UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Canadian energy use and greenhouse gas emissions in the 1980’s and 1990’s : decomposition of changes,… Herbert, Deborah Marie 1999

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

Item Metadata

Download

Media
831-ubc_1999-0265.pdf [ 50.1MB ]
Metadata
JSON: 831-1.0089005.json
JSON-LD: 831-1.0089005-ld.json
RDF/XML (Pretty): 831-1.0089005-rdf.xml
RDF/JSON: 831-1.0089005-rdf.json
Turtle: 831-1.0089005-turtle.txt
N-Triples: 831-1.0089005-rdf-ntriples.txt
Original Record: 831-1.0089005-source.json
Full Text
831-1.0089005-fulltext.txt
Citation
831-1.0089005.ris

Full Text

CANADIAN ENERGY USE AND GREENHOUSE GAS EMISSIONS IN THE 1980*S AND 1990*S: DECOMPOSITION OF CHANGES, EXTRAPOLATION OF TRENDS AND COMPARISON TO OTHER OECD COUNTRIES  by DEBORAH MARIE HERBERT B.A., Carleton University, 1994  A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS in THE FACULTY OF GRADUATE STUDIES (Department of Resource Management and Environmental Studies)  We accept this thesis as conforrning to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA April 1999 © Deborah Marie Herbert, 1999  In  presenting  degree  at  this  the  thesis  in  partial  fulfilment  of  University  of  British  Columbia,  I agree  freely available for copying  of  department publication  this or of  reference  thesis by  this  for  his thesis  and study. scholarly  or for  her  The University of British Columbia Vancouver, Canada  DE-6 (2/88)  purposes  gain shall  requirements that  agree that  may  representatives.  financial  permission.  I further  the  It not  be is  the  permission  granted  allowed  an  advanced  Library shall  by  understood be  for  for  the that  without  make  it  extensive  head  of  my  copying  or  my  written  Abstract This thesis examines energy use in Canada in the early to mid-1980's to mid-1990's to investigate what factors caused energy use and greenhouse gas emissions to rise. Trends from this period in energy use and fuel mix are also projected to the years 2000 and 2010. In addition, Canada is compared to 12 other OECD countries to determine whether differences in climate, geography and industrial structure account for differences in absolute and per capita energy use between Canada and these countries.  Changes in activity were the main drivers of the increases in energy use and greenhouse gas emissions in the 1980's and 1990's. This influence was partially offset by declines in energy intensity. Structural changes tended to have a less profound impact. Based on trendsfromthis period, both energy use and greenhouse gas emissions will continue rising. More positively, there already are trends towards less greenhouse gas-intensive fuels in some sectors.  Climate, geography and industrial structure do not account for differences in per capita energy use between Canada and other industrialized countries. The one exception is the United States. This implies that, with the exception of the U.S., Canada is relatively less energy efficient than other industrialized countries.  ii  Table of Contents Abstract Acknowledgements List of Tables List of Figures  ii . W Yfj XI  Chapter 1: Introduction and Context 1.1 Introduction 1.2 Future Climate Change 1.3 Greenhouse Gases 1.4 Canada's contribution to emissions 1.5 Canada's International Commitments: The Framework Convention on Climate Change 1.6 Canada's Mitigation Policies 1.7 Conclusion and Introduction to Thesis Research  1 1 3 4 7  Chapter 2: Methodology 2.1 Introduction to Decomposition Approach 2.2 Derivation of Decomposition Expression 2.2.1 Subsectoral Activities and the Indicators Pyramid .... 2.2.2 Derivation of Laspeyres Indices 2.2.3 Partial Decompositions 2.2.4 Incorporating Weather Changes 2.2.5 Greenhouse Gas Emission Decompositions 2.3 Alternatives to the Laspeyres Index 2.4 Limitations of this Approach 2.5 Method for OECD Comparisons 2.5.1 Climate 2.5.2 Geography 2.5.3 Industrial Structure 2.5.4 Overall Comparison 2.6 Data Sources  24 24 25 26 27 31 32 37 45 48 51 52 53 55 55 56  Chapter 3: Decomposition of Energy Use Changes 3.1 Overall Results 3.2 Industrial Sector 3.2.1 Industry 1973-1983 3.2.1.1 Activity Changes 3.2.1.2 Digression on Prices 3.2.1.3 Intensity Changes 3.2.1.4 Structure Changes 3.2.2 Industry 1984-1996 3.2.2.1 Activity Changes 3.2.2.2 Intensity Changes 3.2.2.3 Process-level Intensity Changes 3.2.2.4 Structure Changes fi i  12 17 20  57 57 59 59 60 61 63 65 66 67 68 69 71  3.3 Freight Transport 3.3.1 Activity Changes 3.3.2 Structural Changes 3.3.3 Intensity Changes 3.3.4 Freight Truck Intensity Changes 3.4 Passenger Transport 3.4.1 Activity Changes 3.4.2 Structural Changes 3.4.3 Intensity Changes 3.4.4 Car Intensity Changes 3.5 Service Sector 3.5.1 Partial Decompositions By Building Type 3.5.2 Activity Changes 3.5.3 Structural Changes 3.5.4 Intensity Changes 3.6 Residential Sector 3.6.1 Activity Changes 3.6.2 Intensity Changes 3.6.2.1 End-Use Demand 3.6.2.2 Efficiency of Space Heating Equipment 3.6.2.3 Heated Floor Area Per Household 3.6.2.4 Appliance and Air Conditioner Penetration ... 3.6.2.5 Number of Households and Distribution Across Housing Types 3.6.2.6 Summary 3.7 Conclusion Chapter 4: Decomposition of Greenhouse Gas Emissions 4.1 Overall Results 4.1.1 Changes in Emissions 4.1.2 Emission Decompositions 4.1.3 Fuel Mix Changes 4.2 Industry, 1984 to 1996 4.2.1 Changes in Emissions 4.2.2 Decomposition Results 4.2.3 Fuel Mix Changes 4.3 Freight Transport, 1984 to 1996 4.3.1 Changes in Emissions 4.3.2 Emission Decompositions 4.3.3 Fuel Mix Changes 4.4 Passenger Transport, 1984 to 1996 4.4.1 Changes in Emissions 4.4.2 Decomposition Results 4.4.3 Fuel Mix Changes  iv  72 72 73 74 75 77 77 78 79 79 81 82 83 83 84 86 87 88 88 89 90 90 91 92 92 96 97 97 99 101 101 101 102 103 103 103 104 105 106 106 107 108  4.5 Service Sector, 1981 to 1995 4.5.1 Changes in Emissions 4.5.2 Decomposition Results 4.5.3 Fuel Mix Changes 4.6 Residential Sector, 1981 to 1995 4.6.1 Changes in Emissions 4.6.2 Decomposition Results 4.6.3 Fuel Mix Changes 4.7 Summary and Conclusion Chapter 5: Extrapolation of Historical Trends in Energy Use and Greenhouse Gas Emissions 5.1 Canada Overall 5.1.1 Energy Use Extrapolations 5.1.2 Projected Greenhouse Gas Emissions, with Changing Energy Use 5.1.3 Projected Greenhouse Gas Emissions, with Frozen Energy Use 5.2 Industrial Sector 5.2.1 Energy Use Extrapolations 5.2.2 Projected Greenhouse Gas Emissions, with Changing Energy Use 5.2.3 Projected Greenhouse Gas Emissions, with Frozen Energy Use 5.3 Freight Transport 5.3.1 Energy Use Extrapolations 5.3.2 Projected Greenhouse Gas Emissions, with Changing Energy Use 5.3.3 Projected Greenhouse Gas Emissions, with Frozen Energy Use 5.4 Passenger Transport 5.4.1 Energy Use Extrapolations 5.4.2 Projected Greenhouse Gas Emissions, with Changing Energy Use 5.4.3 Projected Greenhouse Gas Emissions, with Frozen Energy Use 5.5 Service Sector 5.5.1 Energy Use Extrapolations 5.5.2 Projected Greenhouse Gas Emissions, with Changing Energy Use 5.5.3 Projected Greenhouse Gas Emissions, with Frozen Energy Use  v  109 109 109 Ill Ill Ill 112 113 114  117 118 118 120 122 123 123 126 127 129 129 131 132 133 133 135 137 138 138 139 140  5.6 Residential Sector 5.6.1 Energy Use Extrapolations 5.6.2 Projected Greenhouse Gas Emissions, with Changing Energy Use 5.6.3 Projected Greenhouse Gas Emissions, with Frozen Energy Use 5.7 Summary and Conclusions 5.7.1 Energy Use Projections — Summary 5.7.2 Greenhouse Gas Emissions Projected with Changing Energy Use — Summary 5.7.2 Greenhouse Gas Emissions Projected with Frozen Energy Use — Summary 5.7.4 Conclusions  142 142  Chapter 6: Comparison of Canada to Other Industrialized Countries 6.1 Climate 6.2 Geography 6.3 Industrial Structure 6.4 Total Energy Use 6.5 Summary and Conclusions  151 156 158 161 162 165  Chapter 7: Conclusion 7.1 Which factors "explain" the changes in energy use and greenhouse gas emissions that have been observed since the early 1980s in Canada? 7.2 Is Canada "wasteful" of energy compared to other industrialized nations? 7.3 What are the future prospects for reducing energy use and related greenhouse gas emissions (focusing on CO2 emissions) in Canada, given current and expected climate change policy and the results of the analyses to address question 1? 7.4 Summary and Conclusions  168  References  176  vi  143 144 146 146 148 149 150  169 171  172 173  List of Tables  Table 1.1: Average Annual C 0 Emissions And Sinks, 1980-89 (GtC/Yr) Table 1.2: C 0 Emissions Corresponding To Stabilization At Different Atmospheric C 0 Concentrations, LPCC Estimates Table 1.3: Summary Of Federal Government Initiatives Announced December, 1996 Table 3.0 Sectoral Energy Use Shares, 1984-1995 Table 3.1 Canada, 1995 Overall Decomposition Results (Base Year 1984) Table 3.2 Manufacturing + Mining, 1983 Decomposition Of Energy Use (Base Year 1973) Table 3.4 Mining + Manufacturing Sectors Change In Energy Intensity 1973 To 1983 Table 3.5 Comparison Of Energy Intensity Decomposition Results Table 3.6 Manufacturing + Mining, 1973-1983 Change In Subsector Activity Shares And Average Energy Intensity Table 3.7 Manufacturing, 1973-1983 Change In Subsector Activity Shares And Average Energy Intensity Table 3.8 Industrial Sector, 1996 Decomposition Of Energy Use (Base Year 1984) Table 3.9 Industry Change In Energy Intensity 1984 To 1996 Table 3.10 Industry, 1984-1996 Change In Subsector Activity Shares And Average Energy Intensity Table 3.11 Manufacturing, 1984-1996 Change In Subsector Activity Shares And Average Energy Intensity Table 3.12 Freight Transport, 1996 Decomposition Of Energy Use (Base Year 1984) Table 3.13 Freight Transport Mode Shares Table 3.14 Freight Transport Energy Shares Table 3.15 Freight Transport Change In Energy Intensity 1984 To 1996 And Average Mode Shares Table 3.16 Freight Trucks, 1984-1995 Truck Stock, Fuel Efficiency, Kilometres Per Truck and Tonne-Km Per Truck Table 3.17 Passenger Transport, 1996 Decomposition Of Energy Use (Base Year 1984) Table 3.18 Passenger Transport, 1984-1996 Mode Shares and Average Energy Intensity Table 3.19 Passenger Transport, 1984-1996 Energy Intensity and Average Mode Shares Table 3.20 Passenger Transport, 1984-1995 Car Energy Use, Stock, Fuel Efficiency, Distance Travelled And Passenger-Kilometres Table 3.21 Service Sector, 1995 Decomposition Results (Base Year 1981) Table 3.22 Service Sector, 1995 Energy Decomposition By Building Type 2  Page 189  2  2  vu  189 190 201 201 201 201 202 202 202 203 203 204 204 205 205 205 205 206 207 207 208 209 210 211  Table 3.23 Service Sector, 1981-1995 Floor Area Shares and Average Energy Intensity By Building Type Table 3.24 Service Sector, 1981-1995 Energy Intensity and Average Floor Area Shares By Building Type Table 3.25 Service Sector, 1981-1995 Energy Use, Floor Area and Energy Intensity By Building Type Table 3.26 Residential Sector, 1995 Decomposition Results (Base Year 1981) Table 3.27 Residential Sector, 1981-1995 Households, Population and Household Size Table 3.28 Residential Sector, 1981-1995 Energy Use And Intensity By End-Use Table 3.29 Residential Space Heating, 1981-1995 Share Of Households and Average Energy Intensity By Equipment Type Table 3.30 Residential Space Heating, 1995 Decomposition Results (Base Year 1981) Table 3.31 Residential Sector, 1981-1995 Households and Heated Floor Area Table 3.32 Residential Sector, 1981-1995 Appliance Penetration and Unit Energy Consumption Table 3.33 Residential Sector, 1981-1995 Distribution Of Households By Building Type Table 4.1 Greenhouse Gas Emissions And Energy Use, 1984-1995 Table 4.2 Canada, 1995 GHG Decomposition Results (Base Year 1984) Table 4.3 Canada, 1995 GHG Decomposition Results (Base Year 1990) Table 4.4 Canada, 1984-1995 Fuel Shares Table 4.5 Greenhouse Gas Emission Factors Table 4.6 Industry, 1996 GHG Decomposition Results (Base Year 1984) Table 4.7 Industry, 1996 GHG Decomposition Results (Base Year 1990) Table 4.8 Industry, 1984-1996 Fuel Shares Table 4.9 Freight Transport, 1996 GHG Decomposition Results (Base Year 1984) Table 4.10 Freight Transport, 1996 GHG Decomposition Results (Base Year 1990) Table 4.11 Freight Transport, 1984-1996 Fuel Shares Table 4.12 Passenger Transport, 1996 GHG Decomposition Results (Base Year 1984) Table 4.13 Passenger Transport, 1996 GHG Decomposition Results (Base Year 1990) Table 4.14 Passenger Transport, 1984-1996 Fuel Shares Table 4.15 Service Sector, 1995 GHG Decomposition Results (Base Year 1981) Table 4.16 Service Sector, 1995 GHG Decomposition Results (Base Year 1990) Table 4.17 Service Sector, 1981-1995 Fuel Shares  viii  212 212 213 214 214 215 216 216 216 217 217 218 219 219 219 220 220 220 221 221 221 222 222 222 223 223 223 224  Table 4.18 Residential Sector, 1995 GHG Decomposition Results (Base Year 1981) Table 4.19 Residential Sector, 1995 GHG Decomposition Results (Base Year 1990) Table 4.20 Residential Sector, 1981 -1995 Fuel Shares Table 5.1 Canada Energy Use Extrapolations Table 5.2 Canada Greenhouse Gas Emissions In 2000 And 2010 Table 5.3 Canada Greenhouse Gas Emissions In 2000 And 2010 No Change In Energy Use From 1995 Level Table 5.4 Industrial Energy Use Extrapolations Table 5.5 Industrial Energy And Activity Annual Average Rates Of Change Table 5.6 Industrial Energy Intensity Annual Average Rates Of Change Table 5.7 Industrial Subsector Shares Annual Average Rates Of Change Table 5.8 Manufacturing Energy Use Extrapolations Table 5.9 Manufacturing Subsector Shares Annual Average Rates Of Change Table 5.10 Industrial Greenhouse Gas Emissions In 2000 And 2010 Table 5.11 Manufacturing Greenhouse Gas Emissions In 2000 And 2010 Table 5.12 Industrial Greenhouse Gas Emissions In 2000 And 2010 No Change In Energy Use From 1996 Level Table 5.13 Manufacturing Greenhouse Gas Emissions In 2000 And 2010 No Change In Energy Use From 1996 Level Table 5.14 Freight Transport Energy Use Extrapolations Table 5.15 Freight Transport Energy And Activity Annual Average Rates Of Change Table 5.16 Freight Transport Mode Shares Annual Average Rates Of Change Table 5.17 Freight Transport Energy Intensity Annual Average Rates Of Change Table 5.18 Freight Transport Greenhouse Gas Emissions In 2000 And 2010 Table 5.19 Freight Transport Fuel Shares Table 5.20 Freight Transport Greenhouse Gas Emissions In 2000 And 2010 No Change In Energy Use From 1996 Level Table 5.21passenger Transport Energy Use Extrapolations Table 5.22 Passenger Transport Energy And Activity Annual Average Rates Of Change Table 5.23 Passenger Transport Mode Shares Annual Average Rates Of Of Change Table 5.24 Passenger Transport Energy Intensity Annual Average Rates Of Of Change Table 5.25 Passenger Transport Greenhouse Gas Emissions In 2000 And 2010 Table 5.26 Passenger Transport Fuel Shares Table 5.27 Passenger Transport Greenhouse Gas Emissions In 2000 And 2010 No Change In Energy Use From 1996 Level Table 5.28 Service Sector Energy Use Extrapolations Table 5.29 Service Sector Energy And Activity Annual Average Rates Of Change iX  224  224 224 225 225 225 226 227 227 227 228 228 228 229 229 229 23 0 231 231 231 232 232 232 233 234 234 234 235 235 235 236 237  Table 5.30 Service Building Floor Area Shares Annual Average Rates OfChange Table 5.31 Service Building Energy Intensity Annual Average Rates OfChange Table 5.32 Service Sector Greenhouse Gas Emissions In 2000 And 2010 Table 5.33 Service Sector Greenhouse Gas Emissions In 2000 And 2010 No Change In Energy Use From 1995 Level Table 5.34 Residential Sector Energy Use Extrapolations Table 5.35 Residential End-Use Energy Intensity Annual Average Rates Of Change Table 5.36 Residential Sector Energy And Activity Annual Average Rates OfChange Table 5.37 Residential Greenhouse Gas Emissions In 2000 And 2010 Table 5.38 Residential Greenhouse Gas Emissions In 2000 And 2010 No Change In Energy Use From 1995 Level Table 6.1 Canada Energy Use With Adjustments For Climate, Geography & Industrial Structure Table 6.2 Comparison Of Per Capita Energy Use (GJ/Person) Table 6.3 Components Of Per Capita Energy Use And Implied Energy Use/Emission Reduction  237 237 238 238 238 239 239 240 240 241 242 243  List of Figures Page Figure Figure Figure Figure  1.1: 1994 Greenhouse Gas Emissions - Canada 1.2: Canadian C02 Emissions by Sector, 1993 1.3: C02 EmissionsfromFossil Fuel Use, Canada and Global 1.4: Annual Percentage Change in C02 Emissions: Canada and Global Figure 1.5: Per Capita C02 EmissionsfromEnergy Use, 1980 Figure 1.6: Per Capita C02 EmissionsfromEnergy Use, 1993 Figure 1.7: Average Annual Growth in C02 Emissions, 1980-1993 Figure 1.8: Final Energy Consumption per Unit of GDP - 1980 Figure 1.9: Final Energy Consumption per Unit GDP - 1993 Figure 1.10: Energy per Unit of GDP, G7 Countries Figure 1.11: Canadian C02 EmissionsfromFossil Fuel Use and Real GDP Figure 1.12: Annual Percent Change in C02 Emissions and Real GDP, Canada 1958-1993 Figure 1.13: Difference in Annual Percentage Rates of Change of C02 Emissions and Real GDP, Canada, 1958-1993 Figure 3.0 Sectoral Shares in Energy Use 1995 Figure 3.1 Canada Decomposition Results (1984 Base Year) Figure 3.2 Manufacturing + Mining Decomposition Results (1973 Base Year) Figure 3.3 Manufacturing + Mining Real GDP, 1973-1983 Figure 3.4 Manufacturing Sector, 1973-1983 Capacity Utilization Figure 3.5 Average World Price of Crude Oil Figure 3.7 Manufacturing Sector, 1973-1983 Fuel Costs as a Percentage of Total Costs Figure 3.8 Manufacturing + Mining, 1973-1983 Energy Intensities Figure 3.9 Industrial Sector, 1973-1983 Energy Intensities Figure 3.10 Manufacturing + Mining, 1973-1983 Activity Shares Figure 3.11 Manufacturing Decomposition Results (1973 Base Year) Figure 3.12 Industrial Sector Decomposition Results (1984 Base Year) Figure 3.13 Manufacturing Decomposition Results (1984 Base Year) Figure 3.14 Industrial Sector Real GDP, 1984-1996 Figure 3.15 Industrial Sector, 1984-1996 Energy Intensities Figure 3.16 Industrial Sector, 1984-1996 Energy Intensities Figure 3.17 Freight Transport Decomposition Results (1984 Base Year) Figure 3.18 Freight Transport Tonne-kilometres and Goods Production Real GDP, 1984-1996 Figure 3.19 Freight Transport, 1984-1996 Energy Intensities Figure 3.20 Marine Freight Transport, 1984-1996 Energy Use and Activity Figure 3.21 Freight Trucks Less Than 4546 kg, 1984-1996 Energy Use and Activity  248 249 250 251 252 252 253 254 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279  Figure 3.22 Freight Trucks Between 4546 and 15000 kg Energy Use and Activity, 1984-1996 Figure 3.23 Freight Trucks Greater Than 15000 kg Energy Use and Activity, 1984-1996 Figure 3.24 Rail Freight Transport, 1984-1996 Energy Use and Activity Figure 3.25 Passenger Transport Decomposition Results (1984 Base Year) Figure 3.26 Passenger Transport, 1984-1996 Passenger-kilometres and Population Figure 3.27 Passenger Transport Mode Shares Figure 3.28 Passenger Transport, 1984-1996 Energy Intensity by Mode Figure 3.29 Service Sector Decomposition Results (1981 Base Year) Figure 3.30 Service Sector, 1981-1995 Floor Area versus Population, Services GDP & Services Labour Force Figure 3.31 Service Sector Floor Area Shares Figure 3.32 Schools, 1981-1995 Share in Service Sector Floor Area Figure 3.33 Health-Related Buildings, 1981-1995 Share in Service Sector Floor Area Figure 3.34 Religious Buildings, 1981-1995 Share in Service Sector Floor Area Figure 3.35 Other Institutional Buildings, 1981-1995 Share in Service Sector Floor Area Figure 3.36 Office Buildings, 1981-1995 Share in Service Sector Floor Area Figure 3.37 Retail Buildings, 1981-1995 Share in Service Sector Floor Area Figure 3.38 Hotels and Restaurants, 1981-1995 Share in Service Sector Floor Area Figure 3.39 Recreation-Related Buildings, 1981-1995 Share in Service Sector Floor Area Figure 3.40 Warehouses, 1981-1995 Share in Service Sector Floor Area Figure 3.41 Service Sector, 1981-1995 Energy Intensity by Building Type Figure 3.42 Residential Sector Decomposition Results (1981 Base Year) Figure 3.43 Residential Sector, 1981-1995 Households, Persons per Household and Population Figure 3.44 Residential Space Heating, 1981-1995 Equipment Shares Figure 3.45 Residential Space Heating Decomposition Results (1981 Base Year) Figure 3.46 Residential Sector, 1981-1995 Appliance and Air Conditioner Penetration Rates Figure 4.1 Canada, 1984-1995 Greenhouse Gas Emissions and Energy Use Figure 4.2 Carbon Dioxide Emissions by Sector Figure 4.3 Methane Emissions by Sector Figure 4.4 Nitrous Oxide Emissions by Sector Figure 4.5 Canada, 1984-1995 C02 Decomposition Results (1984 Base Year) Figure 4.6 Canada, 1984-1995 CH4 Decomposition Results (1984 Base Year) Figure 4.7 Canada, 1984-1995 N20 Decomposition Results (1984 Base Year) Figure 4.8 Canada, 1984-1995 C02 Decomposition Results (1990 Base Year) xii  280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 398 399 300 301 302 303 304 305 306 307 308 309 310 311 312  Figure 4.9 Canada, 1984-1995 CH4 Decomposition Results (1990 Base Year) Figure 4.10 Canada, 1984-1995 N20 Decomposition Results (1990 Base Year) Figure 4.11 Canada Total Energy Use by Fuel Figure 4.12 Industrial Sector Greenhouse Gas Emissions Figure 4.13 Industrial Sector C02 Decomposition Results (1984 Base Year) Figure 4.14 Industrial Sector CH4 Decomposition Results (1984 Base Year) Figure 4.15 Industrial Sector N20 Decomposition Results (1984 Base Year) Figure 4.16 Industry, 1984-1996 C02 Decomposition Results (1990 Base Year) Figure 4.17 Industry, 1984-1996 CH4 Decomposition Results (1990 Base Year) Figure 4.18 Industry, 1984-1996 N20 Decomposition Results (1990 Base Year) Figure 4.19 Industrial Energy Use by Fuel Figure 4.20 Industrial Sector 1984-1996 Energy Use Shares Figure 4.21 Freight Transport Greenhouse Gas Emissions Figure 4.22 Freight Transport C02 Decomposition Results (1984 Base Year) Figure 4.23 Freight Transport CH4 Decomposition Results (1984 Base Year) Figure 4.24 Freight Transport N20 Decomposition Results (1984 Base Year) Figure 4.25 Freight Transport, 1984-1995 C02 Decomposition Results (1990 Base Year) Figure 4.26 Freight Transport, 1984-1996 CH4 Decomposition Results (1990 Base Year) Figure 4.27 Freight Transport, 1984-1996 N20 Decomposition Results (1990 Base Year) Figure 4.28 Freight Transport Energy Use by Fuel Figure 4.29 Freight Transport Energy Use by Fuel Figure 4.30 Passenger Transport Greenhouse Gas Emissions Figure 4.31 Passenger Transport C02 Decomposition Results (1984 Base Year) Figure 4.32 Passenger Transport CH4 Decomposition Results (1984 Base Year) Figure 4.33 Passenger Transport N20 Decomposition Results (1984 Base Year) Figure 4.34 Passenger Transport, 1984-1996 C02 Decomposition Results (1990 Base Year) Figure 4.35 Passenger Transport, 1984-1996 CH4 Decomposition Results (1990 Base Year) Figure 4.36 Passenger Transport, 1984-1996 N20 Decomposition Results (1990 Base Year) Figure 4.37 Passenger Transport Energy Use by Fuel Figure 4.38 Passenger Transport Energy Use by Fuel Figure 4.39 Service Sector Greenhouse Gas Emissions Figure 4.40 Service Sector C02 Decomposition Results (1981 Base Year) Figure 4.41 Service Sector CH4 Decomposition Results (1981 Base Year) Figure 4.42 Service Sector N20 Decomposition Results (1981 Base Year) Figure 4.43 Service Sector C02 Decomposition Results (1990 Base Year) Figure 4.44 Service Sector CH4 Decomposition Results (1990 Base Year) Figure 4.45 Service Sector N20 Decomposition Results (1990 Base Year) Figure 4.46 Service Sector Energy Use by Fuel Figure 4.47 Service Sector Energy Use by Fuel Figure 4.48 Residential Greenhouse Gas Emissions xiif  313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352  Figure 4.49 Residential Sector C02 Decomposition Results (1981 Base Year) Figure 4.50 Residential Sector CH4 Decomposition Results (1981 Base Year) Figure 4.51 Residential Sector N20 Decomposition Results (1981 Base Year) Figure 4.52 Residential Sector C02 Decomposition Results (1990 Base Year) Figure 4.53 Residential Sector CH4 Decomposition Results (1990 Base Year) Figure 4.54 Residential Sector N20 Decomposition Results (1990 Base Year) Figure 4.55 Residential Energy Use by Fuel Figure 4.56 Residential Sector Energy Use by Fuel Figure 5.1 Canada Carbon Dioxide Emissions Figure 5.2 Canada Methane Emissions Figure 5.3 Canada Nitrous Oxide Emissions Figure 5.4 Canada Carbon Dioxide Emissions No Energy Use Change from 1995 Level Figure 5.5 Canada Methane Emissions No Energy Use Change from 1995 Level Figure 5.6 Canada Nitrous Oxide Emissions No Energy Use Change from 1995 Level Figure 5.7 Industrial Sector Carbon Dioxide Emissions Figure 5.8 Industrial Sector Methane Emissions Figure 5.9 Industrial Sector Nitrous Oxide Emissions Figure 5.10 Industry Carbon Dioxide Emissions No Energy Use Change from 1995 Level Figure 5.11 Industry Methane Emissions No Energy Use Change from 1995 Level Figure 5.12 Industry Nitrous Oxide Emissions No Energy Use Change from 1995 Level Figure 5.13 Manufacturing Carbon Dioxide Emissions No Energy Use Change from 1995 Level Figure 5.14 Manufacturing Methane Emissions No Energy Use Change from 1995 Level Figure 5.15 Manufacturing Nitrous Oxide Emissions No Energy Use Change from 1996 Level Figure 5.16 Freight Transport Carbon Dioxide Emissions Figure 5.17 Freight Transport Methane Emissions Figure 5.18 Freight Transport Nitrous Oxide Emissions Figure 5.19 Freight Transport Carbon Dioxide Emissions No Energy Use Change from 1995 Level Figure 5.20 Freight Transport Methane Emissions No Energy Use Change from 1995 Level Figure 5.21 Freight Transport Nitrous Oxide Emissions No Energy Use Change from 1995 Level Figure 5.22 Passenger Transport Energy Use by Fuel Figure 5.23 Passenger Transport Carbon Dioxide Emissions Figure 5.24 Passenger Transport Methane Emissions Figure 5.25 Passenger Transport Nitrous Oxide Emissions Figure 5.26 Passenger Transport Carbon Dioxide Emissions xiv  353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385  No Energy Use Change from 1995 Level  Figure 5.27 Passenger Transport Methane Emissions No Energy Use Change from 1995 Level Figure 5.28 Passenger Transport Nitrous Oxide Emissions No Energy Use Change from 1995 Level Figure 5.29 Service Sector Carbon Dioxide Emissions Figure 5.30 Service Sector Methane Emissions Figure 5.31 Service Sector Nitrous Oxide Emissions Figure 5.32 Service Sector Carbon Dioxide Emissions No Energy Use Change from 1995 Level Figure 5.33 Service Sector Methane Emissions No Energy Use Change from 1995 Level Figure 5.34 Service Sector Nitrous Oxide Emissions No Energy Use Change from 1995 Level Figure 5.35 Residential Sector Carbon Dioxide Emissions Figure 5.36 Residential Sector Methane Emissions Figure 5.37 Residential Sector Nitrous Oxide Emissions Figure 5.38 Residential Sector Carbon Dioxide Emissions No Energy Use Change from 1995 Level Figure 5.39 Residential Sector Methane Emissions No Energy Use Change from 1995 Level Figure 5.40 Residential Sector Nitrous Oxide Emissions No Energy Use Change from 1995 Level Figure 6.1 Residential and Service Sector Energy Use Canada Actual versus Canada with OECD Heating Degree-Days Figure 6.2 Transportation Energy Use Canada Actual versus Canada with OECD Activity Levels Figure 6.3 Industrial Energy Use Canada Actual versus Canada with OECD Subsector Shares Figure 6.4 Total Energy Use Canada Actual versus Canada with OECD Values Figure A. 1 Residential Sector, 1981-1995 Comparison of Normalized Decomposition Results (Base Year 1990) Figure A.2 Residential Sector, 1981-1995 Comparison of Normalized Decomposition Results (Base Year 1990)  xv  386  387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406  Acknowledgements  I wish to extend most sincere thanks to my thesis supervisor, John Robinson, who provided excellent advice, support and enthusiasm. He has helped me raise the rigour of my work considerably. I am also grateful to my thesis committee (Michael Goldberg, Ray Meadowcroft and William Rees) who were encouraging and helpful. Lee Schipper and Ralph Torrie provided data and excellent advice.  I would like to especially thank Les Lavkulich, Chair of RMES, for his tireless efforts on my behalf. Many times he has helped me, sometimes in spite of myself (!), and I will always remember, and be grateful for, his dedication to his students.  xvi  Chapter 1 Introduction and Context  1.1 Introduction This thesis investigates energy use in Canada over the last three decades in order to address three questions:  1. Which factors "explain" the changes in energy use and greenhouse gas emissions that have been observed since the early 1980s in Canada? 2. Is Canada "wasteful" of energy compared to other industrialized nations? 3. What are the future prospects for reducing energy use and related greenhouse gas emissions (focusing on CO2 emissions) in Canada, given current and expected climate change policy and the results of the analyses to address question 1?  Question 1 will be addressed by decomposing changes in sectoral energy use into three components: changes in energy intensity, changes in the structure of the sector and changes in the level of sectoral activity. In the residential and service sectors, the influence of changes in weather will also be isolated. This analysis will provide insights into what has driven changes in energy use in the recent past and indicate in broad terms the areas that should be targeted by greenhouse gas mitigation policy in the future. This analysis and its results are discussed in Chapter 3 of this thesis. Question 1 will also be addressed by decomposing changes in sectoral greenhouse gas (GHG) emissions into four components: changes in fuel mix, changes in energy intensity, changes in the structure of the sector and changes in the level of sectoral activity. Weather will also be  1  considered in the residential and service sectors. This analysis and its results are discussed in Chapter 4 of this thesis.  Question 2 will be addressed by adjusting sectoral energy use in Canada and other OECD countries to account for non-energy "exogenous" factors and then comparing corrected energy use across countries. These non-energy "exogenous" factors are climate, geography and industrial structure and have often been discussed in official Canadian government documents as explanations for Canada's relatively high rate of energy use and greenhouse gas emissions. This comparative analysis will provide estimates of the degree to which these exogenous factors "explain" differences in energy use between Canada and other OECD countries. It is discussed in Chapter 6.  Question 3 is discussed in Chapter 5. It will be addressed by drawing on the results of the analyses of chapters 3 and 4 as well as on the review of Canadian climate change policy which is provided below in this chapter.  The remainder of this chapter sets the context for the analyses in subsequent chapters by reviewing the threat of climate change, the levels and sources of greenhouse gas emissions, Canada's contribution to world greenhouse gas emissions and rank compared to other OECD countries with respect to COa emissions and energy-intensity, and Canada's international commitments and national policy with respect to climate change and greenhouse gas emissions.  2  1.2 Future Climate Change According to the world's scientific authority on climate change, the Intergovernmental Panel on Climate Change (IPCC), human activity, especially fossil fuel combustion, has already had a "discernible influence on climate" (IPCC, 1995a). Theyfindthat  [g]lobal mean surface temperature has increased by between about 0.3 and 0.6°C since the late 19th century, a change that is unlikely to be entirely natural in origin ... Global sea level has risen by between 10 and 25 cm over the past 100 years and much of the rise may be related to the increase in global mean temperature. (IPCC, 1995a: Section 2.4)  According to the best estimates of the IPCC , by the year 2100 global mean surface air 1  temperature could be 2°C higher and sea level couldriseby 50 cm (Bolin, 1996). 2  Temperature and sea level will continue torisebeyond the year 2100 (Bolin, 1996) and, most importantly, "the average rate of wanning would probably be greater than any seen in the last 10,000 years" (IPCC, 1995a: Section 2.7).  This level of temperature increase and sea level rise will have significant, probably largely adverse impacts . The expected rate of change of climate exceeds the rate at 3  which forest ecosystems can adjust, implying significant changes in these ecosystems. Deserts are expected to become hotter, but not much wetter. An intensification of the global hydrological cycle is expected, with major impacts on regional water resources. In  ' T h i s forecast i s based o n the I P C C ' s mid-range scenario for future greenhouse gas emissions, Scenario IS92a ( B o l i n , 1996), w h i c h involves intermediate assumptions about population a n d economic growth, compared to the other five scenarios developed b y the I P C C (Houghton et al, 1995: 261). It should be noted that these estimates are highly uncertain. T h e estimate for global mean temperature rise is actually 1.0 - 3.5°C, w i t h a best estimate o f 2 . 0 ° C , a n d the estimate for sea level rise is 15 - 95 em, w i t h a best estimate o f 50 c m ( B o l i n , 1996). 2  3  particular, the magnitude and timing of runoff and the intensity of floods and droughts are expected to be significantly affected. Agricultural production is not expected to change much at the global level, but there will be significant changes at the local and regional levels (it should be noted that this estimate does not take into account possible changes in the occurrence of agricultural pests). Human health is expected to be adversely affected, due to increases in the frequency and severity of heat waves and changes in vector-borne infectious diseases such as malaria. Overall, the impacts on developing countries are expected to be more severe than the impacts on industrialized countries and the former may have much smaller scope for adaptation than the latter.  1.3 Greenhouse Gases A number of gases cause the greenhouse effect, including carbon dioxide (CO2), methane (CH4), nitrous oxide (N2O), tropospheric ozone , halocarbons (including HCFCs and 4  HFCs) and SF6 (TPCC, 1995a). Of these gases, C 0 is the most important and is 2  estimated to have caused about 60 percent of the radiative forcing that has occurred since the beginning of the Industrial Revolution (IPCC, 1995a). Currently, the atmospheric concentration of CO2 is almost 360 ppmv , an increase of almost 30 percentfromthe pre5  industrial concentration of 280 ppmv (IPCC, 1995a). Atmospheric concentrations of the other greenhouse gases have also increased.  This paragraph is based on the impacts discussed by Bolin (1996). Professor Bolin is the Chairman of the IPCC and these impact estimates are taken from the IPCC's most recent assessment which was finalized in December, 1995. which has the chemical precursors: nitrogen oxides, non-methane hydrocarbons and carbon monoxide parts per million by volume 3  4 5  4  Because CO2 is the most important greenhouse gas, this chapter will focus exclusively on co  6 2  C 0 emissions comefroma number of sources, some natural and some human2  induced. The IPCC has estimated current emissions by source and these estimates are provided in Table 1. As can be seenfromTable 1, fossil fuel combustion and cement production are by far the most important sources of CO emissions, contributing 2  approximately 77 percent of global emissions.  The relationship between emissions and atmospheric concentrations of CO2 is not one-toone because some C 0 emissions are removedfromthe atmosphere by the oceans and by 2  terrestrial vegetation. Therefore, atmospheric concentrations depend on emissions but also on rates of atmospheric uptake by oceans and terrestrial vegetation. In addition, C 0  2  stays in the atmosphere for a long time (on the order of a century), so that today's emissions have a long-lasting effect on climate (IPCC, 1995a). The IPCC has developed a number of emission profiles corresponding to different levels of stabilized atmospheric concentrations of COV These estimates are provided in Table 2, along with 1980s levels of emissions and two of the IPCC's scenarios of future emissions (the scenarios with lowest and highest future emissions levels). The left-hand column of Table 2 contains various levels of stabilized atmospheric concentrations of C 0 and the right-hand column 2  contains the cumulated emissions over the period 1990-2100 that correspond to each different atmospheric concentration.  