Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Forest residues to energy : local air quality, health risks and greenhouse gas emissions Petrov, Olga 2018

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

Item Metadata

Download

Media
24-ubc_2018_may_petrov_olga.pdf [ 4.54MB ]
Metadata
JSON: 24-1.0365819.json
JSON-LD: 24-1.0365819-ld.json
RDF/XML (Pretty): 24-1.0365819-rdf.xml
RDF/JSON: 24-1.0365819-rdf.json
Turtle: 24-1.0365819-turtle.txt
N-Triples: 24-1.0365819-rdf-ntriples.txt
Original Record: 24-1.0365819-source.json
Full Text
24-1.0365819-fulltext.txt
Citation
24-1.0365819.ris

Full Text

  FOREST RESIDUES TO ENERGY:  LOCAL AIR QUALITY, HEALTH RISKS AND GREENHOUSE GAS EMISSIONS  by  Olga Petrov   A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Chemical and Biological Engineering)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  April 2018   © Olga Petrov, 2018  ii  Abstract Local impact assessment of biomass-based district energy systems (DES) is still in its infancy. There has been a lack of appropriate assessment methods for parameters with broad variability on local scale, and lack of DES impact assessments. This study investigates how would: 1) the inclusion of site-specific terrain, land use and microclimatic characteristics, variable population density and breathing rates affect accuracy of assessments on local air quality and health;  2) an incremental increase of PM2.5, NOx and CO concentrations from DES contribute to ambient air quality and population exposure, 3) life-cycle GHG emissions from DES contribute to global warming, and 4) the introduction of biomass affect economics of DES compared to the fossil fuel-based DES.    Utilizing dispersion modeling the study established an assessment approach which confirmed the need for inclusion of population dynamics, site-specific microclimatic characteristics, and diurnal circulation patterns. Otherwise, health risks could potentially be underestimated by more than 20%. Applying this approach on a small-scale biomass gasification plant (BRDF), the study concluded that the health impact was the highest for NO2 (677 DALY) when all energy was produced by biomass, and for PM2.5 (64 DALY) if all energy was produced by natural gas. Complete replacement of Power House (PH) by one biomass plant can result in almost 28% higher impact compared to 513 DALY when both BRDF and PH are operational. NO2 emissions from the BRDF exceeded the air quality objectives (BCAQO) in all seasons except during summer. Although overall incremental contribution of PM2.5 is at least one order of magnitude lower than BCAQO, the maximum PM2.5 emissions from the PH could adversely add to the already high background concentrations.  iii  Meeting energy demand solely by an expanded full-scale BRDF from locally supplied biomass reduces GHG annually to 3.81E+06 kg CO2eq from 7.08E+07 kg CO2eq when energy was produced solely by the current PH. An introduction of biomass increased total costs by $19 M compared to existing PH, but saved $8.4 M in carbon tax over plants’ lifetime.  $3.3 M of societal damages could be avoided over plants’ lifetime in case of combined use of natural gas and biomass.   iv  Lay Summary This research improves current methods for assessing impacts of biomass-based district heating systems. It is confirmed that introducing site specific characteristics such as population dynamics, local meteorological conditions along with outdoor pollutant concentrations could more accurately evaluate local air quality and population health risks. The study further evaluates impacts of a biomass plant located at the University of British Columbia, Vancouver campus which is operational since 2012 and supplies heat to almost 20% of campus heat demand. The study found that the choice of fuel (wood versus natural gas) will have impacts on a global scale in terms of reduced impacts on global warming, whereas the choice of plant location balanced with techno-economic benefits should be a primary consideration for minimizing local impacts (population exposure and local air quality) regardless the fuel type. The development of biomass plants could be costly but savings exist in carbon taxes and societal damages.            v  Preface The topic of this doctoral dissertation was deliberated during the discussion with my academic adviser Dr. Xiaotao Bi and The Bridge Program Director Dr. Michael Brauer.  The entire research reported here was conducted by the author, Olga Petrov and included: developing thesis objectives and research questions, conducting systematic literature review, designing the research program, gathering and evaluating data, developing and applying a novel impact assessment methodology for community-based district energy systems in consort with developing and running airborne pollutant dispersion, population exposure and life cycle modeling scenarios, and analyzing and interpreting the results. The following parties were regularly consulted: UBC’s Engineers at the Bioenergy Research and Demonstration Facility (BRDF), Power House (PH) and Campus + Community Planning; Nexterra Energy Corp. and Cloverdale Fuel Ltd. Parts of this research were published:   A version of chapters 1 and 2 have been published. Petrov, O. (2012). Forest Residues to Energy: Is this a pathway towards healthier communities? National Collaborating Centre for Environmental Health. Evidence Review. Available from: http://www.ncceh.ca/sites/default/files/Forest_%20Residues_to_Energy_Mar_2012.pdf. I conducted systematic literature review and wrote the whole manuscript. Dr. Bi and Dr. Brauer provided invaluable comments and edits.  A version of chapters 2 and 3 has been published. Petrov, O., Bi, X., & Lau, A. (2015). Impact assessment of biomass-based district heating systems in densely populated communities. Part I: Dynamic intake fraction methodology. Atmospheric Environment, 115, 70–78. https://doi.org/10.1016/j.atmosenv.2015.05.036 . I conducted all data vi  collection, modeling and wrote the manuscript. Dr. Bi and Dr. Lau provided invaluable comments and edits.  A version of chapter 4 has been published. Petrov, O., Bi, X., & Lau, A. (2017). Impact assessment of biomass-based district heating systems in densely populated communities. Part II: Would the replacement of fossil fuels improve ambient air quality and human health? Atmospheric Environment, 161, 191–199. https://doi.org/10.1016/j.atmosenv.2017.05.001. I conducted all data collection, modeling and wrote the whole manuscript. Dr. Bi and Dr. Lau provided invaluable comments and edits.   A version of chapters 5 and 6 is under preparation for publishing:  Petrov, O.; Xiaotao Bi, Anthony Lau. (2017). Global impacts assessment and economic analysis of woody biomass as an alternative to fossil fuels in district heating applications. I wrote the whole manuscript; conducted literature search on carbon neutrality; collected and processed consumption data for natural gas, fuel oil and biomass used in the UBC Power House (PH) and Bioenergy Research and Demonstration Facility (BRDF); estimated emissions; performed LCA modeling, and economic analysis. Data on biomass supply are based on the work prepared for the UBC SEEDS project, published at: https://sustain.ubc.ca/courses-teaching/seeds-program/seeds-sustainability-library. Dr. Bi and Dr. Lau provided invaluable comments and edits.  vii  Table of Contents Abstract ................................................................................................................................... ii Lay Summary ......................................................................................................................... iv Preface ......................................................................................................................................v Table of Contents .................................................................................................................. vii List of Tables .......................................................................................................................... xi List of Figures ...................................................................................................................... xiii List of Acronyms ....................................................................................................................xv List of Symbols and Selected Units .................................................................................... xvi Acknowledgements ............................................................................................................. xvii Dedication ........................................................................................................................... xviii Chapter 1: Introduction ..........................................................................................................1 1.1 Background ....................................................................................................................1 1.1.1. Global drivers and perspectives on energy production and utilization ...............1 1.1.2. Environmental concerns and alternatives to fossil fuels for energy production..4 1.1.3. Availability of biomass resources and district energy systems in British Columbia .............................................................................................................6 1.1.4. Public perception and acceptance of biomass systems ........................................9 1.2 Thesis objectives and research questions .....................................................................13 1.3 Case study.....................................................................................................................14 1.4 Thesis structure.............................................................................................................16 Chapter 2: Literature Review , .............................................................................................20 2.1 Biomass classification and characterization .................................................................20 2.1.1 Chemical composition ........................................................................................21 2.1.2 Heating value ......................................................................................................21 2.1.3 Moisture content .................................................................................................22 2.1.4 Ash content .........................................................................................................22 2.2 Characterization and control of emissions from biomass-based energy systems ........23 2.3 Impacts on ambient air quality and climate ..................................................................24 2.4 Population exposure and health risks ...........................................................................27 viii  2.5 Carbon footprint and large scale impacts of district energy systems ...........................36 2.5.1 Bioenergy and carbon neutrality discussions .....................................................37 2.5.2 GHG emission estimates ....................................................................................38 Chapter 3: Integrated impact assessment approach to evaluate community-based district heating systems .........................................................................................................40 3.1 Introduction ..................................................................................................................40 3.2 Methods ........................................................................................................................41 3.3 Local air quality assessment methodology ...................................................................42 3.3.1. Microclimatic conditions and diurnal circulation patterns ................................42 3.3.2. Dispersion modeling: CALPUFF modeling system ..........................................45 3.3.2.1. Model input data ..................................................................................... 48 3.3.2.2. Model output data: ambient pollutant concentrations ............................ 51 3.3.3. Ambient air quality regulation and background pollutant levels ......................52 3.4 Population exposure and health risk assessment methodology ....................................53 3.4.1 Dynamic intake fraction (iF) ..............................................................................53  Input values used for iF calculations ...................................................... 54  Scenarios and resulting iF values ........................................................... 57 3.4.2 Health-related Impact Score (IS) ........................................................................65 3.5 Environmental footprint methodology .........................................................................67 3.5.1. Goal and scope definition ..................................................................................68 3.5.2. Life cycle inventory (LCI) analysis ...................................................................69 3.5.3. Life cycle impact assessment (LCIA) ...............................................................69 3.6 Conclusions ..................................................................................................................71 Chapter 4: Impact assessment of the UBC district heating system on local air quality and associated health risks  ...................................................................................................73 4.1 Introduction ..................................................................................................................73 4.2 District heating at UBC Point Gray campus ................................................................74 4.2.1. Thermal energy demand and supply profile ......................................................74 4.2.2. Biomass supply requirements for fossil fuel replacement .................................75 4.3 Scenarios for evaluating options for district heating at UBC .......................................76 ix  4.4 Emission characteristics and estimates.........................................................................77 4.5 Local air quality assessment .........................................................................................80 4.6 Health impact assessment .............................................................................................80 4.7 Discussion ....................................................................................................................81 4.8 Model performance evaluation .....................................................................................88 4.8.1. Graphical analysis..............................................................................................89 4.9 Conclusions ..................................................................................................................94 Chapter 5: Global impacts of the UBC district heating system ........................................96 5.1 Introduction ..................................................................................................................96 5.2 Quantifying global impacts of UBC district heating ....................................................97 5.2.1. Feedstock sourcing and characterization at the UBC Point Grey campus ........97 5.2.2. Goal and Scope ..................................................................................................99 5.2.3. Life cycle inventory .........................................................................................100 5.3 Global impact assessment of UBC district heating options and discussion ...............107 5.3.1. Impact assessment approach 1 .........................................................................107 5.3.2. Impact assessment approach 2 .........................................................................114 5.4 Conclusion ..................................................................................................................117 Chapter 6: Economic valuation of district heating options .............................................119 6.1 Introduction ................................................................................................................119 6.2 A summary of reported UBC district heating costs and GHG emissions ..................121 6.3 Economic valuation methodology ..............................................................................123 6.3.1. Assessment of costs and benefits associated with the development, operation and maintenance of biomass-based district heating at UBC ...........................124 6.4 Results and discussion ................................................................................................126 6.4.1 Addressing uncertainty .....................................................................................130 6.4.2 Trade-offs associated with the selection of district heating options ................131 6.5 Conclusions ................................................................................................................137 Chapter 7: Conclusions and future research directions ..................................................140 7.1 Conclusions and significance of the research.............................................................140 7.2 Strengths and limitations of the research ...................................................................145 x  7.3 Future research directions ..........................................................................................147 Bibliography .........................................................................................................................149  Literature review ...............................................................................................181  Literature search methodology .........................................................................181  CO2 neutrality overview ...................................................................................183  UBC Fuel characteristics and consumption, and energy calculations ..............186  Conversion of units used in fuel calculations ...................................................186  Fuel consumption and steam produced ............................................................187  Emission estimates used in modeling scenarios ................................................190  Results of ambient air quality and health risks assessment ..............................196  Global impacts data ...........................................................................................197  Emission factors for energy products ...............................................................197  Annual emissions over life cycle stages ...........................................................201  Meeting CAP2020 GHG reduction goals ..........................................................205    xi  List of Tables Table 1.1 Examples of biomass DES projects in British Columbia. ........................................ 9 Table 2.1 Summary of iF evaluation approaches based on the reviewed literature. .............. 30 Table 3.1 Surface and upper-air weather stations considered in the study. ............................ 49 Table 3.2 BRDF Source parameters. ...................................................................................... 51 Table 3.3 Ground-level PM2.5 concentrations, UBC campus, September 2012. .................... 52 Table 3.4 Summary of provincial Air Quality Objectives (AQO) and Canadian Ambient Air Quality Standards (CAAQS) for selected contaminants. ....................................... 53 Table 3.5 UBC Campus population distribution as a function of diurnal dynamics. ............. 57 Table 3.6 Modeling scenarios and calculated iF and IS. ........................................................ 65 Table 4.1 Summary of operational scenarios used in the DH impact assessment. ................. 76 Table 4.2 Estimated emission factors and annual emissions from biomass gasification (BRDF) and natural gas/oil combustion (PH). ....................................................... 78 Table 4.3 Measured and modeled PM2.5 data and Totem station meteorological parameters for July 17, 2012. ................................................................................................... 91 Table 5.1  Greenhouse gas intensity [g GHG/kWh electricity generated] in BC. ................ 102 Table 5.2  Transportation and wood processing data ........................................................... 105 Table 5.3  Wood chips characteristics. ................................................................................. 106 Table 5.4  Mid-point impacts for annual energy output of 1,011 TJ at UBC Point Grey campus.................................................................................................................. 115 Table 6.1 Economic parameters. .......................................................................................... 126 Table 6.2 Summary of calculated parameters [in $2012]. .................................................... 127 Table 6.3. Summary of externalities for district heating options at UBC ............................ 133 Table 6.4 Summary of key findings on local and global impacts for UBC district heating options. ................................................................................................................. 136 Appendices: Table A.2-1 Summary of findings on biomass CO2 neutrality based on the reviewed literature. .............................................................................................................. 183 Table B.2-1 Natural gas and oil consumption (energy input) and steam produced (energy output) at PH and BRDF. ..................................................................................... 187 xii  Table B.2-2 Wood requirements for 1,011TJ energy output. ............................................... 188 Table B.2-3 Seasonal distribution of energy demand of 1,011 TJ for 2012-2013. .............. 189 Table C-1 Scenario 1 Base case: Daytime and nighttime pollutant emissions from PH per month 2012-2013. ................................................................................................ 190 Table C-2 Scenario 1 Base case: Daytime and nighttime pollutant emissions from BRDF per month 2012-2013. ................................................................................................ 191 Table C-3 Scenario 1 Base case: Resulting emissions daytime and nighttime pollutant emissions from PH and BRDF per month 2012-2013. ........................................ 192 Table C-4 Scenario 2: Daytime and nighttime pollutant emissions per month 2012-2013 if only PH is operational. ......................................................................................... 193 Table C-5  Scenario 3 Daytime and nighttime pollutant emissions per month 2012-2013 if only BRDF is operational. ................................................................................... 194 Table C-6 Scenario 5 Daytime and nighttime pollutant emissions per month 2009-2010 when only PH was operational. ..................................................................................... 195 Table D-1 Summary of ambient air quality, iF and IS for five district heating operational scenarios at UBC. ................................................................................................. 196 Table E.1-1 Emission factors for natural gas........................................................................ 197 Table E.1-2 Emission factors for heavy fuel oil. .................................................................. 198 Table E.1-3 Emission factors for middle distillates.............................................................. 199 Table E.1-4 Emission factors for middle distillates for HDV operation. ............................. 200 Table E.2-1 Annual emission by process and transport stages for Scenario 1: NG, fuel oil and biomass. ................................................................................................................ 201 Table E.2-2 Annual emission by process for Scenario 2: Natural gas only. ........................ 202 Table E.2-3 Annual emission by process and transport stages for Scenario 3: Biomass only. .............................................................................................................................. 203 Table E.2-4 Annual emission by process and transport stages for Scenario 3: Biomass only, changed transportation distance. .......................................................................... 204  xiii  List of Figures Figure 1.1 Total annual anthropogenic GHG emissions by gases [GtCO2-eq/yr] for the period 1970 to 2010. ............................................................................................................ 2 Figure 1.2 World energy consumption from 1990 to 2040 projections in quadrillion Btu. ..... 3 Figure 1.3  UBC campus buildings, BRDF and PH emission sources. .................................. 16 Figure 2.1  Atmospheric species on spatial and temporal scales. ........................................... 26 Figure 3.1 Wind patterns for day and night periods at selected Metro Vancouver stations. .. 43 Figure 3.2 Wind rose for daytime (left) and nighttime (right), September 2012, UBC Totem weather station. ...................................................................................................... 44 Figure 3.3 Nested grid receptors, red rectangular depicts an area with removed receptors due to absence of population......................................................................................... 48 Figure 3.4  Scenario 2: iF for each building for September 2012. ......................................... 59 Figure 3.5 Scenario 3: iF for each building with actual occupancy, September 2012. .......... 60 Figure 3.6 Scenario 4: a) daytime iF and b) nighttime iF for each building with actual occupancy for September 2012. ............................................................................. 61 Figure 3.7 Scenario 4: daytime (upper graph) and nighttime (bottom graph) variations of iF for September 2012. ............................................................................................... 62 Figure 3.8. Scenario 5:  UBC campus iF for September 2012 distinguishing day vs night periods with spatial and temporal dynamics, and varying BR. .............................. 63 Figure 3.9 LCA framework. ................................................................................................... 68 Figure 3.10  Scheme of the impact categories dealt with in ILCD Handbook on Life Cycle Impact Assessment at midpoint and at endpoint. ................................................... 71 Figure 4.1  Wind circulation at 10 m altitude and projected PM2.5 concentrations for June 4, 2012 at 1 am (nighttime) for Scenario 3 (a), Scenario 2 (b) and Scenario 1(c), and at 1 pm (daytime) for Scenario 3 (d), Scenario 2 (e) and Scenario 1(f). Arrows present wind fields obtained by CALMET. ........................................................... 83 Figure 4.2  (a) Daytime and (b) nighttime average concentrations per pollutant and modeling scenario. ................................................................................................................. 84 Figure 4.3 Graphical comparison of ambient measured and ambient modeled PM2.5 concentrations for July 17, 2012. ........................................................................... 89 xiv  Figure 5.1 Trends in GHG emissions in BC 1990 – 2014. ..................................................... 96 Figure 5.2 Process stages and transportation segments considered in evaluating global impacts of a) biomass and b) fossil fuels. ............................................................ 100 Figure 5.3 Wood chips at BRDF: a) storage bin b) sizing c) oversized for oversized wood chips. .................................................................................................................... 106 Figure 5.4 Scenario 1: Annual GHG emissions [kgCO2eq] per life cycle stage for natural gas, fuel oil and biomass. ............................................................................................ 109 Figure 5.5 Scenario1: Pollutant emission contributions from different life cycle stage ...... 110 Figure 5.6 Scenario 2: Annual GHG emissions [kgCO2eq] per life cycle stage for natural gas. .............................................................................................................................. 110 Figure 5.7  Scenario 2: Pollutant emission contributions by life cycle stage. ...................... 111 Figure 5.8 Scenario 3: Annual GHG emissions [kgCO2eq] per life cycle stage for biomass.112 Figure 5.9 Scenario 3: Pollutant emissions contributions over different life cycle stages. .. 113 Figure 5.10 Scenario 3: Annual GHG emissions [kgCO2eq] per life cycle stage for biomass with increased transportation distance. ................................................................ 114 Figure 6.1 Energy demand for the UBC campus in 2012-2103. .......................................... 122 Figure 6.2  Cost breakdown for two DH options at UBC. ................................................... 129 Figure 6.3 External costs for option A (natural gas only) and option B (natural gas and biomass) for the period of plants’ life time. ......................................................... 135 Figure F 1 External costs for option A (natural gas only), option B (natural gas and biomass) and potential BRDF expansion, over plants’ lifetime. CO2 costs are excluded. .. 207          xv  List of Acronyms ADES Academic DES AQO Air Quality Objectives (provincial – BC) BC MoE British Columbia Ministry of Environment BRDF Bioenergy Research and Demonstration Facility at UBC campus CAAQS Canadian Ambient Air Quality Standard CALPUFF CALifornia PUFF Model CEC Campus Energy Centre CHP Combined heat and power CH4 Methane CO Carbon monoxide CO2 Carbon dioxide DES District energy system DH District heating EIA Environmental Impact Assessment EPA Environmental Protection Agency  ESP Electrostatic precipitator FU Functional unit (LCA) GIS Geographic Information System GWP Global Warming Potential IPCC Intergovernmental Panel on Climate Change LCA Life Cycle Assessment  LCI Life Cycle Inventory LCIA Life Cycle Impact Assessment LFV Lower Fraser Valley MM5 Mesoscale Meteorological Model, Version 5 MV Metro Vancouver NG Natural gas NMVOC Non-methane volatile organic compounds NO Nitric oxide NO2 Nitrogen dioxide NOx Nitrogen oxides PH Power House (at UBC) PM Particulate matter PM2.5 Particulate matter with a diameter less than 2.5µm PM10 Particulate matter with a diameter less than 10µm PV Present Value (used for economic assessment) RF Radiative Forcing SOx Sulphur oxides UBC University of British Columbia VOC Volatile organic compounds xvi  List of Symbols and Selected Units  Symbol Unit Definition BR m3/person/day Breathing rate C µg/m3 Pollutant ambient concentration at a receptor CO2eq kg Unit for GHG, calculated by multiplying emissions by GWP EFHeatlh DALY/kg Human health toxicological effect factor EFNG g/GJ Emission factor for natural gas combustion (converted from kg/mmBtu) EFOIL g/GJ Emission factor for oil #2 combustion EFWG g/GJ Emission factor for wood gasification HEF m Effective stack height HHV MJ/kg Higher Heating Value iF ppm Impact fraction IS DALY Health-related Impact Score m g Mass of a pollutant emitted from a source MCW % Moisture content of wood, wet basis MCD % Moisture content of wood, dry basis Ϭx m Standard deviation of Gaussian distribution in the downwind direction Ϭy m Standard deviation of Gaussian distribution in the cross-wind direction Ϭz m Standard deviation of Gaussian distribution in the vertical direction Q g/sec Pollutant emission rate      Unit Definition Btu The British thermal unit equal to 1,055 joules DALY Disability-adjusted life years EJ Exajoule, equal to 1018 joules GJ Gigajoule, equal to 109 joules or 103 MJ KLBS Kilopounds KSCF Kilo Standard Cubic Feet (at 21ºC and 101.325 kPa) kWh Kilowatt hour    lb Pound MMBtu 106 Btu , also known as mmBtu MWh Megawatt hour µg Microgram,  equal to 10-6 g ODMT Oven dry or Bone dry (BD)  metric tonne – mass of wood after all moisture has been evaporated ppm Parts per million (v/v),  equal to 10-6 t Tonne, metric, equal to 1,000 kg TJ Terajoule,  equal to  1012 joules xvii  Acknowledgements  Doctoral studies at UBC were my much wanted and much appreciated journey of personal and professional growth. For this achievement I offer my enduring gratitude and admiration to my supervisor Dr. Xiaotao Bi, for sharing his expertise, his guidance, patience and trust over the years. I sincerely thank Dr. Anthony Lau and Dr. Taraneh Sowlati for constructive comments, suggestions and encouragement throughout my studies and research. I appreciate the time and insightful questions and comments from my examining committee: Dr. Zhongchao Tan, Dr. Madjid Mohseni, Dr. Kasun Hewage and Dr. James Brander.  This research study would not have been possible without the financial support from The Bridge/CIHR Doctoral Fellowship Program at UBC and The Engineering Professional Development Fund at British Columbia Institute of Technology, for which I am thankful.   Data and consultation provided by Jeff Giffin, Joshua Wauthy and UBC Campus + Community Planning, Nexterra Systems Corp., Cloverdale Fuels Inc., Metro Vancouver and Lakes Environmental were crucial for my research and were much appreciated.  A special thank you I owe to Bridge fellows: Siduo Zhang, Alicia LaValle, Ther Aung, and Jackie Yip, and BBRG (Biomass and Bioenergy Research Group)  members for cooperation but above all for their friendship, laughs and perseverance which kept me strong during the most difficult and challenging periods of my life.  To my family and friends whose love and support inspired me to advance, I promise undivided attention in years to come. xviii  Dedication Dedicated with love and pride to my husband Aleksandar Petrov, PhD (Mech. Eng.), whose achievements and passion for biomass research were interrupted too soon by illness.     1  Chapter 1: Introduction 1.1 Background 1.1.1. Global drivers and perspectives on energy production and utilization  Climate change has been a focus of environmental research for decades now.  Independent analyses conducted by thousands of scientists across the world undoubtedly confirmed the exceptional changes of the Earth’s climate system since 1950s observed as warming of the atmosphere, sea level rise and diminished amounts of snow and ice (IPCC, 2015).  Concentrations of greenhouse gases (GHG), especially those of carbon dioxide (CO2), methane (CH4) and nitrous oxide (N2O) mainly generated by human activities, are rapidly increasing in the atmosphere. Between 2000 and 2010 these emissions were estimated to be higher than ever and, along with other anthropogenic factors, are claimed to be extremely likely the dominant cause of atmospheric warming since the second half of the 20th century (IPCC, 2015). There is evidence that anthropogenic (human made) pollutants contribute more to overall atmospheric content of gases and particles than it would exist or change at a certain rate naturally. As presented in Figure 1.1 from the latest IPCC report (AR5), anthropogenic emissions of greenhouse gases (GHG), especially those of carbon dioxide (CO2) from industrial practices and burning fossil fuels, continue to increase, reaching 49 ± 4.5 GtCO2eq1 in 2010. It is estimated that 47% of increased GHG emissions between 2000 and 2010 originates from the energy sector while industry and transportation sectors contributed with 30% and 11% respectively. In addition to recognizing CO2 as a major contributor (76% of total GHG in 2010), CH4 contributed 16%,                                                  1 CO2eq – Carbon dioxide equivalent emissions of CO2, CH4, N2O and fluorinated gases based the 100-year Global Warming Potentials (GWP), using IPCC Second Assessment Report (SAR).  2  N2O contributed around 6.0%, and 2.0% came from fluorinated gases (IPCC, 2015). It is determined that steady population growth and steep economic growth based intensively on coal combustion present the major cause of CO2 emissions. Therefore, reducing carbon intensity of the world’s energy supply presents challenges for years to come. Figure 1.1 Total annual anthropogenic GHG emissions by gases [GtCO2-eq/yr] for the period 1970 to 2010.      Adopted from: (IPCC, 2015), Figure 1.6, pg. 46).2  The U.S. Energy Information Administration (EIA) projected increase of energy use by 48% from 2012 to 2040 (Figure 1.2) by using known demographic trends and policies which were in place at the time of their analysis (EIA, 2016). Global energy consumption is projected to be especially pronounced for non-OECD3 countries such as China and India where rapid economic growth requires extensive energy use. Such countries are projected to increase energy demand by                                                  2 FOLU -  Forestry and Other Land Use; F-gases - fluorinated gases covered under the Kyoto Protocol;    CH4 - methane; N2O - nitrous oxide. 3 OECD - Organization for Economic Cooperation and Development. 3  71% by 2040 from 2012 levels, whereas more mature and stable OECD economies are projected to have 18% of increase in energy demand for the same time frame.   Figure 1.2 World energy consumption from 1990 to 2040 projections in quadrillion Btu.4 Source: Based on EIA (2016).  Some of the parameters that will influence the type of energy sources used include energy security and energy prices, as well as impacts on the environment.  Since fossil fuels used for energy production are considered as the main source of GHG emissions, their replacement with cleaner and renewable energy sources are a world-wide policy approach.  Based on the EIA report (EIA, 2016), renewable energies lead the global energy demand with an annual average growth rate of 2.6%, followed by an increase of 2.3% in nuclear energy use and 1.9% increase in natural gas, a least carbon intensive fossil fuel (EIA, 2016).  Canada with its vast non-renewable and renewable resources has a challenge but also the                                                  4 Quadrillion British thermal units (Btu) = 10E+15 BTU which is equal to 1.05587E+19 J. 4  opportunity of selecting cost effective and environment-friendly energy options. Natural gas is a fossil fuel convenient for many energy applications such as process heating and steam generation in industry, water and space heating in buildings and cooking in residential units.  Natural gas accounted for 33% of total primary energy production in 2013, 34.1% in 2014 and reached marketable production of 14.2 Bcf/d 5 in 2014 (Natural Resources Canada (NRC), 2015). Supply of natural gas greatly exceeds domestic demand but Canadian exports are directed to only one market, the United States, which poses challenges due to lack of diversity in markets (Canada’s Natural Gas, 2017). Primary domestic users are industrial and commercial sectors. While industrial consumption depends on economic conditions, natural gas demand and consumption for heating will also depend on weather condition and population growth. Between 1990 and 2008, population grew about 20% as did the number of households and living space, leading to 14% increase in residential energy use but only 8% increase in greenhouse gases (GHG) emissions due to the use of cleaner energy sources and increased energy efficiency (Government of Canada, 2011). According to Fallahi et.al. (2016), Canada still belongs to countries with steady rather than explosive pattern of energy consumption. 1.1.2. Environmental concerns and alternatives to fossil fuels for energy production Although natural gas, which is widely used for heating and electricity production in British Columbia (BC) and Canada, is a relatively clean-burning fossil fuel, it still contributes to greenhouse gases (GHG) emission. Ecosystem deterioration ranging from local- to global-scale due to the extensive use of fossil fuels has been well documented over the past few decades. In                                                  5 Bcf/d = Billion cubic feet per day = 28316846.592 m3/d = approximately equal to one trillion BTUs. 5  recognizing these issues, the province of British Columbia sets a suite of policy actions proposing, amongst others, the use of BC’s plentiful biomass resources for energy generation. One of the most comprehensive strategy is the 2007 BC Bioenergy Plan  which outlines a clean energy vision for the Province, followed by the BC Bioenergy Strategy in 2008 (BC Ministry of Energy, Mines and Petroleum Resources, 2008). In addition to a well-known BC's low carbon electricity generation profile which largely relies on hydropower generation, the mentioned documents further elaborate on the energy-related goals with one of which referring to utilization of biomass through the bioenergy sector development, generation of energy from pine beetle infected wood, development of the BC biomass inventory and investments in bioenergy research and development (BC Ministry of Energy, Mines and Petroleum Resources, 2008). The main goals of BC energy plans and strategies in diversifying energy resources and increasing energy security is to focus on clean energy in order to minimize impacts on climate and the environment, and protect human health.  Biomass refers to all the living matter available in different forms such as:  vegetation, agricultural waste, and residues from forests and industrial operations, animal manure, all of which could be used as energy sources (Searcy and Flynn, 2010). Forest residues refer to a non-merchantable woody biomass, such as tree species and residues from logging practices, including roadside and in-forest wood. In addition, forest residues from industrial operations, such as mill wood waste (sawdust, shavings, bark), are commonly considered as woody biomass - convenient for use as a fuel or energy source. Biomass applications either through district heating or through decentralized heating options with wood pellets are seen as a good solution for Canadian remote  communities in terms of reduction of GHG and heating costs, and increase of energy 6  independence  (Stephen et al., 2016). Uncertainty in forest bioenergy supply chains exists partly due to economic fluctuations, which also affects other energy industries, yet additional complexities exist (Shabani and Sowlati, 2016). 1.1.3.  Availability of biomass resources and district energy systems in British Columbia  British Columbia has abundant forest resources which could be used for energy in many ways.  More than 400 million hectares of Canada’s land (44%) are forests; most under provincial jurisdiction. The largest user of biomass (mostly forest residues) for energy is the Canadian forest products industry which generates almost 60% of its energy from this renewable source (Bradley, 2006). An Inventory of the Bioenergy Potential of British Columbia (Ralevic and Layzell, 2006) identified forest residues from industrial operations, such as mill wood waste (sawdust, shavings, bark), and forest residues from logging practices, as significant woody biomass resources in British Columbia. The same industrial sector, especially pulp and paper industry, utilizes such residues to generate energy (heat and electricity) for its processes. Other studies evaluated pine beetle damaged wood as an additional forest residues-type of feedstock for energy (Mahmoudi et al., 2009) for the next 15years  (Schwab et al., 2009; Envirochem Services Inc., 2008),  and considered  its impacts on the forest sector and province’s economy. All of these resources, if used for energy, have potential to provide many benefits to the province: minimizing wood-waste which would otherwise be either burnt (increasing air pollution) or sent to landfills (increasing GHG emissions, carbon footprint6); development of new biomass-based technologies, and creation of jobs.                                                  6 Carbon footprint – is defined in many ways but in essence accounts for total amounts of CO2 and other GHG emitted over the full life cycle of a process or product. It is expressed as grams of CO2 equivalent per kilowatt hour of energy generation (g CO2eq /kWh) (POST, 2006). 