Loading...
Bainbridge_GHG Inventory Report_FINAL_20191122_201911271016146596CITY OF BAINBRIDGE ISLAND GREENHOUSE GAS EMISSIONS INVENTORY FINAL FINDINGS REPORT 2019 CASCADIA CONSULTING GROUP, INC. Cascadia Consulting Group, Inc. Tel (206) 343-9759 Fax (206) 343-9819 www.cascadiaconsulting.com CITY OF BAINBRIDGE ISLAND | 2019 | 2 Table of Contents Acronyms ___________________________________________________________________________ 3 Executive Summary ___________________________________________________________________ 4 Inventory Results ___________________________________________________________________ 6 Introduction ________________________________________________________________________ 11 Greenhouse Gas Inventory Overview __________________________________________________ 12 Roadmap of this Report _____________________________________________________________ 17 Communitywide Emissions ____________________________________________________________ 18 Results __________________________________________________________________________ 19 Data Sources and Methodology _______________________________________________________ 31 Community Contribution Analysis _____________________________________________________ 40 Municipal Operation Emissions _________________________________________________________ 43 Results __________________________________________________________________________ 43 Data Sources and Methodology _______________________________________________________ 52 Municipal Contribution Analysis ______________________________________________________ 58 Consumption-Based Emissions _________________________________________________________ 61 Results __________________________________________________________________________ 61 Data Sources and Methodology _______________________________________________________ 62 Conclusion and Next Steps _____________________________________________________________ 63 References _________________________________________________________________________ 64 Appendix: Tree Carbon Sequestration ______________________________________________________ i Data Sources and Methodology _________________________________________________________ i Results ___________________________________________________________________________ iii CITY OF BAINBRIDGE ISLAND | 2019 | 3 Acronyms ACS American Community Survey (from the U.S. Census Bureau) BOD Biochemical oxygen demand (a metric of the effectiveness of wastewater treatment plants) EIA United States Energy Information Association EPA United States Environmental Protection Agency CO2e Carbon dioxide equivalent GHG Greenhouse gas (limited to CO2, CH4, N2O, and fugitive gases in this inventory) GWP Global Warming Potential HFC Hydrofluorocarbon MTCO2e Metric tons of carbon dioxide equivalent MOVES Motor Vehicle Emission Simulator model (developed by EPA to quantify emissions from mobile sources) MPG Miles per gallon NONROAD Part of MOVES model developed by EPA to quantify non-road mobile emissions O&M Operations and Maintenance ODS Ozone depleting substance PFC Perfluorocarbon PSCAA Puget Sound Clean Air Agency PSE Puget Sound Energy PSRC Puget Sound Regional Council SF6 Sulfur hexafluoride TCR The Climate Registry USDA United States Department of Agriculture WARM Waste Reduction Model (model developed by EPA to quantify solid waste emissions) WSDOT Washington State Department of Transportation WWTP Wastewater Treatment Plant VMT Vehicle Miles Traveled CITY OF BAINBRIDGE ISLAND | 2019 | 4 Executive Summary Global climate change poses a growing threat and humanitarian climate emergency, and Bainbridge Island shares the responsibility to reduce greenhouse gas emissions (GHG) to stabilize the global climate while preparing for the effects of climate change. The City of Bainbridge Island recently completed a comprehensive greenhouse gas inventory as part of its commitment to reducing emissions and leading on climate action. Greenhouse gas emission inventories quantify the amount of climate pollution produced by an entity—in this case, from the Bainbridge Island community and municipal government operations. As the Bainbridge Island government and community continues to take action to reduce GHG emissions, these inventories will serve as tools for tracking progress and making improvements along the way. Inventory Approaches This report describes findings from three distinct inventories: • A community inventory that estimates GHG emissions produced by activities of the Bainbridge Island community, including residents and businesses. The community GHG inventory includes emissions due to energy used to power and heat homes and businesses; fuel used by vehicles running within Bainbridge Island; solid waste that is generated by the community; agriculture activities; fuel use by off-road equipment and vehicles; and others. • A municipal inventory that accounts for the GHG emissions resulting from City of Bainbridge Island government operations. This inventory can help the City understand GHG emissions stemming from various activities associated with municipal operations, including from municipal building and facility operation, transportation, solid waste, wastewater, and refrigerant leakage. • A consumption-based inventory that estimates GHG emissions associated with the consumption of food, goods, and services within the community, regardless of their origin. For example, this inventory would not include emissions from the production of locally-manufactured goods that are consumed entirely outside the community; however, it would include emissions associated with the production of goods manufactured in another community but consumed by Bainbridge Island residents, visitors, or businesses. This inventory can be examined in context with the community inventory to paint a more comprehensive picture of community emissions. This report also presents findings from additional analyses:  Municipal and communitywide contribution analyses that identify key drivers of observed emission trends. For example, analysis calculates the impact that a hotter summer or colder winter may have had on household energy use, and thus, emissions.  A supplemental carbon sequestration analysis that estimates the amount of carbon dioxide that Bainbridge Island trees absorb—or sequester—from the atmosphere on an annual basis (in Appendix). CITY OF BAINBRIDGE ISLAND | 2019 | 5 Inventory Methodology The GHG inventories summarized in this report account for human-caused emissions of the most prominent and typical greenhouse gases for communities: carbon dioxide (CO2), methane (CH4), and nitrous oxide (N2O). To account for the differences in potency among these gases, all emissions are calculated and reported in units of metric tons of carbon dioxide equivalent (MTCO2e). The inventories were conducted using widely accepted tools and protocols, including the Local Government Operations Protocol and U.S. Community Protocol for Accounting and Reporting of Greenhous Gas Emissions. GHG emissions are inventoried by multiplying annual activity data (e.g. electricity consumption) by emission factors (greenhouse gas emissions produced per kWh of electricity). For the municipal operations inventory, we obtained most data from City staff and local documentation such as Puget Sound Energy's (PSE) billing records, records from local wastewater treatment facilities, and fleet vehicle fuel use records. For the communitywide inventory, key data sources included electricity consumption data from PSE, Puget Sound Regional Council (PSRC) modeling outputs for vehicle miles traveled by fuel type, Bainbridge Disposal tonnage records, and Kitsap County Conservation District estimates of agricultural landowners and acreage. Where local data were not available, we referenced and downscaled national data sources such as from the Federal Transit Administration, U.S. Census Bureau, and the U.S. Department of Agriculture. Inventory Boundaries The activities and sectors included in GHG inventories are often classified into three “scopes,” which represent relative levels of control over an emissions source:  Scope 1 emission sources include those directly caused by an organization’s actions, such as from owned equipment and facilities.1  Scope 2 emissions are those indirectly associated with purchased electricity, steam, heating, or cooling.  Scope 3 includes all other indirect emissions that are not covered in Scope 2. The communitywide and municipal inventories for Bainbridge Island included emissions sources from all three scopes: Scope 1, Scope 2, and Scope 3. The inventories included all sources required by the consulted protocols and additional sources, as relevant. Inventory Years The inventories summarized in this report cover two representative years, 2014 and 2018. To provide Bainbridge Island with the most comprehensive, consistent, and relevant information on their GHG emissions, we selected the most recent year available as an inventory year – 2018. For a comparison year, we selected 2014 because no major changes in organizational structure or infrastructure occurred after that year. Additionally, choosing a more recent comparison year increases the likelihood that all needed data are available and derived using consistent methodologies. 1 Except direct carbon dioxide emissions from biogenic sources. CITY OF BAINBRIDGE ISLAND | 2019 | 6 INVENTORY RESULTS Community Emissions The Bainbridge Island community emitted an estimated 233,998 metric tons of carbon dioxide equivalent (MTCO2e) in 2018, or 9.4 MTCO2e per Bainbridge Island resident. A high-level comparison suggests that this per-capita estimate is consistent with that of Kitsap County (9.9 MTCO2e per person), and lower than per-capita estimates for the U.S., Washington State, King County, and Bellevue. However, Bainbridge Island’s estimated per-capita emissions are almost twice those of Seattle and Tacoma (Figure 6). The majority of Bainbridge Island community emissions stem from consumption of electricity in homes and commercial buildings (Figure 4). Figure 1. Bainbridge Island communitywide emissions in 2018 (total = 233,998 MTCO2e). CITY OF BAINBRIDGE ISLAND | 2019 | 7 Figure 2. Per-capita GHG emissions comparison.2 Overall, communitywide emissions have increased by 9% since 2014 (Figure 5). Per-capita emissions, however, only increased 1% over that period, suggesting that population growth contributed to the observed trends. Changes in electricity fuel source (e.g., renewables versus coal) and growth in employment also pushed emissions upward, while improvements in vehicle fuel economy, reductions in the distance each person drives, and declining per-household and per-business energy consumption on Bainbridge Island softened the extent of those increases. Figure 3. Bainbridge Island community emissions trends, by year and source. 2 Other jurisdictions may use different data sets, methods, and years for their GHG emission inventories. CITY OF BAINBRIDGE ISLAND | 2019 | 8 City Government Emissions Emissions from City of Bainbridge Island activities—which make up about 1% of the total community emissions—increased 11% from 2014 to 2018, totaling 2,291 MTCO2e in 2018 (Figure 7 and Figure 8). Major emissions sources included facility electricity consumption (60%) and on-road fleet vehicles (17%). Emissions from municipal facility electricity and on-road fleet vehicles increased 14% and 7%, respectively. Just four facilities accounted for 80% of all facility electricity use in 2018: Bainbridge Island Waste Water Treatment Plant (WWTP), City Hall, Fletcher Bay: Well Field, and Bainbridge Island Public Works Operations and Maintenance Yard. Among the largest emissions decreases were from streetlights and traffic signals improvements. Figure 4. City operations GHG emissions in 2018 (total = 2,291 MTCO2e). Figure 5. Bainbridge Island city operations emissions trends, by year and source. CITY OF BAINBRIDGE ISLAND | 2019 | 9 Consumption-based Emissions The purchasing decisions we make impact the environment. Some types of foods and materials, such as meat and furniture, can carry a significant GHG emissions burden. For example, meat and dairy cows emit methane—a potent greenhouse gas. Residents in Bainbridge Island who consume beef contribute to the emissions from these cows—even if the cows are raised outside the island. Results from a household-based economic modeling tool suggest that the average Bainbridge Island household emits 52 MTCO2e a year through their purchasing behaviors.3 There were an estimated 9,404 and 9,798 households on Bainbridge Island in 2014 and 2018, respectively, indicating that total consumption-based emissions from all households on Bainbridge Island could have reached approximately 510,000 MTCO2e in 2018. Major drivers include the purchase of meat, furniture, clothing, home energy, and travel-related expenses such as car fuel and air travel. Figure 6. Consumption-based emissions per Bainbridge Island household. 