6  Chapters 4 and 5 examine CQ2 as well as CH4 and N 0. 2  5  To stabilize atmospheric concentrations of C 0 at today's levels (about 350 ppmv), 2  emissions would have to be cut by 50-70 percent from 1980s levels . Alternatively, 7  according to the IPCC,  If net global anthropogenic emissions (i.e., anthropogenic sources minus anthropogenic sinks) were maintained at current levels (about 7 GtC/yr including emissionsfromfossil-fuel combustion, cement production and land-use change), they would lead to a nearly constant rate of increase in atmospheric concentrations for at least two centuries, reaching about 500 ppmv (approaching twice the pre-industrial concentration of 280 ppmv) by the end of the 21st century. (IPCC, 1995a: Section 4.6)  In the absence of mitigation policies, it is unlikely that emissions will continue at today's 8  levels. Even the lowest future emission scenario of the IPCC (Scenario IS92c), based on the assumption of low growth rates of population and economic activity and low availability of fossil fuels, involves emission rates that are roughly equal and perhaps a bit higher than current levels. Clearly, policies to reduce net GHG emissions are necessary if serious climate change is to be avoided: even low rates of population and economic growth and relatively scarce fossil fuels will not be enough to reduce emissions.  Most of historical and current C 0 emissions arefromindustrialized countries and 2  countries with economies in transition (CGCP, 1996). In 1990, average per capita C 0 9  2  Emissions have increased since the 1980s: i n 1991, reported global CO2 emissions from fossil fuel use a n d cement production were 6.2 G t C , up from the 1980s average o f 5.5 G t C (Houghton et al, 1995: 49). 8  M i t i g a t i o n refers to both ways o f reducing atmospheric concentrations o f C C ^ : cutting emissions a n d  enhancing sinks w h i c h absorb atmospheric C C ^ (e.g., planting forests). These countries are countries w h i c h are m a k i n g the transition from centrally-planned to capitalist economic systems a n d include the Eastern European countries a n d Russia. 9  6  emissions " globally were 1.1 tonnes, with per capita emissions in the developed world 1  averaging 2.8 tonnes and in the developing world averaging only 0.5 tonnes (IPCC, 1995a). In other words, average per capita CO2 emissions in 1990 were more than five times higher in the developed world than in the developing world . Clearly, the bulk of 11  the responsibility for emission reduction rests with the industrialized countries This responsibility has in fact been explicitly recognized in the Framework Convention on Climate Change, and the Berlin Mandate excludes developing countriesfromemission reduction requirements (Anonymous, 1995b). The Framework Convention and the Berlin Mandate are discussed below in Section 1.5.  1.4 Canada's contribution to emissions Figure 1.1 shows Canada's GHG emissions by type in 1994. The different kinds of emissions are represented in terms of their 100-year global warming potential for purposes of comparison. CO2 is by far the most important GHG, accounting for almost 80 percent of GHG emissions. The next important gas is methane, CH4, accounting for about 14 percent of total GHG emissions. Nitrous oxide, N 0 , is the only other gas with 2  a significant share of emissions, at about 6 percent.  Figure 1.2 shows the relative contributions to Canadian C 0 emissions of various sectors 2  and activities in 1993. The most significant sources of CO2 emissions are power generation (19.7 percent), passenger transport (16.0 percent), industry (15.5 percent) and  1 0  These figures refer only to CO2 from fossil fuels.  O n a per dollar o f output basis, the difference between the developed a n d developing worlds narrows, w i t h average per US dollar emissions o f 0.26 tonnes i n the developed w o r l d a n d 0.16 tonnes i n the developing w o r l d ( I P C C , 1995a). 1 1  7  other mobile sources (14.5 percent). These are all sectors which depend on fossil fuels for their energy needs and which are important economically.  Figure 1.3 shows Canadian and global CO2 emissionsfromfossil fuel use over the period 1958 to 1992 and Figure 1.4 shows the growth rates of Canadian and global emissions over this period. The trend in Canadian emissions is close to the trend in global emissions. In fact, during the 1980s the rates of growth of Canadian and global emissions were virtually identical. This relationship has broken down in recent years, with the growth rate of Canadian CO2 emissions showing much more variability than the global rate although both show the same general trends. Both global and Canadian emissions appear to have (on average) declined during the late 1970s and early 1980s, increased over much of the 1980s and then declined again since the late 1980s. However, it is of course too soon to determine whether this latter trend will continue.  Canada is a relatively high energy user and emitter of C0 . Canadian C 0 emissions 2  2  account for approximately 2 percent of global emissions, but Canada comprises only about 0.5 percent of the world's population (CGCP, 1996). Even compared with other developed countries Canada is a relatively high energy user and C 0 emitter. Figures 1.5 2  and 1.6 show per capita C 0 emissionsfromenergy use for each of the OECD countries 2  for the years 1980 and 1993, with the countries rankedfromhighest to lowest. In 1980, Canada was the third highest per capita emitter of CO2, behind the US and Luxembourg. In 1993 it was the fourth highest, with Australia overtaking Canada. In both years Canada was a higher emitter than all of the other G7 countries, with the exception of the  8  United States. Figure 1.7 shows the average annual growth rate of CO2 emissions from OECD countries over the period 1980-1993, again rankedfromhighest to lowest. Canada's showing is much better, with a significantly lower rate of emissions growth than most other OECD countries and about half of the other G7 countries. However, this rate of growth was still positive, in contrast to most northern European countries, including three G7 members, which experienced negative rates of growth of emissions.  Canada's high rate of CO2 emissions is largely due to Canada's relatively high rate of energy use, most of which is derivedfromfossil fuels . Approximately 98 percent of 12  C 0 emissions, and 88 percent of total greenhouse gases emissions from anthropogenic 2  sources, in Canada derivefromfossil fuel production and use (Canada, 1994: 36). Figures 1.8 and 1.9 show average energy consumption per unit GDP in OECD countries for the years 1980 and 1993. Canada was the second and third most energy intensive OECD country in 1980 and 1993, respectively. Figure 1.10 shows the evolution of final energy consumption per unit of GDP in the G7 countries over the period 1980-1993. Canada is by far the most energy-intensive G7 country. On a more positive note, the figure shows Canadian energy intensity falling over most of this period, while intensities remainedflatin all of the other G7 countries except the US.  Figure 1.11 shows Canada's CO2 emissions from fossil fuel use and real GDP from 1958 to 1993 and Figure 1.12 shows the annual percentage change in the variables over the period. In general, these two variables tend to move together. Both have exhibited  9  positive rates of growth over most of the period, except in the recession years 1982 and 1991 (GDP) and the years 1975, 1977, 1981-1983, 1986 and 1990-91 (C0 ), some of 2  which are recession years and some of which are years of economic expansion. The growth rates of both variables appear to exhibit a slight downward trend over the period, but the year-to-year variability is so high that establishing a trend is difficult. It appears that this variability has been increasing in recent years, for both variables, with higher variability for the growth rate of C 0 emissions. 2  Figure 1.13 presents the difference between the growth rate of C 0 emissions and that of 2  real GDP. It is clear from this figure that the two variables do not move entirely in concert and that there have been some differences in their relationship over the period since the late 1950s. In the 1970s up until 1986, GDP grew faster than C 0 emissions. 2  This may be (at least partially) due to improvements with respect to emissionsfromfossil fuel use or reductions in the fossil fuel-intensity of GDP (i.e., how much fossil fuel energy is used to produce one dollar of GDP). This will be explored further in Chapter 4. In the 1960s and after 1987, the difference between the growth of the two variables is erratic, with C 0 emissions growing faster than GDP about half the time. 2  The fact that Canada is both a high per capita emitter of C 0 and a relatively energy2  intensive country implies that Canada should reduce its C 0 emissions and should do so 2  by reducing energy use. Even though Canada is a relatively small source of global emissions, it is a large emitter on a per capita basis, in comparison with both the Third World and other industrialized countries. The relatively high rate of energy use in  12  In 1992. 73 percent of Canada'stotal rmmary eifggy demand was met by fossil fuels (Canada, 1994: 23). 10  Canada also suggests that Canada can significantly reduce its CO2 emissions by reducing energy use. Fossil fuel use also contributes to environmental problems other than climate change, such as smog, particulate emissions and other forms of air pollution; water pollution, involving heavy metals, hydrocarbons, sulfuric acid and other dangerous chemicals; pipeline leaks and oil spills; and disruption of natural terrestrial and aquatic ecosystems in the process of extraction (BC Ministry of Environment, Lands and Parks, 1995; Statistics Canada, 1994). This provides even more reason to reduce fossil fuel use . 13  In 1993, the Royal Society of Canada's COGGER Panel surveyed the existing literature 14  on the costs of reducing GHG emissions in Canada and found that Canadian CO2 emissions could be reduced about 20 percentfromcurrent levels by the year 2010 at zero or negative economic cost, primarilyfromimproving the efficiency with which energy is used (COGGER, 1993). This finding confirms that Canada may significantly reduce its C 0 emissions simply by reducing energy use. It also implies that currently Canadians 2  are "wasting" energy and emitting more CO2 than necessary.  Unfortunately, current government and industrial policy in Canada does not appear to be sufficient to achieve this "no-regrets" level of emission reduction . In fact, it appears 15  likely that Canadian energy-related GHG emissions in the year 2000 will be about 9.5  In fact, it i s perplexing w h y activists have not really focused o n these other impacts i n their quest to reduce greenhouse gas emissions. Canadian Options for R e d u c i n g Greenhouse G a s Emissions A "no regrets" level o f emission reduction is the reduction that can be achieved a zero or negative net economic cost. 1 3  1 4  1 5  11  percent higher than 1990 levels (CGCP, 1996) Canada's international commitments 16  and policy with respect to climate change will be discussed in detail below.  1.5 Canada's International Commitments: The Framework Convention on Climate Change In recognition of the potentially serious and adverse effects of greenhouse gas-induced climate change, over 150 governments signed the United Nations Framework Convention on Climate Change (FCCC) in 1992 at the United Nations Conference on Environment and Sustainable Development (informally known as the Rio Conference) (Anonymous, 1994). Canada signed the FCCC at the Rio Conference and ratified it on December 4, 1992, making Canada the eighth national government to do so (Anonymous, 1995a).  The FCCC came into effect on March 21, 1994, ninety days after it was ratified by the fiftieth country (Anonymous, 1994). The "Ultimate Objective" of the FCCC is  to achieve ... stabilization of greenhouse gas concentrations in the atmosphere at a level that would prevent dangerous anthropogenic interference with the climate system. Such a level should be achieved within a time frame sufficient to allow ecosystems to adapt naturally to climate change, to ensure that food production is not threatened and to enable economic development to proceed in a sustainable manner. (United Nations Framework Convention on Climate Change, Article 2)  Industrialized signatories of the FCCC are listed as Annex 1 countries . As part of their 17  commitments under the FCCC, Annex 1 countries pledged to stabilize their emissions of  Total greenhouse gas emissions (i.e., from energy-related and non-energy-related sources) are expected to be about 8 percent higher than 1990 levels by the year 2000 (CGCP, 1996).  12  GHGs at 1990 levels by the year 2000 (hereafter called the stabilization target). At the first meeting of the Conference of the Parties (COP) to the F C C C in Berlin in 1995, it 18  was decided that the stabilization target would not be sufficient to achieve the ultimate objective of the FCCC and the Berlin Mandate was established (Cutajar, 1995). The Berlin Mandate is a process to design and negotiate some kind of mechanism or means to further reduce GHG emissionsfromAnnex 1 countries (Cutajar, 1995).  In December, 1997, the Kyoto Protocol to the FCCC was unanimously adopted by the Third COP. The Kyoto Protocol provides, for thefirsttime, legally binding commitments for industrialized country greenhouse gas emission reductions. The emission reduction commitment for industrialized countries (Annex I countries) overall is a 5.2 percent reductionfrom1990 levels by the "commitment period" 2008-2012. The magnitude of 19  individual country commitments varies. Canada's commitment is a 6 percent reduction from 1990 levels. The European Community and members states' commitment is an 8 percent reduction; the US: 7 percent reduction; and Japan: 6 percent reduction.  The Kyoto Protocol covers six greenhouse gases: carbon dioxide (CO2), methane (CH4), nitrous oxide (N2O), hydrofluorocarbons (HFCs), perfluorocarbons (PFCs) and sulphur hexafluoride (SF ). Industrialized Parties to the Protocol commit to reducing their total 6  C02-equivalent emissions of these gases to varying percentages (as listed in Annex B of  T h e list o f A n n e x 1 countries includes western industrialized countries as w e l l as countries w i t h economies i n transition, i.e., Russia a n d Eastern European countries. i.e.. the signatories o f the F C C C Commitment periods were adopted because they reduce the impacts o f a single anomalous year (e.g., w i t h respect to weather or economic conditions) o n emission requirements. The total emission reduction commitment i s calculated as a percentage o f total CC^-equivalent emissions o f the s i x gases i n 1990 (or 1995, as appropriate) times 5. 1 7  1 8  1 9  13  the Protocol) of 1990 levels in the first commitment period 2008-2012. Parties are permitted to use 1995 as a base year for calculating emission reduction commitments for HFCs, PFCs and SF but they must use 1990 as a base year for C 0 , CH4 and N 0. 6  2  2  The Kyoto Protocol includes three so-called flexibility instruments: Joint Implementation, Emissions Trading and the Clean Development Mechanism. Joint implementation (JI) involves the establishment by one country of greenhouse gas (GHG) emission reduction projects in another country, with the countries somehow sharing the emission reduction credit. Under the Kyoto Protocol, Annex I countries may now receive credit for JI activities in other Annex I countries (under Article 6) and in non-Annex I countries (under the Clean Development Mechanism — see below).  Article 17 permits Annex B Parties to the Protocol to engage in emissions trading for 20  the purposes of fulfilling their emission reduction commitments under the Protocol. However, the trading must be supplemental to domestic actions. Article 17 does not otherwise elaborate rules and guidelines governing emission trading. Rather, it states that the Conference of the Parties to the UNFCCC "shall define the relevant principles, modalities, rules and guidelines, in particular for verification, reporting and accountability for emissions trading".  Article 12 establishes a Clean Development Mechanism (CDM). The CDM is a process for JI between Annex I and non-Annex I Parties. Under the CDM, Annex I countries  A n n e x B Parties are A n n e x I Parties to the U N F C C C w h o agreed to keep their emissions to some percentage o f 1990/1995 levels b y 2008-2012. T h e i r commitments are listed i n A n n e x B o f the Protocol.  2 0  14  may receive credit against their emission reduction requirements for certified emission 21  reductions resultingfromprojects in non-Annex I countries, provided that the projects result in "real, measurable, and long-term" emission mitigation benefits and that the emission reduction is additional to what would have occurred otherwise. Projects may be undertaken by "private and/or public entities". Annex I Parties are permitted to "bank" certified emission reductions resultingfromCDM activities in the years 2000-2007 and use them to help satisfy their emission reduction requirements in the first commitment period 2008-2012.  The CDM will assist in arranging funding of projects. A share of the proceeds from CDM activities will be used to cover its administrative costs as well as to assist particularly vulnerable developing country Parties to meet the costs of adaptation to climate change.  Some observers are concerned that the CDM could become a loophole because it requires emissions additionality but not project additionality and because banking of CDM 22  emission reductionsfrom2000-2007 will reduce required emission reductions in 20082012.  23  2 1  E m i s s i o n reductions w i l l be certified b y "operational entities to be designated b y the" C O P serving as the  MOP. i.e.. emission reductions have to be additional to what w o u l d have occurred otherwise, but there does not need to be a demonstration that the project itself w o u l d not have occurred otherwise; this creates the possibility o f project proponents c l a i m i n g emissions credit for projects that w o u l d have occurred otherwise. T h e purpose o f the C D M i s to encourage new emission reduction or low-emission projects. Chris Rolfe, 1998. Kyoto Protocol to the United Nations Framework Convention on Climate Change: A Guide to the Protocol and Analysis of Its Effectiveness. Vancouver: West Coast Environmental L a w Association. 2 2  2 3  15  In November, 1998, the Conference of the Parties held their Forth Meeting in Argentina. As expected by many observers, this meeting did not result in definitive rules governing the three flexibility instruments. Rather, the Buenos Aires Plan of Action was the outcome of the meeting. The Buenos Aires Plan of Action sets a deadline of the year 2000 for resolution of a number of outstanding issues related to the Kyoto Protocol, including the flexibility mechanisms (Goree, 1998).  Canada has now committed to a six percent reduction in GHG emissionsfrom1990 levels by the period 2008-2012. As mentioned above, it appears unlikely that stabilization of emissions at 1990 levels by the year 2000 will be achieved by Canada and that, in fact, emissions will be probably be significantly above 1990 levels by the year 2000 . The federal Ministers of Environment and Natural Resources acknowledged this 24  failure at the end of 1996 and have promised to work towards improving emission reductions (Environment Canada and Natural Resources Canada, 1996). However, they also admit that policies currently in place and announced in December, 1996 will not be sufficient to achieve the stabilization target (Environment Canada and Natural Resources Canada, 1996). It is further doubtful that Canada's new commitment under the Kyoto Protocol will be reached without significant changes in policy.  Canada is not alone i n its failure to live u p to its commitment under the F C C C . It appears likely that most other A n n e x 1 countries w i l l also fail to achieve stabilization. In fact, according to independent N G O evaluations o f A n n e x 1 countries' national reports to the F C C C , only Germany, L u x e m b o u r g and  16  1.6 Canada's Mitigation Policies As part of their commitments under the FCCC, Annex I countries must produce and submit to the Convention Secretariat reports detailing their emissions and activities with respect to climate change (Anonymous, 1994). Canada released its report, Canada's National Report on Climate Change, in 1994 (Canada, 1994). This report discusses in detail Canada's policies and programs for mitigation, as well as other activities related to climate change such as improving scientific understanding of climate change, educating the public and preparing for climate change.  In 1995, the federal and provincial governments established the National Action Program on Climate Change (NAPCC), which "sets the course for meeting Canada's Convention commitments in the areas of... mitigation, adaptation, research and education and international cooperation" (Canada, 1995: 2). A report on the National Action Program was tabled for the first COP, as a follow-up to Canada's National Report (Canada, 1995). This report also discusses Canada's policies for mitigation, but is less detailed than the 1994 National Report and focuses primarily on "strategic directions" for policy. It does, however, reveal the paucity of government action to stimulate significant emission reductions and it is clearfromthe 1995 report that the federal and provincial governments are reluctant to go beyond an approach which relies on voluntary actions. The report identifies government's primary role as removing barriers to emission reduction, but does not indicate any substantial efforts in this direction beyond establishing various bodies to investigate ways that barriers can be removed (see pp. 14-17 of the report).  Switzerland appear likely to meet the stabilization target, w i t h the United K i n g d o m a " m a y b e " (Climate Network Europe a n d U S C l i m a t e A c t i o n Network, 1996: 5).  17  The mitigation actions discussed in the two reports amount to a collection of ad hoc programs to promote energy efficiency and the use of fossil fuel alternatives and to educate Canadians about climate change. These programs largely involve information provision, education, some small financial incentives and limited regulation (with respect to the energy efficiency of some appliances and some industrial equipment). Activities pursued by all three levels of government (federal, provincial and municipal) as well as industry are discussed in both reports, which makes it difficult to identify a national strategy and actions on climate change. This may be a reflection of the fact that Canada is a federal state, with responsibilities for GHG mitigation resting with both the federal and provincial governments (Canada, 1994).  In addition to the education programs and promotion of energy efficiency and fossil fuel alternatives, the NAPCC has established the Voluntary Challenge Registry (VCR) whereby large energy users commit themselves to emission reductions. The VCR is in fact the central policy instrument with respect to mitigation (Canada, 1995). It is entirely voluntary and suffers from a number of problems including weak commitments, poor reporting, exclusion of important sectors and a focus only on large energy users (Comeau, 1996). In 1996, the Minister of Natural Resources announced a number of measures to address some of these weaknesses, especially with respect to reporting and scope, but these new initiatives still rely on the goodwill of participants (i.e., the measures will not ensure compliance and do not include mandated standards) (see Table 3 below).  18  These changes to the VCR are included in Table 3 which lists the 40 new initiatives announced under the NAPCC in December, 1996. These new initiatives include information and education programs as well extension and strengthening of minimum energy efficiency standards for appliances and industrial machines and improved standards for federal buildings. An amendment to Canada's income tax act has also been made which permits tax savings for investments in alternative energy sources. Unfortunately, as mentioned above, these policies and programs will not be sufficient to achieve stabilization of emissions, let alone reductions in emissionsfrom1990 levels. The Ministers of Environment and Natural Resources promised to "work over the next year to further strengthen the VCR program" and to "work with stakeholders and colleagues ... to develop" further actions (Environment Canada and Natural Resources Canada, 1996.  In April, 1997, the federal government established the National Climate Change Process. As part of this process, several "Issues Tables" were established in which experts (and stakeholders) would prepare reports on various aspects of climate change policy which would then be used by government in formulation of policy. The federal government also established the Climate Change Secretariat to oversee this process. At the Joint Meeting Of Federal, Provincial And Territorial Ministers Of Energy And Environment in October, 1998, the Ministers "asked officials to propose a process, by Spring 1999, which will lead to a strategy to be reviewed by Ministers in late 1999" (Canada, 1998a).  19  In 1998, the federal government has also established the Climate Change Action Fund (CCAF) which will provide $150 for climate change over three years (Canada, 1998b). The CCAF will be allocated as follows:  " 'Technology Early Action Measures (TEAM) ($56 million) to support cost-effective technology projects that will lead to significant reductions in greenhouse gas emissions •Science, Impacts and Adaptation ($15 million) to improve our knowledge of the climate system and to assess the impact of climate change on the regions of Canada and the options for adaptation •Foundation Analysis ($34 million) to support the sound analysis of options for implementing the Kyoto Protocol •Public Outreach ($30 million) to inform and engage Canadians on climate change and to form partnerships with other governments, communities, the private sector and other organizations in early action measures." (Canada, 1998b)  Figure 1.14 shows the main climate change policy-making bodies in Canada as of late 1998 .  1.7 Conclusion and Introduction to Thesis Research While contributing only about 2 percent of global GHG emissions, Canada is a significant emitter on a per capita basis. Canada is also a comparatively high energy user, and a survey of the literature conducted by the Royal Society of Canada's COGGER Panel has found that Canadian emissions can be reduced by up to 20 percent  20  from 1988 levels, largely by improving the efficiency with which energy is used in Canada.  Unfortunately, Canadian federal government policy to date has been unable to motivate the emission reductions necessary to stabilize emissions at 1990 levels by the year 2000 (and thereby honour Canada's international commitment to that goal), and seems unlikely to be sufficient to achieve the further six emission reduction requirement that was established in the Kyoto Protocol. In particular, federal government policy with respect to GHG mitigation is largely confined to the to-date ineffective Voluntary Challenge and Registry Program and an ad hoc collection of programs to educate Canadians about climate change and to promote energy efficiency and the use of alternative energy sources. However, in 1998, significant funds were budgeted to climate change and the federal and provincial governments seem to be committed to establishing a more coherent and effective policy program than has existed so far.  One of the motivations of the analysis conducted in Chapter 5, the extrapolation of historical trends in energy use and GHG emissions in Canada, is the need to test whether a voluntary approach (which has characterized climate change policy in Canada to date) will be sufficient to achieve Canada's international emission reduction commitment. Trendsfromthe mid-1980's to mid-1990's will be used in the extrapolation; this period was marked by relatively few policy or programs to encourage energy conservation (or GHG emission reduction) and projections based on trendsfromthis period may be interpreted as what might be expected in a future without significant policy to reduce  21  GHG emissions - one in which the current voluntary approach, which requires no significant changes on the part of any stakeholder, prevails.  In official government documents such as Canada's National Report on Climate Change and Canada's National Action Program on Climate Change, the fact that Canada is a relatively high energy user is "explained" by references to Canada's cold climate, huge geography, resource-based economy and other non-energy factors. The analysis in Chapter 6, the comparison of Canada to other OECD countries, is motivated by the need to test these "explanations" and will do so by adjusting actual Canadian energy use for differences in climate, geography and industrial structure and then comparing Canadian energy use with that of other OECD countries.  The analyses in Chapters 3 and 4 involve the use of energy indicators and the decomposition methodology developed by Lee Schipper and his colleagues at the Lawrence Berkeley National Laboratories in Berkeley, California. This approach was chosen because it is relatively transparent,rigorousand tractable and because it has been used in previous analyses of energy use in Canada and many other countries . Chapter 25  26  2 describes the methodology in some detail and discusses the time periods and sectors examined as well as the nature and sources of the data used in the analyses in Chapters 3 and 4. These analyses highlight the factors that were of most importance in changing energy use over time and thereby indicate areas which energy conservation and GHG emission reduction policy might target.  2 5  2 6  See. for e.g., M a r b e k Resource Consultants et al (1989) and Natural Resources Canada (1996). See, for e.g., Schipper a n d M e y e r s et al (1992).  22  The results of the energy use and GHG emission decompositions, extrapolations of historical trends and comparisons of Canada to other OECD countries are brought together in Chapter 7 which discusses the three questions which were posed at the beginning of this chapter.  23  Chapter 2: Methodology  2.1 Introduction to Decomposition Approach The analyses in Chapters 3, 4 and 5 are based on the energy indicators/decomposition method developed by Lee Schipper and others at the Lawrence Berkeley Laboratories in California. This methodology is described in some detail in Schippers and Meyers et al (1992). It has also been used by the Energy Efficiency Branch of Natural Resources Canada in their Energy Efficiency Trends in Canada analyses (NRCan, 1996 and 1997a).  The purpose of the decomposition method is to disaggregate total energy use changes over some period of time into component parts. Typically, these component parts are changes in energy use due to:  • changes in the overall level of activity (e.g., manufacturing output) • changes in the structure of activity (e.g., shifts in manufacturing activity from to chemicals to pulp and paper) • changes in the energy intensities of one or more activities (e.g., joules per dollar of pulp and paper output)  In chapter 3,1 will also separate out the change in energy use due to weather changes. In chapter 4, changes in emissions of carbon dioxide (CO2), methane (CH4) and nitrous oxide (N2O) are disaggregated into changes due to:  24  • changes in fuel mix (e.g., shiftsfromoil to natural gas) • changes in the overall level of activity • changes in the structure of activity • changes in the energy intensities of one or more activities • weather  The decomposition method works by calculating what the change in energy use would have been if only one factor (e.g., the level of activity) had changed but the others had remained at their base-year values. In this way, the influence of each factor on the overall change in energy use is determined.  2.2 Derivation of Decomposition Expression The decomposition method is based on the identity:  E = I*A where E is the energy use associated with a particular activity I is the energy intensity of that activity A is the level of the activity For instance, total industrial energy use is equal to industrial energy intensity times the level of industrial activity:  Industrial energy use (GJ) = Intensity (GJ/$) * Output ($)  25  2.2.1 Subsectoral Activities and the Indicators Pyramid At the sectoral level, total activity is comprised of a number of subactivities. For instance, industrial activity is comprised of manufacturing, mining, construction, forestry and (in some cases) agricultural activity. These industrial activities are themselves composed of a number of sub-subactivities. For instance, in Natural Resources Canada's industrial energy use database (NRCan, 1997b), manufacturing is subdivided into cement, chemicals, iron and steel, petroleum refining, pulp and paper, smelting and refining, and other manufacturing. In Natural Resources Canada's residential energy use database, household activity is composed of space heating, water heating, cooking, lighting and appliances, with space heating and cooling and appliance use subdivided into a number of different types of equipment.  To help sort out the subactivities and sub-subactivities, Schipper and his colleagues have devised what they call the "Indicators Pyramid" . The Indicators Pyramid shows the 27  different sub- and sub-subactivities that occur at different levels of aggregation within an energy-using sector. Indicator pyramids will be used for each sector to depict the different levels of aggregation as well as the activities examined in the analysis of that sector. They are shown in Figures 2-1 to 2-5.  Each subactivity (and each sub-subactivity and so on) has its own energy intensity. This means that, even if the different energy intensities remained the same, a shiftfromone kind of activity to another will change overall energy use. For instance, pulp and paper tends to be more energy intensive than chemicals and pulp manufacturing tends to be  26  more energy-intensive than paper manufacturing. So, all things being equal, a shift in manufacturing activity from pulp and paper to chemicals, orfrompulp to paper manufacturing, will decrease overall energy use. Schipper and Meyers et al (1992) call this the "structure effect".  To incorporate the structure effect, the energy use identity can be altered: E = A * S i Silt m  where S; is the share in total activity  of subactivity i  (i.e., Si = A/Aj, with A; is the activity level of the subactivity i) Ii is the energy intensity of subactivity i  In addition to changes in the structure of activity, changes in the energy intensity and the level of activity can also affect overall energy use. The main part of the sectoral analyses presented in Chapter 3 will involve calculating the differential contributions to energy use changes of these changes. To isolate these differential contributions, I will be using Laspeyres indices that have also been developed by Schipper and his colleagues (see, for e.g., Schipper and Meyers et al, 1992) and which are based on the energy use identities described above.  2.2.2 Derivation of Laspeyres Indices  T h e Indicator Pyramids are also used i n N R C a n (1996 a n d 1997a).  27  Following Schipper and Meyers et al, 1992, changes in overall energy use due only to shifts in the structure of activity are called the "structure effect", changes in energy use due solely to changes in energy intensities are called the "intensity effect" and changes in energy use due solely to changes in activity levels are called the "activity effect".  