7  Woody biomass could be thermaly converted to energy in many ways. The most traditional applications include domestic applications in fireplaces and stoves and large scale applications in advanced energy-efficient wood combustion (AWC) systems with air pollution controls in place for heating and  electricity generation. These systems are widely used in Europe and became attractive in North America in applications such as district energy systems. More efficient boilers will result in lower emissions than traditional wood combustion system, however, emissions may not be as low as for oil or natural gas boilers so engineered pollution controls are likely to be needed (Chandrasekaran et al., 2011).    Major biomass energy technologies include: a) Direct combustion of pellets,  briquettes, or wood chips with heat or steam as the major product which can be further directed to turbines to produce electricity; b) Pyrolysis, a high-temperature, anoxic7  thermo-chemical process in the absence of oxygen to produce bio-oil, biochar and combustible gases which can be further used for heat and power generation, and c) Gasification, another high-temperature thermochemical process which converts biomass under lean oxygen conditions into synthetic gas which can be further used for heat and power, and chemical production  (Rubio-Maya et al., 2011).     District energy systems (DES) have potential to provide effective energy solutions.  Configured as centralized production of steam or hot water for heating and in some cases electricity for local community (neighborhood), DES are characterized by lower infrastructure costs, lower overall                                                  7 Anoxic – the absence of oxygen.  8  emissions and reduced cost compared to conventional distributed heating systems mainly based on natural gas.  These systems can use a variety of conventional and renewable sources.   Analysis of Swedish energy policy in terms of its effects on district heating (DH) economic performance and climate change mitigation (Gustavsson et al., 2007) demonstrated that the most cost-effective policy option is the investment in biomass-based combined heat and power (CHP) systems in the case of applicable taxes and policies such as Tradable Green Certificate (TGC).  In the case when national taxes and policies are excluded, natural gas fired DH becomes a superior investment  (Difs et al., 2010). About 80% or 4 million residents in Sweden are connected to district heating systems (Swedish District Heating Association, 2014). More than 400 district heating companies supply 98 % of the district heating or some 1.6 million households in Denmark  (Danish District Heating Association, 2014).   Biomass-based DES (Fiorese et al., 2014), configured as combined heat and power (CHP) or heat only production systems (DH), are rapidly growing in Canada. There were merely 3 such projects in 2009 but it increased to more than 100 projects in the last few years (CIEEDAC, 2015; Bradley, 2012) with a total heating capacity of 121 MWth or 3.4% share of Canadian district energy heating capacity from all energy sources (CIEEDAC, 2015). Utilization of bioenergy could be beneficial to Canada’s and BC’s economy, providing improvements in energy efficiency and reduction in greenhouse gas (GHG) emissions. British Columbia also has a number of DES projects which have either already commenced or under development with special emphases to renewable energy resources (Province of BC, 2012). Table 1.1 outlines some of the biomass energy projects in the province. 9  Table 1.1 Examples of biomass DES projects in British Columbia. Project Location Capacity Benefits Dockside Green (Dockside Green Energy, 2008)) Victoria 2 MWth    Supplies hot water to the Dockside Green community,  Enables Zero Carbon footprint of the site. Kruger Products  (Canadian Biomass), 2009)  New Westminster 40,000 lbs/h of process steam   Displace about 445,000 GJ of natural gas annually  Reduces GHG from the plant for 22,000 t per year UBC Biomass Research and Demonstration Facility (BRDF) (UBC, 2015a) UBC, Vancouver 6 MWth 2 MWel  BRDF provides a quarter of campus heating needs,   Eliminates 14% of campus GHG emissions. UniverCity Sustainable Energy Project (SFU, 2016)   SFU, Burnaby 10 MWth  2,400 t of GHG reduction (85% reduction from heating, 69% reduction from all sources),  Reduces the overall cost of energy to the customer. Revelstoke Community Energy System (FVB Energy Inc., 2017)  Revelstoke 1.5 MWth  Diverts 70,000 t of wood residue annually, from beehive burners and improves air quality,  Reduces GHG for 3,200 t annually,  Supplies heat for several buildings and steam for Downie’s sawmill drying kilns. Prince George District Heating  (FVB Energy Inc., 2017) UNBC, Prince George 7.5 MWth  Reduces GHG for 1,900 t annually,  DES  connection to the new Wood, Innovation and Design Centre (WIDC)  Provides heating for several downtown buildings and enables research at UNBC.  1.1.4.  Public perception and acceptance of biomass systems  Since energy crisis in 1970s, those who were promoting political agenda and economic and environmental priorities influenced energy policies world-wide. While there have been extensive discussions on wood for energy with a focus mainly on resource availability and economic needs, public opinion has been rarely heard. As pointed out by Mittlefehldt (2016), conflicting narratives around competing energy alternatives materialized in the direction of creating ecological and public health risks associated with biomass-based energy systems. While supporters promoted biomass as a decentralized energy resource by nature, opponents feared that 10  the development of biomass-based energy sources would create local centers of power, different from fossil fuel related political structures. Social constrains like using forested areas for recreation or cultural activities where harvesting is not allowed, or other land uses issues can reduce biomass mobilization (Kraxner et al., 2016). Thus,  understanding and considering parameters of social acceptance of a novel technology by including citizen in the decision making process is crucial for the deployment of local district heating systems in communities (Zaunbrecher et al., 2016).  Similar findings were presented by a German longitudinal study on public acceptance of decentralized power generation by biomass and relevant influencing factors (Kortsch et al., 2015). The study emphasized a multi-actor and multi-dimensional character of the acceptance process ranging from individual to regional perspectives and factors. While regional economic development and benefits could be readily accepted as positive factors, individual and local scale factors rather revolve around perceived negative impacts caused by increased noise due to truck traffic, smells etc. However, public involvement in the planning process and increased awareness and information diffusion certainly increase the level of acceptance of biomass projects on the local level. A Swedish study (Kautto and Peck, 2012) gave emphasis to stakeholder involvement and new biomass resource mobilization within regional planning while leaving biomass sources in general to national planning.   In general, it could be deduced that the critical factors for the diffusion of bioenergy for district heating are both economic  and non-economic in nature (Toka et al., 2014; Wright et al., 2014; Aguilar et al., 2011), since main barriers range from economic, technological to cultural and psychological. While the main benefits could be seen in the reduction of CO2 emissions, still the negative image of system’s operation and not well understood impacts of airborne emissions 11  may nurture resistance for their adoption. DES are of a much lower capacity (a few MWth) with consequently lower emission rates from shorter stacks (< 20 m) than large power plants             (> 100 MWth) with tall stacks (> 200 m) (Zhou et al., 2003).  This indicates the need of considering DES impacts on a much smaller spatial/temporal scale than a commonly practiced large and remote power plant in order to address the “Not-In-My-Backyard” health concerns. In addition, better connections between urban planning and energy policy development are necessary for the acceptance of DES (Gabillet, 2015). A successful process of transition to biomass DES in Sweden (Di Lucia and Ericsson, 2014)  can serve as a good guidance for the process of adopting different renewable energy choices and biomass in particular.  The full impact assessment of DES, especially those using biomass, with respect to local air quality and community health has not yet been properly investigated and addressed. Very few studies started recognizing the importance of local and urban health impacts of near-by stationary sources. For example, Jonsson and Hillring (2006),  pointed out that meteorological and topographical conditions need to be considered with small-scale DES due to near-source high pollutant concentrations. Another study (Curci et al., 2012) showed increased NO2 emissions from a proposed biomass plant.   Systematic literature review (Chapter 2) revealed  that  previous studies relied on many assumptions and did not account for dynamic population changes and actual spatial and temporal variations of ambient air quality (Martenies et al., 2015), or relied on selected archetypal environments and emission sources (Humbert et al., 2011), which points to the lack of an appropriate impact assessment method for small-scale stationary sources.  12  The lack of knowledge about biomass-based DES impacts is reflected by many rejected biomass-based DES proposals by communities that are concerned about increased health impacts in recent years. One of the well-known projects was a proposed biomass DES for the Vancouver’s Olympic Village which was abandoned in 2006. According to the “City of Vancouver’s memorandum (Appendix A), regarding the energy source for the Southeast False Creek district heating centre” (Ghafghazi, 2011): “The public process to date  with various stakeholders (including individual residents, resident associations, Southeast False Creek Developer, various non-governmental organizations and others) has identified a number of concerns related to biomass: - Perception that wood combustion generates harmful emissions - Perception that truck delivery of wood pellet would have an undesirable impacts - Concern that environmental impacts have not been adequately assessed.”   Therefore, two major knowledge gaps with respect to biomass-based DES were identified:   Knowledge gap 1: the lack of appropriate and accurate impact assessment methodology for parameters with extensive variability on local scale. Such variability may potentially influence the outcomes (impacts) which otherwise would not be noticed and considered;   Knowledge gap 2: assessment of biomass-based DES impacts on local ambient air quality and human health which will be based on impact assessment methods with higher accuracy and inclusion of local, site-specific characteristics.  The University of British Columbia initiated the development of a small-scale biomass research and demonstration facility (BRDF) at the Point Grey campus in Vancouver to enable not only research and demonstration of biomass conversion technologies but also quantification and 13  potential reduction of air emissions and a range of environmental impacts of biomass applications for community-based energy systems. This research work therefore presents timely and much needed study to contribute to our knowledge on potential impacts and sustainability characteristics of community-based biomass energy systems. 1.2 Thesis objectives and research questions In order to address knowledge gaps and to address community and other stakeholders’ concerns, this study sets the following primary objectives: a) Improve current approach (methodology) for air quality and integrated health impact assessment of community-based biomass district heating systems; b) Investigate, by applying the proposed methodology to a case study,  the impacts of signature pollutants such as airborne fine filterable particles (PM2.5), oxides of nitrogen (NOx), and carbon monoxide (CO) on ambient local air quality, population  exposure potential expressed by inhalation intake fraction (iF) and  health risks expressed by impact score (IS).  Additionally, this study also aims to: c) Update an in-house Life Cycle Inventory database for British Columbia with the foreground fuel supply and conversion data for the UBC Bioenergy Research and Demonstration Facility (BRDF); d) Investigate global impacts of greenhouse gas (GHG) emissions over the entire life cycle in terms of environmental damages such as climate change and human health; e) evaluate sustainability of district heating options connecting their environmental, social and economic characteristics.    14  In doing so this study addresses the following research questions:  1. How would the inclusion of site-specific terrain, land use and microclimatic characteristics, variable population density and breathing rates improve accuracy of local air quality and population health impact assessment of community-based biomass energy systems?  2. How would an incremental increase of PM2.5, NOx and CO concentrations from investigated biomass DES contribute to local effects such as ambient air quality and population exposure? 3. How would life-cycle GHG emissions from the investigated biomass DES contribute to global warming? 4. Considering capital, operational and maintenance (O&M) costs and externalities, how would the introduction of biomass-based DES affect economics compared to fossil fuel-based DES? 1.3 Case study The University of British Columbia (UBC), Point Gray campus in Vancouver was selected as a community for this study from a number of reasons. The term “community” in urban context was generally  recognized as one that occupies certain geographical area, but unlike  cities defined by specific size, communities are  rather characterized by social networks (Huang et al., 2017b; Petersen, 2016). Communities share identity, have common interest and values and therefore planning and implementation of policies could be reached in a more meaningful manner. Examples include community-scale energy system planning incorporated with urban planning as response to climate change (Lin et al., 2017), or those using a risk-based methods (Ioannou et al., 2017) where community members are an important stakeholder. A variety of techno-economic parameters are also commonly used for evaluation of options (Ghafghazi et al., 2010; Arena et al., 2010) and energy planning for sustainable future (Bhowmik et al., 2017). 15  The Bioenergy Research and Demonstration Facility (BRDF) at UBC Vancouver campus is one of the most innovative and inspirational renewable energy developments recognized world-wide (UBC, 2015b). Built in 2012, this permitted biomass DES8 is using the Nexterra gasification-combustion technology for CHP generation, to demonstrate the technology and to allow researchers to study emissions and their dispersion characteristics in a community setting among other projects. Wood waste, a mixture of forest residue and sawmill/planner waste is used as the fuel at the BRDF, supplied daily by Cloverdale Fuels Inc. (Cloverdale). Adding biomass to DH reportedly avoided 5,500 tonnes of fossil CO2 which would have been otherwise emitted  from natural gas combustion during the first year of operation (UBC, 2015c). Stack emissions are closely monitored and correlated to the quality of biomass feedstock and the local meteorological conditions. Since 2012 when it became operational, BRDF produced steam for approximately 20% of the campus’s thermal energy demand of 1,011 TJ over the period 2012-2013 (Petrov et al., 2017). The rest was produced by combusting natural gas (base load) and #2 heating oil (peak load) at the UBC Power House (PH) built in 1925, which is gradually being replaced with the new Academic District Energy System (ADES), positioning “UBC as a Living Lab” with a more efficient hot water instead of steam heating system.   UBC Point Grey campus is a vibrant community of researchers, students, residents, employers and visitors with a pronounced daily and seasonal dynamics.  It is continuously growing and new developments are providing more space and facilities for research and residency on its 4.02 km2 property. Approximately 50,000 people daily work, reside or visit the campus, staying in more                                                  8 Capacity: Thermal mode only 5.8 MWth, 2.8 MWth heat recovery and 1.96 MWel. 16  than 500 buildings of different uses (offices, classrooms, laboratories, libraries, dormitories, etc). Some of the residences such as Marine Drive residential complex are located just across BRDF to the north and north-west. Due to such close proximity of residential buildings to a 20 m tall BRDF stack, both oxides of nitrogen and fine particle monitors were installed on the roof of Marine Drive building 5 to ensure acceptable levels of those pollutants at all times. Configuration and capacity of campus buildings (depicted as yellow rectangular surfaces), PH and BRDF as emissions sources considered in this study (depicted as red stars) are presented in Figure 1.3 and explained in detail in Chapter 3: and Chapter 4: of this study.          Figure 1.3  UBC campus buildings, BRDF and PH emission sources. Source: UBC Campus + Community Planning.  A local weather station, Totem park station, is located to the south of BRDF and data on ambient temperature, humidity and wind parameters could be downloaded. 1.4 Thesis structure The thesis is organized in chapters starting with the introduction chapter and the literature review 17  chapter, followed by four chapters on research results, and a final chapter on overall conclusions drawn from the research work, limitations of the current study and opportunities for the further research. More specifically:  Chapter 2 provides a systematic literature review (detailed in Appendix A.1) with the goal to address the current scientific knowledge about three main topics covered in the subsequent chapters. Sections 2.1 to 2.3 cover biomass classification and characterization and control of resulting emissions in order to evaluate impacts of biomass-based district energy systems on local air quality and climate. Section 2.4 reviews existing literature addressing population exposure and associated health risks due to such exposure with a focus on inhalation intake fraction (iF) and health-related impact score (IS) metrics. Finally, subsection 2.5 recapitulates published results on global effects of biomass energy systems using life cycle assessment.    Chapter 3 is dedicated to developing an improved impact assessment methodological approach, with a focus on improving a dynamic iF method for assessments on local, community scale, environmental impacts (connecting local air quality and human health). The importance of introducing local and site-specific parameters for more accurate quantification of impacts is highlighted. The overarching goal of this chapter is to present a comprehensive methodological approach which could be generalized for assessing community-scale energy systems, while its application is demonstrated in subsequent chapter. Methodologies used for assessing the global issues are also covered.  18  Chapter 4 starts with a section describing a district heating system at the UBC Point Gray campus in Vancouver, BC, which was chosen as a case study. The subsequent two sections explain operational district heating scenarios considered in the study (section 4.3) and emission estimates and measurements in section 4.4. The second part of this chapter applies improved assessment approach (Chapter 3) and investigates in detail impacts of emitted pollutants from two operational plants, BRDF and PH, over five operational scenarios on local air quality (section 4.5) and human health (section 4.6). Where applicable, obtained result were compared with regulatory limits for emission sources and ambient air quality objectives.   Chapter 5 focuses on global impact assessment, in which three main DES operational scenarios were subjected to a streamlined life cycle analysis to quantify the global warming impact. Upstream processes for natural gas and heating oil as well as transportation, electricity and machinery operations were obtained via GHGenius and included in the analysis. Based on actual data on fossil fuel consumption at UBC and biomass feedstock locally supplied and gasified at BRDF, foreground information on energy and material flows as well as wastes generated were summarized and analyzed using two impact assessment approaches: 1) MS Excel for compiling emission inventories to evaluate impacts over different life cycle stages, and 2) SimaPro software with Ecoinvent database and IMPACT 2002+ for global impact analysis.   Chapter 6 tackles economic analysis as an important pillar of sustainability. Assessment of costs associated with the development, operation and maintenance of the UBCdistrict heating system is discussed along with the economics of GHG emissions. Externalities are also discussed.   19  Chapter 7 as the final thesis chapter summarizes findings of this study, outlines strengths and limitations and provides recommendations for future research work. 20  Chapter 2: Literature Review 9,10  The current knowledge on DES with a focus on biomass as a feedstock was reviewed systematically to identify knowledge gaps in possible impacts, methods and approaches used for evaluating such impacts.  Further details are provided in Appendix A1.   2.1  Biomass classification and characterization Extensive investigation into biomass characterization and its potential for sustainable utilization for replacing fossil fuels and combating climate change have been reported in recent years. Generally, biomass could be generated either from natural processes (vegetation through photosynthesis, animal waste and food waste) or from processing of naturally obtained biomass such as municipal solids waste. According to Vassilev and collaborators (Vassilev et al., 2010), biomass could be classified as:  Wood and woody biomass – which includes various wood species such as coniferous or deciduous, parts of a tree such as stem, branches, bark, but also various woody biomass forms such as pellets, briquettes, chips, sawdust;  Herbaceous and agricultural biomass – grasses, straws and plant residues;  Aquatic biomass – algae, seaweed, lake weed;  Animal and human biomass waste – manures, chicken litter and others;                                                  9 A version of this chapter was published. Petrov, O. (2012). Forest Residues to Energy: Is this a pathway towards healthier communities? National Collaborating Centre for Environmental Health. Evidence Review. Available from: http://www.ncceh.ca/sites/default/files/Forest_%20Residues_to_Energy_Mar_2012.pdf. 10 A version of this chapter was published. Petrov, O., Bi, X., & Lau, A. (2015). Impact assessment of biomass-based district heating systems in densely populated communities. Part I: Dynamic intake fraction methodology. Atmospheric Environment, 115, 70–78. https://doi.org/10.1016/j.atmosenv.2015.05.036. 21   Industrial biomass waste – wastes from municipal works such as tree trimming, sewage sludge, demolition wood and others. 2.1.1  Chemical composition Biomass is a heterogeneous mixture of organic and, to a lesser extent, inorganic matter known as ash. The chemical composition of biomass, especially the inorganic portion, varies due to high variation in moisture content, ash yield,  and ultimately due to the biomass origins (Vassilev et al., 2010). Organic compounds are comprised of five main elements: carbon (C), hydrogen (H), oxygen (O), and nitrogen (N). The major elements (>1.0%) are carbon (C), oxygen (O), hydrogen (H), nitrogen (N), calcium (Ca), and potassium (K).  The minor elements (0.1-1.0%) commonly found in biomass are: silica (Si), magnesium (Mg), aluminum (Al), sulphur (S), iron (Fe), phosphorous (P), chlorine (Cl), sodium (Na) (Vassilev et al., 2010), as well as:  cadmium (Cd), chromium (Cr), copper (Cu), nickel (Ni), zinc (Zn), arsenic (As), mercury (Hg) and lead (Pb) (Telmo et al., 2010). The trace elements (<0.1%) are manganese (Mn) and titanium (Ti). Municipal wood waste could also be contaminated with a number of other elements. In order to be used as fuel, some of the main properties to be considered are:  2.1.2  Heating value Structural analysis of biomass (main constituents: cellulose, hemicellulose and lignin) is important in estimating the higher heating value (HHV) through ultimate fuel analysis where the HHV of lignin is reported to be higher than HHV of cellulose and hemicellulose. HHV could be directly measured or estimated as (Vallios et al., 2009): HHV = 34.1C + 123.9H – 9.85O + 6.3N + 19.1S          (2-1) 22  where: HHV is higher heating value in [MJ/kg]; C, H, O, N and S  are carbon, hydrogen, oxygen, nitrogen and sulphur in [weight %]. Typical HHV values of different types of biomass are: green wood 8 MJ/kg, spruce wood 20.5 MJ/kg, softwoods 19.8 MJ/kg, hardwoods 19 MJ/kg, wood bark 20.3 MJ/kg, sawdust 18.4 MJ/kg (Saidur et al., 2011). 2.1.3 Moisture content Moisture content is an important parameter as it directly impacts the combustion performance of biomass fuels. Ideal fuel would have low moisture content (Singh et al., 2017; Singh et al., 2014). While fresh wood may contain more than 50% of moisture (Striūgas et al., 2017; Zeng et al., 2017), pellet are typically between  5.1% and 8.5% moisture content for achieving high pellet density and strength (Huang et al., 2017).    2.1.4 Ash content Ash content is indicative of the presence of inorganic and mineral compounds in biomass. It is one of the most studied biomass characteristics. The ash yield is the inorganic residue (formed from organic, inorganic and fluid components) resulting from the combustion process. Combustion temperature will have a substantial impact on total ash yield resulting in 20 -70% less ash for combustion temperatures above 1,000 ºC  (Vassilev et al., 2010). High ash yields containing Cl, K, Na, P, S as well as other elements forming chlorides, sulphates, carbonates, oxalates, nitrates, to mention some, may cause issues during biomass thermochemical conversion (Vassilev et al., 2017). Furthermore, the composition of ash will depend on the biomass species and part of the biomass plant, with bark having higher ash content than wood (Saidur et al., 2011) . 23  2.2 Characterization and control of emissions from biomass-based energy systems Conventional furnaces (such as cooking stoves) and  open biomass burning (such as forest fires) emit particulate matters (PM) and a wide range of gaseous pollutants such as oxides of sulphur (SOx), oxides of nitrogen (NOx), carbon dioxide (CO2), carbon monoxide (CO), black carbon, free radicals and various organics (Naeher et al., 2007; Gustavsson et al., 2007).  By comparison, advanced thermochemical conversion systems such as gasifiers are characterized by a reduction in the number of pollutant species and the concentration of PM, CO and volatile organic compounds (VOCs) (Sethuraman et al., 2011; Miranda et al., 2010).  In addition, as previously mentioned, woody biomass is usually a heterogeneous fuel and emissions depend on the tree species and moisture content.   Boiler type and operating conditions, as well as the type of biomass, affect particulate and gaseous emissions (Kaivosoja et al., 2013; Kocbach Bølling et al., 2009; Boman et al., 2004;     Boman et al., 2003). For example, when combusted in high efficiency boilers, wood chips (from forest residues and waste wood) emitted significantly higher fine particles with a diameter less than 2.5 microns (PM2.5) and NOx and SO2 gases due to higher sulphur and nitrogen contents in wood chips. These emissions are higher than those emitted from the combustion of wood pellets. Sulphur content in wood pellets and wood chips ranges from 63.6 to 175 mg/kg dry wood, which is much lower than sulfur content in fossil fuels (Chandrasekaran et al., 2011).   The emission of gases and particulates in modern wood boilers is also lower than old-type residential boilers (Johansson et al., 2004). While  the assumption of carbon neutrality of forest biomass is not correct (Röder et al., 2015; Vanhala et al., 2013;  Holtsmark, 2013),  fuel derived 24  from woody biomass indeed has much lower greenhouse gas (GHG) emissions such as CO2 and methane (CH4) when compared to natural gas over the entire life cycle (Pa et al., 2011).  Methane could be formed during biomass gasification/pyrolysis in the reducing zone, together with CO and H2 (Sansaniwal et al., 2017) but is not directly released to the environment.    Air pollution control devices such as electrostatic precipitators (ESP) and selective catalytic reduction (SCR) need to be installed for the removal of particulates and NOx, respectively. In general, air pollution control could be achieved using dry and wet methods. Dry cleaning methods do not use liquid but rather use gravity, centrifugal force, impaction, direct interception, electrostatic attraction and other mechanisms for pollutant removal. Examples of such controls for particle removal from flue gases are: cyclones, filters and ESPs. While cyclones are most efficient for coarse particles, ESPs and filters can achieve high removal efficiency over 99% for fine particles (Asadullah, 2014; Ghafghazi et al., 2011). Wet controls for particle removal include a large selection of scrubbers, wet ESPs and hybrid controls (Singh and Shukla, 2014;  Ghafghazi et al., 2011). Wet methods, e.g. wet scrubbers, are also used for the removal of soluble gases via absorption in addition to adsorption. Some water soluble gases from biomass combustion are sulphur dioxide (SO2), ammonia (NH3), hydrochloric acid (HCl) and hydro-fluoric acid (HF) (Singh and Shukla, 2014). 2.3 Impacts on ambient air quality and climate  Maintaining good air quality is a challenge with population growth and industrial development.  Even switching from fossil fuels to renewables needs to be evaluated beforehand to ensure maintaining air quality within prescribed limits and minimizing health risks.  One study  25  (Jonsson and Hillring, 2006) found, based on dispersion modeling, that conversion to small scale district heating resulted in higher pollutant concentrations closest to the emission source (and then decreasing and spreading over a larger area), than the case of pellet stoves in individual houses. However, in both cases outdoor concentrations still remained within allowable air quality limits. The study, however, indicated that other factors may impact the dispersion of emitted pollutants and consequently ambient air quality (such as terrain, temperature inversion11). So contributing background pollutant levels and site-specific local emissions need to be investigated together.   As depicted in Figure 2.1, spatial and temporal scales of processes, which span eight orders of magnitude, exist in the atmosphere. Four main categories are:  Microscale – which exists up to 100 m in space and minutes to hours in temporal scale; short-lived species such as free radicals and process are described  by turbulent motions;  Mesoscale – which exists on a spatial scale of tens to hundreds of kilometers where processes such as sea- and land-breezes, mountain-valley circulations dominate and oxides of sulphur, tropospheric ozone and aerosols;  Synoptic scale – which is well known for the motion of weather systems over hundreds to thousands of kilometers and which is closely connected to the global scale, impact transport of moderately-lived (hours to years)  species such as oxides of nitrogen,                                                   11 Temperature inversions are defined as an increase of ambient (i.e. outdoor) air temperatures with altitude which leads to stable atmospheric conditions and poor dispersion. 26   Global scale – which is the largest spatial and temporal scale existing on a tens of thousands of kilometers where long-lived species such as methane (CH4), nitrous oxide (N2O), chlorofluorocarbons (CFCs), known as greenhouse gases (GHGs) and ozone depleting species exist for years.            Figure 2.1  Atmospheric species on spatial and temporal scales.  Source: Based on Pandis and Seinfeld (2006).  These scales overlap so do the processes and species which undergo transport, chemical transformations and depositions after being emitted into the atmosphere. While short- and moderately-lived species determine the quality of outdoor (ambient) air on an urban/local scale, long-lived species have a profound impact on global scale, most notable of which is climate change and stratospheric ozone depletion (Pandis and Seinfeld, 2006). Among environmental issues, emissions of fine particles are recognized as ones very hard to predict as they will depend Synoptic toGlobal ScaleCO2CFCsN2O10 years CH4Moderately Long COLived SpeciesSO21day NOx H2O2DMS1hr C5H8Short-lived CH3O2100sec HO2NO31sec OH1m 10m 100m 1km 10km 10,000km MesoscaleC3H6MicroscaleUrban or Regional or Local scaleSpatial scaleTemporal scale100yearsLong-livedSpeciesBoundary layer Mixing Time1yearAerosolsO3 tropospheric100km 1,000kmCH3BrCH3CCl327  not only on biomass characteristics and operational conditions but also on local meteorology (Pantaleo et al., 2014). 2.4 Population exposure and health risks Situated in communities, district energy systems (DES), even with renewable sources, can raise concerns about health risks for local populations especially with respect to fine PM2.5 and NOx levels (Genon et al., 2009; Jonsson and Hillring, 2006). Unlike conventional energy systems located in remote areas away from city centers, proximity of district energy  systems can have direct impact on residents (Pa et al., 2011). Therefore, exposure scenarios in addition to emissions could help evaluate health risks of DES and compare them to conventional systems   (Genon et al., 2009; Heath et al., 2006). Like in cases of conventional wood burning, Intake Fraction (iF) could be used as a metric for district heating systems to evaluate the inhaled portion of airborne pollutants by exposed populations (Ries et al., 2009).   For the assessment of air pollution and public health, inhalation intake fraction (iF), as the fraction of iF which encompasses three routes of exposure (inhalation, ingestion and dermal) should be used. Inhalation intake fraction is also called exposure efficiency (Evans et al., 2000; Lai et al., 2000; Smith, 1993), exposure effectiveness, nominal dose effectiveness (Smith, 1993) intake factor (Čupr et al., 2013) and inhalation transfer factor (Lai et al., 2000) by different authors, and it has been widely used as a key metric for evaluating population exposure to pollutants emitted from a source or source class including stationary (power plants), mobile (vehicular traffic) or other sources. In its simplest form it could be expressed as the incremental intake of a pollutant emitted from a source of interest and summed over exposed individuals of 28  the studied population and exposure time, per unit of pollutant released from that source into the environment (Bennett et al., 2002). Inhalation intake is a product of airborne concentrations, population density at a location of exposure and breathing rate (Evans et al., 2002).  Inhalation iF has been used in evaluating impacts from different emission sources, such as  urban smoke emissions in Canada (Ries et al., 2009), sulfur dioxide (SO2), sulfate (SO4), nitrogen oxides (NOx), nitrate (NO3) and fine primary particle (PM2.5) emissions from industrial stacks in China  (Wang et al., 2006;  Zhou et al., 2006; Zhou et al., 2003), centralized and distributed electricity generation plants (Heath and Nazaroff, 2007; Heath et al., 2006) or other outdoor origins (Marshall et al., 2006)  in  the United States and Czech Republic (Čupr et al., 2013), proposed biomass plant in Italy (Curci et al., 2012), and non-reactive pollutants (Lobscheid et al., 2012; Du et al., 2012; Luo et al., 2010; Greco et al., 2007; Marshall et al., 2005a) or organics of particular concerns for human heath such as benzene from vehicular sources (Manneh et al., 2010; Loh et al., 2009).  iF has also been used for exposure assessment associated with episodic exposures (Russo and Ezzat Khalifa, 2010; Nazaroff, 2008), or cooking in indoor micro-environments (Grieshop et al., 2011).   To estimate iF, a variety of approaches were applied on different spatial and temporal scales. Ambient pollutant concentrations were usually obtained by modeling, ranging from steady-state mass balance models (Manneh et al., 2010; Marshall et al., 2005b), to box models (Stevens et al., 2007a), and more sophisticated dispersion models such as ISC (Panepinto et al., 2014; Wang et al., 2006),  AERMOD (Lobscheid et al., 2012), CALPUFF (Curci et al., 2012;  Zhou et al., 2006; Zhou et al., 2003) and CMAQ (Xu et al., 2013) for stationary sources. Some authors (Zhou and 29  Levy, 2008; Greco et al., 2007)  recommended higher spatial resolution dispersion models, especially for primary conserved pollutants such as PM2.5 due to a significant near-source contribution, and because they can improve iF estimates and increase confidence in results (Manneh et al., 2010). Furthermore, many studies considered static and uniform population distribution based on either census tract population data or region and country average population data. One study (Marshall et al., 2006) introduced population stratified by age, income, ethnicity and 4 micro-environments for vehicular emissions exposure. In most of the reviewed studies using iF, breathing rate was 20 m3/day for an adult during the day regardless the level of activities; only few studies, mostly dedicated to traffic exposures, introduced some kind of variation in breathing rates (Lobscheid et al., 2012; Luo et al., 2010; Loh et al., 2009). According to Wang et al. (2006), using a constant breathing rate of 20 m3/day has long been recognized as a weakness of previous iF studies.  Intake fraction values vary by several orders of magnitude across reviewed studies. For stationary sources iF values for urban, rural, remote areas and ground-level, low and tall stack range from 0.1 to 44 ppm (Levy et al., 2002a) to 260 ppm for residents and 1000 ppm for pedestrians in a street canyon (Zhou and Levy, 2008).  For a biomass-based DES located in densely populated urban areas, there is an expected high degree of variation in population, and possibly in micro-meteorological conditions which imposes extra challenges for accurate estimation of iF values. Summary of reviewed iF related studies is presented in Table 2.1. 30  Table 2.1 Summary of iF evaluation approaches based on the reviewed literature. iF Study Goal/Scope Pollutant Concentration Calculation Method Population Density Method Breathing Rate IF estimates (x10-6) Reference Point source(s)/energy related studies Forecasted the temporal and spatial distribution of PM10 pollution from 3 main sources in Taiyuan City, China (BPANN) model  33 x 16 grids; each grid 500m x 500m PM monitoring data Yearly average data Data from city’s  records 2002-2008 Yearly average population per km2 used as pop. density  temporal resolution   Constant breathing rate of an adult 20 m3/day mean = 8.5 in urban area mean = 4.61 in suburbs    Zhang et al., 2013 Characterization of properties of 6 size fractions of PM(focus on PAHs) an assessment of the human health risks they pose [Brno, Czech Republic] PM Sampling followed by mineralogical and chemical analyses  Exposure scenario, a person body weight 70kg, exposure 8hr/day for 70 years Constant breathing rate of an adult 20 m3/day IF > 1.5 (Statistically significant genotoxic potential, GP) for PM < 0.45µm and 0.95 µm>PM>0.4 µm for 30m3/ml When IF expressed per mg of PM and associated PAHs, then the highest GP is for PM in range 1.5-3 µm;0.95-1.5 µm and 0.45-0.95 µm  Čupr et al., 2013 Intake avoided per unit of SO2 emissions reduced; Beijing-Tianjin-Hebei region, China  CMAQ modeling system Population of 35 sub-areas obtained from provincial statistical yearbooks Constant breathing rate of an adult 20 m3/day Avoided iF per tonne of SO2 Heat & electricity              0.473 Smelting                            0.646 Other                                 0.934   Xu et al., 2013 Impact of SO2 ,NO2 and PM10 emissions from a proposed biomass energy power plant, Italy CALPUFF dispersion model; Domain 40 km×40 km, 250 m resolution, 8 vertical layers City population of 70,000 people Constant breathing rate of an adult 20 m3/day Max predicted  for SO2 and PM10              ≈   25     Curci et al., 2012 Integration of PM-related emissions and PM human exposure into LCA –microenvironments: outdoor (urban, rural & remote), and indoor Source-location framework; 3 emission heights in different microenvironments; regression models from literature Based on average population density for urban, rural, remote areas 13 m3/day For primary PM2.5 Stack       urban    rural     remote High          11        1.6         0.1 Low           15        2.0         0.1 Ground      44        3.8         0.1 Emission   26        2.6        0.1 Weighted average Humbert et al., 2011 31  iF Study Goal/Scope Pollutant Concentration Calculation Method Population Density Method Breathing Rate IF estimates (x10-6) Reference For 32 substances (8 relevant to inhalation iF), evaluation of spatial iF variation within and across the 3 levels of regionalization (LCA),Canada Steady-state mass balance equation 3 spatial resolutions: 15 eco-zones, 13 provinces, and 172 sub-watersheds all with 537 air regions with the same mixing layer and world-level (box model) compartment  varying n/a The highest intake is for long-range transport chemicals and is driven via intake by world-level spatial compartment due to large population For low persistent chemicals higher resolution needed in LCA to capture population density variations   Manneh et al., 2010 iF of winter urban wood smoke  - concentration of PM2.5 and levoglucosan , Canada Mobile monitoring LUR Winter daytime, winter nighttime and shoulder heating season fall/spring Aggregate and census tract population data, 2001 Census Canada Commonly used BR adjusted ±20% for day/night Geom. mean/geom. SD  PM2.5                                 13 (1.9;   6.6 -24)  Levoglucosan        15 (3.3; 4.5-50)   Ries et al., 2009 NOx , PM2.5 and CH2O Compare California’s 25 existing large scale central power stations (CS) and 11 hypothetical distributed electricity generation (DG) plants Gaussian plume modelling system Year 2000 census tract-level population data, no temporal variability 12 m3/day Median:                       CS                      DG NOx             0.66                     11 PM2.5                 0.78                     16                       CH2O          0.66                     13 Heath and Nazaroff, 2007 Conserved     0.8                   16 Primary  pollutants   Heath et al., 2006 Evaluate impacts of emission source location on population exposure in terms of PM and SO2 emissions;  29 plants in China  CALPUFF dispersion model  1999 country-level population data Constant breathing rate of an adult 20 m3/day  Primary PM2.5       10         average SO2                       5         average Sulfate                  4         average Nitrate                  4         average    Zhou et al., 2006 32  iF Study Goal/Scope Pollutant Concentration Calculation Method Population Density Method Breathing Rate IF estimates (x10-6) Reference Inhalation iF of 5 air pollutants of outdoor origin; California’s South Coast air Basin CAMx Eulerian photochemical air pollution model; Resolution – hourly values in 2x2km  grid  cells in a 210x120km domain ~25,000 individuals, stratified by age, income,  ethnicity Time-location activity survey data 4 microenvironments: outdoor, indoor &residence, indoor and non-residence, in/near motor vehicles Age-, gender- and activity –specific; Calculated average to be 13.1 m3/day Inhalation intake rate: Diesel  PM2.5          47 µg/day    Variation in intake rates from  4 -19%  when varying parameters (BR, mobility, location, all parameters)   Marshall et al., 2006 SO2 and total suspended particles (TSP) iF emitted by 590 stacks of 4 industries, China Industrial Source Complex Long Term (ISTLT3) model Within 50 km  1kmx1km grid- densely populated areas;  Country-level pop. data for industrial -rural areas 20 m3/day Sensitivity analysis with 12, 15 and 17 m3/day SO2          4.2 ± 9.16           average TSP       4.4 ± 8.15           average  Wang et al., 2006 Seasonal iF for emissions of sulfur dioxide (SO2), sulfate (SO4), nitrogen oxides (NOx), nitrate (NO3) and fine primary particles (primary PM2.5).  