Minute volume of respiration  VOC  Volatile Organic Compound  VT  Tidal volume  WA  Washington, USA  WDFW  Washington Department of Fish and Wildlife  WHO  World Health Organization  WLAP  British Columbia Ministry of Water, Land, and Air Protection  WWOAN  Whale Watch Operators Association Northwest  xii  ACKNOWLEDGEMENTS I have tremendous gratitude for the countless people who helped me during the course of this thesis, and I offer my sincere apologies to anyone I missed in this list. First off I would like to thank my supervisors Dr. Lance Barrett-Lennard and Dr. Bill Milsom, for their expertise and support. I could not have done this project without the guidance and assistance from my committee members, Dr. Douw Steyn and Dr. Peter Ross. I am extremely grateful to Dr. Rik Blok, Alistair Blanchford, and Atef Abdelkefi for computer programming support, and their patience with me as I learned. Thanks go to Doug Sandilands for providing wonderful maps, and David Bain for data and advice. I thank Mark Malleson for the generous use of his zodiac to collect temperature data in the field, and Anna Lachmuth and Bo Garrett for being excellent field assistants. Thanks go to Nik Dedeluk and Straitwatch for taking me out on the water to collect data. I would also like to thank past and present members of the Milsom Lab at UBC (Cosima Ciuhandu, Angelina Fong, Charissa Fung, Stella Lee, Catalina Reyes, Barb Gajda, Emily Coolidge, Cohn Sanders, and Graham Scott) and the Cetacean Research Lab at the Aquarium (Charissa Fung, Katie Kuker, Valeria Vergara, Doug Sandilands, and Judy McVeigh) for their inspiration, feedback, and friendship. Of course this work could not have been done without funding and I am especially thankful to the Vancouver Aquarium Killer Whale Adoption Program, the Zoology Department at UBC, the Dean Fisher Memorial Scholarship in Zoology at UBC, and the Michael A. Bigg Scientist of the Future Award from the Vancouver Aquarium. I would also like to give special thanks to my family and friends for their remarkable support, encouragement, and ability to keep me balanced.  XIII  1 INTRODUCTION  1.1 INTRODUCTION The population of killer whales (Orcinus orca) known as the southern residents (SRKW) inhabits the waters off southern Vancouver Island, BC during the summer months, and has been studied extensively since the 1970’s. The population experiences intense whalewatching pressure from May to September every year, and are followed on average by 20 vessels for 12-hours a day (Bain, 2002; Baird, 2001; Erbe, 2002; Koski, 2006; Osborne, et al. 2002; 1999). The population is socially, culturally, and genetically distinct from other populations (Barrett-Lennard, 2000) and consists of three maternal subgroups called the J, K, and L pods (Bain, 2002). In 1995 there were 99 whales in the SRKW community, but poor survival and fecundity in 1996 initiated a decline in the population that lasted until 2001 with 81 whales remaining (Krahn et al., 2004; 2002). Because of the population’s small size and genetic isolation from other killer whale groups, the SRKWs were listed as Endangered under the Canadian Species at Risk Act (SARA) in 2001, and the United States Endangered Species Act in 2006 (EC, 2005; NOAA, 2005). Since 2001 the population has fluctuated and as of October 2008, the population numbered 83 whales (Orca Network, 2008). Three anthropogenic factors have been identified as possible causes of the population’s decline: decreased food availability due to the decline of salmon stocks (their primary food source), exposure to toxic chemicals such as polychiorinated biphenyls (PCBs), and vessel disturbance (Bain, 2002; Ross, 2006). The first two threats will require long term solutions to re-establish salmon stocks and phase out toxins; however, the third threat of vessel disturbance is one that could be ameliorated  relatively quickly and easily to minimize one facet of the negative impact humans have on this population. Efforts to quantify vessel disturbance on killer whales has been primarily limited to studying the killer whale’s behavioural responses. Documented short-term behavioural responses to whale-watching vessels are increased dive duration, increased swimming speed, and erratic changes in direction of travel (Jelinski et al., 2002; Kruse, 1991; Williams et al., 2002). It has also been found that vessel noise can be harmful to whales by potentially causing temporary threshold shifts in hearing, permanent hearing loss with prolonged exposure, and by masking their calls and reducing foraging success (Erbe, 2002). While this research demonstrates that whales often adjust their behaviour when vessels are present, it is difficult for researchers to quantify the significance and consequence of those adjustments. Are they seemingly insignificant responses to a mildly irritating stimulus, or are they significant responses to relatively severe disturbances? The present study investigates moredirectly-measurable effects of whale-watching, exposure to exhaust gases, and examines the potential physiological impact of this exposure. The SRKWs are exposed to increasing levels of contaminants in the air they breathe, the water they inhabit, and the food they eat. Recent studies have found that they are some of the most contaminated mammals in the world, and the concentration of so-called persistent organic pollutants (POPs) in their tissues is three times higher than levels that cause damage to the immune and endocrine systems of harbour seals (Ross et al., 2000; Ross, 2006). Toxicological research has focused on the organochlorine family of chemicals, which are found in significantly higher concentrations in SRKWs than St. Lawrence belugas, which were long considered to be the most contaminated marine mammals in the world (Colborn &  2  Smolen, 1996; Ross et al., 2000). While the majority of POPs originate from the diet of killer whales, it has been speculated that chemicals such as unburned fuel and exhaust may be adding to the killer whale’s toxin load through the creation of combustion by-products such as dioxins, furans, and complex hydrocarbons (Bain et al., 2006). However, the potential acute and/or chronic risks to the population from these pollutants and other environmental toxins have not been investigated to date. Air pollutants are often complex mixtures of numerous gases (e.g. hydrocarbons, nitrogen oxides, sulfur oxides, ozone) and particulates, and they predominantly pose a chronic health risk. Air pollutants most commonly affect the lungs, but they also have neurologic, cardiac, gastrointestinal, renal, hematologic, as well as skin pathology effects (BCPHO, 2004; HET, 1999). Marine engines produce more air pollutants per kg of fuel burned than automobile engines, because they usually do not have after-treatment of the exhaust or equivalent pollution control devices. The proximate air pollutants from whale-watching vessels are not the only source of airborne contaminants, however, as the SRKWs are also exposed to regional urban and industrial air pollutants from the large metropolitan centres of Vancouver, Victoria, and Seattle.  1.1.1 Frequency and Duration of Whale-Watching The frequency and duration of whale-watching activities must be quantified in order to determine the concentration of air pollutants the SRKWs are potentially inhaling. Commercial whale-watching operations in Canada and the U.S. were basically non-existent prior to 1976 (Koski, 2006), but by the summer of 2005, 39 whale-watching companies operated 74 vessels in WA and BC, and focused primarily on the SRKWs (Koski, 2006). Of  3  the 74 vessels, Canadian companies owned 55, and 19 were American (Koski, 2006). Canadian commercial whale-watching companies generally use small outboard motor vessels that operate at high revolutions per minute (RPM), while American companies generally use larger inboard motor vessels that operate at a low RPM (Bain, 2002). The vessels used range from 7 m long open boats carrying 6-16 people to 30 m covered vessels carrying up to 280 people (Wiles, 2004). The vessels make two to six trips a day, based on demand (Wiles, 2004). In addition to commercial vessels, large numbers of recreational boaters participate in opportunistic whale-watching approximately 64% of the vessels observing the whales are -  commercially operated, while the rest are privately owned (Osborne et al., 2002). During the summer, commercial whale-watchers and researchers often begin viewing whales at 6:00 a.m. However, the majority view from 9:00 a.m. to 9:00 p.m., with the greatest density occurring between 10:00 a.m. and 5:00 p.m. (Bain, 2002). The commercial season usually starts late April and ends early October, with the peak occurring from May to September (Figure 1.1); if whales are present, some whale-watching will occur throughout the winter and early spring (Bain, 2002). Private vessels engaged in whale-watching have similar seasonal and daily patterns as commercial whale-watchers (Bain, 2002). During peak summer months, the mean number of whale-watching vessels has increased from five in 1990, to 18-26 vessels within 800 m of the whales from 1996-2002 (Bain, 2002; Baird, 2001; Erbe, 2002; Koski, 2006; Osborne et al., 2002; 1999). In total, the SRKWs are exposed to whale-watching vessels approximately 12-hours a day for six months of the year (Trites & Bain, 2000).  4  I 15 I  -a E  z 5.  0—  May  I  I  I  I  June  July  August  September  Month  Figure 1.1: The number of vessels accompanying southern resident killer whale pods per month as measured from the Soundwatch vessel 1998-2005 (Koski, 2006). The bottom of th 25 percentile, the mid line indicates the median, and the top of the box the box indicates th 75 percentile. indicates the From 1998-2002, the annual maximum number of vessels following the SRKW population ranged from 72-120 vessels per day, with the majority being private vessels rather than commercial whale-watching vessels (Wiles, 2004). This annual maximum pales in comparison to the autumn of 1997, when up to 500 private vessels were counted each weekend observing a group of killer whales that remained in Dyes Inlet, WA for one month (Wiles, 2004). Bain et al. (2006) conducted a theodolite study from San Juan Island, WA on SRKW exposure to vessels from 2003-2005. They found that approximately 25% of the killer whale’s time was spent with at least one vessel closer than 100 m, over 50% of their time was spent with vessels within 400 m, over 75% of their time was spent with vessels within  5  1000 m, and there were vessels at further distances on almost 100% of scans (Bain et al., 2006). A study by Jelinski et al. (2002) conducted in Johnstone Strait, BC, found that motorized vessels remained with northern resident killer whales for a longer period of time than non-motorized vessels. The average length of time a charter vessel remained with a group of killer whales was 73 minutes, compared with two minutes for a kayak. Head-on approaches to the whales occurred more often with motorized vessels and there was evidence of aggressive viewing among both motorized and non-motorized vessels (Jelinski et al., 2002). This situation is probably similar for motorized and non-motorized vessels viewing the SRKWs.  1.1.2 Whale-Watching Guidelines The Department of Fisheries and Oceans Canada in conjunction with the National Ocean and Atmospheric Administration (NOAA) created voluntary Be Whale Wise Guidelines (DFO, 2008) for whale-watching vessels to reduce disturbance and manage vessel traffic (Appendix A). The Whale Watch Operators Association Northwest (WWOAN) created more comprehensive guidelines for commercial vessel operators observing the SRKWs, known as the Best Practices Guidelines (WWOAN, 2008). Vessel operators are strongly encouraged to follow the guidelines, however, enforcement is limited. Several charges have been laid in Canada under the Marine Mammal Regulations of the Fisheries Act, with successful prosecutions of both recreational and whale-watching operators being charged. The guidelines recommend that vessels approach whales no closer than 100 m, slow their speed to less than seven knots (3.6 m s’) when within 400 m of whales, and that they travel parallel to the whales rather than in front or behind (DFO, 2008).  6  Commercial whale-watch operators have been quick to adopt the guidelines, however, private vessels are often unaware of them, and unfortunately incidents of non compliance with the guidelines are common (Koski, 2006; Osborne et al., 1999). Additionally, compliance with the guidelines strongly depends on whether or not monitoring and enforcement agencies are on the water (Smith & Bain, 2002). The most frequent incidents are: vessels stopping in the path of whales, vessels under power while inshore of whales, and vessels under power within 100 m of whales (Koski, 2006). Osborne et al. (1999) reported that vessels violated the guidelines in the Haro Strait area 400 times in 1998, and 560 times in 1999, while in 2005 Koski (2006) reported 957 violations. Of the 957 violations in 2005, 10% were from vessels within 100 m of whales (Koski, 2006). In the summer of 2007, the Be Whale Wise Guidelines became law in WA, which has resulted in stronger enforcement in American waters (WDFW, 2008).  1.1.3 Study Site When assessing air quality, an important factor to consider is the occurrence of atmospheric conditions conducive to air pollutant accumulation. The time period of concern is the peak whale-watching season, and the area of concern is the summer habitat of the SRKWs, mainly Haro Strait, Juan de Fuca Strait, and Boundary Pass (Figure 1.2).  7  w  124W  4cf N  4g,  Southern Gull ISlands  11 San Jwrn Islands  A  o .  *OakBay  F  *Hein Bank Buoy  * Christopher Point * Rate Rocks Paiftc  Ocean O  2  4  I II  N  48 N  Figure 1.2: Map of the summer habitat of the southern resident killer whales, and inset of the regional location map. Stars indicate locations mentioned in the text. Calm weather conditions are particularly important in air quality assessment because air pollutant concentrations are greatest under these conditions (EC, 2004). In coastal BC and WA the highest median wind speeds occur in December, the lowest in August (Vingarzan & Thomson, 2004). The Gulf and San Juan Islands commonly experience light winds (Lange, 1998) due to merging airfiows from the Straits of Georgia and Juan de Fuca, and the “wake effect” from Vancouver Island and the Gulf and San Juan Islands (Brook et al., 2004). This has been called the Wake-Induced Stagnation Effect, and it is instrumental in the accumulation and photochemical evolution of air pollutants from Vancouver, Victoria, the Lower Fraser Valley, Whatcom County (WA), and marine vessels. Storms and calm  8  conditions can occur at any time of year along BC’s coast, however, the spring and fall have the most active weather events, whereas winter and summer are calmer. High atmospheric pressure dominates during the summer months, and this reduces airflow to local circulations and produces temperature inversions that can last several days (EC, 2004). Temperature inversions occur when a layer of cold air is trapped near the surface by a layer of warmer air on top, and produces stable atmospheric conditions. The worst air quality events occur during the sun-imer because stagnant air gets trapped close to the surface over large areas of the airshed. The timing of these elevated air pollution events overlap with the commercial whale-watching season. Sea breezes occur most frequently in the summer months (Steyn & Faulkner, 1986), however, they do not generally disperse air pollution in an efficient manner because their speeds are usually less than four m  s’,  and they are closed circulation systems that  experience reversals in the direction of flow diurnally (EC, 2004). This results in limited air exchange, with air pollutants moving back and forth from land to sea in a contained volume of air with the pollutant load increasing from marine sources while over the ocean and from land sources while over land. The extent of air pollution buildup is determined by the duration of the sea breeze and the strength of the inversion layer, and for the system to be flushed, a stronger synoptic system with high winds is required (EC, 2004).  1.2 METHODS This thesis is presented in four chapters. The present chapter introduces the subject matter, and summarizes the rationale and background surrounding the objectives of the study. Chapter two outlines the atmospheric conditions of SRKW habitat, and includes the results  9  from a short study conducted in August 2007 to assess atmospheric stability in SRKW habitat. This information was included in a computer model used to simulate exhaust dispersion to determine killer whale exposure to air pollutants from whale-watching vessels. Data on the number of vessels, engine emission rates, distance of vessels to killer whales, distance between vessels, and length of exposure were also incorporated in the model. Chapter three describes the air pollutants emitted in marine engine exhaust, and their health effects on mammals. An allometric scaling model was used to assess the health impact of the air pollutant exposures predicted by the dispersion model on killer whales. Factors such as age, health, and genetic susceptibility make some individuals more sensitive to air pollution (Van Atten et al., 2004), thus the percentage of sensitive individuals in the SRKW population was determined. Chapter four summarizes the findings in the previous chapters, and concludes the study with suggestions for future research that would increase the confidence of the model predictions. It also provides methods for reducing SRKW exposure to air pollutants from whale-watching vessels. My objective in preparing this thesis was to estimate the quantity of specific air pollutants inhaled by the whales during exposure to whale-watching vessels under various conditions, and predict the health effects. I also wanted to determine the proportion of SRKWs that may be extra sensitive to air pollution. While the results of this study may indicate that vessel exhaust has a negligible physiological impact on killer whales, if the results show that exposure levels are sufficient to pose a health risk, the findings may be utilized as an instrument for further development of the whale-watching guidelines to ensure that whale-watching vessels have the least possible impact on this endangered killer whale population.  10  1.3 REFERENCES Bain, D. E. 2002. A Model Linking Energetic Effects of Whale Watching to Killer Whale (Orcinus orca) Populations. Friday Harbor, WA: Friday Harbor Laboratories, University of Washington. Bain, D. E., Williams, R., Smith, J. C., & Lusseau, D. 2006. Effects ofvessels on behavior ofsouthern resident killer whales (Orcinus spp.) 2003-2005 (NMFS Contract Report No. AB133F05SE3965). Retrieved March 18, 2007, from http ://www.nwfsc.noaa.gov/researchldivisions/cbd!marinemammal/documents/bainn mfsrep2003 -Sfinal.pdf Baird, R. W. 2001. Status of killer whales, Orcinus orca, in Canada. Canadian FieldNaturalist, 115: 676-701. Barrett-Lennard, L. G. 2000. Population structure and mating patterns of killer whales (Orcinus orca) as revealed by DNA analysis. University of British Columbia Ph.D. Thesis. British Columbia Provincial Health Officer (BCPHO). 2004. Every Breath You Take: Provincial Health Officer’s Annual Report 2003. Victoria, BC: Ministry of Health Services. Brook, J. R., Strawbridge, K. B., Snyder, B. J., Boudries, H., Worsnop, D., Sharma, S., Anlauf, K., Lu, G., & Hayden, K. 2004. Towards an understanding of the fine particle variations in the LFV: Integration of chemical, physical and meteorological observations. Atmospheric Environment, 38(34): 5775-5788. Colborn, T. A., & Smolen, M. J. 1996. Epidemiological analysis of persistent organochiorine contaminants in cetaceans. Reviews ofEnvironmental Contamination and Toxicology, 146: 91-172. Department of Fisheries and Oceans Canada (DFO). 2008. Viewing Guidelines. Pacific Region Marine Mammals and Turtles. Retrieved May 14, 2008, from http ://www.pac.dfo-mpo gc.ca!species/marinemammals/viewe.htm .  Environment Canada (EC). 2004. Characterization ofthe Georgia Basin/Puget Sound Airshed. Retrieved September 12, 2007, from http ://www.pyr. ec.gc.cair/gbpsairshed/summary_e.htm Environment Canada (EC). 2005. Killer Whale: Northeast Pacific Southern Resident Population. Species At Risk. Retrieved on November 20, 2005, from http ://www. speciesatrisk. gc.ca!search!speciesDetails_e.cfm?SpecieslD=699 Erbe, C. 2002. Underwater noise of whale-watching boats and potential effects on killer whales (Orcinus orca), based on an acoustic impact model. Marine Mammal  11  Science, 18: 394-418. Health Effects Institute (HEI). 1999. Diesel Emissions and Lung Cancer: Epidemiology and Quantitative Risk Assessment. A Special Report of the Institute’s Diesel Epidemiology Expert Panel. N. Andover, MA: Flagship Press. Jelinski, D. E., Krueger, C. C., & Duffus, D. A. 2002. Geostatistical analyses of interactions between killer whales (Orcinus orca) and recreational whale-watching boats. Applied Geography, 22: 393-411. Koski, K. L. 2006. Soundwatch Public Outreach/Boater Education Project 2004-2005 Final Program Report. Friday Harbor, WA: The Whale Museum. Krahn, M. M., Ford, M. J., Perrin, W. F., Wade, P. R., Angliss, R. P., Hanson, M. B., Taylor, B. L., Ylitalo, G. M., Dahlheim, M. E., Stein, J. E., & Waples, R. 5. 2002. Status review ofSouthern Resident killer whales (Orcinus orca) under the Endangered Species Act (NMFS-NWFSC 54). Washington, DC: United States Department of Commerce. Krahn, M. M., Ford, M. J., Perrin, W. F., Wade, P. R., Angliss, R. P., Hanson, M. B., Taylor, B. L., Ylitalo, G. M., Dahiheim, M. E., Stein, J. E., & Waples, R. S. 2004. 2004 Status review ofsouthern resident killer whales (Orcinus orca) under the Endangered Species Act (NMFSNWFSC 62). Washington, DC: United States Department of Commerce. Kruse, S. 1991. The interactions between killer whales and boats in Johnstone Strait, B.C. In K. Pryor & K. S. Norris (Eds.), Dolphin Societies: Discoveries and Puzzles (pp. 149-159). Berkeley, CA: University of California Press. Lange, 0. S. 1998. The Wind Came All Ways: A Quest to Understand the Winds, Waves and Weather in the Georgia Basin (Cat. No. En56-74/1998E). Victoria, BC: Environment Canada. National Ocean and Atmospheric Association (NOAA). 2005. Final ESA Listing Decision for Killer Whales. Retrieved on November 20, 2005, from http :/Iwww.nwr.noaa.gov/Marine-Mammals/Whales-Dolphins-Porpoise/Killer Whales/ESA-Act-Status/Listing-Final.cfm Orca Network. 2008. Southern Resident Orca Community Births and Deaths since 1998. Retrieved on October 15, 2008, from http://www.orcanetwork.org/news/birthsdeaths.html Osborne, R., Koski, K., & Otis, R. 2002. Trends in Whale Watching Traffic Around Southern Resident Killer Whales. Friday Harbor, WA: The Whale Museum. Osborne, R. W., Koski, K. L., Talimon, R. E., & Harrington, S. 1999. Soundwatch 1999  12  Final Report. Roche Harbor, WA: Soundwatch Boater Education Program. Ross, P. S. 2006. Fireproof killer whales (Orcinus orca): Flame retardant chemicals and the conservation imperative in the charismatic icon of British Columbia, Canada. Canadian Journal ofFisheries andAquatic Sciences, 63: 224-234. Ross, P. S., Ellis, G. M., Ikonomou, M. G., Barrett-Lennard, L. G., & Addison, R. F. 2000. High PCB concentrations in free-ranging Pacific killer whales, Orcinus orca: Effects of age, sex and dietary preference. Marine Pollution Bulletin, 40(6): 504-5 15. Smith, J. C., & Bain, D. E. 2002. Theodolite Study of the Effects of Vessel Traffic on Killer Whales (Orcinus orca) in the Near-Shore Waters of Washington State, USA. Pages 143-145 in Fourth International Orca Symposium and Workshops, September 23-28, 2002, CEBC-CNRS, France. Steyn, D. G., & Faulkner, D. A. 1986. The climatology of sea-breezes in the Lower Fraser Valley, BC. Climatological Bulletin, 20(3): 2 1-39. Trites, A. W. & Bain, D. E. 2000. Short- and Long-Term Effects of Whale Watching on Killer Whales (Orcinus orca) in British Columbia. Vancouver, BC: University of British Columbia. Van Atten, C., Brauer, M., Funk, T., Gilbert, N. L., Graham, L., Kaden, D., Miller, P. J., Rojas Bracho, L., Wheeler, A., & White, R. H. 2004. Honing the Methods: Assessing Population Exposures to Motor Vehicle Exhaust. Montreal, QC: Commission for Environmental Cooperation. Vingarzan, R., & Thomson, B. 2004. Temporal variation in daily concentrations of ozone and acid-related substances at Saturna Island, British Columbia. Journal ofAir & Waste Management Association, 54: 45 9-472. Washington Department of Fish and Wildlife (WDFW). 2008. Wildlife Viewing. Retrieved July 10, 2008, from http://www.wdfw.wa.gov/ Whale Watch Operators Association Northwest (WWOAN). 2008. 2008 Guidelines and Best Practices for Commercial Whale Watching Operators. Retrieved July 15, 2008, from http://www.nwwhalewatchers.org/guidelines.html Wiles, G. J. 2004. Washington State Status Report For The Killer Whale March 2004. Olympia, WA: Washington Department of Fish and Wildlife. Williams, R., Trites, A. W., & Bain, D. E. 2002. Behavioural responses of killer whales (Orcinus orca) to whale-watching boats: opportunistic observations and experimental approaches. Journal ofZoology, London, 256: 255-270.  13  2 EXHAUST EMISSION DISPERSION IN THE MARINE ATMOSPHERIC BOUNDARY  1  2.1.1 INTRODUCTION The first part of this chapter contains background information on the airshed in question, including ambient air quality, air pollution sources, atmospheric processes, and air quality standards. It also contains the results of a study to assess atmospheric stability off southeastern Vancouver Island, BC in August 2007, which was compared to air quality and temperature data from other locations in SRKW habitat. Part two includes background information on air pollution modeling, exhaust emissions, and the whale-watching fleet. It also describes an air pollution dispersion model that I developed to simulate exhaust dispersion from commercial whale-watching vessels, and calculate the killer whale’s exposure under different conditions.  2.1.1.1 Airshed Description and Air Pollution Sources The summer habitat of the SRKWs includes the Georgia Basin Airshed, which comprises the western trans-boundary coastal region of Canada and the U.S., the Georgia Basin in Canada, Puget Sound in the U.S., and Juan de Fuca Strait’s southern coast (Figure 2.1) (EC, 2004). The Georgia Basin Airshed overlaps with the Lower Fraser Valley (LFV) Airshed, which stretches from Horseshoe Bay to Hope, BC, and includes Metro Vancouver (formerly the Greater Vancouver Regional District or GVRD), the south-west portion of the Fraser Valley Regional District (FVRD), and Whatcom County, WA (Figure 2.1) (GVRD, A version of this chapter will be submitted for publication. Lachmuth, C. L., Barrett-Lennard, L. G., and Milsom, W. K. A model-based approach investigating killer whale (Orcinus orca) exposure to marine vessel engine exhaust. 14  2002). The quantity of pollutants emitted into an airshed is often assumed to determine air quality, yet other parameters such as topography, atmospheric conditions, and the air pollution source are very important factors governing air quality (BCME, 2005). The geography and wind currents in the southern section of the Georgia Strait make it an important area for the accumulation, re-distribution, and chemical change of air pollutants (Brook et al., 2004). Atmospheric flow conditions that lead to poor air quality in the LFV Airshed during the summer occur when light morning winds come from the northwest or south (because the coastal mountains and Vancouver Island channel the pollutants), and when there is high-pressure over the eastern Pacific Ocean with a shallow thermal trough over WA and southwestern BC (Ainsley & Steyn, 2007). Elevated air pollution episodes are not associated with sea breeze circulation and they end when the high-pressure ridge shifts eastward, allowing cool marine air into the Juan de Fuca Strait (Ainsley & Steyn, 2007).  15  128 51  50  49  Pacific Ocean 4&.  47 25  0  25  50  75  100 naut,Ca1 mites  5CAthor  125  126’  124  116’  Figure 2.1: The Georgia Basin and Lower Fraser Valley Airsheds (adapted from EC, 2004). The Georgia Basin/Puget Sound region is one of the largest metropolitan centres in North America (EC, 2004). The population of this area was 6.97 million in 2002, and is expected to grow steadily over the next 20 years to the predicted 9 million by 2020 (EC, 2004). Despite the population growth, average and peak levels of nitrogen dioxide 2 (NO ) , carbon monoxide (CO), and sulfur dioxide (SO ) have substantially decreased over the past 2  two decades in the LFV, while the average ozone (03) concentration has increased and the particulate matter (PM) concentration has remained the same (MV, 2006). However, adverse human health and environmental effects can occur at ambient concentrations commonly measured in the LFV (EC, 2004; MV, 2006). The total emissions of air pollutants from all sources during the year 2000 for the LFV Airshed can be seen in Appendix B Table B 1 (GVRD, 2002).  16  The proximate air pollutants emitted by whale-watching vessels are not the only marine sources of air pollutants the SRKWs are exposed to. Their summer habitat experiences high levels of commercial vessel traffic, as the Juan de Fuca Strait, Haro Strait, Boundary Pass, and the Georgia Strait form western Canada’s primary shipping route (The Chamber of Shipping, 2007). Figure 2.2 displays the percentage of air pollutants attributed to marine vessels in the LFV Airshed in the year 2000 emission inventory (GVRD, 2002). 2 emissions, Marine vessels in the Georgia Basin Airshed are the largest single source of SO  but this is primarily from ocean-going vessels due to the high sulfur content of marine grade fuel (EC, 2004). The majority of marine vessel emissions in the LFV Airshed occur in the coastal waters off Metro Vancouver. Over the next decade it is predicted that pollutant emissions from automobiles will decrease, while emissions from marine sources will increase and surpass automobile emissions in the LFV by 2010 (EC, 2004).  35  30  25  20  15  10  5  —r-——  0  PM 10  PM 2.5  —  —  —  NOx  SOx  Smog*  Air pollutant  Figure 2.2: Percentage of air pollutants attributed to all marine vessels in the LFV Airshed in , SO, and 25 the year 2000 emission inventory (GVRD, 2002). *principally NON, VOC, PM . 3 NH  17  The Port of Vancouver is Canada’s busiest; it handles over 70 million tonnes of cargo from more than 90 countries, and almost a million passengers per year (Thomson, 2004). The traffic along the shipping route consists of ocean-going vessels including automobile carriers, bulk carriers, container ships, cargo ships, tankers, and passenger ships (Quan et al., 2002a). In addition to ocean-going vessels, other marine vessels contributing emissions to the Georgia Basin Airshed are: cruise ships, harbour vessels (workboats, tugboats, and charter vessels), ferries, fishing vessels, and recreational vessels (Quan et al., 2002b). The percentage of air pollutant emissions produced by different vessel categories is presented in Figure 2.3 (Quan et a!., 2002b). The large percentage of CO and volatile organic compound (VOC) emissions attributed to recreational vessels is due to the numerous inefficient gasoline (especially two-stroke) outboard engines in use (Quan et al., 2002b). The total amount of air pollutant emissions (in tonnes) by ocean-going vessels in BC during 2005-2006 can be seen in Appendix B Table B2 (The Chamber of Shipping, 2007).  18  NOx  56%  /  22% /  /  -0% 1%  Particulate Matter  Ocean-going vessels Harbour vessels Ferry vessels — Fishing vessels Recreational vessels  Green House Gas Volatile Organic Compounds  (or CO 2 equivalent)  // /  j%i  1%  Figure 2.3: Air pollutant contributions by vessel category in BC outside of Metro Vancouver and FVRD in the year 2000 (Quan et al., 2002b).  Emissions from ocean-going and fishing vessels have seasonal trends: ocean-going vessel emissions peak from May to August, when over 40% of their total emissions are released; fishing vessel emissions peak during July and August, with 40% of their emissions  19  released during those two months (Quan et al., 2002b). Thus the highest emission outputs from ocean-going and fishing vessels occurs during the peak whale-watching season. Ferry vessels maintain relatively constant emissions throughout the year, with the BC Ferry Corporation contributing the majority of emissions to this vessel category (Quan et al., 2002b). Table B3 in Appendix B presents the estimated quantity of air pollutants emitted by marine vessels in the year 2000 in coastal areas outside Metro Vancouver and FVRD, as well as Vancouver Island (Quan et a!., 2002b). The aviation sector also contributes air pollutants to the atmosphere; however, the year 2000 emission inventory of the LFV Airshed determined that aircraft emissions from the Vancouver International Airport only accounted for approximately 1% of all greenhouse gas emissions, and less than 1% of all smog-forming pollutants (GVRD, 2002). However, contributions from aviation to ambient air pollution may be slightly higher in SRKW habitat because the Vancouver International Airport, the Victoria International Airport, the Bellingham International Airport, and the Seattle-Tacoma International Airport are all located within the population’s airshed. Most air quality monitoring stations are located in urban centres; however, the Canadian Air and Precipitation Monitoring Network (CAPMoN) maintains a station on Saturna Island, BC (48°47’32” N 123°08’35” W), which is in the Georgia Strait between Vancouver Island and the BC mainland (Vingarzan & Thomsom, 2004). The station is at an elevation of 178 m and measurements of background gaseous and particulate air pollutants tend to be low indicating that it is relatively isolated from urban sources of air pollution (Brook et al., 2004). However, pollution episodes (periodic pollution increases) do occur at this site and indicate that numerous emission sources from the local to regional are involved.  20  Some of these emission sources are urban pollution from Vancouver (40 km northeast) and Victoria (38 km southwest), as well as industrial pollution from the Crofton pulp mill on Vancouver Island (39 km northwest), an oil refinery at Cherry Point in WA (32 km northeast), two additional oil refineries in Anacortes WA (55 km southeast), and an aluminum smelter and oil refinery in Ferndale WA (45 km northeast) (Vingarzan & Thomsom, 2004).  2.1.1.2 Air Quality Objectives and Standards Government agencies in Canada, the U.S., and the World Health Organization (WHO) in Europe have set ambient air quality standards and objectives to minimize negative human health effects and protect the environment from air pollutants (BCLA, 2005). Canada 5 (particulate matter smaller than 2.5 . 2 has adopted Canada-Wide Standards (CWS) for PM microns in diameter) and 03 (Table 2.1), which are based on the annual  th 98  percentile  concentration averaged over three consecutive years (BCPHO, 2004). Federal, provincial, and territorial governments agreed upon the CWSs, and Canadian jurisdictions must meet these standards by 2010 and show a continuous effort to keep the air clean (BCLA, 2005). The CWSs are minimum targets for all provinces but the standards may not obtain the level of air quality that a particular jurisdiction desires, and in certain areas conforming to the standards may actually lead to air quality deterioration (BCPHO, 2004). Thus Metro Vancouver (MV) established Air Quality Objectives (AQO) for several air pollutants, which are similar to the CWSs (Table 2.2) (MV, 2006). The CWSs and MV AQOs are in , and are for specific averaging periods (durations m ) milligrams per cubic meter of air (mg 3 of exposure).  