1、652 ASHRAE TransactionsABSTRACTBuildings are responsible for approximately 40% of the primary energy use and 36% of greenhouse gas (GHG) emis-sions in the U.S. As we move toward reducing GHG emissions, we need reliable methods for estimating emissions and emis-sion reductions. GHG emissions come fro
2、m all life cycle stages of a building; however, this paper focuses on those associated with energy used in the building. Many data sources and tools are available for calculating the GHG emissions from building activities, but they have different assumptions, different data sources, different system
3、 boundaries, and there are no agreed-upon standards. Therefore, results of GHG emission calcula-tions are neither consistent nor comparable. They often do not include the full life cycle of the energy and fuels and do not account for the regional and temporal variations in power generation and emiss
4、ions. Temporal variations become impor-tant as load-shifting technologies and renewable energy generation are added to buildings. This paper presents the issues associated with estimating buildings-related GHG emissions and estimates the impacts of each issue. Recent and planned projects will provid
5、e more detailed regional and hourly data, but there are still many uncertainties and more work to do to develop accurate, easy-to-use tools. INTRODUCTIONThere are several debates over the existence, causes, and effects of global warming; however, most people agree that anthro-pogenic emissions have
6、increased over the last 100 years. World human-sourced greenhouse gas (GHG) emissions increased 70% from 1970 to 2004, and carbon dioxide (CO2) emissions grew 80% in this same time period and accounted for 77% of the total 2004 GHG emissions (IPCC 2007). GHG emissions are coming increas-ingly under
7、voluntary and regulatory controls as the world moves toward reducing our impact on climate change. Terms such as carbon footprint and carbon offsets are becoming common parts of our daily language, but they are not well understood. There are several on-line carbon calculator tools to help interested
8、 users, but these tools often produce different results and there is little infor-mation about how they should be applied to buildings. We naturally want an easy solution such as a single number with our monthly utility bills. The utility companies are very good at measuring how much energy they sel
9、l, but going from the energy sold back to the primary energy used and calculating the emissions associated with delivered energy are problematic. There are other sources of emis-sions and tracking them all down can be challenging. This paper examines many issues associated with estimating these emis
10、sions and provides recommendations about what should be included in calculating GHG emissions from building operations.EMISSIONS FROM BUILDINGSBuildings in the U.S. account for approximately 70% of the electricity use, 39% of the primary energy consumption, 38% of the CO2emissions, and 36% of the GH
11、G emissions (DOE 2009, EIA 2008a). The CO2 emissions from energy use in U.S. buildings accounted for 8% of the global CO2 emis-sions in 2006 (DOE 2009). In the United States, energy use in and emissions from the buildings sector continue to grow faster than in the other sectors (Figure 1) (DOE 2009,
12、 EIA 2009). The average site energy use intensity of buildings has stayed nearly constant since 1985 (see Figure 2). The increase in emissions from buildings results from an increase in total floor area and from the increased use of electricity relative to on-site natural gas use (Figure 2) (EIA 200
13、8b). The national average emissions rate per unit of energy for electricity is Moving Toward Better GHG Calculations for BuildingsMichael Deru, PhDMember ASHRAEMichael Deru is senior engineer with the Center for Buildings and Thermal Systems at the National Renewable Energy Laboratory in Golden, CO.