power plant, China CALPUFF dispersion model; 3360 km × 3360 km domain with grid spacing of 28 km and 120 receptors County-level population data for the year 1999; ArcGIS was used to match population data with the concentration data Constant breathing rate of an adult 20 m3/day  Feb. May Aug. Nov. SO2 13 5 8 8 SO4 11 3 6 4 NO3 15 2 2 7 PM2.5 25 9 13 14   Zhou et al., 2003 A regression-based  model for iF of primary and secondary PM(LCIA) 40 coal-fired Power plants Based on the case study prepared by Wolff who used CALPUFF model of domain 100 k x100 km 1999 data; Estimate of total population within a fixed radius from the  source As per case study (n/a) Mean Primary PM2.5               2.2 Mean secondary sulfate     0.2 Mean secondary nitrate     0 .035 Primary PM2.5  --   iF greater for power plants with lower stacks, lower near-stack mixing height, higher near-source population     Levy et al., 2002b Traffic related studies IF of non-reactive constituents of motor vehicle exhaust, China Monitoring data of carbon monoxide  Government census data for 1996, 2001 and 2006 interpolated to 12.5 – 20.5 m3/day depending on age groups Average annual      270 For children and adults, exposure to motor vehicle emissions outdoors   Luo et al., 2010 33  iF Study Goal/Scope Pollutant Concentration Calculation Method Population Density Method Breathing Rate IF estimates (x10-6) Reference allocate population to different age groups and 4 micro – environ. in/near vehicles is comparable with indoor exposures iF of primary conserved air pollutants from on-road vehicles, US AERMOD steady-state plume model; 50 km of the centroid of the source census block Census-tract spatially variable to include: county, state and national levels Long term average 14 m3/day Pop. weighted mean    = 8.6 Pop. weighted median = 3.6 to 5.1                      For census regions  Pop. weighted  med    = 2.2 to 7.5 Urban areas  = 14         average Rural areas       =9         average  Lobscheid et al., 2012 iF of NOx and, PM2.5 emissions from vehicles, China  24-hr personal exposure sampling for 114 individuals and concentration monitoring in urban area of Beijing 3 microenvironments (traffic, work, home) for adult and children population groups Data from Beijing Statistics Bureau 2008 0.35 - 2.85 m3/hr  based for 3 micro environments for adults and children (0.0171±x 0.0124)x10-3 ppm  [PM2.5] for an individual over 24 hr; (0.0136±0.0087) x10-3 ppm  for children – average; (0.0199±0.0143) x10-3 ppm for adults – average; Total children popul. = 18 ±11ppm Total adults popul. = 135±96 ppm  Du et al., 2012 Spatial and population-based iF for vehicular benzene emissions, Finland 3 methods:  EXPAND modelling approach (traffic planning model EMME/2, emission modelling CAR_FMI, streets poll. Model OSPM);  Personal monitoring;  Box model From EXPOLIS project. 4 activities;  Average population in the area 1 m3/hr constant rate (for EXPOLIS) and for EXPAND modeling used different BR depending on micro-environm. EXPAND: annual mean = 10 Monitoring: Median = 30;    Mean=39 Box model: Median = 4;    Mean=7 Average=0.01 from measured data for 48-hr   Loh et al., 2009 Evaluation of iF of fine particles PM2.5  from sources (6 categories) in  Europe and Finland The regional-scale dispersion model SILAM. 2 geographical domains- Europe and Northern Europe; Spatial resolution 5 km and 30 km Finland, population data 2004 with resolution 250 x 250m; EU countries, EEA database, 100x100m (2001); Non-EU countries, CIESIN, 2-4km (2000) Constant breathing rate of an adult 20 m3/day Europe                             0.31- 4.42 Finland – traffic                       0.68 (the lowest iF for power plants,  0.5) Winter iF > other seasons Summer iF < other seasons iF is 1.3 times larger for smaller spatial resolution   Tainio et al., 2009 34  iF Study Goal/Scope Pollutant Concentration Calculation Method Population Density Method Breathing Rate IF estimates (x10-6) Reference Exposure of residents to seasonal and annual average PM2.5  and elemental carbon (EC) from diesel trucks, Long Beach, UC CALINE4 line source model Census block, block group and parcel, year 2000 - to evaluate the influence of different spatial resolutions on estimated population exposure Constant breathing rate of an adult 20 m3/day    PM2.5   Average          14 (range 10 - 22) iF in winter is 1.4 times higher than in summer iF of streets traffic is 1.4 times higher than those of freeways traffic  Wu et al., 2009 Evaluates the impact of street canyons (median building heights) to primary conservative and reactive pollutants from traffic, NY, US OSPM model Residents, workers, pedestrians US census data (2000) LANL 250m raster – daytime and nighttime population CHAD,  ACS databases 12-38 m3/day depending on population category;  does  not differentiate BR  day vs night PM2.5 Pedestrians            ~ 1000  Residents                    260 Total iF                     2200  PM10 Pedestrians             ~ 1000  Residents                     150 Total iF                      1700  Zhou and Levy, 2008 Evaluation of primary and secondary PM iFs for Mexico City using 5 different methods -Box models -Atm. dispersion complex model -Emission inventory –  PM  composition model -Regression analysis Population census  data, 2000 Constant breathing rate of an adult 20 m3/day Factor-of-five in variability of iF among different methods  Stevens et al., 2007a; Stevens et al., 2007b  Evaluates the spatial extent of mobile source iF - four mobile source iFs for primary and secondary PM2.5   in 3080 US counties/national  Source-receptor matrix (regression model) based on Climatological Regional Dispersion Model (CRDM) 1990 Census data (Sensitivity analysis for 2000 Census data) County-level data Constant breathing rate of an adult 20 m3/day Primary PM2.5 0.2– 25 (median=1.2;   mean= 1.6) The average across the US = 2.5 The median iF of Secondary sulfates  is a factor of 6 greater than the median iF secondary nitrates Greco et al., 2007 iF for non-reactive vehicle emissions in US urban areas One-compartment steady-state mass balance model Emission-to-concentration relationship Analyzing US NATA 2002 population and area data;  linear population density  12.2 m3/day based on metabolic activity Population-weighted mean varies from  Marshall et al., 2005b Impacts of urban population density and Single compartment model - concentrations Spatial variation of population density n/a4.4 (NATA data) to 21 (one-Smaller-sized areas tend to decrease vehicle emissions while increase Marshall et al., 2005a 35  iF Study Goal/Scope Pollutant Concentration Calculation Method Population Density Method Breathing Rate IF estimates (x10-6) Reference land area changes on per capita inhalation intake of primary pollutants from vehicles are uniform throughout the area compartment model per capita intake; urban sprawl tends to increase vehicle emissions but to reduce per capita intake; iF for carbon monoxide (CO) and C6H6 from vehicles California South Coast  Air Basin, US Ambient monitoring data, period 1996-1999 The average population density of 860/km2 Census US 2001 12.2 m3/day based on metabolic activity CO                     C6H6 32                       36 Marshall et al., 2003 Indoor exposure related studies Cook stove replacement options (PM2.5) Use available exposure and emission data Exposed individual 7.8 m3/day assumed for children and female adults Median     0.18 If are 6 times lower in houses with a chimney than without a chimney Grieshop et al., 2011 Individual iF of PM generated in kitchens Measurements to determine the size dependant emission rate; Computational fluid dynamics (CFD) modeling Individual exposure Air ventilation rates 518.4 m3/hr High exposure to PM even when exhaust hood used as intervention to remove PM Gao et al., 2013 iF of a seated person in the office 2.6m x  2.5m x1.7m with multiple contaminants CFD model under different ventilation and temperature regimes Individual exposure Computer simulated person - Personal ventilation system reduces iF by an order of magnitude, body T changes little effect on iF  Russo and Ezzat Khalifa, 2010    36  Assessment of exposure to airborne pollutants is an essential component of human health risk assessment (HRA) (World Health Organization, 2014). Pollutant concentrations could be either measured or modeled (Branco et al., 2014; Gulliver and Briggs, 2011), and exposure evaluated in conjunction with population activity (Gerharz et al., 2013).  Relatively recently developed methods using remote sensing, land use regression modeling (Lee et al., 2017; Dirgawati et al., 2016; Knibbs et al., 2014), or combined methods (de Hoogh et al., 2014)  improved HRA. Risk assessment can also utilize iF instead of pollutant concentrations (Ji et al., 2011). Subsequently health risks could be estimated by population-weighted health-risk-based air quality index (Shen et al., 2017), or human health-related impact score IS (Jolliet and Fantke, 2015). 2.5 Carbon footprint and large scale impacts of district energy systems Fossil fuels used for energy production are seen by scientists as the main source of GHG emissions so their replacement with cleaner and renewable energy sources is a world-wide policy approach. Among renewables, biomass is an attractive choice perceived as a natural carbon sink due to CO2 uptake by trees, a natural process known as carbon fixation (IEA, 2002). However, the replacement of fossil fuels by biomass may not be a simple-minded solution as CO2 balances depend on many factors such as:  a fossil fuel energy system being replaced versus a technology used for biomass conversion indicating the dependence of GHG emissions on the system efficiency (Schlamadinger and Marland, 1996; Schlamadinger et al., 1995). Moreover, forest growth rates and project time perspectives could be important factors influencing the overall net CO2 emissions which would be lower  in case of  long-term projects and high-efficiency of wood fuels substitution compared to just storing carbon in standing trees (Schlamadinger and Marland, 1996). On the other hand, associated costs (Repo et al., 2015;  37  Levihn, 2014) and sustainability of bioenergy  determined mainly by the biomass type and growing location (Evans et al., 2010) could be foreseen barriers to consideration of bioenergy as a viable replacement alternative for fossil fuels. 2.5.1  Bioenergy and carbon neutrality discussions The notion of carbon neutrality of bioenergy arose from the fact that carbon emitted to the atmosphere as a result of biomass  burning would be offset by trees via absorption of CO2 but  that may lead to an error in accounting for carbon-based emissions (Haberl et al., 2012). One of the reasons lies in missing to account for carbon uptake by plants  which could have occurred had those plants not been harvested at all (Cambero and Sowlati, 2014), or  not accounting for carbon loss due to harvesting of available residues (Repo et al., 2015).     Carbon neutrality of biomass is widely used in literature with a very broad meaning (WBCSD, 2015) such as “life-cycle neutral biomass” representing long-term stored atmospheric carbon which is equal to or greater than emissions associated with the use of such biomass over the entire life cycle. Similarly, “carbon-cycle neutral biomass” refers to biomass for which estimated emissions of biogenic carbon to the atmosphere are completely offset by new growth.  Reviewed literature suggested inaccuracy of  the immediate assumption of forest biomass carbon neutrality (Röder et al., 2015; McKechnie et al., 2011),  as it is a time dependent parameter since forest carbon stocks or sinks, such as soil carbon stocks, could be  reduced over time (Vanhala et al., 2013; Holtsmark, 2013).  In case of wood residues, forest carbon stocks would not be impacted if the rate of harvesting (carbon removal) is equal to the rate of residue decomposition (in case residues were not removed) (McKechnie et al., 2011). Hektor et al. (2016), suggested that 38  assessing carbon neutrality should be performed on the actual case values since the outcome, i.e. CO2 emissions, will largely depend on factors such as: whether biomass originates from sustainably managed forests, biomass characteristics such as moisture content, and applied conversion technologies, all of which can lead to biomass being characterized as both carbon and climate neutral. Another recently published analysis (Nabuurs et al., 2017), which considered realistic case of European sustainably managed forests, pointed out that the use of woody biomass for energy did not reduce large scale average forest carbon stocks but caution should be taken for future estimates due to possible natural disturbances. Among all, carbon debt for removal of harvested residues is the fastest one to be compensated, within a decade.  Appendix A.2 summarizes reviewed studies on carbon neutrality.  2.5.2  GHG emission estimates A comprehensive evaluation of bioenergy benefits with respect to GHG emission reductions should entail emissions evaluation across the entire biomass supply chain, including GHG balance and carbon sinks estimates (van Dam et al., 2010)  for  carbon footprint calculations (Levihn, 2014). Processes such as biomass recovery and removal require energy so such processes contribute to CO2 emissions (Gustavsson et al., 2011).  Furthermore, there are some emissions related to biomass storage so neglecting that fact  GHG savings, utilizing forest biomass in combined heat and power (CHP) and heat production only could be over-rated and be as high as  98% (Jäppinen et al., 2014).  Harvesting scenarios are another parameter causing calculated carbon footprint for wood products to vary widely (Newell and Vos, 2012).  It is worth noting that in comparison to fossil fuels such as coal,  GHG emissions reduction of 83% could be achieved if wood pellets are used instead (Röder et al., 2015). This is especially true for a long-time horizon such as a 100-year 39  period when a significant decrease of 41 Mt of CO2eq was estimated by a study considering wood pellets replacing coal (McKechnie et al., 2014). However,  when accounting for storage emissions, dry matter losses in the supply chain or other biomass-related emissions, it may turn out that pellet co-firing or large-scale biomass electricity generation exceed GHG emissions compared to coal-fired electricity generation, when storage exceeds the period of 4 months (Röder et al., 2015). More specifically, this study claims that still is little known about methane emissions from wood stockpiles and recent studies came out with a large range of results for CH4 emissions, from negligible to over 60%. Drying options also highly influence GHG emissions when fossil fuels are used instead of biomass as a drying fuel.   The importance of biomass feedstock choices is emphasized by a study focusing on climate change mitigation options (Giuntoli et al., 2015). While current CO2-approach is widely used in biomass-related LCA studies for global warming assessments, the authors demonstrated that the impact of bioenergy could be assessed by other parameters such as surface temperature changes and other climate forces. The study concludes that the rate of surface temperature increase by the end of the century as a result of biomass use will depend on the decay rate of the residues used among other parameters. Long-term bioenergy production from a slow-decaying wood will not contribute to climate change mitigation compared to natural gas, unless the biomass residues with the decay rate above 2.7% per year are chosen as feedstock for energy (heat) production. Overall, as for the first decade of CO2 emissions,  similar impacts from biomass and fossil fuels could be noticed but  CO2 emitted from bioenergy use stabilizes over time (Cherubini et al., 2013).  40  Chapter 3: Integrated impact assessment approach to evaluate community-based district heating systems12  3.1 Introduction Systematic review of the literature indicated knowledge gaps in proper environmental assessment of growing community-based district energy systems. There is a need for improving current methods for adequately evaluating impacts of DES at a much smaller spatial and temporal scale than it was commonly used for large, remote power plants.   Impact assessment must therefore integrate different methods: ones that can adequately and more accurately address process impacts on local temporal and spatial scale (short-term local air quality and immediate community exposure) and methods adequate for addressing impacts on large spatial and temporal scale such as climate change (IPCC 5th Report, 2015) and overall human health.  Thus, approaches used to design, propose, justify and apply a comprehensive state-of-the-art methodology for evaluating biomass-based district energy systems in a community setting are presented in this chapter. In addition to explaining micro-climatological characteristics and their importance for the pollutant dispersion close to a pollution source, local impact assessment method uses data for only one (BRDF boiler) stack with an electrostatic precipitator (ESP) for particle control, and one type of pollutant, particulate matter with diameter less than 2.5 micrometers (PM2.5), over a period of one month to investigate the effect of                                                  12 A version of this chapter was published. Petrov, O., Bi, X., & Lau, A. (2015). Impact assessment of biomass-based district heating systems in densely populated communities. Part I: Dynamic intake fraction methodology. Atmospheric Environment, 115, 70–78. https://doi.org/10.1016/j.atmosenv.2015.05.036. 41  dynamic variations of population and micrometeorological conditions on iF. One month data are found to be representative for the purpose of the method development as it takes into account 720 hours of measured wind data from a local surface weather station in addition to modeled prognostic meteorological data MM5 (Fifth-Generation Penn State/NCAR Mesoscale Model), actual plant operating parameters and actual campus population data.  Methods presented in this chapter provide foundation for integrated impact assessment of the Bioenergy Research and Demonstration Facility (BRDF) selected as a case study and carried out in subsequent thesis chapters.  3.2 Methods  This research study utilized quantitative research methods. Methodologies utilized included:   Collecting and analyzing secondary data from the records available at PH and BRDF, professional reports prepared for both plants, permit for BRDF, GIS-based campus planning data for building use, occupancy, building locations and dimensions;  Collecting and analyzing local meteorological data to determine the impacts of locally induced circulation patterns important in the dispersion of pollutants;  Site visits data collection and data processing for an in-house district heating life cycle inventory database;   Improving  methodological approach for assessing the impact of community-based DES by introducing site-specific parameters and applying mathematical modeling and mapping, as well as using dispersion and GIS software packages such as WRPLOT View™, CALPUFF View™  and ArcGIS 10.1;  Statistical analysis of data with inclusion of uncertainties of data.42  3.3 Local air quality assessment methodology 3.3.1. Microclimatic conditions and diurnal circulation patterns With emission sources being located in close proximity to people, local microclimatic diurnal variations play a pivotal role in accurate evaluation of population exposure.  Coupled with diurnal population density dynamics, local air circulation patterns could result in different exposure patterns and consequently different iF during day and night. Coastal areas, such as the UBC Vancouver campus which is located on the Pacific coast and surrounded by the Pacific Spirit Regional Park, are subject to pronounced diurnal variations in wind patterns due to different heating capacities of land and water (Trenberth and Stepaniak, 2004) resulting in wind mostly blowing from the ocean towards land (sea breeze) during the day and from land towards the ocean (land breeze) at night. Such microclimatic conditions as well as their impacts on in-land and orographically induced circulation patterns were well documented in the literature (Fock and Schlünzen, 2012; Azorin-Molina et al., 2011; Buckley and Kurzeja, 1997; Lu and Turco, 1994).   One year daytime and nighttime wind data from six Metro Vancouver surface monitoring stations (Doerksen, 2012) were analyzed  to demonstrate differences in day and night wind patterns and impacts of orographic features which cause upslope (daytime) and downslope (nighttime) circulation.  As presented in Figure 3.1, stations located at the coastline, Horseshoe Bay (T35) and Vancouver International Airport (YVR) (T31), are characterized with prevalent land-breeze circulation during nighttime and sea-breeze circulation during daytime periods. 43  T35 Horseshoe Bay   Nighttime hours only                                     Daytime hours only  (prevailing N and NNE winds)                       (prevailing SW and WSW) T31 YVR  Nighttime hours only                                                Daytime hours only  (prevailing E and ENE,ESE winds)                prevailing NW, WNW and E to SSE) T17 Richmond  Nighttime hours only                                              Daytime hours only   (prevailing E and ENE,ESE winds)        (prevailing NW, WNW and SE,SSE) T 14  North Burnaby      Nighttime hours only                                       Daytime hours only    (prevailing ESE winds)                            (prevailing winds from the SW quadrant)     T12 Chilliwack Airport Nighttime hours only                                              Daytime hours only (prevailing  winds from the NE quadrant)          (prevailing SW, WSW)  T6 Second Narrows     Nighttime hours only                                     Daytime hours only (prevailing winds from the NE quadrant)      (W-E circulation, channeling effect)  Figure 3.1 Wind patterns for day and night periods at selected Metro Vancouver stations.            44  Other stations located at some distance from the shore, Chilliwack Airport (T12), Richmond South (T17), and North Burnaby (T14) at Simon Fraser University (SFU) located at the elevation of 360 m above the sea level, also demonstrated pronounced differences in day and night wind patterns. Second Narrow station (T6) shows the channeling effects caused by daytime circulation influenced by narrow Burrard Inlet situated between mountainous north shore and mainland Vancouver while mountain breeze dominates nighttime circulation.    One month of wind data from the UBC Totem weather station were analyzed to investigate prevailing winds in terms of day-night characteristic wind patterns as hypothesized due to the unique campus location. Wind roses (Figure 3.2) were prepared for a total of 360 daytime hours (8 am to 7 pm) and 360 nighttime hours (8 pm to 7 am) for September 2012 using WRPLOT View™ software from Lakes Environmental (Lakes Environmental, 2012a).         Figure 3.2 Wind rose for daytime (left) and nighttime (right), September 2012, UBC Totem weather station.  Over 77% of daytime hours winds were blowing from the ocean, with prevailing winds 20.3% of time from WNW (west-northwest), 16.9% of time from NW (northwest), and 11.9% of time 45  from each W (west) and WSW (west-southwest) directions.13  Only 20.6% of daytime hours wind was blowing from the north-east quadrant, i.e. from land towards the ocean.  Calms comprised less than 2% of daytime hours.  Nighttime circulation patterns showed the opposite trend. While 71.3% of nighttime hours winds were blowing from land towards the ocean,  predominantly from east (E), east-northeast (ENE) and northeast (NE), with 19%, 16.5% and 13.5% of total nighttime hours respectively,  winds coming from the ocean (the north-west quadrant) were recorded only 23.9% of time. During nighttime, calms were recorded for 3.9% of time. Following the results of this analysis, modeling scenarios were designed to incorporate diurnal wind circulation dynamics to evaluate its effects on iF estimates. 3.3.2. Dispersion modeling: CALPUFF modeling system Dispersion modeling is a convenient approach to evaluate ambient concentrations of emitted pollutants from a source or multiple sources for a variety of purposes. It is becoming a crucial tool in decision-making processes about population exposure, health impacts and environmental justice (Borrego et al., 2015; Maroko, 2012).  This method is commonly used for planned pollution sources to evaluate potential impacts before facility  construction (Vallero, 2014), for modification of existing sources (Todorovic et al., 2015), for evaluating atmospheric fate of a particular pollutant (Holmes and Morawska, 2006) including model validation with observational data (Abril et al., 2016), for urban scale modeling (Pepe et al., 2016) or near-field modeling in urban areas (Tominaga and Stathopoulos, 2016).                                                    13 Wind direction is in meteorology defined as the direction wind is blowing from; eg. “Northerly winds” implies that wind is blowing from north towards south. 46  CALPUFF View™, version 6.4 (Lakes Environmental, 2012b),  a multilayer, non-steady-state Lagrangian Gaussian puff dispersion model, was used in this study to estimate ambient concentrations at different receptors on campus. It is a preferred and verified regulatory model in the United States (US EPA: SCRAM, 2015) and BC (BC MoE, 2015; BC MoE, 2008).  CALPUFF has the capability to cover a large spatial domain with a high resolution to capture microclimatic and atmospheric characteristics conducive to dispersion (Greco et al., 2007), particularly important in urban areas with non-homogenous conditions (Fisher et al., 2005). CALPUFF is  suitable for cases of complex terrain and coastal circulation effects and it has previously been used in iF studies (Curci et al., 2012; Zhou et al., 2006; Zhou et al., 2003; (Jonathan I Levy et al., 2002).    The basic equation for a puff model that connects emitted pollutants with the ambient concentration at a receptor (Scire et al., 2000; Schnelle and Dey, 2000) is: C = 𝑄2𝜋 ϭ𝑥ϭ𝑦     𝑔 exp  [−𝑑𝑎2 /(2ϭ𝑥2)] exp  [−𝑑𝑐2 /(2ϭ𝑦2 )]     (3-1)  With “𝑔” being expressed as:   𝑔 =  2(2𝜋)1/2  ∑ exp[−(𝐻𝑒 ∞𝑛=−∞ + 2𝑛ℎ)2/(2ϭ𝑧2)]        (3-2)  Where: C is the ground-level pollutant concentration [g/m3] per the distance [m] traveled by the puff,  Q is the mass of the pollutant in the puff [g], ϭx is the standard deviation of the Gaussian distribution in the down-wind direction [m], ϭy is the standard deviation of the Gaussian distribution in the cross-wind direction [m], ϭz is the standard deviation of the Gaussian distribution in the vertical direction [m], 47  dc is the distance from the puff center to the receptor in the cross-wind direction [m], g is the vertical term in the Gaussian equation [m], He is the effective height above the ground of the puff center [m], and h is the mixing-layer height [m].  Major CALPUFF features include: possibility of modeling constant or variable emissions for all types of sources (point, volume, area, line); gridded 3-D meteorological fields, vertically and horizontally-varying turbulence and dispersion rates, rural and urban, stability-dependent dispersion coefficients, building downwash effects, plume rise, dry deposition and wet removal, chemical transformation options etc.        The main components of the CALPUFF modeling system in addition to a large number of preprocessing programs are: 1. CALMET – a meteorological program which develops wind and temperature fields within  a three-dimensional modeling domain; 2. CALPUFF – a model which simulates  dispersion of emissions as “puffs” based on spatial and temporal variation of generated meteorological fields by CALMET, producing hourly concentrations or hourly deposition fluxes at selected receptors; 3. CALPOST – processes obtained data, produces tables and identifies the highest and second highest concentrations, produces graphical representations of results such as contours, lines connecting locations with the same values of pollutants. Modeling domain in an initial run was selected to cover an area of 2.6 km x 4 km around the emission source, BRDF boiler stack (EN 02), which was selected to be a reference point for modeling. The domain extended 2 km in each of directions to the north, south and east from the 48  plant and only 0.6 km to the coast at the west (toward the ocean). The selected domain (Figure 3.3) ensured coverage of required receptors (campus buildings) without producing output pollutant concentrations over the ocean since population located on campus was the subject of this analysis.   Figure 3.3 Nested grid receptors, red rectangular depicts an area with removed receptors due to absence of population.   3.3.2.1.  Model input data   CALPUFF modeling system (specifically, CALMET processor) requires the following meteorological data input: hourly surface observation of wind speed and direction, temperature, cloud cover, ceiling height, surface pressure, relative humidity and precipitation (optional). Meteorological input consisted of one hour prognostic MM5 (Fifth-Generation Penn State/NCAR Mesoscale Model) data for 2012 – 2013 prepared by Lakes Environmental, with 50 km x 50 km coverage, 4 km resolution and 11 vertical layers. This initial modeling scenario was 49  carried out only for September 2012. Meteorological grid for CALMET was set at 12.5 km x 12.5 km with 250 m spacing. In addition, data from stations presented in Table 3.1 were analyzed for supplementary entries.   Table 3.1 Surface and upper-air weather stations considered in the study. STATION NAME STATION ID CITY STREET ADDRESS Latitude Longitude Elevation (m) UBC Totem 1108487 Vancouver           Point Gray  49˚15’23.68”N 123˚14’59.92”W 76   decimal 49.26 -123.25    UTM coordinates (X,Y) (m) 481811 5456007  Kitsilano T2 Vancouver 2550 W 19th Ave 49˚15’35.99”N 123˚9’35.99”W 63   decimal 49.26 -123.16    UTM coordinates (X,Y) (m) 488360 5456368  YVR T31 Richmond 3153 Templeton St 49˚11’23.99”N 123˚9’0”W 10 (upper-air)  decimal 49.19 -123.15        UTM zone10, coord. (X,Y) (m) 489070 5448585   YLW 71203 Kelowna Airport  49° 58' 11.99"N 119° 17'59.99"W 454                                            decimal 49.97 -119.30    UTM zone 11, coord. (X,Y) (m) 335073 5537827   Ceiling height and cloud cover data were obtained from METAR14 weather data from Vancouver International Airport (YVR).  Vertical atmospheric data were obtained from twice-a-day sounding data at Kelowna Airport and surface meteorological data from the UBC Totem station. Kitsilano station data were analyzed for comparison purposes.  Terrain data were obtained from GeoBase database assessable through the CALPUFF View™ software (Lakes Environmental, 2012b). Canadian digital elevation data for region 92g were selected with coverage of 1:50,000 (Natural Resources Canada, 2012). In addition, the 1-Degree blocks DEM (Digital Elevation Model) data from WebGIS database for U.S. and Canada were                                                  14 METAR weather data format is mostly used in aviation by pilots and standardized by ICAO (International Civil Aviation organization); available from: http://vortex.plymouth.edu/statlog-u.html. (Accessed January 3, 2018). 50  used for obtaining terrain elevations. Land use data (LULC) were obtained from Global Land Cover Characterization (GLCC) system for North America with 1 km mesh coverage.   Receptors, defined as anything of a value in the environment impacted by pollutants, were selected to be people at campus building locations. Over 500 entries containing: building ID, name, maximum occupancy, geographic coordinates, and heights were provided by the UBC Campus and Community Planning Department which were used to classify buildings into:   Work-related, where residents, students, faculty, and staff reside during their work on campus. A total of 160 (out of 191 existing buildings) with classrooms, labs, administrative and academic offices were included as daytime (8 am to 7 pm) receptors, and  Residences, classified as apartment buildings, high risers or townhouses were separated by individual dwellings resulting in 214 buildings. The occupants of those buildings were receptors during nighttime (8 pm to 7 am) but also during daytime as some residents would likely stay in those building.   A total of 374 campus buildings occupied by people at some point during days and/or nights were considered in this study. Buildings which are under development and/or for which data were not complete were excluded from this analysis. Building parameters for each selected building were entered in excel spreadsheet and used as discrete receptors in CALPUFF modeling and in iF calculations.   Another set of input data included source, i.e. stack parameters as presented in Table 3.2. Boiler stack filterable particle emissions were calculated as an average of 4 replicate emission tests 51  conducted by a third party on July 17, 2012, following the procedures recommended by the BC Ministry of Environment (MoE, 2003).   Table 3.2 BRDF Source parameters. Source ID Description (Stack) Height [m] Diameter [m] Exit T [K] Exit velocity [m/s] Emission rate [g/s] Emission rate [kg/day] PM PM EN-02 Boiler with ESP Measured  20 0.76 477 8.43 0.028  2.419  Reported  Gas exit velocity was calculated as the ratio of the measured flow rate and stack cross-sectional area. The emission rate was calculated as a product of the measured flow rate and measured concentrations previously corrected to 8% O2 as per permit. It was assumed that the emission rate was constant throughout the month selected for modeling, September 2012. 3.3.2.2.  Model output data: ambient pollutant concentrations   CALPOST was set to produce output data as 1-hour and 24-hour average ground-level concentrations at each receptor expressed in micrograms per cubic meter (µg/m3).  Obtained data were imported from model output text files and organized in excel spreadsheets per modeling and iF assessment scenario (described in section 3.4) and for all considered receptors.  All 1-hour data were then separated in daytime and nighttime periods. Mean, maximum, and minimum values were calculated for every daytime and every nighttime period. Summary results were organized in tables to enable comparison with AQO and to be added to background pollutant levels.  In addition, mean values for each scenario and per receptor and daytime/nighttime period were imported in a separated spreadsheet to be used for iF calculations and overall campus pollutant levels and exposure scenarios. Results for September 2012 for PM2.5 are summarized in Table 3.3 below. 52  Table 3.3 Ground-level PM2.5 concentrations, UBC campus, September 2012. Parameter All 374 receptors Daytime 374 receptors  Nighttime 214 receptors  Averaging period 24-hr  PM2.5[µg/m3] 1-hr  PM2.5 [µg/m3] 1-hr  PM2.5 [µg/m3] Mean 0.015 ± 0.031  0.019 ± 0.010 0.012 ± 0.020 Max 0.264 0.230 0.169 Min 0.002 0.001 0.002  Analyses of 1-hour PM2.5 ambient concentrations across campus showed that nighttime mean concentration of PM2.5 were 38% lower than daytime mean and 23% lower than 24-hour average. The maximum nighttime PM2.5 1-hour concentration (0.169µg/m3) was only 27% lower than daytime maximum 1-hour PM2.5 concentration (0.230µg/m3) but 36% lower than maximum PM2.5 concentrations for a 24-hour averaging period. Emission rate is relatively constant for current plant operating conditions. However, due to varying meteorological conditions (other parameters were kept constant), daily dispersion patterns varied. 3.3.3. Ambient air quality regulation and background pollutant levels  The Province of British Columbia adopted more stringent ambient (outdoor) air quality criteria for PM2.5 in 2009 (BC MoE, 2016) due to their harmful potential to human health.  A 24-hour objective is set at 25 µg/m3 while annual objective is set at 8 µg/m3 with planned target of 6 µg/m3.  Table 3.4 summarizes BC Air Quality Objectives (BC AQO) and Canadian Ambient Air Quality Standards (CAAQS) for particles and gases considered in this study as relevant to biomass emissions (BC MoE, 2016).   53  Table 3.4 Summary of provincial Air Quality Objectives (AQO) and Canadian Ambient Air Quality Standards (CAAQS) for selected contaminants. Contaminant Averaging Period Criteria Level Air Quality Objective Date Adopted µg/m3 ppb Carbon Monoxide (CO)   1hour PCOs for Food processing, Agriculturally Orientated, and Other Misc. Industries A B C 14,300 28,000 35,000 13,000 25,000 30,000  1975  8 hour PCOs for Food processing, Agriculturally Orientated, and Other Misc. Industries A B C 5,500  11,000 14,300 5,000 10,000 13,000  1075  Nitrogen Dioxide (NO2) 1 hour Interim Provincial AQO Provincial AQO - - 188 200 100 a 2014  Annual Interim Provincial AQO - 60 32 2014 PM2.5 24 hour Provincial AQO - 25 b - 2009 CAAQS - 28 c - 2013 Annual  Provincial AQO     AAQO Goal 8 6 - - 2009 2009 CAAQS - 10 d - 2013 a Achievement based on annual 98th percentile of daily 1-hour maximum, over one year b Achievement based on annual 98th percentile of daily average, over one year c Achievement based on annual 98th percentile of daily average, averaged over three consecutive years d Achievement based on annual average, averaged over three consecutive years  3.4 Population exposure and health risk assessment methodology 3.4.1 Dynamic intake fraction (iF) Inhalation iF, representing a single-medium approach, is calculated as the portion which is being inhaled by exposed population (Nishioka et al., 2005; Jonathan I. Levy et al., 2002): iF = { ∑ ∑ [𝑃𝑖,𝑗  x 𝐶𝑖,𝑗 x 𝐵𝑅𝑖]} /𝑄𝑖 }𝑛𝑗=1𝑚𝑖=1                                                                   (3-3)  Where:  Qi  is the emission rate of a pollutant [kg/day] in a given time period i [hours] at a geographical area or location j; measured or calculated and as presented in the previous section, 54   Ci,j  is the ambient air pollutant concentration [mg/m3] in time period i at receptor location j; these concentrations were obtained from CALPUFF modeling, Bri  is the breathing rate [m3/person/day] during time period i, and  Pi,j  is the number of people at a specific location and time.   Input values used for iF calculations Emission rates were obtained by measurements performed at the active EN02 stack (Table 3.2) and ambient concentrations at each receptor were obtained from the CALPUFF modeling.   Breathing rate used in previous exposure-related studies was mostly averaged values and uniform for the population considered (as previously presented in Table 2.1).  Some studies (Marshall et al., 2006) suggested 13 m3/person/day for males and females combined (11.3 m3/day for women, 15.2 m3/day for men) based on age and activity.  Most recent Exposure Factors Handbook (US EPA National Center for Environmental Assessment and Moya, 2011) suggested a breathing rate of 14.6 m3/person/day which was estimated as the mean breathing rate for free-living normal-weight males and females combined, between 21 and 31 years old, which corresponded to the majority of UBC campus population. The same handbook suggested a long-term breathing rate of 15.7 m3/person/day for the same category but based on the unweighted average of means from combined key studies. For short-term breathing rate, a person’s activity was taken into consideration. For the “sleep or nap” activity, a mean breathing rate is 0.258 m3/person/hour whereas for the “light-intensity” activity a mean breathing rate is 0.72 m3/person/hour. In this study, improvements were made by separating daytime and nighttime breathing rates which led to more accurate estimates. For the daily breathing rate (dBR) in scenarios where only daytime was considered, 8.64 m3/person/12 hr-daytime (0.72 55  m3/person/hour x 12 hours) was used. Similarly, for the nightly breathing rate (nBR) in scenarios where only nighttime was considered, 3.096 m3/person/12 hr-nighttime (0.258 m3/person/hour x 12 hours) was used. In scenarios where daily iF was calculated over a 24-hr period, the breathing rate (BR) was 11.74 m3/person/day as a sum of daytime and nighttime breathing rates (8.64 m3/person/12 hr-daytime + 3.10 m3/person/12 hr-nighttime).   The number of people at a specific location and time is another parameter directly related to iF which highly influences the intake fraction value. In general, a noticeable variation of population at university campus is due to a larger number of people working or attending classes during daytime versus  a considerably lower number of people residing on campus during nighttime, which could have a significant impact on iF compared to a large city with relatively stable population density (Marshall et al., 2005a). Based on a widely varying criteria (Humbert et al., 2011), UBC, with population between 1500/nighttime and more than 4,000/daytime people/km2, could be characterized as a densely populated urban area.   It was assumed that people would mostly be in work-related buildings and some in residences during daytime but only in residences (for those who live on campus) during nighttime. Time spent while commuting between buildings and exposure duration at locations along the routes were not included since it is considered to be negligible compared to the time spent at certain locations. Exposure potential was evaluated based on the exposure on ambient (outdoor) PM2.5 concentrations, as indoor exposure is beyond the scope of this study. The estimated iF are thus expected to represent the upper limit of the actual values.  56  The number of exposed people is directly related to iF as expressed by equation 3-3. It was assumed that the exposure concentration is equal to the outdoor concentration in each of the two campus-related micro-environments, residential and daytime work-related buildings, while the exposure concentration is equal to zero while people are not on campus but rather in another micro-environment not affected or negligibly affected by the BRDF. Since the indoor fraction of ambient pollutant concentrations is generally lower than the outdoor concentrations (commonly used infiltration factor is 0.7 for PM2.5) but will depend on the ventilation system and building age (Zhou and Levy, 2008), iF calculated as presented, is expected to be higher than the actual value or, in other words, iF from this study is a more conservative version.   All considered buildings were associated with corresponding number of people reported as the maximum building occupancy. Where maximum occupancy is given as a total number of residents in a housing complex, the number of people per building was disaggregated to be uniformly prorated, meaning that an equal number of residents is allocated to each building. As it was assumed that all people were in those buildings (attending classes, working, living) most of the time, occupancy of on-campus restaurants and museums was not considered to prevent double counting of campus population. Temporary workers or visitors to UBC campus were not taken into account as there is no record of such numbers and it is assumed that such number is negligible compared to regular campus inhabitants.   After assigning an appropriate number of people to each identified building, 16,406 persons were considered as the number of campus residents associated with 214 residential buildings occupied during nighttime. During daytime, estimated 49,256 people are distributed in 374 buildings out 57  of which 160 are academic buildings with classrooms and labs and administrative offices (maximum occupancy minus 15% campus residents who are assumed to stay in residences during day) resulting in 46,795 persons and 214 are residential buildings which were assumed to be still 15% (2,461 persons) occupied during daytime. For a 24-hr averaging period, the number of persons on campus was calculated as an average of daytime maximum occupancy and nighttime maximum occupancy proportionally distributed in all 374 buildings (Table 3.5).   