3 As indicated from U.C. Berkeley’s CoolClimate Calculator. Outcomes from the consumption-based inventory analysis are presented at the per-household level because purchasing behavior is typically examined and analyzed at the household—not individual—level. CITY OF BAINBRIDGE ISLAND | 2019 | 10 Next Steps In analyzing the GHG emissions of Bainbridge Island, this report identifies key activities and sectors at both citywide and municipal operations scales that contribute the largest relative amounts of GHG emissions. Drawing from the results in this report, activities and operations can then be targeted for improvement and mitigation. Improving the inventory process, conducting regular inventories, and incorporating these inventory results into decision-making processes will be critical for evaluating progress toward emissions reductions targets and for identifying cost-saving opportunities in the future. CITY OF BAINBRIDGE ISLAND | 2019 | 11 Introduction This GHG inventory report for the City of Bainbridge Island details estimated GHG emissions from activities within the city of Bainbridge Island. GHG inventories quantify emissions associated with a specific “scope”—or boundary condition—over a specific period. By conducting GHG inventories at regular intervals, they can illuminate trends, accomplishments, and opportunities for improvement. They also hold governments and communities responsible for their impact on the environment and keep organizations on track towards reaching climate action goals. This report describes findings from three distinct inventories (Figure 10): • A community inventory that estimates GHG emissions produced by activities of the Bainbridge Island community, including residents and businesses. The community GHG inventory includes emissions due to energy used to power and heat homes and businesses; fuel used by vehicles running within Bainbridge Island; solid waste that is generated by the community; agriculture activities; fuel use by off-road equipment and vehicles; and others. • A municipal inventory that accounts for the GHG emissions resulting from City of Bainbridge Island government operations. This inventory can help the City understand GHG emissions stemming from various activities associated with municipal operations, including from municipal building and facility operation, transportation, solid waste, wastewater, and refrigerant leakage. • A consumption-based inventory that estimates GHG emissions associated with the consumption of food, goods, and services within the community, regardless of their origin. For example, this inventory would not include emissions from the production of locally-manufactured goods that are consumed entirely outside the community; however, it would include emissions associated with the production of goods manufactured in another community but consumed by Bainbridge Island residents, visitors, or businesses. This inventory can be examined in context with the community inventory to paint a more comprehensive picture of community emissions. This report also presents findings from additional analyses:  Municipal and communitywide contribution analyses that identify key drivers of observed emission trends. For example, analysis calculates the impact that a hotter summer or colder winter may have had on household energy use, and thus, emissions.  A supplemental carbon sequestration analysis that estimates the amount of carbon dioxide that Bainbridge Island trees absorb—or sequester—from the atmosphere on an annual basis (in Appendix). CITY OF BAINBRIDGE ISLAND | 2019 | 12 Figure 7. Conceptual relationship among community, government, and consumption-based inventories for Bainbridge Island. GREENHOUSE GAS INVENTORY OVERVIEW Inventory Methodology The GHG inventories summarized in this report account for human-caused emissions of the most prominent and typical greenhouse gases for communities: carbon dioxide (CO2), methane (CH4), and nitrous oxide (N2O). To account for the differences in potency among these gases, all emissions are calculated and reported in units of metric tons of carbon dioxide equivalent (MTCO2e). The inventories were conducted using widely accepted tools and protocols, including The Climate Registry’s Local Government Operations Protocol, ICLEI’s U.S. Community Protocol for Accounting and Reporting of Greenhous Gas Emissions, and U.C. Berkeley’s CoolClimate Calculator. GHG emissions are inventoried by multiplying annual activity data (e.g. electricity consumption) by emission factors (greenhouse gas emissions produced per kWh of electricity). For the municipal operations inventory, we obtained most data from City staff and local documentation such as Puget Sound Energy's (PSE) billing records, records from local wastewater treatment facilities, and fleet vehicle fuel use records. For the communitywide inventory, key data sources included electricity consumption data from PSE, Puget Sound Regional Council (PSRC) modeling outputs for vehicle miles traveled by fuel type, Bainbridge Disposal tonnage records, and Kitsap County Conservation District estimates of agricultural landowners and acreage. Where local data were not available, we referenced and downscaled national data sources such as from the Federal Transit Administration, U.S. Census Bureau, and the U.S. Department of Agriculture. CITY OF BAINBRIDGE ISLAND | 2019 | 13 Inventory Boundaries The activities and sectors included in GHG inventories are often classified into three “scopes,” which represent relative levels of control over an emissions source (see Figure 8 and Figure 9):  Scope 1 emission sources include those directly caused by an organization’s actions, such as from owned equipment and facilities.4  Scope 2 emissions are those indirectly associated with purchased electricity, steam, heating, or cooling.  Scope 3 includes all other indirect emissions that are not covered in Scope 2. The communitywide and municipal inventories for Bainbridge Island included emissions sources from all three scopes: Scope 1, Scope 2, and Scope 3. The inventories included all sources required by the consulted protocols and additional sources, as relevant. Figure 8. Sources and boundaries of municipal GHG emissions.5 4 Except direct carbon dioxide emissions from biogenic sources. 5 The Climate Registry. (2010). Local Government Operations Protocol: For the quantification and reporting of greenhouse gas emissions inventories. CITY OF BAINBRIDGE ISLAND | 2019 | 14 Figure 9. Sources and boundaries of community GHG emissions.6 6 World Resource Institute (2014). Greenhouse Gas Protocol: Global protocol for community-scale greenhouse gas emission inventories. CITY OF BAINBRIDGE ISLAND | 2019 | 15 The tables below summarize sectors included in the municipal and communitywide inventories. Table 1. Community inventory emissions sources and scope categories. Emissions Type Required? Scope 1 Scope 2 Scope 3 Residential Energy Electricity   Propane   Fuel Oil   Commercial Energy Electricity   Propane   Industrial Energy Electricity   Propane  Transportation On-Road Passenger Vehicles   On-Road Freight Vehicles   On-Road Transit Vehicles  Off-Road Vehicles and Equipment  Air Travel  Ferry Travel  Solid Waste, Potable Water, and Wastewater Solid Waste   Potable Water Use Energy*   Wastewater Treatment   Refrigerant Leakage   Agriculture  * Potable water use energy—energy associated with treating and distributing potable water systems on the Island (e.g., from pumping stations)—is included in the non-residential energy consumption sector. Energy used for pumping individual wells is included in the residential energy consumption sector. CITY OF BAINBRIDGE ISLAND | 2019 | 16 Table 2. Municipal inventory emissions sources and scope categories. Emissions Type Required? Scope 1 Scope 2 Scope 3 Buildings and Facility Energy Electricity   Propane   Streetlights and Traffic Signals   Transportation On-Road Fleet Vehicles   Off-Road Vehicles   Employee Commute  Business Travel  Solid Waste, Potable Water, and Wastewater Solid Waste   Potable Water Use Energy   Wastewater Treatment   Refrigerant Leakage   Table 3. Consumption-based inventory emissions sources and scope categories. Emissions Type Required? Scope 1 Scope 2 Scope 3 Consumption-Based Emissions Household Consumption  City Government Consumption  Upstream Energy  Inventory Years To provide Bainbridge Island with the most up-to-date information on their GHG emissions, we selected the most recent year available as an inventory year: 2018. For a comparison year, we selected 2014 because no major changes in organizational structure or infrastructure occurred after that year. For example, the wastewater treatment facility update was completed before 2014, so those changes would be reflected in the 2014 inventory, enabling a more apples-to-apples comparison with 2018. Other factors in that deciding the inventory years include:  Data availability: The availability of City data is scarcer further in the past.  Data consistency: Some data were either modeled differently or organized differently in the past (i.e., methodologies changed over the last decade). Having a more recent baseline better ensures more of accurate comparison over time.  Relevance: Choosing a more recent comparison inventory year allows for a more productive assessment of trends over time to inform climate action planning. For example, if the baseline year of 1990 was chosen it would be difficult to understand distinct drivers of changes between 1990 and 2019 due to the many interim changes that occurred within that time frame. CITY OF BAINBRIDGE ISLAND | 2019 | 17 ROADMAP OF THIS REPORT This report is organized into the following sections to assess GHG emissions associated with Bainbridge Island:  The Communitywide Emissions section presents methodologies and results for community-based inventory.  The Communitywide Contribution Analysis section explores the drivers of communitywide emission trends.  The Municipal Operation Emissions section presents methodologies and results for municipal-based inventory.  The Municipal Contribution Analysis section that explores drivers of municipal operation emission trends.  The Consumption-Based Emissions section presents methodologies and results from the consumption-based inventory. Each inventory section is organized as follows:  Results section presents GHG inventory results for the study years. The results are quantified and, where relevant, summarized for trends between the two inventory years.  Data Sources and Methodology section details the sources of inventory information by sector, with calculation methodologies, where relevant. The section also describes any data limitations or considerations. The report concludes with a summary of the major trends across all the inventories and future considerations for climate action planning. CITY OF BAINBRIDGE ISLAND | 2019 | 18 Communitywide Emissions The communitywide GHG emissions inventory estimates GHG emissions from individuals, businesses, and commercial and industrial processes within Bainbridge Island. It also includes emissions associated with some residential and business activities that occur outside of Bainbridge Island, such as from electricity generation and solid waste transport to an out-of-state landfill. Overall communitywide emissions are provided in units of metric tons of carbon dioxide equivalent (MTCO2e). Table 5 below outlines the various emissions sources included in the communitywide inventory, along with their corresponding scope. The table also shows which sectors are required by the U.S. Community Protocol. Table 4. Community inventory emissions and scope categories. Emissions Type Required? Scope 1 Scope 2 Scope 3 Residential Energy Electricity   Propane   Fuel Oil   Commercial Energy Electricity   Propane   Industrial Energy Electricity   Propane  Transportation On-Road Passenger Vehicles   On-Road Freight Vehicles   On-Road Transit Vehicles  Off-road Vehicles and Equipment  Air travel  Ferry  Solid Waste, Potable Water, and Wastewater Solid Waste   Potable Water Use Energy*   Wastewater Treatment   Other Process & Fugitive Emissions   Agriculture  * Potable water use energy—energy associated with treating and distributing potable water systems on the Island (e.g., from pumping stations)—is included in the non-residential energy consumption sector. Energy used for pumping individual wells is included in the residential energy consumption sector. CITY OF BAINBRIDGE ISLAND | 2019 | 19 RESULTS Overview Bainbridge Island’s communitywide emissions in 2018 totaled 233,998 MTCO2e—equivalent to 9.4 MTCO2e per resident. Residential energy (37%) was responsible for the greatest proportion of communitywide emissions, followed by non-residential energy (14%), air travel (13%), and on-road vehicles (12%) (see Figure 11). Figure 10. Bainbridge Island community emissions by source, 2018 (total = 233,998 MTCO2e) Communitywide GHG emissions increased by 9% from 2014 to 2018 (see Figure 12, Table 6, and Figure 13). These increases are primarily due to changes in electricity fuel sources (e.g., proportion of coal in the utility fuel mix) and growth in population and employment. Improvements in vehicle fuel economy, reductions in the distance each person drives, and declining per-household and per-business energy consumption all reduced the extent of this increase. Although communitywide emissions have increased by 9%, per capita emissions have increased only slightly—by 1% (see Figure 12). CITY OF BAINBRIDGE ISLAND | 2019 | 20 Figure 11. Bainbridge Island community emissions, by year and source and resident. Major sources are labeled. Table 5. Communitywide emissions, by sector. GHG Emissions by Sector (MTCO2e) 2014 2018 Change % Change Residential Energy 84,885 94,810 9,925 12% Electricity 75,363 85,898 10,535 14% Propane & Fuel Oil (Residential, Commercial, and Industrial) 5,912 5,274 -638 -11% Losses from transmission & distribution 3,610 3,638 28 1% Non-Residential Energy 31,162 33,277 2,115 7% Electricity 29,738 31,925 2,187 7% Losses from transmission & distribution 1,424 1,352 -73 -5% Transportation 75,315 80,778 5,463 7% On-Road Passenger & Freight Vehicles 27,448 27,330 -118 0% On-Road Transit Vehicles 590 781 191 32% Air Travel 24,023 31,002 6,979 29% Ferry Travel 14,051 11,334 -2,716 -19% Other Off-Road Vehicles and Equipment 9,204 10,331 1,127 12% Solid Waste & Wastewater Treatment 10,503 11,470 968 9% Solid Waste 8,369 9,289 920 11% Wastewater Treatment 48 51 3 7% Septic Tanks 2,086 2,131 45 2% Other Process & Fugitive Emissions 12,209 13,332 1,123 9% Agriculture 351 331 -20 -6% Enteric Fermentation 319 297 -22 -7% Manure Management 33 34 2 5% TOTAL 214,425 233,998 19,574 9% CITY OF BAINBRIDGE ISLAND | 2019 | 21 Figure 12. Bainbridge Island community emissions by source and scope, 2018. CITY OF BAINBRIDGE ISLAND | 2019 | 22 Residential Energy Emissions from residential use of electricity, propane7, and fuel oil accounted for 41% (94,810 MTCO2e) of total communitywide emissions in 2018, a 12% increase from 2014. Residential electricity use alone makes up 37% of communitywide emissions; it also exceeds non-residential electricity use (see Figure 14). Figure 13. Community building energy use, by sector and year. Electricity Emissions from residential use of electricity increased from 75,363 MTCO2e in 2014 to 85,898 MTCO2e in 2018, a 14% change. One contributing factor is that the consumption of electricity increased slightly from 184,865,169 kWh in 2014 to 189,165,563 kWh in 2018, an increase of 2%. Additionally, Bainbridge Island’s population increased 8% from 23,135 to 24,891. Changes in the electricity fuel mix also contributed to increases in residential electricity emissions (see Table 7). Puget Sound Energy (PSE) provides power to Bainbridge Island. Approximately two-thirds of the power is from coal and hydroelectric generation. In 2017, the fuel mix was 38% coal, 33% hydroelectric, 21% natural gas, 6% wind, and 2% nuclear and other sources.8 In 2014, the fuel mix was 35% coal, 36% hydroelectric, 20% natural gas, 3% wind, and 5% nuclear and other sources. Between 2014 and 2018, Bainbridge Island residents significantly increased their participation in PSE’s Green Power Program. The 7 Propane amounts were derived from records that do not distinguish between residential, commercial, and industrial use. Therefore, all Bainbridge Island propane usage—regardless of sector—is included within the residential summary of emissions. 