21  25 and 03 (BCPHO, 2004). Table 2.1: The Canada-Wide Standards for PM Averaging Standard Air Pollutant Period* ) 3 (mg m 24-hour 0.03 5 . 2 PM 8-hour 0.13 03 *Annual  th 98  percentile averaged over three consecutive years.  Table 2.2: The Metro Vancouver Air Quality Objectives (MV, 2006). Averaging Standard Air Pollutant Period ) 3 (mg m 1-hour 30 CO 8-hour 10 1-hour 0.2 2 NO Annual 0.04 24-hour 0.05 10 PM Annual 0.02 24-hour 0.025 5 . 2 PM Annual 0.012 8-hour 0.13 03 1-hour 0.45 2 SO 24-hour 0.125 Annual 0.03  Air quality standards and objectives are designed to protect human health and are based on human respiratory rates and breathing patterns. Yet the standards in the U.S. (see Appendix C, Table Cl) are less stringent than those in Canada, California, Europe (see Appendix C, Table C2), and several other jurisdictions (Cooper & Alley, 2002). Furthermore, the WHO does not list a standard for PM because they consider there to be no short or long-term exposure to PM below which no harmful effects are expected (WHO, 2000). Despite the variation in air quality standards and the focus on humans, the Metro 2 will be used in this thesis as reference to Vancouver Air Quality Objectives for CO and NO determine what concentration of air pollutants may pose a risk to killer whales.  22  2.1.1.3 Ambient Air Quality  The ambient (background or baseline) air pollutant concentrations of an airshed are not a result of local anthropogenic emissions; instead they arise from a combination of local natural emissions within the airshed, and the long-range transport of natural and/or anthropogenic pollutants (McKendry, 2006). Western North America is subjected to the trans-Pacific transport of Eurasian aerosols and Saharan dust, making background concentrations highly variable both spatially and temporally (McKendry, 2006). The Georgia Basin receives polluted air from long-range transport or air masses from Eurasia for example, and to a lesser extent the Americas. While these pollutants are well diluted on arrival, they add a measurable amount of PM and 03 to ambient concentrations (EC, 2004). Larger diameter particles (> 10 pm) tend to settle out of the atmosphere quickly, smaller diameter particles (< 10 tim) remain suspended for longer durations before deposition, and very small particles (< 1 pm) can remain in the atmosphere for days and can travel much further (BCME, 2006). Because long-range pollutants spend extended periods of time in the atmosphere, they have the opportunity to form secondary pollutants (EC, 2004). The British Columbia Ministry of Water, Land, and Air Protection (WLAP) Air Resources Branch maintains an Atmospheric Data and Air Quality Health Index Web Service for monitoring stations in BC. The data are considered unverified, as they have not been screened for erroneous values due to equipment malfunction or other sampling anomalies (WLAP, 2008). The station that best represents SRKW habitat is the Christopher Point monitoring station (48°18’34” N 123°33’43” W) on the southwestern tip of Vancouver Island, BC, and has a sampling height of 10 m (Figure 1.2). Table 2.3 provides the average and maximum ambient air pollutant concentrations from May to September of 2005-2007 for  23  air pollutants at Christopher Point (WLAP, 2008). The ambient air pollutant concentrations at this site are typical of a remote location and are lower than the CWS and MV AQO5.  Table 2.3: Monthly average ambient air pollutant concentrations (maximums in parentheses) at the Christopher Point, BC air quality monitoring station from 20052007*. August September July June May Air pollutant ) 3 (mg_rn 0.84 0.61 0.91 0.60 0.61 CO (1.31) (1.18) (1.18) (0.92) (0.94) 0.012 0.0072 0.0062 0.0069 0.0078 2 NO (0.064) (0.033) (0.043) (0.046) (0.057) 0.0028 0.0020 0.0020 0.0019 0.0021 NO (0.037) (0.018) (0.034) (0.039) (0.026) 0.043 1 0.05 0.050 0.059 0.076 03 (0.11) (0.10) (0.12) (0.11) (0.13) 0.0052 0.0036 0.0034 0.0033 0.0039 5 . 2 PM (0.018) (0.017) (0.018) (0.021) (0.014) 0.0027 0.0019 0.0017 0.0015 0.0014 2 SO (0.048) (0.016) (0.024) (0.026) (0.032) * 2 and NO data are from May-September of 2006 The CO data are from May-September of 2006-2007; the NO 2 data are from September 2005, and May-September 20065 and SO . 2 and May-July of 2007; and the 03, PM 2007 (WLAP, 2008).  Ambient air pollutant concentrations in the Georgia Basin Airshed display seasonal trends, as outlined in Table 2.4 (Vingarzan & Thomson, 2004). Air pollutant concentrations ) peak during the spring, summer, or fall and this is also when the whale 2 (except SO watching season peaks. Unfortunately, the SRKWs are usually offshore in the winter when the ambient air pollutant concentrations are at their lowest in the Georgia Basin Airshed.  24  Table 2.4: Times of the year when ambient air pollutants reach their maximum and minimum concentrations in the Georgia Basin Airshed (Vingarzan & Thomson, 2004). Minimum Maximum Air Pollutant Summer April November 2 SO Winter Spring & summer 2 4 S0 Winter Summer & fall 3 HNO Summer Spring & fall 3 NO Winter Spring & fall 4 NH Winter summer Spring & 03 Winter Summer & fall 5 . 2 PM Winter September 10 PM -  —  2 concentrations are the exception to the seasonal In the Georgia Basin Airshed, SO  2 has a winter maxima between November and trend displayed by the other pollutants, as SO April, and a summer minima (Vingarzan & Thomson, 2004). The other pollutants have the ) and 03, which are strongly 3 same general seasonal trend, especially nitric acid (HNO 3 peak in late summer/early fall due to affected by solar radiation. Concentrations of HNO 3 (Vingarzan & 2 to I-1N0 hot and dry conditions that promote the transformation of NO Thomson, 2004). 03 is primarily produced under sunny conditions with light winds or stagnant periods, when 03 and its precursors can become trapped next to the surface (EC, 2004). 03 concentrations are especially high during the month of May due to background 03, and in the summer due to elevated local emissions (EC, 2004). The weather patterns over the Georgia Basin are not conducive to 03 production 38% of the time (see Section 2.1.1.1), and the meteorological conditions that produce extreme episodes are usually short lived and arise only 3% of the time (Ainsley & Steyn, 2007; McKendry, 1994). The mean 3 in the background concentration of 03 is estimated to range from 0.03 9-0.069 mg m Georgia Basin, which is already at 50% of the CWS, thus it is possible that the CWS are occasionally exceeded by background sources alone (McKendry, 2006). Yet the current  25  ambient health quality standards for 03 do not protect sensitive individuals, as health impacts occur at levels far below ambient standards (EC, 2004), and the mean background concentration of 03 is increasing at a rate of 0.5-2% per year (McKendry, 2006). Chemical 3 of 03 to background transport models show that Asian sources add 6-20 jig m concentrations in the western U.S. during the spring, and this is estimated to increase due to rising anthropogenic emissions in Asia (McKendry, 2006). Air masses with north Pacific trajectories arriving in BC have mean background 3 and are associated with marine and 5 on the order of 1.5-2 jig m . 2 concentrations of PM 5 mass concentration in the . 2 Eurasian sources (McKendry, 2006). The annual average PM , except in the urban centres of Victoria and Vancouver 3 Georgia Basin varies from 6-8 jig m 3 (EC, 2004). However, in 2006, the small town of Saanich, where it averages 9-10 jig m BC, which is just north of Victoria and borders Haro Strait, had the dubious distinction of 3 5 concentration of 8.9 jig m . 2 being BC’s air monitoring station with the highest annual PM 3 usually occur in the summer and 5 concentrations of 23-27 jig m . 2 (BCLA, 2007). Peak PM fall during the months of October and November, and these concentrations fall just below the 3 for 24-hours exposure (EC, 2004). Due to the implementation 5 of 30 jig m . 2 CWS for PM 5 from . 2 of air pollution standards in urban areas, the background concentrations of PM regional or continental scale transport are decreasing (McKendry, 2006). However, peer 25 effects on human reviewed evidence indicates that there is no lower threshold limit for PM mortality and morbidity, and it has been suggested that background levels are more appropriate as CWS (McKendry, 2006). 10 (particulate matter smaller than 10 microns The annual average concentration of PM , with peak 3 in diameter) in the Georgia Basin Airshed varies from 12-16 jig m  26  3 usually occurring in September (EC, 2004). During stagnant concentrations of 28-30 tg m , which is 3 10 can reach 50-75 tg m summer conditions the daytime concentration of PM 3 (EC, 2004). The Georgia Basin 10 of 50 ig m greater than the MV AQO for PM meteorological patterns conducive to the production of elevated PM occur 46% of the time, most often in the spring and winter (EC, 2004). The United States Environmental Protection Agency (USEPA, 2002a) lists heavily traveled marinas as diesel PM “hotspots” in air, along with major roadways, bus stations, and train stations. There is limited data on concentrations of PM measured at hotspots, however, one study found the concentration of diesel PM at a busy Manhattan bus stop ranged from 3 (USEPA, 2002a). These extremely high PM concentrations indicate that 13.0-46.7 ig m more research is needed at hotspots, because there is potential for many people to be exposed in these areas (USEPA, 2002a). However, diesel PM concentrations in urban areas are 3 (USEPA, 2002a). usually much lower, ranging from 1.7-3.6 tg m  2.1.1.4 Atmospheric Mixing and Boundary Layer Stability The stability of a layer of air is its tendency to either rise or fall in the atmosphere a -  stable layer resists vertical movement, and an unstable layer easily rises or falls. Static stability of air pollutants decreases mixing, and leads to higher local pollutant concentrations (Tj ernström et al., 2005). Thus characterizing atmospheric mixing in SRKW habitat during the peak whale-watching season is essential as it determines whether air pollution will accumulate or disperse. The portion of the troposphere affected by marine vessel emissions is called the boundary layer, and it is directly affected by the Earth’s surface where it responds to surface  27  forcing (pressure and stress forces), such as air pollutant emissions, in an hour or less (Stull, 1988). Because water has a large heat capacity, sea surface temperature (SST) remains relatively constant in a 24-hour period; thus, the depth of the atmospheric boundary layer over the ocean changes relatively slowly spatially and temporally compared to boundary layers over land (Stull, 1988). Air pollutant accumulation is a function of the emission rate, dispersion rate, generation rate, and destruction rate, where dispersion rate is a function of local meteorological conditions (wind speed and direction), humidity, temperature, and atmospheric stability (Cooper & Alley, 2002; Stull, 1988). As expected, air pollutant concentrations are proportional to the source strength, and are inversely proportional to wind speed and the distance to the source (Stull, 1988). The two main differences between pollution boundary layers over land and water are the vertical mixing height and stability (Hanna et al., 1985). The vertical mixing height defines the layer above the surface where mixing due to turbulent motion occurs; above this height turbulent motion is suppressed by a stable capping layer (BCME, 2006). In the Georgia Basin Airshed, shallow atmospheric mixing depths of less than 100200 m near the shoreline are common due to cool water temperatures (Hoff et al., 1997; USNSB, 1996). Whenever the SST is lower than the air temperature, the boundary layer can become stably stratified, with pollutants in the lower layers being effectively unmixed by turbulence due to turbulent energy damping (Stull, 1988). The greatest static stability occurs close to the ocean’s surface, and this stability decreases toward neutral (well mixed) with height in the atmosphere. When the stability near the surface is large enough that temperature increases with height, then that section of the stable boundary layer is described  28  as having a temperature inversion. Pollutant dispersal is severely limited by temperature inversions and stable boundary layers because the tendency to move vertically in the atmosphere is eliminated and turbulence is suppressed (Oke, 1987). Boundary layer stability can be determined by comparing the air temperature gradient or the environmental lapse rate (ELR) to the dry adiabatic lapse rate (DALR). The DALR is the negative rate of temperature change a rising parcel of unsaturated air has under adiabatic 1 (Oke, 1987). The 3 °C m (no heat transfer) conditions, and has a constant value of-9.8x10 dry adiabatic lapse rate rather than the saturated adiabatic lapse rate is used in the boundary layer below cloud base height because it has less than 100% relative humidity (Oke, 1987). The ELR can be determined by the temperature change with height: ELR=  2 z  —  1 z  where ELR is the environmental lapse rate (°C m’), T is the temperature (°C), z is the 1 and altitude (m), with z  Z2  measurements from two different altitudes.  When comparing the environmental lapse rate (ELR) to the DALR, three scenarios can occur (Oke, 1987): 1. The slope of the ELR is more negative than that of the DALR: The layer of air is defined as unstable because an air parcel would rise vertically in the atmosphere due to buoyancy (Figure 2.4a). This is typical of sunny days when the ground heats the near-surface air. 2. The slope of the ELR is more positive than that of the DALR: The layer of air is defined as stable because an air parcel would be colder than the surrounding air and would sink (Figure 2.4b). The greater the difference between the two slopes, the larger the damping tendency. This is typical of an inversion layer when warmer air overlies cooler air. 29  3. The slope of the ELR and the DALR are equal: The layer of air is defined as neutral, because an air parcel would not rise or fall vertically. This is typical of cloudy windy conditions, which minimizes horizontal temperature stratification.  -  C)  a)  I  ELR” (stable) Temperature (°C)  Temperature (°C)  Figure 2.4: Graphs of temperature change with height showing (a) an unstable atmosphere and (b) a stable atmosphere. The solid line is the ELR or the Environmental La?se Rate, which is a measured temperature profile; the dashed line is the DALR (-9.8x10 °C m’). The arrows on the right side of each graph demonstrate the motion an air parcel (indicated by . If the parcel were displaced downwards, it 1 a circle) would have if displaced above height z would be a mirror image of the displacement upwards. Adapted from Oke (1987). A crude alternative method to determine the stability of the lower marine boundary layer is to calculate the air-sea surface temperature difference, AT  =  Tair  —  Tsea. The air  temperature is measured in the boundary layer below mixing height, which is approximately 1-10 m from the surface in a stable atmosphere (D. Steyn, pers. comm., November 2008). When the air-sea surface temperature difference is negative (air colder than water) the atmospheric conditions are unstable, when near zero the conditions are neutral, and when positive (air warmer than water) the conditions are stable. Both methods of determining the atmospheric stability of SRKW habitat during the whale-watching season were employed air temperature gradients were sampled, as were -  30  SSTs to obtain the air-sea surface temperature difference. Since SST is highly conservative in enclosed waters (Steyn & Faulkner, 1986), but air temperatures tend to increase towards the mainland and Vancouver Island due to overland heating, positive air-sea surface temperature differences, or stable atmospheric conditions, would be expected at locations closer to the BC mainland and Vancouver Island.  2.1.2 MARINE ATMOSPHERIC BOUNDARY LAYER MEASUREMENTS Air and sea surface temperatures were collected from locations offshore of southeastern Vancouver Island, BC. Dates sampled in 2007 were on August 21 from 10:54 a.m.-4:00 p.m.; on August 22 from 5:21-6:30 p.m.; on August 23 from 12:00-1:00 p.m.; on August25 from 6:45-7:45 p.m.; and on August27 from 1:15-3:30 p.m. During sampling, . To sample 1 cloud cover ranged from 0-100%, and wind speeds ranged from 1-7.7 m s temperatures, a three-meter aluminum pole was attached vertically to the bow of an 3.35 m zodiac, and four copper-constantan thermocouple wires were attached at 0.78 m intervals along the pole, with the first positioned 0.70 m above the waterline (Figure 2.5). Stacked white plate covers shielded the thermocouples from solar radiation and allowed airflow. A fifth thermocouple wire entered the upper surface layer of the ocean to obtain SST at approximately 0.1 m in depth. The exposed wires on the fifth thermocouple were coated with epoxy to render them waterproof for obtaining SST. The five thermocouple wires were attached to a micrologger (Campbell Scientific 21X(L)) programmed to record the temperature from each thermocouple every 5 seconds, and calculate the average temperature after 15 minutes. During each 15-minute trial the zodiac maintained a speed of two knots (1.03 m s’), and traveled approximately 927 m.  31  ‘J  1 meter  Sea Surface Thermocouple (.  Figure 2.5: Profile view of the thermocouple setup on the zodiac. Four thermocouple wires covered with stacked white plates were mounted on a vertical aluminum pole to sample the air temperature at evenly spaced heights above the water, and a fifth thermocouple entered the water to obtain sea surface temperature. A horizontal wood plank along the centre line of the zodiac and rigging provided stability for the pole.  Forty-one 15-minute trials were conducted under low, medium, and high combinations of wind speed and cloud cover. Due to stability issues with the pole mounted on the bow of the zodiac, trials were not carried out when wind speeds exceeded 15 knots (7.72 m s’). All trials occurred near Oak Bay, Vancouver Island, BC (Figures 1.2 and 2.6), a location where the SRKWs are often sighted. This location was chosen because it is in  32  SRKW habitat, and further distances from shore could not be sampled due to the instability of the zodiac with the experimental setup.  A NORTH Oak Bay  S’  S  48  4  24’  StudyArea Fuca Juan de 123 20’  123  strau 16’  1 nm 123  12’  Figure 2.6: Map of southeast Vancouver Island, BC. The area traversed during temperature sampling is shaded in light grey.  2.1.3 MARINE ATMOSPHERIC BOUNDARY LAYER DATA To determine atmospheric stability, Oak Bay air temperature gradients (ELR5) were plotted and compared to the DALR, and the air-sea surface temperature differences were calculated. All 41 trials displayed the same general trend an increase in air temperature -  with height above the water. The average slope for all 41 trials was 4.72 °C m’ (SEM 0.36), which is more positive than that of the DALR (-9.8x10 3 °C rn”). Figure 2.7 shows the first 10 15-minute trials conducted; for clarity not all trials are shown as they display almost identical profiles. Only nine trials (of the total 41) deviated from the general trend, as one or two temperatures did not increase with height, as seen in Figure 2.8. This is likely because  33  the average wind speed of these nine trials was greater (by 1.3 m s’) than the average wind speed of all 41 trials, and the average cloud cover of the nine trials was less (by 29%) than the average of the 41 trials.  3.5  3.0  2.5 S 0 ——--v———  20  —  —‘— -  —---—  1.5  —  —  —0--—’— —4-— —  —‘-—————  1.0  •.  A  ——--v———  Trial 1 Trial2 Trial3 Trial4 TrialS Trial 6 Trial 7 Trial 8 Trial9 TriallO DALR  0.5 13.5  14.5  14.0  16.0  15.5  15.0  Temperature ç’C)  Figure 2.7: Results from the first ten 15-minute trials at Oak Bay, BC, demonstrating that the average temperature increased with height above the water. The dry adiabatic lapse rate (DALR) is plotted as a solid black line. 3.5  3,0  2.5 S cJ)  Trial 15  2.0 ——---V.-—— —  1.5  —a’—”-  ————  ——D—--’—  ——-4--—— —0--———— A  1.0  Trial24 Trial27 Trial34 Trial 35 Trial 37 Trial 40 Trial 41 DALR  0.5 12.0  13.0  14.0  15.0  16.0  17.0  18.0  19.0  Temperature (DC)  Figure 2.8: Plot of the nine trials at Oak Bay, BC, that had deviations from temperature increasing with height. The dry adiabatic lapse rate (DALR) is plotted as a solid black line. 34  The average difference between the air temperature measured 1.5 m above the water 0.20); all 41 trials had positive  and SST for all 4115-minute trials was 2.45 °C (SEM  differences except two (Figure 2.9). The air temperature at 1.5 m above sea surface was used because that was the air temperature sampling height used at Race Rocks Ecological Reserve (RRER, 2008). The two trials with negative differences were from consecutive samples taken on the same day at 14:00 and 14:15 while traveling around the southern tip of Trial Island. The SSTs of these two trials were approximately 2 °C greater than those of the other 39 trials, making the SST slightly higher than the air temperature, which indicates the zodiac passed through a water mass with different characteristics (e.g. an eddy).  7  .  6 5  . 4  1  3  F-  2  .  H  .  . • . .  .  •.  V  0•  • —1  I  I  I  I  C) U) C U) C) U) C) U) C) U) C) U) C) U) 0 U) C) U) C) U) — 0 C) CI U) C) C) — U) C) — C) U) C) C) — U) U) C) — U)  I U)  U)  U)  —  U)  Time of day trial started  Figure 2.9: Air-sea surface temperature difference (TairTsea) for all 411 5-mm trials, by the time of day the trial started. Air temperatures were measured 1.5 m above the sea surface, and sea surface temperatures were measured approximately 0.1 m below the sea surface.  In order to evaluate the extent to which our pilot study was representative of general air-sea temperature conditions in the area, and to determine the frequency of stable boundary  35  layers during the rest of the year, a comparison was made to data from the Race Rocks Ecological Reserve (480 181 N 123°32’ W) in the Juan de Fuca Strait, BC (Figure 1.2). The air-sea surface temperature difference was calculated from monthly air temperature data gathered in 2007 at the Race Rocks Ecological Reserve (RRER, 2008), and SST data from the lighthouse at Race Rocks maintained by the Department of Fisheries and Oceans Canada (DFO, 2008a). The average monthly difference between air temperature and SST at Race Rocks in 2007 is plotted in Figure 2.10. Air temperature was measured 1.5 m above rock surface and SST was measured daily off the end of the dock 1-hour before high tide, from a bucket sample lowered 1-2 m below sea surface. No air temperature data was available for the month of March; however, based on the trend seen in Figure 2.10, March would likely have an air temperature similar to the SST. 3  2  C) C (0  0  I-  -2  —3  I  Jan.  I  I  I  I  I  Feb. March April May June July  I  I  I  Aug. Sept. Oct. Nov. Dec.  Month  Figure 2.10: Average monthly air-sea surface temperature difference (TairTsea) at Race Rocks, BC in 2007, with standard error of the mean bars. Air temperature was sampled 1.5 m above sea surface, and sea surface temperature was sampled 1-2 m below sea surface.  36  The average monthly air-sea surface temperature difference from June to September in 2007 at Race was positive (average  =  0.42). The positive difference at  0.81 °C, SEM  Race Rocks in August is in agreement with the data collected at Oak Bay in August 2007, and indicates stable atmospheric conditions. To ensure that 2007 was not an anomalous year, air-sea surface temperature differences from 2002-2006 were calculated from Race Rocks data (Figure 2.11). The average monthly air-sea surface temperature difference at Race Rocks was 0.33 °C (SEM  =  0.47). From April to October the differences were positive, with an average of 1.53 °C (SEM = 0.34). Since May 2007 had unstable/neutral atmospheric conditions, a comparison was made to Race Rocks data from 2002-2006. The unstable/neutral atmospheric conditions in May of 2007 were due to unusually low air temperatures (about 2 °C lower than average), as the SSTs in 2007 were consistent with the SST averages from 2002-2006. 4  —-  1  l—t 0 I  -3  I  Jan.  I  I  Feb. March April  I  May June July Aug. Sept. Oct. Nov. Dec.  Month  Figure 2.11: Average monthly air-sea surface temperature difference (TairTsea) at Race Rocks from 2002-2006, with standard error of the mean bars. Air temperatures were measured 1.5 m above the sea surface, and sea surface temperatures were measured 1-2 m below the sea surface.  37  Air-sea surface temperature differences were calculated for two more locations in SRKW habitat: the Halibut Bank Buoy (2002-2005 data), and the Hem Bank Buoy (20042007 data). The Halibut Bank Buoy is situated in the middle of the Georgia Strait between Tsawwassen (on the mainland) and Valdez Island (49°20’2” N 123°43’2” W) (Figure 1.2). The air temperature at the Halibut Bank Buoy was sampled hourly 2-3 m above the sea surface, and SST was sampled hourly 1-2 m below the sea surface by a sensor on the buoy (0. Riche, pers. comm., February 14, 2008). From 2002-2005, the average yearly air-sea surface temperature difference at the Halibut Bank Buoy was -0.64 °C (SEM = 0.13) (Figure 2.12). The average monthly air-sea surface temperature differences were negative at this location all year, but occasionally in July, August, and September the differences became near neutral or positive as indicated by the standard error of the mean bars in Figure 2.12. The instability (negative air-sea surface temperature difference) at this location may be partly due to differences in sampling heights; however, the Halibut Bank buoy is in the plume of Fraser River water as it enters the ocean. The fresh Fraser River water absorbs solar radiation as it floats on top of denser seawater, and the temperature also increases from the tidal signal of water warmed while lying over sand at low tide in the Fraser River delta (Thomson, 1981). Higher SSTs create negative air-sea surface temperature differences, and result in more neutral or unstable atmospheric conditions. Thus, this location has confounding factors that affect the atmospheric stability; however, during the time period of concern the conditions were mostly neutral.  38  0.5  0.0  Jan. Feb. March April May June July  Aug. Sept. Oct. Nov. Dec.  Month  Figure 2.12: Average monthly air-sea surface temperature difference (TairTsea) at Halibut Bank Buoy from 2002-2005, with standard error of the mean bars. Air temperatures were measured 2-3 m above the sea surface, and sea surface temperatures were measured 1-2 m below the surface. The Rein Bank Buoy is situated in Juan de Fuca Strait, southeast of Victoria, BC (48°20’O” N 123°10’O” W) (Figure 1.2), and the air temperature was sampled hourly 4 m above the sea surface, and SST was sampled hourly 0.6 m below the sea surface by a sensor on the buoy (NOAA, 2008). From 2004-2007, the average air-sea surface temperature difference from May to September was 0.88 °C (SEM = 0.16) (Figure 2.13).  39  1.6 1.4 1.2 1.0 C)  s—  0.8  0.2 0.0 -0.2 June  May  July  August  September  Month  Figure 2.13: Average monthly air-sea surface temperature difference (Tajr-Tsea) at Hem Bank Buoy from 2004-2007, with standard error of the mean bars. Air temperatures were measured 4 m above the sea surface, and sea surface temperatures were measured 0.6 m below the surface. 2.1.4 MARINE ATMOSPHERIC BOUNDARY LAYER CONCLUSIONS Two methods were employed to determine the atmospheric stability offshore of Oak Bay, BC: sampling air temperature profiles (ELRs) and calculating the air-sea surface temperature difference. Both methods indicated stable atmospheric conditions during the sampling period, as the ELR slopes (average  4.72 °C m’) were more positive than that of  3 °C m’) for all 41 trials conducted, and positive air-sea surface the DALR (-9.8x10 temperature differences (average  =  2.45 °C) were found in 39 of the trials. Thus, the  atmospheric boundary layer can be classified as stable during peak whale-watching season, and the increase in temperature with height indicates that a near surface temperature inversion formed. A stable lower boundary layer creates the worst conditions for pollution dispersal because turbulence is suppressed and the vertical dispersion of pollutants is eliminated (Oke, 1987).  40  The results from Oak Bay may not represent other areas in SRKW habitat, since they were conducted very close to shore, and had limited geographical and temporal scopes. Therefore, the results from Oak Bay were compared to other locations in SRKW habitat. The 2007 monthly air-sea surface temperature differences at Race Rocks indicate that stable boundary layers occurred from June until September. In addition, data from 2002-2006 at Race Rocks indicates that 2007 was not an anomalous year, as the air-sea surface temperature differences were positive from April until October, which overlaps with the peak whale-watching season that runs from May until September. By comparing Figures 2.10 and 2.11, it can be seen that 2007 had monthly atmospheric conditions that were less stable than those in 2002-2006, and this was mainly due to unusually low air temperatures in 2007. The Halibut Bank Buoy air-sea surface temperature differences indicate that July, August, and September tend to have neutral atmospheric conditions, while at other times of the year atmospheric conditions are unstable. In contrast, the Hem Bank Buoy air-sea surface temperature differences indicate that stable atmospheric conditions predominate from May to September, which overlaps with the whale-watching season. This location is closer to Race Rocks than the Halibut Bank Buoy, and is not influenced by Fraser River water. Thus, the air-sea surface temperature differences more closely resemble those from Race Rocks than those from the Halibut Bank Buoy. Both methods of evaluating atmospheric stability, and all three locations compared indicate that the atmospheric conditions the SRKWs are exposed to during the commercial whale-watching season are predominantly stable. This can result in an accumulation of air pollutants above the surface of the water where the killer whales breathe. Additionally,  41  stable air would quickly dissipate any whale-generated turbulence from exhalation at the surface, and turbulence from vessel wakes.  2.2.1 MODELING DISPERSION IN THE MARINE ATMOSPHERIC BOUNDARY LAYER Empirical roadside studies indicate that dispersion of vehicle exhaust can be rapid  -  emissions are concentrated within 100 m of busy roads and then decline rapidly with distance from the road and reach background levels by 200-300 m downwind (Hitchins et al., 2000; Roorda-Knape et al., 1998; Tiitta et al., 2002; Zhu et a!., 2002a). Hitchins et al. (2000) found that wind blowing directly from the road caused vehicle emission concentrations to decay to about 50% of the original at a distance approximately 100-150 m from the road, whereas wind blowing parallel to the road caused emissions to decay to 50% at 50-100 m from the road. These studies suggest that the concentration of pollutants from vehicle exhaust can remain high at distances less than 100 m from the source, but then decline considerably with increasing distance. Generally at 100 m from a highway vehicle emissions are at concentrations below air quality standards, yet under certain conditions (e.g. low wind speeds, limited vertical mixing) the concentrations can remain high (HEI, 1988). Empirical studies on the dispersal of emissions over coastal waters or open-ocean are rare due to: the complexities inherent in the emission route; uncertainty regarding the behaviour and fate of pollutants in water; and the diversity in environmental conditions of aquatic systems (Rijkeboer et al., 2004). However, a study by Skyllingstad et al. (2007) found that stratified boundary layers tend to be formed over cool seawater, where air temperature profiles are quite stable, there is minimal turbulence, and relatively low wind  42  speeds near the sea surface. Angevine et al. (2004) found that ozone was transported over coastal waters in stable boundary layers at the surface, and while intermittent turbulence occurred, the chemical constituents and concentrations in the layers remained strong because there was limited deposition and shallow vertical mixing, which minimized dilution. It was concluded that pollution transport over water is different than over land due to several factors: vertical mixing and dilution are reduced and plume shearing increases; the deposition of 03 and other precursors are reduced because they are deposited at a much slower rate to water surfaces than to vegetation; local emissions are reduced because there are no fresh inputs for reactions; wind speeds are greater; and pollutants are carried long distances (20200 km) without major losses (Angevine et al., 2004). Smedman et al. (1997) also found a stratified stable boundary near the sea surface, but with increasing altitude a transition occurred to a near-neutral layer capped by an inversion. A literature search did not reveal published air quality monitoring studies focusing on output from recreational marine engines, and the air pollutant concentrations the SRKWs are exposed to have never been quantified. Real-time air quality monitoring has inherent limitations and uncertainties, such as obtaining adequate sample sizes due to the numerous variables involved. Additionally, the equipment required for a marine air quality study is extremely expensive, and the extensive data required would necessitate an entire team of assistants for collection. Instead I developed an emission dispersal model to determine killer whale exposure to air pollutants produced by whale-watching vessels by running numerous computer simulations with different combinations of parameters.  