14、AB-10-0272010, American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc. (www.ashrae.org). Published in ASHRAE Transactions (2010, Vol. 116, Part 2). For personal use only. Additional reproduction, distribution, or transmission in either print or digital form is not permitted w
15、ithout ASHRAEs prior written permission.2010 ASHRAE 653approximately three times higher than the emissions rate for on-site use of natural gas.GHG emissions associated with buildings come from many sources. The largest source is the combustion of hydro-carbon fuels, which releases CO2, methane (CH4)
16、, and nitrous oxide (N2O). The other main sources of GHGs are leakage of halocarbon refrigerants, CH4, and sulfur hexafluoride (SF6). The emissions of refrigerants and SF6are small, but their impact can be significant. The combined impact of various GHGs is often expressed as carbon dioxide equivale
17、nt (CO2e), which combines the impact relative to CO2using the global warming potential (GWP).Every GHG has a different impact, and their relative strengths are often expressed by the GWP, which compares the ability of the gas to trap radiant energy in the atmosphere rela-tive to CO2over a defined pe
18、riod. The 100-year time horizon was adopted early by the Intergovernmental Panel on Climate Change (IPCC) and the Kyoto Protocol, and has since become the de facto standard. There continues to be some debate over the correct time horizon, or even if GWP is the best index (Shine 2009). The 20-year, 1
19、00-year, and 500-year GWPs for several gases of concern in buildings are shown in Table 1. The GWPs of the blended refrigerants are a weighted average of the GWPs of the constituent refrigerants. The GWPs are from the 2007 IPCC Working Group I report, and the refrigerant blend mass fractions are fro
20、m the 2006 ASHRAE Refrigera-tion Handbook (ASHRAE 2006, Forster et al. 2007). There are two interesting points to note about the GWPs in Table 1. Refrigerants and SF6are very strong GHGs with GWPs three to four orders of magnitude greater than CO2. Refrigerant leakage from supermarkets can easily ac
21、count for 50% of the annual GHG emissions, depending on leakage rates and the electricity source. The question of time horizon is most interesting for CH4and the refrigerants. Because these gases have relatively short atmospheric lifetimes compared to CO2, they have higher GWPs for shorter time hori
22、zons. If the 20-year time horizon were the standard for GWPs, there would be more concern over these gases. This may change our prior-ities for mitigating GHG emissions. This paper focuses on the emissions related to energy use in buildings; however, it is important to realize that GHG emis-sions oc
23、cur throughout every phase of a buildings life. A full life cycle assessment (LCA) would include the energy and emissions required to construct a building, manufacture and transport all the products used in the building, use of the build-ing, water use, transport of the occupants to and from the bui
24、lding, and final disposal of the building at the end of its life. The use of LCA for building products and whole buildings is growing and becoming standardized with the development of standard data sources (NREL 2010) and international stan-dards for developing environmental product declarations (IS
25、O 2006, ISO 2007).CALCULATING EMISSIONSThere are no agreed-upon protocols for calculating the GHG emissions specifically for buildings; however, The GHG Protocol for Project Accounting (WRI and WBCSD 2005) provides general guidance for projects and PAS 2050:2008 (BSI 2008) provides a protocol for go
26、ods and services. The GHG Protocol for Project Accounting fails to address many issues that are specific to buildings, but it does present a general approach that can help with buildings-related projects. The first step of any project is to state the goals and objectives to define what will be accom
27、plished and to establish the need for the project. Next, the scope and boundaries should be clearly defined so implementers and users understand the Figure 1 Annual CO2 emissions in the United States by end use sector (EIA 2009).Figure 2 Commercial building site energy use intensities from CBECS (EI
28、A 2008b).2010, American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc. (www.ashrae.org). Published in ASHRAE Transactions (2010, Vol. 116, Part 2). For personal use only. Additional reproduction, distribution, or transmission in either print or digital form is not permitted w
29、ithout ASHRAEs prior written permission.654 ASHRAE Transactionsextent of the project. The scope defines the emission sources and level of detail that will be included in the analysis. The temporal and physical boundaries define the time frame and what will be included. An example of a physical bound
30、ary definition and the energy flows into and out of that boundary is shown in Figure 3. The practitioner needs to know the types of fuels and how they are used to estimate the emission factors and determine the GHG emissions.For buildings connected to the utility grid, it is impossible to know exact
31、ly where the energy comes from every moment; therefore, it is impossible to determine the exact emissions associated with electricity generation. Several assumptions and approximations have to be made to estimate the emissions and the uncertainties associated with these approximations is difficult t
32、o determine. Several on-line calculators and sources of data can be used to calculate the GHG emissions from buildings. A recent survey identified 48 emissions calculators that are available on the Internet (Yazdani et al. 2009). These tools vary in scope, data sources, data coverage, units, definit