8 Puget Sound Energy. Electricity Supply. www.pse.com/pages/energy-supply/electric-supply (accessed June 27, 2019). The emissions inventory for 2017 was used because the 2018 inventory was not yet available during the time of study. CITY OF BAINBRIDGE ISLAND | 2019 | 23 2018 10% participation rate is among the highest in PSE’s service area. However, due to the complexities of the electricity grid, the U.S. Community Protocol discourages consideration of renewable energy participation in calculating communitywide GHG inventories (see text box). Table 6. Puget Sound Energy (PSE) electricity generation fuel mix for 2014 and 2017.9 Electricity Generation Fuel Type 2014 2017 Biomass 0% 0% Coal 35% 38% Cogeneration 4% 0% Geothermal 0% 0% Hydro 36% 33% Landfill Gas 0% 0% Natural Gas 20% 21% Nuclear 1% 1% Other 0% 0% Petroleum 0% 0% Solar 0% 0% Waste 0% 0% Wind 3% 6% 9 Puget Sound Energy. Electricity Supply. www.pse.com/pages/energy-supply/electric-supply (accessed June 27, 2019). The emissions inventory for 2017 was used because the 2018 inventory was not yet available during the time of study. Green Power Purchases Puget Sound Energy allows electricity users to enroll in a Green Power purchasing program to support renewable energy generation in the region. When utility customers enroll in this program, they are not paying for 100% renewable energy to be delivered directly to their home or business, but rather for their utility to buy a certain amount of renewable energy as part of the utility’s overall fuel mix. Because utility fuel mixes are used to calculate the GHG emissions of consumed electricity in a community, the renewable energy purchases made through Green Power programs are already included in the emissions calculations of the community GHG inventory. Therefore, accounting for Green Power purchases separately within a GHG inventory risks double-counting the benefits of a renewable energy system. CITY OF BAINBRIDGE ISLAND | 2019 | 24 Propane and Fuel Oil Due to a lack in locally available usage data, consumption of residential propane and fuel oil were estimated using data from local propane sales tax revenues, the U.S. Energy Information Administration (EIA), and 2017 American Community Surveys (ACS). These estimations suggest that emissions from residential use of propane10 and fuel oil totaled 5,912 MTCO2e in 2014 and 5,274 MTCO2e in 2018, an 11% decrease in four years. This decline is primarily due to a decrease in the number of households using those fuel sources, including for cooking and heating. Electricity Transmission and Distribution Losses As electricity travels from where it was generated, such as a hydroelectric dam or wind farm, to individual homes, businesses, and other buildings, some of the electricity is naturally lost over power lines and in substations and transformers. The proportion of the total electricity initially generated that is lost through these processes in a given year and region is known as the “grid loss” factor. This grid loss proportion can change from year-to-year based on a number of factors, including weather, transmission distance, power line size, and transmission voltage. Although lost electricity is not ultimately delivered to the consumer, it still results in the emission of greenhouse gases. We estimate that emissions from transmitting and distributing residential electricity to Bainbridge Island residents increased slightly from 3,610 MTCO2e in 2014 to 3,638 MTCO2e in 2018—a less than 1% change. These changes are a result of changes to the amount of electricity consumed and the “grid loss” factor of PSE infrastructure. Non-Residential Energy The non-residential energy sector includes greenhouse gas emissions from commercial and industrial electricity; losses due to the transmission and distribution of that electricity; and energy associated with treating and distributing potable water (e.g., from pumping stations). Commercial and industrial propane consumption emissions are included within the residential emissions sector, as the source data did not distinguish between different use sectors. Electricity and Transmission and Distribution Losses Emissions from commercial and industrial electricity accounted for 14% (31,925 MTCO2e) of total communitywide emissions in 2018, making it the second-largest source along with air travel. Emissions from commercial and industrial use of electricity increased from 29,738 MTCO2e in 2014 to 31,925 MTCO2e in 2018, a 7% change. The increase in emissions can be partially attributed to the increased workforce from 8,224 to 8,943 employees, as well as from changes in the electricity fuel mix(see Figure 14 on page 23) Emissions from non-residential electricity transmission and distribution decreased approximately 5%, from 1,424 MTCO2e in 2014 to 1,352 MTCO2e in 2018. These changes are a result of changes in the amount of electricity consumed and the “grid loss factor” associated with PSE infrastructure. 10 Propane amounts were derived from records that do not distinguish between residential, commercial, and industrial use. Therefore, all Bainbridge Island propane usage—regardless of sector—is included within the residential summary of emissions. CITY OF BAINBRIDGE ISLAND | 2019 | 25 Transportation Transportation contributed over a third of Bainbridge Island community emissions in 2018, making it the second-largest contributor to the community’s greenhouse gas emissions. In 2014, most transportation emissions came from passenger and freight vehicles, followed by air travel. However, in 2018, this trend shifted, with most transportation emissions coming from air travel followed by passenger and freight vehicles (Figure 15). The rapid growth of Sea-Tac Airport explains this shift: there was a 30% increase in landings at SeaTac International Airport between 2014 and 2018.11 Figure 14. Community transportation emissions in 2018 (total = 80,778 MTCO2e). On-Road Vehicles Most on-road vehicle emissions in Bainbridge Island stem from passenger cars. Downscaled modeling by the Puget Sound Regional Council suggests that passenger vehicles traveled over 69 million miles in Bainbridge Island in 2018 (Table 8 and Figure 16). Table 7. Annual vehicle miles estimated for Bainbridge Island. Type 2014 2018 Change Passenger vehicle 66,612,870 69,543,904 4% Medium truck 3,011,067 2,993,763 -1% Heavy Truck 708,280 708,118 0% Transit bus* 394,450 397,634 1% *Miles for transit buses are in annual revenue miles. 11 Air travel emissions from Bainbridge residents were attributed by downscaling total fuel use at Sea-Tac Airport to the Bainbridge Island community based on population. More information can be found in the “Data Sources and Methodology” section of this report. CITY OF BAINBRIDGE ISLAND | 2019 | 26 Figure 15. Distribution of on-road vehicle miles traveled, by vehicle type. Transportation data are typically recorded as vehicle miles traveled, or VMT. VMT is the total number of miles driven by motorized vehicles of specific types. Although the model used to estimate GHG emissions by VMT and vehicle type does not separately note hybrids and electric vehicles, these vehicles are subject to the same emissions standards as conventional gasoline or diesel vehicles. As such, hybrid and electric vehicle VMT are incorporated into the average emissions factors used for each vehicle type and model year, despite not being reported separately from general passenger vehicles. Despite a 4% increase in VMT for passenger vehicles between 2014 and 2018, emissions from on-road vehicles increased minimally—by 0.3%. The increasing average fuel economy of passenger vehicles accounts for this minimal change (Figure 16). We estimate that transit vehicle emissions increased 32% between 2014 and 2018, rising from an estimated 590 MTCO2e in 2014 to 781 MTCO2e in 2018. As reported by Kitsap Transit to the Federal Transit Administration, annual revenue miles remained relatively constant between the two inventory years—only increasing 1%. This trend suggests that increased transit emissions may be the result of incomplete reporting, decreased fuel economy, and/or increased use of high emitting fuels. The report indicates that Kitsap Transit activities in 2018 included more gasoline use for vanpool vehicles and a shift to liquified petroleum gas for paratransit (demand response) vehicles. Off-Road Vehicles & Equipment This inventory categorizes off-road vehicles and equipment into three categories: ferry travel, air travel, and other off-road vehicles and equipment. Other off-road vehicles and equipment includes emissions from vehicles and equipment used for agriculture, construction, lawn/gardening, and recreation such as boating. CITY OF BAINBRIDGE ISLAND | 2019 | 27 Ferry Travel Emissions from ferry travel decreased from 14,051 MTCO2e in 2014 to 11,334 MTCO2e in 2018, an overall decrease of 19%. This decrease is a result of a large reduction in fuel consumption by Washington State ferries traveling the Seattle-Bainbridge Island route. In 2014, we estimate that Bainbridge Island’s share of ferry travel accounted for the consumption of 2.7 million gallons of fuel, compared to 2.2 million gallons in 2018. Air Travel We estimate that air travel contributed 13% to communitywide emissions, making it the largest source of off-road GHG emissions and second-largest source of emissions overall (tied with non-residential energy). Emissions from air travel increased 29%, from an estimated 24,023 MTCO2e in 2014 to 31,002 MTCO2e in 2018. This increase is largely attributable to population growth on Bainbridge Island and a 30% increase in total landings at SeaTac International Airport between 2014 and 2018. Our calculations are based on available jet fuel usage data (from 2015), which were scaled by the proportion of landings at SeaTac International Airport. Bainbridge Island’s population and businesses make up approximately 0.5% of the four-county area that SeaTac predominantly serves (King, Kitsap, Pierce, and Snohomish counties). Other Off-Road Vehicles & Equipment We estimate that emissions from other off-road equipment accounted for 4% of overall emissions in Bainbridge Island in 2018 (10,331 MTCO2e). Most of these emissions—which are estimated by scaling county-level model outputs to Bainbridge Island based on population—are generated by construction equipment, lawn and garden equipment, and pleasure watercraft. Solid Waste and Wastewater Solid Waste Most emissions from the disposal of solid waste are associated with the release of methane from decomposing waste in the landfill (see Figure 17 and Table 9). Emissions from waste disposal increased from 5,527 MTCO2e in 2014 to 6,134 MTCO2e in 2018, an 11% change that is largely due to population growth and associated increases in tons disposed. Based on waste characterization studies from Kitsap County, King County, and Washington State, the composition of waste over time remained relatively constant. Approximately two-thirds of disposed waste is mixed solid waste, and about one-quarter is food scraps. All of Bainbridge Island’s waste is transported off-Island, first in trucks to the Olympic View Landfill in Bremerton then via train to the Columbia Landfill near Arlington, Oregon. We estimate that emissions from waste transportation increased from 2,352 MTCO2e to 2,611 MTCO2e. This is associated with the rise in waste collected and transported, which grew from an estimated 22,616 tons in 2014 to 25,102 tons in 2018. Emissions from equipment used at the landfill to process waste and from yard waste composting make up a nominal proportion of the Island’s solid waste emissions. We estimate that the diesel-powered CITY OF BAINBRIDGE ISLAND | 2019 | 28 equipment at Columbia Landfill generated 371 MTCO2e and 412 MTCO2e in 2014 and 2018, respectively. In 2018, Bainbridge Disposal reported 1,900 tons of yard waste were composted, which generated 132 MTCO2e. Figure 16. Solid waste emissions in 2018, by source. Table 8. Summary of solid waste emissions, by process. Process Emissions (MTCO2e) 2014 2018 Difference Solid Waste Disposal & Decomposition 5,527 6,134 +607 Yard Waste Composting 119 132 +13 Collection & Transportation to Landfill 2,352 2,611 +258 Processing with Equipment at Landfill 371 412 +41 Total 8,369 9,289 +920 Process Emissions by Weight (MTCO2e/ton waste) 2014 2018 Difference Solid Waste Disposal & Decomposition 