Table 3.5 UBC Campus population distribution as a function of diurnal dynamics. Building/ Occupancy/period Residential buildings Academic buildings and offices  No of buildings 214 160 Max occupancy 16,406 49,256 Day-time  occupancy 2,461(15% occupancy) 46,795 Night-time  occupancy 16,406 0 Average 24-hr Occupancy 32,831 (8,203 in residences AND 24,628 in academic buildings)    Scenarios and resulting iF values To evaluate the impacts of space, time, population density and breathing rate variations on the estimated iF, five scenarios were considered: Scenario 1: Base case – All averaged. No spatial, temporal or population dynamics was considered but only average values were used for all relevant parameters. This represents a typical box-model approach widely used in the past in dispersion modeling for health impact assessments. While performing dispersion modeling to obtain the ambient PM2.5 concentrations, a nested receptor grid was set in a way to place receptors equally spaced at 50 m over the modeling domain of 2 km around the source. A total 4,859 receptors were included in modeling 58  while 1,701 receptor sites were removed afterwards as they were over the ocean with no human exposure.   Obtained 24-hour average ground-level concentrations at each receptor were averaged over the entire campus area resulting in an overall average concentration of 0.01 µg/m3.  With 32,831 people being present on campus on average over 24 hours, iF was calculated to be 1.59 mg inhaled per kg emitted particles or 1.59 ppm (parts per million in mass).  Scenario 2: Spatial dynamics of receptors.  In this scenario (Figure 3.4), spatial dynamics of receptors was introduced while other parameters remained as in scenario 1. All 374 buildings inhabited by campus population were entered in the model as discrete receptors and 32,831 persons were equally distributed to each building. Obtained average 24-hour ground level PM2.5 concentrations for each receptor for the month of September 2012 were used to calculate iF for each receptor, with the results then plotted using ArcGIS software. The sum of iF for all receptors on the whole campus was found to be 2.28 ppm, based on values ranging from 0.0008 to 0.1115 ppm per receptor.    59    Figure 3.4  Scenario 2: iF for each building for September 2012.        Indicates buildings.     Scenario 3: Spatial and population dynamics. In addition to the spatial distribution of population in different buildings across the campus, this scenario (Figure 3.5), took into consideration population dynamics by assigning the actual number of people per building during the 24-hour period: 8,218 in residences and 24,628 in academic buildings, with 32,846 people in total. Model output, average 24-hour ground level PM2.5 concentrations were then used to calculate iF for each receptor with different number of occupants per building according to building capacity. The sum of iF for all receptors, indicating iF for the whole campus, was 1.77 ppm, based on values ranging from 0 to 0.113 ppm per receptor.     60           Figure 3.5 Scenario 3: iF for each building with actual occupancy, September 2012.  Comparison of scenarios 2 and 3 clearly indicated the importance of considering population dynamics and actual number of people at a certain location when calculating iF, resulting in a  22% lower iF than in scenario 2 but still 11.3% higher than iF obtained for a static model in scenario 1.  Scenario 4: Spatial, population and temporal dynamics. This scenario further introduces temporal dynamics to account for diurnal variations in meteorological parameters and significant diurnal campus population dynamics. This scenario considered separately daytime and nighttime periods when calculating iF. For the day-time period, a total of 374 buildings (160 work-related buildings, 214 residences) with 49,256 persons distributed as per each building’s actual occupancy were entered in the model as discrete receptors. Similarly, a total of 214 residential buildings with 16,406 people representing actual occupancy per building were used in the model as discrete receptors for nighttime period calculations.   61  Model time averaging period was set to 1–hour, and 1–hour average PM2.5 ground level concentrations at each receptor from the model output were used for iF calculations over the daytime hours and the nighttime hours, separately. Since daytime presented a 12-hour period, the breathing rate (which was not varying in this scenario) was accordingly adjusted to 5.87 m3/person/12 hr by dividing daily breathing rate of 11.74 m3/person/day by two. The same value for breathing rate of 5.87 m3/person/12 hr was used for night-time calculations. The same period adjustment was done for emission calculations resulting in 1.21 kg of PM2.5 /12 hours. iF was calculated for each receptor and summed up as per equation 3-3. for the daytime period resulting in iF being 2.19 ppm and the nighttime period when calculated iF was 1.18 ppm or almost half of daytime iF (Figure 3.6).  Figure 3.6 Scenario 4: a) daytime iF and b) nighttime iF for each building with actual occupancy for September 2012.  This scenario demonstrated the strong influence of day vs. night conditions. Taking an average of those two values gave a daily (24-hour) iF of 1.69 ppm which is only 6.2% higher than an 62  average iF indicating some disadvantages of solely using averaging values where parameters such as diurnal population dynamics significantly vary over the averaging period. Daytime and   nighttime variations of iF per date are presented in Figure 3.7.  Figure 3.7 Scenario 4: daytime (upper graph) and nighttime (bottom graph) variations of iF for September 2012. 63  Scenario 5: Spatial, population and temporal dynamics, varying BR. In the final scenario, (Figure 3.8), breathing rates for day and night were used, with 8.64 m3/person/12 hr for the daytime and 3.10 m3/person/12 hr for the nighttime as previously explained. iF was calculated for each receptor, resulting in a total daytime PM2.5 iF of 3.23 ppm and 81% lower iF for nighttime (0.62 ppm), a strong indication of the significance of diurnal variations in parameters used in iF calculations. Subsequently, iF for a 24-hour period, calculated as an average of daytime and nighttime iF, was 1.93 ppm or 21.4% higher than in scenario 1. Figure 3.8. Scenario 5:  UBC campus iF for September 2012 distinguishing day vs night periods with spatial and temporal dynamics, and varying BR.  64  Results from all five scenarios are presented in Table 3.6.   The iF results from the five modeled scenarios tend to emphasize the importance of introducing high resolution spatial, temporal and population dynamics along with varying breathing rate in the assessment of local health impact for DES systems located in densely populated communities.   Although fine resolutions of a range of parameters have previously been recommended by some researchers  (Xu et al., 2013;  Dhondt et al., 2012; Marshall et al., 2006), none of the studies so far has evaluated variations in iF for biomass-based district energy systems in community settings such as university campuses with high spatial and temporal resolutions and population dynamics. It was demonstrated here that the introduction of spatial dynamics (scenario 2) by replacing a nested grid of receptors (which represented a uniform receptor distribution as space-averaged receptors), with actual discrete locations of receptors, resulted in an overall increase in iF by 43%. Introducing day vs night, i.e. temporal dynamics (scenario 4), provided more accurate estimates of pollutant concentrations at receptors as a result of different day and night air circulations and consequent pollutant dispersion.  65  Table 3.6 Modeling scenarios and calculated iF and IS. SCENARIO Parameter SCENARIO 1  SCENARIO 2  SCENARIO 3  SCENARIO 4  SCENARIO 5  Spatial distribution     receptors/buildings  NO Nested  grid YES 374 daily YES 374 daily YES 374  Daytime   214  Nighttime YES 374  Daytime    214  Nighttime Population dynamics  NO 32,831 people  NO 32,831 people  (uniformly prorated/building) YES 32,831 people   (actual occupancy)  YES 49,256 Daytime      16,406 Nighttime  (actual occupancy) YES 49,256 Daytime        16,406 Nighttime  (actual occupancy) Temporal dynamics  Day/Night  NO 24-hr NO 24-hr NO 24-hr  YES 12-hr Daytime  12-hr Nighttime  YES 12-hr Daytime  12-hr Nighttime  BR dynamics Day/Night NO BR=11.74 m3/pers./day  NO BR=11.74 m3/pers./day NO BR=11.74 m3/pers./day NO BR=5.87 m3/pers./12-hr  YES 8.64m3/pers/daytime 3.10m3/pers/nighttime PM2.5 concentrations averaging period  24-hours   24-hours   24-hours   1-hour   1-hour  iF[mg/kg] = Σ iFI,jdaily= Σ iFI,jDaytime= Σ iFI,jNighttime=  1.59 n/a n/a  2.28 n/a n/a  1.77 n/a n/a  Av.=1.69 2.19 1.18  Av.=1.93 3.23 0.62 iF% change  from scenario 1  0  +43.0%  +11.3%  +6.2%  +21.4%   Additional consideration of different breathing rates for daytime and nighttime (scenario 5), demonstrated significant diurnal variations in iF by up to 81%.   More significant temporal variations are expected if the emission rate, which corresponds to the heat and power demands in full scale DES, also varies during day and night because of higher heating demand in the night during winter and higher hot water and electricity demand in the day during summer.  Calculated iF could further be used to estimate human health impacts resulting from a particular source, as presented by Humbert et al. (2011) and outlined in the next section. 3.4.2  Health-related Impact Score (IS) Two indicators, the dynamic intake fraction (iF) and human health-related impact score (IS) were used to evaluate the health impacts from emitted PM2.5 and gaseous pollutants in this study. These metrics relate the environmental fate of pollutants to the exposure, dose–response, and 66  severity of response (Michael Z.  Hauschild., Ed. and Mark A.J. Huijbregts.  Ed., 2015).  IS was expressed in terms of disability-adjusted life years [DALY], obtained from equation 3-4: (Michael Z.  Hauschild., Ed. and Mark A.J. Huijbregts.  Ed., 2015): 𝐼𝑆 = 𝑚 ∙ 𝑖𝐹 ∙ 𝐸𝐹health                              (3-4)  Where:  m  is the mass of emitted pollutant [g],   iF  is the intake fraction per pollutant [ppm or 10-6, µg inhaled/g emitted],  EFhealth is a human toxicological effect factor [DALY/kg inhaled]. EFhealth is 7.00E-04 DALY/kg PM2.5   for particulate matter, 7.31E-07 DALY/kg for CO and 8.91E-05 DALY/kg for NO2  (Quantis, 2012).  A human toxicological factor  is  obtained from the IMPACT2002+vQ2.22 database (Quantis, 2012), which includes the damage or adverse respiratory effects caused by inorganic substances (PM2.5, biogenic and fossil CO and NO2). The combination of iF and EFhealth results in a characterization factor (CF) which expresses the increase in the number of DALY per unit mass of a pollutant emitted into the atmosphere. It should be noted that the EFhealth values from IMPACT2002+ are based on observations relevant to the European countries, which may not represent the situations in North America very well. Therefore, interpretation should be rather focused on relative values among different scenarios. However, including site-specific iF and mass of emitted pollutants in the IS calculations increases accuracy in estimates.  67  3.5 Environmental footprint methodology   Carbon footprint and water footprint are the most common methodologies used to evaluate impacts in urban environments; however, other environmental compartments (such as air) also need attention. The Urban Metabolism (UM) is a  widely used concept in urban environments for evaluating flows of energy and matter in and out of cities but does not assess environmental impacts (Mirabella and Allacker, 2017). Life Cycle Assessment (LCA) is a powerful tool for evaluating all-encompassing environmental footprints of products and activities over their entire life (Hiloidhari et al., 2017). Thus, for bioenergy applications, LCA is widely used to compare impacts of all life stages to fossil fuel utilization and to determine if reducing fossil fuels and substitution with bioenergy can benefit societies.   As presented in the literature review of 94 LCA-related studies by Cherubini and Strømman (2011), half of the reviewed studies directed the assessment towards the climate change impact category by calculating GHG and energy balances but did not consider other impact categories.  A number of locally conducted studies outlined benefits and drawbacks of biomass utilization in Canadian context. For example, Pa and collaborators  (2012) analyzed emission and energy flows for wood pellet produced in British Columbia and estimated environmental footprints of production, conversion and export of such pellets.  The study found that 295 kg of CO2eq is released for every tonne of pallets produced in BC and exported. If such locally produced pellets find their application in BC to replace firewood, human health impacts could be reduced by 61%, ecosystem quality impacts by 66% and climate change impacts could be reduced by 53%.  68  Life cycle assessment (LCA) can be performed following the recommendations of ISO 140044:2006 (ISO, 2006). Schematically, as presented as in Figure 3.9, the LCA framework consists of four stages which are described in detail below.   Figure 3.9 LCA framework. Source: Based on ISO ( 2006).  3.5.1. Goal and scope definition During this stage, since it inherently involves subjectivity, it is very important to carefully define the goal and scope of the study to minimize user’s influence on results. The goal should: a) explicitly describe the application and intended audiences; for example, a study to be used only internally could be structured differently than the one publically available for which weighting step during the impact assessment phase is rather replaced with a peer review process; b) clearly describe the reasons for conducting the study, in other words, whether the study is just an informative one or it aims at providing a proof. The study can serve more than one purpose.  The scope of the study aims at describing methodological choices, assumptions, and limitations. The most important to clearly be defined are: a) Functional unit (FU) and reference flow – a Goal and scope definitionLife cycle inventory analysis (LCI)Life cycle impact assessment (LCIA)Interpretation of data and findings69  comparison basis which is often a difficult task due to different performance of products and/or services which need to be compared; b) initial15 system boundaries – since not all processes and products along the way need or have to be included so it is helpful to clarify what impacts the results; c) criteria for inclusion of inputs and outputs – which refers to the selection of a threshold below which an input or an output is not considered. ISO 14044 recommends using several criteria for such a threshold, for example, defining a percentage below which the mass inflow will not be accounted for; d) dealing with multifunctional processes – which happens when processes end up in more products; some of the  ISO 14044 recommends are system expansion or allocation which is applied in attributional LCA analysis. 3.5.2. Life cycle inventory (LCI) analysis This is the most demanding stage which includes data collection. Some secondary data are already available if a particular software like SimaPro is used or could be obtained from literature. Basically, two types of data need to be collected: a) background data - for the production of generic materials, transport, wastes and energy; b) foreground data – which refer to a specific product and/or a system that is being modeled. In both cases, data collection is a comprehensive process which encompasses: literature review, site visits, interviews, allocation considerations, data quality, confidentiality issues with data providers, etc.   3.5.3.  Life cycle impact assessment (LCIA) LCIA aims at understanding and evaluating the magnitude and significance of the potential environmental impacts of a product or a process over their entire life, either from “cradle-to-                                                 15 The term “initial” is often used as LCA is an iterative process. 70  grave” (from raw material extraction to waste disposal) or “cradle-to-gate“(from raw material extraction to the point of sales). ISO 14040/44 specifies that:  Classification and characterization are mandatory elements of the analysis, while  Normalization, ranking, grouping and weighting are optional elements. Characterization is about assigning an impact category to the elementary flows from the inventory. This process considers the substances’ ability to contribute to different environmental problems. For example, CO2 will have impact on climate change while CFCs will have impact on climate change and stratospheric ozone depletion. However, although CO2 and CFCs contribute to the same category, the magnitude of their impacts is different; in such case IPCC equivalency factors are applied (1 for CO2,  4,660 for CFC-11, for a 100-year time horizon ) (IPCC, 2013). Units of the results will be [kg CO2eq]. Similarly, other substances are dealt with during the characterization stage by applying appropriate characterization factors.   The ISO standard allows the use of impact category indicators that are either “midpoint impacts” or “endpoint impacts”. Generally speaking, indicators that are chosen close to the inventory results (midpoint) have a lower uncertainty but endpoint indicators are a favorable choice for  decision makers. Impact categories used in practice are presented in Figure 3.10.   71   Figure 3.10  Scheme of the impact categories dealt with in ILCD Handbook on Life Cycle Impact Assessment at midpoint and at endpoint. Source: Based on Sala et al. (2012).   A number of impact assessment methods, built on science-based environmental mechanism, have been developed over the years. One of the drawbacks of methods used with LCA methodology in general is that they are mostly developed for northern and middle Europe, the USA and Japan. 3.6 Conclusions The novelty of this work lays in the improvement of an impact assessment approach for the biomass district energy systems, which accounts for all local-scale variations, from actual population density (as opposed to averaged census data commonly used in assessments), to local spatial and temporal micro-climatological and local spatial orographical conditions at the MIDPOINT ENDPOINTClimate change Climate changehuman health/ecosystemsOzone depletion Ozone depletionhuman health/ecosystems Human healthHuman toxicity Human toxicitycancer/non-cancer cancer/non-cancerPM/respiratory inorganics PM/respiratory inorganicsIonising radiation Ionising radiationhuman health/ecosystrems human health/ecosystremsPhotochemical ozone formation Photochemical ozone formation Natural environmenthuman health/ecosystemsAcidification  terrestrial Acidification  terrestrialaquatic aquaticEutrophication aquatic Eutrophication aquatic (freshwater)/marine)terresrial terrestrialEcotoxicity: aquatic (freshwater)/marine) Ecotoxicity: aquatic (freshwater)/marine) Natural resourcesterrestrial terrestrialLand use Land useResource depletion Resource depletion Methodwater water recommendedmineral/fossil mineral/fossil No methodrenewable renewable availableMethod interimEnvironmental cause effect chainElementary flows72  biomass plant site.  It was shown that the proposed comprehensive methodology gave more accurate and realistic estimates of ambient concentrations at receptors stratified to follow the diurnal atmospheric processes which impact the pollutant dispersion, and consequently more accurate estimate of intake fraction accounting for dynamic variations of population and breathing rates.   When the dynamic variation of all parameters is accounted for (Scenario 5), the real dynamic nature of iF is captured. Neglecting microclimatic characteristics such as site-specific diurnal circulation patterns which influence pollutant dispersion or not considering short-term variation of parameters on a local scale such as population dynamics may lead to underestimation of iF by more than 20%. This amplifies the importance of incorporating both spatial and temporal dynamics in estimating the exposure (i.e. iF) in assessing the health impact of district heating systems in densely populated areas. These results confirm that this improved methodology could be generalized, i.e., applied to any source for which the impact assessment is sought in order to realistically evaluate the local impacts. The practical application of this methodology is presented in Chapter 4 for assessing local health impact and in Chapter 5 for assessment of global climate change impacts of biomass plants. 73  Chapter 4: Impact assessment of the UBC district heating system on local air quality and associated health risks 16  4.1 Introduction  The main objective of this chapter is to fill the knowledge gap about local environmental (ambient air quality) and social (human health) impacts of newly and rapidly developing DES in community settings. Analysis presented in this chapter first applies an improved assessment method established and tested  in Chapter 3 and then presents results obtained by evaluating local air quality and human health over five operational DES scenarios for a one-year period so to account for diurnal and seasonal variations of considered parameters. Operational scenarios were selected to feature the combined operation of BRDF and PH since BRDF commencement in June 2012 and then hypothetical, future scenarios in which the entire heating demand would be met by natural gas only or a scenario where the entire heating demand would be met by biomass (clean solid wood waste) only. Additional scenarios were introduced for sensitivity analysis to estimate the impacts of changing population dynamics and emission rates on output vales. Implications of such operational regimes on ambient air quality and subsequently on campus’ population exposure are discussed.                                                    16 A version of this chapter is published: Petrov, O., Bi, X., & Lau, A. (2017). Impact assessment of biomass-based district heating systems in densely populated communities. Part II: Would the replacement of fossil fuels improve ambient air quality and human health? Atmospheric Environment, 161, 191–199. https://doi.org/10.1016/j.atmosenv.2017.05.001. 74  4.2 District heating at UBC Point Gray campus   Energy for campus heating and hot water was generated exclusively by a PH boiler fired by natural gas (NG) at base load and supplemented by heating oil at peak load until June 2012 when a newly constructed biomass gasification plant BRDF became operational. The introduction of biomass was in line with the UBC’s initiatives to reduce GHG by 33%, 67% and 100% by 2015, 2020 and 2050, respectively, from the 2007 level (UBC, 2015d). The plant was designed as a CHP although it has been mostly operated in the thermal mode since commissioned, using commercially proven Nexterra gasification technology.  4.2.1. Thermal energy demand and supply profile The current plant operation in 2012/2013 was set as the base scenario. Hourly records of NG and fuel oil consumption as well as steam production were obtained for the 2009-2013 period. Daytime (8 am to 7 pm) and nighttime (8 pm to 7 am) data were separated for estimating diurnal fuel consumption and subsequently seasonal and annual fuel consumption.  The June 2009 - May 2010 period was then chosen as a typical year of PH operation for meeting campus thermal energy demand, while June 2012 - May 2013 was chosen as a period which marked the first year of BRDF operation using biomass to produce heat for a portion of campus. For the PH, NG consumption was recorded in thousand-standard cubic feet [KSCF] and oil consumption in thousand-pounds [KLBS]. Steam production was also recorded in KLBS. All processed values were converted to SI units and were presented along with Imperial units in places as needed.  To calculate total energy input from different fuels, NG consumption was multiplied by its higher heating value (HHV) (Bossel, 2003), of 39.11 MJ/Nm3  at normal/standard conditions 75  (1050 BTU/SCF), oil consumption was multiplied by its HHV of 46 MJ/kg, while steam produced (which represents energy output) was multiplied by HHV of 2.8 MJ/kg (1197 BTU/lb) at 1138 kPa (165 psig).  In 2009-2010, total energy input was 930 TJ, out of which 910 TJ was attributed to NG and 20 TJ to oil used only at peak load in winter season. A total of 884 TJ of steam was produced by the 3 boilers (mostly boiler # 5) in the PH with an annual average thermal efficiency of 95%. When the BRDF became operational, 823 TJ of steam produced was recorded as heat output by the PH boilers and 188 TJ as heat output from BRDF during the period of June 2012 - May 2013. Thus, almost 20% of total steam production was contributed by the BRDF. Fuel characteristics and consumption, and energy calculations are detailed in Appendix B.  4.2.2. Biomass supply requirements for fossil fuel replacement BRDF utilizes locally collected and preprocessed solid wood residues with an average moisture content of 35% wet basis (or 54% dry basis) and a HHV of 19.3 MJ/kg of dry wood (Cot, 2016) to produce steam at 68% calculated average thermal efficiency, based on steam produced. On average 7,711 kg/hr (reported as 17,000 lb/hr) or 67,549 t/yr (148,920 KLBS/yr) of steam is produced at the BRDF at the current capacity, implying that on average 17,475 tonnes of wood waste are annually utilized by BRDF. Wood consumption was in this study attributed equally to all periods throughout the year.    Calculations showed that if gasification of wood waste were to replace the combustion of fossil fuels at PH in order to produce 823 TJ of energy as presented for 2012-2013 input energy of 76  1,210 TJ or 96,569 t of wood waste would be required. Adding the estimated wood waste consumption required for the BRDF during the same time period, a total 114,043 t of biomass residues would need to be gasified to meet the campus energy demand. 4.3 Scenarios for evaluating options for district heating at UBC As presented in Table 4.1, five scenarios were considered, two of which served as sensitivity analysis.   Table 4.1 Summary of operational scenarios used in the DH impact assessment.  Operational scenario Fuel used Energy input [GJ] Energy output [GJ] Efficiency[%] Scenario1: Base case- both BRDF and PH operational Wood chips at BRDF, Natural gas for PH base load and  fuel oil for peak load     276,560     904,637        13,694    188,061     822,965        68        89  Scenario 2: PH operational only Natural gas     1,133,232  1,011,026       89 Scenario 3: BRDF operational only Wood chips     1,486,803  1,011,026       68 Scenario 4: All BRDF  with changed population dynamics Wood chips     1,486,803  1,011,026       68 Scenario 5: PH operational scenario at 2009/10 level Natural gas for PH base load and  fuel oil for peak load        909,659          20,552     883,813       95  First four scenarios were based on energy input during the one year period June 2012 - May 2013: Scenario 1 is the base case when both PH and BRDF were operational so energy demand was met by NG/oil and biomass;  Scenario 2 assumed the total energy input was provided by NG at PH only, whereas  Scenario 3 evaluated impacts in case of total replacement of fossil fuels with biomass and at  BRDF in the future.  77  The other two scenarios were introduced to address uncertainty in data selection:  Scenario 4 is based on scenario 3 for biomass-related (BRDF) emissions but with varying population as a single most important parameter in calculating iF. Summer time population on campus was changed to reflect a more realistic scenario by assuming that 50% of people are on vacation and only 10% students stay in residences; consequently daytime population during summer was calculated to be 23,648 in 374 buildings while 1,604 persons stayed at nighttime in 214 buildings.  Scenario 5 is based on emissions from PH during 2009-2010, which corresponds to the operation before the BRDF facility was built in order to evaluate the impacts of different emissions (as a result of different energy demand) on ambient air quality and population exposure. As for PH operation, both natural gas and fuel oil were included for the base case scenario (Scenario 1) and for 2009-2010 (Scenario 5) as both periods were based on actual fuel usage data. Scenario 2, as a hypothetical case was based on the assumption that natural gas boilers will provide peak heating as planned for new District Energy Utility at UBC.  4.4 Emission characteristics and estimates  The pollutants to be considered in this part of the study are PM2.5, CH4, CO, CO2, NOx, N2O and non-methane VOCs (NMVOCs). Since there are no prescribed ambient air quality objectives (AQO) for well-recognized GHGs (IPCC, 2015), VOCs, N2O or CO2 were not modeled in local health impact assessments.  Available in-house data were analyzed along with previous studies; emissions of each pollutant were then either calculated or estimated using published pollutant emission factors (EFp) and fuel consumption (Appendix B.2). BRDF and PH boilers’ stack  78  parameters were obtained from air quality permits and reports (Petrov et al., 2015). Estimated emissions are presented in Table 4.2 below.  Table 4.2 Estimated emission factors and annual emissions from biomass gasification (BRDF) and natural gas/oil combustion (PH).   Biomass at BRDF Natural gas and oil at PH Pollutant Estimated Wood waste gasification EF* Estimated emissions 2012/13 Emissions total if all biomass 2012/13 Estimated NG-fired boiler   EF**   Estimated oil-fired boiler EF** Estimated emissions 2012/13 Estimated emissions if all NG/oil  2012/13   EFwg E E EFNG EFOIL E E Units [g/GJ] [t] [t] [g/GJ] [g/GJ] [t] [t] CO2 fossil       49,170 68,478 45,419 55,721 CO2 biogenic 91,700 25,361 136.340     CO fossil    34.40 15.35 31.33 38.98 CO biogenic 14.6 4.038 21.71     CH4 fossil       0.9424 0.66 0.86 1.07 CH4 biogenic 9.03 2.497 13.43     NOx 73.10 20.217 108.69 40.95 30.71 37.46 46.40 N2O 5.59 1.546 8.31 0.9015 0.80 0.826 1.022 PM2.5 40 0.111 0.595 0.7785 6.14 0.788 0.882 NMVOC 4.3 1.189 6.393 2.2536 1.04 2.053 2.554 *Adopted from Pa et al (2011); wood waste includes forest harvesting residues and sawmill residues. ** Adopted from US EPA (1999, corrected 2010).  Emission factors for natural gas, fuel oil and wood waste (a mixture of forest residues and sawmill and planner mills residues), which relate the amount of emitted pollutants with an activity associated with the emissions (US EPA, 2009) are obtained from a recent study (Pa et al., 2011) and US EPA  (US EPA, 2009; US EPA, 2003a; US EPA, 2003b). The estimated emissions were then calculated by equation 4-1 (US EPA, 2009): 𝐸 = 𝐴 ∙ 𝐸𝐹𝑝  ∙ (1 −𝐸𝑅100)         (4-1)  Where: 79   E    is annual emissions [t/yr],  A     is activity rate such as annual energy input [GJ/yr],  EFp  (used as EFNG for natural gas combustion, EFOIL for fuel oil combustion and EFWg for wood gasification) is uncontrolled emission factor [g/GJinput] for pollutant p, and ER   is emission reduction efficiency [%] of the pollution control device. For instance, PM emission reduction efficiency is taken as 99% due to ESP at the BRDF (Pa et al., 2011), and 0% for NOx or CO since no controls were installed for these pollutants.   Estimated annual emissions were therefore obtained by multiplying pollutants’ respective EFp by annual fuel input. Subsequently, monthly emission rates expressed in [g/sec] were calculated for each pollutant as a product of monthly energy production (separated for daytime and nighttime) and corresponding emission factor, and were used as inputs to the CALPUFF dispersion model in order to reflect the varying emission rates, thus generating more accurate estimated results of local air pollutant concentrations. For example, for the month of February 2010, emission rate for filterable PM2.5 was estimated to be 0.029 g/sec for the PH which is slightly lower than the measured emission rate of 0.033 g/sec during a one-day event when boilers were operated at 45% max capacity. Similarly, for September 2012, the estimated PM2.5 emission rate from BRDF is 0.0036 g/sec versus one-day measured emission rate of 0.0065g/sec, indicating that either the assumed ESP efficiency is higher than the actual ESP efficiency or there are variations of emissions throughout the month.   Detailed emission data calculated for daytime and nighttime for each month, used in modeling scenarios, are presented in Appendix C. 80  4.5 Local air quality assessment  A multilayer, non-steady-state puff dispersion model CALPUFF View™, version 7.2.0 (BC MoE, 2015; Lakes Environmental, 2012b), was used to compute the ambient concentrations of selected air pollutants at 374 discrete receptors (buildings with assumed maximum occupancy) on UBC campus within a campus area of 5 km x 3.5 km around BRDF, which was selected to be a reference point for modeling.  The modeling domain extends 2.5 km in each of directions to the north, south and east from the plant and only 1km to the coast at the west (toward the ocean). Terrain, land use, and population data were used as presented in Chapter 3. Receptors’ height was set equal to the breathing zone of 1.5 m. Ambient air pollutant concentrations were calculated as 1-hr averages [µg/m3] and  were imported into excel spreadsheet where daytime hours were separated from nighttime hours for diurnal pattern analyses. Meteorological data were extended to one year period June 2012 - May 2013. Three primary signature pollutants: PM2.5, NOx, and CO were modeled. Chemical transformations were not considered except for NOx, since NO was assumed to be completely converted to NO2 (BC MoE, 2008). Background ambient concentrations for year 2012-2013 were obtained as an average over four air quality monitoring stations (Vancouver Kitsilano, North Delta, Richmond South, and Vancouver International Airport in Richmond) located in Metro Vancouver (Doerksen, 2014; Doerksen, 2013).  4.6 Health impact assessment After ambient concentrations have been obtained, iF and IS were calculated as per methods presented in Chapter 3. A summary of calculated iF and IS for all 5 scenarios along with ambient concentrations (min, max and mean values) are presented in Appendix D. Incorporating dynamic 81  iF in CF calculations brought about more accuracy to heath impact assessments than other methods such as those associated with commercial life cycle assessment software which uses static iF developed for average population density with an average daily breathing rates.  4.7 Discussion  Separating daytime and nighttime hours provides better insights in the effect of local wind circulation patterns, pollutant dispersion directions, and consequently human exposure. For example,  an hour during daytime and one hour during nighttime were randomly selected to illustrate the diurnal wind pattern changes primarily caused by the sea- and land-breeze circulations (Petrov et al., 2015).   According, to Figures 4.1 (a, b, c), where nighttime wind fields at 10 m attitude indicate north and north-east wind direction causing land-breeze, ambient PM2.5 concentrations on June 4, 2012 were a result of the plants’ emissions dispersed across campus towards the southwest corner. The lowest maximum 1-hr concentration of 0.0068 µg/m3 occurred in scenario 3 (Figure 4.1 a), when only BRDF (with installed ESP for particle control) was operational at full capacity compared to nighttime hours for other two scenarios. Besides, due to the locations of the two plants, the pollutant dispersion zone was broader when only PH is operational (scenario 2, Figure 4.1 b) or, when both PH and BRDF are operational (scenario 1, Figure 4.1 c). The more in-land location of the plant, the larger the impacted area of dispersed pollutants form such source.   Figure 4.1 (d, e, f) are pertinent to daytime hours that are characterized by ocean-to-land circulation, resulting in more dispersion of PM2.5 towards the central part of the campus, 82  southeast and east. It appears that the broadest area of particle dispersion characterizes scenario 1 with both plants being operational. Nevertheless, scenario 1 during this hour recorded 1-hr maximum ambient PM2.5 of 0.189 µg/m3, which was lower than the 1-hr maximum PM2.5 of 0.351 µg/m3 and 0.251 µg/m3 for scenario 3 (biomass only) and scenario 2 (NG only), respectively. 83   Figure 4.1  Wind circulation at 10 m altitude and projected PM2.5 concentrations for June 4, 2012 at 1 am (nighttime) for Scenario 3 (a), Scenario 2 (b) and Scenario 1(c), and at 1 pm (daytime) for Scenario 3 (d), Scenario 2 (e) and Scenario 1(f). Arrows present wind fields obtained by CALMET.a)  Scenario 3 - Biomass only - June 4, 2012 at 1 am d) Scenario 3 - Biomass only- June 4, 2012 at 1 pmb) Scenario 2 - NG only -  June 4, 2012  at 1 am e) Scenario 2 - NG only -  June 4, 2012  at 1 pmc) Scenario 1 - Biomass and NG, June 4, 2012 at 1am f) Scenario 1 - Biomass and NG, June 4, 2012 at 1pmPHBRDF84  The mean seasonal concentrations are based on 1-hr averages. These concentrations are higher during daytime for the four scenarios (2012-2013), which could be attributed to the expanded area of dispersion as previously explained rather than variation in the emissions (Figure 4.2).   Figure 4.2  (a) Daytime and (b) nighttime average concentrations per pollutant and modeling scenario.  The mean ambient NO2 concentrations resulting from natural gas combustion at PH are lower in scenario 5 (ranging from 0.311 µg/m3 for fall nighttime to 0.961 µg/m3 in spring, daytime) when compared with scenario 2 (0.369 µg/m3 fall, nighttime to 1.048 µg/m3 spring, daytime), as expected, because of the lower emissions in 2009/2010.  The mean PM2.5 concentrations associated with natural gas combustion are higher than those associated with biomass gasification, and higher in the spring (mean = 0.020 µg/m3, max = 1.85 µg/m3 from NG and mean = 0.012 µg/m3, max = 1.14 µg/m3 from biomass) and winter (mean = 0.019 µg/m3, max= 2.13 µg/m3 from NG; mean = 0.013 µg/m3, but higher max of  2.39 µg/m3 from biomass) than in other seasons. Scenario 3 (only BRDF using biomass operational) has the lowest mean concentrations for PM2.5 (due to installed ESP for particulate control) as well as  CO as compared 00.20.40.60.811.21.41.61.82PM2.5 CO NO2µg/m3DaytimeScenario 1Biomass and NG/oilScenario 2NG onlyScenario 3Biomass only00.20.40.60.811.21.41.61.82PM2.5 CO NO2Nighttime  Scenario 1Biomass and NG/oilScenario 2NG onlyScenario 3Biomass only85  to other scenarios. The overall incremental PM2.5 contribution to local air quality is at least one order of magnitude lower (1-hr values multiplied by 0.4 to obtain 24-hr average values or multiplied by 0.08 to obtain annual average values) (EPA, 1992), than the BCAQO (BC MoE, 2016). However, when the highest calculated 24-hr average PM2.5 concentration of 1.28 µg/m3 (Scenario 1, winter, nighttime) is added to the averaged maximum background 24-hr concentration of 23.8 µg/m3, the resulting value is slightly higher than the BCAQO of 25 µg/m3. The location of this maximum is north of PH, and occurred on December 7, 2012 at 7 am when winds started shifting from north-east to south and south-east which indicates that PH emissions were likely a major contributor. Emissions from UBC can contribute to possible exceedance of the BCAQO if the Vancouver Kitsilano monitoring station measurements are excluded which will lead to higher averaged maximum background levels of 24.17 µg/m3, suggesting that non-compliance with ambient air quality standard is possible in case of northern winds which would add particles emitted from PH to the already higher background maximum levels around YVR and south Richmond of 30 µg/m3, as reported for 2013 (Doerksen, 2014).   Ambient concentrations of NO2 are significantly higher for the biomass scenario than the NG scenario. While the mean NO2 values are between 0.563 and 2.284 µg/m3, there exist noticeable peak concentrations during daytime and nighttime for all seasons except summer for scenario 3. All of these 1-hr maximum hourly concentrations exceeded the 1-hr BCAQO of 200 µg/m3 with the highest being 436.87 µg/m3 on February 9, 2013 at 4 pm when southwest winds directed NO2 to the location just northeast from BRDF (Civil and Mechanical Engineering building). The second highest maximum of 373.73 µg/m3 occurred on November 5, 2012 at 4 pm with similar wind patterns affecting the Wayne and William White Engineering Design Centre, located north-86  east from BRDF. It should be noted that a most conservative approach of 100% conversion of NO to NO2 was applied. This assumption should be verified by comparing the estimated values with measured ambient NOx and NO2 concentrations at surrounding air quality monitoring stations. Background averaged maximum 1-hr NO2 ambient concentration was 94 µg/m3.   Ambient CO concentrations for the scenarios where PH operation is dominant are up to three times higher than when emissions originate from BRDF. Yet the values are still low compared to the 1-hr BCAQO of 30,000 µg/m3. Maximum concentration of 115.18 µg/m3 (Scenario 2, fall, daytime) is 5% of the existing background value of 2,295 µg/m3. The results for all 5 scenarios are summarized and provided in Appendix D.  With respect to health impacts (iF and IS) estimates, it is observed that NOx gives rise to higher impact than PM2.5 although the impact per unit mass NOx (EFhealth) was 10 times lower than PM2.5, because the uncontrolled NOx emission from biomass district heating system was much higher than controlled PM2.5. It should be noted that NOx has been well recognized as a major air pollutant emitted from biomass combustion systems without NOx emission control device.  iF for the UBC campus over the 2012-2013 period was low (70 ppm) for scenario 3 in the case of biomass totally replacing natural gas. It is even lower (59 ppm) for scenario 4 when the population decreases during summer time, which is a logical outcome since iF is proportional to the population. Scenarios 2 and 5 have almost the same iF of 104 and 107 ppm, respectively17                                                  17 This slight difference in values is likely due to rounding numbers throughout multiple calculation stages in order to obtain iF. 87  demonstrating that varying emissions from the same source (only PH is operational) during two different years do not affect iF, although intake amounts will be different. The base case scenario 1 is characterized by different iFs for each pollutant because pollutant emissions are additive from the two sources at different locations on campus.   The total health-related impact score (IS) is the highest for NO2 ranging from 361 DALY (scenario 5) to 677 DALY (scenario 3).  This is followed by PM2.5 with IS ranging from 25 DALY (scenario 4) to 64 DALY (scenario 2) and 62 DALY (scenario 5). Impact score for CO ranges from 1 to 3 DALY across the scenarios. It appears that IS for NO2 is highly influenced by high emissions of this pollutant especially from biomass gasification in scenario 3, which brings the overall IS for scenario 3 to 708 DALY, making the total replacement with biomass the least favorable option. It should be noted that a potential introduction of NOx control device at a NOx reduction around 70% (Babcock Power Environmental, 2008) could bring IS for NOx down to 203 DALY and overall IS for biomass heating to 233 DALY. Similarly, if in scenario 1, NOx biomass-related emissions18 were reduced by 70%, the total emissions would be reduced from 57,521kg to 43,408 kg so IS for NO2 would drop to 339 DALY and overall IS for scenario 1 to 402 DALY, making the total replacement with biomass the most favorable option. PM2.5, impacts are more significant for the scenarios with dominant use of fossil fuels (NG/oil) for PH operation.  Overall, considering IS by combining pollutant impacts for the first three main scenarios, the use of NG appears to have the smallest total health impact of 495 DALY (although the highest iF), followed by the distributed energy supply system (a split between PH and BRDF)                                                  18 In scenario 1, NO2 emissions from BRDF are 20,161 kg and from PH 37,360 kg. 88  of 513 DALY whereas switching completely to biomass (with uncontrolled NOx emissions) increases the human health burden by 28% (IS = 708 DALY, but the lowest iF) compared to the base case and 30% compared to PH as the sole operation.  