43  2.2.1.1 Emission Dispersion Models Air quality dispersion models use empirically based equations that simulate the behaviour of gases and/or particles emitted into the atmosphere to estimate exposures at receptors. Models are especially important in situations where direct measurement is impractical, as they produce a cause-effect link between emissions and the resulting ambient pollutant concentrations (BCME, 2006). Computers are used to run numerous scenarios that provide an unbiased, reproducible, and inexpensive method for assessing existing or future air quality. Because the models are based on numerous inputs and assumptions, they are subject to many possible inaccuracies (Cooper & Alley, 2002); however, variance from reality is expected due to inherent chaotic processes in the atmosphere (BCME, 2006). Often the predictions made by peer reviewed air quality dispersion models are considered to be the “best estimate” available for decision-making, and the results are used extensively for air quality management worldwide (BCME, 2006). A wide range of air quality dispersion models with different levels of complexity and combinations of parameters are available, and there are mobile-source models where the object emitting pollutants is mobile (as is the case with whale-watching vessels), and fixedsource models where the object emitting is stationary. All mobile-source models require the following parameters: the engine types in the fleet; the number of operating engines; the engine emission rates; the atmospheric conditions; and the geophysical characteristics (BCME, 2005; Van Atten et al., 2004). The British Columbia Ministry of Environment (BCME, 2006) recommends the use of extensively tested dispersion models; however, the recommended models are not appropriate for the whale-watching scenario because they are either: designed for terrestrial situations; designed for single point, area, or volume sources;  44  designed for urban locations, highways, or industrial complexes; and/or require hour-by-hour meteorological data. Due to the complex atmospheric layers over coastal areas, the development of marine models is far behind models of land dispersal (BCME, 2005). One marine model called the Offshore and Coastal Dispersion Model was developed for offshore oil and gas platforms; however, it is also inappropriate for the whale-watching scenario because the pollutant transport distance is on the scale of kilometers, it requires extensive data input (e.g. hourly meteorological data), and does not handle sea surface sources well, if at all (BCME, 2006). Multi-agent or agent-based modeling is becoming increasingly popular for simulating the dynamics of complex systems over time (Anwar et al., 2007). Multi-agent models typically include: an environment, objects in the environment, agents (the active entities in the system), links between agents, operations for the agents, and operators that modify behaviour of the agents (Bousquet & Le Page, 2004). The use of multi-agent models for human-wildlife interactions are rare; however, a multi-agent system developed by Anwar et al. (2007) was used to simulate whale-watching tours, and to calculate the “happiness factor” (the ratio of whale observation time over the trip duration). Here, I used a programmable multi-agent based modeling environment called NetLogo (Wilensky, 1999) to simulate the behaviour of exhaust gases emitted from whale-watching vessels, and estimate the concentration of exhaust gases SRKWs are exposed to under varying conditions. I used a sensitivity analysis to determine which variables had the greatest impact on the predicted air pollutant concentration.  45  2.2.1.2 Marine Engines, Fuel, and Emissions The gaseous and particulate phase of exhaust from marine diesel and gasoline engines contains hundreds of chemical compounds, the most abundant of which are carbon oxides (COg), sulfur oxides (SO), nitrogen oxides (NO), hydrocarbons (HC), and PM. There are also small amounts of known (e.g. benzene and toluene) and unknown chemical substances that may be relevant due to their persistence and/or toxicity (Rijkeboer et al., 2004). HC’s in the exhaust are composed of unburned fuel (aromatics, alkanes, and alkenes), and partially oxidized HC’s (phenols and carbonyls) (Rijkeboer et al., 2004). The majority of PM from combustion engines is in the submicrometer range (0.02-0.5 jim), and is composed of elemental carbon, adsorbed organic compounds from the fuel and oil used, sulfates from the sulfur in the fuel, and trace metals (IPCS, 1996). Several factors influence the emissions produced by marine engines: the age of the engine, the engine type, the fuel characteristics, the engine maintenance, the performance of the engine’s pollution control systems, the engine load, the engine temperature, the engine speed, and the RPM (Van Atten et al., 2004). However, Frey and Bammi (2003) found that engine exhaust emissions depend more on fuel type (diesel or gasoline), and engine technology (i.e. two or four-stroke) than engine size, age, or type of aspiration. Automobile engines have catalytic converters that reduce tailpipe emissions, but due to salt-water corrosion of the catalyst they are uncommon in marine engines. Gasoline engines lacking catalytic converters produce quantities of polycyclic aromatic hydrocarbons (PAHs) similar to diesel engines of equivalent power output (Frumkin & Thun, 2001). The two categories of marine engines are outboards and inboards, and they have very different emission characteristics. Outboard engines are mounted on the stern of the vessel,  46  have self-contained drive units, and are typically used in smaller vessels (Coates & Lassanske, 1990). Inboard engines tend to be used on larger vessels because they usually develop greater power than outboards, they are mounted inside the vessel, and the drive can be relayed through different propulsion systems, most commonly a straight shaft and a swiveling propeller unit referred to as a stern drive (Coates & Lassanske, 1990). Inboard engines are usually fuelled by diesel due to safety issues with gasoline, as it is much more flammable and has a greater explosion risk than diesel. Both diesel and gasoline fuel contain possible and known neurotoxic agents (Kirrane et al., 2006). The organic compounds found in diesel and gasoline exhaust are qualitatively similar, yet there are quantitative differences (Frumkin & Thun, 2001). Diesel engines are more fuel-efficient than gasoline engines, and ), CO, and HCs; however, they produce more NO 2 they produce less carbon dioxide (C0 (HEI, 1999) and PM at a rate approximately 20 times greater than gasoline engines (IPCS, 1996). Inboard and outboard engines can be either two or four-stroke, and are typically fuelled by a gasoline-oil mix or gasoline respectively. Older model two-stroke gasoline engines are notorious for producing more airborne volatiles, toxic organics, and PM than four-stroke engines (Kado et a!., 2000). Incomplete combustion in the two-stroke engine produces unburned residual oil and partially burnt oil that enters the marine environment through the exhaust, and leaves a sheen of oil on the water. Four-stroke engines generally , and NOR, but fewer aromatic HCs than two-stroke engines. Fuels 2 produce more CO, C0 other than gasoline and diesel, such as natural gas, petroleum gas, and biodiesel, are increasingly being used in marine engines. Alternative fuels (Appendix Dl) and fuel additives (Appendix D2) affect exhaust emissions but they were not considered in this study.  47  2.2.1.3 Wet and Dry Exhaust Systems  Marine engines can have either a dry exhaust system that emits exhaust directly into the air, or a wet exhaust system that combines the exhaust with water or directs the exhaust into the water to cool, silence, and minimize human exposure (Gabele & Pyle, 2000; Kado et al., 2000). The first type of wet exhaust system is found in sailboats and small inboard powerboats that mix sea water with the exhaust to cool it before it is ejected from the hull, usually just above the water line (Rijkeboer et al., 2004). The second type is found in many modern outboard and outdrive engines which emit below the water line through the propeller hub, and the majority of the exhaust is in a gas phase that is directly bubbled out of the water column into the atmosphere (Rijkeboer et al., 2004). The proportion of wet and dry exhaust systems in the whale-watching fleet is unknown, thus it was assumed there were equal numbers of each. Volatile exhaust gases with poor water solubility bubble out of the water and are introduced into the air; however, less volatile gases with greater water solubility remain primarily in the water (Juttner et al., 1995). Exhaust components in the water condense and either remain suspended in the water colunm or form an emulsion layer on the water’s surface. These suspended and surface emissions have several routes of atmospheric release or degradation through a mixture of chemical, physical, and biological processes (Rijkeboer et al., 2004). Marine vessels can deteriorate water quality by introducing exhaust gases into the water at levels that exceed water quality criteria (Juttner et al., 1995), and killer whales swimming through the emulsion layer to breathe could potentially ingest the pollutants. The method a toxin enters the body  —  through inhalation, ingestion or skin contact  —  determines  the percent absorption of the toxin and the resulting organ toxicity (NESCAUM, 1999). It is  48  unlikely that a killer whale’s skin would absorb pollutants in the emulsion layer since they have a very thick epidermis (Geraci & St. Aubin, 1990); however, the emulsion layer may affect their mucus membranes (e.g. eyes),.and/or could potentially be ingested and inhaled. Fuel spilled during re-fueling or sloshing from vents can also add oil and gas to the surface of the ocean, which can then be mixed by the wind and other disturbances to form a toxic emulsion layer. While this would most likely occur near shore and at marinas, the emulsion layer can disperse on currents and potentially affect killer whales in the vicinity. Toxic emulsion layers were not quantified in this study due to the number of unknowns involved, yet they may prove to be more harmful to killer whales than exhaust emissions in heavily trafficked waters. Further studies are required to determine to extent of emulsion layer formation and dispersion, and their potential health impacts. Marine engines are usually tested/certified with the engine out of the water, which does not consider exhaust retained in the water from wet exhaust systems (Kado et al., 2000), and quantifying emissions retained in the water has proven to be difficult and often unrepeatable (Coates & Lassanske, 1990). For both two and four-stroke engines with wet exhaust systems, Juttner et al. (1995) found that the VOCs in the water were almost exclusively aromatic HCs, and the amount present in the water was only 10% of that emitted in the exhaust. However, others have found that for both gasoline and diesel fuels, approximately 40% of the HC’s emitted end up being retained in the water phase and accumulate on the water surface, while the remaining 60% escapes in gas bubbles to the surface (Clark et al., 2000; Rijkeboer et al., 2004; Warrington, 1999). For modeling purposes, a 40% reduction was applied to the dry HC emissions to obtain wet HC emissions.  49  CO is poorly soluble in water and more than 80% emitted is immediately lost to the atmosphere from vessels with wet exhaust systems (Rijkeboer et a!., 2004); thus for the dispersion model, a 20% reduction was applied to the dry CO emissions to obtain wet CO emissions. Clark et a!. (2000) found that NO emissions in air for dry exhaust tests were 21% higher than in wet exhaust tests, thus I factored in a 21% loss of NO to the water to obtain wet NO emissions. The PM and PM-associated toxic compounds produced by engine cOmbustion are hydrophobic, thus can potentially re-enter the air from the water (Kado et al., 2000). When diesel engine exhaust is expelled beneath the water surface, about 40% of the PM is deposited in the water (Clark et al., 2000). Since a literature search did not produce information on wet exhaust PM retention for gasoline engines, it was assumed that gasoline exhaust experiences the same 40% reduction as diesel. Thus for modeling emissions, a 40% reduction was applied to the dry PM emissions to obtain wet PM emissions.  2.2.1.4 The Whale-Watching Fleet Information on marine engines used by the commercial whale-watching companies that target the SRKWs was obtained from the Soundwatch Boater Education Program in 2005 (unpubi.), and from personal communication with whale-watching companies in 2006. This provided the number of engines per vessel, the horsepower (hp) of each engine, the fuel type, and the engine type (inboard/outboard, two, or four-stoke) for 23 out of 46 companies that operated during the 2005 season. It was assumed that the missing half of the fleet operated similar engines. Air pollution modelers use emission rate averages of the mixed fleet of vehicles in use (L. Frank, pers. comm., February 13, 2007); thus, from the Soundwatch data it was determined that the “average” whale-watching vessel had either twin  50  200 hp (total  400 hp) inboard four-stroke diesel engines or twin 200 hp outboard four-  stroke gasoline engines. Even though 36% of the vessels engaged in whale-watching are recreational (Osborne et al., 2002), no information is available on their engine configurations. Thus only commercial whale-watching vessel engine configurations were considered in the model.  2.2.1.5 Engine Emission Factors When measured engine emission rates are not available, published engine emission factors can be used (BCME, 2006). The USEPA has published emission factors for HCs, CO, NON, and PM in grams per horsepower-hour (g hp’ hf’) for zero-hour (i.e. new engine), steady-state, non-road gasoline and diesel engines of varying hp (USEPA, 2004a; 2004b). To obtain emission factors for the “average” whale-watching vessel engine, this study utilized the USEPA emission factors for non-road diesel engines with a power rating of 175300 hp, and for four-stroke outboard gasoline engines with a power rating greater than 175 hp (TJSEPA, 2004a; 2004b). The USEPA states that all PM emissions are assumed to be smaller than 10 Jtm, and 92-97% of the PM is assumed to be smaller than 2.5 p.m. Since 1998 the US EPA has been gradually phasing in emissions standards for marine diesel and gasoline engines to reduce their emissions over pre-control levels. In Canada, new engines beginning with the model-year 2001 must comply with the USEPA emission standards (EC, 2005). To account for the effect of federal emission standards, the USEPA produced emission factors for both pre-pollution control recreational marine engines (pre 2006), and post-pollution control recreational marine engines (post-2006) (USEPA, 2004a; 2004b). The USEPA emission factors for pre-2006, twin 200-hp diesel, gasoline, wet and  51  dry exhaust engines were averaged to provide the air pollution emission factors for an “average” whale-watching vessel. The post-2006 pollution control emission factors were also calculated to determine how pollution control devices on the engines would affect the air pollutant concentrations the killer whales are exposed to. The diesel and gasoline emission factors used in the dispersion model can be seen in Tables 2.5 and 2.6 respectively. Table 2.5: USEPA non-road model emission factors for pre and post-2006 model recreational marine diesel engines with power ratings less than 175 to 300 hp. Emission factors Air pollutant Post-2006 hp’ ) 1 hr Pre-2006 (g 0.14 0.22 HC 0.95 0.95 CO 0.11 0.16 PM 4.78 6.67 NO -  Table 2.6: USEPA non-road model emission factors for pre and post-2006 model recreational marine gasoline engines. Inboard (all 4-stroke 2-stroke outboard Air ratings) power outboard hp) 175 pollutant (> (g hp’ hr’) (> 175 hp) Post- Pre- Post-2006 PrePost-2006 Post-2006 Pre- Post-2006 (with (with (with (with 2006 2006 2006 2006  HC CO PM NO  128.7 313.3 7.7 4.5  carburator & ignition alterations)  alterations & catalyst)  electronic fuel injection)  115.6 289.4 7.7 8.2  62.7 246.2 7.7 3.7  18.7 242.5 7.7 8.2  electronic fuel injection)  7.5 258.1 0.06 9.0  n/a n/a n/a a 7 n  5.9 153.7 0.06 5.4  3.0 71.8 0.06 8.5  Transient adjustment factors (TAF5) are applied to emission factors to take into account different engine loads and/or speeds; however, the USEPA did not apply TAFs to the recreational marine engine category due to lack of information, thus it is assumed that marine engines operate at steady-state (USEPA, 2004a; USEPA, 2004b). Steady-state engine operating conditions can produce emissions that are quite different from those produced  52  under transient operating conditions, for example acceleration dramatically increases PM emissions (Graskow, 2001; USEPA, 2002a). Thus the PM emissions calculated in the dispersion model may be lower than real world situations where engines operate under transient conditions. 2 The NO emitted by marine engines is made up of nitrogen oxide (NO) and NO ; thus, a 2 (Quan et al., 2002b). However, ambient air quality objectives only apply to NO conversion factor is required (BCME, 2006). The British Columbia Ministry of Environment 2 is typically 5-10% of the NO concentration, while the (BCME, 2006) states that NO USEPA (2000) uses a mean emission rate of NO to NO ratio of 0.94 (SD  =  0.03), as does  Lloyd’s Register of Shipping (Lloyd’s, 1995). Thus NO accounts for approximately 94% 2 6% of the total NO emitted, and this conversion factor was used for the NO and NO concentrations predicted by the dispersion model. Air quality dispersion models are usually used to estimate the incremental change in pollutant concentrations resulting from specific sources (BCME, 2006). However, the atmosphere always has an ambient/background concentration of air pollutants from natural and anthropogenic sources not included in the model because they add unnecessary complexity (BCME, 2006). Thus a simple equation is used to determine total air quality where, Total  =  Background  +  Predicted Increment (contribution from modeled emission)  (BCME, 2006). Typically a single background value is used, and when modeling worst-case situations (where a conservative estimate of the impacts are preferred) a conservative background concentration rather than the maximum should be used (BCME, 2006). The background concentrations used in the dispersion model were from the Christopher Point  53  ambient air quality monitoring station on Vancouver Island, BC, during May to September of ) 3 2005-2007 (Table 2.3) (WLAP, 2008). Average ambient concentrations of CO (0.71 mg m ) simply need to be summed with the concentrations predicted by the 3 2 (0.008 mg m and NO dispersion model to obtain the total air pollution exposure.  2.2.2 NETLOGO DISPERSION MODEL The NetLogo program interface consists of two main components: space and agents (Wilensky, 1999). The space (also called the world) is the physical environment where the agents are situated and interact; in this case it represents the ocean surface. The space is automatically split into positive and negative x and y quadrants, and I further divided the quadrants into patches 2 m by 2 m square. World wrapping was allowed, so that the vessels and whales were replaced on one side of the space as they disappeared on the other; however, pollution did not wrap and the patches on the edge of the domain removed the pollution as if on an infinite plane. The programming code for the NetLogo dispersion model can be found in Appendix E. The two types of agents in the model were the whale-watching vessels and the whale. The movements of the whale and whale-watching vessels were not based on empirical trajectories; instead the whale was instructed to swim in a straight-line trajectory towards the , which is the published average swimming 1 right of the world at a constant speed of 2.85 m s speed of an adult male killer whale (Kriete, 2002). Since each patch was 2 m x 2 m, it resulted in time-steps (tick) of 0.7 seconds. Straight-line trajectories are representative of northern resident killer whale behaviour when more than three vessels are present within  54  1000 m of whales (Williams & Ashe, 2007). It is expected that the SRKWs would behave in a similar manner (Williams & Ashe, 2007). Since the whale and whale-watching vessels were all moving to the right of the space, the vessels were placed above and below the whale in uniformly spaced rows to simulate paralleling, which is the method of whale-watching recommended by the Be Whale Wise Guidelines (DFO, 2008b). As seen in Figure 2.14, the first row of vessels on either side of the whale were set at the buffer distance variable (the distance vessels maintained from the whale), and the distance between vessels was set by the inter-vessel distance variable. The vessels remained in the same position relative to each other and the whale for the duration of the simulation, and the number of vessels remained constant. The vessels moved at the same speed as the whale, 2.85 m s_i (5.5 knots); thus, the vessel speed was slower than the Be Whale Wise Guideline that recommends a vessel speed less than 3.6 m s_i (7 knots) when within 400 m of whales (DFO, 2008b).  55  Figure 2.14: Image of the NetLogo interface. The whale is at the center of the world and is colored green. The 20 whale-watching vessels are shaped as colored triangles, arranged in two uniformly spaced rows on either side of the whale. The vessels emit blue pollution plumes that are moved downwind at an angle of 240°. North (0°) is at the top of the image, and during simulations the whale and vessels moved to the right, or East (90°). Each vessel moved forward one patch per time-step and had its air pollution emission rate set at 70.2 mg per time-step (equivalent to 100 mg s’). The model was programmed to calculate the air pollutant concentration (equation below) in the whale’s patch at each timestep. Because ambient concentrations scale linearly with emission rate the 100 mg s’ is essentially a dummy pollutant emission rate, which can be multiplied by the USEPA emission factors to obtain specific pollutant concentrations. To calculate the emission rate from the emission factor provided by the USEPA, the emission factor (in g hp’ hr’) was multiplied by the rated horsepower of the engine, which provided the grams of pollutant emitted per hour (g hr’). This was then converted to milligrams emitted per second (mg s’)  56  and divided by the dummy air pollutant emission rate of 100 mg s to obtain the multiplication factor for each air pollutant. The multiplication factors for different marine engine configurations can be seen in Table 2.7. These multiplication factors simply need to be multiplied by the dummy air pollutant concentrations predicted by the dispersion model to find modeled air pollutant concentrations from different engine configurations.  Table 2.7: Air pollutant multiplication factors for different marine engine configurations. PM HC 2 NO CO Engine configuration* 0.10 3.41 0.47 129.50 1 0.08 3.38 0.41 129.50 2 0.18 0.24 0.44 1.06 3 0.07 8.29 0.60 286.78 4 0.11 0.15 0.35 0.84 5 0.04 4.97 0.47 229.42 6 0.12 0.16 0.32 Same as pre-2006 7 Same as pre-2006 218.89 0.45 288.22 8 0.07 3.33 0.57 79.78 9 *  1 2 3 4 5 6 7 8 9  Engine configurations: “Average” vessel: pre-2006 twin 200-hp 4-stroke diesel and gasoline, wet and dry exhaust engines = “Average” vessel: post-2006 twin 200-hp 4-stroke diesel and gasoline, wet and dry exhaust engines Diesel, pre-2006 twin 200-hp, inboard engine, dry exhaust only Gasoline, pre-2006 twin 200-hp, 4-stroke engine, dry exhaust only Diesel, pre-2006 twin 200-hp, inboard engine, wet exhaust only Gasoline, pre-2006 twin 200-hp, 4-stroke engine, wet exhaust only Diesel, post-2006 twin 200-hp, inboard engine, dry exhaust only = Gasoline, post-2006 twin 200-hp, 2-stroke engine, dry exhaust only = Gasoline, post-2006, twin 200-hp inboard engine, dry exhaust only  The most widely used air dispersion models are based on the Gaussian dispersion equation, which has a number of assumptions and limitations. Gaussian models calculate hourly air pollutant concentrations with uniform meteorological conditions within the modeling domain; they do not account for curved plume trajectories or variable wind conditions; they assume the emission plume originates from a point source rather than a mobile source; they assume that exposure occurs at the plume centerline; they have an  57  inverse dependency on wind speed, thus a low wind speed limit that imposes unacceptable bias during stagnant atmospheric conditions; they overestimate air pollutant concentrations under stable atmospheric conditions; and the calculated concentrations are only within ± 50% of actual values due to the restrictive conditions under which the dispersion parameters for the equation were developed (Arya, 1999; BCME, 2006; Cooper & Alley, 2002; Mohan & Siddiqui, 1997). Thus instead of using a Gaussian dispersion equation, the exhaust emissions were dispersed in the model by diffusion and advection, which can be treated separately. The “diffuse” function in NetLogo captures isotropic molecular diffusion, and programs each patch containing pollution to share a percent of its pollution with its eight neighboring patches. The percent of pollution shared is called the diffusion constant variable, and varies between zero and one, thus small values produced narrow concentrated pollution plumes and large values produced fanning diluted plumes. The diffuse function spread the pollutants perpendicular to the wind direction, and the amount of diffusion into the neighboring patches was conserved (except the edge patches where the pollution was removed). Advection was captured in the model by adding wind that moved the pollution downwind, and advection and diffusion created a crosswind plume that grew in width over time. A pollution deposition function was not included in the model because deposition is insignificant for time periods of 1-hour or less (D. Steyn, pers. comm., January 2008), and deposition over water for the pollutants under consideration has not been quantified empirically. The dispersion model contained an equation to calculate the dummy air pollutant concentration in the patch of the whale each time-step, expressed as (D. Steyn, pers. comm., October 2007):  58  e g * Zm  * S  , e is the emission rate in mg 3 where C is the concentration in mg m  s  (equal to 100 mg s’),  g is the patch size that the pollutant disperses into each time-step (set at 2 m, which is the transom length of an average whale-watching vessel), Zm is the vertical pollutant mixing ). Thus g, s, and Zm define 1 1 (equal to 2.85 m s height in m, and s is the vessel speed in m s the volume the pollutant disperses into. The formula to calculate air pollutant exposure is simply (NRC, 1991):  E=c*t ), and t is time (hours). 3 ), c is the concentration (mg m 3 where E is the exposure (mg hr m The length of time was 1-hour since most air pollutant standards (e.g. the MV AQOs for CO ) are based on 1 -hour exposures. 2 and NO A sensitivity analysis was conducted on all the variables included in the model, and two versions of the dispersion model were employed: a completely deterministic version without any random elements, and a stochastic version where the wind angle randomly changed every time-step around a normally distributed mean wind angle. The deterministic version was run until the pollutant concentration the whale experienced reached a steady th 47 time-step, and for consistency the concentration at the state, usually around the  th 100  time-step was taken as the value for each simulation. The world size for the deterministic simulations was 500 by 500 patches (equal to 1,000 mby 1,000 m), and because the simulations were of short duration the agents did not wrap around to the other side of the world, and to obtain 1-hour exposures the predicted air pollutant concentrations were multiplied by one. The stochastic simulations ran for 1-hour of real time (2520 time-steps), thus the predicted air pollutant concentrations at each time step were averaged to obtain 1-  59  hour exposures. The world size in the stochastic simulations was set at 600 by 400 patches (1,200 m by 800 m), and because of the longer duration of the stochastic simulations (2520 time-steps versus 100 time-steps) the agents did wrap around to the other side of the world. While the whale was wrapping around the world it was not exposed to pollution because the edge patches deleted the pollution, and this was taken into account when calculating the average concentration by running the simulations for 2870 time-steps and removing the 60 time-step concentrations that occurred during the transitions (the whale wrapped around five times during the 1-hour simulations). Additionally the first 50 time-steps were removed because it took approximately that long for the pollutants to reach the whale once the simulations began. Multiple repetitions of each variable setting were not required for the deterministic version of the model since there were no random elements. A repetition of a stochastic simulation was conducted and the absolute difference between the two trials was 3 while the relative difference was 1.5%; because this was such a small 0.00247 mg m difference further repetitions were not conducted. There are two types of variables in the model: model structural variables (i.e. diffusion constant) and model representations of real world qualities (i.e. wind speed, wind angle, the vertical mixing height of pollutants in the atmosphere, the buffer distance, the inter-vessel distance, and the number of vessels). Each variable was fixed at a default value, except for the variable being analyzed, which experienced a realistic range of values with at least six increments within that range. The default and range of values used in the sensitivity analysis for each variable are listed in Table 2.8. In addition to the sensitivity analysis, simulations were run with the variables at average-case and worst-case whale-watching values.  60  Table 2.8: Variables in the dispersion model with their default and range of values. Range Default Model variables 5.7 0.71 13.54 Wind speed (m s’) 180 90 270 Wind angle (°) 0 1 0.5 Pollutant diffusion constant 9.5 0.5 3 (m) of pollutant height Vertical mixing 2 122 60 Buffer distance to whale (m) 1 121 20 Number of vessels 4 104 50 Inter-vessel distance (m) —  -  -  -  -  -  -  Meteorological data are necessary in all air quality dispersion models, and this can be in the form of hourly data or a matrix of combinations of realistic meteorological conditions, and in this dispersion model a matrix was used (BCME, 2006). Wind angles coming from one direction produced identical pollutant concentrations as the wind angle they mirrored (i.e. 30° was equivalent to 150°); thus, only angles from one side of the compass rose were considered for the deterministic simulations (i.e. 90°, 120°, 150°, 180°, 210°, 240°, 270°). A constant wind speed and direction can be assumed for time periods equal to 1-hour or less (D. Steyn, pers. comm., January 2008), thus the deterministic simulations kept wind speed and direction constant. However, the effect of a changing wind angle was explored in the stochastic simulations, as the wind angle randomly changed every time-step around a normally distributed mean wind angle and within a standard deviation of the wind direction fluctuations (sigma theta, r) equal to 22.5°. This value for the standard deviation of the wind direction fluctuations is expected in stable to slightly stable atmospheres for time periods of 1-hour or less (Hanna et al., 1982). Three mean wind angles were considered in the stochastic simulations, the angle that typically produced the lowest concentrations in the deterministic simulations (90°), the angle that typically produced medium concentrations (150°), and the angle that typically produced the highest concentrations (210°). The British Columbia Ministry of Water, Land, and Air Protection (WLAP) Air Resources Branch Web  61  Service provides measured wind direction fluctuations at the Christopher Point monitoring station on Vancouver Island, BC. In 2007 from May to September, the average wind direction fluctuation (crn) was  150,  with the highest value of 19.4° occurring in September  (WLAP, 2008). Thus the wind direction fluctuation value of 22.5° used in the dispersion model was higher than the average observed at Christopher Point. The average wind direction at Christopher Point ranged from 220°-253° (winds from the southwest) from May to September 2005-2007 (WLAP, 2008). The same was true for the Hem Bank Buoy from May to September from 2004-2007, as the average direction the wind came from varied from 225°-238° (winds from the southwest) (NOAA, 2008). Wind was progranmied in the model to move the pollution from all patches with the same direction and speed. The minimum wind speed was 0.71 m speed of anemometers, and the maximum was 13.54 m  the approximate stall  (equivalent to 26.32 knots or  Beaufort scale 6); at this maximum wind speed the wind is described as a strong breeze that produces large 3 m waves on the sea surface (Barnes-Svarney, 1995). At wind speeds greater than 10 m s’ the lower atmosphere will no longer be stable but wind driven mixing will render it neutral. In these wind and wave conditions it is likely that whale-watching vessels would be unable to find and follow killer whales, thus greater wind speeds were not considered. At Christopher Point the average wind speed during the peak whale-watching months in 2007 was 8.55 m s’, with the highest wind gust of 35 m s’ occurring in May (WLAP, 2008). Because the Christopher Point station is at the interface between the land and water, the wind speeds may be higher than those over water. At the Hem Bank Buoy the average wind speed from May to September from 2004-2007 was 4.7 m s and the daily average maximum wind speed was 17.2 m s (NOAA, 2008). Thus, the default wind speed  62  in the dispersion model (5.7 m s’) was slightly higher than the average wind speed at the Rein Bank Buoy, but less than the average wind speed at the Christopher Point station. Under most conditions it is expected that the vertical air pollutant concentration profile within the first few meters of the surface is reasonably homogeneous due to mechanical mixing (BCME, 2006), and the model assumed that the profile of vertical pollution diffusion initially decreased rapidly with height and then leveled off. The vertical mixing height in the dispersion model was independent of patch size, and instead was 1-10 times the average vessel transom length (2-20 m) due to the aerodynamic drag of the moving vessel. Drag from the vessel produces a turbulent wake in which the pollutants mix, and is influenced by the shape and speed of the vessel (RET, 1988). Because of the reduced vertical mixing over cool water bodies, the maximum mixing height considered was only 9.5 m since heights greater than 5 m are unlikely (D. Steyn, pers. comm., January 2008). Elevated vertical mixing heights quickly disperse air pollutants into large volumes of air, and produce low downwind air pollutant concentrations; whereas low mixing heights can trap pollutants, giving rise to high pollutant concentrations (BCME, 2006). The Be Whale Wise Guidelines (DFO, 2008b) advise vessel operators to maintain at least 100 m from whales, thus the buffer distance variable ranged from 2-122 m. The number of vessels variable ranged from 1-121, as the maximum number of vessels ever observed around the SRKWs is 120 (Osborne et al., 2002). The default number of vessels was set at 20 as that is the average number of vessels around the SRKWs during the summer months (Bain, 2002; Baird, 2001; Erbe, 2002; Koski, 2006; Osborne et al., 2002; 1999). The inter-vessel distance varied from 4-104 m, as vessel operators try to maintain a 100 m distance between vessels (M. Malleson, pers. comm., February 1, 2008), and due to vessel  63  size it was assumed that vessels would not approach each other closer than 4 m. Vessel count data was gathered from the Straitwatch Boater Education vessel during June, July, and August 2008, and was analyzed for inter-vessel distances at the Marine Communications and Traffic Services radar station. However, the resolution of the radar system was not accurate enough to provide this detailed information. Most air quality dispersion models also take surface roughness into account, due to the complex turbulence patterns created by topography; however, models recommended by the British Columbia Ministry of Environment (2006) set the surface roughness over water to 0.000 1 m, and because that is such a small value it was not included in the dispersion model. Additionally, average wave heights recorded at Hem Bank from May to September in 20042007 only varied from 0.29-0.48 m (NOAA, 2008). Since direct validation of the air pollutant concentrations calculated by the dispersion model was not feasible, the assumptions were made as close to reality as possible. Sources of uncertainty in the model were minimized by error checking, and by correcting odd model behaviour. However, it must be recognized that air dispersion model predictions have “inherent uncertainty” from unknown turbulent atmospheric processes that cannot be resolved (BCME, 2006).  