0.244 0.244 — Yard Waste Composting 0.070 0.070 — Collection & Transportation to Landfill 0.104 0.104 — Processing with Equipment at Landfill 0.016 0.016 — Total 0.344 0.344 — CITY OF BAINBRIDGE ISLAND | 2019 | 29 Wastewater Treatment The combined emissions from wastewater treatment (process and effluent emissions) and fugitive emissions12 from septic sources contributed to 1% of communitywide emissions in 2018. This small proportion is associated with an overall 2% increase in emissions from these sources, from 2,134 MTCO2e in 2014 to 2,182 MTCO2e in 2018 (see Table 10). This increase is due to population growth and the associated increase in waste generated and processed (calculations for emissions from septic tanks were scaled based on Bainbridge Island population trends and calculating emissions from WWTP also accounted for population growth). Table 9. Emissions from wastewater treatment and septic tanks (MTCO2e). Source Emissions (MTCO2e) 2014 2018 Difference % change WWTP Process 13 14 +1 +7% WWTP Effluent 35 37 +2 +7% Septic Tanks 2,086 2,131 +45 +2% Total 2,134 2,182 +48 +2% Other Process & Fugitive Emissions Emissions from other processes and fugitive sources—including hydrofluorocarbons (HFCs), perfluorocarbons (PFCs), and sulfur hexafluoride (SF6)–contributed 6% of Bainbridge Island’s communitywide emissions in 2018.13 Emissions from HFCs—a common refrigerant used in refrigeration and air conditioning—comprised nearly all of these emissions (99%; 13,234 MTCO2e). We estimate that emissions from HFCs and PFCs increased 10% between 2014 and 2018, from 12,061 MTCO2e to 13,237 MTCO2e. This increase is driven by population growth, as emissions were scaled down by population from national-level emissions due to lack of locally available data. Fugitive emissions from leaks of SF6 in electricity transmission and distribution operating equipment declined an estimated 36%, from 147 MTCO2e in 2014 to 95 MTCO2e in 2018. Our calculations are based on data reported by PSE in their 2014 and 2017 GHG inventories, scaled down to the 1% of customers Bainbridge Island represents within PSE’s service area. Data on process and fugitive sources such as HFC, PFC, and SF6 emissions at city scales are scarce. The U.S. EPA’s 2019 Inventory of GHG Sources and Sinks was used as a proxy and the data was scaled to Bainbridge Island’s population. Estimating emissions from process and fugitive sources can be highly uncertain. Factors vary between individual pieces of equipment. Even if the leak rate of a piece of equipment has been tracked carefully, that leak rate can change after the leak is repaired or as the 12 Fugitive emissions are from leaks. 13 Fugitive emissions are those that are not physically controlled but result from the intentional or unintentional release of GHGs. They commonly arise from the production, processing, transmission, storage and use of fuels or other substances, often through joints, seals, packing, and gaskets. Examples include HFCs from refrigeration leaks, SF6 from electrical power distributors, and CH4 from solid waste landfills. CITY OF BAINBRIDGE ISLAND | 2019 | 30 equipment ages. For example, equipment can leak for two or more years before requiring a recharge, so emissions over this timeframe are typically not detected until after they occur. Table 10. Summary of emissions from other processes and fugitive sources. GHG Emissions (MTCO2e) 2014 2018 Difference % change HFC 12,058 13,234 +1,176 +10% PFC 3 3 0 0% SF6 147 95 -53 -36% Total 12,209 13,332 +1,123 +9% Agriculture The Kitsap County agricultural census reported 11,483 and 12,539 livestock animals in 2012 and 2017, respectively. To determine what proportion of livestock animals were on Bainbridge Island during these years, we scaled agricultural census data for Kitsap County to Bainbridge Island based on land area. Using this methodology, we estimate that Bainbridge Island had 803 and 875 livestock animals in 2014 and 2018; representing a 9% increase in animals between the two years. While the total animal population increased on the Island, we estimate that the number of cattle decreased by 12%. This decline drove estimated reductions in enteric fermentation emissions—the release of methane (CH4) as livestock digest food—of 7%, from 319 MTCO2e in 2014 to 297 MTCO2e in 2018. Emissions from manure management increased 5%, from 33 MTCO2e in 2014 to 34 MTCO2e in 2018, driven by the overall increase in livestock animal population. Table 11. Agricultural acres and population trends, by livestock. Animal Population 2014 2018 % change Horses 98 100 +2% Sheep 41 82 +100% Goats 35 44 +26% Cattle 92 81 -12% Poultry 517 538 +4% Swine 20 30 +50% Total 803 875 +9% CITY OF BAINBRIDGE ISLAND | 2019 | 31 DATA SOURCES AND METHODOLOGY We followed the U.S. Community Protocol (Version 1.1) to complete the communitywide emissions inventory for 2014 and 2018, using the U.S. EPA’s GHG Inventory Sources and Sinks Annexes for 2014 and 2019 and the IPCC’s 5th Assessment Report (AR5) for updated GWPs and emissions factors where available. All emissions were calculated in ClearPath and were based on verified local data when they were available. When local data were not available, data inputs were scaled down from Kitsap County, King County, or national datasets. Most data were deemed medium quality due to assumptions about local data. Completing this inventory involved acquiring the following data, summarized in Table 13 and detailed in the followed sections:  Activity data that quantifies levels of activity that generate greenhouse gas emissions, such as vehicle miles traveled, and tons of waste generated.  Emission factors that translate activity levels into emissions. Data quality is assessed and reported on a High (H), Medium (M), and Low (L) scale in accordance with GHG inventory best practices:  A High rating indicates data are detailed and specific to the local geography  A Medium rating indicates data are more general or modeled with robust assumptions and may not be specific to the local geography, but are downscaled from a slightly broader geography (e.g., state- level)  A Low rating indicates data are highly modeled, uncertain, or a default value was used based on national characteristics. Table 12. Key data sources for the City of Bainbridge Island's communitywide inventory. Sector Activity Data (AD) AD Quality Emissions Factors (EF) EF Quality Residential Energy Electricity • Overall kWh consumption from the three substations serving the Island • Residential proportion based on sample of customer data in 2014 and 2018 H • PSE reported emissions factors (PSE Greenhouse Gas Inventory, 2014 & 2017, Table 7.1) H Stationary Fuel Combustion • Local propane sales tax information for Bainbridge Island (2014 and 2018) • EIA’s average propane price per gallon of fuel for 2014 and 2018. M • Emissions factors provided in ClearPath program M Electricity Transmission & Distribution Losses • See “Electricity” • eGRID Western region grid gross loss percentage (2014 and 2016) M • PSE-reported emissions factors (PSE Greenhouse Gas Inventory, 2014 & 2017, Table 7.1) H CITY OF BAINBRIDGE ISLAND | 2019 | 32 Sector Activity Data (AD) AD Quality Emissions Factors (EF) EF Quality Non-Residential Energy Electricity • kWh consumption from the three substations serving the Island • Non-residential proportion based on sample of customer data in 2014 and 2018 H • PSE reported emissions factors (PSE Greenhouse Gas Inventory, 2014 & 2017, Table 7.1) H Electricity Transmission & Distribution Losses • See “Electricity” • eGRID Western region grid gross loss percentage (2014 and 2016) M • PSE reported emissions factors (PSE Greenhouse Gas Inventory, 2014 & 2017, Table 7.1) H Transportation On-Road Vehicles • Annual miles traveled, by vehicle and fuel type, provided by EPA MOVES analysis of travel model data, for proportion of Kitsap County VMT occurring in Bainbridge Island • Annual revenue miles traveled and fuel consumption by type provided by National Transit Database for Kitsap County, then scaled down to Bainbridge Island by population M • CO2, CH4, and N2O emission factors are defaults from the U.S. Community Protocol L Off-Road Vehicles & Equipment • Emissions by off-road sector and fuel type, along with days of use per year, provided by nonroad module of the EPA MOVES model, for proportion of Kitsap County activity occurring in Bainbridge Island M • CO2, CH4, and N2O emission factors are defaults from the U.S EPA MOVES nonroad model L Air travel • 2015 Jet fuel consumption from Port of Seattle, scaled to target years by total landings and to the Island by population/employment M • CO2, CH4, and N2O emission factor defaults from ClearPath H Ferry travel • Ferry fuel price, from EIA, and cost, from WSDOT M • CO2, CH4, and N2O emission factor defaults from ClearPath H Solid Waste and Wastewater Solid Waste Generation • Single-family waste tonnages from Bainbridge Disposal • Multi-family + commercial tonnages estimated from ratio of single-family to multi-family + commercial from 2015-2016 WA Waste Characterization Study, for Kitsap County • Landfill methane collection scenario confirmed “typical” by Columbia Landfill • Annual precipitation from National Weather Service, Arlington, OR M • Waste characterizations from 2015-2016 WA Waste Characterization Study, for Kitsap County • Waste characterization was comparable across Port Orchard samples and the Puget Sound-wide data from the aforementioned study; there was also little difference compared to King County compositions, which were assessed as a reference M CITY OF BAINBRIDGE ISLAND | 2019 | 33 Sector Activity Data (AD) AD Quality Emissions Factors (EF) EF Quality Solid Waste Collection & Transportation • Mass of solid waste from Bainbridge Disposal, with multi-family + commercial estimated from Kitsap County ratio of single-family to multi-family + commercial (see above) • Mileage from Google maps M • Default EF from ClearPath L Yard Waste Composting • Yard waste tonnage from Bainbridge Disposal for 2018; extrapolated for 2014 using 2014:2018 single-family solid waste ratio M • Default EF from ClearPath L Landfill Processing • Mass of solid waste from Bainbridge Disposal, with multi-family + commercial estimated from Kitsap County ratio of single-family to multi-family + commercial (see above) • Fuel type for equipment confirmed by Columbia Landfill M • Default EF from ClearPath L Wastewater Treatment Facility • Wastewater treatment facility processes determined based on desktop research and confirmed with City staff • Population served by wastewater treatment facility estimated by multiplying the number of sewer connections by the average household size from ACS M • Default EF from ClearPath • CH4 emissions were not applicable since the system is aerobic L Septic Tanks • Population data estimated based on number of septic systems reported to the Department of Health and the number of sewer connections in the community. M • Default EF from ClearPath • Fugitive emissions calculated using Community Protocol WW.11 Alternative Method for Methane Emissions from Septic Systems if Only the Population is Known L Refrigerant Leakage Refrigerant Leakage from Building Heating and Cooling Equipment, and Fugitive Emissions from Electricity Generation & Transmission • HFC and PFC tonnage from Annex 6 in U.S. EPA GHG Inventory Sources and Sinks, scaled down to Bainbridge Island population • SF6 tonnage and customer data from PSE, scaled down to Bainbridge Island customers L • HFC and PFC GWP from IPCC 5th Assessment Report, with adjustment for climate- carbon feedback • SF6 100-Year GWP from U.S. EPA GHG Sources and Sinks • Emissions calculated using Community Protocol Alternative Method BE.7.1.A H CITY OF BAINBRIDGE ISLAND | 2019 | 34 Sector Activity Data (AD) AD Quality Emissions Factors (EF) EF Quality Agriculture Enteric Fermentation and Manure Management • Animal population and pastureland from USDA Agricultural Census, for Kitsap County • Agricultural acres with livestock on Bainbridge Island and waste management system, from Kitsap County Conservation District • Typical animal mass and Bo from U.S. EPA GHG Inventory Sources & Sinks, 2017 Annexes • Volatile solids and excreted nitrogen from U.S. EPA GHG Inventory Sources & Sinks, 2014 and 2017 Annexes • Methane conversion factors from Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2017, with annual average temperature from Climate Bremerton M • EF, 100-year GWP for CH4 and N2O, and volatilization nitrogen loss from U.S. EPA GHG Inventory Sources & Sinks, 2017 Annexes • EF for nitrogen volatilization and runoff/leaching from U.S. Community Protocol, Appendix G • Indirect N2O emissions calculated using U.S Community Protocol, Appendix G M The U.S. Community & Local Government Operations Protocols The U.S. Community and Local Government Operations Protocols were built to provide easily applicable and accurate community-level and municipal estimates of GHG emissions. These protocols provide a consistent framework in which to compare emissions from a geographic or operational boundary across time. The U.S. Community Protocol was designed for community-scale GHG accounting, making it a valuable tool for counties and cities, and an appropriate choice for the City of Bainbridge Island. The U.S. Community and Local Government Operations Protocols are widely used, understood, and respected. These community and municipal inventories follow the U.S. Community and Local Government Operations Protocol methodologies and deviate from their stated methods only when more precise, local data are available, per Protocol recommendations. The Local Governments for Sustainability (ICLEI) created the U.S. Community Protocol in 2013 and the Local Government Operations Protocol in 2010. The U.S. Community Protocol requires, at a minimum, reporting of the following five activities: 1) Use of electricity by the community 2) Use of fuel in residential and commercial stationary combustion equipment 3) On-road passenger and freight motor vehicle travel 4) Use of energy in potable water and wastewater treatment and distribution 5) Generation of solid waste by the community. These activities are required because they represent the largest sources of GHG emissions for most communities and are activities that can be impacted by local government actions. The Local Government Operations Protocol includes similar reporting requirements with a focus on only those emission sources that are within the government’s control and are the result of day-to-day operations. CITY OF BAINBRIDGE ISLAND | 2019 | 35 Residential Energy  Data on electricity use were provided by the utility serving Bainbridge Island, PSE, and for transmission and distribution losses, by eGRID.  