Derivation of these indices is after Park (1991) and begins with the identity for total energy use at time t: (1)  E t = Atlt  Equation (1) can be expanded to incorporate the structure of activity: m (2)  Et=  AtZ.S*Iit  where i is one of m subsectoral activities Sit is the share in total sectoral activity of subsectoral activity i at time t; Sjt=  Ah/At  lit is the energy intensity of subsectoral activity i at time t The change in total sectoral energy use from time t=0 to time t=n can be written as (3)  AE = En - E o  Substituting equation (2) into equation (3) yields m  (4)  m  AE = A„Z Si„Ii„-Ao2 SioliO 1  1  Equation (4) can be rewritten as m  (5)  m  AE = (Ao + AA)S .(Sio + ASi)(Iio + AL) - AoS .Sato i  i  where X„ = Xo + AX Multiplying and rearranging terms and substituting Xi - Xo for AX in equation (5) yields 28  (6)  A E =  (An- Ao) I  .Siolio  m + A o Z .(Sin - S ) I i O  1  i 0  m .Siofjin " I i o )  + AqL  + Residuals m  Residuals = ( A „ -  Ao)£  m  .(Sm - S i ) I i o + ( A „ - A o ) L 0  m  .Sio(I  - Iio)  m  m  + A o S .(Sin " S i o X I i n - I i o ) + ( A „ - A , ) ! .(Sin - S ) ( I i n - Iffl)  1  l  i 0  The Laspeyres indices used by Schipper and Meyers et al (1992) are derived by dividing both sides of equation (6) by E (i.e., the initial year energy use): 0  m  (7)  A E =  (A„- Ap) S  Eo  E  output effect  jSjoIjo  0  m + A o Z i(Sin  -  S )Iio  structure effect  - Ijp)  intensity effect  i o  Eo m + A p Z iSio(I  m  Eo  + Residuals m  Residuals -  [(An - Ao)S  m  .(S m  S )Iio]/Eo + [ ( A . - A o ) I  S (Iin - I,o)]/E  i 0  i  i 0  0  i  m  m  + [ A o Z . ( S i n - S ) ( I i n " Ii0)]/E + [(K i 0  0  29  - Ao)2  (S  m  " SioXIin - I i o ) ] / E  0  Equation (7) can be further simplified: (8)  4E =  An _ i  Eo  Ao  output effect  + £ iSinljQ _ ]  structure effect  m S iSioIiO  intensity effect  m + £ jSioIjn _ J £  m iSjoliO  + Residuals  Residuals = [(A„ - Ao)2: .(Sin - S )Iio]/Eo + [(A„ - Ao)Z .S (Iin - Iio)]/E io  i0  i  0  i  m  m  + [ A o l (Si„ - SjoXIin - I )]/Eo + [(An - Ao)I .(S« - S )(Ii„ - Iio)]/E i0  i0  Equation (8) is used in the analyses in Chapters 3 and 4 to isolate the effects on energy use of changes in the level of activity, the structure of activity and the energy intensity of activity, respectively. The residual terms occur because equation (8) is a factorization of a discrete change in energy use over time; the residuals represent the interaction of the changes in the three variables A , S and I.  30  0  2.2.3 Partial Decompositions In the residential sector, it was not possible to conduct a full decomposition as represented in equation (8). This is because the activity measure (number of households) cannot be distributed among the various end-uses. Most end-uses are used by all households; for instance, appliances like refrigerators and stoves are virtually universal. Therefore, it was not possible to calculate the shares of each end-use (i.e., the Si). However, it was possible to do a full decomposition for space-heating in this sector since data were available on the number of households using four different kinds of space heating equipment (normal fuel-efficiency, medium fuel-efficiency, high fuel-efficiency and other).  Similarly, it was possible to do a full decomposition for the commercial sector as a whole with m per building type as the share variable), but not for end-uses within each building 2  type.  In these cases where a full decomposition is not possible, a partial decomposition, involving output and intensity effect calculations, was conducted instead. This is represented in equation (9): (9)  An _ j  A E = Eo  output effect  Ao  +  intensity effect  + Residual  31  m  Residual = (An-Ao) S i(Im-Iio) Eo  Derivation of equation (9) is analogous to the derivation of equation (8), but beginning with the expression: m  m  AE = A„S Ijn-AoL Iio I 1  2.2.4 Incorporating Weather Changes In the residential and commercial sectors, weather can influence the amount of energy used for space heating and cooling. To account for the influence of weather in energy use changes in these sectors, modification was made to the Laspeyres indices, using the approach elaborated in Appendix B, Sections 3.5 and 4 of NRCan (1996).  At the end-use level, space heating and cooling energy intensities are modified by dividing the unadjusted energy intensities by the ratio of current year to base year heating degree-days or cooling degree-days, as appropriate.  The weather-adjusted energy intensities for space heating (f it) and space cooling (fat) H  are: I' it = H  laft  (HDDt/HDDo)  I'cit =  k  (CDWCDDo)  After adjusting space heating and cooling intensities in this way for all subsectoral activities, sectoral decompositions may be conducted, where the total weather-adjusted 32  energy intensity for subsectoral activity i in year t is the sum of the weather-adjusted intensities for space heating and cooling plus the sum of k other end use energy intensities:  ha  I'it =  (HDDt/HDDo)  +  k —  + £ i y  k  i j t  (CDDt/CDD ) 0  To derive this new decomposition, define It as the sectoral intensity index in year t: m  I = £ jSjoIin t  m  £ iSioliO Similarly, f is the weather-adjusted sectoral intensity index in year t: t  It  —  m  £ jSjoIh: m  Z iSiol'iO  Define W as the weather index in year t: t  Wt=  intensity index in year t weather-adjusted intensity index in year t  = JL  I't  £ jSjoi Z iSiolio n  m  m  £ iSjol'm m  Z iSiol'iO  33  Therefore the intensity effect, AEL, can be rewritten as: AEI - JL - 1 = W I't t  t  lo  -1  WtI't  The change in energy usefromthe base year to year n that is attributable to the intensity effect is simply the intensity effect times base-year energy use, Eo  AEjntensity Eo AEI =  or  n  E„ - En = En [ W I't - l]  E„ - E refers to intensity-induced energy  t  0  Wo I'o  changes  34  and  E„ = Eo + Eo [ WtTt -1] Wo I'o = EoLt + E o [ W t I t I'o  (10)  E„ - E  0  Wo I'o  -It] I'o  = Eo I i + E [ _%J\  -JA] -Eo  0  I'o  Wo I'o  I'o  = Eo(Lt + WtTt -± -1) I'o Wo I'o I'o  = E [(r 0  t  . i )+ r (w _i)] L  I'o  L  I'o W  0  Adding and subtracting (w /Wo - 1 )fromthe right-hand side of equation (10), dividing t  both sides by Eo and rearranging yields a revised intensity effect:  (11)  AEL = (IA - l ) + (Wt - l ) + (JA - l X ^ L - O I'o  W  0  I'o  W  0  Thefirstterm on the right-hand side of equation (11) is the weather-adjusted intensity effect, the second is the weather effect and the last is a new residual which arisesfromthe interaction of thefirsttwo terms.  Therefore, the weather-adjusted sectoral decomposition is  (12)  AE =  A„ . ]  output effect 35  Eo  Ao  + £ iSinlio . ! m  structure effect  £ iSioliO m + £ i'SjoI'n . j tn £ iSioI'o  weather-adjusted intensity effect  + W„_ -1 Wo  weather effect  + Residuals  Residuals = {(A,, - Ao)£ ( S - S )Iio]/Eo + [(A, - Ao)£ Sio(Im - I»)]/Eo m  i0  i  i  m m + [AoZ .(Sj„ " Sio)(Iin " I )]/Eo + [(A. - Ao)Z .(Si„ - S )(Iin - I )]/Eo i0  i0  -l)(Wn_-l) I'o  Wo  Weather can be incorporated into partial decompositions by following the same procedure, but starting with: m T = £ jljn m £ a*, t  m  i\ = £ j i i , m £ ii'io This results in the Laspeyres indices and residuals: 36  i0  (13)  AE=  An _ i  Eo  A©  output effect  + ^ Jn _ i m Z il'o  weather-adjusted intensity effect  + W _i Wo  weather effect  n  + Residuals  Residuals = [(Ai - Ao)Z (I - IJO)]/EO M  -l)(Wn_-l)  I'o  Wo  2.2.5 Greenhouse Gas Emission Decompositions In Chapter 4, a decomposition approach is used to disaggregate changes over time in sectoral greenhouse gas emissions (carbon dioxide, methane and nitrous oxide) into changes due to changes in fuel mix, activity, intensity and structure (and weather). The decomposition used in chapter 4 is a modified version of that used in chapter 3 and is derived as follows:  Let Gt be emissions of a greenhouse gas in year t. (H)  Gt = Z g Ekt k  k  37  where gk is the greenhouse gas emission coefficient of fuel k and Ekt is the amount of fuel k used in year t (measured in joules). The greenhouse gas emission coefficient is the amount of the greenhouse gas emitted per joule of fuel k that is consumed. For example, the carbon dioxide (GO2) emission coefficient of natural gas is 49.68 tonnes/terajoule of natural gas consumed . 28  The change in emissions of gas G over time can be written as: (15)  AG = G„ - Go = Zk gkEkn - 2 gkEko k  Let Fkt be the share of fuel k in total sectoral energy use in year t: Fkt = Ekt Et  and  Ekt = FktEt  Substituting this into equation (15) yields (16)  AG = 2 gkFknEn - 2k gkFkoEo = 2k gk(FknEn - FkoEo) k  m Substituting At2 iS Iit for Et and (Xo + AX) for X„ (as in equation (5) above), equation n  (16) can be rewritten as: m (17)  m  AG = 2 gk[(F + AFk)(Ao + AA)2 i(S» - ASiXLo - AL) - F Ao2 iSiolio] k  k0  k0  Multiplying and rearranging terms and substituting X„ - Xo for AX yields:  (18)  m m m AG = 2 gklXFkn - F )Ao2 iSiolio + F (A - Ao)2 iS Iio + F Ao2 ^ k  k0  k0  D  io  ko  m + F oAo2 iSio(Im - Iio) + residuals] k  T h e source o f emission coefficients for a l l t h r e e ^ e e n h o u s e gases i s Jaques, 1992.  - S )Iio io  residuals =  gk * [ F k o A o Z  i ( S - Sio)(I m  m  - Iio) + F o ( A „ - A o ) Z i ( S k  m +  m  - Si )Iio 0  m  F o ( A „ - A o ) Z i S i o ( I i n " I i o ) + F k o ( A „ - A o ) Z i ( S b - S i o ) ( I j n " Iio) k  m  m + (Fkn - F o ) A o Z  i ( S - Sj )IiO + ( F k n - F  k  m  0  k 0  )AoZ  k o  )AoS  m  i S i o ( I i n - Iio)  m  + (Fio, - F k o X A n - A o ) E iSiolio + ( F k „ - F  j(Sj„ - S i ) ( I 0  m  -Lo)  m + (Fkn - F  k 0  ) (A„- Ao)E  i ( S - S )IiO m  i0  m + (Fk„" F o)(A„ - A n ) E iSio(I k  - Iio)  m  m + ( F k n - F k o X A , - A o ) Z i ( S - S i o X I i n " IiO>] m  Dividing both sides of equation (18) by G o yields: (19)  A G =  Z  [gk(Fkn - F o ) A p £  k  jSipIjo]  k  Go  fuel mix effect  Go  output effect G„ S k fgkFicoAoZ  m  i(S  m  - Sio)Iio]  structure effect  Go  m Z rgkF oAoZ k  k  iSio(Iin  intensity effect  -Iio)l  Go  + Zkgk  * residuals  Go  where the residuals are the same as in equation (18). Equation (19) can be simplified considerably by multiplying the right-hand side of the m equation by E o / E , by substituting E for A o L 0  rearranging each term:  0  39  iSiolio  in the fuel mix effect term and by  40  (20)  A G=  Eo * Zkgk(Fkn - F  Go  G  k  0  S  k  )  fuel mix effect  0  + ( An/Ao - 1) Eo +  output effect  [gkF o] k  Go  m . i] E o * ^ * f g k F o l  + [ £ iSfaljo  k  iSiolio  2  G  structure effect  °  m + [ Z jSioIjn 2 iSioI  [gkF o1  . ] ] Eo*  k  m  G i o  + S  k  °  * (energy decomposition residuals)*Eo  + £ k gkFico G  intensity effect  0  & ( F i  n  , - F  t  0  )  *(F„-F ) 0  Go  Recall that  Fia = Eia/Et.  Therefore, FiaJEt = E y . Substituting E  k  0  for F  yields: (21)  A G =  Eo *  S  kgk(Fkn - F  Go  G  k  0  fuel mix effect  )  0  output effect  + ( A n / A o - 1 ) * £ k fflcEkol G  0  m  + r£  jSinljo  ^iSiolio  _ ,] * S  k  [gicEko]  G  structure effect  °  m + [ S jSioIjn ^iSiolio  _ j ] * £ k [g E o] k  G  k  o  41  intensity effect  k  0  E  0  in equation ( 2 0 )  + Z  k  gkEico  * (energy decomposition residuals)  Go  + Zicg (Fkn-F k  k 0  *(E -Eo)  )  R  Go  Since G  t  = Z  k  g ^ , the term  [gkEkol  reduces to 1.  Go  Therefore, equation (21) reduces to (22)  A G = E o *£kgk(F -F ) ta  Go  fuel mix effect  k0  Go  +  [ A n -1]  output effect  Ao  m  + [Z  jSjnlio  j]  structure effect  j]  intensity effect  m iSjoIio  Z  m  + [Z  iSiolin  _  m Z  iSjoIio  + (energy decomposition residuals)  + Z k gkffkn- F ) k0  *  . J^)  Go  Equation (22) is identical to the energy use decomposition in equation (8), except for the addition of a "fuel mix effect" term and an additional residual term.  42  Weather can be easily incorporated into equation (22) using the same method by which it was incorporated into the energy decomposition equation (i.e., through the intensity effect term). This amounts to replacing the intensity effect term in (22) with the weatheradjusted intensity term of equation (12) and adding the weather effect expression and residual representing the interaction of the weather-adjusted terms:  fuel mix effect  + [An-1] Ao  output effect  structure effect  weather-adjusted  + m  +  intensity effect -1  W  weather effect  Wo  + (energy decomposition residuals)  -l)(Wt_-i)  + (It  Wo  I'o  + Z g (F -F o)*( .E ) k  k  k n  k  E n  0  Go  43  The equations for partial decompositions of greenhouse gas emissions, with and without the influence of weather, can be derived in a similar fashion:  Partial greenhouse gas decomposition:  (24)  AG =  Eo * £kgk(Ficn - F o) k  Go  G +  foel  m  i  x  e f f e c t  0  [ A n - 1]  output effect  Ao m  + £ jl _ i  intensity effect  n  m  2 ilo m  +  (An - A o ) I i(Iin - Ii0)  E° +  residuals  Stft (F .-F )*(F -B ). c  ta  k0  11  0  Go  Partial greenhouse gas decomposition with weather effects: (25)  AG =  Eo * ^ ( F k n - F k o )  Go  G +  ^  m  i  x  e f f e c t  0  [Ajj-l]  output effect  Ao m  y + ^LJ£n . i T<  weather-adjusted  m  2 jf o  intensity effect  + WJL - 1 Wo  weather effect  44  +  (A,, - Ap) £  m  j(Iin - I i o )  ^° +  residuals  -1)(WL-1) I'D  WO  + £ k gk(Fkn- F ) * ( E „ - E o ) k 0  Go  The decomposition equations (12) and (23) are the same as would have obtained had one simply added the additional weather effect term to the Laspeyres indices and then subtracted itfromthe residuals. The longer derivation was used here to demonstrate that doing so is logically sound, by showing that these decompositions obtainfromusing an energy use identity that is modified to include the influence of weather on space heating and cooling energy use . 29  2.3 Alternatives to the Laspeyres Index It should be noted that the Laspeyres index approach is not the only way of decomposing changes in energy use into their component parts (i.e., into changes due to activity, intensity and structural changes). A popular alternative to the Laspeyres index is the Divisia index, which uses a weighted average of relative growth rates instead of a fixedbase index as the Laspeyres does (Boyd et al, 1988). In any index, the residual terms reflect the interaction of non-marginal changes in the variablesfromthe base year values. If the differences between base year and year-of-interest (i.e., year n) values are large, residuals can also be large. This problem is mitigated in the Divisia index because the  45  weights change (Howarth et al, 1991). However, a drawback with the Divisia approach is that the interaction terms are arbitrarily assigned to the indices, instead of remaining separate as they are in the Laspeyres approach (Howarth et al, 1991). Further, the Laspeyres index is particularly appropriate for the kind of analyses conducted in Chapters 3 and 4 because it measures changefrominitial year values, which is the purpose of the analyses in those chapters.  In any case, a comparison of the Divisia and Laspeyres indices for US manufacturing energy use found that both indices give "closely similar results in the decomposition of annual energy growth" and that while "the difference in the decomposition of aggregate energy intensity are more marked... they are not significantly large" (Howarth et al, 1991: 142).  Another alternative to the Laspeyres index is the Paasche index, which is very similar except that the base is year-of-interest values, rather than initial year values as in the Laspeyres index (Harnett, 1982: 635-637). Informetrica Limited (1995) compare a number of different indices for measuring energy efficiency and find that using the Paasche index in Schipper's decomposition function is problematic because it is impossible to isolate the effects of changing activity in the structure and intensity decompositions and because the activity decomposition reduces to unity. However, this finding seems questionable; it appears that Informetrica Limited simply divided the  In fact, the same is true o f the original energy decomposition, a n d the reason for the longer derivation is the same: to demonstrate that the decomposition is logically sound a n d obtains from the energy use identity.  46  Laspeyres index numerators by year-in-question energy use. This is not exactly a Paasche index, which for the activity effect, for instance, would be:  m m (An/Ap) £ jSinlin = \X jSjnljn = 1/An E„ AflEn  The structure effect would be  m  m  AnS iCSia/Sjo)!,  = A , S iCSfa)!,,  E„  =  EnS i(Sio)  L _ Z i(Sio)  m  m  and the intensity effect would be  m  m  AnE iSin(Iin/Iio) =  A i Z i(S )I  E*  EaS-KIjo)  m  m  =  1 I i(I ) m  i0  Because the purpose of the analyses is to investigate changes in energy use and greenhouse gas emissions over time (i.e., from one particular, earlier year to another particular, later year), a Paasche index approach is not appropriate.  A third alternative is a combination of the Paasche and Laspeyres indices, the Fisher Ideal Index, which is the geometric mean of the two indices . In the Fisher Ideal index, 30  the base is a combination of initial year and year-in-question values. This index may therefore produce lower residuals (which rise proportionately with the difference from  47  base year values) but is less transparent and intuitively appealing than the Laspeyres approach (Informetrica Limited, 1995).  A fourth alternative is the "chained Divisia" in which the base year is the year before the year-of-interest. This is the index proposed by the Agence de TEnvironnment et de la Maitrise de l"Energie (Ademe) of France (Informetrica Limited, 1995).  There are, in fact, a myriad of index possibilities. According to Boyd et al (1988), Fisher (1972) examines more than 100 different indices (combinations of Laspeyres, Paasche and Divisia). Informetrica Limited (1995) evaluate the Laspeyres, Paasche, Fisher Ideal and chained and unchained Divisia indices against 12 criteria. Theyfindthat the Laspeyres index scores best overall, which increases confidence in the use of that index in Chapters 3 and 4 of this thesis.  2.4 Limitations of this Approach There are a number of limitations to this approach that should be noted. First, only national-level data are used. It can be quite convincingly argued that the appropriate level of analysis (geographically) varies from one kind of activity to another. For instance, household energy use for space heating in Canada is not uniform across the country because winter temperatures and the length of the season vary considerably from one region to another. Similarly, energy use for the same kind of manufacturing activity can vary considerably in terms of kinds of energy used and the intensity of use,  I n bis evaluation o f more than 100 indices, Fisher found that the Fisher Ideal Index performed best (Fisher, 1 9 7 2 , a s c i t e d b y B o y d e / a / , 1988: 310).  3 0  48  depending on where it is undertaken. So, using national level data obscures considerable differences in energy use, including reasons for changes in energy use,fromone region of the country to another. Therefore, conclusions drawn about the causes for changes in energy use based on these data should be treated with some caution . 31  Second, any conclusions drawnfromthis analysis are highly dependent on the base year chosen, because the Laspeyres index measures change as the difference in year-ofinterest valuesfrombase year values. If the base year is anomalous or otherwise problematic, the Laspeyres indices will reflect this and will therefore also be problematic. Further, Laspeyres indices for a given year (e.g., 1995) will differ if different base years are used. In this thesis, base years were chosen as the earliest year for which data were available and were not explicitly assessed in terms of their suitability as base years in the Laspeyres indices. This appears to be common practice in energy decomposition analyses.  Third, the results of the analysis depend on the index chosen. As discussed above, there are a number of alternatives to the Laspeyres indices used in this analysis, and each different kind of index will yield different results. However, it is beyond the scope of this analysis to examine these differences and it is hoped that these differences would not be significant. It must be recognized though that the potential for significant differences does exist.  I a m grateful to R a l p h Torrie for bringing this important point to m y attention.  49  Fourth, the period of analysis chosen (i.e., 198x to 199x, depending on the sector ) will 32  have some influence on the results. This period was chosen because exclusively on the basis of data availability. A more robust analysis would cover a larger time period, ideally expanded to include the 1970's so as to capture the large swings in energy prices since 1973 (i.e., the two OPEC price shocks as well as subsequent declines) as well as significant changes in the nature and scope of government policies with respect to energy supply and conservation (for a discussion of these policies in Canada see Marbek et al, 1989). In addition, it has been shown by Schipper and Meyers et al (1992) that part of the decreases in industrial energy intensity that were observed in the United States and West Germany in the 1970's and 1980's were in fact technologically-based and derived from changes in technology that began in the 1960's.  It was originally hoped that the Canadian energy use data set could be extended back as far as the early 1970's, at least in some sectors. Unfortunately, this was not possible. In some cases, the data were simply not collected, or were collected on a different basis. In other cases, inadequate data documentation and lack of transparency in the Natural Resources Canada energy use database (NRCan, 1997b) prevented extension of the data.  In one sector, industry, data were availablefrom1973 onwards. However, the source of these data changed so significantly after 1984 that it was necessary to conduct the decomposition over two distinct periods: 1973-1983 and 1984-1996.  Aggregate national data cover the period 1984 to 1995; industrial sector data cover two periods: 19731983 a n d 1984-1996; transportation data cover the period 1 9 8 4 - 1 9 % ; residential a n d service sector data cover the period 1981 to 1995.  50  Finally, as Park (1991) notes, the identity on which the decomposition equations are based is simply that: an identity. It does not describe a causal relationship. Therefore, the decomposition can show which of the three underlying factors (activity, intensity, structure) was largest or smallest in the observed change in energy use, but it does not explain why that factor was largest or smallest. However, possible reasons are examined in the discussions in Chapters 3 and 4, which mitigates this shortcoming. In light of this, the decomposition should be viewed as a tool to indicate which underlying factors were significant and therefore should be further investigated.  2.5 Method for OECD Comparisons  Canada is compared to other OECD countries in several ways. First, to investigate the validity of claims that per capita energy use in Canada is higher than other countries due to climate, geography and industrial structure, per capita energy use in Canada in 1984, 1990 and 1994 is calculated using climate, geography and industrial structure factors from the United States and various OECD country aggregations: OECD-12, EU-9, Scandinavian-4, Umb-3 (US, Australia &Japan).  The corrections are as follows:  Energy use in Canada in year t is equal to the sum of energy use in each of the sectors (residential, commercial, transport and industry) in that year: E ,t c  = ERc,t + EC ,t + ET ,t + EI , c  c  c  t  51  Adjustments will be made to energy use in each sector in order to "correct for" differences in climate, geography and industrial structure between Canada and other industrialized countries. These adjustments are described below.  2.5.1 Climate Energy use in the residential and commercial sectors is equal to the sum of energy use for end-uses other than space heating and cooling (i), plus space heating (SH) and cooling (SC): (26)  ERc,t + ECc,t  =  E ERc,i,t + ERc,sH,t + ERc,sc,t + E ECc,i,t + ECc,sH,t + ECc,sc,t  = E ERc,i,t + E ECc,i,t + ERc,sH,t + ECc,sH,t + ERc,sc,t + ECc,sc,t = £ ERc,i, + E EC ,i,t + ESHc, + ESC , t  c  t  c  t  where ESHc,t and ESCc,t are energy used for space heating and cooling, respectively, and ESHc,t = ERc,sH,t + ECc,sH,t  ESCc,t ERc,sc,t + ECc,sc,t =  To incorporate climate, equation (26) can be written as: (27)  ERc.t + ECc,t - E ERr-if + E E C r + E S H r , * H D D . t + E S C r • * C D D . t HDD t CDD ,t c  Q  c  c  where F£DDc,t and CDDc,t are heating degree-days and cooling degree-days, respectively, in Canada in year t.  In the comparison, residential and commercial energy use will be calculated as if Canada had the same degree-days as the country or country-grouping to which it is being  52  compared. To do this, multiply space heating and cooling "intensity" (energy per degreeday) in Canada by the comparison country's degree-days:  (28)  ER'ct + EC'ct = 2 ERc,i, + E EC ,i,t + ESHc. * HDDx,t + ESC ,t * CDDx.t HDD CDDct t  c  t  c  c>t  where HDDx,t and CDDx,t are heating degree-days and cooling degree-days, respectively, in country (or country-grouping) X in year t.  2.5.2 Geography Geography is assumed to affect energy use through its impact on passenger and freight transport demand. In the absence of more descriptive data, passenger-kilometres per capita and tonne-kilometres per dollar of GDP were used to adjust Canada's passenger and freight transport energy use, respectively. This assumes that differences among countries in passenger-kilometres per person and freight tonne-kilometres per dollar of GDP are strictly due to differences in average distances which people and goods must travel.  The passenger transport adjustment was made by multiplying Canada's passenger "transport intensity" (terajoules of energy used in passenger transport divided by passenger-kilometres per capita) in a particular year by passenger-kilometres per person in that year in the United States or one of the four country aggregations:  EPT'ct =  EPT .t  * p-km .t  p-kmct  Popx,t  c  x  53  POpc,t where EPT'c,t is adjusted passenger transport energy use, EPTc.t is actual Canadian passenger transport energy use in year t, popc,t is population in Canada in year t, p-kmc,t is passenger-kilometres in Canada in year t, and popxt and p-kmx,t are population and passenger-kilometres in country or country aggregation X in year t.  Similarly, thefreighttransport adjustment was made by multiplying Canada's freight "transport intensity" (petajoules of energy used infreighttransport divided by tonnekilometres per dollar of GDP) in a particular year by tonne-kilometres per dollar of GDP in that year in the United States or one of the four country aggregations:  EFT'ct =  EFT ,t t-kmct GDP ,t c  * t-km , GDPx,  x t  t  c  where EFT'c.t is adjustedfreighttransport energy use, EFT ,t is actual Canadian freight c  transport energy use in year t, GDPc,t is Gross Domestic Product in Canada in year t, tkmc,t is tonne-kilometres in Canada in year t, and GDPjy and t-kmx,t are Gross Domestic Product and tonne-kilometres in country or country aggregation X in year t. For all countries, Gross Domestic Product is measured in constant US 1990 dollars, converted from local currencies using Purchasing Power Parities.  The adjusted passenger andfreightenergy use were added together to derive total adjusted transport energy use: ET'c,t = EPT' ,t + EFT'ct c  54  2.5.3 Industrial Structure Recall that industrial energy use may be calculated as EIc,t — Ac,t 2 Sc,i,tlc,i,t where A is sectoral activity, Si is the share of subsectoral activity i (e.g., pulp and paper) in total industrial activity and I; is the energy intensity of i.  For the comparison, industrial energy use in Canada is calculated as if Canada had the same industrial structure as the country or country-grouping with which it is being compared: EI'c,t Ac,t 2 Sx,i,tlc,i,t =  2.5.4 Overall Comparison These three adjustments may then be aggregated to determine the extent to which they account for the difference in energy use between Canada and the country of interest. This is calculated as the difference between actual energy use (Ec,t) and adjusted energy use (E'c,t) in Canada in year t, divided by the difference between actual energy use in Canada and actual energy use in the country of interest (Ex,t), both in year t: Ec.t - E'ct Eat - Ex,t  where E'C,t = ER' ,t + EC ,t + ET' ,t + El'ct c  c  c  To avoid the influence of differences in population size, the comparison should be of per capita energy use: 55  (Ec.t - E'ctVpopct E ,t/popc,t - Ex,t/popx.t c  2.6 Data Sources Canadian energy use and activity data were drawn without modificationfromthe Natural Resources Canada energy use database (NRCan, 1997b). The sources of these data are described in some detail in NRCan, 1996 and NRCan, 1997a. As mentioned above, however, the description provided was not sufficiently comprehensive in some cases to permit data extension.  In addition to the NRCan (1997b) energy use data, industrial sector energy use data for the 1973-1983 were provided by Ralph Torrie. These data were developed for and used in the analysis in Marbek et al (1989) and are described therein.  The Canadian and international energy use and activity data used in Chapter 6 were provided by Lee Schipper and his colleagues at the Lawrence Berkeley Laboratories/International Energy Agency. It should be noted that the Canadian data used in Chapter 6 werefromthis database, rather than NRCan (1997b) to ensure consistency with the data for other countries. In most cases, Schipper based his data on NRCan (1997b).  56  Chapter 3 Decomposition of Energy Use Changes  This chapter examines changes in energy use in each energy-using sector (industry, freight transportation, passenger transportation, services and residential) and discusses reasons why these changes occurred. In particular, changes in energy use are broken down into changes due to activity changes, changes due to intensity changes, changes due to structural changes and, in the service and residential sectors, changes due to weather.  In general, changesfromthe early/mid-1980's to 1995/6 are examined, except in the industrial sector where two periods are examined: 1973-1983 and 1984-1996. Changes at the aggregate level are discussed first, followed by more detailed analysis of changes in each sector.  Comparisons of the results presented in this chapter with results obtained in other decompositions of energy use in Canada (NRCan, 1996 and 1997b; Schipper et al, 1997) are provided in the Appendix.  3.1 Overall Results Figure 3.0 and Table 3.0 show the shares of each sector in total energy use in Canada. In 1995, industry had the greatest share, at about 40 percent, followed by residential (about 19 percent), passenger transport (18 percent), services (13 percent) andfreighttransport  57  (9 percent). Over the period 1984 to 1995, the share of industry increased by about 3.5 percent, while the shares of the other sectors declined.  Overall, Canadian energy use increased by about 18 percent between 1984 and 1995. This increase was primarily due to increases in activity: if only activity had changed, overall energy use would have increased by about 28 percent. Offsetting the influence of increased activity was a decline in energy intensity which, if only intensity had changed, would have reduced energy use by about 8 percent over the period. Structure and weather had very little influence on energy use; structural changes on their own would have increased energy use by about 1.6 percent and weather by 0.3 percent (see Table 3.1).  Figure 3.1 shows the evolution of energy use and each of the contributing influences. The figure shows significantly increasing activity, which is partially offset by steadily declining intensity, leading to smaller increases in energy use. Structure and weather are insignificant influences, having magnitudes close to that of the residual of the decomposition.  This pattern is reproduced in each of the five sectors. The exception is industry during the period 1973-1983, in which the roles of activity and intensity were reversed: a decline in activity was offset by an increase in intensity, leading to a small overall decline in energy use.  58  3.2 Industrial Sector  Due to the availability of earlier data, decompositions were conducted for the industrial sector over two periods: 1973-83 and 1984-96. Two separate decompositions were conducted because energy use data in each period come from different sources. 1973-83 energy use data are based on Statistics Canada surveys of manufacturers, which have now been discontinued, while 1984-1996 energy use data are based on Statistics Canada surveys of energy producers. In addition, industrial subsectors in each period are slightly different . Results and discussion for each period are provided in separate sections 33  below.  3.2.1 Industry 1973-1983 Because of data limitations, only manufacturing and mining are examined in this period. From 1973 to 1983, manufacturing plus mining energy use declined slightly, by 0.60 percent. This was due to an activity effect of negative 6 percent, offset by a positive intensity effect of about the same amount. Structure (the share of each subsector's GDP in total manufacturing plus mining GDP) had very little influence on the change in energy use; in fact, the structure effect was significantly lower than the residual of the decomposition. The results of the decomposition are provided in Table 3.2.  Figure 3.2 shows the evolution of manufacturing plus mining energy use and contributing factors over the period. The activity effect and the change in energy use tend to move in tandem and are offset by the intensity effect, with the structural effect making a small  59  positive contribution to energy use change until the early 1980s, then a small negative change thereafter.  3.2.1.1 Activity Changes The negative activity effect through most of the 1970s and early 1980s reflects the recessions of those years, while positive activity effects in other years reflect economic recoveries. Figure 3.3 shows manufacturing plus mining real GDP (in $1981) over the period, and it shows an identical pattern to the activity effect, as would be expected. Figure 3.4 shows capacity utilization for several manufacturing sectors over the same period, and this shows the same pattern overall but with some variation among the sectors. Food and beverage production, for instance, weathers both recessions better than most other sectors.  The intensity effect tends to move conversely to the activity effect. This is because of the component of energy demand that is fixed in the short term. Manufacturing plant, and some equipment, must be powered independently of the level of production, so that when recession hits,firmsare only partially able to reduce energy demand in tandem with their level of production. This leads to temporary increases in intensity (which is measured as energy use - to some extentfixedin the short term - divided by output). These increases can be observed in the spikes in intensity effect in 1975 and 1982, the years following the onset of recession. In subsequent years,firmsare able to reduce thefixedcomponents of energy demand in response to lower production levels, by retiring plant and equipment, and intensity falls. The converse is true for economic expansions, again due to  F o r more detail o n these data sources, please s e ^ c t i o n 2.6 o f Chapter 2.  components of energy demand that are fixed in the short term. Another contributing factor is the availability of profits for investment in energy-saving equipment; in economic expansions, profits are higher, providing more possibility for investments of this type than in recessions (Marbek et al, 1989).  3.2.1.2 Digression on Prices In addition, the period 1973-1983 was marked by two significant world energy price increases brought about by OPEC-member production reductions. Figure 3.5 shows the average (nominal) world price of crude oil , which nearly quadrupledfrom1973 to 34  1974, and increased by 1172 percent from 1973 to 1983.  However, the federal government in Canada chose to shelter Canadian energy consumers from this price shock, to avoid importing foreign inflation (Wahby, 1984). It did this by freezing the price of domestically-produced oil and then subsidizing the cost of imported oil with an tax on Canadian oil exports equal to the difference between foreign and domestic prices (Thirsk and Wright, 1977). Thirsk and Wright (1977) estimate the size of the subsidy in 1974 to be on the order of $3 billion and they cite an estimate, made by Energy, Mines and Resources Canada, of $3 billion for 1975. Thirsk and Wright also estimate that if Canadian prices had attained world levels in 1974, energy consumption overall would have been 34 percent lower than it actually was in that year.  