2.2.3 NETLOGO DISPERSION MODEL RESULTS The air pollution concentrations predicted by the dispersion model were standardized as 1-hour exposures, and are from the patch the whale was in. The graphs that follow depict the results of the average-case and worst-case dispersion model simulations, and have been scaled to the engine emission factors for the “average” whale-watching vessel previously  64  described. To see how different engine configurations affect the exposure concentrations refer to the multiplication factors in Table 2.7. All the graphs have the CO concentration on 2 concentration on the right y-axis, along with reference lines that the left y-axis, and the NO ) for 3 2 (0.2 mg m ) and NO 3 indicate the MV AQOs for 1-hour of exposure to CO (30 mg m comparison. The MV AQOs for other air pollutants have longer exposures (8-hour, 24-hour, and annual), thus they were not compared. Human air quality standards were included to put the exposures into context, and are converted into killer whale equivalents in Chapter three.  2.2.3.1 Results of the Sensitivity Analysis All graphs depicting the results of the sensitivity analysis simulations can be found in Appendix F. The main difference between the deterministic and stochastic versions of the dispersion model was that adding random wind angle fluctuations tended to smooth out the peaks and valleys in the air pollutant concentrations (i.e. lower maximum concentrations and higher minimum concentrations). If the whale was exposed to a direct exhaust plume from a vessel in the deterministic version, the plume would remain directly over the whale for the entire simulation, resulting in a high exposure. Whereas in the stochastic version, the wind angle fluctuations caused the exhaust plumes to move every time-step so that the whale never experienced a direct plume for long, resulting in lower average exposures. The concentrations obtained in the stochastic and deterministic versions of the models were within the same order of magnitude. The wind angle variable produced the greatest extremes in air pollution concentration that the killer whale was exposed to, for both the deterministic and stochastic versions of the dispersion model. However, after wind angle, the ranking of variables that had the greatest  65  effect were slightly different for the deterministic and stochastic simulations. The second most important variable was either buffer distance or mixing height, followed by the number of vessels or the inter-vessel distance, and finally the wind speed. When the whale was downwind of the vessels it received the highest air pollutant exposures. Wind angles of 2100 and 240° consistently produced the highest air pollutant concentrations, with 210° being the highest 67% of the time and 240° 33% of the time. When the wind came from angles of 2100 and 240°, the air pollution was pushed downwind in roughly the same direction as the whale (to the right of the world). Thus at these wind angles the whale often experienced both direct exhaust plumes from whale-watching vessels and indirect diffused pollution, which resulted in the whale receiving the highest air pollutant concentrations. The wind angles that produced the highest air pollutant concentrations after 210° were ranked: 240° was second highest 67% of the time; 180° was third highest 67% of the time; 150° was fourth highest 83% of the time; 120° was fifth highest 67% of the time; 270° was sixth highest 83% of the time; and 90° was the lowest 83% of the time. In the deterministic model, when the wind came from directly ahead of the whale (90°) or directly behind the whale (270°), the whale was not exposed to direct exhaust plumes from vessels, and very limited pollution from diffusion reached the whale, thus the air pollutant concentration the whale experienced was virtually zero at these angles. However, when the wind angle randomly changed in the stochastic version of the model, it resulted in higher air pollutant concentrations at wind angles of 90° and 270°, because diffused pollution ended up reaching the whale.  66  As wind speed increased it generally produced a roughly unimodal trend in the modeled air pollution concentration in the patch the whale was in (Figures Fl and F2 in 1 produced air pollutant concentrations that Appendix F). The lowest wind speed of 0.7 m s 1 the air pollution concentration was often much were essentially zero; however, by 2 m s higher, except for wind angles of 900, 1200 and 270°, which had air pollution concentrations that were essentially zero at all wind speeds. The highest air pollution concentrations occurred at wind speeds between 2-7 m s, and as the wind speed increased above 7 m s’ the air pollution concentration decreased (except at a wind angle of 150°, which had a maximum concentration at approximately 11 m  1)  Increasing the wind speed caused the air pollution  plumes to bend over as the pollution got swept down wind at a faster rate, which changed the angle of plume spreading and resulted in higher air pollutant concentrations. Thus the peaks in air pollutant concentration at the different wind angles were due to exhaust plumes from vessels moving directly into the whale’s path as wind speed increased. The diffusion constant variable had opposing effects on the air pollutant concentration the whale experienced, depending whether or not the whale received air pollution from a direct vessel exhaust plume or from diffusion alone (Figures G3 and G4 in Appendix F). A large diffusion constant resulted in a high air pollutant concentration if the whale was not in a direct exhaust plume (i.e. at wind angles of 90°, 120°, 150°, 270°), because a large diffusion constant spread the pollution so that some eventually reached the whale. The opposite was true if the whale was in a direct exhaust plume from a vessel: a low diffusion constant (e.g. 0.2) exposed the whale to highly concentrated plumes and as the diffusion constant increased the plume was diluted, exposing the whale to lower concentrations.  67  The air pollutant concentration initially displayed a rapid decline as the vertical mixing height increased, but it then leveled off without approaching zero (Figure F5 and F6 in Appendix F). The amount of air pollution the whale received was dependent on whether the whale was in a direct exhaust plume from a vessel or if it was only receiving pollution that diffused toward it. The air pollution concentration tended to be inversely proportional to the buffer distance (Figure F7 and F8 in Appendix F). Oscillations in the pollution concentration occurred because the whale was either in a direct exhaust plume or not due to changing vessel positions and spacing as the buffer distance increased. This variable was unusual because the 2700 wind angle produced the second highest air pollutant concentration, whereas with the other variables that angle produced one of the lowest air pollution concentrations. The 150° wind angle was also unusual as it produced the lowest air pollutant concentration, but with the other variables it produced mid-range concentrations. The air pollution concentration generally increased with the number of vessels (Figure F9 and FlO in Appendix F). The stepwise increases in concentration were due to the addition of rows of vessels as the number of vessels increased. The inter-vessel distance variable did not produce a consistent trend at all wind angles; however, the air pollution concentration generally decreased with increasing inter vessel distance (Figure Fl 1 and Fl2 in Appendix F). At inter-vessel distances less than 10 m the air pollutant concentration was very low because the wind moved the pollution away before it had time to reach the whale. The oscillations in air pollution concentration occurred because the whale was either in a direct exhaust plume or not due to changing vessel positions and spacing as the inter-vessel distance increased.  68  The results of the sensitivity analysis illustrated that the dispersion model captured important dynamics of air pollutant dispersion, and the following sections (2.2.3.2 and 2.2.3.3) will investigate specific phenomena/situations of interest.  2.2.3.2 Results of the Average-case Trials Further deterministic simulations were run with the variables set at average-case whale-watching values, rather than the default values used in the previous sensitivity analysis simulations. The buffer distance, inter-vessel distance, diffusion constant, and number of vessels variables remained constant, while the wind speed and mixing height variables experienced their range of values listed in Table 2.8. The buffer distance was set at 100 m, the distance recommended by the Be Whale Wise Guidelines (DFO, 2008b). The intervessel distance was also set at 100 m, which is the distance whale-watch vessel operators try to maintain between vessels (M. Malleson, pers. comm., February 1, 2008). The diffusion constant was set at 0.5, and the number of vessels was set at 20, which is the average number of vessels around the SRKWs during the summer months (Bain, 2002; Baird, 2001; Erbe, 2002; Koski, 2006; Osborne et al., 2002; 1999). This average-case scenario set-up caused the furthest vessels from the whale to be at a distance of 316 m, which is much closer than the average real-world whale-watching conditions of 20 vessels within 800 m (Bain, 2002; Baird, 2001; Erbe, 2002; Koski, 2006; Osborne et al., 2002; 1999). Thus the average-case simulations may have overestimated exposures since the vessels were closer to the whale. Under average-case whale-watching conditions the wind speed variable produced CO concentrations that exceeded the MV AQO only at wind angles of 210° and 240°; while the 2 was never exceeded (Figure 2.15). The CO MV AQO was only exceeded MV AQO for NO  69  when the air pollution concentrations peaked, generally at wind speeds between , thus it is the mid-range wind speeds that are most problematic. The 1 approximately 1-9 m s 3 occurred at a wind angle of 240°, highest dummy air pollutant concentration of 0.33 mg m and wind speed of 2.1 m  s.  This maximum concentration is lower than both the  deterministic and stochastic versions of the model, when run with default values rather than average values. The mean CO concentration predicted by the average-case simulations was 3 (SEM = 2 concentration was 0.04 mg m 3.63), and the mean NO  3 (SEM 11.98 mg m  . 2 0.01). These means are well under the MV AQOs for both CO and NO  60.0 0.20 50.0  c  0.15  40.0 U  r\  /  \  30.0  ill \  \.  /\  /  20.0  !‘  !/  /  \  •I// \‘ \\ j /1 • \ 1/1.7 \  10.0  o.o. 0.0  \\  /  /  C  8 o  !\  \,z  \ / ,• v’-——’ /  U  /  /  /  \  /  \  -  8.0 6.0 Wind speed (mis)  ——-————  ——0---—  —-—4----— —  •O  o  —--—v———  —  N’. N\  \  •  —  0.05 Z  ‘.  ,-‘  4.0  8  \ \  /  • 2.0  0.10  \  —  I  10.0  • 12.0  0.00 14.0  90° 120° 150° 180° 210° 240° 270° ) 3 CO MV AQO (30 mg/rn ) 3 2 MV AQO (0.2 mg/rn NO  2 concentration as a function of wind speed and angle under Figure 2.15: CO and NO average-case whale-watching conditions.  70  Under average-case whale-watching conditions the vertical mixing height variable produced CO concentrations that exceeded the MV AQO at wind angles of 150°, 180°,  2100,  and 2400 when the mixing heights were less than approximately 1.5, 2.3, 6, and 1 m 2 concentrations exceeded the MV AQO at wind angles respectively (Figure 2.16). The NO of 150°, 180°, and 210° when the mixing heights were less than approximately 1, 1.5, 3.7 m 3 occurred at a respectively. The highest dummy air pollutant concentration of 2.87 mg m wind angle of 210° and mixing height of 0.5 m. This maximum concentration is lower than that of the default deterministic version in the model, but higher than the stochastic version.  400.0  1.4  \  •1.2  300.0  \ C  C  0.8 .2  \  200.0  0.6o  C.) C  C  o0  0 0  1  0.4  o.o  Z 0.2 —  0.0  0.0 0.0  6.0 Mixing height (m) 4.0  2.0  0  ———v——— —..—-•—  —--*— —  —0---  —  —---——  —  —  —  —  8.0  10.0  90° 120° 150° 180° 210° 240° 270° ) 3 Co MV AQO (30 mg/rn ) 3 NO, MV AQO (0.2 mg/rn  2 concentration as a function of the vertical mixing height and wind Figure 2.16: CO and NO angle under average-case whale-watching conditions.  71  2.2.3.3 Results of the Worst-Case Trials Additional deterministic simulations were conducted with reasonable “worst-case” values for the variables, which were within the bounds of whale, vessel, and atmospheric behaviour. For these simulations the buffer distance was set at 50 m, which is half the distance recommended by the Be Whale Wise Guidelines (DFO, 2008b). The inter-vessel distance was also set at 50 m, which is half the distance whale-watching vessel operators try to maintain from each other (M. Malleson, pers. comm., February 1, 2008). The number of vessels was set at 40, which is double the average number of vessels (Bain, 2002; Baird, 2001; Erbe, 2002; Koski, 2006; Osborne et al., 2002; 1999). The vertical mixing height was set at 2 m, and the diffusion constant was set at 0.5, and the wind speed varied from 2.6-6.7 . 1 m s  The two wind angles that produced the highest air pollutant concentrations in the  deterministic version of the model, 210° and 240°, were used in the worst-case simulations. 2 were exceeded at all wind speeds, and both wind The MV AQO5 for CO and NO angles in the worst-case simulations (Figure 2.17). The highest dummy air pollutant . 1 3 occurred at a wind angle of 210° and wind speed of 5.4 ms concentration of 1.35mg m 3 (SEM The mean CO concentration predicted by the worst-case simulations was 76.28 mg m 3 (SEM = 0.06). Both of these 2 concentration was 0.27 mg m 15.44), and the mean NO means (especially that for CO) are above the MV AQOs.  72  180.0 0.6 160.0  0.5  0)  g 120.0 C 0  0.4  100.0 80.0 C  60.0 40.0 ::  20.0 0.1  0.0 2.0  5.0  4.0  3.0  6.0  7.0  Wind speed (mis) 2100 240°  0 —  —  — —  ) 3 CO MV AQO (30 mg/rn ) 3 NO MV AQO (0.2 mg/rn 2  2 concentration as a function of the wind speed and angle under Figure 2.17: CO and NO worst-case whale-watching conditions.  2.2.4 DISPERSION MODEL CONCLUSIONS The results from the sensitivity analysis simulations suggest that the wind angle had the largest effect on the concentration of air pollutants the killer whale was exposed to, with downwind angles (210° and 240°) producing the highest concentrations. The second most important variable was either buffer distance or mixing height, followed by the number of vessels or the inter-vessel distance, and finally the wind speed. The deterministic version of the model produced higher maximum concentrations than the stochastic version, except for the inter-vessel distance variable. Under average-case conditions when the 20 vessels maintained the recommended 100 2 could be m distance from the whale and each other, the MV AQOs for CO and NO  73  2 were always exceeded. While in worst-case simulations the MV AQOs for CO and NO exceeded. The average CO concentration predicted by the average-case simulations was . This average 3 , with the highest concentrations ranging from 42.6-372 mg m 3 11.98 mg m 3 range of CO concentrations concentration of CO is five times greater than the 2.0-2.5 mg m measured 30 m from a busy Los Angeles highway (Zhu et al., 2002b). The average , which is 34 3 concentration of CO predicted by the worst-case simulations was 76.28 mg m times greater than the average concentration of CO measured by Zhu et al. (2002b). These results are not that surprising since the zone around non-road engines often have exhaust emissions that are the same or double those measured around busy roadways (USEPA, 2002b; 2004c). 2 concentration predicted by the average-case simulations was 0.04 The average NO 2 . This average NO 3 , with the highest concentrations ranging from 0.15-1.34 mg m 3 mg m ) measured 3 2 concentrations (0.032-0.037 mg m concentration is just above the range of NO 2 at a distance of 115 m from busy motorways (Roorda-Knape et al., 1998). The average NO , which is 7.8 times 3 concentration predicted by the worst-case simulations was 0.27 mg m 2 concentrations measured by Roorda-Knape et al. (1998). This elevated greater than the NO level of NO 2 is not that surprising since 22% of NO in the LFV Airshed in the year 2000 emission inventory was attributed to marine vessels. However, it indicates that under worst case whale watching conditions, the whales are experiencing very poor air quality episodes that are potentially harming their health. Therefore, even the average-case simulations predicted air quality (as indicated by ) to be worse (especially for CO) than that measured along busy highways, with 2 CO and NO worst-case predictions indicating extreme pollution episodes. Generally the MV AQOs for  74  2 were exceeded in the average-case simulations when: the wind came from an CO and NO angle of  1500,  180°, 210° or 240°; the wind speed was between 1-9 m s’; and the mixing  height was less than 6 m. However, the average-case simulations were run with only 20 vessels, and they all maintained 100 m from the whale and each other. The sensitivity 2 were generally exceeded when the analysis indicated that the MV AQOs for CO and NO buffer distance was less than 20 m; the number of vessels was greater than 27 on average; and the inter-vessel distance was less than 50 m. The simulations showed that shallow mixing heights produced high air pollutant concentrations, and stable atmospheric conditions (shallow mixing heights) predominate during the whale-watching season. Additionally, the average wind speed measured at the Hem Bank Buoy and Christopher Point during the summer months was 4.7 m s’ and 8.55 m s’ respectively (NOAA, 2008; WLAP, 2008), which is within the range of wind speeds that produced the highest air pollutant concentrations in the average-case simulations. Since the atmospheric conditions during the whale-watching season are highly conducive to air pollutant accumulation, the potential to exceed the MV AQOs is high. Therefore, to ensure air pollutant concentrations are below the MV AQOs, vessel operators need to maintain adequate distances from each other and the whales, the number of vessels around the whales should not exceed the average too often, and vessels should not be positioned upwind of the whales, especially at wind angles of 210° and 240° to the whale. Empirical roadway studies have shown that air pollutant concentrations decrease to approximately 50% of the original lOOm from the roadway (as explained in section 2.2.1). Therefore, the predicted air pollutant concentrations from the deterministic simulations that manipulated buffer distance were compared at distances of 2 m and 102 m. When the whale  75  was downwind (2100 and 240°), the concentration at 100 m decayed to approximately 38% of the original. When the wind was parallel to the whale (90° and 180°), the concentration at 100 m decayed to 24% of the original. Thus the model predicted emissions to decrease to a greater extent (on average to 31% of original) than those found in empirical roadway studies (—50%). This indicates that the dispersion model may have underestimated concentrations, as higher concentrations were expected because of the low mixing heights used. Obviously improved modeling or on the water air sampling would determine the accuracy of the calculated concentrations. In the dispersion model, the whale-watching vessels paralleled a single whale rather than a pod of whales, which is more representative of watching a lone transient killer whale rather than a pod of SRKWs. The average number of vessels following the SRKWs could have been divided by the number of individuals to find how many vessels watch a single whale on average; however, the number of vessels was a variable that was manipulated during the simulations, therefore this issue was taken into account.  76  2.2.5 REFERENCES Ainsley, B., & Styen, D. G. 2007. Spatiotemporal trends in episodic ozone pollution in the Lower Fraser Valley, British Columbia, in relation to mesoscale atmospheric circulation patterns and emissions. Journal ofApplied Meteorology, 46(10): 16311644. Angevine, W. M., Senff, C. J., White, A. B., Williams, E. J., Koermer, J., Miller, S. T. K, Talbot, R., Johnston, P. E., McKeen, S. A., & Downs, T. 2004. Coastal boundary layer influence on pollutant transport in New England. Journal ofApplied Meteorology, 43: 1425-1437. Anwar, Sk. M., Jeanneret, C. A., Parrott, L., & Marceau, D. J. 2007. Conceptualization and implementation of a multi-agent model to simulate whale-watching tours in the St. Lawrence Estuary in Quebec, Canada. Environmental Modelling & Software, 22: 1775-1787. Arya, S. P. 1999. Air Pollution Meteorology and Dispersion. New York: Oxford University Press. Bain, D. E. 2002. A Model Linking Energetic Effects of Whale Watching to Killer Whale (Orcinus orca) Populations. Friday Harbor, WA: Friday Harbor Laboratories, University of Washington. Baird, R. W. 2001. Status of killer whales, Orcinus orca, in Canada. Canadian FieldNaturalist, 115: 676-701. Barnes-Svarney, P. (Ed.) 1995. The New York Public Library Science Desk Reference. New York, NY: Stonesong Press Inc. Bousquet, F., & Le Page, C. 2004. Multi-agent simulations and ecosystem management: A review. Ecological Modelling, 176: 3 13-332. British Columbia Lung Association (BCLA). 2007. State ofthe Air in British Columbia. Vancouver, BC: British Columbia Lung Association. British Columbia Ministry of Environment (BCME). 2005. A Primer on the Guidelinesfor Air Quality Dispersion Modelling in British Columbia. Victoria, BC: British Columbia Ministry of Environment, Environmental Protection Division. British Columbia Ministry of Environment (BCME). 2006. A Primer on the Guidelines for Air Quality Dispersion Modelling in British Columbia. Victoria, BC: British Columbia Ministry of Environment, Environmental Protection Division. British Columbia Ministry of Water, Land and Air Protection (WLAP) Air Resources  77  Branch. 2008. Atmospheric Data andAir Quality Health Index Web Service. Retrieved May 27, 2008, from http:/!alOO.gov.bc.ca!pub/aqiis/archivel.report British Columbia Provincial Health Officer (BCPHO). 2004. Every Breath You Take. Provincial Health Officer’s Annual Report 2003. Victoria, BC: Ministry of Health Services. Brook, J. R., Strawbridge, K. B., Snyder, B. J., Boudries, H., Worsnop, D., Sharma, S., Anlauf, K., Lu, G., & Hayden, K. 2004. Towards an understanding of the fine particle variations in the LFV: integration of chemical, physical and meteorological observations. Atmospheric Environment, 38(34): 5775-5788. Clark, N. N., Nine, R. D., Mace, B. E., Morrison, R. W., Remcho, V. T., Lowe, P. C., & McLaughlin, L. W. 2000. Use of soy-derived fuel for environmental impact reduction in marine engine applications. Transactions of the ASAE, 43(6): 13 831391. Coates, S. W., & Lassanske, G. G. 1990. Measurement and analysis of gaseous exhaust emissions from recreational and small commercial marine craft. SAE Transactions, 99(3): 1873-1887. Cooper, C. D., & Alley, F. C. 2002. Air Pollution Control: A Design Approach. Prospect Heights, IL: Waveland Press, Inc. Department of Fisheries and Oceans Canada (DFO). 2008a. Data from BC Lighthouses. Pacific Region Ocean Sciences. Retrieved on January 17, 2008, from http://www-sci.pac. dfo-mpo. gc. ca!osap/datalSearchTools/Searchlighthouse_e.htm# Department of Fisheries and Oceans Canada (DFO). 2008b. Viewing Guidelines. Pacfic Region Marine Mammals and Turtles. Retrieved May 14, 2008, from http://www.pac. dfo-mpo.gc.calspecies/marinemammals/view_e.htm Environment Canada (EC). 2004. Characterization of the Georgia Basin/Puget Sound Airshed. Retrieved September 12; 2007, from http :I/www.pyr. ec. gc.caJair/gbps_airshed/summary_e.htm Environment Canada (EC). 2005. Smoke on the Water. The Science and Environment Bulletin. Retrieved on October 23, 2005, from http://www.ec.gc.caJscience/sandemay00/article 1 _e.html Environment Canada (EC). Unpublished. Emissionsfrom 4 Dfferent Light Duty Vehicle Technologies Operating on Low Blend Ethanol Gasoline, Report #04-27 A: Tailpipe 0). Ottawa, ON: Environment Canada, Science & 2 , N 4 , CH 2 Greenhouse Gases (C0 Technology Branch. Environment Canada (EC). Unpublished. Emissionsfrom 4 Dfferent Light Duty Vehicle  78  Technologies Operating on Low Blend Ethanol Gasoline, Report #04-27 B: Tailpipe Regulated & Unregulated Gaseous Emissions (CO, NON, THC, NMHC, NMOG, Ethanol, Carbonyls, VOC). Ottawa, ON: Environment Canada, Science & Technology Branch. Erbe, C. 2002. Underwater noise of whale-watching boats and potential effects on killer whales (Orcinus orca), based on an acoustic impact model. Marine Mammal Science, 18: 394-418. Frey, H. C., & Bammi, S. 2003. Probabilistic nonroad mobile source emission factors. Journal ofEnvironmental Engineering, 129(22): 162-168. Frumkin, H., & Thun, M. J. 2001. Diesel Exhaust. A Cancer Journal for Clinicians, 51: 193-198. Gabele, P. A., & Pyle, S. M. 2000. Emissions from two outboard engines operating on reformulated gasoline containing MTBE. Environmental Science and Technology, 34: 368-372. Geraci, J. R. & St. Aubin, D. J. 1990. Sea Mammals and Oil: Confronting the Risks. San Diego: Academic Press, Inc. Graskow, B. R. 2001. Design and Development ofa Fast Aerosol Size Spectrometer. University of Cambridge Ph.D. Thesis. Greater Vancouver Regional District (GVRD). 2002. 2000 Emission Inventory for the Lower Fraser Valley Airshed. Burnaby, BC: Greater Vancouver Regional District Policy & Planning Department. Hanna, S. R., Briggs, G. A., & Hosker, R. P. 1982. Handbook on Atmospheric Diffusion. Oak Ridge, TN: U.S. Department of Energy. Health Effects Institute (HEI). 1988. Air Pollution, the Automobile, and Public Health, A. Y. Watson, R. R. Bates, & D. Kennedy (Eds.). Washington, DC: National Academy Press. Health Effects Institute (I-LET). 1999. Diesel Emissions and Lung Cancer: Epidemiology and Quantitative Risk Assessment. A Special Report of the Institute’s Diesel Epidemiology Expert Panel. N. Andover, MA: Flagship Press. Hitchins, J., Morawska, L., Wol, R., & Gilbert, D. 2000. Concentrations of submicrometre particles from vehicle emissions near a major road. Atmospheric Environment, 34: 51-59. Hoff, R. M., Harwood, M., Sheppard, A., Froude, F., Martin, J. B., & Strapp, W. 1997. Use  79  of airborne lidar to determine aerosol sources and movement in the Lower Fraser Valley (LFV), BC. Atmospheric Environment, 31(14): 2123-2 134. International Programme on Chemical Safety (IPCS). 1996. Environmental Health Criteria 171: Diesel Fuel and Exhaust Emissions. Geneva: World Health Organization. Juttner, M. F., Backhaus, D., Matthias, U., Essers, U., Greiner, R., & Mahr, B. 1995. Emissions of Two- and Four-stroke Outboard Engines- I. Quantification of Gases and VOC. Water Resources, 29(8): 1976-1982. Kado, N. Y., Okamoto, R., Karim, J., & Kuzmicky, P. 2000. Airborne particle emissions from 2- and 4-stroke outboard marine engines: polycyclic aromatic hydrocarbon and bioassay analyses. Environmental Science and Technology, 34: 2714-2720. Kirrane, E., Loomis, D., Egeghy, P., & Nylander-French, L. 2006. Personal exposure to benzene from fuel emissions among commercial fishers: comparison of two-stroke, four-stroke and diesel engines. Journal ofExposure Science and Environmental Epidemiology, 17: 151-158. Koski, K. L. 2006. Soundwatch Public Outreach/Boater Education Project 2004-2 005 Final Program Report. Friday Harbor, WA: The Whale Museum. Kriete, B. 2002. Bioenergetic Changes From 1986 to 2001 in the Southern Resident Killer Whale Population, Orcinus orca. Friday Harbor, WA: Orca Relief Citizens’ Alliance. Lloyd’s Register of Shipping. 1995. Marine Exhaust Emissions Research Programme. London, UK: Lloyd’s Register Engineering Services. McKendry, I. G. 1994. Synoptic circulation and summertime ground-level ozone concentrations at Vancouver, British Columbia. Journal ofApplied Meteorology, 33: 627-641. 5 and Ozone in British . 2 McKendry, I. G. 2006. Background Concentrations ofPM Columbia, Canada. British Columbia Ministry of Environment: Vancouver, BC. Metro Vancouver (MV). 2006. 2006 Air Quality Reportfor the Lower Fraser Valley. Burnaby, BC: Metro Vancouver. Mohan, M., & Siddiqui, T. A. 1997. An evaluation of dispersion coefficients for use in air quality models. Boundary-Layer Meteorology, 84: 177-206. National Oceanic and Atmospheric Administration (NOAA). 2008. National Data Buoy Center, Station 46088 New Dungeness, WA (Hem Bank). Retrieved on May 26, 2008, from http://www.ndbc.noaa. gov/station_history.php?station4608 8 -  80  National Research Council (NRC). 1991. Human Exposure Assessmentfor Airborne Pollutants. Washington, DC: National Academy of Sciences. Northeast States for Coordinated Air Use Management (NESCAUM). 1999. The Health Effects of Gasoline Constituents Attachment 1. Boston, MA: Northeast States for Coordinated Air Use Management. —  Oke, T. R. 1987. Boundary Layer Climates (2’’ ed.). Cornwall, UK: Methuen & Co. Ltd. Osborne, R., Koski, K., & Otis, R. 2002. Trends in Whale Watching Traffic Around Southern Resident Killer Whales. Friday Harbor, WA: The Whale Museum. Osborne, R. W., Koski, K. L., Talimon, R. E., & Harrington, 5. 1999. Soundwatch 1999 Final Report. Roche Harbor, WA: Soundwatch Boater Education Program. Quan, R. G., Cheng, K. C., Trask, T. C., & Meilleur, R. A. 2002a. Marine VesselAir Emissions in the Lower Fraser Valleyfor the Year 2000. Richmond, BC: Levelton Engineering Ltd. Quan, R. G, Trast, T. C., & Cheng, K. C. 2002b. Marine Vessel Air Emissions in B. C. and Washington State Outside the GVRD and FVRD for the Year 2000 (File number 4010989). Richmond, BC: Levelton Engineering Ltd. Race Rocks Ecological Reserve (RRER). 2008. Average Temperatures at Race Rocks Ecological Reserve. Retrieved on January 17, 2008, from http ://www.racerocks.com/racerock/data/weatherlinlclCurrent_Vantage_Pro.htm Rice, R. W., Sanyal, A. K., Elrod, A. C., & Bata, R. M. 1991. Exhaust gas emissions of butanol, ethanol, and methanol-gasoline blends. Journal ofEngineering for Gas Turbines and Power, 113(3): 377-381. Rijkeboer, R. C., Vermeulen, R. J., Jongbloed, R. H., van der Wal, J. T., Gerretsen, E., Jansen, H. W., Visser, J. A., Quispel, M., & Handley, P. 2004. Final Report: Stocktaking Study on the Current Status and Developments of Technology and Regulations Related to the Environmental Performance ofRecreational Marine Engines (Contract ETD/FIF.20030701). Deift, Netherlands: TNO Automotive. Roorda-Knape, M., Janssen, N., De Harthog, J., Van Vliet, P., Harssema, H., & Brunekreef, B. 1998. Air pollution from traffic in city districts near major motorways. Atmospheric Environment, 32(11): 1921-1930. Skyllingstad, E. D., Vickers, D., Mahrt, L., & Samelson, R. 2007. Effects of mesoscale sea surface temperature fronts on the marine atmospheric boundary layer. Boundary Layer Meteorology, 123: 2 19-237. Smedman, A. -S., Bergstrom, H., & Grisogono, B. 1997. Evolution of stable internal  81  boundary layers over a cold sea. Journal of Geophysical Research, 102: 1091-1099. Steyn, D. G., & Faulkner, D. A. 1986. The climatology of sea-breezes in the Lower Fraser Valley, BC. Climatological Bulletin, 20(3): 2 1-39. Stull, R. B. 1988. An Introduction to Boundary Layer Meteorology. Dordrecht, Netherlands: Kiuwer Academic Publishers. The Chamber of Shipping. 2007. 2005 2006 BC Ocean-Going Vessel Emissions Inventory. Vancouver, BC: The Chamber of Shipping. -  Thomson, B. 2004. Characterization of the Georgia Basin/Puget SoundAirshed (Cat. no. En84-3/2004E-PDF). Ottawa, ON: Environment Canada & the United States Environmental Protection Agency. Thomson, R. E. 1981. The Oceanography ofthe British Columbia Coast. Canadian Special Publication of Fisheries and Aquatic Sciences, Bulletin 56. Ottawa, ON: Fisheries and Oceans Canada. Tiitta, P., Raunemaa, T., Tissari, J., Yli-Tuomi, T., Leskinen, A., Kukkonen, J., Harkonen, J., 5 concentrations near a . 2 & Karppinen, A. 2002. Measurements and modelling of PM major road in Kuopio, Finland. Atmospheric Environment, 36: 4057-4068. Tjernström, M., Zagar, M., & Angevine, W. 2005. New England Coastal Air Pollution Dispersion Modeling. Stockholm, Sweden: Stockholm University. United States Environmental Protection Agency (USEPA). 2000. Analysis of Commercial Marine Vessels Emissions and Fuel Consumption Data (EPA Contract No. 68-C70051, Work Assignment No. 1-10). Washington, DC: Sierra Research by Energy and Environmental Analysis, Inc. United States Environmental Protection Agency (USEPA). 2002a. A Comprehensive Analysis ofBiodiesel Impacts on Exhaust Emissions (EPA42O-P-02-001). Washington, DC: United States Environmental Protection Agency, Office of Air and Radiation. United States Environmental Protection Agency (USEPA). 2002b. Health assessment documentfor diesel engine exhaust (EPA!600/8-90/057F). Washington, DC: United States Environmental Protection Agency, Office of Transportation and Air Quality. United States Environmental Protection Agency (USEPA). 2004a. Exhaust and Crankcase Emission Factors for Nonroad Engine Modeling Compression-ignition (Report No. NR-009c). Washington, DC: United States Environmental Protection Agency, Office of Transportation and Air Quality. —  United States Environmental Protection Agency (USEPA). 2004b. Exhaust Emission  82  Factors for Nonroad Engine Modeling Spark Ignition (Report No. NR-0 1 Od). Washington, DC: United States Environmental Protection Agency, Office of Transportation and Air Quality. —  United States Environmental Protection Agency (USEPA). 2004c. Air Quality, Health, and Welfare Effects. In Final Regulatory Analysis: Control ofEmissionsfrom Nonroad Diesel Engines (Report No. 420-R-04-007) (pp. 56-211). Washington, DC: United States Environmental Protection Agency, Office of Transportation and Air Quality. United States Naval Studies Board (USNSB). 1996. Shipboard Pollution Control: US. Navy Compliance with MARPOL Annex V. Washington, DC: National Academy Press. Van Atten, C., Brauer, M., Funk, T., Gilbert, N. L., Graham, L., Kaden, D., Miller, P. J., Rojas Bracho, L., Wheeler, A., & White, R. H. 2004. Honing the Methods: Assessing Population Exposures to Motor Vehicle Exhaust. Montreal, QC: Commission for Environmental Cooperation. Vingarzan, R., & Thomson, B. 2004. Temporal variation in daily concentrations of ozone and acid-related substances at Saturna Island, British Columbia. Journal ofAir & Waste Management Association, 54: 459-472. Warrington, P. 1999. Impacts ofOutboard Motors on the Aquatic Environment. Maddison, WI: North American Lake Management Society. Wilensky, U. 1999. NetLogo. Center for Connected Learning and Computer-Based Modeling. Northwestern University, Evanston, IL. http://ccl.northwestern.edu/netlogo Williams, R., & Ashe, E. 2007. Killer whale evasive tactics vary with boat number. Journal ofZoology, 272: 390-397. World Health Organization (WHO). 2000. Air Quality Guidelines for Europe (2’ ed.) (European Series, No. 91). Copenhagen, Denmark: WHO Regional Publications. Zhu, J., Cao, X., Pigeon, R., & Mitchell, K. 2003. Comparison of vehicle exhaust emissions from modified diesel fuels. Journal ofAir & Waste Management Association, 53: 67-76. Zhu, Y., Hinds, W. C., Kim, S., Shen, S. & Sioutas, C. 2002a. Study of ultrafine particles near a major highway with heavy-duty diesel traffic. Atmospheric Environment, 36: 4323-4335. Zhu, Y., Hinds, W. C., Kim, S., & Sioutas, C. 2002b. Concentration and size distribution of ultrafine particles near a major highway. Journal ofthe Air & Waste Management Association, 52(9): 1032-1042.  