Stationary fuel combustion consisted of residential fuel oil, with data from EIA and the number of households from ACS.  Propane usage for residential, commercial, and industrial sectors was estimated using local sales tax data and EIA information on average propane fuel prices per gallon. Non-Residential Energy  Data on electricity use were provided by the utility serving Bainbridge Island, PSE, and for transmission and distribution losses, by eGRID. PSE does not track commercial and industrial energy separately, hence they were combined in this analysis. Transportation  Vehicle miles traveled were derived from Puget Sound Regional Council’s MOVES analysis of travel model data and account for all mileage within the Kitsap County boundary regardless of trip origin or destination. Emissions for Bainbridge Island are assumed proportional to the share of Kitsap County VMT occurring in Bainbridge Island city limits. Share of average weekday VMT occurring in Bainbridge Island was calculated from 2014 travel model data and includes all vehicle types. Bainbridge Island VMT shares represents all trips to/from Bainbridge Island as well as those driving through the city boundaries. Year 2018 was extrapolated from the rates between 2014 and 2016. The PSRC data reported VMT and emissions for passenger vehicles, medium trucks, heavy trucks, and transit. In ClearPath, we reported medium trucks as light trucks.  Annual revenue miles and fuel consumption data for Kitsap County were acquired from the National Transit Database from the Federal Transit Administration. These data were scaled down to Bainbridge Island by population.  The Port of Seattle supplied jet fuel consumption data for SeaTac Airport for 2015, of which 0.5-1% was attributable to Bainbridge Island. SeaTac passenger enplaning surveys and population and employment statistics were used to assign the above fraction of the total airshed emissions to the Bainbridge Island community. This allocation methodology means that communities with more residents and business travelers are assigned a greater proportion of travel-related emissions at SeaTac airport. We scaled 2015 jet fuel usage to 2014 and 2018 based on the total landings in those years.  Ferry fuel consumption was derived from EIA data on diesel fuel price and from WSDOT data on the reported fuel cost for the Bainbridge Island route in fiscal year 2015.  Off-road emissions were calculated from the nonroad module of the EPA MOVES model for 2014 and 2018. The model estimates emissions at the county level, which were then scaled down to Bainbridge Island by population. Emissions include common non-road equipment, including construction, agriculture, lawn/gardening, and recreational equipment. CITY OF BAINBRIDGE ISLAND | 2019 | 36 Solid Waste and Wastewater  Bainbridge Disposal single-family tonnages and Kitsap County waste characterization data, along with default emissions factors provided in ClearPath, were used to calculate emissions from waste disposal and composting. Multi-family and commercial tonnages were not directly available and were estimated using a single-family to multi-family + commercial ratio of 32:68, based on the transfer station survey data collected for the 2015-2016 Washington Waste Characterization Study. Waste characterization was comparable across Port Orchard samples and the Puget Sound-wide data from the aforementioned study; there was also little difference compared to King County compositions, which were assessed as a reference.  Landfill emissions assumed an 82.5% capture rate, based on Columbia Landfill’s confirmation of a “typical” collection scenario and ClearPath’s associated capture rates.  Wastewater emission calculations required data from Kitsap County Sewer District #7 and the City of Bainbridge Island wastewater treatment plants. City and District staff provided the data.  Data required for higher quality calculations in ClearPath, such as BOD5 or population specifically using septic systems (including unpermitted systems), were not available. Therefore, fugitive septic tank emissions were estimated based on the number of Bainbridge Island residents not served by a sewer connection. Ferry Emissions Attribution Accounting for emissions related to ferry transportation can be complicated by the fact that ferries typically travel between multiple jurisdictions. To simplify the methodology for calculating these emissions, the U.S. Community Protocol recommends allocating portions of the total emissions related to ferry transportation according to the number of stops located in each jurisdiction. For example, for the Bainbridge Island-Seattle ferry, this would equate to a 50/50 split of emissions since there is one ferry stop in each city. This is the calculation approach that was used in this community inventory. However, another possible method for attributing emissions to a particular jurisdiction is by ridership and examining the proportion of ferry riders from each city. To test that the 50/50 allocation was the correct choice for Bainbridge Island, we performed a sensitivity analysis using a ridership-based methodology. Using data reported by Washington State, we assumed that where travelers were heading in the evening on a weekday was likely where they reside. Based on this assumption and available data, we examined the proportion of total ferry travelers that end in Bainbridge Island in the evening versus elsewhere. Using this technique, we deduced that 47% of weekday ferry commuters are residents of Bainbridge Island—a value that is very close to the original 50% estimate. Given that this proportion strongly aligns with the 50/50 approach recommended by the U.S. Community Protocol, we ultimately followed the U.S. Community Protocol attribution approach to align with the robust and established methodology used by other communities in the region. CITY OF BAINBRIDGE ISLAND | 2019 | 37 Other Process & Fugitive Emissions  Data on fugitive refrigerant HFC and PFC emissions at city scales is scarce. We used the U.S. EPA’s 2019 Inventory of GHG Sources and Sinks and scaled the data to Bainbridge Island by population. GWP were adjusted to the latest available values in 2014 and 2018, which corresponded to the values in the 4th and 5th Assessment Reports, respectively.  PSE’s greenhouse gas inventories from 2014 and 2017 provided fugitive electric transmission and distribution emissions data for SF6. We scaled the data to Bainbridge Island using PSE customer data. Agriculture  The USDA provides publicly available data on the number of animals by county, which was scaled down to Bainbridge Island according to the ratio of livestock to land area. Kitsap County Conservation District provided data on manure management systems. The EPA provides national- level animal enteric and manure emissions factors, and state-level emissions factors for cattle. CITY OF BAINBRIDGE ISLAND | 2019 | 38 Considerations At the time of this inventory, not all communitywide data were available. Many calculations were scaled from national or county estimates by population. ClearPath’s default emissions factors were used in cases where local data were not available. There is always a certain level of uncertainty associated with inventories, given constraints in data availability for most communities. However, the goal of performing inventories is to attempt to accurately and precisely account for all sources of emissions using best available data and unbiased methodologies. While ranges of uncertainty are typically not quantified, we strive to use approaches that produce estimates that are neither conservative nor liberal in their leanings, and are based on the most locally-specific and reputable data sources. A list of data limitations and other considerations specific to Bainbridge Island’s communitywide inventory is provided below.  Multi-family and commercial solid waste tonnage: Since tonnage data was not available, we assumed a 32:68 ratio of single-family to multi-family and commercial waste, based on Port Orchard transfer station surveys in 2015.  Yard waste composting tonnage: Bainbridge Disposal data was available for 2018. To derive 2014 yard waste composting tonnage, we calculated the 2018 ratio of single-family tonnage to yard waste composting tonnage and applied the same ratio to 2014 single-family tonnage.  2018 VMT: PSRC data was available for 2014 and 2016. To derive 2018 VMT, we assumed a linear rate of increase from 2014 to 2016 to 2018.  2014 and 2018 Annual Revenue Miles: We scaled Kitsap County data to Bainbridge Island according to population, which may not accurately reflect public transit ridership on Bainbridge Island.  Septic tank fugitive emissions: In the absence of data on BOD5, we used the U.S. Community Protocol WW.11 Alternative Method to calculate methane emissions when only total population is known.  HFC and PFC tonnage: National data was available for 2014 and 2017 and was scaled down to Bainbridge Island by population. We assumed semiconductor manufacture, and the production of magnesium and aluminum, did not apply to Bainbridge Island and excluded contributions from those sources. The PFC-Substitution of ODS value provided was “does not exceed 0.05”; 0.04 was used as an estimate and may be an overestimate.  Air travel: We were not able to obtain fuel use data for 2018, so the air travel emissions were estimated by using a scaling factor from 2015 fuel use data based on the total landings for 2014 and 2018.  Off-road vehicle and equipment: We used the EPA MOVES model to estimate off-road vehicle and equipment emissions. The model estimates emissions at the county-level; we scaled Kitsap County emissions to Bainbridge Island by population. These considerations should be taken into account when interpreting findings from this inventory. We also recommend that future inventories seek to address these data limitations, if possible. CITY OF BAINBRIDGE ISLAND | 2019 | 39 COMMUNITY CONTRIBUTION ANALYSIS Introduction In 2014, Bainbridge Island’s 23,135 residents and 9,404 emitted approximately 214,425 MTCO2e. In 2018, the population increased 7% (24,891), the number of households increased 4% (9,798), and communitywide emissions increased 9% (to 233,998 MTCO2e). Apart from population growth, what drove that estimated change in emissions? We utilized the Analyzing Drivers of Change in Greenhouse Gas Emissions Inventories tool available from ICLEI USA to attribute changes in the community inventories to the economic, social and technological forces that influenced them.14 Results The communitywide contribution analysis highlighted factors that lead to the most significant changes in emissions between 2014 and 2018. The analysis revealed that changes to the PSE electricity fuel mix, increased activity at SeaTac airport (and, thus, higher estimated air travel emissions), and growth in population and employment all contributed to increases in emissions within the community. Conversely, downward forces on emissions included reduced ferry fuel use and more efficient driving and energy use. Figure 18 below shows the influences of the top three drivers on the two inventory years, plus a category labeled “other”—a compilation of contributors illustrated in Figure 19. Figure 17. High-level summary of major drivers of communitywide inventory increases and decreases. 