The world crude oil price is calculated as the average of the prices of Saudi Arabian Light 34° API, Iranian Light 34° APL Libyan Es Sider 37° API, Nigerian Bonny Light 37° APL Indonesian Minas 34° API and Venezuelan Tia Juana Light 26° API. Prices are as of the first Friday in January of each year. 3 4  61  Since the share of oil products in total manufacturing plus mining energy use is significant (rangingfromabout 19 percent in 1983 to 34 percent in 1973), the influence of the crude oil import subsidy is important. It is likely that if industrial producers had faced rapidly escalating oil prices, they would have diverted some of their energy demand to other sources, which they in fact did do. However, provided that it was not replaced with coal use, this decline in oil products use would have had a beneficial impact on greenhouse gas emissions.  The crude oil import subsidy was gradually phased out over the 1970s, as the Canadian government came to recognize that the rise in world crude oil prices was not a temporary phenomenon . In 1980, the National Energy Program was established, and was to have 35  allowed Canadian oil prices to gradually reach world levels (Margolick, 1997). World prices fell soon after, however.  Figure 3.7 shows the share of fuel costs in total production costs in several Canadian manufacturing sectors. The share rose over the period, with some small decline after 1982, reflecting the movement of overall energy prices, which, on a blended basis, rose by 427 percent from 1973 to 1985 (Marbek et al, 1989). While this price increase may seem significant, it should be noted that world crude oil prices rose about 980 percent over the same period.  Source: U.S. Energy Information Administration, Web Page: http://www.eia.dw.gov/emeu/international/prices.html.  62  Marbek et al (1989)findthat the Canadian manufacturing plus mining sector was not particularly responsive to energy price increases, calculating an aggregate energy price demand elasticity of between -0.08 and -0.11. They note that the business cycle - the presence or absence of recession — was the most important determinant of energy intensity changes.  3.2.1.3 Intensity Changes Figure 3.8 shows energy intensity (MJ per $ of sectoral real GDP in 1981 dollars) for mining and each manufacturing sector from 1973 to 1983. In most sectors, energy intensity increased by less than 10 percent and in some intensity declined. The exception is mining, in which intensity increased by 39 percent (see Table 3.4).  Figure 3.9 shows the evolution of aggregate energy intensity over the period for manufacturing plus mining and manufacturing. There was some variation in intensity over the period, corresponding to the business cycle, as explained above. Intensity rose at the beginning of the period, corresponding to the OPEC price-shock-induced recession, then fell as the economy recovered. Intensity rose again in the early 1980s as the economy went back into recession, and fell at the end of the period as the economy recovered. Over the period as a whole, manufacturing intensity fell by over 6 percent, while manufacturing plus mining intensity rose by almost the same amount, reflecting the large increase in mining energy intensity.  For instance. Energy, Mines and Resources Canada stated in 1976 in^w Energy Strategy For Canada that, "Domestic energy prices must continue to increase, to reinforce efficiency and restraint in energy  63  In their study of manufacturing plus mining energy use over 1973-1987, Marbek et al (1989) decompose changes in energy intensity in mining and each manufacturing sector into changes due to structural change and changes due to energy efficiency. They follow an approach similar to the decomposition method used above, except that the base year is 1987, and they do not calculate an activity effect. Theyfindthat over the period 36  manufacturing energy efficiency improved by 7.2 percent and manufacturing + mining energy efficiency improved by 5.8 percent, after accounting for the influence of structural change. In trying to replicate Marbek et ats results using their data, Ifindthat my results are slightly different. A comparison is provided in Table 3.5.  Marbek et al (1989) cite resultsfromthe Canadian Industry Program for Energy Conservation that show that, of the improvement in industrial energy efficiency 37  between 1973 and 1985,40 percent was due to housekeeping measures, 30 percent due to retrofit, 18 percent due to process change and 12 percent due to product change.  Based on extant analyses of the potential for energy conservation in industry , Marbek et 38  al estimate the economic potential for energy conservation in industry to be 30 percent, of which only 6 percent had been achieved by 1987. They estimate the remaining economic energy conservation potential in each industry as follows:  use..."; p. 146. (as cited b y T h i r s k a n d Wright, 1977: 356). I use 1973 as the base year. Energy M i n e s a n d Resources Canada (1987). "Canadian Industry Program for Energy Conservation", internal memo, Ottawa. A c r e s Consulting Services L t d . (1979). A Study of the Potential for Energy Conservation in Industry, Toronto, December; a n d M i n i s t r y o f State for Science and Technology (1982). Energy Conservation Technologies and Their Implementation. Ottawa. 3 6  3 7  3 8  64  Canadian  Food and beverage  15%  Paper and Allied  10%  Primary metals  15%  Nonmetallic minerals 20% Chemicals  45%  Other manufacturing 20% Mining  35%  3.2.1.4 Structure Changes As mentioned above, the structure effect was not a significant part of the change in energy use in manufacturing plus mining over the period 1973-1983. Figure 3.10 shows the shares of each subsector in total manufacturing plus mining real GDP over the period. While it appears from the figure that activity shares did not change much over the period, in fact they did, as shown in Table 3.6. In general, the activity shares of the most energyintensive subsectors declined over the period. However, the share changes cancelled each other out overall, leading to a very slight positive structure effect.  Structural changes have a positive impact on energy use when mining is included, but when mining is excluded, structural changes have a negative impact on energy use. This is because the share of mining in total activity declined considerably over the period, by almost 27 percent. Mining tends to be less energy-intensive than other subsectors so, all things being equal, a declining role for mining is a shift to more energy-intensive sectors and an increase in industrial average energy intensity.  65  Figure 3.11 shows the decomposition results for manufacturing only (i.e., with mining excluded). Comparing Figure 3.11 to Figure 3.2, the exclusion of mining has the effect of shifting up the activity effect and shifting down the intensity and structure effects. As can be seen in Table 3.2, which shows the decomposition results for 1983 compared to 1973, excluding mining reduces the decline in energy use by half and leads to a positive activity effect, significantly negative structure effect, and half as large intensity effect. In other words, manufacturing sector energy use declined by 0.34 percent, reflecting positive activity and intensity effects that were offset by a relatively large negative structure effect.  The intensity effect was positive, despite the overall decline in manufacturing intensity, because the sectors for which intensity increased (i.e., paper and allied products, primary metals and other manufacturing) together accounted for about 75 percent of manufacturing activity in 1973. Since the intensity effect is calculated using base year (1973) activity shares, the intensity increases in these subsectors outweigh intensity declines in other sectors and result in a positive intensity effect.  Table 3.7 shows the change from 1973 to 1983 in the share of total manufacturing GDP of each manufacturing subsector as well as their average energy intensities. The shares of most of the most energy-intensive subsectors declined over the period, with the exception of chemicals; this is reflected in the relatively large negative structure effect. 3.2.2 Industry 1984-1996  66  Decomposition results for 1996 (compared to 1984) are provided in Table 3.8. Over the period, total industrial energy use increased significantly, by about 23 percent. This was largely due to activity (industrial real GDP in $1986), which increased by a similar amount. Changes in intensity and structure also contributed to the increase in energy use, but were of about the same magnitude as the residual of the decomposition. Manufacturing energy use increased by almost 12 percent over the period, also reflecting positive activity, intensity and structure effects.  Figure 3.12 shows the decomposition of industrial sector energy use over the period 1984 to 1996. Like manufacturing and mining in the previous period, the activity effect and the change in energy use in industry in this period tend to move together. The intensity effect, which is quite small in this period, tends to move in an opposite direction. The structure effect is also quite small, and offsets the activity effect in some years while augmenting it in others.  Figure 3.13 shows the decomposition for manufacturing only, which shows the same patterns as the industrial decomposition, but at a lower level of magnitude for energy use and activity and a higher level of magnitude for the intensity and structure effects.  3.2.2.1 Activity Changes Clearly, in the period 1984-1996 increased activity was the major contributing factor to increased industrial energy use. From 1984 to 1996, industrial sector real GDP (in $1986) increased by about 23 percent. Figure 3.13 shows industrial sector real GDP over the  67  period. As would be expected, the path of real GDP matches the path of the activity effect.  3.2.2.2 Intensity Changes The changes in energy intensity observed in the 1984-1996 period are in general smaller than the economic potential for conservation estimated by Marbek et al (1989). This can be seen by comparing the last column of Table 3.9 with the middle column . The only 39  subsector that came close to realizing its potential for conservation was iron and steel (primary metals) in which intensity fell by almost 15 percent. The energy intensities of cement and chemical production and other manufacturing also fell, but by less than the estimated economic potential. Pulp and paper and mining, activities for which relatively large conservation potentials were estimated, both show significantly increased energy intensity. Aggregate industrial intensity increased by 0.21 percent, rather than falling by the 30 percent that was estimated to be its conservation potential.  The evolution of subsectoral intensities over the period is shown in Figure 3.15 and aggregate level intensities in Figure 3.16. As in the earlier period, intensities in this period tended to increase in times of recession and decrease in times of economic recovery and expansion, for the reasons mentioned above (fixed energy demand in the short term and availability of profits for conservation investments).  It should be recognized that the subsector definitions used by M a r b e k et al (1989) are different from the ones used i n Table 3.9. See section 2.6 o f chapter 2.  68  3.2.2.3 Process-level Intensity Changes NRCan (1996; 1997) discusses several process-level changes which led to the subsectoral intensity changes.  The decline in cement production energy intensity reflects a switch from wet to dry process cement production, which uses about a third less energy. Dry process accounted for about 85 percent of cement production in 1994, compared with 70 percent in 1984 (NRCan, 1996: 32).  Chemical production energy intensity declined due to a shift from heavy fuel oil and steam to natural gas and electricity (the latter have higher conversion efficiency than the former, which means that less energy must be consumed to produce the same level of energy service) as well as a shift in productionfromchlorine and caustic soda to sulphuric acid. Sulphuric acid requires only 0.03 GJ of energy per tonne of product, compared to chlorine and caustic soda, which are co-produced and require 30 GJ per tonne of product (NRCan, 1997: 47).  The decline in iron and steel energy intensity reflects a shift to electric-arc furnace technology, which uses about 13 percent of the energy of an integrated mill; the share of steelfromscrap produced using this technology increasedfrom26 percent in 1984 to 33 percent in 1994 (NRCan, 1996: 31). Other developments in the iron and steel sector which reduced energy intensity were the shiftfromcoal and coke to natural gas (natural gas has more energy content) and the replacement of ingot casting with continuous  69  casting (the latter's share increasedfrom39 percent in 1984 to 97 percent in 1995) (NRCan, 1996: 31; NRCan, 1997: 46).  The increase in mining energy intensity reflects shifts in productionfromdownstream metal and non-metal mining to upstream oil and gas mining ; the latter tends to be much 40  more energy-intensive than the former. It also reflects a shift from metal mines to nonmetal mines; metal mines use about 5 times less energy than non-metal mines (NRCan, 1997: 45).  In pulp and paper production, there was a shift from more energy-intensive chemical pulping to less-intensive mechanical pulping; chemical pulping uses about 20 percent more energy than mechanical pulping. Chemical pulping accounted for 43 percent of production in 1994, compared with 55 percent in 1984 (NRCan, 1996: 32). Another development that decreased energy intensity was the shift to recycled paper production, which uses only 17-23 percent of the energy needed for virgin production. In 1990, the share of recycled paper and board production was 11 percent; it jumped to 22 percent by 1995 (NRCan, 1997a: 44). However, these improvements were more than offset by a shift from oil to self-generated fuels (wood wastes and pulping liquor). Self-generated fuels have a lower conversion efficiency than fossil fuels, which means that more fuel must be consumed to provide the same energy service level. Overall, therefore, there was an increase in pulp and paper energy intensity (NRCan, 1997a: 44).  E.g., oil sands, potash 70  The significant decline in petroleum refining energy intensity reflects energy efficiency improvements (energy use per m of various fuel production declined by about 2 percent 3  between 1990 and 1995) as well as downsizing and restructuring in the industry, in which several refineries and distribution terminals were shut down, leading to an improvement in overall system efficiency (NRCan, 1997: 46).  The energy intensity decline in smelting and refining reflects replacement of older, less energy-efficient aluminum smelters with new energy-efficient smelters: 69 percent of aluminum production capacity in 1994 was based on smelters using less than 18 megawatt-hours per tonne of aluminum, compared with only 5 percent in the early 1980s. Aluminum production accounted for 72 percent of smelting and refining energy use in 1994 (NRCan, 1996: 32).  3.2.2.4 Structure Changes The structure effect at the industrial level was very low, about 0.8 percent. This masks considerable change in subsectoral shares, however, which tended to cancel each other out. The combined activity share of the six most energy-intensive subsectors (pulp, paper and sawmills; iron and steel; smelting and refining; cement; chemicals and petroleum refining) fell by less than two percent. The changes in subsectoral activity shares and average energy intensities are provided in Table 3.10.  The structure effect was larger for manufacturing only, reaching nearly 4 percent. Again, significant changes in subsector shares tended to cancel out, but in the case of  71  manufacturing, the significant increase in smelting and refining's share and the increase in petroleum refining's share were not as offset by declines in other energy-intensive subsectors' shares and increases in less intensive subsectors' shares. Table 3.11 shows the change in manufacturing subsector activity shares.  3.3 Freight Transport From 1984 to 1996, freight transport energy use increased by 20.5 percent, due to increased activity levels (tonne-kilometres) and structural change towards more energyintensive forms of transport. If only activity had changed, freight transport energy use would have been about 14 percent higher in 1996 than in 1984; if only the structure of freight transport (mode shares) had changed, energy use would have been about 20 percent higher. These developments were offset by a decline in energy intensity, if only intensity (GJ/tonne-km) had changed, energy use in 1996 would have been almost 10 percent lower than in 1984 (see Table 3.12).  Figure 3.17 shows the evolution of freight transport energy use and contributing factors over the period. The activity effect tends to move in concert with energy use, while the intensity effect tends to move in an opposite direction. The structure effect rises significantly throughout the period.  3.3.1 Activity Changes Freight activity, measured as tonne-kilometres (t-km), increased significantly over the period, as shown in Figure 3.18. From 1984 to 1996, freight t-km increased by about 14  72  percent. There was some variation over the period, however, as can be seen in Figure 3.18. This variation appears to correspond with the business cycle (represented in Figure 3.18 by Goods Production Real GDP); in periods of recession, freight transport activity falls and then it rises again in periods of recovery and expansion. This is logical: if fewer goods are being produced, as is the case during a recession, there is less to transport and freight activity falls. The path shown by freight transport t-km in Figure 3.18 matches the path of the activity effect in Figure 3.17, as would be expected.  3.3.2 Structural Changes The most significant change in the structure of freight transport over the period is the shift from marine and rail to trucks. The share of trucks in totalfreightactivity increased by 55 percent over the period, while the shares of rail and marine declined by 3 percent and 18 percent, respectively (see Table 3.13). This shift is at least partially in response to the advent of "just-in-time" production/delivery in which intermediate production and final retail goods are ordered and delivered as they are needed, rather than being held in inventory on-site. This kind of production favours trucks over other modes because trucks can offer more flexibility in terms of load size and timing of deliveries. While the share of trucks in totalfreightactivity has grown considerably, it still accounts for less than 25 percent of total activity. Rail continues to enjoy the largest share, at about 46 percent in 1996, while marine accounts for about 31 percent. At the same time, trucks accounted for almost 73 percent of totalfreighttransport energy use in 1996 (see Table 3.14). However, the share of trucks infreightenergy use grew less than 5 percent over the period, compared to the over 55 percent increase in the share of trucks in freight  73  activity. This reflects considerable improvement in truck energy intensity, as discussed in the next section.  3.3.3 Intensity Changes As mentioned above, if only freight intensity had changed, freight energy use would have fallen by nearly 10 percentfrom1984 to 1996. Despite falling energy intensity for almost every mode, overallfreightenergy intensity increased by over five percent. This is because of the large increase (26 percent) in marine transport energy intensity; marine transport accounted for on average 34 percent of totalfreightactivity over the period, making any change in marine intensity important for overallfreightintensity. Figure 3.19 and Table 3.15 show changes in energy intensity by freight mode.  It seems counterintuitive that the intensity effect is negative at the same time that total energy intensity has increased. This is because of the way the intensity effect is calculated, as the sum of changesfrom1984 to 1996 in energy intensity for each mode times the 1984 share of each mode. In this calculation, the negative intensity effects of rail and trucks more than offset the positive intensity effect of marine, resulting in an overall negative intensity effect.  Figure 3.20 shows marinefreighttransport energy use and activity (t-kms) over the period. In the mid- to late-1980s, marine energy use and activity moved together, but after 1988 they began to diverge. Unfortunately, detailed data on marine energy use and activity are not available and so it is not possible to determine why marine transport  74  energy use and activity diverged in the late 1980s and early 1990s. A possible explanation is the recession of the early 1990s, but while this may explain why marine activity levels fell, it does not explain why marine energy use increased.  Figures 3.21 to 3.24 show energy use and activity levels for the other freight transport modes over the period. Truck energy use and activity move together throughout the period, whereas rail energy use and activity levels diverge throughout the period. Again, in the absence of detailed data on rail energy use and activity, it is not possible to determine why this is so.  3.3.4 Freight Truck Intensity Changes Overall, freight truck energy intensity declined by almost 30 percent (see Table 3.15). There was some variation in the intensity improvement among the different weight classes: the intensity of light trucks fell the most, by almost 22 percent, followed by heavy trucks, declining almost 17 percent, and then medium-heavy trucks, declining almost 15 percent.  These changes in energy intensity reflect several factors: truck fuel efficiency (i.e., energy per kilometre, usually in litres per 100 kilometres), the size of the truck stock, the distance each truck is driven on average each year, tonne-kilometers per unit of G D P and the size of the G D P , or:  Energy = t-km  Energy * km * truck stock km truck t-km * GDP $ G D P  75  Freight trucks are fueled by either gasoline or diesel; diesel trucks come in three weight classes (light, medium-heavy and heavy) and gasoline trucks come in two weight classes (light and medium-heavy). Table 3.16 shows changesfrom1984 to 1995 in truck stock, stock fuel efficiency, average distance per truck, average tonne-kilometres per $ of Goods Production real GDP and Goods Production real GDP, for each weight class and each fuel type.  Fuel efficiency improved in every weight class and fuel type; the improvements ranged from almost 10 percent for medium-heavy diesel trucks to over 20 percent for gasoline light trucks. The size of the truck stock increased in almost all weight classes and fuel types, except for medium-heavy gasoline trucks, which fell by over 60 percent. The biggest increase was in the stock of diesel-powered medium-heavy trucks which increased by about 185 percent. Overall, the truck stock increased by about 41 percent, from about 3 million trucks in 1984 to nearly 4.3 million trucks in 1995.  The distance driven per truck increased considerably in every weight class and fuel type; the increases rangedfrom22 percent in light gasoline trucks to 31 percent in mediumheavy diesel trucks. Tonne-kilometres per unit of GDP increased by almost 9 percent and over 50 percent for light and heavy trucks, respectively, and fell by over 25 percent for medium-heavy trucks. Goods production real GDP increased by 22 percent. The share of light trucks in total truck tonne-kilometres fell by over 30 percent, but was insignificant (less than one percent) in any case. The shares of medium-heavy and heavy  76  trucks in total truck tonne-kilometres stayed fairly constant, at 11 percent and 88 percent, respectively.  To summarize, both the size of the truck stock and the activity level per truck (kilometres driven) increased in almost every weight class and fuel type, as did goods production real GDP and tonne-kilometres per unit of goods production real GDP. This increase in activity was more than offset, however, by improvements in truck fuel efficiency, so that overall energy intensity fell over the period.  3.4 Passenger Transport Over the period 1984 to 1996, passenger transport energy use increased by over 20 percent. This increase was largely due to an increase in activity (passenger-kilometres), which was partially offset by improvements in intensity (MJ/passenger-km). If only activity had changed, passenger transport energy use would have increased by nearly 45 percent; if only intensity had changed, energy use would have fallen by about 18 percent. Structure (the share of each mode in total activity) had a very small positive influence on energy use: if only structure had changed, energy use in 1996 would have been only 0.76 percent higher than it was in 1984. Table 3.17 presents the decomposition results for 1996; the evolution of energy use and each of the factors is presented in Figure 3.25.  3.4.1 Activity Changes Passenger transport activity, measured as passenger-kilometres, increased considerably over the period 1984 to 1996, by almost 45 percent. This was partially as a result of  77  population growth (a total of about 15 percent from 1984 to 1995) but also reflected an increase of about 22 percent in passenger-kilometres per person. Figure 3.26 shows the evolution of passenger-kilometres and population over the period.  3.4.2 Structural Changes The shares of total passenger-kilometres by mode changed significantly, for most modes, over the period 1984-1996. Figure 3.27 shows the initial and final distribution of activity shares by mode. In general, the shares of the less energy-intensive modes fell, while the shares of the more intensive modes increased. The share of rail, already small in 1984, fell sharply, so that by 1996 it accounted for only 0.24 percent of total passenger transport activity. The share of bus travel also fell, by nearly 18 percent.  The share of air travel increased by about 18 percent, rising from about 11 percent of passenger-km in 1984 to almost 13 percent in 1996. Overall, the share of light vehicles (cars and light trucks) in total passenger transport activity increased very slightly, by 0.17 percent. Table 3.18 shows the change in activity shares as well as the average energy intensity of each mode. Although the shares by mode changed significantly for some modes, the structure effect was small. Light vehicles account for about 80 percent of passenger transport activity and energy use, so the small change in the share of light vehicles mostly offset the larger changes in the shares of the other modes, resulting in only a small structure effect.  78  3.4.3 Intensity Changes Figure 3.28 and Table 3.19 show the change in energy intensity by passenger transport mode over the period 1984 to 1996. The energy intensity of all modes except buses fell dramatically. The energy intensity of rail and air travel fell by 37 percent and 22 percent, respectively, while the energy intensity of bus travel fell by only two percent. The energy intensity of small cars fell by about nine percent, large cars 26 percent and light trucks 14 percent, resulting in an overall decline of 17 percent in light vehicle energy intensity.  The improvement in air travel energy intensity was largely due to greater fuel efficiency from fleet renewal; there was little change in the ratio of passenger seating utilization to capacity (NRCan, 1996). The decline in rail energy intensity reflects the elimination of low capacity and low profit lines (NRCan, 1997a). The small decline in bus energy intensity reflects fewer riders and lower capacity utilization, among other factors (NRCan, 1997a). Factors contributing to the change in car energy intensity are discussed below.  3.4.4 Car Intensity Changes Detailed data on energy efficiency and activity are not available for most modes, but they are available for cars. Small and large car energy intensity is a function of the size of the car stock, the stock fuel efficiency (usually expressed in litres per 100 kilometres), the average distance driven per car, passenger-km per person and population size, or.  79  Energy = p-km  Energy * km * car stock km car p-km * population person  Data on stock size, stock fuel efficiency and average distance per car are available for both small cars and large cars. Table 3.20 presents changes in these variables, plus changes in passenger-km per person and population size, over the period 1984 to 1995. Car fuel efficiency improved considerably; small cars by about 14 percent and large cars by over 30 percent. Because the share of large cars in total car stock fell, total car fuel efficiency improved by about 24 percent. Almost all of the increase in car stock was small cars, which increased by 43 percent. The number of large cars increased by only 0.15 percent.  The improvement in fuel efficiency was more than offset by the increase in car stock (almost 23 percent) and an increase in average distance driven per car (almost 18 percent), leading to a nine percent increase in car energy demand. However, car energy intensity fell because passenger-kilometres (i.e., the denominator of the energy intensity expression) increased by almost 35 percent; small car passenger-km increased by 56 percent while large car passenger-km increased by about 10 percent. This increase in 41  passenger-km was due to an increase in passenger-km per person of almost 17 percent augmented by a 15 percent increase in population.  O v e r the same period, car energy use increased b y nine percent. Because the increase i n energy use is smaller than the increase i n passenger-kilometres, energy intensity declined.  80  3.5 Service Sector From 1981 to 1995, total service sector energy use increased by 18 percent. This was mostly due to a nearly 60 percent increase in service sector activity (square metres of floor space) offset by a considerable improvement in energy intensity (GJ/m of floor 2  space). If only service sector activity had changed, service sector energy use would have increased by 57 percent; if only intensity had changed, energy use would have fallen by 26 percent. Structural change (the share in total service sector m of each building type) had a small, 2  positive impact on energy use, but the structure effect was much smaller than the residual of the decomposition. If only structure had changed, energy use would have increased by 1.4 percent.  Weather also influenced service sector energy demand, through its effect on space heating and cooling demand. If only weather-adjusted energy intensity had changed, service sector energy use would have fallen by nearly 30 percent from 1981 to 1995. If only weather had changed, service sector energy use would have been about 4 percent higher in 1995 than 1981. Table 3.21 shows the decomposition results for 1995, with base year 1981.  Figure 3.29 shows the evolution of service sector energy use and contributing factors over the period. A strong activity effect is partially offset by a strong intensity effect through all of the period, resulting in a positive, but smaller, increase in energy. Energy use and the activity effect rise throughout the period while the intensity effect has a  81  progressively larger negative influence on energy use. The structure and weather effects are relatively constant over the period.  42  3.5.1 Partial Decompositions By Building Type A partial decomposition of energy use was conducted for each service sector building type, the results of which are presented in Table 3.22. Although energy use increased at the sectoral level, it did not increase in all building types. Energy use declined in schools, religious and health-related buildings and warehouses, and increased in Other Institutional, office, retail and recreation-related buildings and hotels and restaurants.. The increase in energy use rangedfrom20 percent in Other Institutional buildings to almost 60 percent in office buildings. The decline in energy use rangedfromfour percent in health-related buildings to 25 percent in warehouses.  The activity effect was large and positive for all building types except warehouses and was offset in all cases by negative intensity effects. Weather had a positive influence on energy use for all building types, but, at the same time, the weather-adjusted intensity effect was larger than the non-adjusted intensity effect. As in the sectoral-level decomposition, the residuals of the building-level decompositions were relatively large, reflecting the interaction of the activity and intensity effects.  W h a t is mteresting about the evolution o f service sector energy demand and its components is the steadily increasing (negative) residual o f the decomposition. A s shown i n Chapter 2, t i e residual is the interaction o f the Laspeyres indices (i.e., the activity, intensity, structure a n d weather effects) w i t h each other. T h e g r o w i n g residual therefore reflects the growth i n the Laspeyres indices, but i n this case i n particular the growth o f the activity a n d intensity effects. T h e interaction term o f these two effects i s larger than the other interaction terms a n d becomes increasingly large over time, resulting i n a growing residual. 4 2  82  3.5.2 Activity Changes From 1981 to 1995, service sector activity (m of floor area) increased by almost 60 2  percent. Over the same period, population grew by 19 percent, the service sector labour force grew by 27 percent and service sector GDP grew by 40 percent. Each of these 43  factors probably contributed to the growth in service sectorfloorarea: population growth because of its effect on the demand for education, health, religious and other institutional services and therefore facilities, and services sector labour force and GDP for their influence on the demand for commercial facilities. Figure 3.30 shows service sector floor area, GDP and labour force as well as population over the period 1981 to 1995.  3.5.3 Structural Changes From 1981 to 1995 there have been significant changes to the distribution of service sectorfloorarea across building types, as shown in Table 3.23 and Figure 3.31. In general, the share of institutional buildings has declined while the share of commercial buildings has increased. The shares of the most energy-intensive kinds of buildings, hotels and restaurants and health-related buildings, increased slightly or fell. The share of the least energy-intensive kind of buildings, warehouses, fell by almost 40 percent. Overall, the increases in the shares of some building types were mostly offset by decreases in the shares of others, resulting in a small structure effect.  Figures 3.32 to 3.40 show the evolution offloorspace shares for each kind of service sector building. Most non-commercial building types show a steady decline infloorarea  Service sector labour force data for 1981 a n d 1982 were not available, therefore the growth i n labour force i s calculated from 1983 to 1995. 4 3  83  share; the exception is Other Institutional buildings, which show a steady increase in share. Office buildings and recreation-related buildings show a small, but steady increase in share, while warehouses show a fairly large and steady decrease in share. The share of retail buildings increased considerably until the early 1990s and has been falling since. Similarly, hotels and restaurants show an increasing share until 1987 and variable, but trending downward, share thereafter.  NRCan (1996; 1997a) discuss the evolution of service sector floor areas by building types. During the 1980's, there was significant investment in commercial buildings, which tended to increase their shares in total service sector floor area (NRCan, 1996). After 1990, however, recession in the economy reduced demand for additional commercial building space and the rate of investment fell (NRCan, 1997a). Commercial floor space increased at a rate of 2 percent per yearfrom1990 to 1995, compared to 4.8 percent per yearfrom1984 to 1995 (NRCan, 1997a: 30).  