83  3 PHYSIOLOGICAL EFFECTS ASSOCIATED WITH EXPOSURE TO AIR 2 POLLUTION  3.1 INTRODUCTION Air pollution is an issue that currently receives a lot of attention, but the 1952 London Smog was the first air pollution episode to cause major public concern because of its causal link with a sharp rise in human mortality (Bell & Davis, 2001). The association between air pollution and human mortality is generally independent of location or climate, and by 1956 the United Kingdom established clean air legislation, and open coal burning was banned (Bates, 1994). The statistics concerning air pollution are troubling it is estimated that air -  pollutants kill approximately three million people worldwide each year, and 50% of chronic respiratory illnesses are associated with air pollutants (Pimentel et al., 2007). In Canada approximately 21,000 deaths will be attributed to air pollution in 2008, with 88% due to the chronic effects of long-term exposure, and 12% due to acute short-term recent exposure (Canadian Medical Association, 2008). The compounds in engine exhaust can produce short-term, reversible effects as well as long-term irreversible effects (NESCAUM, 1999). The majority of direct health effects attributed to air pollution are respiratory illness (e.g. asthma, acute respiratory infection, and lung cancer), cardiovascular illness, as well as premature death (BCPHO, 2004; Pimentel et al., 2007). Adverse health effects from long-term exposure to air pollution usually originate in the alveolar region of the lung (USEPA, 2002); however, increases in mortality from air pollution exposure often arise from heart conditions rather than respiratory disorders 2  A version of this chapter will be submitted for publication. Lachmuth, C. L., Barrett-Lennard, L. G., and Milsom, W. K. A model-based approach investigating killer whale (Orcinus orca) exposure to marine vessel engine exhaust.  84  (BCPHO, 2004). Positive associations have been found between respiratory symptoms, such as asthma, and traffic volume (USEPA, 2004), and the concentration and duration of exposure correlates with the start and severity of adverse health effects to some extent, but the relationship is not linear (Kalberlah et al., 2002). There is a statistically significant risk of mortality for individuals living within 200 m of a busy road, with the risk decreasing with distance from the road (USEPA, 2004). Epidemiology, human laboratory research, and animal and in-vitro toxicology are used to study the effects of air pollution on human health (Koenig, 2000). Human epidemiological studies show that exposure to low levels of air pollution produce chronic adverse health effects, but exposure measurements are often vague (Bates, 1994). Studies on acute human exposure to single gases establish minimum-effects levels; however, animal studies are currently the only way long-term effects are currently studied (Bates, 1994). All known human carcinogens also cause cancer in one or more animal species, and this similarity in response is used as the basis for extrapolation of animal studies to humans (Goddard & Krewski, 1992). Animal-to-human extrapolations have given assumptions and mathematical methods, including route of exposure, dose, and response (BCPHO, 2004). An individual’s exposure to air pollution depends on their pattern of activity in relation to the source of the air pollution, and their response to air pollution depends on their age, health, and genetic predisposition (Van Atten et al., 2004). Individuals that tend to be especially sensitive to air pollution are infants, children, the elderly, individuals already suffering from respiratory or cardiovascular illness, and individuals who engage in frequent exercise outdoors (HEI, 1988; Koenig, 2000). Thus this chapter includes an estimate of the proportion of SRKWs that may be extra sensitive to air pollution.  85  3.1.1 Compounds in Diesel and Gasoline Exhaust Both diesel and gasoline fuels are mixtures of highly toxic chemicals that when C0 NOR, hydrocarbons (HCs; composed of , combusted produce air pollutants such as CO, 2 over 150-260 aliphatic and aromatic HC compounds including potential neurotoxicants such as benzene, n-hexane, toluene, xylenes, naphthalene, and 1,3-butadiene), formaldehyde, and PM (composed of elemental carbon, sulfates, trace metals, and adsorbed organic compounds that usually comprise 10-30% of the particle mass) (BCPHO, 2004; IPCS, 1996; Ritchie et al., 2001; Yu et al., 1991). For a list of all the classes of compounds in diesel exhaust, see Appendix G. Unburned HCs and NO in gasoline and diesel exhaust can combine and undergo photochemical reactions to form secondary pollutants in the atmosphere, such as 03 and smog, which often lead to shortness of breath, chest pain, wheezing, and coughing in humans (Frumkin & Thun, 2001). Gasoline contains 25-30% aromatic HCs and is very volatile; diesel contains no aromatic HCs because it is distilled, and has very low volatility (Kirrane et al., 2006). Diesel PM is 75% elemental carbon and has a lower fraction of organic matter than gasoline PM, which is 25% elemental carbon (USEPA, 2002). Diesel engines also emit more benzo[ajpyrene, but less VOC, CO, and HC than gasoline engines (USEPA, 2002). The six principle air pollutants referred to as Criteria Air Pollutants by the USEPA are PM, ground level 03, CO, S0, N0, and lead (Koenig, 2000). There are also over 189 Hazardous Air Pollutants (e.g. benzene, mercury, asbestos), but little is known of their health effects or ambient concentrations (Koenig, 2000). PM in diesel exhaust is a pollutant of primary concern because it is easily inhaled due to its small size (the mass median aerodynamic diameter is approximately 0.2 tim), and because it has a high capacity to bind to  86  harmful chemicals (HEI, 1999; Yu et al., 1991). 03 and PM have been identified as the two Criteria Air Pollutants associated with the most serious health effects (MV, 2006); however, the most carcinogenic compounds in engine exhaust are 1 ,3-butadiene, benzene, and formaldehyde (NESCAUM, 1999). Animal studies have shown that there is potential for synergistic, additive, and/or antagonistic interactions between the individual components in fuel exhaust (Ritchie et al., 2001). The zone around non-road diesel engines (e.g. all terrain vehicles, lawnmowers, and marine vessels) frequently has air pollutant concentrations that are the same or nearly double those found along busy roads, and full-time operators of non-road engines have a significantly increased incidence of lung cancer (USEPA, 2002; 2004). Therefore, the health effects from on-road diesel engine exhaust can be applied to non-road engine exhaust (USEPA, 2004). Non-road engine technology usually lags behind that of on-road engines, which have had emission standards since 1988 (USEPA, 2004). While new technology often reduces exhaust emissions, some research has shown that there may be unintended consequences. Increases in fuel injection pressure creates higher rates of fuel atomization and evaporation, which can produce PM in the exhaust with much smaller diameters, in the nanometer size range (USEPA, 2002). This is very worrisome since ultrafine PM produces the most adverse health effects (USEPA, 2002). Very little information exists on ultrafine ) and less than or 10 particles, and only PM with a diameter less than or equal to 10 im (PM ) are currently regulated with air quality standards. 5 . 2 equal to 2.5 jim (PM A study by Kirrane et al. (2006) found that fishermen exposed to vessel exhaust received higher concentrations of benzene from vessels with two-stroke gasoline engines compared to four-stroke gasoline or diesel engines. However, even though the benzene  87  exposure was greater than ambient levels, the effects of benzene and other gasoline constituents at these concentrations on human health is unknown, but they are below levels to which health effects are reliably attributed (Kirrane et al., 2006). The simple presence of pollutants does not equal a health risk, and this issue will be explored further below.  3.1.2 Health Effects from Exposure to Diesel and Gasoline Exhaust The toxicological database for diesel exhaust is considerable, especially when compared to other toxins, and critical health effects have been derived from numerous longterm chronic exposure studies (USEPA, 2002). Research on diesel exhaust exposure has focused on PM, which has historically been used as a proxy for exposure to the entire mixture of diesel exhaust; however, there may be additive or synergistic effects from the vapor-phase of the exhaust (USEPA, 2002). It has been suggested that diesel PM is likely no 5 (USEPA, 2002). Long. 2 more toxicologically potent than other constituents of ambient PM term animal studies show that diesel exhaust poses a chronic respiratory hazard to humans (USEPA, 2002). Diesel exhaust has little acute toxicity (IPCS, 1996), but has been identified as a probable human carcinogen due to inhalation from environmental exposure by several national and international agencies (i.e. National Institute for Occupational Safety and Health, International Agency for Research on Cancer, World Health Organization, California USEPA, and U.S. Department of Health and Human Services); this classification is based on broad evidence from epidemiologic, toxicological, and experimental studies on animals and humans (HEI, 1999; USEPA, 2002). For a description of the health effects attributed to specific pollutants in exhaust see Appendix H. The database on the health effects unique to gasoline exhaust is far sparser than that for  88  diesel, and there are few epidemiology studies on gasoline exhaust (USEPA, 2002). There have been limited animal studies (again, not as comprehensive as the numerous and lengthy diesel exhaust studies) on gasoline exhaust done in prior years. These studies suggest that gasoline exhaust may have similar health effects as diesel exhaust (USEPA, 2002). For example, a study by Seagrave et al. (2002) evaluated the toxicity of diesel and gasoline exhaust from engines that were either normal or high emitters, by using a panel of assays that incorporated several categories of biological responses. They found that normal-emitter diesel and gasoline engines had very similar toxicity per unit of mass, but high emitters were much more toxic; cold conditions produced emissions that were more toxicologically potent; and adverse responses occurred even at low doses (Seagrave et al., 2002). The engine emission rates used in the dispersion model in Chapter two were from both diesel and gasoline engines. But diesel and gasoline PM is not equivalent in terms of toxicity, so instead of using PM as a proxy for exposure to the mixture of exhaust, I will use 2 from the dispersion model. the predicted exposures of CO and NO Adverse health effects from exposure to exhaust observed in animals are: formation of DNA adducts; lung mass increases up to 400%; pulmonary inflammation; impairment of lung mechanics; presence of PM-laden macrophages; increasing changes in epithelial cells; and a high incidence of lung tumors (Bond et al., 1984; 1990; IPCS, 1996; Kaplan et al., 1982; Nikula et al., 1995; Steerenberg et al., 1998; USEPA, 2002; Yu et al., 1991). The mode of action for exhaust-induced lung cancer in humans is not fully understood, and it is possible that there is a non-threshold mode of action involving mutagenic events (USEPA, 2002). In addition, the latent period for lung cancer development in humans is 20-30 years or more, making the causal link difficult to prove (Bruske-Hohlfeld et al., 1999). The lower end  89  of human occupational exposure concentrations overlap with engine exhaust and lower environmental exposure concentrations, thus current environmental exposures have the potential to be hazardous to human health (USEPA, 2002). Since there is no known level of carcinogen exposure that is risk-free, exposure to diesel and gasoline exhaust should be minimized as much as possible. Experimental data from one species is often used to predict effects in another, and this is prone to a level of uncertainty, because parameters such as uptake, deposition, biotransformation, mode of toxic action, and clearance can differ among species in qualitative and quantitative ways (Blaauboer, 1996). Additional uncertainties arise with: extrapolations from high-dose animal studies to low-dose human exposures; differences in health effects that occur in short-term versus long-term studies; similarities in toxicological response between animals and humans; and knowledge gaps regarding human exposure levels (Kalberlah et al., 2002; USEPA, 2002). Furthermore, laboratory studies are conducted on small, homogeneous groups of animals, and the results are usually extrapolated to the heterogeneous human population (Blaauboer, 1996). Even though human occupational exposure studies on exhaust usually lack detailed exposure information, the observed health effects support animal-to-human extrapolation (USEPA, 2002). The uncertainties with extrapolation present difficulties in risk estimation, yet animal studies further the understanding of biological plausibility and mechanisms of action (Bates, 1994).  3.1.3 Retention and Clearance of Air Pollutants in the Lungs Quantifying the impact of air pollution exposure requires an understanding of the processes involved, such as: accumulation, partitioning, and vertical transfer (Klanj scek et  90  al., 2007). Lipid-soluble pollutants are stored in fat, and this is the primary way pollutants deemed to be persistent and bioaccumulative are transferred vertically in marine mammal food webs (Klanjscek et al., 2007). However, most air pollutants do not bioaccumulate, as they can be cleared or metabolized, making vertical pollutant transfer negligible (Klanjscek et al., 2007). For terrestrial mammals the nose is the first line of defence against air pollution as it cleans inhaled air by removing water-soluble gases. However, it has a low capacity to filter PM from 0.1-0.5 jim in diameter (Koenig, 2000). Inhaled gases quickly come into contact with airway surfaces via molecular diffusion, but there is limited uptake for compounds that are insoluble in water (i.e. 03), and the greatest uptake often occurs in the peripheral areas of the lung because of longer residence times and larger surface areas (Lippmanri, 2000). Compounds that are more water-soluble dissolve in and/or react with fluids on the surface of the airways thus removing them from the air. In this manner very water-soluble compounds ) barely penetrate into the lung because they are almost completely removed by 2 (i.e. SO airways in the head (Lippmann, 2000). Thus, the solubility of the compound can affect where regional health effects will occur. The majority of organic compounds are reasonably soluble in lung tissue, thus can be cleared from the lungs by direct absorption into the blood (Gerde et al., 1991; HEI, 1988). From the blood, compounds can be retained in extrapulmonary tissues, excreted, or biotransformed (by metabolic activation) in the nose, lung, skin, intestine, placenta, kidney, testes, adrenals, and liver (HEI, 1988). When rodents and humans inhale diesel PM, approximately 15-20% of it is initially deposited in the lungs and respiratory tract (RET, 1988). PM deposition is affected by the geometry of the respiratory tract in several ways, the diameter of the airway determines the  91  displacement necessary for a particle to contact a surface, the cross-section of the airway determines the airflow velocity, and differences in branch lengths affects regional deposition (HEI, 1988). However, particle size and the convective flow of inhaled air are the most important parameters for inhaled PM deposition (Lippmann, 2000). The total deposition of PM increases with decreasing particle size, and deposition strongly depends on breathing frequency (FR) and weakly depends on tidal volume (VT) (Tu & Knutson, 1984). High inspiratory flow rates produce more turbulence and flatter velocity profiles in the large conducting airways, which decreases axial dispersion (Palmes et al., 1973). Alveoli with the shortest connecting branch lengths receive higher deposition of PM with diameters greater than 1 .tm (RET, 1988). Submicrometer PM tends to deposit evenly in all lobes and is not affected by branch length (HEI, 1988), but there can be preferential deposition at airway bifurcations (Lippmann, 2000). Inertial impaction is the primary deposition mechanism in the nasal, extrathoracic, and tracheobronchial regions of the lungs, especially for PM with a diameter greater than 2.5 tm, due to high airflow velocities and abrupt changes in airflow direction in these regions (Stber et al., 1993). Sedimentation is deposition of PM due to gravity, and increases with increasing residence time in the airways, increasing particle size, and density, but decreases with increasing FR (HET, 1988). Sedimentation occurs to PM exceeding 0.5 tm in diameter in airways with relatively low air velocity; however, diffusion dominates for PM less than 0.2 im (HEI, 1988). PM with a diameter less than 0.5 im (ultrafine) tends to penetrate the lungs past the extrathoracic and tracheobronchial regions, and is subject to diffusive deposition in the alveolar region of the respiratory tract (Stöber et al., 1993).  92  The following discussion of air pollutant clearance from the lungs assumes that exposure is an acute event. Removal of PM from the lungs occurs by either dissolution or mechanical clearance (Yu et al., 1991). Compounds that dissolve in the conducting airway mucus or alveolar surfactant can quickly diffuse into epithelial cells and into the blood, thus can be spread to other organs and tissues (Lippmann, 2000). If the mucus thickness lining the trachea is halved, the dose is increased by a factor of 10 (HEI, 1988). Inhaled compounds can also undergo chemical and metabolic processes in the fluids and cells of the lung, which can limit their entrance into the blood and can create products that differ in solubility and toxicity (Lippmann, 2000). Mechanical clearance occurs by either mucociliary transport (in the nasal passages, ciliated airways, and tracheobronchial region), or macrophage phagocytosis and migration in nonciliated airways (Yu et a!., 1991). Mucociliary transport is accomplished by rhythmically beating cilia lining the respiratory tract from the terminal bronchioles to the trachea, which move poorly soluble PM and alveolar macrophages in a mucous layer toward the larynx (Robertson, 1980). The rate of mucociliary transport is species dependent and is determined by the flow of mucous, which is slow in the distal airways and increases proximally (Stöber et al., 1993). Clearance mechanisms in the lungs are believed not to be especially particle-sized dependent (Snipes et al., 1983). Studies on large diameter PM (> 2.5 i.im) have found that it is quickly eliminated by mucociliary transport (almost complete within 24-hours in humans) and dissolution (requires a few hours in humans), however some long-term retention occurs (HEI, 1988; IPCS, 1996; USEPA, 2004; Yu et al., 1991). Macrophages act as a defence mechanism by engulfing foreign matter in the lungs and secreting inactivating enzymes (Koenig, 2000); however, organic chemicals in exhaust  93  emissions can produce toxic effects in alveolar macrophages (HEI, 1988). PM in the alveolar region is primarily cleared by macrophage phagocytosis (approximately 6-hours in humans) and migration (several weeks in humans) to the tracheobronchial area, lymph nodes, and blood (Yu et aL, 1991). PM not removed by macrophages is incorporated into epithelial and interstitial cells that clear PM slowly by dissolution and/or lymphatic drainage (half-time is approximately 30-1000 days or more), but particles can be retained indefinitely in interstitial sites (HEI, 1988; IPCS, 1996; Lippmann, 2000; USEPA, 2004; Yu et al., 1991). Electrostatic charges on PM can affect regional deposition, due to repulsion and attraction, and this effect is inversely proportional to PM size and airflow velocity (Cohen et al., 1998). For humans, deposition and clearance rates are affected by age. PM deposition is higher in infants and children than in adults while nasal breathing at rest, with maximum deposition occurring at approximately two years of age (HEI, 1988). Clearance rates for the elderly and children are slower than clearance rates for adults (Yu et aL, 1991). Clearance mechanisms and patterns in the respiratory tract are similar for humans and most other mammals, as diffusion across epithelial barriers occurs at approximately the same rate, but they may differ if mediated by mucous transport or macrophages (HEI, 1988; USEPA, 2002). However, the clearance rate of insoluble particles is species dependent and not fully understood, with the retention half-time for rats, mice, and hamsters about 50-100 days, and several hundred days for dogs, guinea pigs, and humans (Yu et al., 1991). Humans have a greater lung burden of PM than rats because they inhale greater quantities of PM and have slower clearance rates (Yu et al., 1991). Evolutionary selection for PM clearance is likely higher for terrestrial than marine mammals because of inherent differences in exposure rates, which would suggest that retention times may be greater in killer whales than humans.  94  The discussion of clearance mechanisms above is mainly based on studies that exposed experimental animals to acute doses of air pollutants. When animals (Yu et al., 1991) and humans (StOber et al., 1967) are chronically exposed to high doses of diesel exhaust the clearance of particles deep in the lungs can be impaired, a pattern referred to as the overload effect. Because killer whales are chronically exposed to whale-watching vessels for 12-hours a day for six months of the year, they are potentially experiencing the overload effect and accumulating PM at a rate faster than it can be removed. Accumulation may be especially severe if their clearance rates are significantly slower than humans.  3.1.4 Killer Whale Respiratory Anatomy and Physiology While little has been documented on the respiratory physiology of killer whales, lessons may be learned from another better-studied dephinid, the bottlenbse dolphin (Tursiops truncatus). The Deiphinidea are nested in the suborder Odontoceti (toothed whales), and they have the most extreme lung modifications of all marine mammals (Perrin et al., 2002). Delphinidae lungs are larger relative to body size than most other marine mammals (Perrin et al., 2002). Their lungs have no external lobulation, and are pyramidal in shape with the base situated dorsally and caudally (Fanning & Harrison, 1974). Like other cetaceans, the peripheral airways of deiphinids are highly reinforced with cartilage and smooth muscle to keep the conducting airways open during deep dives while allowing the alveoli to collapse (Kooyman, 1 989a). Delphinid lungs have lost respiratory bronchioles, they have bronchial sphincters, and they have very flexible (compliant) chest walls (Perrin et al., 2002). The trachea is lined with microvillous surface cells, while distal parts of the trachea and bronchi are lined with ciliated and goblet cells (Fanning, 1977). The terminal  95  airways and alveoli typical of the mammalian pattern have additional connective tissue support, have an abundance of free macrophages (Fanning, 1974; Fanning & Harrison, 1974), and the alveolar surfactant has high fluidity and rapid expansion capabilities (Foot et al., 2006). In the terminal bronchus, the thickness of the blood-air barrier averages 150-250 nm, with the thinnest recorded barrier being 120 nm (Fanning & Harrison, 1974). Cetaceans are obligate nasal breathers, and the nostrils of Odontocetes have joined to form one blowhole (Perrin et aL, 2002). Cetaceans do not have facial sinuses or conchae (turbinate bones in the nasal cavity), which aids in accelerating inhalation and exhalation (Perrin et a!., 2002). The lack of facial sinuses in cetaceans may render them more , because instead of being removed by 2 vulnerable to water-soluble air pollutants like SO airways in the head (Lippmann, 2000), they would penetrate further into the lungs. Countercurrent heat exchange and induced turbulence in the nares and nasal sac system of bottlenose dolphins extracts most of the water vapor in the exhalation, and results in a 70% reduction in water loss compared to terrestrial mammals (Perrin et al., 2002). In Odontocetes there is complete regression of the olfactory system by birth (Marino, 2004). A consequence of this in the context of this study is that killer whales are unable to smell, and thereby avoid, engine exhaust. The adaptations in the respiratory system of marine mammals provide greater elastic recoil of the lungs, chest cavity, and diaphragm (Perrin et a!., 2002), and the larger conducting airways allow extremely fast ventilation rates, with most of the VT exchanged within a fraction of a second (Kooyman, 1 989a). Marine mammals are able to use a much greater proportion of their total lung capacity (TLC) while breathing than terrestrial mammals, and their VT is usually greater than 75% of TLC, and their vital capacity (Vc) can  96  exceed 90% of TLC (Perrin et al., 2002). This means that they exchange almost all of their respiratory gases every breath, but the large VT means they also have a reduced reserve lung volume and prevents them from significantly increasing VT further during strenuous exercise (Perrin et al., 2002). A killer whale’s respiratory cycle begins with a rapid active expiration or blow ( 0.38-0.59 s) that starts just before surfacing and clears the upper airway of water, followed by a slower passive inhalation  (-- 0.75-0.78  s), and a variable period of apnea that typically lasts  between 2-10 minutes (Kooyman & Cornell, 1981; Kooyman & Simiett, 1979; Kooyman et al., 1975; Milsom, 1989). A high flow rate is maintained for the entire expiration, whereas in humans and other terrestrial mammals the expiratory flow rate peaks early and then decreases substantially throughout the end of the breath (Perrin et al., 2002). Kooyman and Cornell (1981) measured the peak expiratory flow rate of a 4.4 m killer whale to be 180 L s (approximately 2.5 Vc s’), while that for a normal human male is approximately 10 L s (1.5-2.0 Vc s’) (Gregg & Nunn, 1973; Kooyman & Cornell, 1981). Breathing patterns are affected by behaviour, and while resting at the surface the FR of killer whales is about 0.9 breaths per minute (b m’) (Mortola & Limoges, 2006), and during strenuous exercise is 1 (Kriete, 1995). When traveling at low speed the blowhole barely approximately 7.3 b m clears the water during exhalation and inhalation, but at higher speeds there is greater clearance as the animal porpoises above the sea surface (Perrin et al., 2002). The anatomy and physiology of the respiratory system of cetaceans affects particle retention in the lungs. Due to the structure of the blowhole and nasal cavities, the epithelium of the proximal trachea is exposed to higher levels of PM in the air (Fanning, 1977). The large VT of cetaceans allows PM and gases to penetrate deep into the lungs, and increases  97  deposition by sedimentation in smaller conducting airways and alveolar regions where airway diameters and velocities are smaller (RET, 1988; Martin & Finlay, 2006). For example, a rat that doubles its VT increases aerosol deposition seven times (RET, 1988). Deposition by impaction does not depend on VT because impaction mainly occurs in the upper and central airways (Martin & Finlay, 2006). Fast inhalation velocities, however, do increase deposition by impaction, but decrease deposition by sedimentation and diffusion, and high velocities may also create turbulence, which increases PM deposition by impaction (RET, 1988). Increasing the minute volume (Vmin) of respiration causes significantly greater absorption of gases in the pulmonary region, but the absorption in the tracheobronchial region does not increase as much (HEI, 1988). Thus, compared to humans and other terrestrial mammals, killer whales would be expected to have increased PM deposition and absorption of gases in the distal airways. Simpson and Gardner (1972) examined cetacean lungs (including those of killer whales) from whaling vessels worldwide and found that the animals very rarely suffered from overt lung disease or pulmonary histopathology, which was attributed to their relatively clean and microbe-free environment. However, lung disease in porpoises was reasonably common, and usually a result of nematode infestation (Simpson & Gardner, 1972). In contrast, a recent study by Raverty et al. (unpubl.), obtained data on killer whale strandings worldwide, which included 222 post mortem findings from 1944-2003, and 309 tissue samples from 48 killer whales. They found that 50% of the 46 animals had histopathological abnormalities in the lung tissue (Raverty et al., unpubl.). Some of the effects observed by Raverty et al. (unpubl.) are not inconsistent with the likely effects of air pollutants (i.e. disease processes in the endocrine and metabolic systems, and neoplastic tissue changes) but  98  other important factors affect killer whale health in the same manner, such as persistent organic pollutants, macroparasites, disease, nutrition!condition, and age. However, the results from Simpson and Gardner (1972) are very different than those of Raverty et al. (unpubi.), and may reflect the deterioration in air quality since the 1970’s.  3.1.5 The Effects of Diving and Breath Holding Every 10 m in depth under water adds one atmosphere of hydrostatic pressure. This increase in pressure with depth decreases lung volume, causing a higher partial pressure of gasses in the lungs, which increases gas solubility, and the amount of gas dissolved in tissues (Fahiman et al., 2006; Kooyman, 1973). Blood flow through tissue determines when gas equilibrationlsaturation is reached, and vasoconstriction and bradycardia increase the time to saturation; thus the heart saturates rapidly, and fat saturates slowly (Fahlman et al., 2006). To avoid nitrogen narcosis and decompression sickness, seals and dolphins experience lung collapse while diving at approximately 40-80 m, and air in the lungs is pushed into conducting airways where no gas exchange occurs (Fahlman et al., 2006; Ridgway & Howard, 1979). However, the animal is vulnerable to decompression hazards when repeatedly diving to depths shallower than those at which lung collapse occurs (Ridgway & Howard, 1979). Many marine mammals exhibit adaptations for diving to depth, and the “mammalian diving reflex” allows maintenance of a constant blood pressure, and reduced metabolic rate while diving (Hastie et al., 2006; Kooyman, 1989b; Noren et al., 2004). Upon submergence killer whales reduce their resting heart rate by 50% (Spencer et al., 1967). Since killer whales live their entire lives in water they likely have “normal” heart rate and blood flow  99  distribution while shallow diving, and an “abnormal” or elevated distribution while surfacing (Bill Milsom, pers. comm., October 2008). Thus killer whales would only be expected to experience depressed metabolic rates and bradycardia during very deep dives, and this would result in slower rates of pollutant conversion to metabolites. Since circulation to the skin and splanchnic organs virtually stops during deep dives, 02 is channeled to the organs that require it most (i.e. heart and brain), and during very deep dives killer whales may shift to anaerobic metabolism (Perrin et al., 2002). When the lungs are collapsed at depth no gas uptake occurs, but the shift of blood flow from organs that detoxify blood could allow toxins already in systemic circulation to concentrate in sensitive tissues like the heart and brain (Reynolds et a!., 2005). Diving mammals have large blood volumes with a high 02 carrying capacity, allowing them to store more 02 in their blood, and they are very tolerant to low arterial 02 tensions (Kooyman, 1 989b). Compared to terrestrial mammals, cetaceans tend to have fewer red blood cells but those cells have a higher hemoglobin concentration (Dhindsa et al., 1974), they have muscle myoglobin concentrations 10-30 times greater for additional 02 transport (Kooyman, 1989b), and they generally have a greater blood 02 affinity (Dhindsa et al., 1974). Killer whale blood also has a high buffering capacity, which restricts the development of respiratory and metabolic acidosis during dives (Lenfant et al., 1968), and 2 (Mortola & Limoges, 2006). Delphinids reduces their ventilatory response to inhaled CO dive with 22% of their total oxygen stores in the air in their lungs, with the rest sequestered in muscle myoglobin and blood hemoglobin (Perrin et al., 2002). All cetaceans have rapid, high velocity breathing, allowing them to exchange a large percentage of their lung volume during the brief period of inhalation and exhalation at the  100  surface (Reed et al., 2000). Hyperventilation often occurs before and after diving to increase 2 tension (Perrin et al., 2002). Cetaceans are able to fully restore 02 tension and decrease CO 2 02 stores in the first three to four breaths after a long dive; however, built-up levels of CO must also be eliminated, and a further two to three breaths are often required (Reed et al., 2 than to replenish 02, there may be occasions 2000). Since it takes longer to eliminate CO 2 stores before beginning a dive sequence and when cetaceans do not fully readjust their CO 2 load (Reed et al., 2000). Just before surfacing, there have to manage with an increased CO 2 is often an increase in heart rate, which decreases 02 levels in the blood and increases CO 2 elimination by the lungs (Perrin et al., 2002). Deep and allows rapid 02 uptake and CO 2 exhaled in the first breath upon surfacing than shallow dives, because dives produce less CO there is no gas exchange while the lungs are collapsed (Perrin et al., 2002). Killer whales often react to vessels by increasing their dive duration and speed (Kruse, 1991; Williams et 2 concentrations are high in . If CO 2 al., 2002) and during this time they may accumulate CO the air they are inhaling due to the exhaust from whale watching vessels and if they just take 2 load may become exaggerated. This could have serious one breath on surfacing, their CO impacts on gas exchange and recovery time, and may result in future health problems. The SRKWs primarily occupy near-surface waters and only 2.4% of their time is spent below 30 m in depth, but deep dives last for much longer than shallow dives (Baird et al., 2005; 2003). Baird et al. (2003) suggested that deep dives are primarily for foraging, while time at the surface is for breathing, socializing, resting, traveling, and some foraging. Adult males make significantly more deep dives than adult females, but both sexes dive equally deep (Baird et al., 2005). Greater dive rates and swim speeds occur during the day compared to night, and dive depths greater than 150 m occur regularly, with 264 m the  101  maximum recorded depth for a SRKW (Baird et al., 2005). Since the SRKWs spend most of their time in the first 30 m of the water column (Baird et al., 2005; 2003), they are likely not experiencing lung collapse (Ridgway & Howard, 1979). However, at 30 m in depth the atmospheric pressure is three times that at the surface, thus the whales have greater amounts of gas dissolving into their tissues (Fahlman et al., 2006; Kooyman, 1973). When an aerosol is inhaled, its probability of being exhaled decreases exponentially with time (Goldberg & Smith, 1981; Palmes et aL, 1973); thus as breath holding time increases, deposition in the lungs increases exponentially (Goldberg & Smith, 1981; Tu & Knutson, 1984). Aerosol persistence decreases as the lung volume of the held breath decreases, due to smaller intrapulmonary air spaces in smaller lungs (Palmes et al., 1973). Therefore, even though non-breath holders breathe more frequently, killer whales experience greater levels of deposition in the distal areas of lungs because of breath holding, low FR, large VT, and high air flow rates (Goldberg & Smith, 1981; HEI, 1988; Palmes et al., 1973; Tu & Knutson, 1984). In addition, their large lung size (which is greater than expected allometrically) increases the area for deposition (Perrin et al., 2002). Because of these issues, it appears that killer whales are likely more sensitive to air pollution than humans.  