4.8 Model performance evaluation  In spite of the high performance and sophistication of dispersion models nowadays used in practice, it is important that they are properly validated due to possible economic, environmental and public health implications of predicted results. Model validation can be scientific, statistical or operational. Scientific and operational validation of CALPUFF View™ has been safeguarded by US EPA (US EPA: SCRAM, 2015) and Lakes Environmental (Lakes Environmental, 2012b). Statistical validation is the most common and appropriate where model results are being compared against measured values. Such process addresses uncertainty associated with factors such as input values  (Chang and Hanna, 2004). It should be noted that randomness of natural processes (such as atmospheric turbulence and dispersion) leads to inherent uncertainty and makes validation and verification of atmospheric models very difficult. Nevertheless, evaluation of model results via comparison with measured values gives better insight in upper and lower limits of possible values, i.e. pollutant concentrations. Evaluation can be performed as graphical (time series plots or scattered plots of modelled vs observed hourly concentrations) or statistical. Due to the scarce number of ambient air quality measurements on campus (only one monitoring station was installed for monitoring the impacts of BRDF emissions), graphical analysis is selected in this study.  89  Since BRDF became operational in 2012, instruments are set on the roof of an adjacent Marine Drive #5 residential building to monitor hourly concentrations of (NO2) and fine particles (PM2.5) which are selected for this analysis. Comparisons between ambient concentrations obtained as model output and continuous monitoring data of PM2.5 were performed on an hourly basis.  4.8.1.  Graphical analysis Based on recommendations from US Environmental Protection Agency (US EPA, 2007), a period of interest (a day, a week, a season, etc.) of hourly values for PM2.5 on July 17, 2012 was selected, as stack emissions were monitored on that day by a third party.  Fine particle measured emission rate was 0.028 g/sec, flue gas exit temperature 477 K and gas exit velocity from the 20 m high EN02 boiler stack was 8.43 m/sec. Dispersion modeling was then performed using these measured emission data (assumed to be constant throughout the day) so ambient modeled PM2.5 concentrations were checked against ambient monitoring data (Figure 4.3).    Figure 4.3 Graphical comparison of ambient measured and ambient modeled PM2.5 concentrations for July 17, 2012. 0246810121416181 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24Concentrations [ug/m3]PM2.5Ambient measuredPM2.5 [ug/m3]ModeledPM2.5 based on measured emission [ug/m3]90  The receptor was first set to be the top of the Marine Drive #5 residential building (45.38 m roof top height based on UBC data) where the ambient air quality monitoring instrument was installed and then at 25 m to check on concentrations in close vicinity of the stack. The horizontal ground distance between EN02 boiler stack and the Marine Drive building is estimated to be 80 m.    Modeled ambient concentration appeared to be noticeable only at 2 am and 3 am (values number 3 and 4 on the graph), and were lower than ambient monitoring data. In the recommendations of US EPA, performance evaluation priority may be given to those days with 24-hour average PM2.5  > 65 µg/m3 (US EPA, 2007). However, this analysis was not performed, as measured PM2.5 concentrations never reached that level during the period selected for analysis in this study (June 2012 to May 2013).  Data from the Totem campus weather station were considered to provide better understanding of local circulation patterns and consequently dispersion direction on hourly basis, with the results summarized in Table 4.3 below. In addition, MM5 wind fields19 at10 m height (corresponding to anemometer height at the weather station) were also recorded as those data were used in modeling. It appears that calms were recorded at 2 am and 3 am with wind speed below 0.5 m/sec and wind directions for these hours being ENE and NNE, respectively as per Totem station. This could be interpreted as that any emissions from the stack at that time were likely lingering around the source, possibly in the vicinity of the nearby Marine Drive #5 residential                                                  19 MM5 data are based on a number of meteorological stations and satellite data which have been reanalyzed and gridded into a format suitable for input to the meteorological model. Calculations are also carried out for wind behaviour between grid cells (Source: Lakes Environmental).  91  building. At the same time, MM5 data indicated southern winds during these hours which could have dispersed pollutants towards Marine Drive building. Based on the wind directions recorded, any additional pollution reflected in a slightly higher measured concentrations of PM2.5 than modelled would likely come from nearby sources located in the north-eastern quadrant20 from the source.   Table 4.3 Measured and modeled PM2.5 data and Totem station meteorological parameters for July 17, 2012. Month  Day  Hour  Ambient measured PM2.5 [µg/m3]  Modeled PM2.5 H=45.38m [µg/m3] Modeled PM2.5 H=25m [µg/m3]  Wind speed [m/s] Wind direction Totem [deg] Wind direction MM5 7 17 0 10.3 0.0535 0.0363 1.6 91.7 (E) S 7 17 100 10.6 0.0000 0.0000 0.62 60.2 (ENE) S 7 17 200 11.6 7.9849 0.2612 0.27 72.7 (ENE) S 7 17 300 11 1.0658 0.0304 0.58 35.4 (NE) S 7 17 400 11.7 0.0000 0.0000 0.24 249.7 (WSW) SSW 7 17 500 13.5 0.0000 0.0000 1.1 109.7 (ESE) WNW 7 17 600 16.9 0.0000 0.0000 1.2 79.4 (E) WNW 7 17 700 13.3 0.0000 0.0000 0.98 100.1 (E) WNW 7 17 800 12.6 0.0000 0.0000 1 248.4 (WSW) WNW 7 17 900 14 0.0000 0.0000 0.4 157 (SSE) WNW 7 17 1000 15.4 0.0000 0.0000 1 243.8 (WSW) WNW 7 17 1100 14.3 0.0000 0.0000 1.83 277.1 (W) WNW 7 17 1200 15.2 0.0000 0.0000 1.72 237.1 (WSW) WNW 7 17 1300 11.6 0.0000 0.0000 1.52 242.9 (WSW) W 7 17 1400 10.2 0.0000 0.0000 1.52 243 (WSW) WSW 7 17 1500 11.3 0.0000 0.0000 1.48 203 (SSW) SW 7 17 1600 13.5 0.0000 0.0000 1.5 164.7 (SSE) SW 7 17 1700 8.4 0.0000 0.0000 1.37 170.3 (S) SSW 7 17 1800 8.4 0.0000 0.0000 1.15 155.6 (SSE) SSW 7 17 1900 10.8 0.0000 0.0000 1.07 153.2 (SSE) SSW 7 17 2000 13.3 0.0000 0.0000 0.69 137.6 (SE) N 7 17 2100 13.3 0.0000 0.0000 1.53 103.7 (ESE) NW 7 17 2200 15.1 0.0000 0.0000 1.85 112.3 (ESE) SW 7 17 2300 12.7 0.0000 0.0000 3.07 96.3 (E) WNW                                                  20 North-eastern quadrant refers to locations from 0 degrees to 90 degrees, meaning north to east. 92  Likewise, for hours when wind was not blowing towards the receptor (Marine Drive #5 residential building), the expectation is that the modeled values will show zero concentrations, ie. there were no emission impacts from the source on particular receptor.  CALPUFF was used in many ‘near-field’ applications (distance from a source < 10 km). The  better  performance is demonstrated for predicting mean annual  concentrations than short-term ones (Holnicki et al., 2016), and for larger distances (Rood, 2014).  Inspite the possible issues with underestimating concentrations compared to Gaussian plume models (U.S. EPA, 2008), its use is justified in cases of complex wind fields (sea-land breeze), calms, lack of measured meteorological data (only one meteorological near-by station at UBC) as it can fully treat variation of meteorology in space and time unlike steady-state Gaussian models.  Although some experiments confirmed that some other Gaussian models (such as ADMS21 developed in UK) can better perform in the built environments than CALPUFF and SCREEN (Tominaga and Stathopoulos, 2016), other studies (Hajra et al., 2010) showed that such models cannot treat complex plume behavior due to turbulence caused by buildings as obstacles to the flow and as such may cause considerable errors in estimates of effects such as short-term exposure and unsteady processes. Vieira de Melo et al. (2012) concluded based on wind tunnel experiments that AERMOD will predict higher near-field concentrations than CALPUFF; the latter can also under-predict concentrations by a factor of two or more, depending on conditions. Another study (Cui et al., 2011) showed that CALPUFF can simulate flow in near-by complex terrain but can underestimate peak concentrations. A larger number of measured outdoor (ambient)                                                  21 ADMS - Atmospheric Dispersion Modelling System. 93  concentrations would be ideal to evaluate model performance using numerical methods such as in Holnicki et al. (2016).   When meteorological data were used for interpretation of results for a longer period of time such as for the whole month of July 2012, the following was observed:   When the wind was blowing over 393 hours from 45 deg. to 125 deg. i.e., from NE, E to SE, the average wind speed was 1.6 m/s and maximum measured PM2.5 concentration was 23.1 µg/m3 while maximum modeled concentrations were 7.98 µg/m3 based on measured emission rates.  Wind direction and speed indicate contributions from sources located to the NE, E and SE from the source.  When the wind was blowing over 157 hours from 135 deg. to 225 deg. i.e., from SE to SW, meaning that the plume from the BRDF stack should be carried towards the building, with the average wind speed of 1.4 m/s, measured maximum PM2.5 concentration was 25.7 µg/m3 while maximum modeled concentrations were 2.08 µg/m3. There is an assumption that building (sources) located SE from BRDF and vehicular traffic to the S and SW of BRDF contribute to higher measured concentrations than modelled.  When the wind was blowing over 166 hours from 226 deg. to 315 deg. i.e., from SW, W and NW, the average wind speed was 1.5 m/s and measured maximum PM2.5 concentration was 26.7 µg/m3, the highest of all measured over this period. Modeled maximum concentrations were 0.039 µg/m3. This indicates a low likelihood of contributions from the BRDF stack but increased particulate levels could be rather originating from the vehicular traffic in Marine Drive. 94  4.9 Conclusions Based on the results of this study, it appears that the health impact from a biomass-based energy system installed with an efficient PM control device mainly results from the uncontrolled NOx emission, followed by PM and CO emissions, among all criteria air pollutants. The lowest iF for this option indicates the importance of the plant location relative to community setting where the smallest number of people would be affected by plant emissions since iF is mostly influenced by the number of people exposed. On the other hand, it appears that a distributed district heating system with combined NG and biomass may have an advantage over a community-based centralized heating system in terms of overall health impact. This option can have smaller iF, depending on plant locations compared to the location of a single plant (like in case of PH) and also lower overall health impacts compared to a single biomass energy supply system. Further research is needed to confirm the initial findings presented in this study that multiple emission sources and combined use of NG and biomass could lower health impact compared to community-based centralized biomass plant.   It is worth of noting that considering locally obtained dynamic iF for calculating CF may also bring more accuracy in assessing local impact in life cycle assessment studies instead of using the CFs based on consensus data and other population density data averaged over a large area, e.g. the European continent. The “emission factor method” was used to estimate emissions in this study. Future research on impact assessment should focus on conducting direct measurements of the emissions in order to support scale-up and draw conclusions on the basis of seasonal and annual variations. It would be useful that a community installs ambient air monitoring stations on several “hot spots” for obtaining real-time concentration data. This would be especially 95  important for monitoring pollutant concentrations during periods of possible increased concentrations (since concentrations obtained by CALPUFF could be underestimated) which can violate air quality objectives.  Depending on the capital and operating costs (including the cost for emission control) as well as energy efficiency being acceptable on a local scale, either splitting emissions into more than one source at different locations and different fuel types or a single source at the least-impact-based location with biomass as a fuel and emission control could be a viable option. In the decision making process about community-based energy systems, associated costs for different options should be balanced with lowering ambient air pollutant concentrations and hence reducing the risk of exceeding the BCAQO, as well as reducing population exposure.    96  Chapter 5: Global impacts of the UBC district heating system  5.1  Introduction The presence of greenhouse gasses (GHG) in the atmosphere is responsible for the global warming impact and consequently affects the Earth’s climate. Evaluation of GHGs is regularly performed and reported on the national and provincial levels in Canada. The Ministry of Environment and Climate Change Strategy BC reported a 2.7% increase in GHG emissions from 2011 to 2014 but a 9% decrease from 2004.  Although population in the province is steadily growing, GHG intensity (GHG emissions per person) is in decline during the last decade with a small peak in 2013 as shown in Figure 5.1. The same figure illustrates the steady decline in GHG intensity when measured with respect to GDP (BC Ministry of Environment and Climate Change Strategy, 2016).   Figure 5.1 Trends in GHG emissions in BC 1990 – 2014. Source: BC Ministry of Environment and Climate Change Strategy (2016).97  The largest amounts of GHG in the province come from the energy sector where transportation and stationary combustion sources, namely space heating, were identified as the major sources of GHG emissions  (BC Ministry of Environment and Climate Change Strategy, 2016).  Thus, it is of a paramount importance to evaluate GHG emissions from existing and new energy sources.    Assessment of environmental impacts on large spatial and temporal scales, such as global scale, requires different methodologies from those used for assessments on a local scale. Life Cycle Assessment (LCA) is an effective tool commonly used for those purposes, which is widely used to evaluate the impacts of different fuels over the entire life cycle. Many studies used this method to evaluate impacts of bioenergy systems but high variability of results among studies points towards a need for harmonization in assumptions and the selection of a functional unit,  system boundaries, allocation methods, carbon cycle modeling etc. (Muench and Guenther, 2013).  5.2 Quantifying global impacts of UBC district heating  To estimate the energy use and the emissions and impacts associated with the supply and use of biomass and fossil fuels (natural gas and oil) for district heating, two assessment approaches are used, one of which is an attributional life cycle assessment (LCA) methodology following the recommendations of ISO 140044:2006  (ISO, 2006), as explained in Chapter 3. 5.2.1. Feedstock sourcing and characterization at the UBC Point Grey campus Wood feedstock for the BRDF is supplied by a recycling company Cloverdale Fuels Ltd. (Cloverdale) located in Langley, BC. A number of visits to Cloverdale served as the first step to collect data on source industries, their locations and quantities of woody biomass collected, 98  processed and stored at the Cloverdale’s site in Langley, BC.  Data were organized in a spreadsheet and distances traveled from each site to Cloverdale and from Cloverdale to BRDF were calculated. The company receives woody biomass from many different places and produces wood fuels for many different customers, but only the wood residue retrieved and sent to BRDF was studied here.   Once received at UBC, a variety of biomass properties are regularly measured. Wood bulk density is determined following the CEN/TS 15103 method (CEN (European Committee for Standardization), 2005a). Moisture content of received wood-fuel samples was measured following the CEN/TS 14774 method (CEN (European Committee for Standardization), 2005b). Moisture content can be expressed on wet basis as:  MCW = (Wwet - Wdry)/Wwet ∙100)                                                         (5-1)  where MCW is the moisture content on wet basis [%], Wwet  is the mass of the sample before drying [g] and Wdry  is the mass of the sample after drying [g].  Conversion to moisture content on dry basis (MCd, %) is performed using the following equation:   MCd = MCw/(100 –MCw) ∙100                                                       (5-2)  The high heating value (HHV) of the sample was then measured and recorded following CEN/TS 14918 method (CEN European Committee for Standardization), 2005) using a bomb calorimeter  (Model 6300, Parr Instrument Company).  99  5.2.2. Goal and Scope The primary goal of this analysis is to quantify the global warming impact as it is an important global impact for energy systems (IEA Bioenergy, 2011).  For the reference of energy system, the UBC campus energy production scheme was used over the period of June 2012 – May 2013 and it was evaluated against the two other energy system schemes: 1) if all energy demand for campus heating was met entirely by biomass; and 2) if all energy demand was met entirely by fossil fuels. Global warming impact results from LCA were then used together with the local health impact to discuss the possible trade-offs between local and global impacts for the selection of district heating systems. The functional unit was selected to be MJ of energy (for each fuel) produced to enable comparison among scenarios. The amount of heat produced in the period June 2012 – May 2013 which is equal to 1,011 TJ, is marked as the annual energy output which is equal for all considered scenarios.  System boundaries for unit processes and transportation segments for LCA are presented in Figure 5.2. Since biomass feedstock supplied to UBC is waste material, upstream processes associated with plantation, harvesting and processing of trees to generate biomass residues were excluded. Fossil fuel-related processes include extraction and refining, transmission and combustion at UBC Power House. Global warming impacts were evaluated using two commercially available software packages: GHGenius, version 4.03 ((S&T)2 Consultants, 2013) and SimaPro, version 8.2.0.0 (PRé, 2016).   Foreground processes related to biomass for which site-specific data were collected include waste wood transportation from industrial sites to Cloverdale site and from Cloverdale to UBC, 100  processing at the Cloverdale site, i.e. estimating energy and materials input and emissions output due to the use of wood transportation, machinery fuels, and gasification of wood residues at BRDF (Figure 5.2 a). As wood was treated as a waste material, emissions associated with upstream from tree plantation to wood waste generation were excluded from the system boundary in the current study.  Figure 5.2 Process stages and transportation segments considered in evaluating global impacts of a) biomass and b) fossil fuels.   For natural gas (base load) and oil #2 (peak load) combustion, background data included upstream processes: extraction, refining and transmission whereas site-specific data were collected for combustion of natural gas and oil at UBC (Figure 5.2 b).  5.2.3.  Life cycle inventory  Site-specific data, referring to actual Cloverdale and BRDF operations, were collected during the Cloverdale site visits through company’s records and direct laboratory analysis of received wood feedstock. Emissions from BRDF wood gasification and natural gas/oil combustion at Power House (PH) were estimated based on published emission factors for natural gas (EFNG) 101  and oil (EFOIL) combustion, wood gasification (EFwg), and calculated actual fuel consumption for year 2012-2013 as presented in Chapter 4.  Data for natural gas and oil upstream processes from fossil fuel extraction to transmission to PH as well as transportation data (fuel-energy used and emissions factors expressed as [g/t-km]) were obtained from GHGenius v. 4.03 and entered in excel spreadsheet to calculate actual emissions (impact assessment approach 1) and entered in SimaPro v.8.2.0.0 for impact assessment (impact assessment approach 2). The types of energy consumption, both primary and secondary, considered in this study are: electricity, natural gas, fuel oil, diesel (middle distillate), and wood waste. Electricity supply  mix in BC changed over the years shifting to increased use of natural gas and non-hydro renewables (Natural Resources Canada (NRC), 2015; Government of Canada, 2014; NEB, 2013), and 2013 was used as the base year in this study. The largest contribution to BC energy mix came from hydro (91.82%), natural gas and combustion of other fuels contributed with 1.62% and 1.68% respectively, steam from waste heat 0.15%, renewables (wind, tidal and solar) 0.28% and other generation 4.4%.   BC electricity generation intensity is presented in Table 5.1. The mass of methane per kWh electricity produced decreased since 1990s but CO2 intensity increased compared to 2001 and 2012 whereas N2O intensity varied over the years being higher in 2013 than in 2012 but lower compared to 2005-2010 period.    102  Table 5.1  Greenhouse gas intensity [g GHG/kWh electricity generated] in BC. Year 1990 2000 2005 2010 2011 2012 2013 2014 CO2 intensity [gCO2/kWh] 17 35 24 23 13 11.1 14.9 14.3 CH4 intensity [gCH4/kWh] 0.004 0.009 0.007 0.007 0.004 0.003 0.003 0.003 N2O intensity [gN2O/kWh] 0.0006 0.001 0.0015 0.0015 0.0011 0.0007 0.0009 0.0009 Generation Intensity [gCO2eq/kWh] 17 35 25 24 14 11.4 15.2 14.7 Source: (Environment Canada, 2017).  Biomass collection. Cloverdale collects waste wood from a number of locations, all situated in the Lower Mainland. The analysis of Cloverdale’s client records, which were predominantly manufacturing companies dealing with import-export of wood products, identified 114 locations in radius of 100 km from the company (Cot, 2016).   Biomass transportation. Transportation considered delivering wood waste to the Cloverdale site and delivering processed wood waste to UBC by heavy duty vehicles (HDV), trucks which use diesel (middle distillate). GHGenius was used for transportation segment calculations. Input data included truck characteristics, namely, load and fuel consumption. The average trucks’ fuel consumption in liters per distance traveled (L/100km) was calculated as:    𝐴𝑣𝑔. 𝐶𝑜𝑛𝑠 (𝐿100 𝑘𝑚) =   𝐴𝑣𝑔.𝐶𝑜𝑛𝑠 𝑤𝑒𝑒𝑘 (𝐿)⁄𝐴𝑣𝑔.𝑇𝑟𝑎𝑣𝑒𝑙 𝑤𝑒𝑒𝑘⁄ (𝑡𝑟𝑎𝑣𝑒𝑙)∗𝐴𝑣𝑔.𝐷𝑖𝑠𝑡𝑎𝑛𝑐𝑒(𝑘𝑚𝑡𝑟𝑎𝑣𝑒𝑙)  ∗ 100               (5-3)  This could also be calculated through energy consumption intensity expressed by kJ/tonne.km-shipped indicating the biomass load:   Avg. Cons (kJtonne.km−shipped) =  Avg.Cons (Lkm)∗Avg.km (kmtravel)∗Ediesel (kJL)Load 〈tonne〉∗Avg.km〈kmtravel〉                              (5-4)  103  A total of eight trucks with 30 m3 carrying capacity transport wood waste from designated sites to Cloverdale’s site daily. For an average wood residue density of 187.4 kg/m3 fully loaded trucks can transport 5.7 t of wood for each trip. The assumption of fully loaded trucks for each trip and an average of 3 trips per day was applied in calculations. Based on equation 5-3 and equation 5-4 the average truck diesel fuel consumption was 52.4 L/100 km. This consumption appeared to be higher than the average HDV consumption in Canada for 2013 of 40 L/100km ((S&T)2 Consultants, 2013). Older trucks and urban area routes could be some of the reasons for the higher fuel consumption rates.    Biomass processing. Once delivered to the Cloverdale site, three different types of machinery, a diesel excavator, a diesel loader and an electrical grinder, were used to process wood waste. An excavator feeds a grinder yielding in wood chips smaller than 3 inches as required by UBC. A loader brings wood chips to the sheds to be stored before being loaded on trucks and transported to UBC. The energy consumption of each machine was expressed per tonne of wood. Cloverdale provided data for the average energy consumption of the loader and the excavator as well as the average wood chips production per day so that the consumption was calculated as: Avg. Cons (Lt) =Avg.Consper day (L/day)Avg.tonneproduced per day (t/day)                                                    (5-5)  The data for the electrical grinder were not directly available from Cloverdale.  Instead, the grinder energy consumption was calculated from the Cloverdale’s electricity bills and daily grinder productivity.  Cloverdale operations are 7 days per week, but the processing units are 104  running only 5 days a week or on average 21 days per month. The final consumption per tonne of wood was calculated to be 51.3 kWh/t of wood waste  (Cot, 2016).  GHGenius was used for estimating emissions from transportation and machinery fuels, since it uses the North America specific database. Output data from GHGenius included upstream emissions for each considered gaseous and particulate pollutant (here considered CO2, CO, CH4, N2O, NOx as NO2, SOx, NMVOC, PM). GHGenius also does not report biogenic emissions separately. Instead, biogenic EF were taken from another study (Pa, 2010) for subsequent  impact analysis. Electricity mix and fuel characteristics for British Columbia in the year 2013 were used. Data of indirect emissions linked to the use of electric power are already included in GHGenius. Input and calculated data for the segment of collecting, processing (at Cloverdale) and transporting biomass from sites to Cloverdale and from Cloverdale to UBC as previously explained are presented in Table 5.2. Storage of wood waste and associated emissions were not considered.  105  Table 5.2  Transportation and wood processing data                                                   22 (UBC, 2015a). 23 Source: http://hydrogen.pnl.gov/hydrogen-data/lower-and-higher-heating-values-hydrogen-and-other-fuels (Accessed October 4, 2016). 24 Source: http://www.world-nuclear.org/information-library/facts-and-figures/heat-values-of-various-fuels.aspx (Accessed October 4, 2016).  Input Parameter Value Units Output/Calculated Value Units Transport HDV Truck transport data Distances & frequency travelled  3 trips/day Average distance for 1 trip Industry to Cloverdale  27.2 km 3 trips/day Average distance for 1trip Cloverdale to UBC  51.6  Truck capacity 40 Cubic Yards Truck capacity 30.4 m3 Number of trucks/trips  to UBC 2-422 3 trucks average Total capacity for 3 trucks 91.2 m3 Mass of wood received 91.2 x 187.4 m3 x kg/m3 Average mass of wood received/day 17.1 t    Mass of wood received/truck 5.7 t HDV fuel (diesel) consumption 0.5 L/km HDV fuel (diesel) consumption 52.4 L/100km HHV diesel 45.6 39 MJ/kg 23 MJ/L 24 Fuel efficiency 2.271 MJ/t-km-shipped Wood processing at the Cloverdale site (biomass production)                           5 days/week = 21days/month Electricity for grinder per day 4,104 kWh  Consumption/ t (@80t/day) 51.3 kWh/t Loader production/day 80 t Loader consum./t (diesel) 1.25 L/t Excavator production/day 80  t Excavator consum./t (diesel) 1.25 L/t 106  Biomass and fossil fuel conversion. Once at the UBC site, the BRDF, wood chips are unloaded from the trucks and placed in one of the two bins (Figure 5.3 a). The second step is sorting wood chips with desired size for gasification (Figure 5.3 b) so that oversized wood chips could be separated and placed in a wood waste bin (Figure 5.3 c) and returned to Cloverdale. Figure 5.3 Wood chips at BRDF: a) storage bin b) sizing c) oversized for oversized wood chips.  Wood characteristics, determined in the UBC lab are presented in Table 5.3. Table 5.3  Wood chips characteristics.  Parameter Value and units MCd    54% HHV 19.3 MJ/kg (dry wood) Average wood density 187.4 kg/m3    UBC-owned natural gas distribution system25 enables regular supply of natural gas from Shell Energy North America via FortisBC pipelines.  Gasification of wood chips is carried out in BRDF and combustion of natural gas and oil in the PH. Emissions data for both plants (BRDF                                                  25 Source: http://energy.ubc.ca/ubcs-utility-infrastructure/natural-gas/ (Accessed April 10, 2017).  107  and PH) were calculated as presented in Chapter 4. Upstream (production of energy) and downstream (usage of energy) emission factors and emissions generated and obtained via GHGenius and UBC reports, are presented in Appendix E1 and Appendix E2, respectively. Emission factors for transportation stages are expressed in kg of pollutant emitted per tkm (traveling 1 km with the load of 1 t), [kg/tkm] while for processes emission factors are expressed in kg of pollutant emitted per MJ of energy input [kg/MJ].  Emissions are obtained by multiplying energy consumption and a corresponding emission factor for each pollutant.  5.3 Global impact assessment of UBC district heating options and discussion 5.3.1. Impact assessment approach 1 A spreadsheet model in MS Excel was used to calculate emissions based on upstream and downstream emission factors (EF) for different fuels used for both PH and BRDF which are presented in Appendix E1, Table E1-1 to E1-4.  Based on the fuels consumption emissions were calculated for process and transportation stages using equation 4-1 and other parameters. For example, the annual emission of each pollutant resulting from the operation of a loader at the Cloverdale site is calculated as a product of fuel consumption (1.25 L per tonne of wood processed), mass of wood processed (t/day), 21 working days per month, 12 months per year (252 days) and HHV of diesel (39 MJ/L) which resulted in a factor characteristic for a particular scenario depending on the mass of wood processed. This factor is used as a multiplier for each respective pollutant EF [kg/MJ] upstream and downstream, as presented in Table E1-3. It should be noted that fuel efficiency was included and considered in GHGenius. The same approach was taken to calculate emissions of excavator and grinder as well as transportation stages (from industry to Cloverdale and from Cloverdale to UBC) for waste wood utilization. Natural gas and 108  fuel oil combustion as well as waste wood gasification emissions were calculated in a similar way although presented upstream and combustion EF in Appendix E1 were multiplied by energy input calculated in Chapter 4. The analysis of emissions per processing and transportation stage was done for each scenario and it is expressed as annual GHG emissions in kg of CO2eq.    Scenario 1 included the base case when both biomass at BRDF and natural gas and fuel oil at PH were used to meet the campus energy demand of 1,011 TJ in the period of 2012-2013.  As depicted in Figure 5.4 the major contributor to annual GHG emissions (76.9 %) is natural gas combustion with 4.48E+07 kgCO2eq followed by upstream natural gas processing  with 1.17E+07 kgCO2eq (20.1%). Oil combustion is responsible for 1.6% of total GHG emissions with 9.42E+05 kgCO2eq while wood gasification contributes only 0.8% with 4.88E+05 kgCO2eq of total emissions since CO2 emissions are attributed mostly to biogenic  CO2. Other biomass-related processes contributed with less than 1% share of total emissions:  wood processing at Cloverdale (1.12E+04 kgCO2eq), and total wood transport (1.41E+05 kgCO2eq).  GHG emissions from upstream processing for fuel oil are negligible, 0.4% contribution to total GHG emissions (2.11E+05). It is clear that fossil fuel, namely natural gas usage, produces the largest amount of GHG. In addition, gasification/combustion of biomass does not account for net CO2 emissions because they are regarded as neutral over life cycle, so the regulatory practices require reporting of CO2biogenic separately (BC Ministry of Environment and Climate Change Strategy, 2016). 109   Figure 5.4 Scenario 1: Annual GHG emissions [kgCO2eq] per life cycle stage for natural gas, fuel oil and biomass.  When contribution of pollutants from each stage is considered (Figure 5.5), major greenhouse gas, CO2 fossil, originated almost completely from NG combustion followed by natural gas upstream processing, oil combustion, wood transport and wood processing.  While N2O and NOx emissions could be attributed to biomass gasification and NG combustion, SOx emissions of 1.61E+04 kg/year originated from oil combustion and upstream natural gas processing. Wood gasification N2O emissions (1.55E+03 kg/year) are an order of magnitude higher than emissions from natural gas combustion (8.16E+02 kg/year). Particles which are of primary concern for health impacts are mostly emitted from natural gas combustion (7.04E+02 kg/yr) and wood gasification with ESP in place (1.10E+02 kg/year) followed by oil combustion (84.1kg/year) and wood processing (46.9 kg/year). A summary of numerical vales are presented in Table E2-1.    110    Figure 5.5 Scenario1: Pollutant emission contributions from different life cycle stage  Scenario 2, which considers a district heating option with natural gas meeting the campus’ energy demand, includes emissions from upstream processing and the combustion process (Figure 5.6).  It is obvious that upstream processing with 1.47E+07 kgCO2eq/year contributes to GHG emissions less than combustion with 5.61E+07 kgCO2eq/year.        Figure 5.6 Scenario 2: Annual GHG emissions [kgCO2eq] per life cycle stage for natural gas. 111  The analysis of stage-wise emissions per pollutant (Figure 5.7) indicted that the major contributor to CO2 (5.57E+07 kg/year), N2O (1.02E+03 kg/year) and CO (3.90E+04 kg/year) emissions is natural gas combustion whereas upstream processes are mostly associated with emissions of CH4 (1.81E+05 kg/year), NOx (5.16E+04 kg/year), SOx (1.17E+04 kg/year ) and NMVOCs ( 3.92E+03 kg/year). Emissions of particulate matters are associated with both process stages, 6.75E+02 kg/year from upstream processes and 8.82E+02 kg/year from natural gas combustion. Numerical values are presented in Table E2-2.  Figure 5.7  Scenario 2: Pollutant emission contributions by life cycle stage.   Scenario 3 considered biomass as the only fuel used for district heating at UBC campus to meet energy demand of 1,011 TJ per year. This implies a larger amount of wood (as calculated and presented in Table E2-3) to be processed and delivered to UBC. While 68.8% of emitted GHG (2.62E+06 kgCO2eq) is attributed to the gasification stage with mainly biogenic emissions, wood transport (1.12E+06 gCO2eq) shares 29.4% GHG emissions and wood processing (6.69E+04 kgCO2eq) is responsible for  remaining 1.8%  and (Figure 5.8). 112   Figure 5.8 Scenario 3: Annual GHG emissions [kgCO2eq] per life cycle stage for biomass.  The total amount of emitted GHG, here calculated to be 3.81E+06 kg CO2eq (Table E2-3), is  one order of magnitude smaller than in Scenario 2 where total annual GHG of  7.08E+07 kg CO2eq were emitted when the same amount of energy was produced solely by natural gas. The complete replacement of natural gas with wood waste could therefore annually reduce GHG emissions for   67 ktCO2eq which is more than 90 % reduction in GHG emissions.26  One of the major contributors to GHG, CO2, is released wherever fossil fuels are used such as for wood transport and wood processing with equipment utilizing diesel. Wood processing also contributes to emissions of CH4, CO and SOx, but emissions of CH4, SOx and CO also come from transportation. The portion of the emitted pollutants are biogenic in nature which helps to minimize GHG emissions and global warming impacts.                                                   26 1 kilogram [kg] = 1.00E-06 kiloton (metric) [kt]. 113  Pollutant contributions to emissions by process stages for scenario 3 are depicted in Figure 5.9.  Figure 5.9 Scenario 3: Pollutant emissions contributions over different life cycle stages.  To address uncertainty, a sensitivity analysis was performed by changing the wood waste transport distance. When the distance was set to be 150 km per trip instead of 78.8 km per trip (27.2 km per trip  from industry to Cloverdale and 51.6 km per trip from Cloverdale to UBC), GHG result just for transportation segment doubled,  from 1.12E+06 kg CO2eq to 2.13E+06 kgCO2eq (Figure 5.10). The gasification stage still remained the main contributor to GHG mainly by N2O and biogenic emissions of CH4 (see Table E2-4). The reduction of GHG compared to scenario 2 is 98.37 ktCO2eq, implying that increased transportation distance added 1.01 ktCO2eq/year. 114   Figure 5.10 Scenario 3: Annual GHG emissions [kgCO2eq] per life cycle stage for biomass with increased transportation distance.  5.3.2. Impact assessment approach 2  All energy consumption, emission data for unit processes and transportation (related to energy and fuels production and use) were input into SimaPro software for analysis.  Impact assessment was conducted using IMPACT2002+ v 2.12 methodology. The focus of this assessment is global warming. Mid-point is a convenient impact category as  further pathway (damage categories) may be associated with higher uncertainties (Olivier et al., 2003), although they are being commonly used in a number of studies (Pa et al., 2011; McManus, 2010), and lately with improved assessment methods (Weldu et al., 2017; Notter, 2015).   Characterization factors (CF) for global warming reflects only the emissions into the air. A separate damage category climate change is identical to global warming midpoint and also expressed in [kgCO2eq/kg]. The version of SimaPro used in this study (8.2.0.0) included IMPACT 2002+ assessment methodology with GWP for 500-year time horizon so the method 2.13E+066.69E+042.62E+06Scenario 3: total transportation changed to 150 kmWood transport Wood processing wood gassification115  was adapted for this study with a 100-year time horizon for global warming potentials. GWP presented in the IPCC 5th Assessment report (IPCC, 2013), and  suggested in IMPACT2002+ User Guide (Quantis, 2012),  for fossil CH4 characterization factor (CF) of 27.75 kg CO2eq /kg  and for biogenic CH4 of 25 kg CO2eq /kg to reflect the fact that CO2 produced from biogenic CH4 in the atmosphere is neutral.    LCA method is generally more appropriate for evaluation of global impacts since generic characterization factor, and therefore iF, represent average conditions (such as population density) of a broader area or a region, therefore the method is not sensitive to variation  of site-specific local conditions for accurate assessment of community-related impacts such as exposure and human health. Thus, impacts related to global human health were not evaluated by this method since the local health impacts have been already covered in Chapters 3 and 4 of this study. Mid-point results for global warming/climate change comparing the three selected scenarios in this study are presented in Table 5.4.  Table 5.4  Mid-point impacts for annual energy output of 1,011 TJ at UBC Point Grey campus. Mid-point Category Units Scenario 1: NG, oil and biomass Scenario 2: NG, oil  Scenario 3: Biomass Global  warming kg CO2 eq 6.18E+07 7.57E+07 1.95E+06  It should be noted that both upstream and downstream processes are included in the life cycle analysis. However, since biomass was utilized from local sources as waste wood, stages such as harvesting and long-distance transportation are avoided, which could be noticeable contributors 116  to overall impacts (Pa, 2010). Utilizing locally sourced wood waste can reduce GHG emissions by approximately 97% compared to heat produced by natural gas only or a combination of natural gas, fuel oil and biomass.  The use of biomass in combination with natural gas and oil (Scenario 1) lowers the global warming impacts which would otherwise be more significant in case of only fossil fuels utilization (Scenario 2). The use of biomass in Scenario 3 achieves noticeable reductions in global warming impact from 7.57E+07 to 1.95E+06 kg CO2eq compared to the use of natural gas only.  Previous case studies for UBC campus heating (Pa, 2010) reported GHG emissions reduction between 79% and 83% when fossil fuels are replaced with biomass. The previous study, however, considered upstream emissions from harvesting and sawmill operation, used longer transportation distances for wood residues delivery to UBC which could have added to GHG emissions, and estimated BRDF emissions as the plant was not built yet. In addition, the study considered different period with calculation based on total energy demand equal to 974 TJ. This study used actual performance data for fuel consumption for both plants and the 2012-2013 period of operation with 1,011 TJ energy produced.  As  most LCA studies confirmed, when biomass replaces fossil fuels, a significant net reduction in GHG could be achieved (Cherubini and Strømman, 2011).  A study by Parajuli et al. (2014) evaluated DES in Denmark where straw was used for district heating. They found the reduction 117  of GHG27 to be 187g/CO2eq per MJ heat production when gasification technology is used instead of combustion. A review (Patel et al., 2016) which compared biomass conversion technologies for energy production outlined studies where 8.8 to 10.5 gCO2eq was achieved as the reduction in GHG emissions per MJ of energy produced in case of CHP gasification and 80 -110 gCO2eq/MJ in case of biomass combustion for heat only production. A Finish study (Havukainen et al., 2018) which investigated small-scale biomass CHP, concluded that the replacement of fossil fuels with biomass for heat production can result in 59–66 gCO2eq./MJ energy reduction with biogenic emissions included. The findings are in line with this study where reduction of GHG of 66 gCO2eq/MJ heat production is estimated for total replacement of natural gas by biomass.  It should be noted that different boundary framework, different softwares (such as GREET, SimaPro, TEAM, GHGenius, GaBi ) are used across studies along with different impact assessment methods so  direct comparison may be a challenge. 5.4 Conclusion The release of greenhouse gasses into the atmosphere is responsible for global atmospheric warming and consequently the changes of Earth’s climate. One of the major contributors to greenhouse gas emissions in BC is energy sector. In this study assessment of greenhouse gas emissions from a district heating system at the UBC campus was studied. Three scenarios were considered: Scenario 1, the base case scenario as existed in 2012-2013 where 80% of base load system produced energy by natural gas combustion at the Power House with an addition of a small peaking demand met by fuel oil and approximately 20% was supplied by a biomass                                                  27 Study (Parajuli et al., 2014) uses different GWPs for CH4 (25) and N2O (298) for a 100-years horizon than this study which used GWP for CH4 fossil (27.75) and N2O (265) for both assessment approaches. 118  gasification plant. Scenario 2 considered the same annual energy demand but met only by natural gas, whereas scenario 3 investigated global impacts in terms of GHG emissions in case that the whole energy demand was met by biomass gasification plant on campus. Upstream and downstream life cycle stages were considered for the production and use of natural gas and oil whereas only collection, transportation and use stages were considered for biomass with wood waste collected locally.   It was concluded that the total amount of emitted GHG from Scenario 3  (3.81E+06 kg CO2eq) is one order of magnitude  smaller than in Scenario 2 where total annual GHG of  7.08E+07 kg CO2eq were emitted when the same amount of energy was produced solely by natural gas. The replacement of natural gas with wood waste could therefore reduce GHG emissions for more than 90% in case of wood waste being sourced locally. The analysis of stage-wise emissions per pollutant in case of natural gas being the only fuel used (Scenario 2) indicted that the major contributor to CO2, N2O and CO emissions is natural gas combustion whereas upstream processes are less intense in emissions and are associated with emissions of CH4, NOx, SOx and NMVOCs. Scenario 3 with biomass indicated that CO2, a major contributor to GHGs, is released wherever fossil fuels are used for wood residue processing and transport. Increasing transportation distances from 78.8 km to 150 km for biomass scenario could double GHG emissions from transportation segment and add 1.01 kt per year of GHG to the atmosphere. 