3.5.4 Intensity Changes While service sector activity increased considerably, service sector energy intensity fell significantly. Table 3.24 and Figure 3.41 show the evolution of energy intensity by building type over the period 1981 to 1995. Energy intensity fell in all building types; the decline in intensity ranged from 15 percent in hotels and restaurants to 36 percent in schools. Most of the decline in energy intensity occurred in the 1980's; after 1990, energy intensity is essentially flat in all building types. In all buildings, adjusting  84  intensity for weather increased the decline in intensity by about three percent, because 1981 was a milder year than 1995.  Table 3.25 shows energy use, floor area and energy intensity by building type in 1981 and 1995. At the sectoral level, the decline in energy intensity is due to the fact that floor area increased by more than energy use. In some building types, there was an absolute decline in energy use, despite increasing floor area. These building types are schools, health, religious and warehouses. In all the other kinds of buildings, energy intensity declined for the same reason it declined at the aggregate level: floor area increased by a larger amount than did energy use.  At the end-use level, there were energy efficiency improvements but also increases in the penetration of some kinds of office equipment. Changes to space heating equipment have improved energy efficiency by up to 10 percent since the early 1980s (NRCan, 1996). Improvements in fluorescent lighting (which accounts for about 70 percent of office lighting energy use) have reduced lighting energy intensity, and new, more efficient systems (such as T-8 systems, which were not commercially available in 1984) are now capturing a large share of the market (NRCan, 1996). The energy efficiency of space cooling systems improved by about 25 percent since 1984; this was offset, however, by increased penetration of some kinds of office equipment, the heat discharge of which led to increased space cooling demand (NRCan, 1996).  85  There was a 500 percent increase in sales of microcomputers in Canada between 1985 and 1994; about 75 percent of microcomputers are sold to business, government and education, so most of the increase in computer-related energy demand was in the service sector (NRCan, 1996). Sales of fax machines increased by a factor of 10from1986 to 1994 and laser printer sales increasedfromalmost nothing in the mid-1980's to more than 250,000 units in 1994 (NRCan, 1996). Increased penetration of these kinds of office equipment has been augmented by trends towards more powerful machines, leading to significant increases in office equipment-related energy demand (NRCan, 1996).  3.6 Residential Sector From 1981 to 1995, residential energy use increased by about 13 percent. This was due to a large increase in activity (number of households), partially offset by a reduction in energy intensity (TJ/household). Because many end-use applications are used in all or most Canadian households, it was not possible to calculate a structure effect at the sectoral level (see section 2.2.3 of Chapter 2 for a derivation of the partial 44  decomposition without a structure effect; see the Appendix for a discussion of how a structure effect was calculated in other studies).  If only activity had changed, residential sector energy use would have increased by almost 32 percent. If only energy intensity had changed, residential energy use would have fallen by about 14 percent.  Detailed end-use data by housing type are available for Canada from at least 1981 onwards, i f not earlier. Unfortunately, I was not able to obtain these data despite repeated requests to Natural Resources Canada, w h i c h produces the data. Clearly, this analysis w o u l d benefit from the use o f these more detailed data.  86  As in the service sector, incorporation of weather increases the intensity effect by about three percent. Therefore, if only weather-adjusted intensity had changed, residential energy use would have fallen by about 16 percent. If only weather had changed, residential energy use would have increased by over three percent. This is because both heating degree-days and cooling degree-days were higher in 1995 than they were in 1981 Table 3.26 shows the change in energy use and contributing factors from 1981 to 1995.  Figure 3.42 shows the evolution of residential sector energy use and contributing factors from 1981 to 1995. The change in energy use shows some variation over the period, and appears to coincide with changes in weather until about 1990, after which they diverge. Activity shows a steady and increasing positive effect on energy use, while intensity shows a more variable, but overall increasing, negative influence on energy use. The residual of the decomposition grows over time, but remains at less than five percent.  3.6.1 Activity Changes The number of households in Canada increased by over 31 percent from 1981 to 1995. This reflects an increase in population but also a decline in the average size of households. Population grew by almost 19 percent from 1981 to 1995, while the number of persons per household fell by nearly 10 percent over the same period. Table 3.27 and Figure 3.43 show the evolution of the number of households, population and persons per household over the period.  87  3.6.2 Intensity Changes Overall, residential sector energy intensity (TJ/household) declined by almost 14 percent from 1981 to 1995. Adjusted for weather, the change in intensity is about 16 percent. The change in energy intensity reflects changes in several underlying factors, including end-use energy demand, the efficiency of space heating equipment, heatedfloorarea per household, the penetration of various appliances and air conditioners, the number of households and the distribution of households between various housing types (single family attached, single family attached and multiple-family). Each of these underlying factors will be discussed in more detail below.,  \ 3.6.2.1 End-Use Demand There arefivedifferent categories of residential end-use demand: space heating, space cooling, water heating, lighting, and appliances . Table 3.28 shows the change in total 45  energy demand and energy demand per household for each end-use over the period 19811995. The energy intensities of all end-uses but space heating increased over the period, due to larger increases in end-use energy use than in the number of households. The increase in energy intensity was fairly low, however, for most end-uses. The exception was space cooling, which registered an increase in intensity of almost 75 percent (and an increase in weather-adjusted intensity of over 110 percent, due to a 21 percent increase in cooling degree-days in 1995 compared to 1981).  Overall, sectoral energy intensity fell because the decrease in space heating energy intensity of over 22 percent more than offset the increases in the intensities of the other  88  end-uses. Space heating accounted for almost 64 percent of residential sector energy use on average over the period so any change in space heating intensity will have a significant influence on sectoral efficiency. Space cooling accounted for only 0.3 percent, on average, of residential energy use; because of this, the large increase in space cooling intensity did not have much impact on aggregate-level intensity.  3.6.2.2 Efficiency of Space Heating Equipment There was a significant improvement of the average efficiency of space heating equipment over the period. In 1981, about 73 percent of households used the most energy-intensive kind of heating equipment surveyed: normal efficiency heating equipment fueled by natural gas or oil. The remaining households used Other Equipment types (coal, propane, wood, electric baseboard, heat pumps and dual systems - electricity and oil, wood and oil, and wood and electricity). By 1995, less than half of households used normal efficiency equipment, over six percent used medium-efficiency natural gasor oil-fueled systems and four percent used high-efficiency natural gas- or oil-fueled systems. Over 40 percent of households used Other Equipment types. Table 3.29 shows the change in the percentage of households using equipment type, as well as the average energy intensity of each kind of equipment. Figure 3.44 shows the evolution of equipment shares over the period.  Table 3.30 shows the results of a decomposition of the change in space heating energy use over the period. Overall, space heating energy use increased only slightly, by less than two percent. If only activity had changed (the number of households), space heating  45  Some analysts list cooking as a sixth category, ggthis analysis, cooking is included under appliances.  energy use would have increased by over 30 percent. As discussed above and shown in Table 3.29, there was an increase in the use of more energy-efficient kinds of space heating equipment over the period. If only the distribution of space heating among different equipment types had changed, space heating energy use would have fallen by over six percent. If only the energy intensity of space heating equipment would have changed, space heating energy use would have fallen by 14 percent (or 18 percent if adjusted for weather). If only weather had changed, space heating energy use would have increased by overfivepercent, since 1995 was colder than 1981 (heating degree-days were about 5 percent higher in 1995 than in 1981). Figure 3.45 shows the evolution of space heating energy use and contributing factors over the period.  3.6.2.3 Heated Floor Area Per Household Heatedfloorarea per household increased from 105 m per household in 1981 to 111 m 2  2  per household in 1995, an increase of 5.8 percent. When combined with the 31 percent increase in the number of households, this results in a nearly 40 percent increase in total residential heated floor area,fromalmost 900 million m in 1981 to 1.25 billion m in 2  2  1995 (see Table 3.31). This is reflected in space heating energy demand, and acts to counteract gains in equipment efficiency and improvements in the thermal envelopes of buildings.  3.6.2.4 Appliance and Air Conditioner Penetration From 1981 to 1995 there was an increase in penetration of air conditioners and all appliance types. Increases in penetration rates rangedfromabout 5 percent for clothes  90  washers to almost 74 percent for air conditioners (see Table 3.32). By 1995, every Canadian household had a refrigerator or combination refrigerator/freezer, more than half had freezers, 78 percent had clothes washers, 73 percent had clothes dryers, almost half had dishwashers and more than a quarter had air conditioners. Figure 3.46 shows the evolution of penetration rates over the period.  This increase in penetration rates was partially offset by improvements in unit energy consumption for all appliances except refrigerators, which became larger and therefore show an increase of nearly 25 percent in unit energy consumption (NRCan, 1996) . The 46  improvements in unit energy consumption were considerable, rangingfrom11 percent in clothes washers to over 30 percent in dishwashers.  3.6.2.5 Number of Households and Distribution Across Housing Types As mentioned above, the number of households increased by about 31 percent between 1981 and 1995. Table 3.33 shows the distribution of households among four different building types: single family attached, single family detached, apartments and mobile homes. Of the four kinds of buildings, single family detached are the most energy intensive (in terms of energy use per household). Single family attached and apartments are less energy-intensive because heatfromadjacent households displaces some of the need for space heating. Mobile homes tend to be less energy-intensive than other single family detached homes because they are smaller in area.  Unit energy consumption data were not available for air conditioners. 91  Over the period 1981 to 1995, there was a small decline in the share of single detached houses and apartments, a larger decline in the share of mobile homes and a large increase in the share of single attached homes. The decline in the share of single detached homes should have a beneficial impact on residential sector energy use; in the absence of data on energy use per building type, it is impossible to say more about the impact of these shifts in building types on residential sector energy use.  3.6.2.6 Summary Improvements in the energy efficiency of space heating equipment and appliances (unit energy consumption), as well as shifts towards more energy-efficient kinds of heating equipment and housing types reduced energy intensity over the 1981-1995 period. These developments were partially offset, however, by increases in heated dwelling area, the penetration of appliances and air conditioners and the energy intensity of end-uses other than space heating.  3.7 Conclusion Energy use in Canada rose by 18 percent from 1984 to 1995, due primarily to increased activity, which was partially offset by declines in energy intensity. At the aggregate level, structure and weather had minor effects on energy use.  Energy use increased in all sectors (industry,freighttransportation, passenger transportation, services and residential). In most sectors, the rise in energy use reflects  92  increased levels of activity partially offset by declines in energy intensity. The exception is industry, where energy intensity increased in both periods (1973-1983 and 1984-1996).  The influence of structural changes on energy use varied somewhat from one sector to the next. Structural changes, in the form of intermodal shifts, had a large, positive impact on energy use in freight transportation. They had small, positive effects on energy use in passenger transportation and the service and industry sectors. The impact of structure on energy was more pronounced in manufacturing compared to industry as a whole; in the early period (1973-1983), the structure effect was 0.05 percent for industry as a whole and -9.1 percent for manufacturing; in the later period (1984-1996), the structure effect was 0.8 percent for industry as a whole and 3.8 percent for manufacturing. It was not possible to isolate the influence of structural change in the residential sector.  The impact of weather on energy use was calculated in the residential and service sectors. In both sectors, adjusting the intensity effect for weather increased the magnitude of the effect by about three percent. The effect of weather itself was to increase energy use in both sectors, since heating and cooling degree-days were both higher in 1995 than they were in 1981. The impact of weather was more pronounced in the service sector than in the residential sector, but in neither sector was it close to the magnitude of the activity and intensity effects. As mentioned in Chapter 1, GHG emissions are expected to have significant, and uncertain, impacts on the Earth's climate system. One possibility is more severe seasons: harsher winters and hotter summers. Canada's northern geographic position will likely mean that climate changes will be more severe than in countries in  93  more temperate latitudes. This could have significant impact on the demand for space heating and space cooling. Demand for the former may fall (if winters become milder) while demand for the latter rises (due to hotter summers). Alternative possibilities include higher space heating demand and lower cooling demand, rising demand for both, falling demand for both or changes in the timing, duration and frequency of peak demand periods.  Based on the results in this Chapter, one may conclude that activity has been the main driver of the increase in energy use that has been observed in every sector in Canada in the period since the early- to mid-1980's. In all sectors but industry, declines in energy intensity have helped keep the rise in energy use below what it might have been otherwise. Unfortunately, these improvements were not sufficient to keep energy use from growing. In addition, in all sectors but manufacturing, structural changes have been towards more energy-intensive activities, augmenting the positive influence of activity increases on energy use.  This suggests that future policy directed toward energy conservation in Canada should attempt to strengthen the recent, largely "autonomous" trend toward lower energy intensity. Whether policy should attempt to dampen activity or change the structure of that activity clearly depends on more than just the consequent impact on energy use. For instance, lower industrial activity would mean lower GDP, which is, for now anyway, not a goal of government or business. Shifting industrial production awayfromenergyintensive activities like pulp and paper production may be beneficial in terms of energy  94  conservation, but may have other, more socially-costly effects like unemployment and community decline.  In light of the need to reduce Canada's GHG emissions and the other benefits to be derived from reducing energy use, it would be advantageous to evaluate economic policies, which generally encourage growth, sometimes in the most-energy-intensive activities, in terms of their implications for energy use. The analysis presented in this chapter shows that increased activity will, all things being equal, mean increased energy use, and that changes in the structure of that activity also have important implications for energy use. Since the 1980's, both have led to increased energy use in Canada.  Changes in fuel mix were not considered in this chapter. They are, however, examined in the next chapter which discusses the decomposition of changes in greenhouse gas emissions in each sector.  95  Chapter 4 Decomposition of Greenhouse Gas Emissions  This chapter examines changes in emissions of three greenhouse gases (GHGs), carbon dioxide (CO2), methane (CH4) and nitrous oxide (N2O). These three gases, and not all GHGs, are considered because they are the ones for which emission coefficients are readily available. These gases accounted for 98.5 percent of Canada's C02-equivalent GHG emissions in 1994 (see Figure 1.1), and therefore account for the majority of GHG emissions.  Changes in emissions of these gases are decomposed into contributing factors in much the same way that changes in energy use were decomposed in Chapter 4. In fact, changes in emissions are equal to the energy use activity, structure, intensity and weather effects plus a new factor that considers the influence of fuel mix changes on emissions (the "fuel mix effect"), as well as new residuals that reflect the interaction of the fuel mix effects with the other effects. Changes in emissions are therefore equal to the fuel mix effect plus the changes in energy use plus (new) residuals.  47  Greenhouse gas emissions were calculated by multiplying energy use in each fuel type by emission factors for that fuel type (the emission factors are listed in Table 4.5). The source of the emission factors is Table S3 in Jaques, 1992. Unfortunately, emission factors for electricity were not provided in this table. Therefore, they were estimated by dividing electricity energy use in 1990 by the relevant emissions reported in Table S. 1 of  96  Jaques, 1992 for power generation. Electricity energy use was taken from NRCan, 1997b. This is obviously a simplification. Because the relative contribution of fossil fuels compared to other energy sources for electricity generation is crucial to the size of 48  the emission coefficients for electricity, a more rigorous analysis would have calculated the emission factor (in each year) as a weighted average of emission factors (weighted according to the share of each fuel source in electricity generation in each year). However, this was beyond the scope of this analysis and it is hoped that this back-ofenvelope approach will produce estimates that are roughly consistent with thosefroma more rigorous analysis.  In this chapter, overall changes in emissions are consideredfirst,followed by analyses of emission changes in each of the sectors.  4.1 Overall Results 4.1.1 Changes in Emissions Table 4.1 shows the changesfrom1984 to 1995 in emissions of each gas and in total energy use by sector and for Canada as a whole. From 1984 to 1995, emissions of CO2, CH4 and N2O increased by 14.6 percent, 5.8 percent and 4.2 percent, respectively, while energy use increased by almost 17 percent. Figure 4.1 shows the changes in each year 49  See section 2.2.5 o f Chapter 2 for a derivation o f the decomposition equation for emissions. as w e l l as the relative contributions o f o i l , natural gas a n d coal, since each has different emission coefficients. T h e change i n energy use is different here than the change shown i n Chapter 4. T h i s i s because "Other Fuels" are excluded from energy use i n the calculations o f emission changes a n d the associated decomposition. T h i s was necessary because the nature o f "Other Fuels" is not specified a n d therefore it is not possible to determine emission factors. 4 7  4 8  4 9  97  over the period. Emissions and energy use follow the same overall pattern, rising in periods of economic expansion and falling in the recession of the early 1990's.  Figure 4.2 shows CO2 emissions by sector in 1984, 1990 and 1995. CO2 emissions rose in every sector over the period. Figure 4.3 shows CH4 emissions, which fell significantly in all sectors but passenger transport. However, as shown in Figure 4.3 and Table 4.1, passenger transport accounted for the bulk of CH4 emissions over the period , so the rise 50  in passenger transport CH4 emissions more than offset the fall in emissionsfromother sectors, resulting in an overall rise in emissions.  Figure 4.4 shows N2O emissions by sector. Like CH4, N2O emissions fell significantly in most sectors. The exceptions were industry, where emissions rose slightly, and passenger transport, where emissions rose more significantly. Also like CH4, passenger transport accounts for the bulk of N2O emissions and, combined with industry, accounted for 75-85 percent of total N2O emissions. Therefore, the rise in emissionsfromthese two sectors more than offset the significant declines in emissionsfromthe other sectors, resulting in an overall increase in N2O emissions.  Canada has committed internationally to reducing its GHG emissions to 94 percent of 1990 levels by 2008-2012. In light of this commitment, it is important to examine how emissions have changed since 1990. As can be seen from the second-last column of Table 4.1, emissions of all three gases increased between 1990 and 1995. CO2 emissions  This is because there are relatively large CH4 emission coefficients for gasoline and aviation gasoline, two fuels which are important components of passenger transport energy use. 98  rose by 6.5 percent, CH4 emissions by almost 4 percent, and N2O emissions by 3.4 percent. Emissions of C 0 , the most important GHG, rose in all sectors. CH4 and N2O 2  emissions rose in some sectors, but fell in others, as discussed above. Emissions rose by slightly less than energy use, indicating that there was a switch to fuels with lower emissions per joule. This is discussed in more detail in the next section.  4.1.2 Emission Decompositions Table 4.2 presents the results of the decomposition of changes in emissions of each of the three GHGs, from 1984 to 1995. CO2 emissions rose by almost 15 percent from 1984 to 1995. This was primarily due to an increase in activity-induced energy use: if only activity had changed, emissions would have been 28 percent higher in 1995 than they were in 1984. Structural change also had a positive impact on CO2 emissions: if only structural factors had changed, emissions would have been about two percent higher in 1995 than they were in 1984. Intensity and fuel mix changes partially offset the activity and structure effects. If only fuel mix had changed, CO2 emissions would have fallen by 2.3 percent and if only intensity had changed, CO2 emissions would have fallen by over 17 percent.  The same is true for CH4 and N2O emissions, except that fuel mix changes were relatively more effective in reducing the increase in emissions. If only fuel mix had changed, CH4 emissions would have fallen by 10.4 percent, instead of the observed increase of 5.8 percent. Similarly, if only fuel mix had changed, N2O emissions would have fallen by 11.6 percent, instead of increasing by 4.2 percent.  99  Figures 4.5 to 4.7 show the decomposition results, with base year 1984, for each gas over the period. As can be seen in Figure 4.5, activity and structure had a positive and increasing influence on CO2 emissions while intensity had a negative and increasing effect on emissions. Fuel mix had a negative, but fairly constant effect on CO2 emissions. Similarly, activity and structure had a positive impact on CH4 and N2O emissions and were partially offset by intensity. Fuel mix had a larger, and increasing, negative impact on CH4 and N2O emissions.  Table 4.3 shows the results of a decomposition of changes in emissions of the three GHGs, this time with base year 1990. Between 1990 and 1995, C 0 emissions rose by 51  2  about 6.5 percent. This rise was due to a strong activity effect, augmented by a small, but positive structure effect and partially offset by negative intensity and fuel mix effects. If only fuel mix had changed (i.e., if energy use would have remained at 1990 levels), CO2 emissions would have fallen slightly, by about half a percent.  CH4 emissions also rose between 1990 and 1995, by almost four percent, again due to increased energy use. If energy use would have remained at 1990 levels and only fuel mix changed, CH4 emissions would have fallen by about 3.3 percent. Similarly, N2O emissions rose, by about 3.4 percent, from 1990 to 1995, but would have fallen by about the same amount if energy use had not risen and only fuel mix changed.  I.e.. Table 4.3 shows changes from 1990 to 1995. 100  Figures 4.8 to 4.10 show the decomposition results for each gas, with base year 1990. Emissions of all three gases actually fell between 1990 and 1992, apparently largely due to the recession-induced decline in activity in those years. Thereafter, however, emissions increased. The influences of the various effects are similar to the case with base year 1984: activity and structure tended to have a positive impact on emissions and were partially offset by intensity and fuel mix, which tended to have a negative influence on emissions.  4.1.3 Fuel Mix Changes Over the period 1984 to 1995, there was a shift towards fuels with lower emissions per unit of energy. In particular, there was a shiftfromoil and coal to natural gas (see Table 4.4). In 1984, oil products accounted for about 12.3 percent of total energy use; by 1990 their share had fallen to less than ten percent and by 1995 to about 7.6 percent. Coal and associated products accounted for about 6.8 percent of total energy use in 1984; by 1990, their share had fallen slightly and by 1995 their share was about 5.8 percent. Similarly, the share of fuel wood, while accounting for less than two percent of energy use in 1984, also fell over the period. At the same time, the shares of natural gas, liquid petroleum gases and electricity rose. Figure 4.11 shows the distribution of energy use across fuel types in 1984, 1990 and 1995, and Table 4.5 lists the GHG emission coefficients by fuel type.  4.2 Industry, 1984 to 1996 4.2.1 Changes in Emissions  101  Industrial sector emissions of all three gases rose over the period 1984 to 1995. CO2 emissions rose by over seven percent in each of the periods 1984-89 and 1990-95, and rose by over 15 percent from 1984 to 1995 (see Table 4.1). Industrial CH4 emissions fell by about 1.4 percent from 1984 to 1989, but rose by almost four percent from 1990 to 1995, leading to a small rise of just over two percent over the entire period. Similarly, industrial N 0 emissions fell slightly from 1984 to 1989, then rose by almost ten percent 2  from 1990 to 1995, leading to an overall rise of about 6.6 percent between 1984 and 1995. Figure 4.12 shows industrial emissions of each of the three GHGs in 1984, 1990 and 1995.  Over the same period, industrial energy rose by over 18 percent, indicating that fuel switching helped reduce emissionsfromthis sector.  4.2.2 Decomposition Results Table 4.6 shows the results of a decomposition of industrial sector GHG emissions, with base year 1984. From 1984 to 1996, industrial CO2 emissions rose by over 17 percent , 52  largely due to increased activity and some structural shift towards more energy-intensive industrial sectors. This was partially offset by reduced energy intensity. If energy use had not changed and only fuel mix had changed, industrial CO2 emissions would have fallen by about two percent. Similarly, if only fuel mix had changed, industrial CH4 emissions would have fallen by over ten percent, rather than rising by almost eight percent, and industrial N 0 emissions would have fallen by almost 15 percent, rather than 2  102  rising by over two percent. Figures 4.13 to 4.15 show the decomposition results over the period for each of the gases (with base year 1984).  Table 4.7 presents the decomposition results for each of the gases, this time with base year 1990. From 1990 to 1995, industrial sector CO2 emissions rose by just over four percent, and emissions of CH4 and N2O fell, by about one percent and six percent, respectively. If energy use had not increased and only fuel mix had changed, emissions of all three gases would have fallen, by 0.7 percent for C 0 , 5.6 percent for CH4 and 10.6 2  percent for N2O. Figures 4.16 to 4.18 show the decomposition results with base year 1990, for each of the three gases.  4.2.3 Fuel Mix Changes Table 4.8 and Figures 4.19 and 4.20 show the shares of various fuels in industrial energy use in 1984, 1990 and 1996. Like Canada as a whole, industrial energy users switched from fuels with high GHG emissions per unit of energy, such as oil and coal products, to fuels with lower emissions, such as electricity, natural gas and liquid petroleum gases. The exception was petroleum coke and distilled gas, which accounted for 7.25 percent of industrial energy use in 1984 and 8.7 percent in 1995.  4.3 Freight Transport, 1984 to 1996 4.3.1 Changes in Emissions  It should be noted that Tables 4.6 a n d 4.7 present the results o f the decomposition i n 1996, w h i l e Table 4.1 shows changes only u n t i l 1995. Therefore, the emission changes reported i n Tables 4.6 a n d 4.7 are different from those reported i n Table 4.1.  103  Freight transport CO2 emissions rose significantly over the period 1984 to 1995, rising by almost nine percent in each of the periods 1984-89 and 1990-95 and by over 18 percent over the whole period (see Table 4.1). By contrast, freight CFLt emissions fell sharply, by over 37 percent over the whole period, reflecting decreases of about 18 percent from 1984 to 1989 and about 23 percent from 1990 to 1995. Freight emissions of N 0 also fell 2  significantly, by almost 15 percent from 1984 to 1989, over 20 percent from 1990 to 1995 and 32 percent over the whole period. Figure 4.21 showsfreighttransport emissions of the three gases in 1984, 1990 and 1995.  At the same time,freightenergy use also increased, by about the same amount as CO2 emissions, indicating that fuel switching, while not as significantly reducing C 0  2  emissions as was the case in other sectors, very considerably reduced CH4 and N2O emissions.  4.3.2 Emission Decompositions Table 4.9 shows the results of a decomposition offreighttransport emissions of each of the three gases, in 1996 with base year 1984. Emissions of all three gases changed considerably from 1984 to 1996: CO2 emissions rose by more than 20 percent, CH4 emissions fell by more than a third and N 0 emissions fell by about 30 percent. Unlike 2  other sectors, fuel mix changes had a small but positive influence onfreighttransport C 0 emissions: if only fuel mix had changed,freightCO2 emissions would have 2  increased very slightly. As it was, emissions actually increased by 20 percent, indicating that increased energy use (due primarily to increased activity and structural shifts towards  104  more energy-intensive modes, which were partially offset by reduced intensity) was the main reason for the increase in freight transport CO2 emissions.  Changes in fuel mix had a strong negative influence onfreighttransport emissions of CH4 and N2O: if only fuel mix had changed,freightemissions of CH4 and N2O would have fallen by 45 percent and 41 percent, respectively, which is more than they actually did fall. This indicates that increased energy use partially offset the beneficial impact of fuel mix changes. Figures 4.22 to 4.24 show the decomposition results (with base year 1984) over the entire period.  The changes in freight emissionsfrom1990 to 1996 and contributing factors are shown in Table 4.10. Emissions of CO2 rose by about 11 percent, while emissions of CH4 and N2O fell by 19 percent and 17 percent, respectively. Again, increased energy use partially offset the beneficial impact of fuel mix changes on CH4 and N 0 emissions, and 2  was the main reason for increased CO2 emissions. Figures 4.25 to 4.27 show the decomposition results with base year 1990 for each of the three gases.  4.3.3 Fuel Mix Changes Table 4.11 shows the shares of various fuels infreighttransport energy use in 1984, 1990 and 1986. There were some significant shifts in fuel mix over the period, most notably a large increase in the share of diesel (which grewfrom55 percent of total energy use in 1984 to 70 percent in 1996) and a large decrease in the share of gasoline (which fell from 34 percent of total energy use in 1984 to 18.6 percent in 1996). Diesel and gasoline have  105  similar CO2 emission coefficients, although the coefficient of gasoline is slightly lower (see Table 4.5). This explains the small but positive fuel mix effect on CO2 emissions. Gasoline has much larger emission coefficients for CH4 and N 0 , which explains the 2  much larger fuel mix effects for these gases . 53  There were also small shifts towards fuels with lower emission coefficients. The shares of propane and natural gas increased over the period, but in 1996 still accounted for less than three percent of total freight energy use. Figures 4.28 and 4.29 show the distribution of freight energy use by fuel in 1984, 1990 and 1996.  4.4 Passenger Transport, 1984 to 1996 4.4.1 Changes in Emissions Passenger transport emissions of all three GHGs rose significantly from 1984 to 1995 (see Table 4.1). C 0 emissions rose by 10 percent from 1984 to 1989 and by seven 2  percent from 1990 to 1995, leading to an 18 percent increase over the entire period. CH4 emissions rose by 6.4 percent from 1984 to 1989, 8.5 percent from 1990 to 1995 and 15.5 percent over the period as a whole. N 0 emissions rose by 16.5 percent over the whole 2  period, reflecting increases of 7.3 percent from 1984 to 1989 and 8.6 percent from 1990 to 1995. Figure 4.30 shows passenger transport emissions of the three GHGs in 1984, 1989 and 1995.  The CH4 a n d N 2 0 emission coefficients o f gasoline may seem high; the coefficients reported i n Table 4.5 a n d used i n this analysis are actually the mid-points o f a range o f values reported i n the source document. T h e reported ranges were 6.92 to 121.11 kilograms o f C H , a n d 6.6 to 47.6 kilograms o f N 2 0 per terajoule o f gasoline.  106  Passenger transport energy use rose by about 18 percent over the period, which is about the same magnitude as the increase in CO2 emissions, and only slightly higher than the increases in CH4 and N2O emissions. This indicates that fuel switching did not play a significant role in reducing passenger transport emissions.  4.4.2 Decomposition Results Table 4.12 shows the results of decompositions of passenger transport emissions in 1996, with base year 1984. Passenger transport CO2 emissions increased by over 20 percent from 1984 to 1996. This was largely due to increased energy use, reflecting increased activity partially offset by improved intensity. Fuel mix changes had little impact on CO2 emissions: if energy use had not increased and only fuel mix had changed, CO2 emissions would have fallen by only a tenth of a percent from 1984 to 1996.  