14 ICELI USA Analyzing Drivers of Change in Greenhouse Gas Emissions Inventories tool. http://icleiusa.org/ghg- contribution-analysis/ (Accessed July 1, 2019). CITY OF BAINBRIDGE ISLAND | 2019 | 40 Figure 18. Detailed depiction of major drivers of inventory increases and decreases. **Includes effects of population on residential energy, VMT and waste generation. **After accounting for weather. This change is the net effect of factors that may include occupant behavior, changes to building types and uses, federal appliance standards, utility programs and new electronic devices. Specifically, contributors to inventory changes included the following, in order of highest increase to lowest decrease: • Electricity fuel mix (+10,542) describes changes to types of resources used to generate electricity for the community. For example, increased use of renewable sources such as hydroelectricity and solar and wind power would drive emissions decreases in the electricity fuel mix. For Bainbridge Island, changes in the PSE electricity fuel mix contributed to increases in electricity emissions. • Growth in population (+6,581) includes the impacts of increased housing, driving, and waste generation from Bainbridge Island’s growing population. Population grew 7.1% from 23,135 in 2014 to 24,891 in 2018. • Not analyzed (+5,418) describes the emissions from multiple sources—including increases in emissions from off-road vehicles, air travel, agriculture, wastewater, and refrigerant loss— that are not explicitly linked to a variable available in the tool. Due to the calculation CITY OF BAINBRIDGE ISLAND | 2019 | 41 methodology used to estimate emissions from these sources, many of these increases can be linked to population growth, which was used as a scaling factor to estimate emissions.15 • Growth in employment (+2,583) describes the effect of increased job growth and GDP per person on emissions. Economic growth leads to larger consumption of goods and services, therefore resulting in emissions increases. • Hotter summer (+499) is the increase emissions associated the enlarged strain on various systems due to hotter temperatures during the summer, such as increased electricity and energy consumption for residential and commercial cooling. The summer in 2018 was hotter than in 2014. • Colder winter (+366) is the increase emissions associated with the amplified strain on various systems due to colder temperatures during the winter, such as increased electricity and energy use for residential and commercial heating. The winter in 2018 was colder than in 2014. • Waste generation per person (-386) describes the impacts from change in the amount of waste disposed per person. In Bainbridge Island, the growth of waste generation was slower than that of population, which resulted in reduced per-person waste generation. • Decreased VMT per person (-919) describes changes from people driving more or less in the community. On average, each Bainbridge Island resident drove fewer miles, resulting in emissions decreases. • Decreased on-road emissions per mile (-1,061) represents changes in the amount of emissions per vehicle mile driven. These changes could be due to better driving behavior or more fuel-efficient vehicles. • Decreased energy use per household (-1,687) denotes a decline in the amount of energy used by the average Bainbridge Island household. This change could be due to more energy efficient behaviors or appliances. • Decreased commercial energy use per job (-3,754) incorporates the fact that job growth has increased, but a large conversion of energy efficient standards has been made throughout the commercial sector resulting in a decrease in emissions. 15 These sources are classified as “not analyzed” because population is used as a scaling factor to estimate emissions for these sources, and therefore population growth is already “baked into” the estimation. Thus, it would be inappropriate to group these sources with the “growth in population” driver. CITY OF BAINBRIDGE ISLAND | 2019 | 42 Municipal Operation Emissions The municipal operation emissions inventory accounts for the GHG emissions resulting from Bainbridge Island City government operations. Bainbridge Island’s government provides a range of municipal services including police, streets, planning, zoning, and general administration services. The City also operates the water and wastewater utilities for a portion of the island. Table 14 below provides an outline of the various emissions sources included in the municipal inventory, paired with their respective Scope. The table also specifies which sectors are required by the Local Government Operations Protocol. Table 13. Municipal inventory emissions and scope categories. Emissions Type Required? Scope 1 Scope 2 Scope 3 Buildings and Facilities Energy Electricity   Stationary Fuel Consumption   Streetlights and Traffic Signals   Transportation On-Road Fleet Vehicles   Off-Road vehicles   Employee Commute  Business Travel  Solid Waste, Potable Water, and Wastewater Solid Waste  Wastewater Treatment   Refrigerant Leakage   RESULTS Overview Bainbridge Island’s 2014 municipal-based GHG inventory totaled 2,067 MTCO2e. Municipal emissions in 2018 increased 11% to 2,291 MTCO2e. Major emissions sources in 2018 included facility electricity consumption (60%) and on-road fleet vehicles (17%) (see Table 15). Transportation and electricity also contributed some of the largest increases in emissions from 2014 to 2018. Municipal facility electricity and on-road fleet vehicles increased by 14% and 7%, respectively. Just four facilities accounted for 80% of all facility electricity use: Bainbridge Island Waste Water Treatment Plant (WWTP), City Hall, Fletcher Bay: Well Field, and Bainbridge Island Public Works Operations and Maintenance Yard. There were a few sectors with decreases in emissions, the largest of which was from streetlight and traffic signal improvements. CITY OF BAINBRIDGE ISLAND | 2019 | 43 Table 14. Municipal GHG emissions by year, sector, and scope. GHG Emissions (MTCO2e) Sector Scope 2014 2018 Building Propane 1 40 40 On-Road Fleet Vehicles 1 359 385 Off-Road Vehicles 1 77 63 Refrigerant Loss 1 18 18 Wastewater Treatment Plant 1 40 43 Electricity 2 1,212 1,383 Street Lights & Traffic Signals 2 103 91 Solid Waste Generation 3 59 84 Employee Commute 3 160 184 Business Travel 3 - - Total 2,067 2,291 Figure 19. Municipal emissions, by source and scope, for 2018 (total = 2,291 MTCO2e). CITY OF BAINBRIDGE ISLAND | 2019 | 44 Figure 20. Bainbridge Island municipal emissions, by year and source. Buildings and Facilities Energy The City of Bainbridge Island operates several government administration buildings, including City Hall, the Police Station, the Municipal Court, and a maintenance yard, as well as several community spaces, a wastewater treatment facility, and the streetlights and traffic signals throughout the City. This section details the GHG emissions resulting from the consumption of energy at these buildings and facilities. The types of energy consumed at these locations include electricity, provided by Puget Sound Energy, and propane, which is combusted in stationary equipment on-site. In 2018, energy consumption at City of Bainbridge Island buildings and facilities accounted for 66% of total municipal GHG emissions. Building electricity usage accounted for the bulk of emissions, comprising 89% and 91% of the energy sector’s emissions in 2014 and 2018, respectively (see Figure 22). From 2014 to 2018, building and facility electricity emissions grew by 14%, primarily due to fluctuations in the electricity utility fuel mix (see Electricity section below). The following sections provide greater detail on trends for the three sources of building and facility energy use: electricity, stationary fuel combustion, and streetlights and traffic signals. CITY OF BAINBRIDGE ISLAND | 2019 | 45 Figure 21. Bainbridge Island municipal buildings and facilities energy emissions. 2014 (Total = 1,354 MTCO2e) 2018 (Total = 1,513 MTCO2e) Electricity Electricity usage contributed the largest portion of GHG emissions within the City’s buildings and facilities sector, resulting in the emission of 1,212 MTCO2e in 2014 and 1,383 MTCO2e in 2018. Between 2014 and 2018, electricity emissions grew by 14%. However, electricity usage in kWh only grew by 4% within this same time period. This indicates that the growth in emissions between 2014 and 2018 was primarily due to fluctuations in the electricity utility fuel mix (i.e., changes in the proportion of fossil fuel sources used by the utility to generate electricity in a given year) rather than substantial increases in electricity usage. The two facilities with the greatest electricity usage in both years were the wastewater treatment facility (WWTP) and City Hall. These two facilities also experienced the greatest reductions in electricity usage between 2014 and 2018, demonstrating that the City is continuing to make progress on energy efficiency and conservation in the highest consumption facilities (see Figure 23 Figure 24 below). One way the City has reduced electricity usage at City Hall is through on-site solar electricity production: the solar panels at City Hall produced 71,300 kWh and 74,500 kWh of solar electricity in 2014 and 2018, respectively. Figure 22. Top electricity-using facilities in 2018. CITY OF BAINBRIDGE ISLAND | 2019 | 46 Figure 23. Facilities with the greatest electricity consumption increases between 2014 and 2018. Figure 24. Facilities with the greatest electricity consumption decreases between 2014 and 2018. Stationary Fuel Combustion Stationary fuel (propane) combustion was the lowest contributor to building-related municipal GHG emissions in both 2014 and 2018. Between both years, emissions remained constant at 40 MTCO2e. The only source for stationary fuel combustion is the Public Works Operations and Maintenance shop, which consumed an estimated 7,077 gallons of propane for heating. The number of gallons consumed remained constant from 2014 to 2018 and therefore resulted in no change in emissions. Streetlights and Traffic Signals The second largest contributor to GHG emissions 2018 were streetlights and traffic signals. Between 2014 and 2018, the City of Bainbridge Island increased the number of streetlights from 328 to 343. However, the emissions produced and electricity consumed by the streetlights has decreased by 12%. This decrease can be attributed to the City’s increased use of energy efficient bulbs. CITY OF BAINBRIDGE ISLAND | 2019 | 47 Transportation Transportation sector emissions are the third largest contributor to the City of Bainbridge Island’s municipal emissions. This section details the GHG emissions resulting from the consumption of gasoline, diesel, biodiesel, and propane in various vehicles. These vehicles are a part of numerous city operations, such as emergency services, landscaping, and government travel. In 2018, the transportation sector contributed 28% of all emissions, totaling 663 MTCO2e. The City of Bainbridge Island utilizes fuel for different vehicles such as boats, pickup-trucks, dump trucks, construction equipment, and patrol vehicles (see Figure 26). Municipal fuel consumption and emissions increased between 2014 and 2018, and there was a small decrease in fuel economy. Between 2014 and 2018, average municipal fuel economy decreased from 13 mpg to 12 mpg (see Figure 27). Figure 25. City of Bainbridge Island fleet vehicle emissions, by type. 2014 (Total = 436 MTCO2e) 2018 (Total = 449 MTCO2e) * Refers to police department fuel purchased outside of O&M. This data is not connected to specific vehicle types. CITY OF BAINBRIDGE ISLAND | 2019 | 48 Figure 26. Municipal fleet fuel consumption and average fuel economies, by year. On-Road Fleet Vehicles On-road vehicles made up the majority of transportation emissions in 2014 and 2018. Emissions come from on-road fleet vehicles driving on roadways within the community. In 2018, on-road vehicles consisted of 61% of all transportation sector emissions. We estimate that municipal on-road vehicles produced 359 MTCO2e and 385 MTCO2e in 2014 and 2018, respectively. The 2003 Volvo Vactor Sewer Cleaner consumed the greatest amount of fuel both years (1,518 gallons of diesel in 2014 and 2,673 gallons in 2018). The Patrol/Interceptor vehicles sector contained the largest number of fuel-consuming vehicles. Figure 28 presents the top-10 fuel consuming vehicles in 2018. Figure 27. Top 10 fuel-consuming fleet vehicles, 2018. CITY OF BAINBRIDGE ISLAND | 2019 | 49 Off-Road Vehicles Off-road equipment, such as lawnmowers, boats, and forklifts, contribute another 10% of transportation sector emissions in 2018. There was a 17% reduction in emissions between 2014 and 2018, from 77 MTCO2e to 63 MTCO2e. Reductions in emissions are from reduced fuel consumption, decreasing from 7,730 gallons in 2014 to 6,474 gallons in 2018. Employee Commute Employee commute includes single-occupancy vehicles (SOV), vanpools and carpools. In 2014, employee commuting produced an estimated 160 MTCO2e from 486,922 VMT. There were 106 full-time equivalent municipal employees, resulting in 1.51 MTCO2e emitted per employee. The emissions increased in 2018 to 184 MTCO2e. This was a consequence of an increase in VMT to 646,027. In 2018, there were 119 full- time equivalent employees for the city with only a small increase to 1.55 MTCO2e per person. Solid Waste and Wastewater Waste sector includes emissions from solid waste generation and biological process emissions related to wastewater treatment—detailed below. Solid Waste Solid waste is comprised of municipal landfill waste and municipal diverted waste (recycling, cardboard, and yard waste). We estimate that emissions from waste increased from 59 MTCO2e in 2014 to 84 MTCO2e in 2018. Much of this increase can be attributed to additional service at the Waterfront Park and downtown areas. An increase in foot traffic in the Winslow downtown area, as well as the reconstruction of the Waterfront Park and City Dock, required the installation of more waste receptacles in these areas. We estimate that 147 tons of waste was sent to the landfill in 2014 and that amount increased to 209 tons in 2018. Between 2014 and 2018, the amount of diverted waste increased from 8.6 tons to 28.8 tons. Yard waste in particular, increased from 2.1 tons in 2014 to 3.9 tons in 2018. This increase in yard waste composting contributed to a slight increase in emissions within the diverted waste category, from 0.15 MTCO2e to 0.27 MTCO2e (Figure 29). CITY OF BAINBRIDGE ISLAND | 2019 | 50 Figure 28. Municipal solid waste disposal trends. Wastewater Treatment The City owns and operates the Winslow Wastewater Treatment Plant and operates collection for the South Island Sewer System. The Winslow Sewer System serves the historic downtown Winslow area and the South Island Sewer System serves the Lynwood Center, Point White, Pleasant Beach, Emerald Heights, Blakely School, and Rockaway Beach neighborhoods. The collection systems have a combined infrastructure that includes emergency generators, manholes, residential grinder pumps and sewage pump stations. In 2018, emissions from wastewater treatment totaled an estimated 43 MTCO2e. This was an 7% increase from 2014 when emissions were 40 MTCO2e. This can be attributed to the increase in population served from 4,743 in 2014 to 5,079 in 2018.16 Refrigerant Leakage Leaks in refrigeration systems cause emissions of potent greenhouse gases. These losses can occur in domestic refrigeration, commercial refrigeration, industrial refrigeration, chillers, and residential/commercial air conditioning units. Refrigerant gases differ in their Global Warming Potential (GWP)—the amount of heat a greenhouse gas traps in the atmosphere relative to carbon dioxide. Emissions from R-22 leaks, for example, are especially impactful: they have a GWP 1,760, which means that they are over one thousand times more damaging than carbon dioxide. The City’s heating and cooling systems largely use R-22, which we estimate produces 18 MTCO2e per year. 16 Service populations were estimated at 25% of the total Bainbridge Island population, as estimated by Bainbridge Island City staff. CITY OF BAINBRIDGE ISLAND | 2019 | 51 DATA SOURCES AND METHODOLOGY The municipal emissions inventory involved acquiring the following data types, summarized in Table 16 and detailed in the following sections:  Activity data that quantifies levels of activity that generate GHG emissions, such as miles traveled, and kWh of electricity consumed.  