3.1.6 Physiological Models Used to Estimate Internal Pollutant Dose and Health Effects Many different models are used to predict the effect of a pollutant dose on the physiology of a species, and simplified mathematical formulas, correlation of data, assumptions, simplifications, and extrapolation all play a role (HEI, 1988). Several methods are used to overcome uncertainties with extrapolation, such as the introduction of safety factors, uncertainty factors, and default values (Blaauboer, 1996). Uncertainty factors are  102  used to take into account inter-individual variability, and animal-to-human variability. The uncertainty factor (UF) for inter-individual variability is usually set at 10, as is that for animal-to-human extrapolation (Renwick & Lazarus, 1998). This implies that the effects seen in animals would occur in humans at concentrations 10 times lower because of higher doses and/or greater sensitivity of human tissue (Renwick & Lazarus, 1998). The sensitivity of killer whales is currently unknown, but can be estimated by using extrapolation and a weight of evidence approach for pollutant effects in marine mammals (Ross, 2000). However, based on the previous information one would assume that killer whales could tolerate a concentration ten times higher than humans. However, the UF of 10 is applied when scaling up from rodent data to humans and may not be applicable when scaling up from humans to killer whales. Lung compartment models are the most basic, and are used to predict gas uptake by the blood and tissues (HEI, 1988). Dosimetric models (either mathematical or experimental) are used to predict gas uptake and distribution in specific regions of the respiratory system for different exposure levels in different species, and to estimate toxicological effects (HEI, 1988). Sophisticated physiologically-based pharmacokinetic (PBPK) models can be used to predict ultimate health effects by simulating concentrations of pollutants in the blood and tissues that result as a function of duration and exposure (Béliveau et al., 2005; NRC, 1991). However, all of these models require detailed information, such as: the anatomy/geometry of the airways, tissue and blood spaces; ventilation and perfusion limitations; pulmonary function parameters (e.g. respiratory and pulmonary blood flows); convection in respiratory tract fluids (i.e. blood and mucus); lipid and water content of blood and tissues; enzyme concentrations; metabolic constants; material balance equations that describe time-dependent  103  and spatial distributions of the pollutant; thermodynamic equilibration, diffusional flux, and chemical reaction rate equations; particle characteristics (i.e. size, shape, density, and electrostatic attraction); partition coefficients; and binding capacities (Béliveau et al., 2005; HEI, 1988; USEPA, 2002). It would be possible to use one of the above model frameworks to develop a model for killer whales by estimating parameters with virtually no experimental data. However, the use of a sophisticated model would result in a level of uncertainty in proportion to the accuracy of the parameters included (Preston, 2005). Unfortunately much of the necessary input information required is lacking for killer whales, rendering the level of uncertainty unacceptably high. I have therefore opted for using a simple allometric scaling model to predict the effect of air pollutants on killer whale physiology.  3.1.7 Allometric Scaling to Estimate Internal Pollutant Dose and Health Effects Allometric scaling is a commonly used extrapolation method that allows quantitative comparison of function between or within species, since physiological functions (such as respiratory mechanics) are often related to body size or mass (Mb) (HEI, 1988; West et al., 1999). A routinely used simple allometric conversion modifies the external dose concentration based on the quantity of pollutant inhaled per unit body mass per day, whereas more complicated conversions can include body surface area, which accounts for differences in metabolism, biotransformation, and degradation (McColl et al., 2000). Allometric scaling occurs most often between rodents and humans, and these extrapolations have many potential difficulties and uncertainties, but when dose-response assessment includes adequate allometric conversions, the interspecies UF can be reduced (McColl et al., 2000).  104  Allometric scaling uses parameters that are either constant with body size or are related to body size by a proportion (HEI, 1988; West et al., 1999). Metabolic rate (02 consumption) in vertebrates tends to scale with a three-quarters power of body mass (Kleiber, 1961; Sample & Arenal, 1999; West et al., 1999). Respiratory variables related to gas exchange (i.e. the respiratory minute volume, which is the volume of air that is inhaled or , but size-related variables of the 75 exhaled in one minute) also tend to scale to Mb° respiratory system (i.e. VT) tend to scale to Mb’ (Milsom, 1989; West et al., 1999). 25 while the ratio of dead space volume (VD) to VT is Breathing frequency (FR) scales to M’o° independent of body size (HEI, 1988). The FR of most aquatic mammals is below the allometric curve of teffestrial mammals, and this difference increases in larger aquatic mammals (Mortola & Limoges, 2006). In addition, the resting metabolic rate (RMR) of 75 (Kooyman, marine mammals is about 1.5-3 times greater than that predicted by Mb° I 989b); however, others have found no real difference in RMR for marine mammals (Lavigne et al., 1982; 1986). 75 for the extrapolation of test animal The USEPA recommends scaling to Mb° carcinogenicity data to humans (USEPA, 1992), and for wildlife risk assessment Sample et 75 using mammalian toxicity data. Since al. (1996) also recommend scaling to Mb° , species-specific carcinogenicity and toxicity scale 75 respiratory rates also scale to Mb°’ directly with respiratory rate (Schneider et al., 2004). If the toxicity value for a test animal (Ar), an allometric scaling exponent (b), and the body mass of the test animal (Mb) and wildlife species (Mb) are known, one can calculate the toxicity value for the wildlife species (A) by the equation (Sample et al., 1996): A W =A  Mb Mb  105  Sample and Arenal (1999) investigated the allometric relationships for acute mammalian toxicity data for a number of chemicals, and concluded that unless a chemical94 should be used for mammals. specific scaling exponent is known, scaling to Mb° However, the 0.94 scaling exponent was for acute rather than chronic toxicity data, and since the modes of action for acute and chronic effects differ for many chemicals the 0.94 scaling exponent is not appropriate for the whale-watching scenario. Schneider et al. (2004) found ) is much more accurate than 75 that for the majority of toxins, caloric demand scaling (Mb° body mass scaling predictions (Mb’), even for readily metabolized substances. Caloric demand scaling predicts that smaller species are less susceptible to toxins than larger species if the dose is per kg body mass (Schneider et al., 2004). Thus, I used a scaling exponent of 0.75 since it accounts for uncertainty in interspecies extrapolation with allometric rules (Schneider et al., 2004). Differences in pollutant sensitivity between humans and other animals was evaluated by Kalberlah et al. (2002), who used data from the Agency for Toxic Substances and Disease Registry (ATSDR). For gases and liquids, humans were more sensitive in 62% of the cases, and for particles, humans were again more sensitive in 53% of the cases (Kalberlah et al., , 2 2002). However, humans and animals had no quantitative differences in sensitivity to NO and sensitivity levels were similar for 03 (Kalberlah et al., 2002). The study highlighted the fact that there is only limited data on quantitative interspecies comparison even for well known respiratory toxicants, but on average humans tend to be more sensitive (Kalberlah et al., 2002). The studies evaluated by Kalberlah et al. (2002) primarily utilized body mass scaling from rodents to humans to estimate sensitivity, which suggests that the larger the animal the more sensitive it is. When this logic is applied to killer whales, it suggests that  106  they would be even more sensitive than humans to air pollutants.  3.2 METHODS An individual’s ultimate exposure to a pollutant is a function of the concentration they are exposed to, their breathing pattern, and their particle retention pattern (USEPA, 2002). All sources of exposure need to be considered (i.e. ingestion, skin penetration, inhalation), as does the individual’s activity level because ventilation rate is used to determine dose (Koenig, 2000). Studies investigating the effects of oil spills on cetaceans have found that their epidermis is relatively impermeable to petroleum, because of tight intercellular bridges, vitality of the superficial cells, and extreme epidermal thickness (Geraci & St. Aubin, 1990). Therefore, it is unlikely that skin contact is a significant source of toxicity (O’Shea & Aguilar, 2001; Wiles, 2004). Killer whales could be exposed to exhaust pollutants such as HCs by ingestion while feeding, but marine mammals can generally metabQlize and excrete HCs (Wiles, 2004). Thus, it was assumed that inhalation was the only route of exposure to exhaust pollutants for the SRKWs. A basic estimate of a pollutant dose is: dose (in minutes)  *  =  ) 3 concentration (in mg m  *  duration  ventilation rate (in litres per minute’), which assumes the pollutant is  completely absorbed internally (Koenig, 2000). The biologically effective dose is the amount of pollutant or its metabolites that interacts with a target organ during a given time period to alter physiological functioning (NRC, 1991). It is often not practical to use the biologically effective dose when determining the overall effect of exposure to air pollution because of a lack of information about uptake, distribution, metabolism, and modes of action of contaminants (NRC, 1991). Thus I used the basic dose equation, and converted the  107  concentration to mg per litre, which resulted in a dose in mg during 60 minutes. The mean 2 concentrations predicted by the dispersion model in Chapter two for averageCO and NO 3 respectively), and worst-case simulations 3 and 0.04 mg m case simulations (11.98 mg m 3 respectively) were used in the basic dose equation. The dose 3 and 0.24 mg m (67.91 mg m was then divided by the mass of a male or female killer whale to obtain the dose per kg of body mass for each gender. A free-swimming breathing rate for killer whales was used, because in the dispersion model the killer whale was traveling at a constant speed of 2.85 m s’ after Kriete (2002). Kriete (2002) measured the respiratory variables of free-ranging SRKWs, and the FR for . Kriete (1995) estimated the 1 , and for females was 1.74 b min 1 males was 1.63 b min maximum VT of captive adult killer whales, and found that males exchange approximately 210 litres per breath (L b’), and females exchange approximately 100 L b’. Thus the . Kriete (1995) also 1 , and for females is 174 L min 1 predicted Vmin for males is 342.3 L min estimated the mass of four female and two male killer whales in captivity, by measuring their body lengths and using Bigg and Wolman’s (1975) equation for the relationship between body length and mass for killer whales. The average mass for female killer whales was 2,427.5 kg, and for males was 3,766.5 kg. Effective ventilation (Veff) is the ventilation in the alveolar region (VA  =  VT VD) multiplied by FR, which provides a more accurate internal -  dose estimate than simply calculating dose using Vmin, which includes VD. There are no published values of Veff or VD for killer whales; however, VD can be calculated by using the allometric equation formulated by Stahl (1967), where VD  . Thus the VD for a 96 2.76 Mb°  , and for a 1 3,766.5 kg male killer whale is predicted to be 7.5 L with a Veff of 330.08 L min  108  2,427.5 kg female killer whale the VD is 4.9 L, with a  Veff  of 165.47 L miii’. These values  for male and female killer whales were used to determine the dose per kg body mass, by 2 per kg body using the basic dose equation. For comparison, the basic dose of CO and NO mass was calculated for a 70 kg human during mild exercise (requiring less than 60% maximum 02 uptake), with a  Veff  of 18.0 L miii’ (ACSM, 2006).  2 The allometric equation of Sample et al. (1996) was used to calculate CO and NO toxicity values (A) for male and female killer whales. The MV AQO for 1-hour exposure to ) were used as the 3 , and 8-hour exposure to CO (10 m g m m ) 2 (0.2 mg 3 , NO m ) CO (30 mg 3 test species toxicity values (At), rendering humans the test species. The MV AQOs are based on expected actual exposure conditions of human populations in British Columbia, and if humans are exposed long-term to mean concentrations above the standards they would potentially exhibit premature mortality, increased admissions to hospitals, respiratory symptoms, and decreased lung function (USEPA, 2002). The calculated toxicity values for killer whales were then used in the basic dose equation, to calculate a toxicity dose for killer whales. This toxicity dose is the dose below which no adverse health effects would be expected in an average adult killer whale, but above it adverse health effects would be expected. Thus, the toxicity dose can be compared to the basic dose that was calculated by using the average-case and worst-case exposures predicted by the dispersion model. For comparison a toxicity dose for humans was also 2 as the exposure values in calculated by using the MV Air Quality Objectives for CO and NO the equation.  109  3.3 RESULTS The basic dose per kg body mass for male and female killer whales and humans using 2 predicted by the dispersion 1-hour and 8-hour exposures to CO and 1-hour exposure to NO equation are presented in Table 3.1.  2 per kg body mass that male and female killer whales Table 3.1: The dose of CO and NO and humans receive during average-case, worst-case, 1-hour. and 8-hour exposures. Average Average Worst-case Average Worst-case Species & 1-hour 1-hour 1-hour 8-hour 1-hour gender 2 NO 2 NO CO CO CO ) 1 kg (mg kg’) (mg (mg kg’) (mg kg’) (mg kg’) 0.0014 0.00021 0.40 0.50 0.063 Killer whale, male 0.0011 0.00016 0.31 0.39 0.049 Killerwhale, female 0.0042 0.00062 1.17 1.48 0.19 Human  Even though killer whales have a much lower FR, much greater Veff, and extremely 2 per kg body mass that male and large VT compared to humans, the doses of CO and NO female killer whales receive were much lower than that of humans (Table 3.1). The average2 for male killer whales were three times lower than case and worst-case doses of CO and NO those for humans, and 3.8 times lower for female killer whales than humans. The difference in values between killer whales and humans is because dose was calculated on a per kg body mass basis, and because killer whales have a much greater body mass, their dose is much smaller. The worst-case doses for 1-hour exposure to CO were very similar to the average case doses for 8-hours exposure to CO. Male killer whales had doses slightly higher than female killer whales (Table 3.1). 2 toxicity values (A) for male and female killer whales can be seen The CO and NO in Table 3.2, and they were calculated by using the equation derived by Sample et a!. (2006). 110  2 (1-hour) toxicity values (A) for male and Table 3.2: CO (1-hour and 8-hour) and NO female killer whales.  The toxicity values (A) for male and female killer whales (Table 3.2) are much 3 for 3 for a 1-hour exposure and 10 mg m lower than the human MV AQOs for CO (30 mg m 3 for a 1-hour exposure). The toxicity values for 2 (0.2 mg m an 8-hour exposure) and NO killer whales in Table 3.2 were then used in the basic dose equation (Koenig, 2000), calculate the toxicity dose (Table 3.3). Table 3.3 also includes the toxicity dose for humans. 2 per kg body mass for male and female killer Table 3.3: Toxicity doses of CO and NO whales and humans, using toxicity values (A). 2 (mg kg’) NO ) 1 CO (mg kg’) CO (mg kg Species & gender 1-hour 8-hour 1-hour 0.00039 0.16 0.058 Killer whale, male 0.00034 0.13 0.049 Killer whale, female 0.0031 1.23 0.46 Human  The toxicity doses (Table 3.3), which are based on the MV AQOs, are very similar to those predicted by the average-case simulations in the dispersion model (Table 3.1). For male killer whales the average-case dose of CO for 1-hour of exposure is slightly higher than the CO toxicity dose, but for female killer whales the doses were equal. The average-case 8hour CO dose was approximately 3 times higher than the 8-hour CO toxicity dose for both 2 for both male and female male and female killer whales. The average-case doses of NO 2 toxicity doses. killer whales were slightly lower than the NO 2 for male and female killer whales were much The worst-case doses of CO and NO higher than the toxicity doses. The worst-case CO dose was 6.9 and 6.3 times higher than the 111  CO 1-hour toxicity dose, for male and female killer whales respectively. The worst-case 2 toxicity dose, for male and female killer 2 dose was 3.6 and 3.2 times higher than the NO NO whales respectively. 2 for male killer whales were on average The 1-hour toxicity doses of CO and NO 2 for female killer 33% that of humans; whereas the 1-hour toxicity doses of CO and NO whales were on average 11% that of humans. ) 3 The ambient average concentrations (from May to September) of CO (0.71 mg m ) measured at Christopher Point, BC were added to the average-case 3 and NO 2 (0.008 mg m and worst case concentrations to obtain the total exposure and total dose (Table 3.4). 2 per kg body mass that male and female killer whales Table 3.4: Total dose of CO and NO are estimated to receive under average-case or worst-case whale-watching conditions. 2 2 Worst-case NO Worst-case CO Average NO Average CO Gender (mg kg’) ) 1 (mg kg (mg kg’) (mg kg’) 0.0015 0.00025 0.40 0.068 Male 6.0011 0.0002 0.31 0.053 Female -  2 are relatively low Because the average ambient concentrations of CO and NO 2 that male and female killer compared to the exposure doses, the total doses of CO and NO whales receive are almost identical to the predicted doses in Table 3.1. Infants, children, and the elderly are especially sensitive to air pollution (HEI, 1998; Koenig, 2000), and current urban levels of air pollution result in chronic, adverse effects on lung development in children (Gauderman et al., 2004). Gauderman et al. (2004) found that , 2 children living in communities exposed to numerous air pollutants like 03, acid vapor, NO elemental carbon, and PM had significant reductions in respiratory system growth, resulting in smaller lung volume. As of October 2008, 36 members or 43% of the SRKW population  112  were post-reproductive females, calves or juveniles, and may therefore have had lower threshold toxicities than those calculated above.  3.4 CONCLUSIONS The model simulations based on average whale-watching conditions predicted a CO 3 for 1-hour of exposure, which is lower than the female killer concentration of 11.98 mg m whale’s threshold toxicity value for CO but higher than the male’s (Table 3.2). If the average-case simulation exposure to CO lasts for up to 8-hours, then the dose the killer whale receives (Table 3.1) would be three times greater than the 8-hour toxicity value (Table 3.3). This suggests that under average whale-watching conditions, the doses of CO that killer whales receive are just at levels predicted to cause adverse health effects for 1-hour of exposure, but if the exposure lasts 8-hours then the threshold for adverse health effects is greatly exceeded. Considering that a typical cigarette emits 67 mg of CO (Löfroth et al., 1989), during 1-hour of average whale-watching conditions the male killer whale would experience the equivalent of 3.5 cigarettes, and the female 1.8 cigarettes. The worst-case doses of CO that male and female killer whales receive are much higher, by factors of 6.9 and 6.3 respectively, than those expected to cause adverse health effects in killer whales. Thus the male killer whale potentially experiences the equivalent of 22.5 cigarettes, and the female 11.2 cigarettes during 1-hour of worst-case whale-watching conditions. 2 that male and female killer whales receive are just The average-case doses of NO below levels that are predicted to cause adverse health effects in killer whales Yet peaks in 2 concentrations (Table 2.3) can approach levels predicted to cause adverse ambient NO 2 that male and female health effects in killer whales (Table 3.2). The worst-case doses of NO  113  killer whales receive are also higher, by 3.6 and 3.2 times respectively, than those expected to cause adverse health effects in killer whales. 2 that Even though the modeled average-case and worst-case doses of CO and NO killer whales receive are on average 3.4 times lower than what a human would receive, their 2 are much lower. When these toxicity values were used to toxicity values for CO and NO , it was determined that the toxicity doses 2 calculate the 1-hour toxicity doses of CO and NO for killer whales are on average 12% of those for humans. This suggests that killer whales are potentially much more sensitive to air pollutants in the atmosphere than humans, because humans require much higher concentrations to produce adverse health effects. This is primarily due to the effect of allometric scaling with an exponent equal to 0.75 as it -  produces an inverse relationship between body size and toxicity value/dose. In addition killer whales may be more sensitive to air pollution due to their respiratory anatomy and physiology alone. Compared to humans and other terrestrial mammals, killer whales would experience increased PM deposition, persistence, and absorption of gases in the distal airways because of breath holding, low FR, large VT, and fast respiratory rates (Goldberg & Smith, 1981; HEI, 1988; Palmes et al., 1973; Tu & Knutson, 1984). As breath holding time during a dive increases, deposition in the lungs increases exponentially (Goldberg & Smith, 1981; Tu & Knutson, 1984), and their large lung size increases the surface area for deposition (Perrin et al., 2002). Even though the SRKWs spend the majority of their time in the first 30 m of the water column, 30 m in depth is three times the atmospheric pressure at the surface. The increase in pressure with depth increases gas solubility, thus the killer whales experience greater amounts of gasses dissolving into tissues while diving above depths at which lung  114  collapse occurs (Fahiman et al., 2006; Kooyman, 1973). Since the calculated toxicity doses 2 do not account for the effect of increased gas solubility during shallow for CO and NO dives, they may be misleading, and much lower concentrations may actually pose adverse health effects for killer whales. However, during deep dives when lung collapse occurs, blood is re-directed to 02 sensitive tissues causing the heart and brain to saturate rapidly (Fahiman et al., 2006), and the reduced metabolic rate would result in slower rates of pollutant conversion to metabolites. Thus the calculated doses per kg body mass for killer whales do not apply to deep diving situations. These results are cause for concern as whalewatching is potentially having a greater effect on the SRKWs than previously recognized by behavioural and acoustic studies alone. Life history also plays a role in an individual’s sensitivity to pollutants. Almost half of the SRKW population (43%) is comprised of calves, juveniles, and post-reproductive females, which are members of the population considered sensitive to air pollution. This underscores the importance of regulating emissions in their vicinity since air quality standards only consider the general population and certain individuals may be more susceptible to adverse health effects (USEPA, 2002). When air quality standards are set, the size and nature of sensitive populations are considered (USEPA, 2002). Thus the large proportion of sensitive individuals in the SRKW population indicates that strict standards are required to protect the population’s health. Safety and uncertainty factors applied to air pollution thresholds may provide a level of protection to more sensitive individuals; however, differences in the sensitivity of individual killer whales has not been quantified, and these differences may be larger than what safety and uncertainty factors account for.  115  3.5 REFERENCES Agency for Toxic Substances and Disease Registry (ATSDR). 1993. Toxicological Profile for 1,3-butadiene. Atlanta, GA: U.S. Department of Health and Human Services, Public Health Service. American College of Sports Medicine (ACSM). 2006. Advanced Exercise Physiology, C. M. Tipton (Ed.). Baltimore, MD: Lippincott, Williams & Wilkins. Baird, R. W., Hanson, M. B., Ashe, E. E., Heithaus, M. R., & Marshall, G. J. 2003. Studies ofForaging in “Southern Resident” Killer Whales During July 2002: Dive Depths, Bursts in Speed, and the Use ofa “Crittercam” System for Examining Sub-Surface Behavior (Report Order Number AB133F-02-SE-1744). Seattle, WA: National Marine Fisheries Service, National Marine Mammal Laboratory. Baird, R. W., Hanson, M. B., & Dill, L. M. 2005. Factors influencing the diving behaviour of fish-eating killer whales: sex differences and diel and interannual variation in diving rates. Canadian Journal ofZoology, 83: 257-267. Bates, D. V. 1994. Environmental Health Risks and Public Policy: Decision Making in Free Societies. Vancouver, BC: UBC Press. Bates, D. V., Koenig, J. Q., & Brauer, M. 2003. Health andAir Quality 2002—Phase] Methods for Estimating and Applying Relationships between Air Pollution and Health Effects. Vancouver, BC: British Columbia Lung Association. Beck, B. D., & Clewell, H. J. 2001. Uncertainty/safety factors in health risk assessment: Opportunities for improvement. Human and Ecological Risk Assessment, 7(1): 203207. Béliveau, M., Lipscomb, J., Tardif, R., & Krishnan, K. 2005. Quantitative structureproperty relationships for interspecies extrapolation of the inhalation pharmacokinetics of organic chemicals. Chemical Research in Toxicology, 18(3): 475-485. Bell, M. L., & Davis, D. L. 2001. Reassessment of the lethal London Fog of 1952: Novel indicators of acute and chronic consequences of acute exposure to air pollution. Environmental Health Perspectives Supplements 109(S3): 389-394. Bigg, M. A., & Wolman, A. A. 1975. Live-capture killer whale (Orcinus orca) fishery, British Columbia and Washington waters, 1962-73. Journal ofFisheries Research Board of Canada, 32: 1213-1221. Blaauboer, B. J. 1996. In vitro and other non-animal experiments in the biomedical sciences. ANZCCART News, 9(4): 1-4.  116  Bond, J. A., Butler, M. M., Medinsky, M. A., Muggenburg, B. A., & McClellan, R. 0. 1984. Dog pulmonary macrophage metabolism of free and particle associated [14C]benzo[a]pyrene. Journal of Toxicology and Environmental Health, 14: 181189. Bond, J. A., Johnson, N. F., Snipes, M. B., Mauderly, J. L., & Sega, G. A. 1990. DNA adduct formation in rat alveolar type II cells: Cells potentially at risk for inhaled diesel exhaust. Environmental and Molecular Mutagenesis, 16: 64-69. British Columbia Provincial Health Officer (BCPHO). 2004. Every Breath You Take: Provincial Health Officer ‘s Annual Report 2003. Victoria, BC: Ministry of Health Bruske-Hohlfeld, I., Möhner, M., Abrens, W., Pohiabein, H., Heinrich, J., Kreuzer, M., Jöckel, K. H., & Wichmann, H. E. 1999. Lung cancer risk in male workers occupationally exposed to diesel motor emissions in Germany. American Journal of Industrial Medicine, 36: 405-4 14. Canadian Medical Association. 2008. No Breathing Room, National illness Costs ofAir Pollution (CMA ICAP Technical Report). Retrieved November 25, 2008, from http ://www.cma.calindex.cfm!ci_idJ8683 0/la_id/i .htm Cohen, B. S., Xiong, J. Q., Fang, C. P., & Wei, L. 1998. Deposition of charged particles on lung airways. Health Physics, 74: 554-560. Dhindsa, D. S., Metcalfe, J., Hoversiand, A. S., & Hartman, R. A. 1974. Comparative studies of the respiratory functions of mammalian blood. X. Killer whale (Orcinus orca Linnaeus) and beluga whale (Deiphin apterus leucas). Respiration Physiology, 20: 93-103. Fahlman, A., Olszowka, A., Bostrom, B., & Jones, D. R. 2006. Deep diving mammals: Dive behavior and circulatory adjustments contribute to bends avoidance. Respiratory Physiology and Neurobiology, 153: 66-77. Fanning, J. C. 1974. Proceedings: The structure of the terminal airways and alveolar septum of the dolphin. Journal ofAnatomy, 118(2): 393. Fanning, J. C., & Harrison, R. J. 1974. The structure of the trachea and lungs of the South Australian Bottle-nosed dolphin. In R.J. Harrison (Ed.) The Functional Anatomy of Marine Mammals. London: Academic Press. Fanning, J. C. 1977. The Submicrosopic Structure ofthe Dolphin Lung. PhD Thesis, University of Adelaide. Foot, N. J., Orgeig, S., & Daniels, C. B. 2006. The evolution of a physiological system: The pulmonary surfactant system in diving mammals. Respiratory Physiology and Neurobiology, 154: 118-138.  117  Frumkin, H., & Thun, M. J. 2001. Diesel Exhaust. A Cancer Journalfor Clinicians, 51: 193-198. Gauderman, W. J., Avol, E., Gilliland, F., Vora, H., Thomas, D., & Berhane, K. 2004. The effect of air pollution on lung developmerft from 10 to 18 years of age. New England Journal ofMedicine, 351(11): 1057-1067. Geraci, J. R. & St. Aubin, D. J. 1990. Sea Mammals and Oil: Confronting the Risks. San Diego: Academic Press, Inc. Gerde, P., Medinsky, M. A., Bond, J. A. 1991. Particle-associated polycyclic aromatic hydrocarbons a reappraisal of their possible role in pulmonary carcinogenesis. Toxicology and Applied Pharmacology, 108: 1-13. -  Godard, C. A. J., Wise, S. S., Kelly, R. S., Goodale, B., Kraus, S., Romano, T., O’Hara, T., & Wise Sr., J. P. 2006. Benzo[a]pyrene cytotoxicity in right whale (Eubalaena glacialis) skin, testis and lung cell lines. Marine Environmental Research, 62: S20S24. Goddard, M. J., & Krewski, D. 1992. Comment: Interspecies extrapolation of toxicity data. RiskAnalysis, 12(2): 315-317. Goldberg, I. S., & Smith, R. B. 1981. Settling and diffusion of aerosol particles in small airways during breath holding. Annals ofBiomedical Engineering, 9: 5 57-575. Gregg, I., & Nunn, A. J. 1973. Peak expiratory flow in normal subjects. British Medical Journal, 3: 282-284. Hastie, G. D., Rosen, D. A. S., & Trites, A. W. 2006. The influence of depth on a breathhold diver: Predicting the diving metabolism of Steller sea lions (Eumetopias jubatus). Journal ofExperimental Marine Biology and Ecology, 336: 163-170. Health Effects Institute (HEI). 1988. Air Pollution, the Automobile, and Public Health, A. Y. Watson, R. R. Bates, & D. Kennedy (Eds.). Washington, DC: National Academy Press. Health Effects Institute (HEI). 1999. Diesel Emissions and Lung Cancer: Epidemiology and Quantitative RiskAssessment. A Special Report of the Institute’s Diesel Epidemiology Expert Panel. N. Andover, MA: Flagship Press. International Programme on Chemical Safety (IPCS). 1996. Environmental Health Criteria 171: Diesel Fuel and Exhaust Emissions. Geneva: World Health Organization. Invernizzi, G., Ruprecht, A., Mazza, R., Rossetti, E., Sasco, A., Nardini, S., & Boffi, R. 2004. Particulate matter from tobacco versus diesel car exhaust: an educational perspective. Tobacco Control, 13: 219-221.  118  Kalberlah, F., Föst, U., & Schneider, K. 2002. Time extrapolation and interspecies extrapolation for locally acting substances in case of limited toxicological data. Annals of Occupational Hygiene, 46(2): 175-185. Kirrane, E., Loomis, D., Egeghy, P., & Nylander-French, L. 2006. Personal exposure to benzene from fuel emissions among commercial fishers: comparison of two-stroke, four-stroke and diesel engines. Journal ofExposure Science and Environmental Epidemiology, 17: 151-158. Klanjscek, T., Nisbet, R. M., Caswell, H., & Neubert, M. G. 2007. A model for energetics and bioaccumulation in marine mammals with applications to the Right Whale. Ecological Applications, 17(8): 2233-2250. Kleiber, M. 1961. The Fire ofLife: An Introduction to Animal Energetics. New York: Wiley. Koenig, J. Q. 2000. Health Effects ofAmbient Air Pollution. Norwell, MA: Kiuwer Academic Publishers. Kooyman, G. L. 1973. Respiratory adaptations in marine mammals. American Zoologist, 13(2): 457-468. Kooyman, G. L. 1989a. Diving physiology: marine mammals. In S. C. Wood (Ed.), Comparative Pulmonary Physiology: Current Concepts. New York: Marcel Dekker, Inc. Kooyman, G. L. 1 989b. Diverse Divers: Physiology and Behavior. Berlin: SpringerVerlag. Kooyman, G. L., & Cornell, L. H. 1981. Flow properties of expiration and inspiration in a trained bottle-nosed porpoise. Physiological Zoology, 54(1): 55-61. Kooyman, G. L., Norris, K. S., & Gentry, R. L. 1975. Spout of the gray whale: its physical characteristics. Science, 190: 908-910. Kooyman, G. L., & Sinnett, E. E. 1979. Mechanical properties of the harbour porpoise lung. Respiratory Physiology, 36: 287-300. Kriete, B. 1995. Bioenergetics in the Killer Whale, Orcinus orca. Ph.D. Thesis. Vancouver, BC: University of British Columbia. Kriete, B. 2002. Bioenergetic Changes From 1986 to 200] in the Southern Resident Killer Whale Population, Orcinus orca. Friday Harbor, WA: Orca Relief Citizens’ Alliance. Kruse, S. 1991. The interactions between killer whales and boats in Johnstone Strait, B.C. In K. Pryor & K. S. Norris (Eds.), Dolphin societies: Discoveries and puzzles (pp.  119  149-159). Berkeley, CA: University of California Press. Lavigne, D. M., Barchard, W., limes, S., & Oritsiand, N. A. 1982. Pinniped bioenergetics. In G. S. Clark (Ed.), Mammals ofthe Seas. Small Cetaceans, Seals, Sirenians and Others, vol IV. Rome: FAO Fisheries Series 5. Lavigne, D. M., Times, S., Worthy, G. A. J., Kovacs, K. M., Schmitz, 0. J., & Hickie, J. P. 1986. Metabolic rates of seals and whales. Canadian Journal ofZoology, 64: 279284. Law, R. J., & Whinnett, J. A. 1992. Polycyclic aromatic hydrocarbons in muscle tissue of harbour porpoises (Phocoena phocoena) from UK waters. Marine Pollution Bulletin, 24(11): 550-553. Lenfant, C., Kenney, D.W., & Aucutt, C. 1968. Respiratory function in the killer whale Orcinus orca (Linnaeus). American Journal of Physiology, 215(6): 1506-15 11. Le Tertre, A., Medina, S., Samoli, E., Forsberg, B., Michelozzi, P., Boumghar, A., Vonk, J. M., Bellini, A., Atkinson, R., Ayres, J. G., Sunyer, J., Schwartz, J., & Katsouyanni, K. 2002. Short-term effects of particulate air pollution on cardiovascular disease in eight European cities. Journal ofEpidemiology and Community Health, 56(10): 773779. Lippmann, M. (Ed.). 2000. Environmental Toxicants: Human Exposures and Their Health nd 2 ed.). New York, NY: John Wiley & Sons, Inc. Effects ( Löfroth, G., Burton, R. M., Forehand, L., Hammond, S. K., Seila, R. L., Zweidinger, R. B., & Lewtas, J. 1989. Characterization of environmental tobacco smoke. Environmental Science and Technology, 23: 610-614. Marino, L. 2004. Cetacean brain evolution: Multiplication generates complexity. International Journal of Comparative Psychology, 17: 1-16. Martin, A. R., & Finlay, W. H. 2006. Technical note: A general, algebraic equation for predicting total respiratory tract deposition of micrometer-sized aerosol particles in humans. Aerosol Science, 38: 246-253. Matkin, C., & Saulitis, E. 1997. Restoration Notebook: Killer Whale. Anchorage, AK: Exxon Valdez Oil Spill Trustee Council. Mauderly, J. L. 1992. Diesel exhaust. In M. Lippmann (Ed.), Environmental Toxicants: Human Exposures and Their Health Effects. New York: Van Nostrand Reinhold. McColl, S., Hicks, J., Craig, L., & Shortreed, J. 2000. Environmental Health Risk Management: A Primer for Canadians. Waterloo, ON: Network for Environmental Risk Assessment and Management.  