The same is true for passenger transport CH4 and N2O emissions. Increased energy use was primarily responsible for the 16.6 percent and 18 percent increases in CH4 and N2O emissions, respectively, from 1984 to 1996. Fuel mix changes were more effective in reducing emissions of these gases: if only fuel mix had changed, CEU emissions would have fallen by 3.6 percent and N2O emissions by 2.5 percent.  Decomposition results (with base year 1984) for each of the three GHGs are shown in Figures 4.31 to 4.33. These figures show steadily rising emissions over the period, with the exception of the early 1990s, reflecting increased activity which was only partially offset by fuel mix changes and falling intensity.  107  Table 4.13 shows the decomposition results for passenger transport GHG emissions in 1996, this time with base year 1990. From 1990 to 1996, passenger transport emissions of each of the three GHGs increased by almost 10 percent. Again, this was primarily due to increased energy use. Fuel mix changes were not particularly effective in reducing emissions from 1990 to 1996. If only fuel mix had changed, CO2 emissions would have fallen by less than a tenth of a percent and C H emissions would have fallen by just over 4  a quarter of a percent. N2O emissions would have increased, by 0.14 percent. Figures 4.34 to 4.36 show the decompositions with base year 1990 for each of the three GHGs.  4.4.3 Fuel M i x Changes As discussed above, fuel mix changes were not especially effective in reducing passenger transport GHG emissions. Table 4.14 and Figures 4.37 and 4.38 show the distribution of passenger transport energy use by fuel in 1984, 1990 and 1996. Fuel shares did not change much over the period, except for slight increases in the shares of natural gas, propane and aviation turbo fuel, at the expense of motor gasoline and aviation gasoline. None of the fuel shares changed by more than three percent, and motor gasoline accounted for the bulk of passenger transport energy use throughout the period.  Motor gasoline, diesel and aviation turbo fuel produce relatively more emissions than propane, natural gas, electricity and aviation gasoline. Because the shares of the latter fuels did not increase significantly, the fuel mix effect was small.  108  4.5 Service Sector, 1981 to 1995 4.5.1 Changes in Emissions From 1984 to 1995, service sector emissions of C 0 increased while emissions of CH4 2  and N2O decreased (see Table 4.1). Service sector CO2 emissions increased by almost 14 percent, reflecting increases of almost four percent from 1984 to 1989 and almost 10 percent from 1990 to 1995. CH4 emissions fell by nearly 19 percent from 1984 to 1989, but then increased by just over five percent from 1990 to 1995, resulting in a decrease of almost 15 percent over the entire period. N2O emissions fell by nearly 30 percent from 1984 to 1989 and increased very slightly (by less than one percent) from 1990 to 1995, leading to a decline of about 29 percent over the period. Figure 4.39 shows service sector emissions of each of the three GHGs in 1984, 1990 and 1995.  Service sector energy use increased by over 15 percent over the 1984 to 1995 period, which is larger than the increase in emissions of any of the three GHGs. This indicates that fuel switching acted to reduce service sector GHG emissions.  4.5.2 Decomposition Results Table 4.15 shows the decomposition of service sector GHG emissions in 1995, with base year 1981. Service sector CO2 emissions rose by about 14 percent from 1981 to 1995, largely reflecting a considerable increase in activity which was augmented by weather and structural changes but partially offset by a decline in intensity. Fuel mix changes had a fairly small impact on CO2 emissions: if only fuel mix had changed, CO2 emissions would have fallen by 3.6 percent.  109  Service sector CH4 and N2O emissions fell dramatically from 1981 to 1995. CH4 emissions fell by over 25 percent, reflecting an increase in energy use which was more than offset by fuel mix changes. If only fuel mix had changed, CH4 emissions would have fallen by about 37 percent. Similarly, NjO emissions, which fell by almost 45 percent, would have fallen by almost 54 percent if only fuel mix had changed.  Figures 4.40 to 4.42 show the decomposition of service sector emissions of each of the three GHGs (compared to 1981) over the period 1981 to 1995. Thesefiguresshow, for CH4 and N2O, the strong offsetting influence of fuel mix changes on emissions compared to the relatively weak impact of fuel mix changes on CO2 emissions.  Table 4.16 shows the decomposition of service sector GHG emissions in 1995, this time with base year 1990. Emissions of all three GHGs increasedfrom1990 to 1995. CO2 emissions increased by almost 10 percent, reflecting increased energy use partially offset by fuel mix changes. Fuel mix changes were not especially effective in reducing service sector CO2 emissions. If only fuel mix had changed, service sector CO2 emissions would have fallen by less than half a percent from 1990 to 1995.  Service sector CH4 and NjO emissions increased by just overfivepercent and less than one percent, respectively, from 1990 to 1995, reflecting, like CO2, increased energy use partially offset by fuel mix changes. Fuel mix changes were more effective in reducing emissions of these gases. If only fuel mix had changed, CH4 emissions would have fallen  110  by about 4.5 percentfrom1990 to 1995 and N 0 emissions would have fallen by about 2  8.4 percent from 1990 to 1995.  Figures 4.43 to 4.45 show the decompositions, with base year 1990, of service sector emissions of the three GHGs.  4.5.3 Fuel Mix Changes Table 4.17 and Figures 4.46 and 4.47 show the distribution of service sector energy use across fuel types in 1981, 1990 and 1995. There was considerable change in fuel shares; the shares of electricity and natural gas increased steadily over the 1981-1995 period, at the expense of the shares of light and heavy fuel oils These changes acted to reduce emissions of all three GHGs because natural gas and electricity produce fewer emissions per unit of energy than light fuel oil and heavy fuel oil (see Table 4.5). The weakness of the fuel mix effect for CO2 emissions compared to CH4 and N2O emissions is largely explained by the increase in electricity: the CO2 emission factors for electricity, light fuel oil and heavy fuel oil are somewhat similar while the emission factors for CH4 and N2O are much lower for electricity than they are for light fuel oil and heavy fuel oil (see Table 4.5).  4.6 Residential Sector, 1981 to 1995 4.6.1 Changes in Emissions As in the service sector, residential sector emissions of CO2 increased from 1981 to 1995 while emissions of CH4 and N2O decreased (see Table 4.1). Residential CO2 emissions  111  increased by seven percent from 1981 to 1989 and by about 1.5 percent from 1990 to 1995, resulting in an increase of 8.6 percent over the 1981 to 1995 period. CH4 emissions decreased by about 4.5 percent from 1981 to 1989 and 4.75 percent from 1990 to 1995, leading to a decline of just over nine percent over the period as a whole. Residential N2O emissions fell by over 11 percent from 1981 to 1989 and by almost 13 percent from 1990 to 1995, leading to a decline of almost 30 percent from 1981 to 1995. Figure 4.48 shows residential sector emissions of each of the three GHGs in 1981, 1990 and 1995.  At the same time, residential sector energy use increased, by almost five percent from 1981 to 1989, just over 10 percent from 1990 to 1995 and over 15 percent over the whole period. The increase in energy use is considerably larger than the changes in GHG emissions, indicating that fuel mix changes were effective in reducing residential sector emissions.  4.6.2 Decomposition Results Table 4.18 shows the decomposition of residential sector emissions of the three GHGs in 1995, with base year 1981. CO2 emissions increased by almost seven percent from 1981 to 1995, reflecting increased energy use partially offset by fuel mix changes. If only fuel mix had changed, residential C 0 emissions would have fallen by almost six percent. 2  Residential CH4 emissions fell by almost 30 percentfrom1981 to 1995, reflecting increased energy use which was more than offset by fuel mix changes. If only fuel mix had changed, CH4 emissions would have fallen by almost 38 percent. Similarly,  112  residential N2O emissions fell by almost 35 percent from 1981 to 1995, but would have fallen even more (by over 42 percent) if only fuel mix had changed.  Figures 4.49-4.51 show the decomposition of residential GHG emissions, with base year 1981, over the 1982-1995 period.  Table 4.19 shows the decomposition of residential GHG emissions in 1995, but with base year 1990. Residential CO2 emissions increased slightly from 1990 to 1995, by 1.5 percent. This reflects increased activity augmented by weather, but almost completely offset by fuel mix changed and reduced intensity. CH4 emissions fell by almost five percent from 1990 to 1995, reflecting fuel mix changes which more than offset increased energy use. N2O emissions fell by almost 13 percent, again reflecting fuel mix changes which more than offset increased energy use.  Figures 4.52-4.54 show the decomposition of residential GHG emissions, with base year 1990, over the 1982-1995 period.  4.6.3 Fuel M i x Changes The largest change in residential fuel shares was a considerable decline in oil use, which was made up for by increased natural gas and electricity use (see Table 4.20 and Figures 4.55 and 4.56). Because oil is relatively more GHG-intensive than natural gas or electricity, this shiftfromoil acted to reduce emissions (as shown in the fuel mix effects discussed above). The fuel mix effect was stronger for CH4 and N2O emissions than for  113  CO2 emissions because of the shift from oil to electricity. Electricity has much smaller emission factors than oil for CH4 and N2O; while the CO2 emission factor for electricity is also lower than that for oil, it is not as substantially lower as the CH4 and N2O emission factors (see Table 4.5).  4.7 Summary and Conclusion From 1984 to 1995, national-level emissions of carbon dioxide, methane and nitrous oxide increased. CO2 emissions increased by more in the earlier part of this period (1984 to 1989) than in the latter part of the period (1990 to 1995); the reverse is true for CH4 and N2O emissions. At the same time, energy use grew by more than emissions, reflecting the beneficial influence of fuel mix changes on emissions. These patterns were also observed at the sectoral level.  In all sectors, increased energy use was the primary reason for emission increases. In most sectors, fuel mix changes mitigated the impact of increased energy use, by varying degrees. Infreighttransport and the service and residential sectors, fuel mix changes were significant enough to lead to reductions in CH4 and N 0 emissions, and lower 2  increases in CO2 emissions. In industry and passenger transport, fuel mix changes were less effective in reducing emissions. In general, fuel mix changes had a larger impact on CH4 and N2O emissions than on CO2 emissions. This is because the CH4 and N2O emission factors for electricity were much smaller than for most other fuels, especially the more GHG-intensive fuels.  114  In the post-1990 period, CO2 emissions grew in all sectors. Emissions of CH4 and N2O fell in industry, freight transport and the residential sector, and increased in passenger transport and the service sector. In almost all cases, emission increases were due to increased activity, partially offset by intensity reductions and fuel mix changes. The exception is passenger transport N2O emissions, where there was a small but positive fuel mix effect. Emissions in every sector declined in the early 1990's, reflecting the downturn in activity due to the recession.  Overall, CO2 emissions increased the least in the residential sector and the most in freight transportation (the increase in passenger transport C 0 emissions was nearly as high as it 2  was for freight transport). At the same time, CH4 and N2O emissions fell the most in freight transport, and the least in passenger transport. Unfortunately, passenger transport accounted for the majority of CH4 and N2O emissions, and about 20 percent of C 0  2  emissions, throughout the 1984 to 1995 period, so the failure in reducing emissions from this sector contributed to aggregate-level increases in emissions.  More positively, emissions of CH4 and N2O fell in most sectors. From 1984 to 1995, CH4 emissions fell infreighttransport and the service and residential sectors and increased by only 2.25 percent in industry. N2O emissions fell infreightand the service and residential sectors.  Unfortunately, however, aggregate emissions of all three GHGs have increased since 1990, which raises some concern about the likelihood of Canada achieving the  115  commitment of a six percent reduction in emissions from 1990 levels by 2008-2012 that is now formalized in the Kyoto Protocol. The next chapter, Chapter 5, extrapolates the historical trends examined in Chapters 3 and 4 to determine the magnitude of energy use and GHG emissions in the years 2000 and 2010.  116  Chapter 5 Extrapolation of Historical Trends in Energy Use and Greenhouse Gas Emissions  In this chapter, trends in energy use and greenhouse gas emissions over the 1980's and 1990's are extrapolated to derive estimates of energy use and emissions in the years 2000 and 2010. This extrapolation is conducted at the aggregate and sectoral levels. The purpose of the extrapolation is to estimate whether a "Business-As-Usual" approach, involving no significant policy changes, would be sufficient to meet Canada's commitment to reduce GHG emissions to 94 percent of 1990 levels by 2008-2012. Business-As-Usual is assumed to be represented by average annual growth rates in energy use by fuel and overall, as well as in the factors which drive energy use: activity, structure and intensity.  This approach assumes that these historical growth rates represent what might be expected in a future without government or other policy to effect significant changes in fuel shares or in the factors which drive energy use. It is believed that this is a reasonable assumption because the periodfromthe mid-1980's to the mid-1990's was marked by a relative paucity of government policy measures and programs to induce energy conservation and switching to less GHG-intensive fuels  54  Therefore, trends in activity,  structure, intensity and fuel shares in this period may be interpreted as what could prevail  F o r a discussion o f policy during the 1980's, see M a r b e k et al (1989); for a discussion o f more recent policy, see Chapter 1 o f this thesis and C O G G E R (1993) a n d M a r g o l i c k (1998).  117  in the absence of a concerted policy effort to reduce Canada's GHG emissions (and enery use). At the aggregate level and in each sector, three different average annual growth rates are used: 198x to 1989, 1990 to 199x, and 198x to 199x. These different growth rates are 55  used in order to compare energy and GHG emission projections based on pre-1990 growth rates versus post-1990 growth rates. This gives an indication of whether recent (i.e., post-1990) trends are towards higher or lower energy use and emissions.  Results for Canada as a whole are presentedfirstand are followed by results for each sector. The extrapolation results for energy use are presented first, followed by extrapolations of emissions of three greenhouse gases (GHGs), carbon dioxide (CO2), methane (CH4) and nitrous oxide (N2O). Two GHG emissions projections are made. The first is based on energy use and fuel mix both changing at historical rates and the second is based on frozen energy use (at 1995 or 1996 levels, depending on the sector) and fuel mix changing at historical rates. The latter extrapolation isolates the influence of fuel mix changes and highlights the contribution of energy use to GHG emissions.  5.1 Canada Overall 5.1.1 Energy Use Extrapolations In 1995, 7140 petajoules (PJ) of energy were consumed in Canada. If energy use grows at the average annual rate of growth that prevailed during the 1984-89 period, energy use  The years which begin and end the periods (as represented by the "x") differ from one sector to another and depend on data availability. Industry and freight and passenger transport data begin in 1984 and end in 1996; service and residential sector data begin in 1981 and end in 1995. For Canada overall, the period 1984 to 1995 is used as data are available for all sectors for these years. 55  118  in the year 2000 will be over 8000 PJ. In the year 2010, it will be more than 10,300 PJ. If, on the other hand, energy use grows at the lower average annual rate of growth that prevailed in the 1990-95 period, energy use will be 7750 PJ in 2000 and 9130 PJ in 2010.  56  As discussed in previous chapters, energy use depends on the level of activity, the structure of activity and the energy intensity of activity. Table 5.1 shows the results of calculations of future energy use based on changes in all three underlying factors, as well as changes in activity only, intensity only and structure only.  In all cases, projected energy use is considerable higher than actual 1995 energy use if it is calculated solely on the basis of changes in activity, relatively constant if calculated solely on the basis of intensity changes and considerably lower if calculated solely on the basis of structural changes. This shows that, in the past, activity has tended to increase energy use, intensity changes have tended to have a small but positive impact on energy use and structural changes have tended to reduce energy use. This also indicates that policy to reduce energy use might include measures to reduce activity levels and enhance trends in intensity and structural change.  Also in all cases, projected energy use is higher if calculated on the basis of the (combined) rates of change in activity, structure and intensity than if it were calculated on  In some cases, projected energy use is lower when the 1984-1996 rate is used. This is an artefact of the method used to calculate the annual rate of change, which is: r=(X,/Xo) - -l The projections corresponding to growth rates that prevailed during the entire 198x-199x period are included in reported results, but this is only for interest. Comparisons will be made only between projections based on growth rates prevailing in the 198x-1989 period and those based on 1990-199x growth rates. 5 6  1/(n  0)  119  the basis of the rate of change of energy use itself. This indicates that the combined positive influence of the three underlying factors on energy use grows with time.  Most importantly, this analysis indicates that "Business-As-Usual" will in most cases involve increased energy use. The exception would be if only structural trends continued, and all other factors remained at 1995 levels.  5.1.2 Projected Greenhouse Gas Emissions, with Changing Energy Use Emission projections were calculated by applying the emission factors provided in Table 4.5 of Chapter 4 to estimated fuel shares in each sector. The fuel shares were estimated by calculating energy use by fuel in 2000 and 2010 based on historical annual average growth rates in use of that fuel, summing this projection across fuel types and then determining the share of each fuel in this total. These fuel shares are then multiplied by projected total energy use (based on historical growth rates in total energy use) to derive the estimated energy use by fuel in the years 2000 and 2010. Calculations of emissions were conducted at the sectoral level, and the summed across sectors to derive the aggregate national estimates of emissions.  Table 5.2 shows the projections of total national emissions of the three greenhouse gases in the years 2000 and 2010. Actual values for 1990 and 1995 are also provided for comparison. The most important outcome of these projections is that in no case do emissions decline from 1990 levels. There is an apparent exception in the projections of emissions in 2010 based on 198x-1989 trends. As explained below in section 5.4 on  120  passenger transportation, this appears to be the consequence of an error in the passenger transportation natural gas energy use data. It also appears to be a consequence of a large increase in freight transport natural gas use from 1986 to 1987 (from about 0.5 PJ in 1986 to just over 1 PJ in 1987), which may be a data error or may simply be an increase in actual use. Therefore, aggregate national and passenger transport emission projections based on the 198x-89 trend should be treated with caution and, in fact, are assumed to be in error. Because the increase in freight transport natural gas use does not seem to be as anomalous as the passenger transport natural gas use increase, projections for freight transport are assumed to be legitimate, but should be treated with caution.  Figure 5.1 shows projected CO2 emissions as well as actual emissions in 1990 and 1995. The smallest increase in CO2 emissions occurs if fuel shares change at their 198x-95 average annual rates; the largest occurs if fuel shares change at their 198x-89 rates. This indicates that CO2 emissions grew at a larger rate in the 1980's than they did in the 1990's, which bodes well for the future. Unfortunately, keeping the rate of change in fuel mix to the rate that prevailed in the 1990's will still involve an increase in CO2 emissions, rather than the six percent decrease to which Canada has committed.  The reverse is true for CH4 and N2O emissions, as shown in Figures 5.2 and 5.3. The smallest increases in emissions of these gases would occur if fuel mix changed at the average annual rate that prevailed during the 1980s, and the largest increase would occur if fuel mix changed at the rate that prevailed during the 1990s. This does not bode well  121  for future emissions, and indicates that fuel switching would be a profitable area for GHG mitigation policy.  5.1.3 Projected Greenhouse Gas Emissions, with Frozen Energy Use If energy use is kept frozen at the 1995 level, projected GHG emissions would be lower than if energy use grew at historical rates. CO2 emissions do not fall below 1990 levels under any scenario, but CH4 and N2O emissions do fall below 1990 levels under some scenarios. Table 5.3 shows the projections of total national emissions of the three greenhouse gases in the years 2000 and 2010, based on no change in energy use from 1995 levels.  Figure 5.4 shows projected CO2 emissions. While emissions are above 1990 levels in each case, they also fall below 1995 levels, indicating that that the 1995 level is a kind of peak in CO2 emissions and that "autonomous" fuel mix changes will help reduce emissions in the future. CH4 emissions do not change much from 1990 levels under any scenario (see Figure 5.5). When calculated on the basis of 198x-1989 rates of change in fuel shares, CH4 emissions fall from 1990 levels. However, this result must be treated with caution as discussed above. When emissions are calculated on the basis of 1990199x rates of change they rise above 1995 levels in the year 2000, but then fall below 1990 levels by the year 2010. This is encouraging, and again suggests that autonomous changes in fuel mix will (eventually) lead to a decline in CH4 emissions.  122  N2O emissions, as shown in Figure 5.6, do not fall below 1990 levels, except under the scenario involving 198x-1989 rates and these projections must be treated with caution. When calculated on the basis of 1990-199x rates of change in fuel shares, N2O emissions rise above 1995 levels in 2000, but then fall below 2000 levels in the year 2010. Unfortunately, the level in 2010 is still above the 1990 and 1995 levels. This indicates that autonomous fuel switching may eventually lead to a decline in emissions, but that this decline will not bring emissions below 1990 levels before the year 2010.  As mentioned above, projected emissions in this analysis, with energy use frozen at 1995 levels, are lower than emissions projected with energy use growing at historical rates, which is to be expected. This indicates the importance of energy conservation in reducing emissions. However, even when energy use does not change, there is still growth in some emissions and this indicates that conservation alone may not be sufficient to reduce emissions below 1990 levels, and quite likely will not be sufficient to meet Canada's commitments under the Kyoto Protocol. While "autonomous" changes in fuel shares appear to eventually bring about a decline in emissions, this takes some time and does not produce dramatic declines.  5.2 Industrial Sector 5.2.1 Energy Use Extrapolations Table 5.4 shows the results of extrapolating growth rates in industrial energy use and contributing factors. If energy use grows at the rate that prevailed during the 1984-1989 period, it will reach nearly 3300 PJ in the year 2000 and just over 4400 PJ in the year  123  2010. If industrial energy use grows at the slightly lower rate that prevailed during the 1990-1996 period, it will reach about 3150 PJ in 2000 and nearly 3800 PJ in 2010. Table 5.5 shows historical growth rates in industrial and manufacturing energy use and activity. Industrial energy use grew at nearly three percent per year on average during the 19841989 period; projected energy use is highest when this growth rate is used.  Extrapolations of energy use using historical growth rates in the factors that contribute to energy use shows the relative roles of each. If only activity changes, energy use will reach between 3050 and 3350 PJ in 2000 and between 3390 and 4700 PJ in 2010. The lower points of these ranges are close to actual 1996 energy use (2911 PJ) and correspond to activity growth rates that prevailed during the 1990-1996 period. This is encouraging, since, as can be seenfromcomparing the extrapolations presented in Table 5.4, activity has the strongest positive influence on energy use. Activity increased on average by 3.45 percent per year from 1989 to 1989, but only about 1.3 percent per year from 1990 to 1996 (see Table 5.5).  If only intensity changes, energy use in 2000 and 2010 will remain near actual 1995 levels. The same is true for the case when only structure changes. Projected energy use is highest if either of these two contributing factors grows at the levels that prevailed during the 1990-1996 period. Table 5.6 shows average annual changes in the intensities of industrial subsectors. Total industrial energy intensity fell by nearly half a percent per year from 1984 to 1989, and grew by 0.8 percent per year from 1990 to 1996. Table 5.7 shows annual average changes in industrial subsector activity shares. In general, shares  124  of the most energy-intensive subsectors fell over the 1984-1989 period but grew, or fell less sharply, over the 1990-1996 period. The fact that the highest projected energy use results when intensity and structure grow at their average 1990-1996 rates is somewhat discouraging; however, this still results in energy use changes that are fairly low.  If all three underlying factors change at historical rates, projected energy use in 2000 and 2010 is slightly higher than energy projectedfromhistorical energy use growth rates. This indicates that the combined effect of the underlying factors grows with time. The lowest projections of energy use are those that obtain when the underlying factors grow at their average rates in the 1990-1996 period, which is encouraging.  Table 5.8 shows manufacturing energy use extrapolations. The same kinds of patterns hold for changes in activity only and changes in all three underlying factors combined. As can be seen in Table 5.5, manufacturing activity grew by 3.24 percent per year from 1984 to 1989 and 1.85 percent per yearfrom1990 to 1996.  Changes in manufacturing intensity only or structure only result in the lowest projections of energy use which, in all but one case, are slightly below actual 1996 energy use. There is only slight differences in projected energy use based on the various historical growth rates in these factors, which indicates that growth rates did not change much over the 1984-1996 period. The lowest energy use projections occur when these factors grow at their average rates of the 1990-1996 period. This is confirmed by Tables 5.6 and 5.9. Manufacturing energy intensity fell by 0.18 percent per year on average from 1984 to  125  1989 and by 0.56 per yearfrom1990 to 1996, indicating an acceleration in the rate of improvement in the efficiency of manufacturing energy use. While the activity shares of some of the more energy-intensive manufacturing subsectors started to fall or continued falling over the 1990-1996 period, the shares of other energy-intensive subsectors grew over this period. Overall, however, there was a shift in manufacturing activity in the 1990-1996 period towards less energy-intensive subsectors.  5.2.2 Projected Greenhouse Gas Emissions, with Changing Energy Use Table 5.10 shows projected industrial greenhouse gas emissions in the years 2000 and 2010, based on historical rates of changes in the shares of various fuel types and historical growth rates in energy use. Under no scenario do industrial GHG emissions fall below 1990 levels, nor do they fall below 1995 levels.  Figure 5.7 shows projected industrial CO2 emissions in 2000 and 2010 as well as actual emissions in 1990 and 1995 for comparison. Under no case do CO2 emissions fall below 1990 levels, but the lowest increase in emissions occurs when they are projected using 1990-1996 rates of change in fuel types, which is encouraging and indicates that beneficial fuel switching has already been occurring, in the absence of focused GHG mitigation policy. It is possible that this improvement may be accelerated in future, perhaps leading to an ultimate reduction in emissions.  126  Figure 5.8 shows projected industrial emissions of CH4. Again, emissions of these gases are not projected to fall below 1990 levels under any of the scenarios. Further, projected emissions are lowest when based on 1984 to 1989 rates of change in fuel types, indicating that fuel switching in the post-1990 period has been contributing to increased emissions of this gas. More positively, while projected N2O emissions do not fall below 1990 levels, they are lowest when based on 1990-1996 rates of change in fuel mix (Figure 5.9).  Table 5.11 shows projected greenhouse gas emissions for manufacturing only. These projections are more encouraging. While CO2 emissions do not fall below 1990 levels under any scenario, CH4 and N2O emissions do fall below 1990 levels under most scenarios. Specifically, CH4 emissions fall below 1990 levels under projections based on rates of change in fuel shares that prevailed in the 1990-1996 period. N2O emissions in 2000 are below 1990 levels under all scenarios, and in 2010 are below 1990 levels if projected based on 1990-1996 rates of change in fuel shares.  5.2.3 Projected Greenhouse Gas Emissions, with Frozen Energy Use When industrial energy use is frozen at 1996 levels, projected GHG emissions do not change significantly under any scenario and in some cases they fall below 1990 levels (see Table 5.12).  Figure 5.10 shows projected industrial CO2 emissions in 2000 and 2010. Under no case do CO2 emissions fall below 1990 levels, but the lowest increase in emissions occurs when they are projected using 1990-1996 rates of change in fuel types. Under each  127  scenario, CO2 emissions grow to slightly above 1995 levels in the year 2000, then fall below 1995 levels, but not below 1990 levels, by the year 2010. This indicates that autonomous fuel switching in the industrial sector will eventually lead to declining CO2 emissions.  Figure 5.11 shows projected industrial emissions of CH4. Emissions of these gases are projected to fall below 1990 levels when calculated on the basis of 1984-1989 rates of change in fuel shares. Unfortunately, when calculated on the basis of the more recent (1990-1996) rates of change, emissions are projected to rise above 1990 levels and show a steady upward trend. This indicates that fuel switching in recent years has been towards more CPU-intensive fuels and that even freezing energy use at the 1996 level will not bring about a decline in emissions, if trends in fuel mix continue as they have been. The same holds true for N2O emissions (see Figure 5.12), which, while declining from 1995 levels in 2000, rebound to nearly 1995 levels by 2010.  To summarize, autonomous fuel switching in industrial energy use has in recent years been towards fuels which involve lower C 0 emissions, but higher N2O and CH4 2  emissions, which demonstrates a trade-off in fuel mix policy and the need to consider all GHGs in formulating that policy. It also indicates that even if energy use remains at the 1996 level, recent trends in fuel mix will keep emissions of all three GHGs above 1990 levels (assuming that these trends continue).  128  Table 5.13 shows emission projections for manufacturing, again with energy use frozen at the 1996 level. The results for manufacturing are very promising: under every scenario and in both 2000 and 2010, emissions of all three GHGs fall well below 1990 levels. This indicates that there has been switching towards less GHG-intensive fuels and that if these trends continue, manufacturing emissions will fall to levels that meet or exceed Canada's international commitments.  Figure 5.13 shows manufacturing C 0 emissions with frozen energy use. Emissions fall 2  below 1990 levels under each scenario, but fall the most when calculated on the basis of 1990-1996 rates of change in fuel shares. The same is true for manufacturing CH4 and N2O emissions (see Figures 5.14 and 5.15). This indicates that recent trends in manufacturing fuel use have been towards even less GHG-intensive fuels than in the 1980s, and this is a promising development.  5.3 Freight Transport 5.3.1 Energy Use Extrapolations Table 5.14 shows the results of extrapolations of freight transport energy use. If energy use continues growing at historical levels, it will be between 693 PJ and 715 PJ in the year 2000 and 809 PJ and 905 PJ in the year 2010. As can be seen in Table 5.15, freight transport energy use grew by 2.4 percent per year on average from 1984 to 1989 and by 1.75 percent on average from 1990 to 1996. This is encouraging, sincefreighttransport energy use, while not falling, seems at least to be growing at a reduced rate.  129  Table 5.14 also shows freight transport energy use extrapolations based on changes in the three factors which underlie freight energy use (i.e., activity, structure and intensity). The lowest change in energy use occurs if the structure of freight transport is the only factor to change, and if mode shares change at the annual average rates that prevailed in the 1990-1996 period. Under this scenario, freight transport energy use actually falls from 1996 levels: in 2000 it is about 20 PJ lower and in 2010 it is about 90 PJ lower. Table 5.16 shows changes in the activity shares of freight transport modes. While the shares of the two least energy-intensive modes, marine and rail, fell during the entire 1984-1996 period, trucking activity shifted from light trucks to heavy trucks. Since heavy trucks are the least energy-intensive kind of trucks, this leads to a decline in projected energy use when structure (only) grows at 1990-1996 rates.  The highest extrapolated energy use occurs when all three factors change at rates that prevailed during the 1984-1989 period. Under this scenario, energy use is almost 100 PJ higher in 2000 than it actually was in 1996; by 2010, energy use is more than 400 PJ higher.  Table 5.15 shows that freight activity grew at an annual average rate of half a percent from 1984 to 1989 and then grew at almost 5 times that ratefrom1990 to 1996. Table 5.17 shows changes in energy intensity byfreightmode. From 1984 to 1989, the energy intensities of light and medium-heavy trucks fell, but the intensities of all other modes rose. In the 1990-1996 period, the energy intensities of all modes fell, at rates ranging from about 0.8 percent for light trucks to over 3.5 per year for rail.  130  5.3.2 Projected Greenhouse Gas Emissions, with Changing Energy Use Table 5.18 shows projectedfreighttransport greenhouse gas emissions in the years 2000 and 2010, based on historical rates of changes in the shares of various fuel types. While CO2 and N2O emissions do not fall below 1990 levels, or even 1996 levels, under any scenario , CH4 emissions do fall below 1990 levels under all scenarios. 