Emissions factors that translate activity levels into emissions (e.g., MgCO2e per kWh). Data quality is assessed and reported on a High (H), Medium (M), and Low (L) scale in accordance with GHG inventory best practices:  A High rating indicates data are detailed and specific to the local geography  A Medium rating indicates data are more general or modeled with robust assumptions and may not be specific to the local geography, but are downscaled from a slightly broader geography (e.g., state- level)  A Low rating indicates data are highly modeled, uncertain, or a default value was used based on national characteristics. Table 15. Key data sources for the City of Bainbridge Island's municipal inventory. Sector Activity Data (AD) AD Quality Emissions Factors (EF) EF Quality Buildings and Facilities Energy Electricity • kWh consumption compiled from PSE bills H • PSE reported emissions factors (PSE Greenhouse Gas Inventory, 2014 & 2017, Table 7.1) H Stationary Fuel Combustion • Gallons of propane used for heating compiled based on municipal purchasing information provided by O&M M • Emissions factors provided in ClearPath program M Streetlights and Traffic Signals • Quantity, type, and wattage of each streetlight and traffic signal transcribed from PSE bills • Installed wattage used to estimate electricity consumption in kWh M • PSE reported emissions factors (PSE Greenhouse Gas Inventory, 2014 & 2017, Table 7.1) H Transportation CITY OF BAINBRIDGE ISLAND | 2019 | 52 Sector Activity Data (AD) AD Quality Emissions Factors (EF) EF Quality On-Road Fleet Vehicles • Fuel volumes by vehicle and type, along with annual miles traveled, provided by Public Works staff • Where fuel volumes and miles traveled were unavailable for specific vehicles, the total quantity of fuel purchased was provided (with no vehicle information associated) H • CO2 emission factors provided in ClearPath program • CH4 and N2O average emissions factors for each vehicle and fuel type were calculated based on the model year composition of the municipal fleet vehicles • Emissions factors used in these calculations were derived from the EPA (2017) Inventory of U.S. Greenhouse Gas Emissions and Sinks (Climate Leadership Resource) H Off-Road Vehicles • Fuel volumes by vehicle and type, along with annual hours of use, provided by Public Works staff • Where fuel volumes were unavailable for specific vehicles, the total quantity of fuel purchased was provided (with no vehicle information associated) H • Emissions factors provided in ClearPath program • Emissions factors for off-road vehicle/equipment propane use were missing from the ClearPath program; these emissions were calculated separately using emissions factors derived from the EPA (2017) Inventory of U.S. Greenhouse Gas Emissions and Sinks (Climate Leadership Resource) M Employee Commute • MTCO2e obtained from Commute Trip Reduction survey conducted by WSDOT M • N/A; value converted to MTCO2e by WSDOT N/A Solid Waste and Wastewater Solid Waste Generation • Bin sizes, pickup frequency, and waste type were transcribed from solid waste bills • Estimated cubic yards of waste were then converted to tons using EPA solid waste weight conversions • This methodology assumes all waste pickup containers were full L • Waste characterizations from King County multifamily and commercial buildings used to estimate Bainbridge Island municipal waste composition • ClearPath default emissions factors used to estimate all future methane emissions resulting from the decomposition of each waste type L Wastewater Treatment Facility • Wastewater treatment facility processes determined based on desktop research and confirmed with city staff • Population served by wastewater treatment facility estimated by multiplying the number of sewer connections (city staff) by the average household size (ACS) M • Local Government Operations Protocol equations 10.8 and 10.10 used to calculate N2O emissions • CH4 emissions were not applicable since the system is aerobic L Refrigerant Leakage CITY OF BAINBRIDGE ISLAND | 2019 | 53 Sector Activity Data (AD) AD Quality Emissions Factors (EF) EF Quality Refrigerant Leakage from Building Heating and Cooling Equipment • List of heating/cooling equipment provided by city staff • Equipment categorized by type and refrigerant used L • Emissions calculated using Equation 6.35 and the default factors in Table 6.4 of the Local Government Operations Protocol L CITY OF BAINBRIDGE ISLAND | 2019 | 54 Buildings and Facilities Energy  We acquired electricity consumption data by hand transcribing individual monthly PSE electricity bills for the City of Bainbridge Island’s municipal operations. City staff photocopied and provided these bills.  In 2018, gallons of propane used to heat the maintenance shop were tracked in City records. However, only cost information was tracked in 2014. Based on the comparison of cost information between the two years, City staff estimated the propane use at this facility was the same in 2014 and 2018.  Propane usage in an emergency generator at the Senior Center was omitted from the 2014 and 2018 inventories. In 2014, the generator was not maintained by the City and in 2018 the generator was not used.  We estimated the total electricity usage of streetlights and traffic signals operated by the City of Bainbridge Island by counting the number, wattage, and types of streetlights and traffic signals included in the individual monthly PSE electricity bills. We used Equation 6.15 in the Local Government Operations Protocol to convert wattage to estimated electricity consumption in kWh. We assumed an average daily operating time of 12 hours for streetlights and 8 hours for traffic signals. Transportation  We categorized the list of on-road and off-road vehicles and equipment into standard vehicle types (i.e., passenger car, light-duty truck, heavy-duty truck, etc.) based on provided vehicle descriptions and make/model information. We totaled fuel gallons and vehicle miles traveled by each vehicle category and type of fuel. CO2 emissions were derived based on usage volumes by type of fuel and ClearPath default emissions factors. CH4 and N2O emissions were derived based on miles traveled by vehicle category and fuel type. We used CH4 and N2O emissions factors for each vehicle category-fuel type combination originally derived from the EPA (2017) Inventory of U.S. Greenhouse Gas Emissions and Sinks and modified these factors to reflect the municipal fleet age composition.  One forklift within the municipal fleet used propane. However, propane is not an available fuel type for off-road equipment in the ClearPath program. Therefore, we calculated propane emissions outside of ClearPath using emissions factors derived from the EPA (2017) Inventory of U.S. Greenhouse Gas Emissions and Sinks.  We obtained total emissions (in MTCO2e) associated with employee commuting from the Commute Trip Reduction Survey conducted by WSDOT. Solid Waste and Wastewater  We acquired waste generation data by hand transcribing individual monthly Bainbridge Disposal bills for the City of Bainbridge Island’s municipal operations. City staff photocopied and provided these bills.  Waste bills generally indicated the disposal bin volume, the pickup frequency, and the waste stream (i.e., recycling, yard waste, trash, etc.). We assumed that waste bins were full at time of pickup and CITY OF BAINBRIDGE ISLAND | 2019 | 55 that waste described as recycling or cardboard was recycled, waste described as yard waste was composted, and waste described as trash was landfilled. We converted annual volumes of waste by type into tons of waste by type using EPA solid waste weight conversions. We then used King County waste composition information and ClearPath default emissions factors to estimate emissions associated with waste generation and composting.  We estimated wastewater treatment N2O emissions using Equation 10.7 and 10.10 of the Local Government Operations Protocol. Population served was estimated by Bainbridge Island City staff. Default nitrogen load information was used since local data was not available. Refrigerant Leakage  We acquired a list of heating/cooling equipment used in municipal buildings from Bainbridge staff. We did an online search of the equipment models and serial numbers to determine how to best categorize each piece of equipment. Using these categorizations, we estimated emissions from refrigerant leakage using Equation 6.35 and Table 6.4 of the Local Government Operations Protocol. CITY OF BAINBRIDGE ISLAND | 2019 | 56 Considerations There is some degree of uncertainty in any GHG inventory. This uncertainty can come from incomplete data, but it can also result from uncertainty in the methodology or emissions factors used in calculating units of activity (e.g. gallon of fuel, kilowatt-hour of electricity, short ton of solid waste) into CO2-equivalent emissions. These considerations should inform future inventory and reporting efforts, including prioritization of additional data collection, understanding of inventory results, and in the development of mitigation goals and monitoring systems. Considerations for the municipal operations inventory include:  Electricity consumption: The hand transcribing of electricity and waste bills introduces a greater potential for human error.  Streetlight and traffic signal daily operations: Assumptions made on average daily operating time could affect electricity consumption amounts.  Employee commuting emissions: WSDOT modeled the Commute Trip Reduction Survey, therefore the methodology, such as emissions factors assumed for passenger vehicles, could be different from those used for community passenger vehicles.  Municipal vehicle fleet: Some municipal feel vehicles and equipment are fueled by bulk fuel purchases or mobile fueling tanks. In these instances, we had to estimate vehicle miles traveled and vehicle composition, potentially affecting the CH4 and N2O calculations.  Solid waste generation: Disposal frequency can be difficult to determine from solid waste bills— which was the only data available to deduce solid waste disposal amounts. We also lacked data on the amount of waste included in each disposal bin at the time of pickup, we assumed all bins were full. The hand transcribing of waste bills introduces a greater potential for human error.  Solid waste characterization: Bainbridge Island lacked a local waste characterization study, therefore, we applied King County waste characterization data to estimate the relative proportions of different materials in the waste stream. The waste composition for Bainbridge Island may be different than for King County.  Wastewater treatment: Due to limitations in local data availability, the calculations used were default national values based on the treatment method. CITY OF BAINBRIDGE ISLAND | 2019 | 57 MUNICIPAL CONTRIBUTION ANALYSIS Introduction In 2014, total municipal emissions were 2,067 MTCO2e. In 2018, total emissions increased to 2,291 MTCO2e, resulting in a 11% (223 MTCO2e) increase from the 2014 inventory. We employed the Analyzing Drivers of Change in Greenhouse Gas Emissions Inventories tool available from ICLEI USA to attribute changes in the municipal inventories to the economic, social and technological forces that caused them.17 The primary goal of this tool is to provide a methodology for discovering the drivers for change between the two inventory years. Results The contribution analysis revealed the factors that led to the net increase in emissions between 2014 and 2018. In summary, a lower-renewables electricity fuel mix as reported by Puget Sound Energy, increased water treatment electricity use, and increased use of fleet vehicles all contributed to emissions increases. Reductions were due to more carbon-efficient employee commuting, more energy-efficient streetlights, and increased fuel economy of the vehicle fleet. Figure 30 shows the influences of major forces on the inventory between 2014 and 2018, and Figure 31 presents a more detailed depiction. Figure 29. High-level summary of major drivers of government operations inventory increase and decreases. 