119  Chapter 6: Economic valuation of district heating options 6.1 Introduction Air pollution costs global economy more than US $5.11 trillion in welfare losses each year. This metrics incorporates costs associated with health and consumption. Monetized losses due to absence from work (lost income) alone cost global economy US $225 billion annually (The World Bank and Institute for Health Metrics and Evaluation, 2016). North America’s welfare losses are 3% of GDP,28 2013 equivalent, while at the same time the greatest losses are in East Asia and the Pacific where costs of premature death from air pollution reached 7.5% of GDP.  Policy actions in many countries which target air pollution reduction focus primarily on reduction of greenhouse gases (GHG) and fossil fuels.  Systematic literature review by Akhtari et al. (2014) emphasizes policies and government incentives such as CO2 taxes and tradable carbon credits as ones which can play a substantial role in making biomass an attractive choice for district heating. Economics of a variety of GHG abatement options including biomass as a fuel to replace fossil fuels is extensively covered in literature.    An Austrian study (Kalt and Kranzl, 2011),  suggested that the abatement costs associated with GHG mitigation and fossil fuel replacement  will depend on the technology selection, feedstock type,  plant size and site-specific combined heat and power (CHP) plants operating conditions. The authors showed that when oil-fired boilers and gas-fired heat generating plants are being                                                  28 GDP – Gross Domestic Product. 120  replaced with wood-based heat generating technologies abatement costs ranged from  - 45 €/t CO2eq ( – 11 €/MWh-HHV) to 93 €/t CO2eq (24 €/MWh-HHV), respectively.  The authors concluded that using biomass for the heat generation and CHP are the most cost-effective solutions for Austria in terms of GHG mitigation and fossil fuel replacement. In addition, the study showed that wood-based heating systems are more economic and have a lower GHG abatement cost if they operate at the higher annual operating hours at the full load.  Heating systems with 50 kW capacity have the best economic efficiency (€/t CO2eq).    Another study conducted in Portugal estimated 17,981 t CO2eq/year avoided emissions in case of investing in biomass power plants based on dedicated energy crops. However, the financial viability for such projects may be difficult to estimate as the costs for energy crops supply chains could be higher than for the power plants (Carneiro and Ferreira, 2012).   Some costs of air pollution have not been included regularly in economics of technology selection, such as external costs or externalities (Li et al., 2015). One of the most important externalities is human health which when monetized can demonstrate a burden of disease caused by air pollution impacts and costs due to morbidity or premature deaths. Economic valuation of human health can well express the interest of a society for trade-offs (what people are willing to give up in alternative to choices related to consumption) for benefits in environmental quality (Bell et al., 2008). For example, reducing daily average PM2.5 levels to prescribed air quality (AQ) standards in China could reduce emergency departments visits and deaths from respiratory 121  diseases for 23 - 42 M yuan (approximately $4.4 to 8.2 M)29  and 25 - 670 M yuan (approximately $4.9 to 130 M)  per year in 2015 yuan, respectively (Chen et al., 2017). A non-monetized approach such as physical health impact indicator DALY (Disability-Adjusted Life Year) is highly recommended (Bachmann and van der Kamp, 2017) for policy communication and quantified in many studies related to biomass applications (Jana and De, 2017; Martenies et al., 2015; Pa et al., 2013; Perilhon et al., 2012; Pa et al., 2011).   The objective of this chapter is to estimate the Net Present Value (NPV) of the biomass district heating system so to evaluate if a sum of discounted cash flows associated with considered benefits (savings in taxable GHG emissions, avoided fossil fuel procurement costs, etc.) overweight cost associated with such system (e.g. new capital investment, variable operational and maintenance costs, wood fuel costs). Costs and benefits of health-related impact are also discussed. 6.2 A summary of reported UBC district heating costs and GHG emissions As mentioned in Chapter 4, energy production to meet the UBC campus energy demand for the base year of 2012-2013 was calculated to be 1,011 TJ comprising of natural gas for base load and fuel oil for peak load. In addition, BRDF (Bioenergy Research Demonstration Facility) started operation contributing approximately 20% of total energy production in that year. During the first year of BRDF operation, steam production from BRDF was reported to be 148,920 KLBS which corresponds to 188 TJ of thermal energy. A 100% plant’s uptime as a very                                                  29 $ denotes Canadian dollars; $1M denotes $1,000,000; $1K denotes $10,000.  122  conservative approach and equal monthly steam production are assumed in this study.  During the same period, a total of 907 TJ energy was produced by PH, with 893 TJ using natural gas for base load and almost 14 TJ from fuel oil for peak load over the period of December 2012. (Figure 6.1).   Figure 6.1 Energy demand for the UBC campus in 2012-2103.  The total thermal capacity of BRDF is 5.8 MWth from steam and 2.8 MWth from heat recovery when operating in heating mode, and 1.96 MWel when operating as CHP (UBC, 2015a). This implies that at the full capacity BRDF can generate 75,336 MWhth or 271 TJ30 of heat annually. Based on the latest UBC report (UBC, 2017), the use of natural gas (and therefore UBC’s costs) is decreasing since 2015 whereas the use of biomass slightly increased. That could likely be attributed to the new ADES (Academic District Energy System) project and more efficient use of fuel.                                                  30 1MWh = 3.6 GJ; I TJ = 10E+03 GJ. 123  6.3 Economic valuation methodology A simplified economic analysis presented here is solely based on the biomass conversion at the UBC campus and only for thermal mode. Input data were obtained from UBC reports (Wauthy and Giffin, 2017; UBC, 2015b; UBC, 2014; UBC, 2011; UBC, 2010), and literature sources. All available costs reported were summarized, input in MS excel spreadsheet as capital cost and operational and maintenance cost (O&M) so future values (FV) were calculated for the assumed 20-year life period of the considered plants (NREL, 2016). An annual inflation rate of 2.04% was used based on the Bank of Canada31 historic data as a 10-year average with a commercial interest of 6.5% adjusted for inflation32 to give an effective interest rate of 4.37%. Since the focus of this part of study (Chapters 5 and 6) is GHG emissions and costs and benefits related to abatement of GHG emissions, for calculating the present value, PV (2012 $), of operating costs the following equation was used (Field and Olewiler, 2005):  PV = ∑ (𝑂𝑀𝑛)/𝑛=20𝑛=0 (1 + 𝑟)𝑛                                                                          (6-1)  Where:  PV     is the present value of annual costs,  OMn  is the total operational and maintenance cost for period n, referring to thermal mode,  r         is the effective commercial interest rate adjusted for inflation, n         is the time period (n = 0 for 2012 and n = 20 for the end of life time).                                                   31Source:  http://www.bankofcanada.ca/rates/indicators/capacity-and-inflation-pressures/inflation/historical-data/ (Accessed August 28, 2016). 32 Source: http://www.calculatorsoup.com/calculators/financial/investment-inflation-calculator.php (Accessed August 28, 2016). 124  The base year for assessment (n=0) is 2012 and different operational options were considered as specified in section 6.4. Additional economic analysis was performed to discuss external costs which refer specifically to air pollution.  6.3.1. Assessment of costs and benefits associated with the development, operation and maintenance of biomass-based district heating at UBC BRDF was constructed in the period 2010 - 2012 and became operational in 2012. The construction was completed on time with the contribution of the following funding sources:  Sustainable Development Technology Canada, NRCan - Canadian Wood Council, NRCan Clean Energy Fund, BC Bioenergy Network, FP Innovations/Ministry of Forests, BC Innovative Clean Energy Fund, Western Economic Diversification, Nexterra (In kind), UBC Building Operations/Energy and Water Services.  Capital investment for building the BRDF was reported to be $27.4 M.33  It included  Nexterra plant equipment procurement and installation (roughly $16.4 M), plant building construction (over $5M), utilities connections, planning and design fees, permits, insurance, project management, retained risk fee. However, since only thermal mode was considered in this study some assumptions are made below along with a summary of parameters used in calculations:                                                  33 Planned budget was $26M (UBC, 2010). 125  a) Capital cost of the equipment associated with thermal mode is estimated to be $8.2 M and this cost was used as capital cost in the calculation, which is 50% of the capital cost for the CHP Nexterra equipment.  b) Total capital cost for heating mode, TCCh, (including the building and other costs cited above) used in calculations is $19.2 M  obtained by subtracting the engine cost:  TCCh = $27.4 M - ($16.4 M/2)                   (6-2)  c) Annual operation and maintenance cost (O&M) during the first year of BRDF operation was estimated as 5% (Delivand et al., 2015) of the TCCh cost which is $96 K.  d) The delivered cost of biomass is $69/ODMT ($62/OMDT+GST+PST) in 2012 cost as contracted and supplied by Cloverdale.  In addition, as a public sector organization UBC pays carbon offsets34 on all commodities including biomass (only for CH4 and N2O emissions) at $0.06/GJ  (for emission factor of 2.24 kg CO2eq/GJ for wood fuel) whereas carbon tax is only paid for natural gas (Wauthy and Giffin, 2017; UBC, 2010).  e) All-in cost of fossil fuel of $10.47/GJ included the cost of fossil fuel delivered and all taxes (equal to $7.72/GJ), cost of carbon tax of $1.5/GJ and carbon offset of $1.25/GJ (for natural gas GHG emission factor 49.87 kg CO2eq/GJ for combustion only), (Wauthy and Giffin, 2017; UBC, 2010).  f) Since there were no new capital investments in the PH, capital cost will remain equal for both options considered in economic assessment and therefore it was excluded in comparison for simplicity; this postulation does not impact the difference in total cost among scenarios.                                                  34 Carbon taxes of CAN $30/t CO2eq are paid according to the Carbon Tax Act and Carbon Offsets of $25/t CO2eq are purchased according to the Carbon Neutral Government (CNG) Regulation. 126  g) Ash generation is 119 t/year but 50% is used on campus whereas the other 50% is disposed at a cost of $100/t which equals $5,950 annually (UBC, 2010). h) Fuel consumption used in this analysis for all scenarios is calculated in Chapter 4.   Economic parameters used in the current analysis are summarized in Table 6.1 below.  Table 6.1 Economic parameters.  Parameter Value  Capital investment   a) Capital cost for Nexterra equipment for thermal mode [$] 8.2 M35                    b)  Cost of land, building construction, installation, permits) [$] 11 M Operating and Maintenance (O & M)  Maintenance [$/year] 96 K Fuel biomass [$/ODMT+taxes] 69 Carbon offset purchase for wood fuel [$/GJinput] 0.06 All-in fuel natural gas cost [$/GJinput]a 10.47 Ash disposal [$/t] 100 Operators’  salary a [$/year] 60.2 K DHS useful service life [year] 20 Annual inflation rate [%] 1.88 Nominal commercial interest rate [%] 6.5 Effective commercial interest rate adjusted for inflation 4.37 Equipment depreciation rate [%] 30 a Source: (UBC, 2010), basic NG price based on 3-year average. b Source: http://www.fin.gov.bc.ca/tbs/tp/climate/A4.htm (Accessed October 8, 2017).  As for the steam plant power house (PH) total O&M costs for base year (2012) were $2.9 M. 6.4 Results and discussion The following options are considered:                                                   35 Cogen plant was sold as package with a fixed price of $16.4 M from Nexterra. Thermal mode only specification was not available so 50% of total cost was assumed. 127  A) Operation of only PH where natural gas was used to meet energy demand of 1,011 TJ for 2012.  Since the building and equipment already existed, no capital investment was considered in this case. Fuel all-in costs for meeting entire energy demand were estimated to be $11.9 M out of which $3.1 M was spent on carbon costs alone. B) Operation including both PH and BRDF as of 2012 (scenario 1 in previous chapters) where 822,965 GJ was produced by PH and 188,061 GJ by BRDF requiring 9,304 ODMT of wood which costed $64.2 K; $1.7 K was spent on purchased carbon offsets; in addition, all-in fossil fuel costs were $9.6 M out of which $2.5 M was spent on carbon taxes; Capital cost for the BRDF (heating mode only) of $19.2 M was included in total costs.  The main calculated costs are summarized in Table 6.2. Present value was calculated for each option, for option A only O&M costs and for option B total PV included capital cost and O&M costs. For option A where only O&M were considered due to already existing PH infrastructure, annual O&M was $14.8 M and calculated total PV $209 M.  For option B, capital cost of $19.2 M was added to O&M PV of 208.7 M which resulted in a total PV for this option of $227.9 M.   Table 6.2 Summary of calculated parameters [in $2012]. Parameter Option A Option B  PH only             PH  and BRDF Annual energy output [GJ] 1,011,026 822,965 188,061 Annual energy input [GJ]  1,133,232 918,332 276,560 Annual fuel cost [$/year] 8.7 M                7.1M  64.2 K Annual Carbon Tax [$/year] 1.7 M                1.4 M      0 Annual Carbon Offset [$/year] 1.4 M 1.1 M   1.7 K Annual other O&M costs [$/year] 2.9 M 4.5 M PV (O&M) costa  [$ ] 209 M 208.7 M Capital cost -   19.2 M a There are other O&M parameters (like ash disposal, salaries,  maintenance etc.) which are included in calculation but not presented in this table. 128  With respect to O&M costs, option A included fuel costs and carbon costs which for natural gas included a carbon tax of $30/t CO2eq and a carbon offset of $25/t CO2eq. With an addition of BRDF to DES, as in option B, costs associated with carbon tax are avoided for the portion of biomass used (about 20% of heat generated). However, since old steam power plant PH was still producing about 80% of heat, high O&M costs associated with aging PH were still substantial. Option B also had a substantial capital expenditure due to new BRDF infrastructure, resulting in  total PV of $227.9 M.    The economic impact of introducing biomass in DES can be demonstrated through the Net Present Value (NPV), calculated as the difference  between options A (NG only) and option B (combined NG and biomass). The net present value (NPV) between options A and B is $ -18.9 M which indicates that the introduction of a biomass plant (option B) increased the cost of district heating at UBC campus.  With the parameters as selected in this study, the results in Figure 6.2 show that carbon tax and carbon offset are dominating the life cycle costs followed by the O&M costs which are higher for option B due to combined costs for both plants. The use of wood fuel leads to savings in carbon offsets and carbon tax (which indicates benefits of introducing biomass), because carbon offset is much higher for natural gas ($1.25/GJ) than for biomass ($0.06/GJ). Savings in carbon taxes is an incentive that can support biomass based DES providing that other conditions and associated costs are well justified. 129   Figure 6.2  Cost breakdown for two DH options at UBC.  A techno-economic study was carried out in Italian context (Arena et al., 2010) to evaluate two design configurations of a biomass-to-energy small scale plant. It concluded that the variation in biomass cost is less important than all-inclusive feed-in tariff which crucially affects economic parameters such as averaged discounted cash flow. However, the authors emphasized that site-specific variables such as heat demand and the costs of the waste treatmant and disposal should be taken into consideration. Another study (Börjesson and Ahlgren, 2010) evaluated cost-effectivness of different applications of biomass gasification and found that CHP plant is a cost-competetive choice in situations of high heat demand. An important parameter is the availability of  local low-cost biomass. A North American study by Young et al. (2018), investigated driving factors for small scale biomass applications in 3142 counties. Among the most important are PH only PH and BRDFPV Other O&M 41 63PV Carbon cost biomass 0 0.23PV Biomass fuel 0 9PV Carbon cost NG 44 36PV Fosill fuel cost 124 100Capital cost 0 19050100150200250[$  M]PV [$ M]130  Heating Degree Days (HDD),36 natural gas processing and available biomass as well as regional government initiatives and national financing policies. 6.4.1 Addressing uncertainty Economic estimates are associated with a number of uncertainties which range from fuel prices to inflation rates. To address uncertainty in fuel price, natural gas prices were analyzed from historical trends to forecasted prices. Historical data show that natural gas reached the highest price in December 2005 of US $15.39/MMBtu (equal to US $13.8/GJ).37  Forecast data predict an increase in the next few decades. For example, based on EIA forecast for Henry Hub  as presented in Wauthy and Giffin (2017), natural gas price can increase to $12/GJ by 2035. If other parameters are kept the same as previously explained, the all-in price for natural gas would come to $16.19/GJ which is an increase of 35% compared to current price. This increase will affect total cost over plant life time resulting in a present value of $301M for option A (natural gas only) and $302 M for option B (combined NG and biomass). It appears that an increase in natural gas price alone can bring total costs of option A close to option B.  Sensitivity analysis was also performed to investigate how carbon tax changes would influence the PV of proposed operating scenarios at present fuel prices. In 2016 Government of Canada proposed a carbon pricing framework to reduce GHG emissions and grow green economy (Canada, 2017). By 2022 price should reach $50/t CO2eq. This amount was used to re-evaluate                                                  36 HDD is the number of degrees [ºC] that a day’s average temperature is below 18 ºC indicating that a building requires heating; this metrics quantifies energy demand. Source: https://www.investopedia.com/terms/h/heatingdegreeday.asp (Accessed April 5, 2018). 37 Source: https://tradingeconomics.com/commodity/natural-gas (Accessed March 30, 2018).    1 MMBtu = 1.055056 GJ. 131  presented scenarios. Since carbon tax is applied only to natural gas, it will result in an all-in price of $11.47/GJ provided that other price components remain unchanged. It was found that an increase of $20/t CO2eq in carbon tax from the current $30/t CO2eq will increase the PV by $16 M (to $225 M) for option A but only $13 M (to $241 M) for option B which indicates benefits in the form of saved expenditures on carbon tax when biomass was replacing fossil fuel even partially. It appears that savings in carbon taxes at a higher tax rate will offset part of the capital investments for bioenergy plant. 6.4.2 Trade-offs associated with the selection of district heating options Switching to biomass for district heating applications have numerous advantages, the reduction of GHG and consequently global warming, and savings in carbon-related taxes as demonstrated by this study (Chapters 5 and 6). At the same time, local air quality and human health may be compromised due to proximity of a biomass plant to local population (Chapter 4).   External costs of selected pollutants are presented in Table 6.3. External costs are taken from Pa et al. (2013), and total emissions are calculated in this study and presented in Tables E2-2 and E2-3. Since health impacts are more of a local character, only emissions from the plants were considered. Biogenic CO2 from wood combustion is assumed to have zero contribution to net GHG emissions as previously explained. It should be noted that both generic and biogenic components of CO and CH4 were taken into account although biogenic components of these compounds have slightly lower impacts on climate change. Here, emissions of each CH4 and CO are presented as a sum of each compound’s generic and biogenic component, so external costs 132  for these compounds were assumed to be an average value of their respective components’ values.  Calculations indicated that the total annual external costs for option A (when PH is operational only) is $2.08 M/year and $29 M over plant’s life time, most of which, over $25 M is attributed to CO2. An addition of biomass to natural gas for DES (option B) decreases external costs to $25.7 M over plant’s life time (which could be considered as benefits) with $20.5 M attributed to CO2 solely.    133  Table 6.3. Summary of externalities for district heating options at UBC  Pollutant $/kg emitted Option A [kg/year] Option A [$/year] Option A PV[$] Option B [kg/year] Option B [$/year] Option  B PV [$]  NG  NG NG NG+biomass NG+ biomass NG+biomass CO2 fossil 0.032 5.57E+07 1.78E+06 2.52E+07 4.45E+07 1.42E+06 2.02E+07 CO2 biog. - - 0.00E+00 0.00E+00 2.50E+04 0.00E+00 0.00E+00 CH4 0.24 1.07E+03 2.57E+02 3.64E+03 2.58E+04 6.18E+03 8.75E+04 N2O 4.5 1.02E+03 4.59E+03 4.59E+03 2.38E+03 1.07E+04 1.51E+05 NOx as NO2 5.23 4.64E+04 2.43E+05 3.44E+06 5.75E+04 3.01E+05 4.26E+06 SOx 4.01 0.00E+00 0.00E+00 0.00E+00 6.24E+03 2.50E+04 3.54E+05 PM2.5 25.60 8.82E+02 2.26E+04 3.20E+05 8.97E+02 2.30E+04 3.25E+05 CO 0.68 3.90E+04 2.65E+04 2.57E+04 3.53E+04 2.40E+04 3.40E+05 NMVOC 1.47 2.55E+03 3.75E+03 5.31E+04 3.24E+03 4.76E+03 6.74E+04 SUM 4.17E+01 5.58E+07 2.08E+06 2.90E+07 4.47E+07 1.82E+06 2.57E+07  134  Analysis per pollutant (Figure 6.3) revealed that the largest externalities are emissions of fossil CO2   from natural gas burning, $1.78 M/year if all energy demand is met by natural gas (option A). This cost could be interpreted as a monetary damage due to global warming from the combustion of natural gas (base load).   Introduction of biomass to DES (option B) resulted in decrease of total costs associated with CO2 emissions to $1.42 M/year, and savings of $5 M over plant’s life time.  While biomass has a benefit in terms of global warming, emission of fine particles and nitrogen oxides are of concern for local air quality and human health. $2.26 K/year could be the expected costs for human health impacts from fine particulate emissions from natural gas combustion, whereas those costs for fine particles increase to $2.30 K/year when biomass on-site gasification plant is operational in addition to PH (option B). On a 20-year (plant life time) basis, expected costs for human health impacts  due to PM2.5 emanations could be $32.00 K and $32.50 K  for natural gas (option A) and natural gas and biomass (option B), respectively. Since uncontrolled NOx emissions from biomass use are higher than from natural gas, external costs are also around 20 % higher for option B where biomass was introduced in addition to fossil fuels ($30.10 K/year) than the natural gas only scenario ($24.30 K/year). On a 20-year basis, NOx external costs from natural gas were estimated to be $3.44 M and $4.26 M from combined natural gas and biomass option. These pollutants would make impact on local air quality and fine particulates in particular should be of concern with respect to human health impacts.    135                Figure 6.3 External costs for option A (natural gas only) and option B (natural gas and biomass) for the period of plants’ life time.   It appears that both technologies have advantages and shortfalls when it comes to emissions and impacts. Even when emissions are monetized, the same conclusions appear to be valid for trade-offs between these two options as the ones presented in Chapter 4: biomass is a superior choice when global impacts especially global warming and consequently climate change are of primary concern. Policy incentives of non-taxable emissions originating from renewable sources favor such outcome.  However, caution should be exercised with choosing the location in order to minimize population exposure to air pollutants. On the other hand, fossil fuels, natural gas in particular as a clean-burning fossil fuel would impose less local impacts but will cause damages to the global environment in terms of global warming and climate change.   The following table (Table 6.4) summarizes the key findings with respect to DES options considered for UBC. The use of natural gas to meet UBC heat demand entirely could be a preferable choice in terms of impacts on overall community exposure as IS can increase with the 0510152025CO2fossilCO2biog.CH4 N2O NOx asNO2SOx PM2.5 CO NMVOC$ MPV [$ M]Option A Option B136  addition of biomass mainly due to uncontrolled NOx emissions but fine particles as well, at the current plant locations. However, introducing biomass could reduce life cycle GHG emissions by 12,490 t (17.6%) and total external costs over the plants’ life time by more than $3 M which will contribute to climate change and global warming mitigation. On the other hand, total costs (over plants’ life time) associated with the inclusion of biomass in DES expressed as total PV (capital and O&M costs) are higher by almost $19 M than the existing steam power plant run on natural gas.  Table 6.4 Summary of key findings on local and global impacts for UBC district heating options. Parameter NG only (PH operational) NG and biomass (PH and BRDF operational) ∑ IS scenario [DALY] 495 513 Life Cycle GHG emissions [t CO2eq] 70,809 58,319 Total PV [$] 209 M 228 M Total PV externalities [$] 29 M 26 M Note: highlighted are advantageous parameters.  This analysis points out the importance of including societal costs into analyses of DES. Typically, techno-economic analysis serves for decision making on technology choices and location selection (to minimize infrastructure and maintenance costs) neglecting the externalities which could inform decisions in order to better protect human health and the environment. For example, biomass-related projects give priority to location assessable for trucks to bring biomass, a spatial lot for truck maneuver, closeness to distribution system (steam or hot water), etc. Permit application requires assessment of air quality but does not require assessments in terms of location selection with least impacts to particular community. This study demonstrated that such 137  analysis is important to avoid exposure to fine particles in the first place and should be equally and regularly evaluated along with techno-economic factors.  Attempt was also made to tackle a possible expansion of the existing BRDF as one of the options UBC is considering for meeting the University’s Climate Action Plan (CAP) for GHG reductions goals of 67% below 2007 level by 2020. It appears that such option, where biomass would solely meet UBC’s energy demand, may lead to increased costs (Appendix F).   6.5 Conclusions It is widely accepted that many factors affect economics of biomass energy systems, but published studies only emphasized policies and government incentives such as CO2 taxes and tradable carbon credits which can play a significant role in making biomass an attractive choice for district heating systems. Some social costs of air pollution, like health impacts, have not been regularly included in economics of technology selection and as such they are considered to be external costs or externalities. One of the objective of this study was to tackle economics as part of the assessment of biomass district heating systems as a third pillar of sustainability, in harmony with environmental (ecological) and social in a path towards sustainable development.     A simplified economic analysis which focused on operational and maintenance costs and fuel procurement was performed to obtain O&M PV and total PV which included capital investment of district heating system at UBC Point Grey campus. Economics of two options is considered following the assumptions from previous chapters of this study: an originally existed option with PH producing heat for the campus with energy demand as of 2012-2013 and an option (scenario 138  1 as named throughout the study) as it was in period 2012-2013 when most of the steam (for heat and hot water on campus) was produced by natural gas and peak fuel oil at PH and almost 20% steam by BRDF using biomass. It is concluded that an introduction of biomass to DES increased total costs (total PV included capital and O&M costs) by $19 M compared to existing PH although some savings in carbon tax were generated at $8.4 M over the period of plants’ life time (20 years).    When externalities are considered, namely, assigning monetary values to air pollutants emitted at the site both by PH and BRDF, $26 K could be avoided annually and $3.3 M over plants’ life time in terms of its societal damages38 when switching fossil fuel use to combined use of natural gas and biomass. With respect to individual pollutant damages, $2.30 K/year and $32.50 K over plants’ life time could be expected as external costs for human health impacts from fine particulate emissions from combined operation of an on-site biomass gasification plant and PH plant. The same combined operation resulted in $30 K/year and more than $4 M over plants’ life time in external costs from uncontrolled NOx emissions. These pollutants would make impact to local air quality and fine particulates in particular should be of concern with respect to human health impacts. Their respective external costs are lower for the natural gas operation originated solely from PH.  Emissions of CO2 from solely using natural gas (option A) would cost the society $1.78 M annually and more than $25 M over plants’ life time. Savings of $5 M could be expected with an introduction of biomass to DES.                                                    38 Damages avoided represent benefits.  139  It appears that biomass is a superior choice when global impacts especially global warming and consequently climate change are of primary concern. However, caution should be exercised in choosing the location in order to minimize population exposure to air pollutants. On the other hand, fossil fuels, natural gas in particular, have less local impacts but they will cause more damages to the global environment in terms of global warming and climate change. A proper compromise or trade-off should be considered in developing such district heating systems, based on a careful evaluation of local air quality impacts and global impacts as illustrated in this study. Sustainable cities and communities call for sustainable solutions where all aspects must be evaluated and balanced. Inclusion of externalities could inform policy makers of damages that could not otherwise be acknowledged in a typical techno-economic analysis.       140  Chapter 7: Conclusions and future research directions 7.1 Conclusions and significance of the research  Local impact assessments of biomass-based systems on air quality and the resulting community exposure are still in its infancy. Very few studies started recognizing the importance of local and urban health impacts of near-by stationary sources. Due to either lack of data or project purposes, those previous studies relied on many assumptions and did not account for dynamic population changes and actual spatial and temporal variations of ambient air quality (Martenies et al., 2015), or relied on selected archetypal environments and emission sources. Systematic literature review identified that there has been a lack of appropriate and accurate impact assessment methodology for parameters with extensive variability on local scale, and lack of assessment of biomass-based DES impacts on local ambient air quality and human health based on methods with higher accuracy and inclusive of local, site-specific characteristics.  To address these knowledge gaps, a systematic study has been conducted with UBC BRDF bioenergy facility as a case to address the following research questions. (1). How would the inclusion of site-specific terrain, land use and microclimatic characteristics, variable population density and breathing rates improve accuracy of local air quality and population health impact assessment of community-based biomass energy systems? (2). How would an incremental increase of PM2.5, NOx and CO concentrations from investigated biomass DES contribute to local ambient air quality and population exposure? (3). How would life-cycle GHG emissions from the investigated biomass DES contribute to global warming? (4). Considering capital, 141  operational and maintenance (O&M) costs and externalities, how would the introduction of biomass-based DES affect the economics compared to fossil fuel-based DES?  An analysis on local air quality and human health impact by varying the spatial distribution of receptors, population dynamics, temporal population dynamics, and diurnal variations in people’s breathing rates revealed that when accounting for all local-scale variations, from actual population density (as opposed to averaged census data commonly used in assessments), to local spatial and temporal micro-climatological and local spatial orographical conditions at the biomass plant site, more realistic, site-specific results could be obtained.  When the dynamic variation of all parameters is accounted for, the real dynamic nature of iF is captured. Neglecting microclimatic characteristics such as site-specific diurnal circulation patterns which influence pollutant dispersion or short-term variation of parameters on a local scale such as population dynamics may lead to underestimation of iF by more than 20%. This amplifies the importance of incorporating both spatial and temporal dynamics in estimating the exposure (i.e. iF) in assessing the health impact of district heating systems in densely populated areas.  The improved methodology was then applied to the UBC district heating system with 2012-2103 operation as the base case and two other scenarios when all demanded heat would be produced only by PH using natural gas and all heat would be produced by an expanded BRDF using biomass, respectively. The results showed that the health impacts from a biomass-based energy system installed with an efficient PM control device mainly resulted from the uncontrolled NOx emission, followed by PM and CO emissions, among all criteria air pollutants. The lowest iF for this option indicates the importance of the plant location relative to community setting where the 142  smallest number of people would be affected by plant emissions since iF is mostly influenced by the number of people exposed. On the other hand, it appears that a distributed DES with combined NG and biomass may have an advantage over a centralized DES in terms of overall health impact. Depending on plant locations compared to the location of a single plant (like in case of PH), distributed DES may have lower overall health impacts compared to a single biomass energy supply system.   With respect to pollutant contributions to air quality and health risks, it was found that the overall incremental contribution of fine particles (PM2.5) was at least one order of magnitude lower than the provincial air quality objectives (BCAQO). However, the maximum PM2.5 emission from the natural gas fueled PH could adversely add to the already high background concentrations.  Nitrogen dioxide (NO2) emissions from the BRDF with no engineered pollution controls in place exceeded BCAQO in all seasons except during the summer. It should be noted that CALPUFF predictions could be lower than actual outdoor concentrations originating from the considered sources so regular measurements can provide better insights in possible concentration exceedances. The impact score, IS, was the highest for NO2 (677 DALY) when biomass entirely replaced fossil fuels, and the highest for PM2.5 (64 DALY) if all energy was produced by natural gas. Complete replacement of fossil fuels by one biomass plant can result in almost 28% higher health impacts (708 DALY) compared to 513 DALY when both BRDF and the PH are operational mostly due to uncontrolled NO2 emissions.   Global impacts of emitted pollutants from BRDF and PH were investigated in terms of life cycle greenhouse gas emissions. It was concluded that the total amount of emitted GHG (3.81E+06 kg 143  CO2eq) from Scenario 3 where the entire energy demand is met by biomass is  one order of magnitude smaller than in Scenario 2 where total annual GHG of  7.08E+07 kg CO2eq were emitted when the same amount of energy was produced solely by natural gas. The replacement of natural gas with wood waste could therefore reduce GHG for more than 90% in case of wood waste being sourced locally. The analysis of stage-wise emissions per pollutant in case of natural gas being the only fuel used (Scenario 2) indicted that the major contributor to CO2, N2O and CO emissions is natural gas combustion whereas upstream processes are less intense in emissions and are associated with emissions of CH4, NOx, SOx and NMVOCs. Scenario 3 with biomass indicated that CO2, a major contributor to GHGs, is released wherever fossil fuels are used for wood residue processing and transport. Increasing transportation distances from 78.8 km to 150 km for biomass scenario could double GHG emissions from transportation segment and add 1.01 kt of GHG annually to the atmosphere.  Economics of the original PH producing heat for the campus with energy demand as of 2012-2013 and an option (scenario 1 as named throughout the study) and the setting when most of the steam was produced by natural gas and peak fuel oil at PH and almost 20% steam by BRDF using biomass, was evaluated. It is concluded that an introduction of biomass to DES increased total costs (total PV included capital and O&M costs) by $19 M over the plants’ life time compared to existing PH although some savings in carbon tax were generated at $8.4 M over the same period.     Introducing external costs into consideration, namely, monetizing emissions with respect to their impacts on global warming and human health, $26 K could be avoided annually and $3.3 M over 144  the plants’ life time in terms of societal damages when switching fossil fuel use to combined use of natural gas and biomass. With respect to individual pollutant damages, $2.3 K/year and $32.5 K over the plants’ life time could be expected as external costs for human health impacts from fine particulate emissions from combined operation of an on-site biomass gasification plant and PH plant. The same combined operation resulted in $30 K/year and more than $4 M over the plants’ life time in external costs from uncontrolled NOx emissions. These pollutants would make impact to local air quality and fine particulates in particular should be of concern with respect to human health impacts. Their respective external costs are lower for the natural gas operation originated solely from PH.  Emissions of CO2 from solely using natural gas would cost the society $1.78 M annually and more than $25 M over the plants’ life time. Savings of $5 M could be expected with an introduction of biomass to DES.   Overall, it appears that biomass is a superior choice when global impacts especially global warming and consequently climate change are of primary concerns. However, caution should be exercised in choosing the location in order to minimize population exposure to air pollutants. On the other hand, fossil fuels, natural gas in particular, have less local impacts but they will cause more damages to the global environment in terms of global warming and climate change. A proper compromise or trade-off should be considered in developing such district heating systems, based on a careful evaluation of local air quality impacts and global impacts as illustrated in this study. Sustainable cities and communities call for sustainable solutions where all aspects must be evaluated and balanced. Inclusion of externalities could inform policy makers of damages that could not otherwise be acknowledged in a typical techno-economic analysis.  145  7.2 Strengths and limitations of the research This study contributes to knowledge by developing an improved impacts assessment methodology for community-based biomass plants which is suggested for use by city planners, regulators and public health practitioners. A special focus is on local impacts which were scarcely covered in literature as the majority of published studies covered high-level global impacts.  Even though some studies indicated the importance of local impacts, inclusion of detailed local micrometeorology, population dynamics and varying breathing rates has not been covered before for district energy systems. This approach enabled  locally obtained dynamic iF and demonstrated that developed site-specific characterization factor CFhelth may bring more accuracy in assessing local impacts compared to the CFs based on consensus data and other population density data averaged over a large area, e.g. the European continent available in commercial life cycle analysis software packages such as SimaPro. Therefore the method proposed in this study (Chapter 3) is among the few studies that offer the accurate impact assessment on local air quality and human health.    This approach could be generalized and applied to communities and regions with complex settings, microclimatic conditions and varying population density, such as Metro Vancouver and Lower Fraser Valley districts in order to accurately evaluate impacts of growing biomass district energy systems, protect human health and ensure air-shed planning within the ecological carrying capacity. By analyzing in-depth the local impacts of a community-based operational biomass plant at UBC (which was selected as a case study) utilizing a newly proposed improved impact assessment methodology, this study led the way for future impact assessment approaches. The study also contributes to higher accuracy in the global impact assessment by utilizing BC 146  specific electricity mix and transportation data via GHGenius, a locally developed software ((S&T)2 Consultants, 2013).   Finally, by covering, local, global, economic and social aspects of biomass-based DES, this study demonstrated that all sustainability pillars should be included in an integrated impact assessment, which is especially important for the future development as trade-offs may be needed in order to protect environment and human health in a cost-effective manner.  Since BRDF was constructed and became operational in 2012, data analyzed in this study were used with a number of assumptions due to the limited data on emission testing. In this study, plant uptime was considered to be 100% and monthly steam production to be equal over the period 2012-2013 so to keep consistency in calculations. This assumption is a conservative approach and has not impacted the results or methodological approach.  Emission estimates from BRDF and Power House relied on published emission factors such as US EPA which are described as factor of varying quality in terms of number of measurements they were drawn from so those factors carry inherited uncertainties. Whenever needed, those factors were supplemented with ones published in literature originating from similar emission sources. Nevertheless, periodic emission tests at the plant were considered for comparison. The same applies for the ambient air quality tests which were available just from one station located on the roof of an adjacent building to monitor the impacts of emitted pollutants to buildings in close proximity of the plant. Measurements from one location present limitation in terms of further exploring local air quality based on monitoring data. 147  Life cycle assessment was carried out to quantify the global warming impacts due to the appropriateness of such methods for assessing global rather than local impacts. The importance of increased accuracy in assessment of local impacts associated with population exposure and health risks, which is the main strength and contribution of this study, is the main retraction in LCIA methodology so only global impacts were considered here using two approaches, both of which included foreground actual performance data wherever possible.   