120  Metro Vancouver (MV). 2006. 2006 Air Quality Reportfor the Lower Fraser Valley. Burnaby, BC: Metro Vancouver. Milsom, W. K. 1989. Comparative aspects of vertebrate pulmonary mechanics. In S.C. Wood (Ed.), Comparative Pulmonary Physiology: Current Concepts (pp. 587-6 19). New York: Marcel Dekker, Inc. Mortola, J. P., & Limoges, M. -J. 2006. Resting breathing frequency in aquatic mammals: A comparative analysis with terrestrial species. Respiratory Physiology & Neurobiology, 154: 500-514. National Research Council (NRC). 1991. Human Exposure Assessmentfor Airborne Pollutants. Washington, DC: National Academy of Sciences. Nikula, K. J., Snipes, M. B., Barr, E. B., Griffith, W. C., Henderson, R. F., & Mauderly, J. L. 1995. Comparative pulmonary toxicities and carcinogenicities of chronically inhaled diesel exhaust and carbon black in F344 rats. Fundamental and Applied Toxicology, 25(1): 80-94. Noren, S. R., Cuccurullo, V., & Williams, T. M. 2004. The development of diving bradycardia in bottlenose dolphins (Tursiops truncatus). Journal of Comparative Physiology B, 174: 139-147. Northeast States for Coordinated Air Use Management (NESCAUM). 1999. The Health Effects of Gasoline Constituents —Attachment 1. Boston, MA: Northeast States for Coordinated Air Use Management. Orca Network. 2008. Southern Resident Orca Community Births and Deaths since 1998. Retrieved October 15, 2008, from http://www.orcanetwork.org/news/birthsdeaths.html O’Shea, T. J., & Aguilar, A. 2001. Cetaceans and Sirenians. In R. F. Shore & B. A. Rattner (Eds.), Ecotoxicology of Wild Mammals. Chichester, UK: John Wiley & Sons Ltd. Palmes, E. D., Wang, C. -S., Goldring, R. M., & Altshuler, B. 1973. Effect of depth of inhalation on aerosol persistence during breath holding. Journal ofApplied Physiology, 34(3): 356-360. Perrin, W. F., Wursig, B., & Thewissen, J. G. M. (Eds.) 2002. Encyclopedia ofMarine Mammals. San Diego: Academic Press. Pimentel, D., Cooperstein, S., Randell, H., Filiberto, D., & Sorrentino, 5. 2007. Ecology of increasing diseases: Population growth and environmental degradation. Human Ecology, 35(6): 653-668. Preston, R. J. 2005. Reflections in mutation research: Extrapolations are the Achilles heel of  121  risk assessment. Mutation Research, 589: 153-157. Quan, R. G., Cheng, K. C., Trask, T. C., & Meilleur, R. A. 2002. Marine VesselAir Emissions in the Lower Fraser Valleyfor the Year 2000. Richmond, BC: Levelton Engineering Ltd. Raverty, S., Gaydos, J., Hanson, B., & Wong, S. Unpublished. An Overview ofPost Mortem Findings ofKiller Whales (Orcinus orca) Stranded 1994-2003. Reed, J. Z., Chambers, C., Hunter, C. J., Lockyer, C., Kastelein, R., Fedak, M. A., & Boutilier, R. G. 2000. Gas exchange and heart rate in the harbour porpoise Phocoenaphocoena. Journal of Comparative Physiology [B], 170: 1-10. Renwick, A. G., & Lazarus, N. R. 1998. Human variability and noncancer risk assessment an analysis of the default uncertainty factor. Regulatory Toxicology and Pharmacology, 27: 3-20.  —  Reynolds III, J. E., Perrin, W. F., Reeves, R. R., Montgomery, S., & Ragen, T. J. (Eds.) 2005. Marine Mammal Research: Conservation Beyond Crisis. Baltimore: The John Hopkins University Press. Ridgway, S. H., & Howard, R. 1979. Dolphin lung collapse and intramuscular circulation during free diving: Evidence from nitrogen washout. Science, 206: 1182-1183. Ritchie, G., Still, R., Alexander, W., Nordholm, A., Wilson, III, J., & Mattie, D. 2001. A review of the neurotoxicity risk of selected hydrocarbon fuels. Journal of Toxicology and Environmental Health Part B, 4(3): 223-3 12. Robertson, B. 1980. Basic morphology of the pulmonary defence system. European Journal ofRespiratory Diseases, 61(suppl. 107): 21-40. Ross, P. S. 2000. Marine mammals as sentinels in ecological risk assessment. Human and Ecological Risk Assessment, 6: 29-46. Sample, B. E., & Arenal, C. A. 1999. Allometric models for interspecies extrapolation of wildlife toxicity data. Bulletin ofEnvironmental Contamination and Toxicology, 62: 653-663. Sample, B. E., Opresko, D. M., & Suter, II, G. W. 1996. Toxicological Benchmarks for Wildlfe: 1996 Revision. Oak Ridge, TN: Oak Ridge National Laboratory. Schneider, K., Oltmanns, J., & Hassauer, M. 2004. Allometric principles for interspecies extrapolation in toxicological risk assessment empirical investigations. Regulatory Toxicology and Pharmacology, 39: 334-347. -  Seagrave, J., McDonald, J. D., Gigliotti, A. P., Nikula, K. J., Seilkop, S. K., Gurevich, M., &  122  Mauderly, J. L. 2002. Mutagenicity and in Vivo toxicity of combined particulate and semivolatile organic fractions of gasoline and diesel engine emissions. Toxicological Sciences, 70: 2 12-226. Simpson, J. G., & Gardner, M. B. 1972. Comparative microscopic anatomy of selected marine mammals. In S. H. Ridgway (Ed.), Mammals of the Sea Biology and Medicine. Springfield, IL: Chas. C. Thomas Publishing. -  Snipes, M. B., Boecker, B. B., McClellan, R. 0. 1983. Retention of monodisperse or polydisperse aluminosilicate particles inhaled by dogs, rats, and mice. Toxicology and Applied Pharmacology, 69: 345-362. Society of Environmental Toxicology and Chemistry (SETAC). 1996. Environmental Contaminants in Wildlfe. Interpreting Tissue Concentrations, N. Nelson Beyer, G. H. Heinz, & A. W. Redmon-Norwood (Eds.). Boca Raton, FL: CRC Press, Inc. Spencer, M. P., Gornall, T. A., & Poulter, T. C. 1967. Respiratory and cardiac activity of killer whales. Journal ofApplied Physiology, 22: 974-981. Stahl, W. R. 1967. Scaling of respiratory variables in mammals. Journal ofApplied Physiology, 22: 453-460. Steerenberg, P. A., Zonnenberg, J. A. J., Dormans, J. A. M. A., Joon, P. N. T., Wouters, I. M., Van Bree, L., Scheepers, P. T. J., & Van Loveren, H. 1998. Diesel exhaust particles induced release of interleukin 6 and 8 by (primed) human bronchial epithelial cells (BEAS 2B) in vitro. Experimental Lung Research, 24: 85-100. Stöber, W., Einbrodt, H. J., & Klosterkötter, W. 1967. Quantitative studies of dust retention in animal and human lungs after chronic inhalation. In C. N. Davies (Ed.), Inhaled Particles and Vapours II: Proceedings ofan International Symposium, SeptemberOctober 1965 (pp. 409-418). Oxford, United Kingdom: Pergamon Press. Stöber, W., McClellan, R. 0., Morrow, P. E. 1993. Approaches to modeling disposition of inhaled particles and fibers in the lung. In D. E. Gardner, J. D. Crapo, & R. 0. McClellan (Eds.), Toxicology of the Lung (pp. 527-601). New York: Raven Press. Tu, K. W., & Knutson, E. 0. 1984. Total deposition of ultrafine hydrophobic and hygroscopic aerosols in the human respiratory system. Aerosol Science and Technology, 34(4): 453-465. United States Environmental Protection Agency (USEPA). 1992. Draft Report: A crossspecies scaling factor for carcinogen risk assessment based on equivalence of /day. Federal Register, 57(109): 24152-24173. 4 ” 3 mg/kg United States Environmental Protection Agency (USEPA). 2002. Health assessment documentfor diesel engine exhaust (EPA/600/8-90/057F). Washington, DC: United  123  States Environmental Protection Agency, Office of Transportation and Air Quality. United States Environmental Protection Agency (USEPA). 2004. Air Quality, Health, and Welfare Effects. In Final Regulatory Analysis: Control ofEmissionsfrom Nonroad Diesel Engines (Report No. 420-R-04-007) (pp. 56-2 1 1). Washington, DC: United States Environmental Protection Agency, Office of Transportation and Air Quality. Van Atten, C., Brauer, M., Funk, T., Gilbert, N. L., Graham, L., Kaden, D., Miller, P. J., Rojas Bracho, L., Wheeler, A., & White, R. H. 2004. Honing the Methods: Assessing Population Exposures to Motor Vehicle Exhaust. Montreal, QC: Commission for Environmental Cooperation. West, G. B., Brown, J. H., & Enquist, B. J. 1999. The fourth dimension of life: Fractal geometry and allometric scaling of organisms. Science 284: 1677-1679. Wiles, G. J. 2004. Washington State Status Report For The Killer Whale March 2004. Olympia, WA: Washington Department of Fish and Wildlife. Williams, R., Trites, A. W., & Bain, D. E. 2002. Behavioural responses of killer whales (Orcinus orca) to whale-watching boats: Opportunistic observations and experimental approaches. Journal of Zoology, London, 256: 255-270. World Health Organization (WHO). 2000. Air Quality Guidelines for Europe, (2w’ ed.) (European Series, No. 91). Copenhagen, Denmark: WHO Regional Publications. Yu, C. P., Yoon, K. J., & Chen, Y. K. 1991. Retention modeling of diesel exhaust particles in rats and humans. Journal ofAerosol Medicine, 4(2): 79-115. Zhu, Y., Hinds, W. C., Kim, S., Shen, S. & Sioutas, C. 2002. Study of ultrafine particles near a major highway with heavy-duty diesel traffic. Atmospheric Environment, 36: 4323-4335.  124  4 CONCLUSIONS AND FURTHER STUDIES  4.1 GENERAL CONCLUSIONS The summer habitat of the southern resident killer whale (SRKW) population is subject to heavy vessel traffic, including commercial shipping, fishing, recreation, tourism, and ferries, and whales are almost always in the presence of some type of vessel (Bain et al., 2006). These vessels not only affect the behaviour of the animals (Jelinski et al., 2002; Kruse, 1991; Williams et al., 2002), but they also impact their acoustic abilities (Erbe, 2002). Prior to this study there was no information on how vessels may be affecting the air quality near the SRKWs or on potential adverse health effects resulting from this exposure. The present study used modeling techniques based on highly-predictable dispersal properties of gases to estimate exposures of killer whales to exhaust pollution under average and worstcase weather and vessel scenarios. These exposures were used. to predict internal pollutant doses in killer whales, which were compared with toxic threshold estimates extrapolated from humans. This summary chapter provides recommendations for limiting killer whale exposure to harmful levels of exhaust pollutants and identifies areas for further research. Chapter two established that the atmospheric conditions the SRKWs are exposed to during the commercial whale-watching season are predominantly stable, which can cause an accumulation of air pollutants above the surface of the water where the whales breathe. The results from the dispersion model indicated that the wind angle had the largest effect on killer whale exposure to air pollutants, with downwind angles (2100 and 240°) producing the highest concentrations. The ranking of variables that followed was: the distance of the vessels to the whale (buffer distance), which was equally as important as the mixing height,  125  followed by the number of vessels, which was equally as important as the inter-vessel distance, and finally the wind speed. Generally the Metro Vancouver Air Quality Objectives 2 were exceeded when: the wind came from an angle of 1500, (MV, 2006) for CO and NO 180°, 210° or 240°; the wind speed was between 1-9 m s’; and the mixing height was less than 6 m; the buffer distance was less than 20 m; the number of vessels was greater than 27; and the inter-vessel distance was less than 50 m. Even under average-case conditions with 20 vessels maintaining the recommended 2 could be 100 m distance from the whale and each other, the MV AQOs for CO and NO exceeded. The mean CO concentration predicted by the average-case simulations (11.98 mg ) at 100 m from the source was five times greater than that measured 30 m from a busy 3 m Los Angeles highway (Zhu et al., 2002). Based on measurements of cigarette emissions by Löfroth et al. (1989), during 1 -hour of average whale-watching conditions the male killer whale would potentially experience the equivalent of 3.5 cigarettes, and the female 1.8 2 concentration predicted by the average-case simulations (0.04 mg cigarettes. The mean NO ) at 100 m from the source was just above that measured at a distance of 115 m from busy 3 m motorways (Roorda-Knape et al., 1998). Therefore, the model predicted that even averagecase whale watching scenarios can produce worse air quality than along busy highways, and worst-case scenarios can produce extreme pollution episodes that greatly exceeded the MV AQOs. The average-case dispersion model simulations in Chapter two were essentially “best 2 case” whale-watching conditions, and the mean exposure concentrations of CO and NO from these simulations were used to predict the internal pollutant dose male and female killer whales receive. It was determined that the doses of CO (Table 3.1) from average-case  126  whale-watching conditions were equal to or slightly higher than those predicted to cause 2 were adverse health effects in killer whales (Table 3.3), while average-case doses of NO slightly lower than those predicted to cause adverse health effects. Unfortunately, the Be Whale Wise Guideline (DFO, 2008) that vessels must maintain 100 m from the whales is violated frequently (Koski, 2006), thus this “best-case” scenario does not often hold true. Since the atmospheric conditions during the whale-watching season are highly conducive to air pollutant accumulation, the whale-watching conditions can easily become “worst-case”. 2 calculated from worst-case exposure simulations (Table 3.1) were The doses of CO and NO on average 6.6 and 3.4 times greater respectively than those expected to cause adverse health effects in killer whales. This suggests the male killer whale can potentially experience the equivalent of 22.5 cigarettes, and the female 11.2 cigarettes during 1-hour of worst-case whale-watching conditions. The monthly average and maximum ambient concentrations of CO (Table 2.3) measured at the Christopher Point, BC air quality monitoring station from 2005-2007 are all well below the calculated toxicity values for male and female killer whales (Table 3.2). 2 2 are just below the NO However, the monthly maximum ambient concentrations of NO 2 toxicity values for both male and female killer whales. Thus occasionally peak ambient NO concentrations alone are almost at the level predicted to be a health hazard to killer whales. The dispersion model estimated 1-hour exposures to air pollutants, but whale watching occurs on average 12-hours a day during the peak season (Bain, 2002). When exposed to pollutants for longer than 1-hour, the exposure concentration considered safe is , which is 3 lowered. For example the MV AQO for an 8-hour exposure to CO is 10 mg m 3 (MV, 2006). However, it is unlikely that much lower than the 1-hour exposure of 30 mg m  127  the whales would be exposed to the situation modeled for 12-hours per day, but the situation may apply for several hours during the day. Thus a toxicity dose for 8-hour exposure to CO was calculated (Table 3.3), and it was much lower than the average-case 8-hour dose of CO (Table 3.1), which suggests that long exposures are a health hazard for the whales. Since the SRKWs are being exposed to air pollutants for longer durations than were modeled, the calculated 1-hour toxicity doses are likely conservative. Chapter three also included factors that may affect killer whale sensitivity to air pollutants. The SRKWs spend 97.6% of their time within 30 m of the surface (Baird et a!., 2005; 2003), which is above the depth that lung collapse occurs (Ridgway & Howard, 1979). However, even shallow dives increase pressure in the lungs, and would cause greater gas solubility and uptake into the systemic circulation. During deep dives (which do not occur that often) metabolic rate is depressed and would result in a slower rate of pollutant conversion to metabolites. Furthermore, during deep dives blood flow is preferentially directed to 02 sensitive tissues such as the heart and brain, and if pollutants are present in the systemic circulation, then those tissues could potentially receive higher doses (Fahlman et al., 2006; Kooyman, 1973). In addition, 43% of the SRKW population is post-reproductive females, juveniles, and calves that are likely to experience adverse health effects at concentrations lower than the calculated toxicity values (Table 3.2). The dose equation did not take the effects of diving or extra-sensitive individuals into account, which also makes the calculated toxicity values and doses conservative. However, it may be that safety and uncertainty factors make up for this deficiency. The main results from this thesis are summarized in bullet form below.  128  •  The atmospheric conditions in SRKW habitat during the commercial whale-watching season are predominantly stable, which can cause an accumulation of air pollutants above the surface of the water where the whales breathe.  •  Wind angle had the largest effect on killer whale exposure to air pollutants, and the highest exposures occurred when the whale was downwind of vessels.  •  2 concentrations that Average-case whale-watching simulations produced CO and NO occasionally exceeded the Metro Vancouver Air Quality Objectives (i.e. with low wind speeds, and mixing heights).  •  2 Worst-case whale-watching simulations produced extreme CO and NO concentrations that greatly exceeded the Metro Vancouver Air Quality Objectives.  •  2 calculated from average-case whale-watching conditions were Doses of CO and NO approximately equal to and lower respectively than those predicted to cause adverse health effects in killer whales.  •  2 calculated from worst-case exposure simulations were on Doses of CO and NO average 6.6 and 3.4 times greater respectively than those expected to cause adverse health effects in killer whales.  •  2 concentrations in SRKW habitat are just below the level predicted Peak ambient NO to be a health hazard to killer whales.  •  SRKWs are exposed to air pollutants for longer than the 1-hour duration modeled, thus the calculated 1-hour toxicity doses are likely conservative.  •  During shallow dives there would be greater gas uptake into the systemic circulation than while at the surface, due to increases in gas solubility.  129  •  During deep dives there may be slower rates of pollutant conversion to metabolites due to metabolic depression, and the heart and brain could potentially receive higher doses of pollutants because of the re-direction of blood flow.  •  43% of the SRKW population can be considered sensitive to air pollution, and would likely experience adverse health effects at concentrations lower than the calculated toxicity values.  4.2 UNCERTAINTY AND ASSUMPTIONS IN THE MODELS Mathematical models never describe the real world completely, and the dispersion and allometric models included in this thesis have many simplifying assumptions. In some cases the model parameters could not be measured, and were calculated using formulae with their own simplifying assumptions (e.g. using the equation by Stahl (1967) to estimate respiratory dead space volume), further increasing uncertainty. The central assumptions and most significant sources of uncertainty in the models are described below. This study helped identify data gaps and lack of suitable models necessary for characterizing marine recreational vessel emissions. No published information exists on the emissions of recreational marine engines engaged in whale-watching, which prevented their inclusion in the dispersion model. Therefore only commercial whale-watching vessel engine configurations were considered, yet recreational marine engines may be very different on average than those used by commercial whale-watching companies. The engine configurations for half of the whale-watching fleet were unknown and were assumed to be identical to those that were known, and this assumption may not be valid in the real world. The emission rates for the vessels were not adjusted for age or other factors. The percent  130  retention of wet exhaust constituents in the water column was obtained from a few published studies, which were prone to inaccuracies. The dispersion model only considered a very specific situation where the whales and vessels were continuously moving at a constant speed (steady-state engine operation), yet whale-watching vessels commonly shut down their engines when viewing the whales. However, transient operating conditions (during arrival, repositioning, and exiting) often have increased emissions compared to steady-state operation (Graskow, 2001; USEPA, 2002a). The dispersion model assumed that the profile of vertical pollution diffusion initially decreased rapidly with decreasing height in the atmosphere and then reached a plateau close to the surface. The dispersion model predicted that air pollutant concentrations decreased to 31% of the original at a distance of 100 m from the source, whereas empirical roadway studies have measured a 50% decrease over the same distance. Thus the dispersion model may have underestimated exposure concentrations. As discussed in Chapter three, the uncertainty factor from the allometric model alone is 100, due to inter-individual variability and animal-to-human extrapolation (Renwick & Lazarus, 1998). It was assumed that the health effects from exposure to air pollutants that occur in small mammals and humans also occur in killer whales. It was also assumed that inhalation was the only route of exposure to exhaust pollutants, which may not be the case (especially with wet exhaust systems). Some of the respiratory rates, volumes, and body masses for male and female killer whales used to calculate dose were obtained from allometric scaling of data from captive killer whales, and are likely inappropriate for all animals in the SRKW population. In typical risk assessments, when there is a high degree of uncertainty with the model and/or data included, there is a corresponding low exposure limit (toxicity value) set as a  131  safeguard against any underestimation of the potential health effects from the pollutant in question (NRC, 1991). This reasoning should also apply to the SRKWs, and only improved information, data, or models should result in raising the exposure limit. Furthermore the SRKWs are not only being exposed to air pollution from whale-watching vessels, as there are numerous pollutant sources in their summer habitat, which could produce additive or synergistic effects. For example, when diesel exhaust is inhaled along with 03, there is a significant increase in lung inflammation in rats (USEPA, 2002b).  4.3 FUTURE RESEARCH Future research to improve this study would include a more comprehensive dispersion model that captures: vessels exiting and entering the scene, vessels operating independently of one another so that the distances from the whale and each other can vary during the simulation, vessels shutting down their engines, and changing killer whale and vessel speeds. It would also be helpful to obtain more accurate representations of time-spent whale-watching from whale-watching vessel operator logs. Information on all the engines used by whale-watching companies would also help, however, that could not be done for recreational vessels engaged in whale-watching because those vessels are constantly changing. Measurements of all the variables included in the model would also be helpful, such as determining emission rates from the fleet of whale-watching vessels, calculating the amount of time killer whales are downwind of vessels (especially at the worst wind angles of exposure 2100 and 240°), and measuring actual vertical mixing heights. To validate the dispersion model, empirical measurements of air pollutant concentrations around the SRKWs  are necessary.  132  Improved estimates and measurements of killer whale respiratory anatomy and physiology would allow the use of more sophisticated physiological models to determine more accurate pollutant doses and toxicity values. Future studies to empirically determine killer whale pollutant doses are biopsy sampling to determine if pollutant metabolites are present, and collection of exhaled air to determine presence of pollutants. Lowering engine emissions standards reduces air pollutant emissions (Table 2.6), but the full benefits of these standards are not realized until there has been a significant turnover in the fleet, and this takes about 12 years for a 90% turnover to occur (NESCAUM, 1999). There is potential for alternative fuels to lower air pollutant emissions (Appendix Dl), and the benefits begin immediately upon use of the fuel and apply to all engines, regardless of their age, technology, or operating conditions (NESCAUM, 1999). However, new technology and alternative fuels can have unintended consequences on air quality as discussed in Appendix D2 and section 3.1.1 in Chapter three. Thus relying on engine and fuel improvements to reduce killer whale exposure to air pollutants may not be an effective solution, and further research in this area is required. This study is the first investigation of whale-watching vessel exhaust emissions, and has demonstrated that in certain situations the SRKWs may be inhaling concentrations of air pollutants that have the potential to cause serious health effects. It can be argued that the “average” whale-watching conditions modeled may not arise frequently in the real world, and may not be very realistic due to the number of assumptions involved. However, these conditions were based upon published averages of whale-watching and atmospheric behaviour. The sensitivity analysis of the dispersion model determined that the wind angle had the largest effect on the whale’s air pollution exposure, and this variable can be  133  examined in further detail. In average-case simulations with wind approaching the whale from the worst angles of 210° and 240°, only emission plumes from the vessels paralleling the whale on one side had the potential to reach the whale (i.e. from ten vessels), and of those, only vessel emissions upwind of the whale had the potential to reach the whale (i.e. from five vessels). This suggests that only five vessels can deteriorate air quality to a point where it is harmful to the whale, thus any whale-watching situation with at least five vessels upwind of whales with their engines running is potentially problematic. Furthermore, the worst-case simulations only considered 40 vessels, whereas the highest number of vessels counted around the SRKWs is 120, thus the worst-case simulations may be conservative. Until further studies are conducted that provide more reliable estimates of exposures and health effects in killer whales, the precautionary principle should be adhered to. However, it is up to decision makers (in Canada, the Ministry of Fisheries and Oceans) to determine if the SRKWs involuntary exposure to exhaust emissions are an acceptable or tolerable risk to the population, based on the probability of harmful health effects, the means of controlling emissions, and the expected costs and benefits of doing so (McColI et al., 2000). Yet, this is one threat to the population that can be easily managed by imposing limits on the number of vessels whale-watching, limits on the amount of time whale-watching vessels are allowed to remain with the whales, critical habitat areas where vessels are not allowed, air pollution standards on the engines in use, and tougher enforcement of the Be Whale Wise Guidelines (DFO 2008). In addition, it must be recognized that air pollution from the vessels is not only a health threat to this endangered species, but it also threatens the health of vessel operators, naturalists, and tourists on board the vessels.  134  4.4 REFERENCES Bain, D. E. 2002. A Model Linking Energetic Effects of Whale Watching to Killer Whale (Orcinus orca) Populations. Friday Harbor, WA: Friday Harbor Laboratories, University of Washington. Bain, D. E., Williams, R., Smith, J. C., & Lusseau, D. 2006. Effects of Vessels on Behavior ofSouthern Resident Killer Whales (Orcinus spp.) 2 003-2005 (NMFS Contract Report No. AB133F05SE3965). Retrieved March 18, 2007, from http ://www.nwfsc.noaa.gov/researchldivisions/cbdlmarinemammal/documents/bainn mfsrep2003 -5final.pdf Baird, R. W., Hanson, M. B., Ashe, E. E., Heithaus, M. R., & Marshall, G. J. 2003. Studies ofForaging in “Southern Resident” Killer Whales During July 2002: Dive Depths, Bursts in Speed, and the Use of a “Crittercam” System for Examining Sub-Surface Behavior (Report Order Number AB133F-02-SE-1744). Seattle, WA: National Marine Fisheries Service, National Marine Mammal Laboratory. Baird, R. W., Hanson, M. B., & Dill, L. M. 2005. Factors influencing the diving behaviour of fish-eating killer whales: Sex differences and diel and interannual variation in diving rates. Canadian Journal ofZoology, 83: 257-267. Department of Fisheries and Oceans Canada (DFO). 2008b. Viewing Guidelines. Pacific Region Marine Mammals and Turtles. Retrieved May 14, 2008, from http ://www.pac. dfo-mpo gc.calspecies/marinemammals/view_e.htm .  Erbe, C. 2002. Underwater noise of whale-watching boats and potential effects on killer whales (Orcinus orca), based on an acoustic impact model. Marine Mammal Science, 18: 394-418. Fahlman, A., Olszowka, A., Bostrom, B., & Jones, D. R. 2006. Deep diving mammals: Dive behavior and circulatory adjustments contribute to bends avoidance. Respiratory Physiology and Neurobiology, 153: 66-77. Graskow, B. R. 2001. Design and Development ofa Fast Aerosol Size Spectrometer. University of Cambridge Ph.D. Thesis. Jelinski, D. E., Krueger, C. C., & Duffus, D. A. 2002. Geostatistical analyses of interactions between killer whales (Orcinus orca) and recreational whale-watching boats. Applied Geography, 22: 393-411. Kooyman, G. L. 1973. Respiratory adaptations in marine mammals. American Zoologist, 13(2): 457-468. Koski, K. L. 2006. Soundwatch Public Outreach/Boater Education Project 2004-2005 Final Program Report. Friday Harbor, WA: The Whale Museum.  135  Kruse, S. 1991. The interactions between killer whales and boats in Johnstone Strait, B.C. In K. Pryor & K. S. Norris (Eds.), Dolphin Societies: Discoveries and Puzzles (pp. 149-159). Berkeley, CA: University of California Press. McColl, S., Hicks, J., Craig, L., & Shortreed, J. 2000. Environmental Health Risk Management: A Primer for Canadians. Waterloo, ON: Network for Environmental Risk Assessment and Management. Metro Vancouver (MV). 2006. 2006 Air Quality Reportfor the Lower Fraser Valley. Burnaby, BC: Metro Vancouver. National Research Council (NRC). 1991. Human Exposure Assessment for Airborne Pollutants. Washington, DC: National Academy of Sciences. Northeast States for Coordinated Air Use Management (NESCAUM). 1999. The Health Effects of Gasoline Constituents —Attachment 1. Boston, MA: Northeast States for Coordinated Air Use Management. Renwick, A. G., & Lazarus, N. R. 1998. Human variability and noncancer risk assessment an analysis of the default uncertainty factor. Regulatory Toxicology and Pharmacology, 27: 3-20.  —  Ridgway, S. H., & Howard, R. 1979. Dolphin lung collapse and intramuscular circulation during free diving: evidence from nitrogen washout. Science, 206: 1182-1183. Roorda-Knape, M., Janssen, N., De Harthog, J., Van Vliet, P., Harssema, H., & Brunekreef, B. 1998. Air pollution from traffic in city districts near major motorways. Atmospheric Environment, 32(11): 1921-1930. Stahl, W. R. 1967. Scaling of respiratory variables in mammals. Journal ofApplied Physiology, 22: 453-460. United States Environmental Protection Agency (USEPA). 2002a. A Comprehensive Analysis ofBiodiesel Impacts on Exhaust Emissions (EPA42O-P-02-00l). Washington, DC: United States Environmental Protection Agency, Office of Air and Radiation. United States Environmental Protection Agency (USEPA). 2002b. Health Assessment Documentfor Diesel Engine Exhaust (EPAJ600/8-90/057F). Washington, DC: United States Environmental Protection Agency, Office of Transportation and Air Quality. Williams, R., Trites, A. W., & Bain, D. E. 2002. Behavioural responses of killer whales (Orcinus orca) to whale-watching boats: Opportunistic observations and experimental approaches. Journal ofZoology, London, 256: 255-270.  136  Zhu, Y., Hinds, W. C., Kim, S., & Sioutas, C. 2002. Concentration and size distribution of ultrafine particles near a major highway. Journal of the Air & Waste Management Association, 52(9): 1032-1042.  137  S 3 APPENDICE  Appendix A: Be Whale Wise Guidelines (DFO 2008).  1. BE CAUTIOUS and COURTEOUS: approach areas of known or suspected marine wildlife activity with extreme caution. Look in all directions before planning your approach or departure. 2. SLOW DOWN: reduce speed to less than 7 knots when within 400 metres/yards of the nearest whale. Avoid abrupt course changes. 3. KEEP CLEAR of the whales’ path. If whales are approaching you, cautiously move out of the way. 4. DO NOT approach whales from the front or from behind. Always approach and depart whales from the side, moving in a direction parallel to the direction of the whales. 5. DO NOT approach or position your vessel closer than 100 metres/yards to any whale. 6. If your vessel is not in compliance with the 100 metres/yards approach guideline (#5), reduce your speed and cautiously move away from the whales 7. STAY on the OFFSHORE side of the whales when they are traveling close to shore. 8. LIMIT your viewing time to a recommended maximum of 30 minutes. This will minimize the cumulative impact of many vessels and give consideration to other viewers. 9. DO NOT swim with, touch or feed marine wildlife.  References cited within appendices are listed in the reference section of the chapter they were mentioned in. 138  Appendix B: Air Pollutant Emissions Table Bi: Total emissions of air pollutants from all sources during the year 2000 for the Lower Fraser Valley Airshed (GVRD, 2002). Emissions (metric tonnes) Air Pollutant 15,228 10 PM 9,021 25 PM 100,090 N0 19,015 SO, 112,360 VOC 483,083 Co 23,022,272 2 Co 2,754 0 2 N 258,508 Principal smog-forming pollutants (NO, VOC, ) 3 , SO,,, NH 5 . 2 PM Table B2: Total emissions of air pollutants from ocean-going vessels during 2005-2006 in British Columbia* (The Chamber of Shipping, 2007). Emissions (tonnes per year) Air Pollutant 1,604 10 PM 1,438 5 . 2 PM 26,500 NO 18,413 SO, 2,236 CO 1,278,084 2 Co 36 N20 934 HC 128 4 CH 28 3 NH *This area includes all inland and territorial waters along the BC coast, the U.S. and Canadian portions of the Strait of Juan de Fuca, and oceanic waters extending 50 nautical miles offshore. Results were also compiled for the LFV region, which includes all of the MV and the southwestern portion of the Fraser Valley Regional District (FVRD).  139  Table B3: Emissions (tonnes per year) for ocean-going vessels, harbour vessels, ferries, fishing vessels and recreational vessels in BC outside of Metro Vancouver and FVRD* and Vancouver Island for the year 2000jQuan et al., 2002b. Vancouver Island Emissions B.C. and WA Emissions Air Pollutant (tonnes per year) (tonnes per year) 2,095 4,145 CO 840,930 1,350,024 2 CO 649 1,332 VOC 23,867 38,404 Total NO 14,944 28,047 NO 954 1,535 2 NO 1,037 1,592 5 . 2 10 & PM PM *Canadjan portion included Vancouver Island and the coast of BC, and the U.S. portion included Washington State, Whatcom County, Puget Sound, and the coast of Washington.  140  Appendix C: Air Quality Standards Table Cl: The United States USEPA National Ambient Air Quality Standards (Cooper & Alley, 2002). Averaging Period ) 3 Standard (mg m Air Pollutant 8-hour 10 CO 1-hour 40 Annual mean 0.1 2 NO Annual mean 0.05 10 PM 24-hour 0.15 Annual mean 15 0.0 5 . 