57  Figure 5.16 shows projectedfreighttransport CO2 emissions in 2000 and 2010 as well as actual emissions in 1990 and 1995 for comparison. Under no case do CO2 emissions fall below 1990 levels and the lowest increase in emissions occurs when they are projected using 1984-1989 rates of change in fuel types. This is discouraging since it indicates that fuel switching in recent years has led to increased CO2 emissions.  Figure 5.17 shows projected CFL4 emissions. Emissions of these gases are projected to fall below 1990 levels under all of the scenarios. What is interesting is that emissions risefromthe actual 1996 level in 2000 and then fall below this level in 2010. This is a result of the way in which fuel shares are calculated (see section 5.1.2 above for description). Projected fuel shares are shown in Table 5.19. The share of motor gasoline falls by about half between 2000 and 2010. Because motor gasoline has the largest CH4 and N2O emission factors by far of all the fuel types, this change in its fuel share explains the fall in projected emissions from 2000 to 2010. The share of motor gasoline also falls between 1996 and 2000, but the influence of this on emissions appears to be offset by  N 2 o emissions do fall below 1990 levels i n 2010 w h e n projected o n the basis o f 1984-89 growth rates i n fuel shares. A s discussed elsewhere, this result may be suspect. 5 7  131  increased energy use plus increases in the shares of emission-intensive fuels like diesel. The largest projected CH4 emissions occur when they are based on 1990-1996 rates of change in fuel shares. This may be due to data errors, as mentioned above, but if it is not, it indicates that changes in fuel mix have in recent years been leading to increased emissions.  Figure 5.18 shows projected N2O emissions. Emissions are projected to rise well above 1990 and 1996 levels. This appears to be due to the increases in shares of N 0-intensive 2  fuels like propane and especially diesel. The largest projected emissions occur when they are based on 1990-1996 rates of change in fuel shares, which, again, may indicate trends in recent years towards a more GHG-intensive fuel mix.  5.3.3 Projected Greenhouse Gas Emissions, with Frozen Energy Use When energy use is frozen at 1996 levels, C 0 and N 0 emissions do not fall below 1990 2  2  levels under most scenarios. More positively, CH4 emissions fall considerably below 1990 levels under all scenarios (see Table 5.20).  Figure 5.19 shows projected C 0 emissions in 2000 and 2010. Under no case do C 0 2  2  emissions fall below 1990 levels and it appears that they are highest when calculated on the basis of the more recent 1990-1996 rates of change in fuel shares. However, as discussed above, this may the result of an error in fuel use data. If this is not an error, this indicates that freight transport fuel switching in recent years has been towards more C0 -intensive fuels. 2  132  Figure 5.20 shows projected CFLj emissions. Emissions of these gases are projected to fall below 1990 levels under all scenarios, but appear to fall the least when calculated on the basis of recent (1990-1996) rates of change in fuel shares. If this is not the result of a data error, then fuel switching in recent years has also been towards more CFLt-intensive fuels than in the 1980's. Fortunately, there has still been sufficient switching to fuels with lower CH4 emissions factors that emissions nevertheless fall below 1990's levels.  The same pattern holds true for freight transport N2O emissions, but in this case emissions in 2000 and 2010 are above 1990 levels when calculated on the basis of 19901996 rates of change in fuel shares (see Figure 5.21). In fact, as Figure 5.21 shows, there is a reversal of the 1990-1995 downward trend in emissions, so that emissions in 2000 and 2010 are well above both 1990 and 1995 levels.  To summarize, fuel switching in freight transport energy use in recent years appears to have been towards more GHG-intensive fuels and, even without any change in energy use from 1996 levels, results in increased CO2 and N2O emissions. However, even with the trend towards more GHG-intensive fuels,freighttransport CH4 emissions fall below 1990 levels.  5.4 Passenger Transport 5.4.1 Energy Use Extrapolations  133  Table 5.21 shows extrapolations of passenger transport energy use. If energy use continues growing at historical levels, it will be between 1395 PJ and 1455 PJ in the year 2000 and 1630 PJ and 1890 PJ in the year 2010. As can be seen in Table 5.22, passenger transport energy use grew by 2.6 percent per year on average from 1984 to 1989 and by 1.6 percent on average from 1990 to 1996. This is encouraging, since passenger transport energy use, while not falling, seems at least to be growing at a reduced rate.  Table 5.21 also shows passenger transport energy use extrapolations based on changes in the three factors which underlie energy use. The lowest change in energy use occurs if the structure of passenger transport is the only factor to change, and if mode shares change at the annual average rates that prevailed in the 1990-1996 period. Under this scenario, energy use falls from 1996 actual levels: in 2000 it is about 40 PJ lower and in 2010 it is about 200 PJ lower. Table 5.23 shows changes in the activity shares of passenger transport modes. Falling energy use under the structure-only scenario appears to comefromthe decline in the rate of increase in the share of air travel, which is the most energy-intensive passenger transport mode. In the 1984-1989 period, the share of air travel in total passenger travel activity was growing at about 3.6 percent per year; in the 1990-1996 period, air travel's share was growing at only about 0.3 percent per year. The share of rail, also a relatively energy-intensive mode, has been falling, at an increasing rate. At the same time, the share of the least energy-intensive mode, buses, has also been falling, at an accelerating rate. Overall, however, structural changes have acted to reduce passenger transport energy use.  134  The highest extrapolated energy use occurs when activity changes at rates that prevailed during the 1984-1989 period. Under this scenario, energy use is about 250 PJ higher in 2000 than it actually was in 1996, by 2010, energy use is nearly 1000 PJ higher. Table 5.22 shows that passenger transport activity (passenger-kilometres) increased by about four percent per year from 1984 to 1989. Over the 1990-1996 period, passengerkilometres were still increasing, but at a slightly reduced rate of 2.8 percent per year.  Changes in intensity only and all three factors combined produce projections that lie somewhere between these two extremes, with the intensity-only energy projections lower than projections with all three factors changing. Table 5.245 shows changes in energy intensity by mode. The combined influence of these changes leads to slightly higher projected energy use when intensity (only) grows at 1990-1996 rates than when it grows at 1984-1989 rates.  5.4.2 Projected Greenhouse Gas Emissions, with Changing Energy Use Table 5.25 shows projected passenger transport greenhouse gas emissions in the years 2000 and 2010, based on historical rates of changes in the shares of various fuel types and in energy use. Under no scenario do GHG emissions fall below 1990 levels, nor do they fall below 1996 levels.  Passenger transport CH4 and N 0 emission projections in 2010 based on 1984-89 trends 2  are very small. This is because of what appears to be an error in the data. Natural gas energy use is between one and three petajoules in the years before and after 1989, but is  135  reported as 10.01 petajoules in 1989. This is shown in Figure 5.22 as the spike in natural gas use and is assumed to be an error. Because the 1989 value for natural gas energy use is an order of magnitude higher than other years, the average annual growth rate of natural gas use in the period 1984 to 1989 is calculated to be almost 100 percent (see Table 5.26). As a consequence of this very high annual growth rate, natural gas is calculated to account for almost 98 percent of passenger transport energy use in the year 2010. This does not seem realistic and therefore the estimates for passenger transport GHG emissions in 2010 based on the 1984 to 1989 trends should be treated with caution. Further, this apparent error in the emission estimates for passenger transport is also contained in the emission estimates for Canada as a whole and these, therefore, should also be treated with caution.  Figure 5.23 shows projected passenger transport CO2 emissions in 2000 and 2010 as well as actual emissions in 1990 and 1995 for comparison. Under no case do CO2 emissions fall below 1990 levels, but it appears that the lowest increase in emissions occurs when they are projected using 1990-1996 rates of change in fuel types, which is encouraging and indicates that beneficial fuel switching has already been occurring.  Figure 5.24 shows projected CH4 emissions. Again, emissions of these gases are not projected to fall below 1990 levels under any of the scenarios. Further, projected emissions appear to be lowest when based on 1984 to 1989 rates of change in fuel types, indicating that fuel switching in the post-1990 period has been contributing to increased emissions of this gas. The same may be true for N2O emissions (see Figure 5.25), but  136  with the suspected error in the fuel use data, it is impossible to be sure, for any of the three GHGs  5.4.3 Projected Greenhouse Gas Emissions, with Frozen Energy Use When passenger transport energy use is kept frozen at 1996 levels, projected emissions of the GHGs do not change significantly from 1995 levels, but they are above 1990 levels. Table 5.27 shows projected passenger transport greenhouse gas emissions in the years 2000 and 2010, based on historical rates of changes in the shares of various fuel types and energy use frozen at 1996 levels. Because of the apparent error in fuel use data (see above), only the emission projections calculated on the basis of 1990-1996 rates of change in fuel shares will be considered.  Figure 5.26 shows projected CO2 emissions. There is a slight downward trend, with emissions in 2000 and 2010 below the 1995 level. Unfortunately, this trend is not sufficient to bring emissions below 1990 levels by 2010. The same is true for CH4 emissions (see Figure 5.27). N2O emissions show a slightly different pattern: rising from 1995 to 2000 and then falling again (see Figure 5.28). Emissions are above 1990 levels in 1995, 2000 and 2010.  In summary, it appears that there is a slight downward trend in passenger transport GHG emissions, indicating a minor switch to less GHG-intensive fuels in the 1990-1996 period. However, even in the absence of an increase in energy use, this fuel switching is not sufficient to reduce emissions below 1990 levels.  137  5.5 Service Sector 5.5.1 Energy Use Extrapolations Table 5.28 shows the results of extrapolations of service sector energy use. If energy use continues growing at historical levels, it will be between 990 PJ and 1020 PJ in the year 2000 and 1115 PJ and 1215 PJ in the year 2010. As can be seen in Table 5.29, service sector energy use grew by 1.3 percent per year on average from 1981 to 1989 and by 1.8 percent on average from 1990 to 1995. This is discouraging, since the rate of increase in service sector energy use has been increasing over time.  Table 5.28 also shows service sector energy use extrapolations based on changes in the three factors which underlie energy use. The lowest change in energy use occurs if structure is the only factor to change, and if building shares change at the annual average rates that prevailed in the 1981-1989 period. Under this scenario, service sector energy use actually falls from 1995 levels: in 2000 it is about 110 PJ lower and in 2010 it is nearly 300 PJ lower. Table 5.30 shows changes in the floor area shares of the various kinds of service sector buildings. The shares of the two most energy-intensive building types, health and hotels and restaurants, fell during the 1981-1989 period but began to increase after 1990 (but at low annual rates). The increase in the shares of these building types in the 1990-1995 period is largely responsible for the higher energy use projections based on 1990-1995 rates of change in structure compared to 1981-1989 rates . The 58  shares of the two least energy-intensive building types, religious and warehouses, also  The other b u i l d i n g types have fairly equal energy intensities, so changes i n the shares o f these kinds o f buildings w o u l d not have as large an impact o n energy use.  138  fell during the entire 1981-1995 period; however, the rate of decline in the shares of these two building types decreased over time, so that in the 1990-1995 period the shares of each were declining at a rate of less than half a percent per year.  The highest extrapolated energy use occurs when activity changes at the annual average rate that prevailed during the 1981-1989 period. Under this scenario, energy use is almost 200 PJ higher in 2000 than it actually was in 1995; by 2010, energy use is more than 700 PJ higher. Table 5.29 shows annual average rates of change in service sector activity (m of floor area). From 1981 to 1995, floor area grew by almost four percent 2  per year;from1990 to 1995, it grew approximately half as fast, at about two percent per year.  Energy use projections based on changes in intensity only and all three factors combined lie between these two extremes, with intensity-only-based projections lower than projections based on all three factors. Table 5.31 shows changes in service sector energy intensities by building type. Offices, retail, schools and warehouses had the largest shares of service sector floor area. The energy intensities of these four building types show a downward trend over the 1981-1995 period; however, only schools and warehouses show negative average annual rates of change in energy intensity over the period and the energy intensity of offices, which account for the most service sector floor space, grew by 2.24 percent per year over the period.  5.5.2 Projected Greenhouse Gas Emissions, with Changing Energy Use  139  Table 5.32 shows projected service sector greenhouse gas emissions in the years 2000 and 2010, based on historical rates of changes in the shares of various fuel types and in energy use. Under no scenario do CO2 emissions fall below 1990 levels, or 1995 levels, but CH4 and N2O emissions do fall below these levels under all but one scenario.  Figure 5.29 shows projected service sector CO2 emissions in 2000 and 2010 as well as actual emissions in 1990 and 1995 for comparison. Under no case do CO2 emissions fall below 1990 levels and the lowest increase in emissions occurs when they are projected using 1981-1989 rates of change in fuel types, which indicates that there has been a shift in recent years to more COi-intensive fuel types. This is discouraging and indicates that, in the absence of policy which encourages shifts in fuel mix towards less GHG-intensive fuel types, there is reason to expect that emissions from this sector will continue growing.  Figure 5.30 shows projected CH4 emissions. If fuel shares changed at rates that prevailed during the 1981-89 period, CH4 emissions in 2000 and 2010 would be below 1990 levels. Unfortunately, if fuel shares change at rates that prevailed during the more recent 19901995 period, CH4 emissions in 2000 and 2010 will be considerably above 1990 levels. The same is true for N2O emissions (see Figure 5.31), except that emissions in 2000 and 2010 projected on the basis of 1990-1995 rates are not much above 1990 levels. Again, projected emissions of these gases indicate that in the absence of policy measures, there is reason to expect increased emissionsfromthis sector.  5.5.3 Projected Greenhouse Gas Emissions, with Frozen Energy Use  140  When energy use is keptfrozenat 1995 levels, projected CO2 emissions remain close to 1995 levels (but above 1990 levels) under every scenario, while CH4 emissions fall below 1990 levels under some scenarios and N2O emissions fall below 1990 levels under all scenarios (see Table 5.33).  Figure 5.32 shows projected CO2 emissions in 2000 and 2010. Under every scenario C 0  2  emissions remain very close to 1995 levels, about 5 million tonnes above 1990 levels. Unfortunately, emissions increase from 1995 levels when calculated on the basis of 19901995 rates of change in fuel shares, but fall when calculated on the basis of 1981-1989 rates of change in fuel shares. The change is small in any case, but this again indicates that there has been fuel switching in recent years towards more CCh-intensive fuels.  Figure 5.336 shows projected emissions of CH4. Emissions of these gases are projected to fall below 1990 levels when calculated on the basis of 1981-1989 rates of change in fuel shares. Unfortunately, when calculated on the basis of the more recent (1990-1995) rates of change, emissions are projected to rise above 1990 levels. This indicates that fuel switching in recent years has been towards more CFLj-intensive fuels. The same holds true for N 0 emissions (see Figure 5.34), except that emissions fall considerably below 2  1990 levels and show downward trends under every scenario. The fact that the decline is less pronounced when the emission projections are based on 1990-1995 rates of change in fuel shares means that there has been fuel switching in recent years towards more N2Ointensive fuels.  141  5.6 Residential Sector 5.6.1 Energy Use Extrapolations Table 5.34 shows the results of extrapolations of residential sector energy use. If energy use continues growing at historical levels, it will be between 1430 PJ and 1475 PJ in the year 2000 and 1540 PJ and 1700 PJ in the year 2010. As can be seen in Table 5.36, residential sector energy use grew by 1.4 percent per year on average from 1981 to 1989 and by 0.76 percent on average from 1990 to 1995. This is encouraging, since, while residential energy use is not falling, it has at least been growing at a reduced rate over time.  Table 5.34 also shows residential sector energy use extrapolations based on changes in the two factors which underlie residential energy use (i.e., activity and intensity; it was not possible to calculate shares by end-use - see section 3.6 for an explanation). The lowest change in energy use occurs if intensity is the only factor to change, and if end use intensities change at the annual average rates that prevailed in the 1990-1995 period. Under this scenario, residential sector energy use actually falls from 1995 levels: in 2000 it is almost 80 PJ lower and in 2010 it is more than 200 PJ lower. Table 5.35 shows changes in the energy intensities of the various residential end-uses. The energy intensities of all end-uses except for space heating increased over the 1981-1995 period. Because space heating accounts for a large share of residential energy use (see Section 3.6 of Chapter 3), the decline in space heating energy intensity more than offset the increases in the intensities of the other end-uses.  142  The highest extrapolated energy use occurs when activity changes at the rate that prevailed during the 1981-1989 period. Under this scenario, energy use is about 150 PJ higher in 2000 than it actually was in 1995; by 2010, energy use is nearly 500 PJ higher. Table 5.36 shows the annual average rates of change in residential sector activity (number of households). From 1981 to 1989, the number of households increased by just over two percent per year; from 1990 to 1995, the number of households increased by just under two percent per year.  Table 5.34 also shows projected energy use with adjustments for weather. Adjusting for weather leads to significantly higher energy use projections, and, in fact, the highest projections of any of the scenarios. As shown in Table 5.36, both heating and cooling degree-days grew over the 1981-1995 period. In other words, winters tended to get colder and summers warmer over time; this can create increased demand for space heating and cooling.  5.6.2 Projected Greenhouse Gas Emissions, with Changing Energy Use Table 5.37 shows projected residential greenhouse gas emissions in the years 2000 and 2010, based on historical rates of changes in the shares of various fuel types and in energy use. Under no scenario do CO2 emissions fall below 1990 levels, or even 1995 levels, but CFLj and N2O emissions are near or below 1990 levels under all scenarios.  Figure 5.35 shows projected residential CO2 emissions in 2000 and 2010 as well as actual emissions in 1990 and 1995 for comparison. Under no case do C 0 emissions fall below 2  143  1990 levels, but the lowest increase in emissions occurs when they are projected using 1990-1995 rates of change in fuel types, which is encouraging and indicates that beneficial fuel switching has already been occurring, in the absence of focused GHG mitigation policy.  Figure 5.36 shows projected CH» emissions. When based on 1981-89 rates of change in fuel shares, emissions of these gases are projected to fall below 1990 levels. However, when based on more recent (1990-1995) rates of change in fuel shares, projected emissions are slightly above 1990 levels and somewhat more above 1995 levels. Although the projected increase in emissions is small, it is still worrisome because it indicates that there have been shifts in recent years towards more CFLt-intensive fuel types.  More positively, N 0 emissions are projected to fall below 1990 levels under all 2  scenarios, and the largest decline occurs when 1990-1995 rates of change in fuel shares are used (see Figure 5.37). Because recent trends in fuel mix have tended to reduce C 0  2  and N 0 emissions, but increase C H emissions, it is difficult to determine whether these 2  4  shifts are (net) beneficial or not, without converting CH4 and N 0 to their C 0 2  2  equivalencies (which is beyond the scope of this work). It does indicate a potential tradeoff in policy which focuses on fuel mix and underlines the need to consider all GHGs when formulating policy.  5.6.3 Projected Greenhouse Gas Emissions, with Frozen Energy Use  144  When residential sector energy use is kept frozen at 1995 levels, CO2 emissions remain close to 1990 levels under most scenarios, dipping below 1990 levels in some cases. CH4 and N2O emissions both are below 1990 levels under every scenario and for N2O the decline is significant. Table 5.38 shows projected residential greenhouse gas emissions in the years 2000 and 2010, based on historical rates of changes in the shares of various fuel types and energy use frozen at 1995 levels.  Figure 5.38 shows projected CO2 emissions in 2000 and 2010. When emissions are projected using 1990-1995 rates of change in fuel types, they fall below 1990 levels. This is encouraging and indicates that beneficial fuel switching has already been occurring, in the absence of focused GHG mitigation policy.  Figure 5.39 shows projected residential emissions of CH4, which fall below 1990 levels under every scenario. The decline in emissions is larger when they are calculated on the basis of 1981-1989 rates of change in fuel shares than when calculated on the basis of more recent (1990-1995) rates. This indicates that fuel switching in recent years has been towards more CFLt-intensive fuels. Fortunately, emissions still decline below 1990 levels.  N2O emissions fall well below 1990 levels and the decline is virtually identical under all scenarios (see Figure 5.40). Emissions in the year 2000 are very slightly higher when calculated on the basis of 1990-1995 rates, but the reverse is true in the year 2010. Therefore, it is not possible to draw a conclusion about fuel switching in recent years compared to the 1980's.  145  In summary, under every scenario, residential CO2 emissions remain close to 1990 levels, while CH4 and N2O emissions fall below 1990 levels. CO2 emissions fall below 1990 levels when calculated on the basis of recent (1990-1995) rates of change in fuel shares. It appears the residential sector fuel switching in recent years has been towards fuels that are less CCvintensive but more CHt-intensive. This indicates that there is a potential trade-off in policy which focuses on fuel switching in this sector and demonstrates the importance of considering all GHGs.  5.7 Summary and Conclusions This chapter discussed the results of three extrapolations of historical trends. The first extrapolation, of energy use, was based on historical rates of change in overall energy use as well as in the three factors (activity, intensity and structure) that underlie energy use. The second extrapolation involved extending historical trends in energy use and fuel shares to derive projections of emissions of three greenhouse gases: C 0 , CH4 and N2O. 2  The last extrapolation used extended historical trends in fuel shares and energy use frozen at 1995 or 1996 levels to derive emission projections.  The results of each extrapolation are summarized below and are followed by conclusions that may be drawn from these results.  5.7.1 Energy Use Projections — Summary  146  When energy use is projected on the basis of historical growth rates in energy use, it does not fall below the 1995 level. The exception is when structure is the only factor to change, except in the residential sector where energy use falls if intensity is the only factor to change and in the industrial sector where energy use does not fall in any case. At the aggregate level and in all sectors, projected energy use is lowest when it is based on energy use growth rates that prevailed in the 1990-199x period. The exception is the service sector, when energy projections are lowest when based on 1981-1989 energy use growth rates.  Of the factors that determine energy use, changes in structure only generally produce the lowest energy projections in all sectors but the industrial and residential sectors, where changes in intensity only produce the lowest energy projections. In all sector but the service sector, energy projections are lowest when based on 1990-199x rates of change in structure or intensity. In all sectors but freight transport, changes in activity only produce the highest projected energy use, and the highest projections occur in all sectors when this factor grows at rates that prevailed during the 1980's. In freight transport, projected energy use is highest when projected on the basis of all three factors changing, and the very highest projection occurs when all three factors change at rates that prevailed during the 1984-1989 period. In the residential sector, projections based on all three factors changing, with adjustments for weather, produce the highest projected energy use.  In all sectors, energy use projected on the basis of historical growth rates is lower than when projected on the basis of all three contributing factors (structure, activity and  147  intensity) changing at historical rates. This indicates that the combined positive influence of the underlying factors on energy use grows with time.  5.7.2 Greenhouse Gas Emissions Projected with Changing Energy Use — Summary In general, greenhouse gas emissions rise above 1990 levels when they are projected on the basis of historical rates of change in fuel shares and energy use. In particular, this is true at the aggregate national level, indicating that the continuation of past trends in fuel switching and energy use will result in increased GHG emissions over time.  There are some exceptions. In manufacturing and the service and residential sectors, CHU and N2O emissions drop below 1990 levels in some cases; freight transport CH4 emissions drop below 1990 levels in all cases.  At the aggregate national level, CO2 emissions rise by less when projected on the basis of more recent (1990-199x) rates of change in fuel shares, but the reverse is true for CH4 and N 0 emissions. In the industrial and residential sectors, CO2 and N2O emissions rise 2  by less, and CH4 emissions rise by more, when projected on the basis of more recent rates of change in fuel shares. In freight transport and the service sector, emissions of all three GHGs rise by more when projected on the basis of more recent rates of change in fuel shares. It is not possible to make this comparison for passenger transport emissions, due to an apparent error in the data.  148  5.7.3 Greenhouse Gas Emissions Projected with Frozen Energy Use — Summary Emissions of the three GHGs are lower when projected on the basis of changing fuel shares but frozen energy use, showing the importance of energy conservation in emission reduction. At the aggregate national level, CO2 and N2O emissions still rise above 1990 levels, but CH4 emissions fall below 1990 levels by 2010 (and fall below 1990 levels by the year 2000 if based on rates of change in fuel shares that prevailed in the 1980's).  Emissions of some or all of the GHGs fall below 1990 levels in almost all sectors under at least one scenario (i.e., historical rates of change in fuel shares in the 1980's or 1990's). Either CH4 or N2O or both falls below 1990 levels under one or more scenarios in most sectors and residential CO2 emissions fall when they are projected on the basis of 19901995 rates of change in fuel shares. The exception is passenger transport, where emissions of all gases remain above 1990 levels. However, due to an error in the fuel use data, emission projections based on 1984-1989 rates in this sector were not considered and so the analysis was restricted to projections based on 1990-1996 rates.  The manufacturing subsector has the most promising results of all the sectors, emissions of all three GHGs fall below 1990 levels under all scenarios. This indicates that manufacturing is well on its way to meeting GHG reduction commitments. The residential sector also has promising results: emissions of all three GHGs fall below 1990 levels when based on recent trends, and N 0 emissions fall considerably. Results from 2  the other sectors are less positive as CO2 emissions increase under all scenarios and CH4  149  and N2O emissions are higher when based on more recent trends, although freight CH4 and service sector N2O are still below 1990 levels. 5.7.4 Conclusions Extending historical trends into the future gives a rough but perhaps still enlightening glimpse of what Business-As-Usual holds in store for Canada. With some exceptions, Business-As-Usual will involve increased energy use and increased emissions of three important greenhouse gases (CO2, CH4 and N 0). There have been some positive trends 2  in the 1990's, including structural changes towards less energy-intensive activities, declines in the energy intensity of most activities and shifts to less GHG-intensive fuels in some sectors (but, unfortunately, not in all). With the exception of the manufacturing subsector, continuing with Business-As-Usual will mean growth in energy use or emissions or emission reductions that are insufficient to meet Canada's international commitments.  150  Chapter 6 Comparison of Canada to Other Industrialized Countries  In this chapter Canada is compared to other OECD (Organisation for Economic Cooperation and Development) countries in order to determine if climate, geography and industrial structure are sufficient reasons why per capita energy use in Canada exceeds that of the other countries considered. The methods used for the comparisons are discussed in detail in Section 2.5. Essentially, they involve calculating what energy use in Canada in a particular year would have been if Canada had had the same climate, geography or industrial structure as either the United States or one of four groupings of OECD countries.  The four groupings of OECD countries are • OECD-12 which consists of Australia, Denmark, Finland, France, Germany, Holland, Italy, Japan, Norway, Sweden, the United Kingdom and the United States; • EU-9 which consists of Denmark, Finland, France, Germany, Holland, Italy, Norway, Sweden and the United Kingdom; •  Scand-4 which consists of Denmark, Finland, Norway and Sweden; and  • PacRim-3 which consists of Australia, Japan and the United States.  OECD-12 is the largest grouping and is intended to represent all industrialized countries, EU-9 is intended to represent Europe, Scand-4 is intended to represent Scandinavia and 151  PacRim-3 is intended to represent the non-European industrialized countries, all of which are in the Pacific Rim.  Comparisons are made only to other industrialized countries because lifestyles and economic activity in these countries are roughly similar to Canada's. Further, it is the industrialized nations that have thus far committed to reducing their greenhouse gas (GHG) emissions and therefore face the same challenges as Canada.  These comparisons are made because the claim is made in Canada's official climate change policy documents that Canadians are high energy users because climate, geography and industrial structure require us to be so (Canada, 1995; Environment Canada, 1997). The implication is that Canadians will continue to be relatively high 59  energy users and GHG emitters because of these non-energy factors; that the non-energy factors reduce the scope for energy conservation and GHG emission reduction.  Unfortunately, this argument is spurious. To see this, one may divide per capita energy use in 1990 (the base year) into three components:  1. The portion that is equal to per capita energy use in the country or country aggregation to which Canada is being compared (EU-9, for instance)  Population growth i s also mentioned as a factor. I n its official submission under the B e r l i n Mandate regarding criteria for differentiated responsibilities (i.e., factors unique to each country w h i c h should be considered w h e n establishing its G H G emission reduction commitment), Canada stated that four factors are relevant to Canada: population growth, the emissions intensity o f its exports, the contribution o f fossil fuel to its exports a n d the already relatively large role o f renewables i n its energy supply ( A G B M , 1997). These factors are analysed i n Canada's submission. 5 y  152  2. The amount of per capita energy use that can be accounted for by differences in climate, geography and industrial structure 3. The amount of per capita energy use that cannot be accounted for by differences in climate, geography and industrial structure  As an example, assume that per capita energy use in Canada in 1990 was 100 PJ and in EU-9 it was 70 PJ. Correcting Canadian per capita energy use for differences in climate, geography and industrial structure reduces per capita energy use to 80 PJ. Therefore, the 100 PJ can be divided into the three components:  1. 70 PJ which is equal to EU-9 per capita energy use 2. 20 PJ which was accounted for by differences in climate, geography and industrial structure (i.e., the difference between 100 PJ and 80 PJ) 3. 10 PJ which was not accounted for by differences in climate, geography and industrial structure (i.e., the difference between 80 PJ and 70 PJ)  Component #1 should have at least the same energy conservation potential as it does in the EU-9 countries. This is because this component represents the amount of per capita energy use that Canada would have had but for differences in climate, geography and industrial structure plus the waste component #3. Therefore, this amount of energy use should have the same efficiency potential in Canada as it would in the EU-9 countries. Under the Kyoto Protocol to the UN Framework Convention on Climate Change, the EU9 countries committed to reduce their GHG emissions by 8 percent of 1990 levels by the  153  2008-2012 commitment period.  Assuming that this will necessitate an 8 percent  reduction in per capita energy use, then component #1 has an energy conservation potential of 8 percent.  