Economic analysis focused just on parameters that were of interest in this study, namely costs and benefits associated with GHG emissions and reductions; externalities were also covered based on monetized pollutant impacts from literature. The economic analysis does not include all costs and benefits associated with any of technologies but since the objective of this study is the consideration of GHG and airborne pollutants, the assumption does not affect the obtained results.  7.3 Future research directions Impact assessment of biomass-based community energy systems with a focus on heat generation was studied here. The study resulted in a number of implications that could be further explored:  Local air quality and human health risks due to exposure will largely depend on the plant’s location. It appears that a distributed DES with combined NG and biomass may have an advantage over a community-based centralized single DES in terms of overall health impacts. Further research is needed to confirm the initial findings presented in this study and to explore impacts in cases of multiple locations and multiple plants on local air quality and population exposure.  148   Indoor air quality, which is influenced by ambient air, could be of interest for further research with an addition of building specific parameters such as ventilation rate. People generally spend considerable time indoors and impacts of DES on indoor air quality has not yet been addressed comprehensively.   An improvement in data availability with increased number of direct source emission tests as well as air quality monitoring at multiple locations would be important in future research which will aim at larger number of locally collected data while decreasing dependence on general average emission factors and other parameters. More site-specific characterization of emissions will increase public confidence and acceptance of DES and will change their perception about associated risks.  This study presented the significance of locally obtained iF, but the future research could explore the inclusion of site-specific iF and CFhealth in LCIA methodology which would bring more accuracy in local impact segment of life cycle assessments.   Comprehensive assessment of sustainability in addition to consideration of environmental, social and economic aspects of technological solutions, embraces sustaining provisioning, regulating, supporting and cultural ecosystem services. Therefore, consideration of other bio-geophysical components should be addressed in future research so to comprehensively evaluate sustainability of energy systems which are fast growing in Canada and other countries and seem to present viable energy solutions for urban areas. 149  Bibliography Abril, G.A., Diez, S.C., Pignata, M.L., Britch, J., 2016. Particulate matter concentrations originating from industrial and urban sources: Validation of atmospheric dispersion modeling results. Atmospheric Pollut. Res. 7, 180–189. https://doi.org/10.1016/j.apr.2015.08.009 Aguilar, F.X., Song, N., Shifley, S., 2011. Review of consumption trends and public policies promoting woody biomass as an energy feedstock in the U.S. Biomass Bioenergy, PROCEEDINGS OF A WORKSHOP OF IEA BIOENERGY TASK 31 ON “SUSTAINABLE FORESTRY SYSTEMS FOR BIOENERGY: INTEGRATION, INNOVATION AND INFORMATION” 35, 3708–3718. https://doi.org/10.1016/j.biombioe.2011.05.029 Akhtari, S., Sowlati, T., Day, K., 2014. Economic feasibility of utilizing forest biomass in district energy systems – A review. Renew. Sustain. Energy Rev. 33, 117–127. https://doi.org/10.1016/j.rser.2014.01.058 Arena, U., Di Gregorio, F., Santonastasi, M., 2010. A techno-economic comparison between two design configurations for a small scale, biomass-to-energy gasification based system. Chem. Eng. J. 162, 580–590. https://doi.org/10.1016/j.cej.2010.05.067 Asadullah, M., 2014. Biomass gasification gas cleaning for downstream applications: A comparative critical review. Renew. Sustain. Energy Rev. 40, 118–132. https://doi.org/10.1016/j.rser.2014.07.132 Azorin-Molina, C., Tijm, S., Chen, D., 2011. Development of selection algorithms and databases for sea breeze studies. Theor. Appl. Climatol. 106, 531–546. https://doi.org/10.1007/s00704-011-0454-4 150  BabcockPower Environmental, 2008. Operating Experience with High Efficiency Emissions Reduction System for Biomass Boilers. Bachmann, T.M., van der Kamp, J., 2017. Expressing air pollution-induced health-related externalities in physical terms with the help of DALYs. Environ. Int. 103, 39–50. https://doi.org/10.1016/j.envint.2017.03.020 BC Ministry of Energy, Mines and Petroleum Resources, 2008. BC Bioenergy Strategy: Growing Our Natural Energy Advantage [WWW Document]. URL http://www2.gov.bc.ca/assets/gov/farming-natural-resources-and-industry/electricity-alternative-energy/bc_bioenergy_strategy.pdf (accessed 1.18.12). BC Ministry of Environment and Climate Change Strategy, 2016. GHG Emissions - Environmental Reporting BC [WWW Document]. URL http://www.env.gov.bc.ca/soe/indicators/sustainability/ghg-emissions.html (accessed 11.30.17). BC MoE, 2016. British Columbia Ambient Air Quality Objectives. BC MoE, 2008. Guidelines for Air Quality Dispersion Modelling in British Columbia [WWW Document]. URL http://www.env.gov.bc.ca/epd/bcairquality/reports/air_disp_model_08.html (accessed 12.30.13). Bell, M.L., Davis, D.L., Cifuentes, L.A., Krupnick, A.J., Morgenstern, R.D., Thurston, G.D., 2008. Ancillary human health benefits of improved air quality resulting from climate change mitigation. Environ. Health 7, 41. https://doi.org/10.1186/1476-069X-7-41 151  Bennett, D.H., McKone, T.E., Evans, J.S., Nazaroff, W.W., Margni, M.D., Jolliet, O., Smith, K.R., 2002. Defining Intake Fraction. Env. Sci Technol 36, 206A–211A. https://doi.org/10.1021/es0222770 Bhowmik, C., Bhowmik, S., Ray, A., Pandey, K.M., 2017. Optimal green energy planning for sustainable development: A review. Renew. Sustain. Energy Rev. 71, 796–813. https://doi.org/10.1016/j.rser.2016.12.105 Boman, C., Nordin, A., Boström, D., Öhman, M., 2004. Characterization of Inorganic Particulate Matter from Residential Combustion of Pelletized Biomass Fuels. Energy Fuels 18, 338–348. https://doi.org/10.1021/ef034028i Boman, C., Nordin, A., Thaning, L., 2003. Effects of increased biomass pellet combustion on ambient air quality in residential areas—a parametric dispersion modeling study. Biomass Bioenergy 24, 465–474. https://doi.org/10.1016/S0961-9534(02)00146-0 Börjesson, M., Ahlgren, E.O., 2010. Biomass gasification in cost-optimized district heating systems—A regional modelling analysis. Energy Policy 38, 168–180. https://doi.org/10.1016/j.enpol.2009.09.001 Bossel, U., 2003. Well-to-Wheel Studies, Heating Values, and the Energy Conservation Principle. Eur. Fuel Cell Forum. Bradley, D., 2012. Economic Impact of Bioenergy in Canada- 2011. CanBio. Bradley, D., 2006. Canada Biomass-Bioenergy report, ON: Climate Change Solutions; Buckley, R.L., Kurzeja, R.J., 1997. An Observational and Numerical Study of the Nocturnal Sea Breeze. Part I: Structure and Circulation. J. Appl. Meteorol. 36, 1577–1598. https://doi.org/10.1175/1520-0450(1997)036<1577:AOANSO>2.0.CO;2 152  Cambero, C., Sowlati, T., 2014. Assessment and optimization of forest biomass supply chains from economic, social and environmental perspectives – A review of literature. Renew. Sustain. Energy Rev. 36, 62–73. https://doi.org/10.1016/j.rser.2014.04.041 Canada, E. and C.C., 2017. Pricing carbon pollution in Canada: how it will work [WWW Document]. gcnws. URL https://www.canada.ca/en/environment-climate-change/news/2017/05/pricing_carbon_pollutionincanadahowitwillwork.html (accessed 9.9.17). Canada’s Natural Gas, 2017. Natural Gas Supply, Demand, and Markets [WWW Document]. Can. Nat. Gas. URL http://www.canadasnaturalgas.ca/en/explore-topics/natural-gas-demand-markets (accessed 3.24.17). Canadian Biomass), 2009. Syngas to power tissue plant [WWW Document]. Can. Biomass. URL https://www.canadianbiomassmagazine.ca/combustion/syngas-to-power-tissue-plant-1155 (accessed 10.9.17). Carneiro, P., Ferreira, P., 2012. The economic, environmental and strategic value of biomass. Renew. Energy 44, 17–22. https://doi.org/10.1016/j.renene.2011.12.020 CEN (European Committee for Standardization), 2005a. CEN/TS 15103: Solid Biofuels - Methods for the Determination of Bulk Density. CEN (European Committee for Standardization), 2005b. CEN/TS 14774: Solid Biofuels - Methods for the Determination of Moisture Content: Oven Dry Method. CEN European Committee for Standardization), 2005. CEN/TS 14918: Solid Biofuels – Method for the Determination of Calorific Value. 153  Chandrasekaran, S.R., Laing, J.R., Holsen, T.M., Raja, S., Hopke, P.K., 2011. Emission Characterization and Efficiency Measurements of High-Efficiency Wood Boilers. Energy Fuels 25, 5015–5021. https://doi.org/10.1021/ef2012563 Chang, J., Hanna, S., 2004. Air quality model performance evaluation [WWW Document]. ResearchGate. URL https://www.researchgate.net/publication/225361827_Air_quality_model_performance_evaluation (accessed 5.22.17). Chen, L., Shi, M., Li, S., Bai, Z., Wang, Z., 2017. Combined use of land use regression and BenMAP for estimating public health benefits of reducing PM2.5 in Tianjin, China. Atmos. Environ. 152, 16–23. https://doi.org/10.1016/j.atmosenv.2016.12.023 Cherubini, F., Guest, G., Strømman, A.H., 2013. Bioenergy from forestry and changes in atmospheric CO2: Reconciling single stand and landscape level approaches. J. Environ. Manage. 129, 292–301. https://doi.org/10.1016/j.jenvman.2013.07.021 Cherubini, F., Strømman, A.H., 2011. Life cycle assessment of bioenergy systems: State of the art and future challenges. Bioresour. Technol. 102, 437–451. https://doi.org/10.1016/j.biortech.2010.08.010 CIEEDAC, 2015. Canadian District Energy Heat Capacity by Baseload Energy Source as of 2014 [WWW Document]. URL https://www.cieedac.sfu.ca/DB_DEnew/index.php?l=fuel_heat (accessed 9.2.15). Cot, B., 2016. CO2 Neutral Biomass Fuel: Life Cycle Analysis of waste to energy systems, a case study [WWW Document]. SEEDS Sustain. Libr. URL https://sustain.ubc.ca/courses-teaching/seeds-program/seeds-sustainability-library (accessed 4.5.16). 154  Cui, H., Yao, R., Xu, X., Xin, C., Yang, J., 2011. A tracer experiment study to evaluate the CALPUFF real time application in a near-field complex terrain setting. Atmos. Environ. 45, 7525–7532. https://doi.org/10.1016/j.atmosenv.2011.08.041 Čupr, P., Flegrová, Z., Franců, J., Landlová, L., Klánová, J., 2013. Mineralogical, chemical and toxicological characterization of urban air particles. Environ. Int. 54, 26–34. https://doi.org/10.1016/j.envint.2012.12.012 Curci, G., Cinque, G., Tuccella, P., Visconti, G., Verdecchia, M., Iarlori, M., Rizi, V., 2012. Modelling air quality impact of a biomass energy power plant in a mountain valley in Central Italy. Atmos. Environ. 62, 248–255. https://doi.org/10.1016/j.atmosenv.2012.08.005 Danish District Heating Association, 2014. EnergyMap. de Hoogh, K., Korek, M., Vienneau, D., Keuken, M., Kukkonen, J., Nieuwenhuijsen, M.J., Badaloni, C., Beelen, R., Bolignano, A., Cesaroni, G., Pradas, M.C., Cyrys, J., Douros, J., Eeftens, M., Forastiere, F., Forsberg, B., Fuks, K., Gehring, U., Gryparis, A., Gulliver, J., Hansell, A.L., Hoffmann, B., Johansson, C., Jonkers, S., Kangas, L., Katsouyanni, K., Künzli, N., Lanki, T., Memmesheimer, M., Moussiopoulos, N., Modig, L., Pershagen, G., Probst-Hensch, N., Schindler, C., Schikowski, T., Sugiri, D., Teixidó, O., Tsai, M.-Y., Yli-Tuomi, T., Brunekreef, B., Hoek, G., Bellander, T., 2014. Comparing land use regression and dispersion modelling to assess residential exposure to ambient air pollution for epidemiological studies. Environ. Int. 73, 382–392. https://doi.org/10.1016/j.envint.2014.08.011 Delivand, M.K., Cammerino, A.R.B., Garofalo, P., Monteleone, M., 2015. Optimal locations of bioenergy facilities, biomass spatial availability, logistics costs and GHG (greenhouse 155  gas) emissions: a case study on electricity productions in South Italy. J. Clean. Prod. 99, 129–139. https://doi.org/10.1016/j.jclepro.2015.03.018 Dhondt, S., Beckx, C., Degraeuwe, B., Lefebvre, W., Kochan, B., Bellemans, T., Int Panis, L., Macharis, C., Putman, K., 2012. Health impact assessment of air pollution using a dynamic exposure profile: Implications for exposure and health impact estimates. Environ. Impact Assess. Rev. 36, 42–51. https://doi.org/10.1016/j.eiar.2012.03.004 Di Lucia, L., Ericsson, K., 2014. Low-carbon district heating in Sweden – Examining a successful energy transition. Energy Res. Soc. Sci. 4, 10–20. https://doi.org/10.1016/j.erss.2014.08.005 Difs, K., Wetterlund, E., Trygg, L., Söderström, M., 2010. Biomass gasification opportunities in a district heating system. Biomass Bioenergy 34, 637–651. https://doi.org/10.1016/j.biombioe.2010.01.007 Dockside Green Energy, 2008. How the energy system works [WWW Document]. URL http://docksidegreenenergy.com/the_energy_system.html (accessed 10.9.17). Doerksen, G., 2014. 2013 Lower Fraser Valley Air Quality Monitoring Report. Doerksen, G., 2013. 2012 Lower Fraser Valley Air Quality Monitoring Report Summary. Du, X., Wu, Y., Fu, L., Wang, S., Zhang, S., Hao, J., 2012. Intake fraction of PM2.5 and NOX from vehicle emissions in Beijing based on personal exposure data. Atmos. Environ. 57, 233–243. https://doi.org/10.1016/j.atmosenv.2012.04.046 EIA, 2016. International Energy Outlook 2016-World energy demand and economc outlook [WWW Document]. URL https://www.eia.gov/outlooks/ieo/world.cfm (accessed 5.28.17). 156  Envirochem Services Inc., 2008. Feasibility Study : Identifying Economic Opportunities for Bugwood and Other Biomass Resources in Alberta and BC.; 2008 [WWW Document]. URL http://eipa.alberta.ca/media/29567/bugwood_study_final_report.pdf (accessed 10.19.10). Environment Canada, 2017. 2017NIR-A13-Electricity- Canadian Provinces and Teritories. EPA, U., 1992. Screening Procedures for Estimating the Air Quality Impact of Stationary Sources, Revised. EPA-454/R-92-019. Evans, A., Strezov, V., Evans, T.J., 2010. Sustainability considerations for electricity generation from biomass. Renew. Sustain. Energy Rev. 14, 1419–1427. https://doi.org/10.1016/j.rser.2010.01.010 Evans, J.S., Thompson, K.M., Hattis, D., 2000. Exposure Efficiency: Concept and Application to Perchloroethylene Exposure from Dry Cleaners. J. Air Waste Manag. Assoc. 50, 1700–1703. https://doi.org/10.1080/10473289.2000.10464199 Evans, J.S., Wolff, S.K., Phonboon, K., Levy, J.I., Smith, K.R., 2002. Exposure efficiency: an idea whose time has come? Chemosphere 49, 1075–1091. https://doi.org/10.1016/S0045-6535(02)00242-4 Fallahi, F., Karimi, M., Voia, M.-C., 2016. Persistence in world energy consumption: Evidence from subsampling confidence intervals. Energy Econ. 57, 175–183. https://doi.org/10.1016/j.eneco.2016.04.021 Feliciano, D., Slee, B., Smith, P., 2014. The potential uptake of domestic woodfuel heating systems and its contribution to tackling climate change: A case study from the North East Scotland. Renew. Energy 72, 344–353. https://doi.org/10.1016/j.renene.2014.07.039 Field, B., Olewiler, N., 2005. Environmental Economics. McGraw-Hill, canada. 157  Fiorese, G., Catenacci, M., Bosetti, V., Verdolini, E., 2014. The power of biomass: Experts disclose the potential for success of bioenergy technologies. Energy Policy 65, 94–114. https://doi.org/10.1016/j.enpol.2013.10.015 Fisher, B., Sylvain, J., Kukkonen, J., Piringer, M., Rotach, M., Schatzmann, M. (Eds.), 2005. Meteorology applied to urban air pollution problems, Final report COST ction 715. Demetra Ltd Publishers. FVB Energy Inc., 2017. Revelstoke Community Heating System [WWW Document]. FVB Energy Inc. URL http://www.fvbenergy.com/projects/revelstoke-community-heating-system/ (accessed 10.9.17). Gabillet, P., 2015. Energy supply and urban planning projects: Analysing tensions around district heating provision in a French eco-district. Energy Policy 78, 189–197. https://doi.org/10.1016/j.enpol.2014.11.006 Gao, J., Cao, C., Xiao, Q., Xu, B., Zhou, X., Zhang, X., 2013. Determination of dynamic intake fraction of cooking-generated particles in the kitchen. Build. Environ. 65, 146–153. https://doi.org/10.1016/j.buildenv.2013.04.006 Genon, G., Torchio, M.F., Poggio, A., Poggio, M., 2009. Energy and environmental assessment of small district heating systems: Global and local effects in two case-studies. Energy Convers. Manag. 50, 522–529. https://doi.org/10.1016/j.enconman.2008.11.010 Gerharz, L.E., Klemm, O., Broich, A.V., Pebesma, E., 2013. Spatio-temporal modelling of individual exposure to air pollution and its uncertainty. Atmos. Environ. 64, 56–65. https://doi.org/10.1016/j.atmosenv.2012.09.069 Ghafghazi, S., 2011. Multi Criteria Evaluation of Wood Pellet Utilization in District Heating Systems. The University of British Columbia, Vancouver, BC., Canada. 158  Ghafghazi, S., Sowlati, T., Sokhansanj, S., Bi, X., Melin, S., 2011. Particulate matter emissions from combustion of wood in district heating applications. Renew. Sustain. Energy Rev. 15, 3019–3028. https://doi.org/10.1016/j.rser.2011.04.001 Ghafghazi, S., Sowlati, T., Sokhansanj, S., Melin, S., 2010. Techno-economic analysis of renewable energy source options for a district heating project. Int. J. Energy Res. 34, 1109–1120. https://doi.org/10.1002/er.1637 Giuntoli, J., Caserini, S., Marelli, L., Baxter, D., Agostini, A., 2015. Domestic heating from forest logging residues: environmental risks and benefits. J. Clean. Prod. 99, 206–216. https://doi.org/10.1016/j.jclepro.2015.03.025 Government of Canada, N.E.B., 2014. NEB – Canada’s Energy Future 2013 - Energy Supply and Demand Projections to 2035 - An Energy Market Assessment [WWW Document]. URL https://www.neb-one.gc.ca/nrg/ntgrtd/ftr/2013/index-eng.html (accessed 11.28.17). Government of Canada, N.R.C., 2011. Energy Efficiency Trends in Canada, 1990 to 2008 | Office of Energy Efficiency [WWW Document]. URL http://oee.nrcan.gc.ca/%20/832 (accessed 1.18.12). Greco, S.L., Wilson, A.M., Spengler, J.D., Levy, J.I., 2007. Spatial patterns of mobile source particulate matter emissions-to-exposure relationships across the United States. Atmos. Environ. 41, 1011–1025. https://doi.org/10.1016/j.atmosenv.2006.09.025 Grieshop, A.P., Marshall, J.D., Kandlikar, M., 2011. Health and climate benefits of cookstove replacement options. Energy Policy 39, 7530–7542. https://doi.org/10.1016/j.enpol.2011.03.024 159  Gulliver, J., Briggs, D., 2011. STEMS-Air: A simple GIS-based air pollution dispersion model for city-wide exposure assessment. Sci. Total Environ. 409, 2419–2429. https://doi.org/10.1016/j.scitotenv.2011.03.004 Gustavsson, L., Eriksson, L., Sathre, R., 2011. Costs and CO2 benefits of recovering, refining and transporting logging residues for fossil fuel replacement. Appl. Energy 88, 192–197. https://doi.org/10.1016/j.apenergy.2010.07.026 Gustavsson, L., Holmberg, J., Dornburg, V., Sathre, R., Eggers, T., Mahapatra, K., Marland, G., 2007. Using biomass for climate change mitigation and oil use reduction. Energy Policy 35, 5671–5691. https://doi.org/10.1016/j.enpol.2007.05.023 Haberl, H., Sprinz, D., Bonazountas, M., Cocco, P., Desaubies, Y., Henze, M., Hertel, O., Johnson, R.K., Kastrup, U., Laconte, P., Lange, E., Novak, P., Paavola, J., Reenberg, A., van den Hove, S., Vermeire, T., Wadhams, P., Searchinger, T., 2012. Correcting a fundamental error in greenhouse gas accounting related to bioenergy. Energy Policy 45, 18–23. https://doi.org/10.1016/j.enpol.2012.02.051 Hajra, B., Stathopoulos, T., Bahloul, A., 2010. Assessment of pollutant dispersion from rooftop stacks: ASHRAE, ADMS and wind tunnel simulation. Build. Environ. 45, 2768–2777. https://doi.org/10.1016/j.buildenv.2010.06.006 Havukainen, J., Nguyen, M.T., Väisänen, S., Horttanainen, M., 2018. Life cycle assessment of small-scale combined heat and power plant: Environmental impacts of different forest biofuels and replacing district heat produced from natural gas. J. Clean. Prod. 172, 837–846. https://doi.org/10.1016/j.jclepro.2017.10.241 160  Heath, G.A., Granvold, P.W., Hoats, A.S., W Nazaroff, W., 2006. Intake fraction assessment of the air pollutant exposure implications of a shift toward distributed electricity generation. Atmos. Environ. 40, 7164–7177. https://doi.org/10.1016/j.atmosenv.2006.06.023 Heath, G.A., Nazaroff, W.W., 2007. Intake-to-delivered-energy ratios for central station and distributed electricity generation in California. Atmos. Environ. 41, 9159–9172. https://doi.org/10.1016/j.atmosenv.2007.07.055 Hektor, B., Backéus, S., Andersson, K., 2016. Carbon balance for wood production from sustainably managed forests. Biomass Bioenergy 93, 1–5. https://doi.org/10.1016/j.biombioe.2016.05.025 Hiloidhari, M., Baruah, D.C., Singh, A., Kataki, S., Medhi, K., Kumari, S., Ramachandra, T.V., Jenkins, B.M., Thakur, I.S., 2017. Emerging role of Geographical Information System (GIS), Life Cycle Assessment (LCA) and spatial LCA (GIS-LCA) in sustainable bioenergy planning. Bioresour. Technol., Special Issue on International Conference on Current Trends in Biotechnology & post ICCB-2016 conference on Strategies for Environmental Protection and Management (ICSEPM-2016) 242, 218–226. https://doi.org/10.1016/j.biortech.2017.03.079 Holmes, N.S., Morawska, L., 2006. A review of dispersion modelling and its application to the dispersion of particles: An overview of different dispersion models available. Atmos. Environ. 40, 5902–5928. https://doi.org/10.1016/j.atmosenv.2006.06.003 Holnicki, P., Kałuszko, A., Trapp, W., 2016. An urban scale application and validation of the CALPUFF model. Atmospheric Pollut. Res. 7, 393–402. https://doi.org/10.1016/j.apr.2015.10.016 161  Holtsmark, B., 2013. Boreal forest management and its effect on atmospheric CO2. Ecol. Model. 248, 130–134. https://doi.org/10.1016/j.ecolmodel.2012.10.006 Huang, Y., Finell, M., Larsson, S., Wang, X., Zhang, J., Wei, R., Liu, L., 2017. Biofuel pellets made at low moisture content – Influence of water in the binding mechanism of densified biomass. Biomass Bioenergy 98, 8–14. https://doi.org/10.1016/j.biombioe.2017.01.002 Humbert, S., Marshall, J.D., Shaked, S., Spadaro, J.V., Nishioka, Y., Preiss, P., McKone, T.E., Horvath, A., Jolliet, O., 2011. Intake Fraction for Particulate Matter: Recommendations for Life Cycle Impact Assessment. Env. Sci Technol 45, 4808–4816. https://doi.org/10.1021/es103563z IEA, 2002. Greenhouse Gas Balances of Biomass and Bioenergy Systems (Task 38). IEA Bioenergy, 2011. Using a Life Cycle Assessment Approach to Estimate the Net Greenhouse Gas emissions of Bioenergy. Ioannou, A., Angus, A., Brennan, F., 2017. Risk-based methods for sustainable energy system planning: A review. Renew. Sustain. Energy Rev. 74, 602–615. https://doi.org/10.1016/j.rser.2017.02.082 IPCC, 2015. Climate Change 2014: Synthesis Report.-Contribution of Working Groups I, II and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [WWW Document]. URL http://www.ipcc.ch/report/ar5/syr/ (accessed 12.19.15). IPCC, 2013. Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA. IPCC 5th Report, 2015. Global-Warming-Potential-Values [WWW Document]. URL (accessed 8.11.16). 162  ISO, 2006. ISO 14040:2006, Environmental management — Life cycle assessment — Principles and framework [WWW Document]. URL https://www.iso.org/obp/ui/#iso:std:iso:14040:ed-2:v1:en Jäppinen, E., Korpinen, O.-J., Laitila, J., Ranta, T., 2014. Greenhouse gas emissions of forest bioenergy supply and utilization in Finland. Renew. Sustain. Energy Rev. 29, 369–382. https://doi.org/10.1016/j.rser.2013.08.101 Ji, S., Cherry, C.R., J. Bechle, M., Wu, Y., Marshall, J.D., 2011. Electric Vehicles in China: Emissions and Health Impacts. Env. Sci Technol 46, 2018–2024. https://doi.org/10.1021/es202347q Johansson, L.S., Leckner, B., Gustavsson, L., Cooper, D., Tullin, C., Potter, A., 2004. Emission characteristics of modern and old-type residential boilers fired with wood logs and wood pellets. Atmos. Environ. 38, 4183–4195. https://doi.org/10.1016/j.atmosenv.2004.04.020 Jolliet, O., Fantke, P., 2015. Human Toxicity, in: Hauschild, M.Z., Huijbregts, M.A.J. (Eds.), Life Cycle Impact Assessment, LCA Compendium – The Complete World of Life Cycle Assessment. Springer Netherlands, pp. 75–96. Jonsson, A., Hillring, B., 2006. Planning for increased bioenergy use—Evaluating the impact on local air quality. Biomass Bioenergy 30, 543–554. https://doi.org/10.1016/j.biombioe.2006.01.002 Kaivosoja, T., Jalava, P.I., Lamberg, H., Virén, A., Tapanainen, M., Torvela, T., Tapper, U., Sippula, O., Tissari, J., Hillamo, R., Hirvonen, M.-R., Jokiniemi, J., 2013. Comparison of emissions and toxicological properties of fine particles from wood and oil boilers in small (20–25 kW) and medium (5–10 MW) scale. Atmos. Environ. 77, 193–201. https://doi.org/10.1016/j.atmosenv.2013.05.014 163  Kalt, G., Kranzl, L., 2011. Assessing the economic efficiency of bioenergy technologies in climate mitigation and fossil fuel replacement in Austria using a techno-economic approach. Appl. Energy 88, 3665–3684. https://doi.org/10.1016/j.apenergy.2011.03.014 Kautto, N., Peck, P., 2012. Regional biomass planning – Helping to realise national renewable energy goals? Renew. Energy 46, 23–30. https://doi.org/10.1016/j.renene.2012.03.024 Knibbs, L.D., Hewson, M.G., Bechle, M.J., Marshall, J.D., Barnett, A.G., 2014. A national satellite-based land-use regression model for air pollution exposure assessment in Australia. Environ. Res. 135, 204–211. https://doi.org/10.1016/j.envres.2014.09.011 Kocbach Bølling, A., Joakim Pagels, Yttri, K.E., Barregard, L., Gerd Sallsten, Schwarze, P.E., Boman, C., 2009. Health effects of residential wood smoke particles: the importance of combustion conditions and physicochemical particle properties. Part. Fibre Toxicol. 6, 29–29. https://doi.org/10.1186/1743-8977-6-29 Kortsch, T., Hildebrand, J., Schweizer-Ries, P., 2015. Acceptance of biomass plants – Results of a longitudinal study in the bioenergy-region Altmark. Renew. Energy 83, 690–697. https://doi.org/10.1016/j.renene.2015.04.059 Kraxner, F., Aoki, K., Kindermann, G., Leduc, S., Albrecht, F., Liu, J., Yamagata, Y., 2016. Bioenergy and the city – What can urban forests contribute? Appl. Energy 165, 990–1003. https://doi.org/10.1016/j.apenergy.2015.12.121 Lai, A.C.K., Thatcher, T.L., Nazaroff, W.W., 2000. Inhalation Transfer Factors for Air Pollution Health Risk Assessment. J. Air Waste Manag. Assoc. 50, 1688–1699. https://doi.org/10.1080/10473289.2000.10464196 164  Lakes Environmental, 2012a. WRPLOT View - Wind Rose Plots for Meteorological Data [WWW Document]. URL https://www.weblakes.com/products/wrplot/index.html (accessed 5.23.17). Lakes Environmental, 2012b. CALPUFF View - Long Range Puff Air Dispersion Model [WWW Document]. URL http://www.weblakes.com/products/calpuff/index.html (accessed 1.24.12). Levihn, F., 2014. CO2 emissions accounting: Whether, how, and when different allocation methods should be used. Energy 68, 811–818. https://doi.org/10.1016/j.energy.2014.01.098 Levy, J.I., Spengler, J.D., Hlinka, D., Sullivan, D., Moon, D., 2002. Using CALPUFF to evaluate the impacts of power plant emissions in Illinois: model sensitivity and implications. Atmos. Environ. 36, 1063–1075. https://doi.org/10.1016/S1352-2310(01)00493-9 Levy, J.I., Wolff, S.K., Evans, J.S., 2002. A regression-based approach for estimating primary and secondary particulate matter intake fractions. Risk Anal. Off. Publ. Soc. Risk Anal. 22, 895–904. Li, H., Sun, Q., Zhang, Q., Wallin, F., 2015. A review of the pricing mechanisms for district heating systems. Renew. Sustain. Energy Rev. 42, 56–65. https://doi.org/10.1016/j.rser.2014.10.003 Lin, Q.G., Zhai, M.Y., Huang, G.H., Wang, X.Z., Zhong, L.F., Pi, J.W., 2017. Adaptation planning of community energy systems to climatic change over Canada. J. Clean. Prod. 143, 686–698. https://doi.org/10.1016/j.jclepro.2016.12.057 165  Lobscheid, A.B., Nazaroff, W.W., Spears, M., Horvath, A., McKone, T.E., 2012. Intake fractions of primary conserved air pollutants emitted from on-road vehicles in the United States. Atmos. Environ. 63, 298–305. https://doi.org/10.1016/j.atmosenv.2012.09.027 Loh, M.M., Soares, J., Karppinen, A., Kukkonen, J., Kangas, L., Riikonen, K., Kousa, A., Asikainen, A., Jantunen, M.J., 2009. Intake fraction distributions for benzene from vehicles in the Helsinki metropolitan area. Atmos. Environ. 43, 301–310. https://doi.org/10.1016/j.atmosenv.2008.09.082 Lu, R., Turco, R.P., 1994. Air Pollutant Transport in a Coastal Environment. Part I: Two-Dimensional Simulations of Sea-Breeze and Mountain Effects. J. Atmospheric Sci. 51, 2285–2308. https://doi.org/10.1175/1520-0469(1994)051<2285:APTIAC>2.0.CO;2 Luo, Z., Li, Y., Nazaroff, W.W., 2010. Intake fraction of nonreactive motor vehicle exhaust in Hong Kong. Atmos. Environ. 44, 1913–1918. https://doi.org/10.1016/j.atmosenv.2010.02.016 Mahmoudi, M., Sowlati, T., Sokhansanj, S., 2009. Logistics of supplying biomass from a mountain pine beetle-infested forest to a power plant in British Columbia. Scand. J. For. Res. 24, 76–86. https://doi.org/10.1080/02827580802660397 Manneh, R., Margni, M., Deschênes, L., 2010. Spatial Variability of Intake Fractions for Canadian Emission Scenarios: A Comparison between Three Resolution Scales. Environ. Sci. Technol. 44, 4217–4224. https://doi.org/10.1021/es902983b Maroko, A.R., 2012. Using air dispersion modeling and proximity analysis to assess chronic exposure to fine particulate matter and environmental justice in New York City. Appl. Geogr. 34, 533–547. https://doi.org/10.1016/j.apgeog.2012.02.005 166  Marshall, J.D., Granvold, P.W., Hoats, A.S., McKone, T.E., Deakin, E., W Nazaroff, W., 2006. Inhalation intake of ambient air pollution in California’s South Coast Air Basin. Atmos. Environ. 40, 4381–4392. https://doi.org/10.1016/j.atmosenv.2006.03.034 Marshall, J.D., McKone, T.E., Deakin, E., Nazaroff, W.W., 2005a. Inhalation of motor vehicle emissions: effects of urban population and land area. Atmos. Environ. 39, 283–295. https://doi.org/10.1016/j.atmosenv.2004.09.059 Marshall, J.D., Riley, W.J., McKone, T.E., Nazaroff, W.W., 2003. Intake fraction of primary pollutants: motor vehicle emissions in the South Coast Air Basin. Atmos. Environ. 37, 3455–3468. https://doi.org/10.1016/S1352-2310(03)00269-3 Marshall, J.D., Teoh, S.-K., W. Nazaroff, W., 2005b. Intake fraction of nonreactive vehicle emissions in US urban areas. Atmos. Environ. 39, 1363–1371. https://doi.org/10.1016/j.atmosenv.2004.11.008 Martenies, S.E., Wilkins, D., Batterman, S.A., 2015. Health impact metrics for air pollution management strategies. Environ. Int. 85, 84–95. https://doi.org/10.1016/j.envint.2015.08.013 McKechnie, J., Colombo, S., Chen, J., Mabee, W., MacLean, H.L., 2011. Forest Bioenergy or Forest Carbon? Assessing Trade-Offs in Greenhouse Gas Mitigation with Wood-Based Fuels. Environ. Sci. Technol. 45, 789–795. https://doi.org/10.1021/es1024004 McKechnie, J., Colombo, S., MacLean, H.L., 2014. Forest carbon accounting methods and the consequences of forest bioenergy for national greenhouse gas emissions inventories. Environ. Sci. Policy 44, 164–173. https://doi.org/10.1016/j.envsci.2014.07.006 167  McManus, M.C., 2010. Life cycle impacts of waste wood biomass heating systems: A case study of three UK based systems. Energy 35, 4064–4070. https://doi.org/10.1016/j.energy.2010.06.014 Michael Z.  Hauschild., Ed., Mark A.J. Huijbregts.  Ed., M.A.J. (Eds.), 2015. LCA Compendium – The Complete World of Life Cycle Assessment. Springer. Mirabella, N., Allacker, K., 2017. The Environmental Footprint of Cities: Insights in the Steps forward to a New Methodological Approach. Procedia Environ. Sci., Sustainable synergies from Buildings to the Urban Scale 38, 635–642. https://doi.org/10.1016/j.proenv.2017.03.143 Miranda, T., Román, S., Arranz, J.I., Rojas, S., González, J.F., Montero, I., 2010. Emissions from thermal degradation of pellets with different contents of olive waste and forest residues. Fuel Process. Technol. 91, 1459–1463. https://doi.org/10.1016/j.fuproc.2010.05.023 Mittlefehldt, S., 2016. Seeing forests as fuel: How conflicting narratives have shaped woody biomass energy development in the United States since the 1970s. Energy Res. Soc. Sci. 14, 13–21. https://doi.org/10.1016/j.erss.2015.12.023 MoE, B.C., 2003. British Columbia Field Sampling Manual: For Continuous Monitoring and the Collection of Air, Air-Emission, Water, Wastewater, Soil, Sediment and Biological Samples. Muench, S., Guenther, E., 2013. A systematic review of bioenergy life cycle assessments. Appl. Energy 112, 257–273. https://doi.org/10.1016/j.apenergy.2013.06.001 168  Nabuurs, G.-J., Arets, E.J.M.M., Schelhaas, M.-J., 2017. European forests show no carbon debt, only a long parity effect. For. Policy Econ., Special section on The economics of carbon sequestration in forestry 75, 120–125. https://doi.org/10.1016/j.forpol.2016.10.009 Naeher, L.P., Brauer, M., Lipsett, M., Zelikoff, J.T., Simpson, C.D., Koenig, J.Q., Smith, K.R., 2007. Woodsmoke Health Effects: A Review. Inhal. Toxicol. 19, 67–106. https://doi.org/10.1080/08958370600985875 Natural Resources Canada, 2012. GeoGratis [WWW Document]. URL http://www.nrcan.gc.ca/earth-sciences/geography/topographic-information/free-data-geogratis/11042 Natural Resources Canada (NRC), 2015. Energy Fact Book 2015-2016. Nazaroff, W.W., 2008. Inhalation intake fraction of pollutants from episodic indoor emissions. Build. Environ. 43, 269–277. https://doi.org/10.1016/j.buildenv.2006.03.021 NEB, 2013. Energy Briefing Note: Canadian Energy Overview 2012. Newell, J.P., Vos, R.O., 2012. Accounting for forest carbon pool dynamics in product carbon footprints: Challenges and opportunities. Environ. Impact Assess. Rev., Trends in biogenic-carbon accounting 37, 23–36. https://doi.org/10.1016/j.eiar.2012.03.005 Nishioka, Y., Levy, J., Norris, G.A., Bennett, D., Spengler, J., 2005. A Risk-Based Approach to Health Impact Assessment for Input-Output Analysis, Part 2: Case Study of Insulation (8 pp). Int. J. Life Cycle Assess. 10, 255–262. https://doi.org/10.1065/lca2004.10.186.2 Notter, D.A., 2015. Life cycle impact assessment modeling for particulate matter: A new approach based on physico-chemical particle properties. Environ. Int. 82, 10–20. https://doi.org/10.1016/j.envint.2015.05.002 169  NREL, 2016. Energy Analysis - Useful Life [WWW Document]. URL https://www.nrel.gov/analysis/tech-footprint.html (accessed 8.30.16). Olivier, J., Alan Brent, Mark Goedkoop, Norihiro Itsubo, Ruedi Mueller-Wenk, Claudia Peña, Rita Schenk, Mary Stewart, Bo Weidema, 2003. Life Cycle Impact Assessment Programme of the Life Cycle Initiative: Final Report of the LCIA Definition Study. Pa, A.A., 2010. Development of British Columbia Wood Pellet Life Cycle Inventory and its Utilization in the Evaluation of Domestic Pellet Applications. The University of British Columbia, Vancouver, B.C., Canada. Pa, A., Bi, X.T., Sokhansanj, S., 2013. Evaluation of wood pellet application for residential heating in British Columbia based on a streamlined life cycle analysis. Biomass Bioenergy 49, 109–122. https://doi.org/10.1016/j.biombioe.2012.11.009 Pa, A., Bi, X.T., Sokhansanj, S., 2011. A life cycle evaluation of wood pellet gasification for district heating in British Columbia. Bioresour. Technol. 102, 6167–6177. https://doi.org/10.1016/j.biortech.2011.02.009 Pa, A., Craven, J., Bi, X., Melin, S., Sokhansanj, S., 2012. Environmental footprints of British Columbia wood pellets from a simplified life cycle analysis. Int. J. Life Cycle Assess. 17, 220–231. https://doi.org/10.1007/s11367-011-0358-7 Pandis, S.N., Seinfeld, J.H., 2006. Atmospheric chemistry and physics: from air pollution to climate change, 2nd ed. Hoboken, N.J. : J. Wiley. Panepinto, D., Viggiano, F., Genon, G., 2014. Evaluation of Environmental Compatibility for a Biomass Plant. Waste Biomass Valorization 5, 759–772. https://doi.org/10.1007/s12649-014-9300-0 170  Pantaleo, A.M., Giarola, S., Bauen, A., Shah, N., 2014. Integration of biomass into urban energy systems for heat and power. Part II: Sensitivity assessment of main techno-economic factors. Energy Convers. Manag. 83, 362–376. https://doi.org/10.1016/j.enconman.2014.03.051 Parajuli, R., Løkke, S., Østergaard, P.A., Knudsen, M.T., Schmidt, J.H., Dalgaard, T., 2014. Life Cycle Assessment of district heat production in a straw fired CHP plant. Biomass Bioenergy 68, 115–134. https://doi.org/10.1016/j.biombioe.2014.06.005 Patel, M., Zhang, X., Kumar, A., 2016. Techno-economic and life cycle assessment on lignocellulosic biomass thermochemical conversion technologies: A review. Renew. Sustain. Energy Rev. 53, 1486–1499. https://doi.org/10.1016/j.rser.2015.09.070 Pepe, N., Pirovano, G., Lonati, G., Balzarini, A., Toppetti, A., Riva, G.M., Bedogni, M., 2016. Development and application of a high resolution hybrid modelling system for the evaluation of urban air quality. Atmos. Environ. 141, 297–311. https://doi.org/10.1016/j.atmosenv.2016.06.071 Perilhon, C., Alkadee, D., Descombes, G., Lacour, S., 2012. Life Cycle Assessment Applied to Electricity Generation from Renewable Biomass. Energy Procedia, Terragreen 2012: Clean Energy Solutions for Sustainable Environment (CESSE) 18, 165–176. https://doi.org/10.1016/j.egypro.2012.05.028 Petersen, J.-P., 2016. Energy concepts for self-supplying communities based on local and renewable energy sources: A case study from northern Germany. Sustain. Cities Soc. 26, 1–8. https://doi.org/10.1016/j.scs.2016.04.014 Petrov, O., Bi, X., Lau, A., 2017. Impact assessment of biomass-based district heating systems in densely populated communities. Part II: Would the replacement of fossil fuels improve 171  ambient air quality and human health? Atmos. Environ. 161, 191–199. https://doi.org/10.1016/j.atmosenv.2017.05.001 Petrov, O., Bi, X., Lau, A., 2015. Impact assessment of biomass-based district heating systems in densely populated communities. Part I: Dynamic intake fraction methodology. Atmos. Environ. 115, 70–78. https://doi.org/10.1016/j.atmosenv.2015.05.036 PRé, 2016. SimaPro 8 Introduction To LCA.pdf [WWW Document]. URL https://www.pre-sustainability.com/download/SimaPro8IntroductionToLCA.pdf (accessed 8.8.16). Province of BC, 2012. District Energy Systems | BC Climate Action Toolkit [WWW Document]. URL http://www.toolkit.bc.ca/tool/district-energy-systems (accessed 5.14.12). Quantis, 2012. IMPACT 2002+,vQ2.22 [WWW Document]. URL http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.454.741&rep=rep1&type=pdf (accessed 5.3.16). Ralevic, P., Layzell, D.B., 2006. An Inventory of the Bioenergy Potential of British Columbia. Repo, A., Ahtikoski, A., Liski, J., 2015. Cost of turning forest residue bioenergy to carbon neutral. For. Policy Econ. 57, 12–21. https://doi.org/10.1016/j.forpol.2015.04.005 Ries, F.J., Marshall, J.D., Brauer, M., 2009. Intake Fraction of Urban Wood Smoke. Env. Sci Technol 43, 4701–4706. https://doi.org/10.1021/es803127d Röder, M., Whittaker, C., Thornley, P., 2015. How certain are greenhouse gas reductions from bioenergy? Life cycle assessment and uncertainty analysis of wood pellet-to-electricity supply chains from forest residues. Biomass Bioenergy, The 22nd European Biomass Conference and Exhibition held in Hamburg, June 2014 79, 50–63. https://doi.org/10.1016/j.biombioe.2015.03.030 172  Rood, A.S., 2014. Performance evaluation of AERMOD, CALPUFF, and legacy air dispersion models using the Winter Validation Tracer Study dataset. Atmos. Environ. 89, 707–720. https://doi.org/10.1016/j.atmosenv.2014.02.054 Rubio-Maya, C., Uche-Marcuello, J., Martínez-Gracia, A., Bayod-Rújula, A.A., 2011. Design optimization of a polygeneration plant fuelled by natural gas and renewable energy sources. Appl. Energy 88, 449–457. https://doi.org/10.1016/j.apenergy.2010.07.009 Russo, J.S., Ezzat Khalifa, H., 2010. CFD assessment of intake fraction in the indoor environment. Build. Environ. 45, 1968–1975. https://doi.org/10.1016/j.buildenv.2010.01.017 Saidur, R., Abdelaziz, E.A., Demirbas, A., Hossain, M.S., Mekhilef, S., 2011. A review on biomass as a fuel for boilers. Renew. Sustain. Energy Rev. 15, 2262–2289. https://doi.org/10.1016/j.rser.2011.02.015 Sala, S., Pant, R., Hauschild, M., Pennington, D., 2012. Research Needs and Challenges from Science to Decision Support. Lesson Learnt from the Development of the International Reference Life Cycle Data System (ILCD) Recommendations for Life Cycle Impact Assessment. Sustainability 4, 1412–1425. https://doi.org/10.3390/su4071412 Sansaniwal, S.K., Pal, K., Rosen, M.A., Tyagi, S.K., 2017. Recent advances in the development of biomass gasification technology: A comprehensive review. Renew. Sustain. Energy Rev. 72, 363–384. https://doi.org/10.1016/j.rser.2017.01.038 Schlamadinger, B., Marland, G., 1996. The role of forest and bioenergy strategies in the global carbon cycle. Biomass Bioenergy 10, 275–300. https://doi.org/10.1016/0961-9534(95)00113-1 173  Schlamadinger, B., Spitzer, J., Kohlmaier, G.H., Lüdeke, M., 1995. Carbon balance of bioenergy from logging residues. Biomass Bioenergy 8, 221–234. https://doi.org/10.1016/0961-9534(95)00020-8 Schnelle, K., Dey, P., 2000. Atmospheric Dispersion Modeling Compliance Guide. McGraw-Hill. Schwab, O., Maness, T., Bull, G., Roberts, D., 2009. Modeling the effect of changing market conditions on mountain pine beetle salvage harvesting and structural changes in the British Columbia forest products industry. Can. J. For. Res.-Rev. Can. Rech. 39, 1806–1820. https://doi.org/10.1139/X09-099 Scire, J., Strimaitis, D., Yamartino, R., 2000. A User’s Guide for the CALPUFF Disoersion Model. Searcy, E., Flynn, P.C., 2010. A criterion for selecting renewable energy processes. Biomass Bioenergy 34, 798–804. https://doi.org/10.1016/j.biombioe.2010.01.023 Sethuraman, S., Huynh, C.V., Kong, S.-C., 2011. Producer Gas Composition and NOx Emissions from a Pilot-Scale Biomass Gasification and Combustion System Using Feedstock with Controlled Nitrogen Content. Energy Fuels 25, 813–822. https://doi.org/10.1021/ef101352j SFU, 2016. SFU signs energy deal that will see 85% reduction in greenhouse gas emissions at Burnaby Campus [WWW Document]. SFU Univ. Commun. URL https://www.sfu.ca/university-communications/media-releases/2016/sfu-signs-energy-deal-that-will-see-85-reduction-in-greenhouse-gas-emissions-at-burnaby-campus.html (accessed 10.9.17). 174  Shabani, N., Sowlati, T., 2016. A hybrid multi-stage stochastic programming-robust optimization model for maximizing the supply chain of a forest-based biomass power plant considering uncertainties. J. Clean. Prod. 112, Part 4, 3285–3293. https://doi.org/10.1016/j.jclepro.2015.09.034 Shen, F., Ge, X., Hu, J., Nie, D., Tian, L., Chen, M., 2017. Air pollution characteristics and health risks in Henan Province, China. Environ. Res. 156, 625–634. https://doi.org/10.1016/j.envres.2017.04.026 Singh, R., Shukla, A., 2014. A review on methods of flue gas cleaning from combustion of biomass. Renew. Sustain. Energy Rev. 29, 854–864. https://doi.org/10.1016/j.rser.2013.09.005 Singh, Y.D., Mahanta, P., Bora, U., 2017. Comprehensive characterization of lignocellulosic biomass through proximate, ultimate and compositional analysis for bioenergy production. Renew. Energy 103, 490–500. https://doi.org/10.1016/j.renene.2016.11.039 Smith, K.R., 1993. Fuel Combustion, Air Pollution Exposure, and Health: The Situation in Developing Countries. Annu. Rev. Energy Environ. (S&T)2 Consultants, 2013. GHGenius [WWW Document]. URL http://www.ghgenius.ca/ (accessed 12.21.16). Stephen, J.D., Mabee, W.E., Pribowo, A., Pledger, S., Hart, R., Tallio, S., Bull, G.Q., 2016. Biomass for residential and commercial heating in a remote Canadian aboriginal community. Renew. Energy 86, 563–575. https://doi.org/10.1016/j.renene.2015.08.048 Stevens, G., de Foy, B., West, J.J., Levy, J.I., 2007a. Developing intake fraction estimates with limited data: Comparison of methods in Mexico City. Atmos. Environ. 41, 3672–3683. https://doi.org/10.1016/j.atmosenv.2006.12.051 175  Stevens, G., de Foy, B., West, J.J., Levy, J.I., 2007b. Corrigendum to “Developing intake fraction estimates with limited data: Comparison of methods in Mexico City”: [Atmos. Environ. 41 (2007) 3672–3683]. Atmos. Environ. 41, 6688–6689. https://doi.org/10.1016/j.atmosenv.2007.07.034 Striūgas, N., Vorotinskienė, L., Paulauskas, R., Navakas, R., Džiugys, A., Narbutas, L., 2017. Estimating the fuel moisture content to control the reciprocating grate furnace firing wet woody biomass. Energy Convers. Manag. 