2 PM 24-hour 0.065 Table C2: The World Health Organization Air Quality Guidelines for Europe (WHO 2000). Averaging Period ) 3 Air Pollutant Standard (mg m 30-minute 60 CO 1-hour 30 1-hour 0.2 2 NO nla Dose response PM  141  Appendix D: Alternative Fuels and Fuel Additives  Appendix Dl: The Effect of Alternative Fuels on Engine Emissions The advantages of using natural and petroleum gasses are that they emit approximately 50% less CO and VOCs, very few photochemically active VOCs, and no toxins like benzene, while the disadvantages of natural and petroleum gasses are problems with on-board fuel storage, handling, and refueling (Cooper & Alley, 2002). Biodiesel is made from vegetable or animal fats; it is non-toxic and biodegradable and is increasingly being used in marine engines as it can be blended with regular diesel fuel (BCLA, 2005). Compared to conventional diesel fuel, biodiesel is relatively clean burning as it produces , PM, and has less 03 forming potential, but produces 2 fewer emissions of HCs, CO equivalent or slightly higher emissions of NO (USEPA, 2002a).  Appendix D2: The Effect of Fuel Additives on Engine Emissions Fuel additives are used to reduce engine emissions of HCs, CO and PM, and often benzene, formaldehyde, and 1 ,3-butadiene are reduced as well (Zhu et al., 2003). However, additives often decrease emissions of certain pollutants while increasing emissions of others (Zhu et al., 2003). Methyl Tertiary Butyl Ether (MTBE) is a gasoline oxygenate and octane enhancer additive used in automobile and marine engines; however, due to water contamination issues its use is declining in the United States (NESCAUM, 1999). The use of MTBE decreases emissions of benzene, 1 ,3-butadiene, and acetaldehyde, but it increases formaldehyde emissions (NESCAUM, 1999). MTBE in air and water has harmful health effects: it is an animal (and possibly human) carcinogen; it produces harmful neurological  142  effects in sensitive individuals; it has harmful reproductive and developmental effects; and at high concentrations it can be toxic to the kidney, liver, and endocrine system (NESCAUM, 1999). However, MTBE is one of the least toxic compounds in gasoline and is much less toxic than either 1 ,3-butadiene or benzene, yet even at extremely low concentrations humans can smell and taste it in air and water (NESCAUM, 1999). Ethanol and ether are often used instead of MTBE to oxygenate gasoline as they reduce emissions of CO. NO and photochemically reactive compounds (Cooper & Alley, 2002; NESCAUM, 1999; Rice et al., 1991). Unfortunately ethanol and ether increase the amount of acetaldehyde and formaldehyde in engine emissions, which are both probable human carcinogens (NESCAUM, 1999). Ethanol added to gasoline may also increase the volatility of other compounds like benzene (NESCAUM, 1999). In addition, two studies by 4 2 and CH Environment Canada (unpubL) found that as ethanol content in fuel increased: CO emission rates remained essentially unchanged; decreases in CO were not always statistically significant; there were consistent increases in NO emissions; HC emissions increased in some engines but decreased in others; and VOC emissions decreased. While fuel additives play a role in decreasing certain emissions, the health effects from other compounds produced by the additives must be considered. MTBE, formaldehyde and acetaldehyde are probable human carcinogens, and this almost certainly applies to marine mammals; therefore, the impacts of using fuel additives in the marine environment should be carefully examined.  143  Appendix E: Programming Code for the NetLogo Dispersion Model  breed [whales whale] breed [boats boat] ;; adapted from “Models> Code Examples> Diffuse Off Edges Example” patches-own pollution new-pollution  ;; this is the quantity we will be diffusing ;; this is the quantity diffused/moved by wind  ] globals [edge-patches main-patches polluter-patches num-boat-in-a-row factor yboat index whale-direction  ;; border patches where pollution should remain 0  ;; patches not on the border ;; patches creating pollution ;; number of boats in a row before stacking to next row ;; multiplication factor for row length ;; equal to buffer distance ;; multiplication factor for row length ;; the random angle that the whale moved at  ] to setup ;; Create boats & whales (both are turtle agents) clear-all set-default-shape whales “shark” create-whales 1 [ set color green set size 4 set heading 90 ;; start facing east  ]  create-boats boatnumber [ ;; number of boats is set by the slider set size 2 set heading 90 ;; start facing east like the whale  ;; the following determines the distribution of boats on either ;; side of the whale (in rows), so that the boats are at the buffer ;; distance from the whale and at the interboat-dist from each other set factor 0 set num-boat-in-a-row 0 set yboat buffer set index 1 repeat boatnumber [ if num-boat-in-a-row / 2 >= (number-of-buffers * buffer) / interboat-dist set yboat (yboat + interboat-dist) set num-boat-in-a-row 0  [  144  set factor 0  I  set num-boat-in-a-row num-boat-in-a-row + 1 if num-boat-in-a-row mod 4 1 [ set factor factor + 1 ifindexmod4= 1 [ if-elsefactor=1 [ set [xcor] of boat index 0  ][  set [xcor] of boat index ( interboat-dist set [ycor] of boat index yboat set [color] of boat index white  *  (factor 1 ))j -  ]  if index mod 4 = 2 [ set [xcor j of boat index (interboat-dist set [ycor] of boat index yboat set [color] of boat index orange  *  (- factor))  if index mod 4 = 3 { if-else factor = 1 [ set [xcor] of boat index 0 set [xcor] of boat index (interboat-dist set [ycor] of boat index (- yboat) set [color] of boat index blue  *  (factor 1))] -  ]  if index mod 4 = 0 [ set [xcor I of boat index (interboat-dist set [ycor] of boat index (- yboat) set [color] of boat index red  *  (- factor))  set index index + 1  I  identify edge and non-edge patches. code - only worked with world wrap checkboxes cleared original ;; ;; otherwise all patches have 8 neighbors ;; set edge-patches patches with [count neighbors != 8] ;; set main-patches patches with [count neighbors 8] ;; this new code works independently of world wrap conditions as long as ;; origin is at center. With world wrapping on turtles are allowed to wrap ;; around boundaries (so they don’t get stuck on edges) but pollution won’t. ;; Can think of it as a “conserved” system: as one boat leaves one side, another ;; comes in on the opposite edge. set edge-patches patches with [abs pxcor = max-pxcor or abs pycor = max-pycor] set main-patches patches with [ abs pxcor < max-pxcor and abs pycor < max-pycor] ;;  145  set polluter-patches n-of boatnumber boats ;; only patches w/boats produce pollution recolor end to go move-whales move-boats blow-wind diffuse-off-edges tick end to move-whales ask whales [ set whale-direction random ( 2 * max-degrees + 1) max-degrees right whale-direction ;; movement of whale is random forward 1 -  end to move-boats ask boats [ right whale-direction forward 1 set pollution pollution + 70.2  ]  ;; by the same speed of the whale ;; dummy pollution emission rate set at ;; 70.2 mg/tick = 100 mg/second  end to blow-wind ;; moves ‘pollution’ from all patches in the same direction. ;; Works by finding the source square (where is the wind moving ;; air from?) and depositing it in variable new-pollution. Once all ;; new-pollutions are set then it’s safe to replace old ‘pollutions’. ;; Kind of tricky because source “square” will probably overlap 4 ;; source patches so we have to sum over all parts of overlapping source ;; patches. ask patches [ ;; find center of square where wind is coming from let source-x pxcor + (wind-speed + le-6) * sin wind-angle let source-y pycor + (wind-speed + le-6) * cos wind-angle set new-pollution weighted-mean-pollution source-x source-y  ]  ask patches [ set pollution new-pollution  146  if random-wind [change-wind] end to diffuse-off-edges ;; diffuse pollution but remove any that has reached the edges diffuse pollution diffusion-constant ask edge-patches [set pollution 0] recolor end to recolor ;; color patches proportionately to pollution ask patches [set pcolor scale-color blue pollution 0 100] end to-report weighted-mean-pollution [x y] ;; reports mean ‘pollution’ in unit square with top-left corner at (x,y) contributed-pollution x y -0.5 0.5 let sum-pollution ;; top left corner set sum-pollution sum-pollution + contributed-pollution x y 0.5 0.5 ;; top right corner set sum-pollution sum-pollution + contributed-pollution x y 0.5 -0.5 ;; bottom right corner set sum-pollution sum-pollution + contributed-pollution x y -0.5 -0.5 ;; bottom left corner report sum-pollution end to-report contributed-pollution [x y deltax deltay] ;; find how much area of unit square at (x,y) overlaps into ;; patch at offset (deltax,deltay) and calculate contribution ;; of ‘pollution’ to unit square from overlap ;; find corner of square let corner-x x + deltax let corner-y y + deltay let corner-patch patch corner-x corner-y ;; override edge-wrapping if ( [pxcor] of corner-patch ! = round corner-x) or ( [pycor] of corner-patch ! round corner-y) [ ;; this condition will only be true if edge wrapping has ;; occurred in finding corner-patch report 0 if-else corner-patch = nobody report 0  [  147  find opposite corner of patch that square corner lands on let patch-x [pxcor] of corner-patch deltax let patch-y [pycor] of corner-patch deltay let area ( abs (corner-x patch-x)) * (abs (corner-y patch-y)) report area * [pollution] of corner-patch ;;  -  -  -  -  end to change-wind ;; randomly shift wind speed and direction set wind-speed wind-speed + 0.01 * (random 3 1) set wind-speed median lput wind-speed [0 1]; must be set wind-angle (wind-angle + random 3 1) mod 360 end -  >=  0 and  <=  1  -  to-report concentration ;; of the whale’s patch ;; 2 is the patch size (2m x 2m) ;; mixing-height set by slider ;; 2.85 is the boat speed in rn/s report [(new-pollution / ( 2 * mixing-height * 2.85))] of whale 0 end  148  Appendix F: Sensitivity Analysis Results from the Dispersion Model. 0.30 80.0 70.0  0.20 -  )(  o o0C  !,I 1 /  /  /  !I fj\  40.0  8  /  /  60.0 E 50.0  0.25  D  1  \/  :\  \. v \.  0.0  2.0  0.05  \  •  •  0.0  0:100  _‘.  /\\  /_ \\ / /7  10.0  C 0  II  \  \\  0.15  \  \  \:,  II  20.0  \.  6.0  4.0  ê  —r  8.0  10.0  1 12.0  0.00 14.0  Wind speed (mis) . ——--v--—— ——-— ——s-— 0  90° 120° 150° 180° 210°  ——0-——  240°  — —+— —  270° CO MV AQO (30 2 MV AQO (0.2 NO  —  —  —  —  mg/rn ) 3 ) 3 mg/rn  2 concentrations as a function of Figure Fl: Deterministic simulation results with CO and NO angle. wind wind speed and  149  0.18  50.0  0.16 / /  40.0  S  Y—  C)  /  2  0.14  /  04  \ 30.0  0  —.  /  2  /  —..  0.12  ‘.  I  \ I  0.10  A  0.08  2 C 0 (U  (‘3 C C 0 0  0)  20.0  0.06  0  I / I  C)  10.0  /  I. ,.  0.0 0.0  H  k\\  ...  (N  \  \. ...  0  o  0.04  O.  z  .  0.02  . Wind speed —..—  10.0  8.0  6.0  4.0  2.0  C () C 0 0  12.0  0.00 14.0  (mis)  90°  150°  —— 210° ) 3 co MV AQO (30 mg/rn  2 concentrations as a function of Figure F2: Stochastic simulation results with CO and NO the wind speed and random mean wind angle of either 90°, 150° or 210°. 0.5  140.0  120.0 0.4  /  2 100.0 0)  S. 0)  /  S  £ 0.3  /  80.0  0  /  Ct’  CU C  -  60.0  -  0 C  -—7——— /  0 0  o C-)  0.2  40.0  0 0 (N  0  I  0.1  p  ,  I  20.0  a)  z  /  /,/ 0.0  , I  0.0  —  —----.—I  0.6  0.4  0.2  —V.—  —  0.8  0.0  1.0  Diffusion constant 90°  ———-—— —.-.—‘-—•.-  150°  ——•——  180° 210° 240° 270°  —  mg/rn ) Co MV AQO (30 3 ) 3 2 MV AQO (0.2 mg/rn NO  ——*—  —.—o—-•—  —  —  —  2 concentrations as a function of Figure F3: Deterministic simulation results with CO and NO the pollutant diffusion constant and wind angle. 150  80.0 70.0  0.25  60.0  V-v  0.20  E 500  C  -  C 0  —-----------  .2 0.15  40.0 C  30.0  0.10  320.0 0.05 0  10.0  0  0 00  1.0  0.4  Diffusion constant 900  _..__  1500 2100  —v—  ) 3 CO MV AQO (30 mg/rn ) 3 2 MV AQO (0.2 mg/rn NO  —  — .  2 concentration as a function of the Figure F4: Stochastic simulation results with CO and NO diffusion constant and random mean wind angle of either 90°, 150° or 2 10°. 500.0  1.8  .  1.6 400.0  \  ‘  1.4  E 1.2 300.0  .1.0  o  \\ o  .2  0.8 o  2000  0.68  o  z 4 . 0  \.\ 100.0  ___0 ..  —  0.0  —  -  0.0  —  —  -  —.._  -ozj  ,.WV—r—-——6.0  4.0  2.0  8.0  0.2  0.0 10.0  Mixing height (m) 900  0’ ___,  120 1500  180  2l0 240  —  — —  —  —  — —  270 Co MV AQO (30 mgim3) 2 MV AQO (0.2 mgim3) NO  2 concentration as a function of Figure F5: Deterministic simulation results with CO and NO the vertical mixing height of the air pollutant and wind angle.  151  350.0  1.2  300.0 1.0  2  E 250.0  0)  3)  0.8  2  C 0  200.0 Ca  0  0.6 a) 0  150.0  0 0  o  0.4 100.0  0  \ 0.2  50.0  0. .  -  p..  0.0  f__I  6.0  4.0  2.0  0.0  c_I  ...  8.0  c_f  0.0 10.0  Mixing height (m) 90°  —•.——  .o•• 150° 210° ) 3 CO MV AQO (30 mg/rn NO, MV AQO (0.2 mgim3)  ——  — —.  2 concentration as a function of the Figure F6: Stochastic simulation results with CO and NO angle of either 90°, 150° or 210°. mean wind vertical pollutant mixing height and random 800.0 700.0  2.5 0)  600.0  2  2.0  2  500.0  0  C 0  1.5  400.0 C Cl)  0  g  300.0  1.0  8 0  z  200.0 0.5 100.0 -  __.—,,-.  0.0  0.0  10.0  20.0  30.0 40.0  my  0.0  50.0 60.0 70.0 80.0 90.0 100.0 110.0 120.0  Buffer distance (m) 0 ———f-—— —. —.—•-  ———  Q•  ——-+—— —  —  —  —  90° 120° 150° 180° 210° 240° 270° ) 3 CO MV AQO (30 mg/rn 2 MV AQO (0.2 mg/rn NO  2 concentration as a function of Figure F7: Deterministic simulation results with CO and NO the buffer distance the vessels maintained from the whale, and wind angle.  152  350.0  1.2  300.0 1.0 2  250.0  g 8 . 0  0)  2  C 0  200.0  CU  CU  C-) C 0 C-)  o C-)  0.6 U) C-) C 0  150.0  0.4° 0  \  100.0  z  0.2  50.0  -V ——-.  p0.0 120.0  0.0  100.0  80.0  60.0  40.0  20.0  0.0  Buffer distance (m) 90 150 210 ) 3 co MVAQO (30 mg/rn — —. NO, MV AQO (0.2 mgfm3) —.——  ....p —‘v—  2 concentration as a function of the Figure F8: Stochastic simulation results with CO and NO buffer distance and random mean wind angle of either 90°, 1500 or 2100. 180.0 0.6 160.0  — —A  0.5 ç 2  140,0 2  /  120.0  /  100.0 80.0  -  C) C  —  —A  °  / —  _.e _—L_ — /  ,  40.0  /  20.0  / ///  . .....<>.  _.  .  0.3 ..._  —  0  —  -D  —  -  —  -  ”  —‘-a—  ....  I.  20  ...  —.  40  ‘2U)  -a  0.2 0  z  T  0.1  • 120  o.o  ““  . . 80 60 Number of vessels  •  .  .  C 0  C) C 0  .  ./  F———-  0.0  .0)  c’,/.*  2 (U  /  o  o  0.4  ———v——— —.—c’—.——A—  ——-0——  — — — —  100  90 120 150 180 210 240 270 ) 3 co MV AQO (30 mg/rn ) 3 2 MV AQO (0.2 mg/rn NO  2 concentration as a function of Figure F9: Deterministic simulation results with CO and NO the number of vessels and the angle the wind came from. 153  120.0 0.4 100.0 0)  N  0)  80.0  E C 0  C 0  0  60.0  C  a)  a) C) C  8  C) C 0 0  40.0  N  0  0 C.)  z 20.0  0.0 100.0  80.0 60.0 Number of vessels —.-— 90° 0 150° —v— 210° ) 3 CO MV AQO (30 mg/rn ) 3 NO, MV AQO (0.2 mg/rn  20.0  0.0  40.0  —  120.0  —  2 concentration as a function of Figure FlO: Stochastic simulation results with CO and NO the number of vessels and random mean wind angle of either 90°, 1500 or 2 10°. 0.6 160.0 140.0  0.5 N  2  E 120.0  0)  0)  E  U.  c  100.0  0  0  0.3  80.0  C  —  C  a) C)  a)  / 60.0  0  0 C.)  0.2  / •/  40.0  N  \  .  /  C.) C 0  0 z  \ 0.1  / 20.0 . 0.0  -  -  0.0  10.0  20.0  40.0 50.0 60.0 70.0 80.0 Inter-vessel distance (m) 90° 120° 0 150° ——--——— —. 180° 210° ——*— 240° ——0—— 270° ——-4—— ) 3 CO MV AQO (30 mg/rn —— — ) 3 2 MV AQO (0.2 mg/rn NO 30.0  .  90.0  0.0 . 100.0 110.0  —  2 concentration as a function Figure F 11: Deterministic simulation results with CO and NO of the inter-vessel distance and the angle the wind came from.  154  180.0 0.6 160.0 \ \\  140.0  0.5  E \\  120.0  0.4 C  0.3  \\  80.0  C  C  8  C  .2  .Q ioo.o  60.0  S  ‘----..  0  0 Z  -,  40.0 0.1 20.0  V  .0  0... VV__0—  0.0  0.0  ---—  0.0  20.0  40.0  60.0  80.0  100.0  Inter-vessel distance (m) 900 •QV 1500  -.-  —v— 2100 — —.  ) 3 CO MV AQO (30 mg/rn ) 3 2 MV AQO (0.2 mg/rn NO  2 concentration as a function of Figure F12: Stochastic simulation results with CO and NO the inter-vessel distance and random mean wind angle of either 900, 150° or 210°.  155  Appendix G: Classes of Compounds in Diesel Exhaust (from Mauderly, 1992). Particulate Phase: • Elemental carbon • Heterocyclics, hydrocarbons (C14-C35), polycyclic aromatic hydrocarbons and derivatives (acids, alcohols, aldehydes, anhydrides, esters, ketones, nitriles, quinones, sulfonates, halogenated and nitrated compounds) • Inorganic sufates and nitrates • Metals Gas and Vapor Phases: • Acrolein • Ammonia • Carbon dioxide • Carbon monoxide • Benzene • 1,3-Butadiene • Formaldehyde • Formic Acid • Heterocyclics, hydrocarbons (Cl-C18), and derivatives (as listed above) • Hydrogen cyanide • Hydrogen sulfide • Methane • Methanol • Nitric and nitrous acids • Nitrogen oxides • Sulfur dioxide • Toluene • Water  156  Appendix H: Health Effects from Exposure to Air Pollutants in Exhaust.  Particulate Matter (PM) Particulate matter represents a broad class of chemically and physically diverse substances that form discrete particles that exist in the condensed phase (USEPA, 2004). PM is of great concern because it is easily inhaled, chemicals such as sulphates, nitrates, and heavy metals (e.g. chromium, manganese, mercury, and nickel) easily attach to the surface, and often the attached chemicals are known or suspected mutagens and carcinogens that persist in the environment (BCPHO, 2004; HEI, 1999; USEPA, 2004). Many of the attached chemicals (and other carcinogens associated with engine exhaust) are not directly toxic and only produce harmful effects when activated metabolically (HEI, 1988). Even short-term elevations in the ambient concentration of PM contributes to numerous harmful health effects such as: cough; lower respiratory symptoms; decrements in lung function; chronic bronchitis; asthma; chronic obstructive pulmonary disease; lung cancer; increases in cardio-respiratory mortality; impairment of lung clearance mechanisms; cough; labored breathing; chest tightness; wheezing; inflammation; cell proliferation; ischemic heart disease; heart failure; and metaplasia (cell transformation from a normal to an abnormal state) (Bates, 1994; HEI, 1999; Invernizzi et al., 2004; USEPA, 2004). Whenever 3 there is an observable short-term the ambient PM concentration increases by 10 mg m health burden, which can be calculated via human morbidity and mortality (Invernizzi et al., 2004). Acute or short-term exposure to diesel exhaust can result in acute irritation to the eyes, throat, and bronchioles, can cause neurophysiological symptoms such as lightheadedness and nausea, and can cause respiratory symptoms like cough and phlegm  157  (USEPA, 2004). Immunologic effects are also produced, such as the aggravation of allergenic responses to allergens, and asthma-like symptoms (USEPA, 2004). , and 25 10 and PM Land-based engines emit 95% of their PM in the fractions of PM marine vessels also emit small diameter PM mostly below 1 im (Quan et al., 2002). Several 5 at concentrations of . 2 studies have found a consistent relationship between inhalation of PM 3 and higher with hospitalizations from cardiac and respiratory diseases like asthma, 15 p.g m bronchitis, emphysema, and even death (BCPHO, 2004). Zhu et al. (2002) conducted a literature survey on the toxicity of PM, and found that PM with a diameter less than 100 nm (ultrafine) is more toxic than larger particles of the same chemical composition and concentration, and the risk of adverse health effects increases proportionately with exposure (BCPHO, 2004). However, a no-effects threshold for PM has never been achieved, thus even concentrations below air quality standards require precaution (BCPHO, 2004).  Nitrogen Oxides (NOt) Nitrogen oxides are byproducts of fuel combustion and are composed primarily of 2 is chemically reactive, water 2 (BCPHO, 2004). NO NO that reacts quickly to form NO ) and 3 soluble, corrosive gas, and when combined with water vapor forms nitric acid (HNO ), that can in turn form carcinogenic nitrosamines (BCPHO, 2004; RET, 2 nitrous acid (HNO  3 is reactive with other organic chemicals and produces 1988; Koenig, 2000). HNO secondary particles like ammonium nitrate, which bind to PM contributing to its toxicity 2 can also react with HCs in sunlight, producing 03 and other (BCPHO, 2004). NO photochemical byproducts that form smog (BCPHO, 2004). NO damages cells that line the  158  lungs, reducing lung function and intensifying health problems like asthma, bronchitis, coughing, and chest pain (BCPHO, 2004). , thus it can penetrate further into the lungs, where 2 NO is not as water-soluble as NO 2 into tissue but is not transported far due to its reaction with it diffuses quicker than NO oxyhemoglobin (Lippmann, 2000). NO has low direct toxicity, but reactions with other compounds can produce potent toxic oxidants (Lippmann, 2000). When NO enters the blood, it binds to hemoglobin to produce nitrosythemoglobin (NOHb) because hemoglobin has a higher affinity for NO than 02 (Lippmann, 2000). Enzymes quickly oxidize NOHb, and as long as the enzyme activity is maintained, potential toxicity from NO-related oxygen 3 transport effects are eliminated, at least with NO concentrations less than 12.3 mg m (Lippmann, 2000). 2 is inhaled up to 90% can be removed, with the majority taken up by the When NO lungs and the rest by the upper respiratory tract (Lippmann, 2000). Increased ventilation 2 to be delivered and absorbed in the alveolar region (Lippmann, 2000). causes more NO 2 is absorbed into the blood where it is likely converted to nitrite (Koenig 2000). Inhaled NO 2 have been related to increased mortality, but generally only Ambient concentrations of NO 3 are problematic (Lippmann, 2000). Animal studies concentrations greater than 1880 çig m 3 over 2 at concentrations of less than 1880 ig m have demonstrated that exposure to NO several weeks to months causes numerous effects to the lungs, spleen, liver and blood (WHO, 2000). In the lungs, the effects are both reversible and irreversible, but structural changes can occur in cell types with emphysema-like effects (WHO, 2000). Increased susceptibility to bacterial and viral lung infections occurs at exposure levels as low as 940 jig 2 (WHO, 2000). 3 of NO m  159  Carbon Monoxide (CO) Carbon monoxide is odorless, colorless, tasteless, nonirritating and is produced by the incomplete combustion of fossil fuels (USEPA, 2004). High levels of CO are recorded near roadways, and peak in city centers and at intersections during rush hour (BCPHO, 2004). However, the respiratory system is not considered the primary target for CO effects, as it mostly affects the cardiovascular system (Koenig, 2000). When CO is inhaled, it enters the bloodstream through the lungs and inhibits hemoglobin’s capacity to carry 02 to the tissues and organs due to the preferential binding of CO over 02, (hemoglobin’s affinity for CO is 240-250 times that of 02) (Lippmann, 2000; USEPA, 2004). This has a significant negative impact on human health because it interferes with 02 release at the tissues (Lippmann, 2000), which causes toxic effects in the blood and tissues, impairs organ functioning, and if exposure is long enough it results in death (USEPA, 2004). Because the heart and brain require more blood flow than other organs and tissues, they experience the hypoxic effects of CO much more rapidly (Lippmann, 2000). CO can also modify electron transport in nerve cells (Lippmann, 2000), thus exposure to CO can impair exercise capacity, visual perception, manual dexterity, learning functions, and the ability to perform complex tasks (USEPA, 2004). The human circulatory system takes 8-12 hours to eliminate CO due to its high hemoglobin affinity (Lippmann, 2000). Epidemiological studies have found a link between CO exposure and premature morbidity such as angina, congestive heart failure, and other cardiovascular diseases (USEPA, 2004). Even ambient CO exposure has been associated with increased hospital admissions due to cardiovascular issues, especially congestive heart failure (USEPA, 2004). Chronic exposure to CO can induce adaptations such as increases in the number of red blood  160  cells, blood volume, heart size, heart rate, stroke volume, and systolic blood pressure (Lippmann, 2000).  ) 2 Sulfur Dioxide (SO Sulfate emissions are approximately proportional to the quantity of sulfur in the fuel, , but 1-4% is oxidized to sulfuric acid in the exhaust 2 and most of the sulfur is oxidized to SO 2 can produce chronic bronchitis and potentially asthma (USEPA, 2002). Exposure to SO 2 is highly water-soluble, thus is easily removed by the upper (Bates, 1994). However, SO airways, which is the primary site of respiratory defence.  Hydrocarbons (HC) Volatile Organic Compounds (VOCs) consist mainly of HCs and other organic gasses, many of which are potentially toxic at ambient levels (i.e. benzene, 1 ,3-butadiene, acetaldehyde, and formaldehyde) (Bates et al., 2003). Aromatic HCs are the most toxic major class of compounds in fuel exhaust, with acute toxicity correlated to light aromatic HCs, and chronic effects correlated to four- and five-ring aromatic and heteroaromatic HCs (Geraci & St. Aubin, 1990). HCs evaporate slower in cooler waters than warmer waters, due to the inverse relationship between temperature and vapor pressure (Geraci & St. Aubin, 1990). HC vapors can irritate and damage the sensitive membranes of the eyes, mouth, and respiratory tract (Geraci & St. Aubin, 1990). HCs with high volatility (e.g. benzene and toluene) are easily inhaled, can displace oxygen in the lungs, are quickly transferred to the blood from the lungs, can accumulate in tissues like the brain and liver, which can lead to  161  neurological disorders and liver damage (Geraci & St. Aubin, 1990). Chemical pneumonitis can be caused by small amounts of insoluble HCs that easily penetrate deep into the bronchopulmonary tree, resulting in bronchospasm and inflammatory response (Le Tertre et al., 2002). Acute and chronic central nervous system and peripheral nervous system toxicity can arise from the systemic absorption of HCs (Le Tertre et al., 2002). Inhalation of HCs can also cause inflammation of mucous membranes, lung congestion, pneumonia, neurological damage, and liver disorders (Matkin & Saulitis, 1997). Therefore, the inhalation of HCs can result in direct mortality or indirect mortality and morbidity. Oil spill research has shown that volatile HCs can cause sudden death in cetaceans that inhale them while traveling quickly or when stressed because their breathing is rapid and explosive (Matkin & Saulitis, 1997). However, for death to occur the HCs must be present in 3 and greater) or the time of exposure needs to be significant high concentrations (100 mg m (Matkin & Saulitis, 1997). Exposure to chronic moderate concentrations of HCs could pose a health risk to killer whales; however, marine mammals have liver enzymes that can metabolize and excrete HCs, which limits their accumulation and potential harm (Geraci & St. Aubin, 1990).  Polycyclic aromatic hydrocarbons (PAHs) Polycyclic aromatic hydrocarbons (PAils) are derived from condensed benzene rings, and are formed via combustion processes (Godard et al., 2006). The USEPA has classified several PAR compounds as carcinogens, and probable human carcinogens based on animal studies (i.e. fish, amphibians, rats) and in human cell culture assays (USEPA, 2004). Many PAHs have carcinogenic or mutagenic potential via their metabolites, which affect the  162  reproductive system, developmental systems, immunological systems, the endocrine system, and can enhance plaque formation in the arteries (Frumkin & Thun, 2001; Godard et al., 2006; USEPA, 2004). The most widely studied PAll is Benzo[a]pyrene; it is the only PAll that has been used in inhalation experiments to test for carcinogenicity, and in these inhalation experiments it produced lung tumors in hamsters (WHO, 2000). PAHs present in diesel soot strongly bind to the surface of PM, which greatly slows their clearance rate (IPCS, 1996). The human lung can clear approximately 50% of the PAHs on diesel PM in one day, but the remaining PAHs have retention half times of 18-36 days (IPCS, 1996). Pulmonary macrophages can metabolize some PAHs by oxidation (IPCS, 1996). PAHs have many routes into the marine environment: oil/fuel spills, ship discharge, oil seepage, road runoff, industrial effluent, forest fires, and atmospheric deposition from the incomplete combustion of fossil fuels (SETAC, 1996). After entering the marine environment PAHs are weathered by physical, chemical, and biological processes, yet the highest PAH concentrations within the water column are right at the surface of the water (SETAC, 1996). PAHs and PAH-DNA adducts have been measured in brain and liver tissue of belugas in the St. Lawrence River estuary (Godard et al., 2006). PAHs taken up by cetaceans are biotransformed/metabolized by oxidation and conjugation to form more poiar and water-soluble metabolites that are either retained or excreted (Law & Whinnett, 1992; SETAC, 1996). PAHs are stored short-term in the liver, and in the long-term they bioaccumulate in muscle tissue and this represents the portion retained (Law & Whinnett, 1992; SETAC, 1996). However, in harbour porpoises (Phocoenaphocoena), accumulation of PAH seems to be low, and most likely primarily comes from their food (Law & Whinnett,  163  1992). Fish in areas polluted with PAHs usually have low concentrations of PAH, as they can quickly convert them to metabolites such as dihydrodiols and phenols (Law & Whinnett, 1992).  Benzene Benzene is a volatile aromatic HC, and makes up one to two percent of the exhaust emitted from gasoline and diesel engines (USEPA, 2004). Benzene is a known human carcinogen and causes leukemia in humans via all routes of exposure (USEPA, 2004). Benzene also causes chromosomal changes in human and animal cells, and blood disorders such as aplastic anemia (USEPA, 2004). However, benzene is volatile and short-lived, thus it is likely involved more with acute toxicity rather than chronic (Geraci & St. Aubin, 1990).  Aldehydes Aldehydes make up an important fraction of the gas phase of exhaust emissions, they are probable human carcinogens, and produce non-cancer health effects (USEPA, 2002). Formaldehyde in exhaust is a result of the incomplete combustion of gasoline and diesel, it is the most abundant aldehyde present in engine exhaust, making up over 10% of the total HC emissions (USEPA, 2004), and makes up 65-80% of the aldehyde emissions in diesel exhaust (USEPA, 2002). Formaldehyde is very reactive (even with itself), but is non-toxic in small quantities as it is a by-product of metabolism (MEl, 1988). Methanol added to gasoline significantly increases formaldehyde emissions, while ethanol increases acetaldehye emissions (Lippmann, 2000).  164  Formaldehyde is very water-soluble, and when inhaled essentially 100% of it is absorbed/deposited in the upper and lower respiratory tract (HEI, 1988; Lippmann, 2000). The long-term inhalation of fonnaldehyde can produce tumors in the sinuses and nasal cavity of humans (USEPA, 2004). Although formaldehyde is a probable human carcinogen since it causes cancer in animals, there is limited evidence of carcinogenicity in humans (USEPA, 2004). Studies have also found that formaldehyde causes mutagenic activity in cell cultures, it is toxic to the kidney at moderately high doses, it has harmful neurological effects, it can reduce pulmonary function, and can initiate skin sensitization (Lippmann, 2000; USEPA, 2004). Other aldehydes can be genotoxic and produce tumors in vivo, and can also produce similar respiratory effects as formaldehyde (Lippmann, 2000).  1,3-Butadiene 1,3-Butadiene is a colorless gas found in small quantities in gasoline vapor and in engine exhaust (ATSDR, 1993). Even though it breaks down quickly in air, especially in the 3 presence of sunlight, the average concentration of 1 ,3-butadiene in urban air is 0.67 tg m (ATSDR, 1993). Exposure to 1,3-butadiene usually occurs from breathing contaminated air, and low levels are not expected to result in adverse health effects (ATSDR, 1993). However, when humans inhale 1 ,3-butadiene at high concentrations for short periods of time it results in central nervous system damage, blurred vision, nausea, fatigue, headache, decreased blood pressure, decreased pulse rate, and unconsciousness (ATSDR, 1993). Studies with experimental animals have found that inhaling 1 ,3-butadiene can increase the number of birth defects, cause kidney and liver disease, produce tumors, damage the lungs, and even cause death of some individuals (ATSDR, 1993). In the United States, 1 ,3-butadiene has been  165  listed as a probable carcinogen based on animal studies, and the occupational exposure limit 3 (ATSDR, 1993). is 2,212mg m  Ozone (03) Ozone occurs naturally in the environment at low levels; however, when engines emit 2 and VOCs, they can react photochemically to produce 03 when air pollutants such as NO exposed to sunlight (BCPHO, 2004). Ozone is typically found several km downwind from the source of primary pollutants, because 03 in ambient air tends to lag the emissions of the primary pollutants required for its formation (Koenig, 2000). Out of all the oxidizing air pollutants, 03 is the most toxic due to its oxidative properties (HEI, 1988). 03 is a highly reactive gas of moderate solubility, thus it can penetrate into the tracheobronchial tree and will react with the mucous layer of the small bronchioles, damaging the tissue underneath (HEI, 1988). 03 is an intense irritant and damages lung epithelial cells, which reduces lung function and aggravates other health problems like: asthma, bronchitis, coughing, pneumonia, and chest pain (BCPHO, 2004). 3 during an 8-hour average (BCPHO, 2004). When young The CWS for 03 is 0.13 mg m 3 of 03 for 6-hours, it adversely affects normal exercising humans are exposed to 0.16 mg m their lung functioning and causes inflammation in the lung (Bates, 1994).  Other Compounds in Exhaust Diesel and gasoline engines emit numerous compounds that are known or suspected to be human or animal carcinogens, and/or have other non-carcinogenic effects when inhaled (USEPA, 2004). Examples are: acetaldehyde, acrolein, dioxin, furans, polychiorinated  166  biphenyls (PCBs), and polycyclic organic matter (USEPA, 2004). While exposure to these compounds usually occurs via food consumption, inhalation and absorption through the skin can also occur (BCPHO, 2004). Some dioxins have been classified as carcinogenic, and some of the other effects of exposure are: skin lesions, liver enzyme changes, damage to the immune system, and damage to the reproductive system (BCPHO, 2004).  167  THE UMVERSITY OF BRITISH COLUMBIA  42  ANIMAL CARE CERTIFICATE Application Number: A06-. 1548 Investigator or Course Director: Lance Barrett-Lennard Department: Zoology Animals:  Whales Southern resident killer whale (Orcinus orca) 85  Start Date:  May 1, 2007  Approval Date:  March 20, 2007  Funding Sources: Funding Agency: Funding Title:  Vancouver Aquarium Marine Science Centre BC Wild Killer Whale Adoption Program  Unfunded title:  Assessing the health implications of marine engine exhaust gas on southern resident killer whales (Orcinus orca)  The Animal Care Committee has examined and approved the use of animals for the above experimental project. This certificate is valid for one year from the above start or approval date (whichever is later) provided there is no change in the experimental procedures. Annual review is required by the CCAC and some granting agencies.  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0067192/manifest

Comment

Related Items