33、ions of terms, and boundaries. These variations make it difficult to compare results between studies. Table 2 presents some items to look for and understand concerning what may or may not be included in a GHG calculation. The main issues are discussed further following the table.Table 1. Global Warm
34、ing Potentials20-Year 100-Year 500-YearCO21 1 1CH472 25 7.6N2O 289 298 153SF616,300 22,800 32,600R11 6,730 4,750 1,620R12 11,000 10,900 5,200R22 5,160 1,810 549404A 6,010 3,922 1,328407A 4,538 2,107 655410A 4,340 2,088 653Figure 3 Example of energy flows and analysis boundary.2010, American Society
35、of Heating, Refrigerating and Air-Conditioning Engineers, Inc. (www.ashrae.org). Published in ASHRAE Transactions (2010, Vol. 116, Part 2). For personal use only. Additional reproduction, distribution, or transmission in either print or digital form is not permitted without ASHRAEs prior written per
36、mission.2010 ASHRAE 655Table 2. Potential Issues with GHG CalculationsIssue Description CommentUnits English or metric (1 metric ton = 1.1 ton). Adds confusion.Carbon or CO21 kg of CO2contains 0.273 kg of carbon. CO2is the more common metric.CO2or CO2eCO2eincludes other GHGs according to their GWP.5
37、0% effect depending on fuel and refrigerant use (Deru and Torcellini 2007).Generated vs. deliveredDelivered electricity includes the transmission and distri-bution (T Liu and Zobian 2002).Precombustion effects (upstream emissions)Emissions from extracting, processing, and transporting fuel to the po
38、wer plant or building, construction of power plants, and manufacturing of power generating devices. 2%-25% effect for fossil fuels depending on fuel type, source, and location; 70%-90% effect for RE; and 50%-80% for nuclear (Deru and Torcellini 2007; Weisser 2007).Downstream emissionsEmissions from
39、handling and storing waste and decom-missioning of power plants. 2% effect for fossil fuels, and 6%-16% for nuclear (Weisser 2007).National vs. regional dataThe electricity grid is divided into interconnects and regions and energy flows between regions to balance the load.Selection of geographical r
40、egion for emission factors can have large effect on the results.Time of useMost emissions data are annual averages; however, real-time emissions vary continuously.The mix of generators varies continuously, and it is impossible to know the source of the electricity and the associated emissions.Averag
41、e vs. marginal dataMarginal plants operate to meet changes in demand bring-ing about continuous variations in the emissions. Should emission offsets be calculated with marginal data and carbon footprints calculated with average data?Temporal adjustmentsMost emissions data are historical from a speci
42、fic period (e.g., 2007), which is probably different from the analysis period. Emissions vary from year to year, and the age of the data should be accounted for in analyses.Weather adjustmentsVariations in the weather affect the demand for electricity, the performance of the power generators, availa
43、ble hydro power, and T however, the mix and operation of gener-ators supplying the grid vary continuously, making it difficult to know the exact source of delivered electricity. The major issues associated with determining emissions at this level are discussed below.Regional Effects. North America i
44、s divided into three main electricity grids (or interconnects) and subdivided into eight regions, which are managed by the North American Electric Reliability Corporation (NERC) (see Figure 4) (NERC 2009). The electricity generation resources are managed by balancing authorities of approximately 150
45、 control areas across the United States and Canada. Electricity is transferred between control areas continuously to balance supply with demand, but relatively little energy is transferred between the three main interconnects. The U.S. Environmen-tal Protection Agency (EPA) has defined 26 subregions
46、 for the Emissions and Generation Resource Integrated Database (eGRID) in an attempt to define logical boundaries for rela-tively self-contained power generation areas (see Figure 5). There are at least seven levels of electricity generation aggregation from largest to smallest: national, interconne
47、ct, NERC region, eGRID subregion, state, control area, and util-ity. The choice of which level to use depends on the objectives of the analysis and can have large impacts on the results; there-fore, it is important to understand the differences when select-ing emission factors. The larger the area,
48、the more inclusive the analysis, but there is a loss of information from averaging, which removes local effects (mix of generators, operation, weather, etc.). At the lowest level, many electricity transfer effects are missed. State boundaries are typically unreliable, because the electricity grid bo
49、undaries rarely follow geopolit-ical boundaries. Variability with the level of aggregation in CO2emission factors on an annual basis for electricity generation is illus-trated in Figure 6, which shows the CO2emission factors from eGRID2007 v1.1 for several cities at the state, eGRID subre-gion, NERC region, and national levels. These emission factors are based on the electricity generated within the bound-aries and do not include any imported electricity. For Miami, all four emission factors are similar, but there can be large vari-ations in the emission factors for other locations