149, 937–949. https://doi.org/10.1016/j.enconman.2017.04.014 Swedish District Heating Association, 2014. District Heating in Sweden [WWW Document]. URL http://www.svenskfjarrvarme.se/In-English/District-Heating-in-Sweden/District-Heating/ (accessed 1.17.12). Tainio, M., Sofiev, M., Hujo, M., Tuomisto, J.T., Loh, M., Jantunen, M.J., Karppinen, A., Kangas, L., Karvosenoja, N., Kupiainen, K., Porvari, P., Kukkonen, J., 2009. Evaluation of the European population intake fractions for European and Finnish anthropogenic primary fine particulate matter emissions. Atmos. Environ. 43, 3052–3059. https://doi.org/10.1016/j.atmosenv.2009.03.030 Telmo, C., Lousada, J., Moreira, N., 2010. Proximate analysis, backwards stepwise regression between gross calorific value, ultimate and chemical analysis of wood. Bioresour. Technol. 101, 3808–3815. https://doi.org/10.1016/j.biortech.2010.01.021 The World Bank, Institute for Health Metrics and Evaluation, 2016. The Cost of Air Pollution: Strengthening the Economic Case for Action. 176  Todorovic, D., Jovovic, A., Petrov, O., Radic, D., Obradovic, M., Karlicic, N., Stanojevic, M., 2015. Using air dispersion modeling to evaluate stack characteristics. Processing 27, 28–36. Toka, A., Iakovou, E., Vlachos, D., Tsolakis, N., Grigoriadou, A.-L., 2014. Managing the diffusion of biomass in the residential energy sector: An illustrative real-world case study. Appl. Energy 129, 56–69. https://doi.org/10.1016/j.apenergy.2014.04.078 Tominaga, Y., Stathopoulos, T., 2016. Ten questions concerning modeling of near-field pollutant dispersion in the built environment. Build. Environ. 105, 390–402. https://doi.org/10.1016/j.buildenv.2016.06.027 Trenberth, K.E., Stepaniak, D.P., 2004. The flow of energy through the earth’s climate system. Q. J. R. Meteorol. Soc. 130, 2677–2701. https://doi.org/10.1256/qj.04.83 UBC, 2017. Energy & Water Department at UBC [WWW Document]. URL http://energy.ubc.ca/ (accessed 12.13.17). UBC, 2015a. Bioenergy Research Demonstration Facility (BRDF) | Energy [WWW Document]. Energy Water Serv. URL http://energy.ubc.ca/projects/brdf/ (accessed 7.18.15). UBC, 2015b. BioEnergy Research and Demonstration Facility - Board 4 project close-out report [WWW Document]. URL http://bog2.sites.olt.ubc.ca/files/2015/03/1.13_2015.04_B4-BioEnergy-Research-Facility.pdf (accessed 8.30.16). UBC, 2015c. Clean Energy Fund:  Advanced Biomass  Gasification For Combined Heat And Power  Demonstration [WWW Document]. URL http://energy.sites.olt.ubc.ca/files/2015/11/UBC-EN-Outreach-Report.pdf (accessed 8.29.16). 177  UBC, 2015d. Campus Initiatives. Climate Action Plan. [WWW Document]. URL http://sustain.ubc.ca/campus-initiatives/climate-energy/climate-action-plan (accessed 7.18.15). UBC, 2010. Request for decision: Bioenergy Research and Demonstration Project. US EPA, 2009. Emissions Factors & AP 42 | Clearinghouse for Emission Inventories and Emissions Factors | Technology Transfer Network | US EPA [WWW Document]. URL https://www.epa.gov/air-emissions-factors-and-quantification/ap-42-compilation-air-emission-factors (accessed 7.21.15). U.S. EPA, 2008. Technical Issues Related to CALPUFF Near-field Applications. US EPA, 2007. Guidance on the Use of Models and Other Analyses for Demonstrating Attainment of Air Quality Goals for Ozone, PM2.5, and Regional Haze. US EPA, 2003a. AP-42: Natural Gas Combustion. Chapter 1.4. US EPA, 2003b. AP-42: Wood Residue Combustion in Boilers; Chapter 1.6. US EPA National Center for Environmental Assessment, W.D., Moya, J., 2011. Exposure Factors Handbook [WWW Document]. URL http://cfpub.epa.gov/ncea/risk/recordisplay.cfm?deid=236252 (accessed 12.18.13). US EPA: SCRAM, 2015. Dispersion Modeling | TTN - Support Center for Regulatory Atmospheric Modeling | US EPA [WWW Document]. URL https://www3.epa.gov/scram001/dispersionindex.htm (accessed 8.15.15). Vallero, D., 2014. Chapter 27 - Modeling Applications, in: Fundamentals of Air Pollution (Fifth Edition). Academic Press, Boston, pp. 683–753. Vallios, I., Tsoutsos, T., Papadakis, G., 2009. Design of biomass district heating systems. Biomass Bioenergy 33, 659–678. https://doi.org/10.1016/j.biombioe.2008.10.009 178  van Dam, J., Junginger, M., Faaij, A.P.C., 2010. From the global efforts on certification of bioenergy towards an integrated approach based on sustainable land use planning. Renew. Sustain. Energy Rev. 14, 2445–2472. https://doi.org/10.1016/j.rser.2010.07.010 Vanhala, P., Repo, A., Liski, J., 2013. Forest bioenergy at the cost of carbon sequestration? Curr. Opin. Environ. Sustain., Terrestrial systems 5, 41–46. https://doi.org/10.1016/j.cosust.2012.10.015 Vassilev, S.V., Baxter, D., Andersen, L.K., Vassileva, C.G., 2010. An overview of the chemical composition of biomass. Fuel 89, 913–933. https://doi.org/10.1016/j.fuel.2009.10.022 Vassilev, S.V., Vassileva, C.G., Song, Y.-C., Li, W.-Y., Feng, J., 2017. Ash contents and ash-forming elements of biomass and their significance for solid biofuel combustion. Fuel 208, 377–409. https://doi.org/10.1016/j.fuel.2017.07.036 Vieira de Melo, A.M., Santos, J.M., Mavroidis, I., Reis Junior, N.C., 2012. Modelling of odour dispersion around a pig farm building complex using AERMOD and CALPUFF. Comparison with wind tunnel results. Build. Environ. 56, 8–20. https://doi.org/10.1016/j.buildenv.2012.02.017 Wang, S., Hao, J., Ho, M.S., Li, J., Lu, Y., 2006. Intake fractions of industrial air pollutants in China: Estimation and application. Sci. Total Environ. 354, 127–141. https://doi.org/10.1016/j.scitotenv.2005.01.045 Wauthy, J., Giffin, J., 2017. Long-Term Carbon and Commodity Price Forecast Report, UBC Energy & Water Services. UBC, Vancouver, BC. WBCSD, 2015. Recommendations on Biomass Carbon Neutrality [WWW Document]. - World Bus. Counc. Sustain. Dev. URL 179  http://www.wbcsd.org/Pages/EDocument/EDocumentDetails.aspx?ID=15347&NoSearchContextKey=true (accessed 9.29.15). World Health Organization, 2014. WHO Expert Meeting Methods and tools for assessing the health risks of air pollution at local, national and international level [WWW Document]. URL http://www.euro.who.int/en/health-topics/environment-and-health/air-quality/publications/2014/who-expert-meeting-methods-and-tools-for-assessing-the-health-risks-of-air-pollution-at-local,-national-and-international-level (accessed 5.6.16). Wright, D.G., Dey, P.K., Brammer, J., 2014. A barrier and techno-economic analysis of small-scale bCHP (biomass combined heat and power) schemes in the UK. Energy 71, 332–345. https://doi.org/10.1016/j.energy.2014.04.079 Wu, J., Houston, D., Lurmann, F., Ong, P., Winer, A., 2009. Exposure of PM2.5 and EC from diesel and gasoline vehicles in communities near the Ports of Los Angeles and Long Beach, California. Atmos. Environ. 43, 1962–1971. https://doi.org/10.1016/j.atmosenv.2009.01.009 Xu, J., Wang, X., Zhang, S., 2013. Risk-based air pollutants management at regional levels. Environ. Sci. Policy 25, 167–175. https://doi.org/10.1016/j.envsci.2012.09.014 Young, J.D., Anderson, N.M., Naughton, H.T., Mullan, K., 2018. Economic and policy factors driving adoption of institutional woody biomass heating systems in the U.S. Energy Econ. 69, 456–470. https://doi.org/10.1016/j.eneco.2017.11.020 Zaunbrecher, B.S., Arning, K., Falke, T., Ziefle, M., 2016. No pipes in my backyard?: Preferences for local district heating network design in Germany. Energy Res. Soc. Sci. 14, 90–101. https://doi.org/10.1016/j.erss.2016.01.008 180  Zeng, J., Xiao, R., Zhang, H., Chen, X., Zeng, D., Ma, Z., 2017. Syngas production via biomass self-moisture chemical looping gasification. Biomass Bioenergy 104, 1–7. https://doi.org/10.1016/j.biombioe.2017.03.020 Zhang, H., Liu, Y., Shi, R., Yao, Q., 2013. Evaluation of PM10 forecasting based on the artificial neural network model and intake fraction in an urban area: A case study in Taiyuan City, China. J. Air Waste Manag. Assoc. 63, 755–763. https://doi.org/10.1080/10962247.2012.755940 Zhou, Y., Levy, J.I., 2008. The impact of urban street canyons on population exposure to traffic-related primary pollutants. Atmos. Environ. 42, 3087–3098. https://doi.org/10.1016/j.atmosenv.2007.12.037 Zhou, Y., Levy, J.I., Evans, J.S., Hammitt, J.K., 2006. The influence of geographic location on population exposure to emissions from power plants throughout China. Environ. Int. 32, 365–373. https://doi.org/10.1016/j.envint.2005.08.028 Zhou, Y., Levy, J.I., Hammitt, J.K., Evans, J.S., 2003. Estimating population exposure to power plant emissions using CALPUFF: a case study in Beijing, China. Atmos. Environ. 37, 815–826. https://doi.org/10.1016/S1352-2310(02)00937-8    181  Appendices  Literature review   Literature search methodology Literature search was based on mostly on electronic resources available from the University of British Columbia library. The following databases were searched to retrieve literature for this research:   Web of Science: is an on-line database with multidisciplinary content updated weekly. It is linked to UBC for access of full text. http://resources.library.ubc.ca/page.php?id=138  Elsevier Science Direct: is an on-line data base with full text articles from 3,800 journals, versatile in content.  https://www.elsevier.com/solutions/sciencedirect.  PubMed: is an on-line database containing articles in life sciences subject areas. It is produced by the National Center for Biotechnology Information (NCBI) at the U.S. National Library of Medicine (NLM) http://www.ncbi.nlm.nih.gov/pmc/; http://resources.library.ubc.ca/571   Google Scholar (GS): An on-line un-restricted database containing a broad range of scholarly literature including peer-reviewed papers, theses, books, preprints, abstracts, technical reports. http://resources.library.ubc.ca/943   Search concepts such as, biomass, emissions, and health, as well as alternative terms for each concept, for example: for biomass I used: wood, forest residues, wood feedstock, were organized in an excel spreadsheet as a matrix. “Wildcards” were used where appropriate in order to retrieve variants on terms (e.g., wood*). Search terms were then combined using Boolean logic (AND, OR) to reduce the search results to those considered to be most relevant to the topic. The selection of retrieved material was restricted to those published since year 2000, unless older 182  literature sources were of significant importance for understanding the later literature which builds on such baseline studies. Only articles and reports published in English language were considered.    Additional grey literature was searched by accessing: Intergovernmental Panel on Climate Change (IPCC)), (Environmental Protection Agency (US EPA), Natural Resources Canada (NRC), BIOCAP Canada, BC Bioenergy Network as well as recent media releases and professionally prepared reports for government of Canada or provincial governments. Some publications and articles were recommended by my supervisor and reviewers of draft articles I submitted for publishing.   Reference software and literature storage Bibliographic data for the obtained electronic literature was saved using online citation management software ZOTERO (http://www.zotero.org/) which enables access from other computers by its “sync” function. Additionally, electronic copies of cited literature were stored on the hard drive of my computer.     Articles selected for detailed consideration and inclusion in literature review were summarized in an excel spreadsheet to enable better understanding and to synthesize information gained for literature review Chapter 2 of this thesis and for journal articles while prepared for publishing.   183    CO2 neutrality overview Table A.2-1 Summary of findings on biomass CO2 neutrality based on the reviewed literature. Biomass utilization stage Method of estimate Findings Reference Supply chain pathways LCA for forest and sawmill residues in the form of wood pellets; electricity production Different drying fuels, storage emission and dry matter losses could result in 73% higher emissions when wood pellets are used instead of coal; emissions during wood fuel storage are particularly significant Roder et al., 2015 The management options studied included forest fertilization, elongated rotation periods, varying the type of forest residues extracted, and leaving high stumps. Simulation-modeling and calculating costs for different scenarios The sooner carbon neutrality is required, the greater are the costs. The smallest carbon loss occurred when only quickly decomposing branches were collected, whereas the largest carbon loss resulted from harvesting all the residues Repo et al., 2015 Supply chain pathways LCA for domestic heating; 3 pathways of  forest residues (loose residues, district heating utilizing chips and pellets in domestic stoves) Supply chain GHG reduction could be beneficial in case of biomass except for the long-term slow decaying biomass – an important factor is biomass feedstock choices Giuntoli, 2015 Woody biomass a land-based option for fuel poverty reduction  -estimates the availability of wood to produce wood fuel in the region and identifies the barriers to the expansion of the wood fuel market for space and water heating purposes. Scenarios:  1) no existing/ projected houses would adopt wood fuel; 2) new and existing houses go off the gas grid to adopt wood fuel systems; 3) existing houses off the gas grid and 15% of the new houses to adopt wood fuel. Carbon dioxide emissions reduction from the adoption of wood fuel systems would be significant compared to non-adoption Some of the barriers for the adoption of wood pellet boilers could possibly be mitigated if some additional thought and finance are made available. Feliciano et al., 2014 Sustainability considerations in the design and planning of forest biomass supply chains for the production of bioenergy and bio products. A review of literature The major environmental issues of forest biomass utilization are related to a) carbon balance and GHG emissions, b) PM emissions, and c) the forest ecosystem health. Carbon neutrality will be achieved in the long term, when the new tree generation has reached a harvestable size Cambero and Sowlati, 2014 DH system, an annual load, annual variable heat-generating costs and technical parameters Case study – Stockholm DH system; investment optimization software It is a complex issue to allocate the emissions from alternative DH options, however: 1)investing in new production, energy efficiency/conservation, only direct or local emissions should be Levihn, 2014 184  Biomass utilization stage Method of estimate Findings Reference accounted for internally; 2)important to understand changed consumption and production and in addition to the marginal perspective in carbon footprint calculations, LCA should be considered Supply chain pathways and different conversion methods  LCA of comminuted forest biomass Most supply-chain GHG emissions arise from soil carbon stocks  changes and possible emissions from storage of biomass Jäppinen, 2014 Supply chain pathways and comparison to reference fossil fuels LCA, Case study, wood pellet production for electricity, domestic use and export The forest carbon accounting methods important, cumulative GHG reduction over longer periods (41MtCO2eq) McKechnie,  2014 Harvest-residue-based bioenergy   A synthesis paper Forest bioenergy is not carbon neutral if forest carbon stocks or sinks are reduced. The intensified removals of the logging residues would decrease the annual carbon sink of these forest soils by 3.1 million tons of CO2eq. Net reductions in the emissions will be achieved only in a longer term. Vanhala,  2013 Biomass harvesting stage  Adjustments to the previous studies  Carbon capturing continues with mature stands not being harvested which should be accounted for Holtsmark,  2013 The assessment of the climate impacts from biogenic CO2 fluxes from single stand to landscape level;  the resulting effects on atmospheric CO2 concentration. A case study – harvest practice which utilizes collection of wood logs with forest residues left on site The change in atmospheric CO2 concentration as a result of biogenic CO2 from regenerative biomass is reversible; at the landscape level similar increase and impacts from biomass CO2 like from fossil fuels for the first decades but later, CO2 from bioenergy stabilizes. Cherubini et al., 2013 Supply chain comparison LCA, carbon footprint modeling Forest type significant factor in carbon footprint which will vary depending on the harvesting scenarios Newell, 2012 Errors in GHG accounting; recommendations for policy makers A viewpoint article discusses the scientific background of an Opinion on bioenergy by the Scientific Committee of the European Environment Agency (EEA). Baseline error caused by assuming carbon neutrality on the basis of returned carbon to the atmosphere during the biomass burning; missed C absorptions should the plants had not been harvested; Policies  should  encourage bioenergy use  from  biomass that reduces GHG emissions,  biomass by-products, wastes,  residues without displacing other ecosystems services.   Haberl et al., 2012 Electricity production from biomass (combustion) of residues and dedicated energy crops Assessment based on: price, efficiency, GHG emissions, availability, limitations, land use, water use and social impacts The type and growing location of the biomass source determine its sustainability; Electricity generation produces low net carbon emissions, mostly in the form of CO2, Evans et al., 2010 185  Biomass utilization stage Method of estimate Findings Reference Overview of ongoing initiatives in biomass and bioenergy certification until 2009; the differences and similarities between these initiatives. A review of literature Certification may influence direct, local impacts with respect  to environmental and social effects of direct bioenergy production;  variation in methodologies and default values for calculating GHG balance and carbon sinks exists van Dam et al., 2010 Full fuel cycle Case study, modeling Major factor in evaluation: forest growth rate, conversion efficiency, fossil fuel energy system replaced Schlamadinger et al.,1996a  Net flux of C to the atmosphere through 4 mechanisms including  storage of C in the biosphere and the use of biofuels to displace fossil-fuel use Mathematical model GORCAM; 16 scenarios Longer time periods and higher efficiency of replacement of fossil fuels by biofuels favor using trees for bioenergy than for C sequestration  Schlamadinger et al., 1996b  Carbon storage in 3 soil carbon pools and carbon fluxes from these pools Model development  The time dependent “Carbon Neutrality” (CN) is the ratio of net emission reduction to the “saved” carbon emissions from the substituted energy system; for bioenergy (from logging residues), CN starts as very low at the beginning (eg. between 0.49 and 0.82 after 20 years) and approaches one at infinity. Schlamadinger et al., 1995   186  UBC Fuel characteristics and consumption, and energy calculations  Conversion of units used in fuel calculations 1 BTU =    1055 J 1 pound =   0.45359237 kg 1 foot =    0.3048 m SCF - A standard cubic foot for measuring natural gas is defined as:  The amount of natural gas contained at standard temperature and pressure, 60 [ºF] equal to 15 [ºC] and 14.73 [psi] equal to 101.325 [kPa]. In industry, the amount of natural is usually expressed as KSCF (103 SCF) or MMSCF (106 SCF) 1  ton =     200 lbs = 0.9072 tonnes 1 tonne [t] =    1,000 kg  Heat content of fuels and steam: Natural gas   heat content/SCF  1050 BTU = 1.107 MJ or 1.107 GJ/KSCF Fossil fuel oil   heat content/kg 46 MJ/kg Wood chips (BRDF)  heat content/kg 19.3 MJ as measured (average) at BRDF         (dry wood)   equals to 19.3 GJ/t (at 35% moisture content, wet basis = 54% moisture content, dry basis) Steam @165psi  heat content/lb 1197 BTU = 1.2628 MJ or 1.2628 GJ/KLBS        187   Fuel consumption and steam produced  Daily data for natural gas and oil consumption and steam produced were summarized by month as presented here, and day/night periods to enable detailed estimates of emissions used later in modeling scenarios as presented in Appendix C.  Table B.2-1 Natural gas and oil consumption (energy input) and steam produced (energy output) at PH and BRDF. MonthSteam from PH [KLBS]Steam to campus [GJ]NATURAL GAS[GJ]HeatingOIL [GJ]Thermal Efficiency [%]MonthSteam from PH [KLBS]Steam to campus [GJ]NATURAL GAS[GJ]Heating OIL [GJ]Thermal Efficiency [%]June 34,366 43,399 43,895 0 99 June 39,867 50,345 55,014 0 91July 32,653 41,236 45,305 0 91 July 28,257 35,684 37,904 0 94August 30,559 38,591 40,856 0 94 August 20,957 26,465 37,309 0 71September 36,581 46,195 49,010 0 94 September 31,197 39,397 42,279 0 93October 57,833 73,033 78,783 0 92 October 52,604 66,430 73,267 0 90November 73,514 92,836 82,292 18,112 92 November 69,456 87,712 94,428 0 93December 99,985 126,264 131,739 2,440 94 December 88,120 111,281 119,793 13,694 83January 77,225 97,523 100,910 0 96 January 92,282 116,537 125,758 0 92February 68,670 86,719 89,743 0 96 February 75,172 94,930 103,429 0 92March 74,719 94,358 97,847 0 96 March 69,305 87,521 96,134 0 91April 63,936 80,741 83,831 0 96 April 51,470 64,999 72,090 0 90May 49,822 62,917 65,449 0 96 May 32,993 41,665 47,231 0 88TOTAL 699,864 883,813 909,659 20,552 TOTAL 651,681 822,965 904,637 13,694Aver. eff 95% Aver. eff 89%Energy output=     699,864 KLBS steam x 0.0012628 TJ/KLBS Energy output= 651,681 KLBS steam x 0.0012628 IJ/KLBS =   823 TJequals to 884 TJ of total  energy output plus from BRDF* 148,920 KLBS steam x 0.0012628 TJ/KLBS =  188 TJequals to 1,011 TJ of total energy outputYear 2009 - 2010  PH Year 2012 - 2013    PH (and BRDF operational*)930 TJ 918 TJ Energy input NG +oil  Energy input NG +oil188  Energy input needed if wood completely replaced fossil fuel assuming the same energy output of 1,011 TJ considering efficiency of wood conversion of 68% and moisture content 54% dry basis as reported by  BRDF measurements To calculate energy from wood needed to produce the same amount of steam:  Energy from wood [GJ] = Steam demand [KLBS]/68 ·100  and to calculate the mass of wood with 54% MCD needed to produce the required energy: wood needed [t] = {wood energy needed [GJ] / 19.3 GJ/t}· 1.54  Table B.2-2 Wood requirements for 1,011TJ energy output.MonthSteam from PH [KLBS]Energy from PH steam [GJ] from wood@68%eff [GJ] Wood @ 54% MCD[t]June 39,867 50,345 74,037 5,908July 28,257 35,684 52,477 4,187August 20,957 26,465 38,920 3,106September 31,197 39,397 57,937 4,623October 52,604 66,430 97,691 7,795November 69,456 87,712 128,988 10,292December 88,120 111,281 163,648 13,058January 92,282 116,537 171,378 13,675February 75,172 94,930 139,602 11,139March 69,305 87,521 128,707 10,270April 51,470 64,999 95,586 7,627May 32,993 41,665 61,272 4,889TOTAL = 1,210,243 96,569plus from BRDF*  = 219,000 17,475114,043Year 2012 - 2013    PH (and BRDF operational*)TOTAL wood needed =189   Based on seasonal and diurnal ratios for steam obtained for 2010-2013 from both plants (PH and BRDF):                  2012/13 steam generation = 651,681 [KLBS]  at PH 7.80062E+11 [BTU]  = 822,965.18 [GJ]    from PH     148,920 [KLBS]  at BRDF 1.78257E+11 [BTU]  = 188,061.39 GJ]     from [BRDF   TOTAL= 800,601 KLBS = 9.58319E+11 [BTU]  = 1,011,026.96 [GJ]             is 1,011.03 TJ            output energy = energy demand   Table B.2-3 Seasonal distribution of energy demand of 1,011 TJ for 2012-2013. Parameter DAYTIME  [DT] NIGHTTIME [NT] Season total ratio DT/NT ratio season/year Units [TJ] [TJ] [TJ]    [%] summer 2012 78 74 151.7 1.06 15 fall 2012 125 107 232.5 1.16 23 winter 2012/13 223 172 394.3 1.29 39 spring 2013 125 107 232.5 1.16 23                 year total= 1,011 TJ               190  Emission estimates used in modeling scenarios SCENARIO 1: Base case as of 2012-2013, both PH and BRDF were operational, total energy produced 1,011 TJ [188,061.39 GJ of energy was produced by BRDF and 822,965.18 GJ by PH] Energy input = 1,194,802 GJ = 1,195 TJ (918,332 GJ from NG and oil and 276,560 GJ from biomass)    Table C-1 Scenario 1 Base case: Daytime and nighttime pollutant emissions from PH per month 2012-2013.  PM DAY [g] PM NIGHT [g] CO DAY [g] CO NIGHT [g] CH4 DAY [g] CH4 NIGHT [g] NO2 DAY [g] NO2 NIGHT [g] Jun 21,415 21,296 946,248 940,991 25,923 25,779 1,126,338 1,120,080 July 14,754 14,675 651,946 648,440 17,861 17,765 776,024 771,851 Aug 14,523 14,445 641,722 638,272 17,581 17,486 763,855 759,748 Sep 16,458 16,366 727,202 723,162 19,922 19,812 865,603 860,794 Oct 28,520 28,367 1,260,196 1,253,421 34,524 34,339 1,500,037 1,491,972 Nov 36,757 36,553 1,624,170 1,615,147 44,496 44,248 1,933,282 1,922,541 Dec 88,673 88,196 2,165,545 2,153,902 60,967 60,639 2,662,861 2,648,544 Jan 48,952 48,689 2,163,031 2,151,402 59,258 58,940 2,574,698 2,560,856 Feb  40,261 40,021 1,778,981 1,768,392 48,737 48,447 2,117,556 2,104,951 Mar 37,421 37,220 1,653,508 1,644,618 45,299 45,056 1,968,203 1,957,621 Apr 28,062 27,906 1,239,943 1,233,054 33,969 33,781 1,475,928 1,467,729 May 18,385 18,286 812,373 808,005 22,256 22,136 966,984 961,785   191   Table C-2 Scenario 1 Base case: Daytime and nighttime pollutant emissions from BRDF per month 2012-2013.  PM DAY [g] PM NIGHT [g] CO DAY [g] CO NIGHT [g] CH4 DAY [g] CH4 NIGHT [g] NO2 DAY [g] NO2 NIGHT [g] Jun 4,609 4,584 168,241 167,306 104,056 103,478 842,357 837,677 July 4,609 4,585 168,241 167,336 104,056 103,496 842,357 837,828 Aug 4,609 4,585 168,241 167,336 104,056 103,496 842,357 837,828 Sep 4,609 4,584 168,241 167,306 104,056 103,478 842,357 837,677 Oct 4,609 4,585 168,241 167,336 104,056 103,496 842,357 837,828 Nov 4,609 4,584 168,241 167,306 104,056 103,478 842,357 837,677 Dec 4,609 4,585 168,241 167,336 104,056 103,496 842,357 837,828 Jan 4,609 4,585 168,241 167,336 104,056 103,496 842,357 837,828 Feb  4,609 4,582 168,241 167,239 104,056 103,436 842,357 837,342 Mar 4,609 4,585 168,241 167,336 104,056 103,496 842,357 837,828 Apr 4,609 4,584 168,241 167,306 104,056 103,478 842,357 837,677 May 4,609 4,585 168,241 167,336 104,056 103,496 842,357 837,828      192  Table C-3 Scenario 1 Base case: Resulting emissions daytime and nighttime pollutant emissions from PH and BRDF per month 2012-2013. Period  PM DAY [g] PM NIGHT [g] CO DAY [g] CO NIGHT [g] CH4 DAY [g] CH4 NIGHT [g] NO2 DAY [g] NO2 NIGHT [g] Jun 26,024 25,880 1,114,489 1,108,297 129,979 129,257 1,968,694 1,957,757 July 19,364 19,260 820,186 815,777 121,916 121,261 1,618,380 1,609,679 Aug 19,132 19,030 809,963 805,609 121,636 120,982 1,606,211 1,597,576 Sep 21,067 20,950 895,443 890,468 123,978 123,289 1,707,960 1,698,471 Oct 33,129 32,951 1,428,437 1,420,758 138,580 137,835 2,342,393 2,329,800 Nov 41,367 41,137 1,792,411 1,782,453 148,551 147,726 2,775,638 2,760,218 Dec 93,282 92,781 2,333,786 2,321,238 165,023 164,135 3,505,217 3,486,372 Jan 53,562 53,274 2,331,272 2,318,738 163,314 162,436 3,417,055 3,398,684 Feb  44,870 44,603 1,947,222 1,935,631 152,793 151,883 2,959,912 2,942,294 Mar 42,031 41,805 1,821,749 1,811,954 149,355 148,552 2,810,560 2,795,449 Apr 32,671 32,489 1,408,184 1,400,360 138,025 137,258 2,318,285 2,305,406 May 22,994 22,871 980,614 975,342 126,312 125,632 1,809,340 1,799,612     193  SCENARIO 2 and 4:  Case when PH would be operational only, total energy produced 1,011 TJ Energy input = 1,133,232 GJ =1,133 TJ  Table C-4 Scenario 2: Daytime and nighttime pollutant emissions per month 2012-2013 if only PH is operational.  Period PM DAY [g] PM NIGHT [g] CO DAY [g] CO NIGHT [g] CH4 DAY [g] CH4 NIGHT [g] NO2 DAY [g] NO2 NIGHT [g] Jun 28,422 28,264 1,255,848 1,248,871 34,405 34,214 1,494,861 1,486,556 July 21,742 21,625 960,686 955,521 26,319 26,177 1,143,523 1,137,375 Aug 21,510 21,395 950,462 945,352 26,039 25,899 1,131,354 1,125,272 Sep 23,464 23,334 1,036,802 1,031,042 28,404 28,246 1,234,126 1,227,270 Oct 35,527 35,336 1,569,796 1,561,357 43,006 42,775 1,868,560 1,858,514 Nov 43,764 43,521 1,933,770 1,923,027 52,977 52,683 2,301,805 2,289,017 Dec 58,968 58,651 2,605,585 2,591,576 71,382 70,999 3,101,479 3,084,804 Jan 55,959 55,658 2,472,631 2,459,337 67,740 67,376 2,943,221 2,927,398 Feb  46,878 46,599 2,071,381 2,059,051 56,747 56,410 2,465,605 2,450,929 Mar 44,428 44,189 1,963,108 1,952,554 53,781 53,492 2,336,726 2,324,163 Apr 35,068 34,873 1,549,543 1,540,934 42,451 42,215 1,844,451 1,834,205 May 25,392 25,255 1,121,973 1,115,941 30,737 30,572 1,335,507 1,328,326     194  SCENARIO 3:  Fossil fuels are completely replaced with wood, BRDF would be operational only, total energy produced 1,011 TJ Energy input = 1,486,803.20 GJ = 1,487 TJ  Table C-5  Scenario 3 Daytime and nighttime pollutant emissions per month 2012-2013 if only BRDF is operational. Period PM DAY [g] PM NIGHT [g] CO DAY [g] CO NIGHT [g] CH4 DAY [g] CH4 NIGHT [g] NO2 DAY [g] NO2 NIGHT [g] Jun 19,417 19,309 708,713 704,775 438,334 435,899 3,548,418 3,528,705 July 15,105 15,023 551,320 548,356 340,988 339,155 2,760,377 2,745,536 Aug 12,393 12,327 452,355 449,923 279,778 278,274 2,264,872 2,252,695 Sep 16,197 16,107 591,180 587,896 365,641 363,609 2,959,950 2,943,505 Oct 24,148 24,018 881,385 876,646 545,130 542,200 4,412,960 4,389,234 Nov 30,407 30,238 1,109,851 1,103,685 686,435 682,622 5,556,857 5,525,986 Dec 37,339 37,138 1,362,872 1,355,544 842,927 838,395 6,823,692 6,787,006 Jan 38,885 38,676 1,419,301 1,411,670 877,828 873,108 7,106,227 7,068,021 Feb  32,530 32,336 1,187,337 1,180,270 734,360 729,989 5,944,819 5,909,433 Mar 30,351 30,188 1,107,803 1,101,847 685,168 681,485 5,546,602 5,516,781 Apr 23,727 23,595 866,019 861,208 535,627 532,651 4,336,026 4,311,937 May 16,864 16,773 615,528 612,219 380,700 378,653 3,081,857 3,065,288    195  SCENARIO 5:  Case when only PH would be operational, total energy produced 884 TJ as in 2009-2010.  Energy input =  930,211.17 GJ = 930 TJ Table C-6 Scenario 5 Daytime and nighttime pollutant emissions per month 2009-2010 when only PH was operational. Period PM DAY [g] PM NIGHT [g] CO DAY [g] CO NIGHT [g] CH4 DAY [g] CH4 NIGHT [g] NO2 DAY [g] NO2 NIGHT [g] Jun 17,087 16,992 754,992 750,798 20,684 20,569 898,682 893,689 July 17,635 17,541 779,243 775,054 21,348 21,233 927,549 922,562 Aug 15,904 15,818 702,729 698,951 19,252 19,148 836,472 831,975 Sep 19,078 18,972 842,974 838,291 23,094 22,966 1,003,409 997,834 Oct 30,667 30,502 1,355,075 1,347,790 37,124 36,924 1,612,973 1,604,301 Nov 87,637 87,150 1,554,431 1,545,796 44,754 44,505 1,962,914 1,952,009 Dec 58,772 58,456 2,284,643 2,272,360 62,882 62,544 2,734,631 2,719,928 Jan 39,280 39,069 1,735,644 1,726,313 47,550 47,294 2,065,972 2,054,864 Feb  34,933 34,725 1,543,571 1,534,383 42,288 42,036 1,837,343 1,826,407 Mar 38,088 37,883 1,682,962 1,673,914 46,106 45,858 2,003,263 1,992,492 Apr 32,632 32,451 1,441,886 1,433,876 39,502 39,282 1,716,306 1,706,771 May 25,477 25,340 1,125,723 1,119,671 30,840 30,674 1,339,970 1,332,766     196  Results of ambient air quality and health risks assessment Table D-1 Summary of ambient air quality, iF and IS for five district heating operational scenarios at UBC.  PM2.5 CO NO2 PM2.5 CO NO2 PM2.5 CO NO2 PM2.5 CO NO2 PM2.5 CO NO2Mean [µg/m3] 0.012 0.499 0.892 0.014 0.607 0.723 0.008 0.299 1.495 0.008 0.299 1.495 0.011 0.471 0.561Max [µg/m3] 1.74 76.88 91.52 2.31 102.04 121.45 1.03 37.46 187.59 1.03 37.46 187.59 1.40 61.73 73.47∑iF (ppm) 26.35 26.76 23.00 29.74 29.74 29.74 18.20 18.20 18.20 8.62 8.62 8.62 32.74 32.74 32.74Mean [µg/m3] 0.007 0.302 0.480 0.009 0.390 0.464 0.003 0.118 0.593 0.003 0.118 0.593 0.007 0.301 0.359Max [µg/m3] 0.46 20.18 29.51 0.61 26.79 31.89 0.65 23.80 119.17 0.65 23.80 119.17 0.43 19.04 22.67∑iF (ppm) 1.36 1.34 1.55 1.24 1.24 1.24 2.10 2.10 2.10 0.61 0.61 0.61 1.35 1.35 1.35Mean [µg/m3] 0.012 0.536 0.836 0.015 0.640 0.761 0.008 0.289 1.449 0.008 0.289 1.449 0.017 0.529 0.640Max [µg/m3] 2.09 92.46 110.06 2.60 115.18 137.10 2.05 74.65 373.73 2.05 74.65 373.73 3.65 99.43 118.34∑iF (ppm) 20.44 20.76 17.43 23.48 23.48 23.48 12.14 12.14 12.14 12.14 12.14 12.14 23.48 23.48 23.48Mean [µg/m3] 0.006 0.257 0.384 0.007 0.310 0.369 0.003 0.112 0.563 0.003 0.112 0.563 0.009 0.255 0.311Max [µg/m3] 1.79 79.20 97.41 2.23 98.66 117.44 1.42 51.84 259.58 1.42 51.84 259.58 1.93 85.17 101.37∑iF (ppm) 0.77 0.78 0.74 0.81 0.81 0.81 0.77 0.77 0.77 0.77 0.77 0.77 0.81 0.81 0.81Mean [µg/m3] 0.022 0.776 1.156 0.019 0.838 0.998 0.013 0.456 2.284 0.013 0.456 2.284 0.015 0.645 0.769Max [µg/m3] 2.28 82.89 101.11 2.13 94.02 111.91 2.39 87.25 436.87 2.39 87.25 436.87 1.50 65.99 78.56∑iF (ppm) 15.76 15.75 15.32 15.93 15.93 15.93 13.55 13.55 13.55 13.55 13.55 13.55 15.93 15.93 15.93Mean [µg/m3] 0.012 0.413 0.577 0.011 0.312 0.371 0.005 0.167 0.834 0.005 0.167 0.834 0.008 0.351 0.419Max [µg/m3] 3.19 82.38 98.10 2.13 94.15 112.07 1.48 53.93 270.03 1.48 53.93 270.03 2.12 82.23 98.43∑iF (ppm) 0.66 0.65 0.66 0.44 0.44 0.44 0.67 0.67 0.67 0.67 0.67 0.67 0.65 0.65 0.65Mean [µg/m3] 0.018 0.767 1.294 0.020 0.881 1.048 0.012 0.449 2.247 0.012 0.449 2.247 0.018 0.807 0.961Max [µg/m3] 1.55 62.56 81.84 1.85 81.00 97.16 1.14 41.67 208.63 1.14 41.67 208.63 1.73 76.64 91.23∑iF (ppm) 29.96 29.37 27.93 31.28 31.28 31.28 21.43 21.43 21.43 21.43 21.43 21.43 31.29 31.29 31.29Mean [µg/m3] 0.009 0.382 0.607 0.010 0.451 0.537 0.005 0.172 0.859 0.005 0.172 0.859 0.009 0.415 0.495Max [µg/m3] 1.33 53.44 69.91 1.58 69.72 83.00 1.50 54.88 274.77 1.50 54.88 274.77 1.35 59.77 71.15∑iF (ppm) 1.02 0.99 1.10 0.97 0.97 0.97 1.28 1.28 1.28 1.28 1.28 1.28 0.97 0.97 0.9796.32 96.40 87.73 103.89 103.89 103.89 70.14 70.14 70.14 59.07 59.07 59.07 107.22 107.22 107.22EFhealth[DALY/kg] 0.0007 7.31E-07 8.91E-05 0.0007 7.31E-07 8.91E-05 0.0007 7.31E-07 0.0000891 0.0007 7.31E-07 0.0000891 0.0007 7.3E-07 8.91E-05m [kg] 897 35,270 57,521 880 38,876 46,275 593 21,648 108,387 593 21,648 108,387 832 31,521 37,77560 2 450 64 3 428 29 1 677 25 1 570 62 2 361*1-hour averaging period **iF expressed in per million , ppmExceedances of Air Quality Objectives presented in boldScenario 5: NG/oil  with 2009/10 emissions ScenarioScenario 1: Base case: Biomass and NG/oilScenario 2: NG onlyScenario 3: Biomass onlyScenario 4: Biomass and varying population Period / Parameter *SUMMERDAYtimeNIGHTtimeFALL DAYtimeNIGHTtimeWINTER DAYtimeNIGHTtimeSPRING DAYtimeNIGHTtime∑iF annual [ppm]∑ IS annual [DALY]**∑ IS scenario [DALY] 513 495 596 426708197  Global impacts data  Emission factors for energy products   Table E.1-1 Emission factors for natural gas.   Pollutant Upstream [kg/MJ] Combustion [kg/MJ] TOTAL [kg/MJ] CO2 fossil 8.41E-03 4.92E-02 5.76E-02 CO2 biogenic 7.98E-05 - 7.98E-05 CH4 1.60E-04 9.42E-07 1.61E-04 CH4 biogenic - - - N2O 2.15E-07 9.02E-07 1.12E-06 NOx as NO2 4.55E-05 4.01E-05 8.65E-05 SOx 1.03E-05 2.58E-07 1.05E-05 PM 5.96E-07 7.79E-07 1.38E-06 CO 6.60E-06 3.44E-05 4.10E-05 CO biogenic - - - NMVOC 3.46E-06 2.25E-06 5.72E-06 198    Table E.1-2 Emission factors for heavy fuel oil.    Pollutant Upstream [kg/MJ] Combustion [kg/MJ] TOTAL [kg/MJ] CO2 fossil 1.13E-02 6.85E-02 7.98E-02 CO2 biogenic 4.75E-04 - 4.75E-04 CH4 1.43E-04 6.60E-07 1.44E-04 CH4 biogenic - - - N2O 3.67E-07 8.0E-07 1.17E-06 NOx 4.17E-05 3.71E-05 7.24E-05 SOx 3.34E-05 4.56E-05 4.89E-04 PM 1.99E-06 6.14E-06 8.13E-06 CO 1.09E-05 1.54E-05 2.63E-05 CO biogenic - - - NMVOC 3.99E-06 1.04E-06 5.03E-06 199   Table E.1-3 Emission factors for middle distillates. *GHGenius based on AP-42 emission factors.  Pollutant Upstream [kg/MJ] Combustion* [kg/MJ] TOTAL [kg/MJ] CO2 fossil 1.82E-02 7.05E-02 8.88E-02 CO2 biogenic 6.20E-04  6.20E-04 CH4 1.57E-04 1.6E-07 1.57E-04 CH4 biogenic - - - N2O 1.0E-06 2.86E-05 2.93E-05 NOx 5.50E-05 4.0E-05 9.45E-05 SOx 6.80E-05 6.54E-07 6.84E-05 PM 5.50E-06 3.8E-06 8.55E-06 CO 1.80E-05 2.15E-05 3.90E-05 CO biogenic - - - NMVOC 4.91E-06 - 4.91E-06 200   Table E.1-4 Emission factors for middle distillates for HDV operation. The totals were calculated before rounding upstream and vehicle operation emissions.      Pollutant Upstream [kg/tkm] Vehicle operation [kg/tkm] TOTAL [kg/tkm]* CO2 fossil 3.76E-02 1.43E-01 1.81E-01 CO2 biogenic 1.33E-06 - 1.33E-06 CH4 3.25E-04 8.87E-06 3.34E-04 CH4 biogenic - - - N2O 1.44E-06 6.17E-06 7.61E-06 NOx 1.13E-04 5.39E-05 1.66E-04 SOx 1.40E-04 5.35E-06 1.45E-04 PM 9.81E-06 2.64E-06 1.24E-05 CO 3.62E-05 2.47E-05 6.09E-05 CO biogenic - - - NMVOC 1.01E-05 1.49E-05 2.50E-05 201   Annual emissions over life cycle stages Table E.2-1 Annual emission by process and transport stages for Scenario 1: NG, fuel oil and biomass.            *Includes both transportation segments – from industry to Cloverdale and from Cloverdale to UBC. ** ESP in place.    Pollutant NG  upstream [kg/yr] Oil  upstream [kg/yr] NG  combustion   [kg/yr] Oil   combustion [kg/yr] Wood transport* [kg/yr] Wood processing Cloverdale [kg/yr] Wood gasification [kg/yr] TOTAL   [kg/yr] CO2 fossil 7.61E+06 1.55E+05 4.45E+07 9.38E+05 1.32E+05 8.93E+03 - 5.33E+07 CO2 bio. 7.22E+04 6.50E+03 - - 9.75E-01 4.33E+03 2.54E+07 2.54E+07 CH4 1.45E+05 1.96E+03 8.53E+02 9.04E+00 2.44E+02 7.50E+01 2.50E+03 1.48E+05 N2O 1.94E+02 5.03E+00 8.16E+02 1.10E+01 5.58E+00 3.75E-01 1.55E+03 2.58E+03 NOx (NO2) 4.12E+04 5.71E+02 3.70E+04 4.21E+02 1.22E+02 2.19E+01 2.02E+04 9.96E+04 SOx 9.32E+03 4.57E+02 0.00E+00 6.24E+03 1.06E+02 1.80E+01 - 1.61E+04 PM 5.39E+02 2.73E+01 7.04E+02 8.41E+01 9.12E+00 4.69E+01 1.11E+02** 1.25E+04 CO 5.97E+03 1.49E+02 3.11E+04 2.10E+02 4.47E+01 3.75E+04 4.04E+03 3.75E+04 NMVOC 3.13E+03 5.46E+01 2.04E+03 1.42E+01 1.83E+01 1.81E+00 1.19E+03 6.45E+03 CO2 eq  [kg/year] 1.17E+07 2.11E+05 4.48E+07 9.42E+05 1.41E+05 1.12E+04 4.88E+05 5.83 E+07 (CO2eq) 202  Table E.2-2 Annual emission by process for Scenario 2: Natural gas only.                 Pollutant NG  upstream [kg/yr] NG  combustion   [kg/yr] TOTAL   [kg/yr] CO2 fossil 9.53E+06 5.57E+07 6.53E+07 CO2 biogenic 9.04E+04 - 9.04E+04 CH4 1.81E+05 1.07E+03 1.82E+05 N2O 2.44E+02 1.02E+03 1.27E+03 NOx (NO2) 5.16E+04 4.64E+04 9.80E+04 SOx 1.17E+04 0.00E+00 1.17E+04 PM 6.75E+02 8.82E+02 1.56E+03 CO 7.48E+03 3.90E+04 4.65E+04 NMVOC 3.92E+03 2.55E+03 6.47E+03 CO2 eq   [kg/year] 1.47E+07 5.61E+07 7.08E+07  (CO2eq) 203    Table E.2-3 Annual emission by process and transport stages for Scenario 3: Biomass only.          *Includes both transportation segments – from industry to Cloverdale and from Cloverdale to UBC. ** ESP in place. Pollutant Wood transport* [kg/yr] Wood processing Cloverdale [kg/yr] Wood gasification [kg/yr] TOTAL   [kg/yr] CO2 fossil 1.05E+06 5.36E+04 0.00E+00 1.11E+06 CO2 bio. 7.76E+00 2.60E+04 1.36E+08 1.36E+08 CH4 1.95E+03 4.49E+02 1.34E+04 1.58E+04 N2O 4.44E+01 2.25E+00 8.31E+03 8.36E+03 NOx (NO2) 9.71E+02 1.31E+02 1.09E+05 1.10E+05 SOx 8.47E+02 1.08E+02 0.00E+00 9.55E+02 PM 7.26E+01 2.81E+02 5.93E+02** 9.47E+02 CO 3.56E+02 4.62E+01 2.17E+04 2.21E+04 NMVOC 1.46E+02 1.09E+01 6.39E+03 6.55E+03 CO2 eq  [kg/year] 1.12E+06 6.69E+04 2.62E+06 3.81E+06 (CO2eq) 204   Table E.2-4 Annual emission by process and transport stages for Scenario 3: Biomass only, changed transportation distance.                *Includes both transportation segments – from industry to Cloverdale and from Cloverdale to UBC. ** ESP in place.  Pollutant Wood transport* [kg/yr] Wood processing Cloverdale [kg/yr] Wood gasification [kg/yr] TOTAL   [kg/yr] CO2 fossil 2.01E+06 5.36E+04 - 2.06E+06 CO2 bio. 1.48E+01 2.60E+04 1.36E+08 1.36E+08 CH4 3.70E+03 4.49E+02 1.34E+04 1.76E+04 N2O 8.46E+01 2.25E+00 8.31E+03 8.40E+03 NOx (NO2) 1.85E+03 1.31E+02 1.09E+05 1.11E+05 SOx 1.61E+03 1.08E+02 0.00E+00 1.72E+03 PM 1.38E+02 2.81E+02 5.93E+02** 1.01E+03 CO 6.77E+02 2.33E+03 2.17E+04 2.25E+04 NMVOC 2.78E+02 1.09E+01 6.39E+03 6.68E+03 CO2 eq  [kg/year] 2.13E+06 6.69E+04 2.628E+06 4.83E+06 (CO2eq) 205  Meeting CAP2020 GHG reduction goals According to the latest UBC report (Wauthy and Giffin, 2017), two energy supply options are being considered for reaching the University’s Climate Action Plan (CAP) for GHG reductions goals of 67% below 2007 level by 2020: - Displacement of natural gas with carbon neutral Renewable Natural Gas (RNG) at the newly constructed and operational CEC (Campus Energy Centre), and  - Expansion of the existing BRDF with an addition of a biomass boiler.    The second option, an expansion of the existing BRDF, is discussed here. It should be noted that the intention of this discussion is not to provide a detailed economic analysis which is beyond the scope of this study, but rather to reveal some economic aspects of this possibly future option.  The new ADES (Academic District Energy System) center at UBC which included CEC (Campus Energy Centre) costed $88.3 M (UBC, 2013) and was designed with natural gas as the fuel. With respect to fuel choices as previously discussed BRDF which utilizes biomass is a good contribution to GHG reduction. However, costs associated with its construction and maintenance could pose obstacles for its adoption. In order to meet heating demand target of 1011 TJ used in this study, capital cost of BRDF expansion is calculated using a cost scaling factor of 0.6 based on the ratio of plant’s potential and current output as per the following equation: Capital investment for the new plant =       (7-1) = (Total energy demand / Current BRDF heat output capacity)^0.6 x initial capital  investment for BRDF  206  Where: total energy demand is 1,011 TJ, BRDF heat output as of 2012-2013 is 188 TJ and capital investment for the heating portion of plant of $19.2 M. The investment for an expanded BRDF in order to meet the UBC campus heating demand is thus estimated to be $52.7 M.  It is assumed that biomass would be mostly sourced locally as it is at present, since supply analysis indicated a large surplus of solid wood waste in the region at a pretty stable cost compared to other commodities (Wauthy and Giffin, 2017). Here, $79.59/OMDT (a 5-year fixed price of $71/OMDT plus GST and PST) as of 2017 is considered in O&M calculations based on the latest Commodity report (Wauthy and Giffin, 2017). According to this report, biomass prices are expected to remain stable over a period of time due to increased supply forecasts. With a thermal efficiency of 68% (calculated based on 2012-2013 data), 74,054 ODMT of wood is needed to meet energy demand, which would cost $5.9 M annually. Costs of other commodities were not included in this analysis.  Ash disposal ($3.3 K) and plant operation employee salaries ($1.8 M) are estimated arbitrarily39 using factors of 5.5 and 3, respectively whereas other O&M costs are estimated as earlier stated as 5% of capital investments (equals to $2.6 M). Carbon offset for wood purchased at $0.06/GJinput would result in an annual carbon cost of $8.9 K. The total annual O&M cost would in such case be $10.5 M and $184 M over the plants’ lifetime (20 years). The total PV, which includes both capital and O&M costs for expanded BRDF, is estimated to reach $237 M which indicates increased costs of $28 M  compared to option A (PH only) and almost $8 M compared                                                  39 Estimates based on the expansion factor and assuming the same operating conditions. 207  to option B (PH and BRDF). It should be noted that those expenditures may be even higher when other costs such as other commodities and their respective carbon taxes, biomass storage and other parameters are included in the economic analysis.  When externalities in case of using biomass to meet the total energy demand for UBC campus are considered, there exist noticeable savings in the total PV of external costs: $18 M compared to  option A (natural gas only) and almost $15M compared to option B (combined biomass and natural gas). This is largely due to the avoided costs of CO2 when biomass is used.  However, external costs associated with fine particles increase to $2 M and oxides of nitrogen to $8 M over the plants’ lifetime. Figure 6.4 illustrates PV costs of considered pollutants excluding CO2 for previously discussed options and potential expanded BRDF plant.      Figure F 1 External costs for option A (natural gas only), option B (natural gas and biomass) and potential BRDF expansion, over plants’ lifetime. CO2 costs are excluded.  Preliminary investigation into possible BRDF expansion indicated that costs calculated as total PV will be higher than options A and B, external costs of NOx and PM2.5 over plants’ lifetime 0123456789CH4 N2O NOx as NO2 SOx PM2.5 CO NMVOC[$ M]PV [$ M]Option A Option B Expanded BRDF208  will increase by $8 M and $2 M respectively whereas savings in carbon tax and offsets will be noticeable at $25 M compared to option A and $20 M compared to option B. 

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}]}"
                            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.24.1-0365819/manifest

Comment

Related Items