17 ICELI USA Analyzing Drivers of Change in Greenhouse Gas Emissions Inventories tool. http://icleiusa.org/ghg- contribution-analysis/ (Accessed July 1, 2019). CITY OF BAINBRIDGE ISLAND | 2019 | 58 Figure 30. Detailed depiction of contributions to change for Bainbridge Island government operations GHG emissions. Contributors to inventory changes included the following, in order from highest increases to lowest decreases: • Electricity fuel mix (+131 MTCO2e) describes changes to types of resources used to generate electricity for the community. For example, increased use of renewable sources such as hydroelectricity and solar and wind power would drive emissions decreases in the electricity fuel mix. For Bainbridge Island, changes in the PSE electricity fuel mix contributed to increases in electricity emissions. • Water treatment kWh/gallon (+66 MTCO2e) includes changes in the energy efficiency of processing wastewater. • Vehicle fleet total VMT (+61 MTCO2e) describes the increased emissions due to a rise in total vehicle miles traveled for municipally owned vehicles. • Other increases (+103 MTCO2e) is a compilation of changes in: population served by water treatment, number of employees commuting, waste generation per employee, and number of streetlights. • Employee commute MTCO2e/mile (-25 MTCO2e) represents more efficient commuting behaviors, possibly by changes in mode, distance, or vehicle efficiency. • kWh per street light (-25 MTCO2e) incorporates the fact that streetlight and traffic signal numbers have increased, but a large conversion to LED prior to 2018 reduced electricity use and emissions. CITY OF BAINBRIDGE ISLAND | 2019 | 59 • Vehicle fleet CO2e/mile (-48 MTCO2e) is the reduction in emissions associated with reduced gasoline consumption in newer vehicles meeting more stringent federal standards, as well as the retiring of older, less fuel-efficient vehicles. • Other decreases (-21 MTCO2e) includes decreased emissions from changes in the type of heating fuels used and less wastewater produced per-person. CITY OF BAINBRIDGE ISLAND | 2019 | 60 Consumption-Based Emissions The purchasing decisions we make impact the environment. Some types of foods and materials, such as meat and furniture, can carry a significant GHG emissions burden. For example, meat and dairy cows emit methane—a potent greenhouse gas. Residents of Bainbridge Island who consume beef contribute to the emissions from these cows—even if the cows are raised outside the island. This section presents estimated emissions associated with the purchasing decisions made by Bainbridge Island households. Consumption-based emissions are provided on a per-household basis because a household is the economic unit used in the underlying analyses (in other words, the analytical approach is based on household, not individual, purchasing trends and decisions). For example, food and home goods are generally purchased by one person for the use of their entire household. RESULTS Overview Results from a household-based economic modeling tool suggest that the average Bainbridge Island household emits 52 MTCO2e a year through their purchasing behaviors.18 There were an estimated 9,404 and 9,798 households on Bainbridge Island in 2014 and 2018, respectively, indicating that total consumption-based emissions from all households on Bainbridge Island could have reached approximately 510,000 MTCO2e in 2018. Major drivers include the purchase of meat, furniture, clothing, home energy, and travel-related expenses such as car fuel and air travel (see Figure 32 and Figure 33). Figure 31. Estimated 2018 consumption-based emissions for Bainbridge Island (total = 509,496 MTCO2e). 18 As indicated from U.C. Berkeley’s CoolClimate Calculator. Outcomes from the consumption-based inventory analysis are presented at the per-household level because purchasing behavior is typically examined and analyzed at the household—not individual—level. CITY OF BAINBRIDGE ISLAND | 2019 | 61 Figure 32. Estimated annual individual household consumption-based emissions for Bainbridge Island residents (total = 52 MTCO2e/year). DATA SOURCES AND METHODOLOGY Because consumption-based inventories are inevitably coarse in scale compared to other inventory sectors, we used readily available sources to estimate Bainbridge Island’s consumption-based emissions. Specifically, we used UC Berkeley’s CoolClimate Calculator, which employs an Economic Input-Output Life Cycle Assessment model (EIO-LCA), designed by Carnegie Mellon University, and the Comprehensive Environmental Database Archive to calculate consumption-based emissions. This model employs an economy-wide model of cradle-to-grave emissions of all major greenhouse gases for over 400 economic sectors of the economy. Considerations The CoolClimate Calculator largely uses national-level data to estimate emissions associated with purchased goods and services. This means that more locally-specific emissions factors—such as those associated with the electricity we purchase—are not incorporated. For Bainbridge Island, that may mean that the calculator is slightly overestimating household emissions, since the Puget Sound Energy electricity fuel mix is different than the national average. However, the calculator does use regionally- specific economic data to characterize spending habits. CITY OF BAINBRIDGE ISLAND | 2019 | 62 Conclusion and Next Steps These GHG inventories provide a solid and informative foundation for taking climate action:  The Bainbridge Island community now understands its major contributors of GHG emissions— residential electricity, non-residential electricity, on-road vehicles, and air travel—revealing major sectors to focus climate action efforts. Conducting inventories for two years also revealed emissions trends: a 9% increase between the two inventory years as well as increased emissions in almost all sectors. The community could consider approaches for addressing emissions sources that are increasing, such as increasing electric vehicle uptake, incentivizing alternatives to air travel, continuing to increase building use efficiency, and reducing growth. The contribution analysis also provided insight into the underlying drivers of these trends: we learned that continuing to push for lower-emissions electricity sources will be important for lowering Bainbridge Island’s climate impact.  The City of Bainbridge Island now understands its key drivers of greenhouse gas emissions— transportation, electricity, and wastewater treatment—which highlight key areas for reduction. The inventories for the two years demonstrated an upward trend in emissions, similar to the community inventory. Between 2014 and 2018, a 11% increase in emissions occurred. The municipal contribution analysis highlighted the drivers of these trends: changes in electricity fuel mix, water treatment operations, and vehicle fleet usage. The City of Bainbridge Island did see small reductions in streetlight and traffic signal use and fleet vehicle fuel economy. The City can consider actions for reducing emissions from increasing sources, such as transferring to a low-carbon electricity fuel mix and decreasing the amount of vehicle travel.  Bainbridge Island households learned the impact of spending habits—an estimated 52 MTCO2e of emissions per year, on average. The consumption-based inventory also exposed key contributors to that footprint: meat consumption, travel expenses, and the purchase of furniture, clothing, and energy. Bainbridge households could consider behavior changes to lower these emissions, such as switching to a low-carbon diet-which includes more vegetables, fruits and whole grains; utilizing alternative forms of travel; and purchasing second-hand goods. This foundational work also illuminated opportunities for improving and expanding future inventories. We were not able to obtain data related to commercial propane use, for example. The City can continue to work with local propane vendors to start collecting these data and, eventually, begin incorporating this important emissions source into the city’s inventory. These and other improvements will not only paint a fuller picture of the community’s overall environmental impact but can also motivate businesses and individuals to better understand their personal carbon footprint and identify pathways for reducing that impact. CITY OF BAINBRIDGE ISLAND | 2019 | 63 References Intergovernmental Panel on Climate Change. (2014). Fifth Assessment Report. https://www.ipcc.ch/assessment-report/ar5/ (accessed June 29, 2019). King County Waste Monitoring Program. (2015). 2015 King County Waste Characterization and Customer Survey Report. Puget Sound Energy. (2015). 2014 Greenhouse Gas Inventory. Puget Sound Energy. (2018). 2017 Greenhouse Gas Inventory. The Climate Registry. (2010). Local Government Operations Protocol: For the quantification and reporting of greenhouse gas emissions inventories. U.S. Department of Agriculture. (2012). National Agricultural Statistics Service. https://www.nass.usda.gov/AgCensus/ (accessed June 29, 2019). U.S. Department of Agriculture. (2017). National Agricultural Statistics Service. https://www.nass.usda.gov/AgCensus/ (accessed June 29, 2019). U.S. Environmental Protection Agency. (2014). Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2014. U.S. Environmental Protection Agency. (2017). Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2017. Washington State Department of Ecology. (2018). 2015-2016 Washington Statewide Waste Characterization Study. Publication 16-07-032. 157 pp. Washington State Department of Transportation. (2017). Commute Trip Reduction Survey. https://www.wsdot.wa.gov/sites/default/files/2007/03/05/2018-07-12_Calculating_GHG.PDF (accessed June 29, 2019). World Resource Institute (2014). Greenhouse Gas Protocol: Global protocol for community-scale greenhouse gas emission inventories. CITY OF BAINBRIDGE ISLAND | 2019 | i Appendix: Tree Carbon Sequestration While inventories focus on emissions of greenhouse gases into the atmosphere, vegetation such as trees and grasslands can sequester—or absorb—carbon dioxide from the atmosphere. Land use changes, such as development of open and working forest and agricultural lands, can affect this process. To illustrate and better understand this benefit, we sought to quantify the extent to which existing trees in Bainbridge Island sequester carbon from the atmosphere. It is important to note that while vegetation on Bainbridge Island is taking carbon dioxide out of the atmosphere, this process does not offset or counteract emissions of greenhouse gases from the community, as quantified in the other sections of this report. Such land carbon “offsets” can only be achieved through rigorous and verifiable changes in land management that result in a net increase in carbon sequestration as compared to a pre-established baseline. This analysis does not quantify such activities, but rather offers some insight into the ecosystem services the trees and natural and working forests of Bainbridge Island provide, and the potential losses of services that could result from their development or removal. DATA SOURCES AND METHODOLOGY We utilized the i-Tree Canopy software to estimate annual tree carbon sequestration on Bainbridge Island. The software facilitates a supervised random sampling using Google Maps aerial photography. Specifically, the software walks users through the following steps:19  Define the project area using Google Maps aerial photography (see Figure 35).  Select project location using dropdown menus. This step characterizes per-acre carbon sequestration based on anticipated vegetation types (e.g., types of trees) typical in that location. For Bainbridge Island, the selected location was Kitsap County (see Figure 36).  Conduct random sampling. i-Tree Canopy shows a magnified image from the project area with a point identified with crosshairs (see Figure 37). The user then identifies whether the crosshairs are on a tree or “non-tree” area.  Repeat random sampling. The tree/non-tree identification is repeated multiple times to increase statistical precision of the estimated percent tree canopy cover of the project area. For this exercise, we classified 100 random samples.  Analyze results. i-Tree Canopy uses the random sampling to provide an estimate of overall percent tree canopy cover and resultant tree carbon benefits. The estimate carries a level of uncertainty (results are indicated by a mean value +/- one standard error). This uncertainty stems from limitations associated with the sampling approach. 19 More information regarding the methodology used by i-Tree Canopy can be found at https://canopy.itreetools.org/resources/iTree_Canopy_Methodology.pdf. CITY OF BAINBRIDGE ISLAND | 2019 | ii Figure 33. Defining the project area within i-Tree Canopy. CITY OF BAINBRIDGE ISLAND | 2019 | iii RESULTS Outcomes from the i-Tree Canopy analysis suggest that 51% of the Bainbridge Island land mass was covered with trees in 2018.20 Those trees sequester an estimated 58,727 MTCO2e from the atmosphere every year.21 This value appears to have decreased slightly since 2014; sampling of historical aerial imagery suggests a tree canopy cover of 55% in 2014 (equivalent to an estimated 63,334 MTCO2e annual sequestration rate).22 This change in tree canopy cover would result in an estimated net loss of 4,607 MTCO2e in carbon sequestration benefits annually. Figure 34. Estimated tree canopy cover on Bainbridge Island in 2018, using random sampling from the i-Tree Canopy software.23 20 Value depicted as mean estimate, with 95% confidence interval of 41.2% to 60.8%. 21 Assumes a sequestration rate of 10,010 lbs. CO2/acre/year. Source: i-Tree Canopy v.6.1. 22 Value depicted as mean estimate, with 95% confidence interval of 45.3% to 64.7%. 23 https://canopy.itreetools.org/ CITY OF BAINBRIDGE ISLAND | 2019 | iv Figure 35. Selected project location for the Bainbridge Island study. Figure 36. Example of random sampling classification exercise. CITY OF BAINBRIDGE ISLAND | 2019 | v Considerations This carbon sequestration analysis represents a high-level estimate of annual land carbon sequestration on Bainbridge Island. Data limitations and other considerations include the following:  Omission of non-tree vegetation: This approach assumes that non-tree vegetation does not sequester carbon, which is not the case. This analysis does not include carbon benefits from non- tree vegetation such as agriculture, pasture, and shrubs.  Tree generalization: This approach does not explicitly differentiate between tree types, but assumes that all trees sequester an average, representative amount of carbon every year.  Statistical sampling: This approach extrapolates a statistical sampling of an area, rather than analyze the area in its entirety, which inevitably results in some level of statistical uncertainty and imprecision.