The GHG emission reduction commitment associated with component #2 is ambiguous, since this is the amount that is "due to" Canada's unique climate, geography and industrial structure and is the basis, therefore, of the argument that Canada "must" have a higher rate of per capita energy use than other countries. However, and contrary to this argument, the fact that Canada has a cold climate, huge land area and resource-based industrial structure means that there may be more opportunities to reduce energy use and GHG emissions in Canada than in countries in which these factors are not as prevalent. For instance, buildings in Canada may be better insulated; transportation systems can be redesigned to favour public/low-intensity transportation forms; and secondary and tertiary manufacturing, using the most advanced and energy- and materials-efficient technology, can be promoted. Because Canada is startingfroma position of being a relatively high energy user, it may well be that there are more inexpensive options for energy conservation in Canada than there are in countries which have already implemented the low-cost options (i.e., which are already comparatively low energy users).  Therefore, the energy conservation potential associated with component #2 should be above zero. As discussed in Chapter 1, the potential for CO2 emission reductions at zero  The exception is Norway, which committed to increase its emissions by 1 percentfrom1990 levels.  154  or negative-cost in Canada was estimated to be 20 percent, largelyfromenergy efficiency improvements (COGGER, 1993).  Component #3 represents "wasted" energy because it is not accounted for after adjusting for the non-energy factors. Its conservation potential is therefore 100 percent.  Therefore, even if the non-energy factors completely account for differences between Canada and other industrialized countries (i.e., if component #3 is zero), they do not provide justification for the higher rate of per capita energy use in Canada. The only situation in which they could justify a higher level of energy use is if they reduced per capita energy use in Canada below levels prevailing in other industrialized countries. In this case, Canada would be a lower per capita energy user (and GHG emission producer) but for the non-energy factors.  This chapter shows comparisons of per capita energy use in Canada versus other countries, after correcting the former for the non-energy factors. This will show:  1. the magnitude of each component 1-3 above 2. that even if the argument that Canada must be a high energy user because of nonenergy factors was logically sound, the numbers do not support it (i.e., even after correcting for non-energy factors, Canada is still a higher per capita energy user than most other industrialized countries)  155  Comparison of GHG emissions was not conducted because energy use is the root of most GHG emissions and therefore comparison of energy use implies proportional comparisons of emissions. In other words, if it is found that climate, geography and industrial structure account for the difference in energy use, then, to the extent that emissions resultfromenergy use, they also account for differences in per capita GHG emissions.  Comparisons were conducted for three years: 1984, the earliest year for which data for all countries were available; 1990, the year against which emission reductions are to be measured; and 1994, the latest year for which data are available for all countries. Comparing 1984 to 1994 gives an indication of how the strength of the explanatory factors (climate, geography and industrial structure) have changed over time. Emission reduction commitments under the Kyoto Protocol are expressed as a percentage reduction from 1990 levels. Therefore, comparisons for the year 1990 investigate the validity of Canada's existing emission reduction commitment (i.e., should it have been higher or lower).  Each of the comparisons (climate, geography and industrial structure) are presented first and are followed by presentation and discussion of aggregated results.  6.1 Climate Climate is assumed to affect energy use through its impact on space heating and cooling. Average climate in a particular year can be expressed through Heating Degree-Days  156  (HDD) and Cooling Degree-Days (CDD). A HDD is one day in which the mean temperature is below a certain base (usually 18° Celsius) by one degree; for example, if in one year there were three days in which mean temperature was 15°C, there would be nine (3x3) HDDs in that year. Similarly, one CDD is one day in which the mean temperature is above certain base (usually 18°C or 21°C) by one degree; in a year in which there were three days with mean temperatures of 21°C would have nine CDDs.  Ideally, the climate adjustment would have included both HDDs and CDDs. Unfortunately, CDD data were not available for the other OECD countries and so only the heating adjustment was made. The adjustment was made by multiplying Canada's "heating intensity" in a particular year by HDDs in that year in the United States or one of the four country aggregations. Canada's heating intensity is calculated as space heating energy use in a particular year divided by (Canadian) HDDs in that year. The result is space heating energy use in Canada if Canada had had the same climate (as expressed by HDDs) as the country or aggregation of interest. Section 2.5 of Chapter 2 provides a more detailed discussion of this calculation.  The results of this adjustment are provided in Table 6.1 and Figure 6.1. Adjusting for weather does reduce residential and service sector energy use below actual levels in each of the three years considered. This means that if Canada had had the same climate as the various country aggregations or as the United States, space heating energy use (and therefore total residential and service sector energy use) would have been lower. The reduction in energy use is highest when average HDDs of the Pacific Rim countries 157  (Australia, Japan and US) are used and lowest when average HDDs of the Scandinavian countries (Denmark, Finland, Norway and Sweden) are used.  Table 6.2 show the adjusted residential and service sector energy use divided by Canada's population as well as average per capita residential and service sector energy use in the United States and the four country aggregations. Adjusting Canada's residential and service sector space heating energy use for climate differences brings per capita energy use in Canada to below actual per capita energy use in the United States and the Pacific Rim countries. This means that climate differences more than explain the differences between per capita residential and service sector energy use in Canada and these countries. However, adjusting for climate differences does not bring Canadian per capita residential and service sector energy use below actual levels in Europe, the Scandinavian countries and the 12 OECD countries overall. This means that climate does not "explain" all of the differences in per capita residential and service sector energy use between Canada and these countries. 6.2 Geography Geography is assumed to affect energy use through its impact on passenger and freight transport demand. Passenger-kilometres per capita and tonne-kilometres per dollar of GDP were used to adjust Canada's passenger and freight transport energy use, respectively. This assumes that differences among countries in passenger-kilometres per person and freight tonne-kilometres per dollar of GDP are strictly due to differences in average distances which people and goods must travel. Clearly, this is a simplification. Passenger-kilometres reflect distances people must travel (i.e., the distancefromone  158  population centre to another) but also reflect the frequency of travel, the choice of motorized transport over walking or bicycling, the number of passengers per vehicle and the choice of modes. Similarly, freight tonne-kilometres reflect the distance which goods must be transported but also reflect mode choice, siting of industrial activity and the composition of industrial activity. A more accurate comparison would be one in which these factors have been removed, but this is beyond the scope of the current analysis. Nonetheless, the current comparison gives a rough indication of the degree to which geography "explains" differences in transport energy use between Canada and other industrialized countries.  The passenger transport adjustment was made by multiplying Canada's passenger "transport intensity" (terajoules of energy used in passenger transport divided by passenger-kilometres per capita) in a particular year by passenger-kilometres per person in that year in the United States or one of the four country aggregations. The freight transport adjustment was made by multiplying Canada'sfreight"transport intensity" (petajoules of energy used infreighttransport divided by tonne-kilometres per dollar of GDP) in a particular year by tonne-kilometres per dollar of GDP in that year in the United States or one of the four country aggregations. The adjusted passenger and freight energy use were added together to derive total adjusted transport energy use. The result is transport energy use in Canada if Canada had had the same level of transport activity as the country or aggregation of interest. Section 2.5.2 of Chapter 2 provides a more detailed discussion of this calculation.  159  The results of this adjustment are provided in Table 6.1 and Figure 6.2. Adjusting for transport activity reduces transportation energy use below actual levels in each of the three years considered. The exception is when activity levelsfromthe United States are used; in this case, adjusted energy is slightly higher than actual energy use. The reduction in energy use is highest when average activity levels in Europe is used and lowest when average activity levels in the Pacific Rim countries is used.  Table 6.2 show adjusted transport energy use divided by Canada's population as well as average per capita transport energy use in the United States and the four country aggregations. Adjusting Canada's transport energy use for activity differences brings per capita energy use in Canada to below actual per capita energy use in the United States and the OECD-12 and PacRim-3 aggregations. However, adjusted Canadian per capita transport energy use is not lower than actual per capita energy use in the EU-9 and Scandinavia-4 aggregations. This means that even if Canada had the same transport activity levels as the European and Scandinavian countries, per capita transport energy use would still be higher than in these countries. On the other hand, if Canada had the same transport activity levels as the United States, Australia and Japan, per capita transport energy use would be lower than in these countries.  These results suggest that geography may be a valid factor for explaining differences in (per capita) transport energy use between Canada and the United States, Australia and Japan but not for differences between Canada and Europe. There are likely a number of explanations for this, including increased density of urban centres in some European  160  countries, closer siting of industry to population centres, industrial structures in which secondary and tertiary manufacturing figure more prominently and the North American and Australian "love affair" with the automobile. Unfortunately, investigation of these potential explanations is beyond the scope of this thesis.  6.3 Industrial Structure It is often claimed part of the reason why energy use is higher in Canada than in other industrialized countries is that Canadian industrial production involves relatively more primary resource production than in the other countries. Primary resource production generally is more energy intensive (when measured as energy use per dollar of output) than secondary or tertiary production.  Canadian industrial energy use was therefore adjusted to determine what energy use would have been if Canada had the same industrial structure as the United States or one of the four country aggregations. Industrial structure is expressed as the share of each subsector in total industrial output (value-added in constant 1990 US dollars). Section 2.5 of Chapter 2 provides a more detailed discussion of this calculation.  The results of this adjustment are provided in Table 6.1 and Figure 6.3. Adjusting for industrial structure in the United States and three of four aggregations does reduce industrial energy use below actual levels in each of the three years considered. However, it does not reduce industrial energy use below European levels, indicating that even on an  161  absolute level, industrial structure does not explain differences in energy use between Canada and Europe. Where there is a reduction in energy use, it is highest when average industrial structure of the Pacific Rim countries (Australia, Japan and US) is used and lowest when average industrial structure of all of the 12 OECD countries is used.  Table 6.2 show adjusted industrial energy use divided by Canada's population as well average per capita industrial energy use in the United States and the four country aggregations. Adjusting Canada's industrial energy use for structural differences does not reduce per capita energy use in Canada below actual per capita energy use in the United States or any of the country aggregations. This means that industrial structure does not "explain" all of the difference in per capita industrial energy use between Canada and other industrialized countries.  6.4 Total Energy Use After adjusting residential, service, transport and industrial energy use for the climate, geographical and industrial structure differences (as described above) they were summed together to derive total adjusted energy use. This is shown at the bottom of Table 6.1 and in Figure 6.4. Overall, the adjustments "explain" differences in total energy use between Canada and the other countries. In fact, the adjustments reduce energy usefromactual levels by anywherefromabout 11 percent to over 26 percent.  On a per capita basis, the adjustments reduce Canadian energy use below actual levels in  162  the United States in all three years considered and below the Pacific Rim aggregation in 1994 (see Table 6.2). This means that the three adjustments considered (climate, geography and industrial structure) do "explain" differences in per capita energy use between Canada and the United States and, to a lesser extent, between Canada and the United States, Australia and Japan.  However, climate, geography and industrial structure do not explain differences between per capita energy use between Canada and the European and Scandinavian countries, and, to a lesser extent, Australia and Japan. This is indicated by the fact that adjusted Canadian per capita energy use is higher than actual per capita energy use in these countries in all three years considered. Other factors which are more amenable to choice and therefore influence by climate change policy may explain the remaining differences in per capita energy use. Possibilities include, for example, dwelling size and type, appliance penetration per household, household size, the prevalence of central and water heating, integrity of building envelopes, household equipment fuel efficiencies and service levels, choice of fuels, service sector floor area per employee or per dollar of service sector output, penetration of office equipment and the average power and fuel efficiency of that equipment and industrial equipment types and fuel efficiencies. It is, however, beyond the scope of this analysis to examine the influence of these and other factors.  For the OECD-12 and EU-9, the percentage amount by which the adjustments reduce energy use increased from 1984 to 1994. In other words, the power of the three factors  163  (climate, geography and industrial structure) to account for differences in energy use between Canada and Europe increased from 1984 to 1994. For the other aggregations and for the US, the amount by which the adjustments reduce energy use was approximately the same in 1984 and 1994, but somewhat lower in 1990. The same patterns generally hold in the case of per capita energy use comparisons.  The top half of Table 6.3 shows per capita energy use in Canada, the United States and the four country aggregations. It also shows Canadian per capita energy use broken down into the three components discussed in the introduction to this chapter. The bottom half of Table 6.3 shows the energy conservation potential for Canada that is implied by the comparisons. The calculation of conservation potential was based on the assumption that each country or country aggregation's GHG emission reduction commitment under the Kyoto Protocol reflected its potential for energy conservation, which is probably not an entirely realistic assumption. This calculation is conducted by multiplying 61  component #1 associated with a particular country aggregation by the aggregation's Kyoto Protocol commitment (expressed as a percentage of 1990 emission levels). In this calculation, the energy conservation potential associated with component #2 is assumed to be zero. This is generous, since, as discussed above, the energy conservation potential associated with component #2 should be greater than zero, and could be quite high for Canada. The energy conservation potential associated with component #3 is 100 percent, since, as discussed above, this component represents waste. There is one exception. In the comparison with the United States, component #3 was negative. In other words,  164  adjusting for non-energy factors reduced Canadian per capita energy use below US levels. Therefore, it is assumed that zero percent energy conservation potential would apply to component #3 in this case.  Compared to the 12 other OECD countries considered, Canada's energy conservation potential is 16 percent, more than double the six percent GHG emission reduction to which Canada committed under the Kyoto Protocol. Compared to Europe and Scandinavia, Canada's energy conservation potential is as high as one-third. However, compared to the Pacific Rim countries, Canada's implied conservation potential is just above 6 percent, and compared to the US alone, it is negative (i.e., Canadian per capita energy use would grow by 4.5 percent from 1990 levels).  Clearly, the implied energy conservation potential depends on the country grouping to which Canada is being compared. Using the OECD-12 aggregation as a proxy for all of the industrialized countries, one might conclude that Canada's commitment in Kyoto might have been much higher, as high as 16 percent (keeping in mind the strong assumptions made in this calculation).  6.5 Summary and Conclusions In all cases, climate, geography and industrial structure "explain" differences in absolute levels of energy use between Canada and other industrialized countries. However, these three factors only "explain" differences in per capita energy use between Canada and the  I.e. it is assumed that the relationship between the reduction in per capita energy use and in GHG emissions is one-to-one. However, the relationship is still proportional, so this assumption, while being 61  165  United States. Adjusted per capita energy use in Canada is still higher than actual per capita energy use in all of the other country groupings.  Climate does not completely account for differences in per capita residential and service sector energy use between Canada and the European and Scandinavian countries and the 12 OECD countries overall. Industrial structure does not completely account for differences in per capita industrial energy use between Canada and any of the other countries. Geography (in its very rough approximation here) accounts for differences in per capita transport energy use between Canada and the United States, Australia and Japan but not between Canada and the European countries. Overall, these three factors fail to account for differences in per capita energy use between Canada and all the other countries except the United States in all three years considered (1984, 1990 and 1994) and the Pacific Rim countries in 1994.  These results suggest that climate, industrial structure and geography provide inadequate explanations of differences in per capita energy use, except when Canada is compared to the United States. Nor do the three factors do much better in accounting for differences in absolute levels of energy use. Canada is more energy-intensive than most other industrialized countries, even after accounting for differences in climate, industrial structure and geography. From this, one may tentatively conclude that other factors, including lifestyle and business decisions that may be influenced by policy, are important in determining the level of energy use in Canada, and that, therefore, there is scope for  heroic, is perhaps not entirely unreasonable. 166  reducing energy use (and, by extension, greenhouse gas emissions)fromcurrent levels in Canada.  A back-of-envelope calculation of what Canada's energy conservation potential could be shows that it could rangefromnegative 4.5 percent (when compared to the United States) to about one-third (when compared to Scandinavia). Assuming that there is a one-to-one relationship between energy conservation and GHG emission reduction, a comparison of Canada with 12 other OECD countries suggests that Canada's GHG emission reduction commitment under the Kyoto Protocol - a six percent reduction from 1990 levels - might have been too low and might have been as high as 16 percent.  167  Chapter 7 Conclusion  This thesis addressed three questions:  1. Which factors "explain" the changes in energy use and greenhouse gas emissions that have been observed since the early 1980s in Canada? 2. Is Canada "wasteful" of energy compared to other industrialized nations? 3. What are the future prospects for reducing energy use and related greenhouse gas emissions (focusing on CO2 emissions) in Canada, given current and expected climate change policy and the results of the analyses to address question 1?  Question 1 was addressed by decomposing changes in sectoral energy use into three components: changes in energy intensity, changes in the structure of the sector and changes in the level of sectoral activity. In the residential and service sectors, the influence of changes in weather was also isolated. Question 1 was also addressed by decomposing changes in sectoral greenhouse gas (GHG) emissions into four components: changes in fuel mix, changes in energy intensity, changes in the structure of the sector and changes in the level of sectoral activity.  Question 2 was addressed by adjusting sectoral energy use in Canada to account for differences between Canada and other OECD in climate, geography and industrial structure and then comparing adjusted energy use in Canada to energy use in other OECD  168  countries. If adjusted Canadian energy use is equal to or less than actual energy use in the other countries, then the factors may be said to "explain" differences in energy use . 62  These three factors were selected because they are often used in justification of the relatively high levels of energy use and GHG emissions in Canada.  Question 3 was addressed by extrapolating historical trends in energy use and fuel mix to the years 2000 and 2010. The year 2010 is the mid-point of the "commitment period" 2008-2012 in which Canadian GHG emissions must be no more than 94 percent of 1990 levels, according to Canada's commitment in the Kyoto Protocol.  The conclusions reached for each of the questions are discussed in turn below.  7.1 Which factors "explain" the changes in energy use and greenhouse gas emissions that have been observed since the early 1980s in Canada?  From 1984 to 1995, energy use in Canada rose by 18 percent. This increase was largely due to increased activity, which was partially offset by improved energy intensity. This pattern was observed at the aggregate national level and in all sectors but the industrial sector where energy intensity had a positive impact on energy use in both periods considered (1973-1983 and 1984-1996). Weather and the structure of activity had only minor impacts on aggregate energy use.  On a per capita basis, energy use in Canada is higher than all other countries considered. 169  The impact of changes in the structure of activity on energy use was generally positive (i.e., increasing energy use) but the magnitude of the influence variedfromone sector to another. Intermodal shifts infreighttransport had a large, positive impact on energy use; in passenger transport, they had a smaller but also positive impact on energy use. Changes in the share of floor area by building type in the service sector and in the share of industrial output by subsector also had small, positive impacts on energy use. The influence of structure on energy use was much more pronounced for manufacturing than for industry as a whole, and in the 1973-1983 period, structural change in manufacturing had a negative impact on energy use.  The influence of weather was considered in the residential and service sectors. In both sectors the weather-adjusted intensity effect was about three percent higher than the unadjusted intensity effect. Because 1995 was both colder in the winter and warmer in the summer than 1981, weather had a positive impact on energy use in both sectors; the influence of weather was more pronounced in the service sector than in the residential sector.  From 1984 to 1995, national-level emissions of carbon dioxide, methane and nitrous oxide increased. In all sectors, increased energy use was the primary reason for emission increases. In most sectors, fuel mix changes mitigated the impact of increased energy use, by varying degrees. Infreighttransport and the service and residential sectors, fuel mix changes were significant enough to lead to reductions in CH4 and N2O emissions, and lower increases in CO2 emissions. In industry and passenger transport, fuel mix  170  changes were less effective in reducing emissions. In general, fuel mix changes had a larger impact on CH4 and N2O emissions than on CO2 emissions. This is because the CH4 and N2O emission factors for electricity were much smaller than for most other fuels, especially the more GHG-intensive fuels.  In the post-1990 period, CO2 emissions grew in all sectors. Emissions of CH4 and N2O fell in industry, freight transport and the residential sector, and increased in passenger transport and the service sector. In almost all cases, emission increases were due to increased activity, partially offset by intensity reductions and fuel mix changes. The exception is passenger transport N2O emissions, where there was a small but positive fuel mix effect. Emissions in every sector declined in the early 1990*s, reflecting the downturn in activity due to the recession.  7.2 Is Canada "wasteful" of energy compared to other industrialized nations? Energy use in Canada was adjusted to account for differences in climate (expressed as heating degree-days), geography (expressed as passenger-km per capita and freight tonne-km per dollar of real GDP) and industrial structure between Canada and 12 other OECD countries. In all cases, climate, geography and industrial structure "explain" differences in absolute levels of energy use between Canada and other industrialized countries.  However, these three factors do not always explain differences in per capita energy use. Overall, they fail to account for differences in per capita energy use between Canada and  171  the European and Scandinavian countries and, to a lesser extent, Australia and Japan. Climate does not completely account for differences in per capita residential and service sector energy use between Canada and the European and Scandinavian countries and the 12 OECD countries overall. Industrial structure does not completely account for differences in per capita industrial energy use between Canada and any of the other countries. On the other hand, geography more than accounts for the differences in per capita transportation energy use between Canada and the United States, Australia and Japan, but it does not completely account for differences between Canada and Europe.  These results suggest that Canada is more energy-intensive than most other industrialized countries, even after accounting for differences in climate, industrial structure and geography. From this, one may tentatively conclude that other factors, including lifestyle and business decisions that may be influenced by policy, are important in determining the level of energy use in Canada, and that, therefore, there is scope for reducing energy use (and greenhouse gas emissions) in Canada.  7.3 What are the future prospects for reducing energy use and related greenhouse gas emissions (focusing on CO2 emissions) in Canada, given current and expected climate change policy and the results of the analyses to address question 1? Extending historical trends into the future gives a glimpse of what Business-As-Usual holds in store for Canada. With a few exceptions, Business-As-Usual will involve increased energy use and increased emissions of three important greenhouse gases (CO2, CH4 and N 0). There have been some positive trends in the 1990's, including structural 2  172  changes towards less energy-intensive activities, declines in the energy intensity of most activities and shifts to less GHG-intensive fuels in some sectors (but, unfortunately, not in all). With the exception of the manufacturing subsector, continuing with Business-AsUsual will mean growth in energy use or emissions or emission reductions that are insufficient to meet Canada's international commitments.  The extrapolation of historical trends also highlighted two important considerations for GHG emission reduction policy. First, at the aggregate level and in every sector, projected GHG emissions were lower when they were based on changing fuel shares but frozen energy use (at 1995 or 1996 levels, depending on the sector). This indicates the important role of energy conservation in reducing emissions. Further, as was discussed in Chapter 1, there are other social and environmental costs that are associated with energy use and which would therefore be reduced if energy use declined. Second, in some sectors, trends in fuel shares were towards a mix that which involved relatively high emissions of one or more GHGs but lower emissions of others. This indicates a potential trade-off in policy that focuses on changing fuel mix and demonstrates the importance of considering all GHGs in formulating such policy.  7.4 Summary and Conclusions Based on the results of the analyses conducted in this thesis, tentative answers to the three questions are as follows.  173  1. Which factors "explain" the changes in energy use and greenhouse gas emissions that have been observed since the early 1980s in Canada?  Changes in activity levels, structural change and weather all tended to increase energy use and GHG emissions while changes in energy intensity and weather-adjusted energy intensity act to reduce energy use and emissions. Changes in fuel mix generally acted to reduce GHG emissions and tended to have a larger impact on CH4 and N2O emissions than on CO2 emissions.  2. Is Canada "wasteful" of energy compared to other industrialized nations?  Compared to other OECD countries, Canada is a high energy user and high emitter of GHG emissions. None of the three non-energy-related contributing factors considered adequately accounted for differences in per capita energy use between Canada and 12 other OECD countries. Differences in climate accounted for differences in per capita energy use between Canada and the United States, Australia and Japan, but not between Canada and the European and Scandinavian countries. Differences in industrial structure failed to account for differences in per capita energy use between Canada and any of the 12 other OECD countries. Differences in geography account for differences in per capita energy use between Canada and the United States, Australia and Japan, but not between Canada and Europe.  174  Overall, climate, geography and industrial structure accounted for differences in total per capita energy use between Canada and the United States (and between Canada and the US, Australia and Japan in 1994). These three factors did not, however, adequately account for differences in per capita energy use between Canada and the European and Scandinavian countries. Therefore, based on the results of this limited and back-ofenvelope analysis, it might be concluded that Canada is indeed "wasteful" of energy compared to Europe and Scandinavia but not compared to the United States and, to a lesser extent, Australia and Japan.  3. What are the future prospects for reducing energy use and related greenhouse gas emissions (focusing on CO2 emissions) in Canada, given current and expected climate change policy and the results of the analyses to address question 1?  The prospects are not encouraging. If past trends continue, both energy use and GHG emissions will continue to grow and Canada will be unable to meet its emission reduction commitment under the Kyoto Protocol. There have been some positive trends, especially in terms of structural shifts, but these will be largely insufficient. The one exception is the manufacturing subsector, in which extrapolation of historical trends yielded projections of declining GHG emissions.  175  References Ad Hoc Group On The Berlin Mandate (AGBM) (1997). "Information Submitted By Parties On Possible Criteria For Differentiation. Note by the secretariat". Document FCCC/AGBM/1997/MISC.3. United Nations. Anonymous (1994). "Climate Change Convention enters into force". United Nations Climate Change Bulletin. Issue 3:1-2 (approximate). Anonymous (1995a). "Ratification Update: The Berlin COP". United Nations Climate Change Bulletin. Issue 6: 10-11 (approximate). Anonymous (1995b). "The Berlin Mandate: Decision 1/CP.l [verbatim version of Berlin Mandate text]". United Nations Climate Change Bulletin. Issue 7: 9-11. Bolin, Bert (1996). "A brief overview of the LPCC Second Assessment Report". United Nations Climate Change Bulletin. Issue 1.0: 1-3 (approximate). Boyd, Gale A., Donald A. Hanson and Thomas Sterner (1988). "Decomposition of changes in energy intensity: A comparison of the Divisia index and other methods". Energy Economics. October: 309-312. BC Ministry of Environment, Lands and Parks (1995). Clean Vehicles and Fuels for British Columbia: A Policy Paper. Victoria, BC: BC Ministry of Environment, Lands and Parks. April. Canada (1994). Canada's National Report on Climate Change: Actions to Meet Commitments Under the United Nations Framework Convention on Climate Change. Ottawa: Environment Canada. Canada (1995). Canada's National Action Program on Climate Change. Canada (1997). Canadian Submission to AGBM 6: Elaboration ofPolicies and Measures in a Protocol or Other Legal Instrument. March 4. Canada (1998a). " Joint Meeting of Energy and Environment Ministers, Halifax, Nova Scotia - October 19-20, 1998". News Release Ref: 830-633/50. October 20. Canada (1998b). "Federal Government Takes Concrete Action On Climate Change". News Release. October 19. Canadian Global Change Program (CGCP) (1996,). Canada and Climate Change: Responding to Challenges and Opportunities. A submission to Canada's Provincial and Federal Ministers of Energy and Environment. November. Canadian Options for Greenhouse Gas Emission Reduction (COGGER) Panel (1993). 176  Canadian Optionsfor Greenhouse Gas Emission Reduction: Final Report of the COGGER Panel to the Canadian Global Change Program and the Canadian Climate Program Board. Canadian Global Change Program Technical Report Series No. 93-1. Ottawa: Royal Society of Canada. September. Climate Network Europe and US Climate Action Network (coordinators) (1996). Independent NGO Evaluations ofNational Plansfor Climate Change Mitigation: OECD Countries. Collas, Pascale (1997). Climate Change International, Global Air Issues Branch, Environmental Protection Service, Environment Canada, personal communication, March 18, 1997. Comeau. Louise (1996). "Canada [Country Report]". In Climate Network Europe and US Climate Action Network (coordinators): pp. 12-13. Cutajar, Michael Zammit (1995). '"Berlin Mandate' opens way for post-2000 commitments". United Nations Climate Change Bulletin. Issue 7: 1-3 (approximate). Energy, Mines and Resources Canada (1976). An Energy Strategy for Canada: Policies for Self-Reliance. Ottawa: Supply and Services Canada. Cited in Thirsk and Wright (1977). Environment Canada (1996). 'Tederal actions on climate change: next steps". Series of Backgrounders on the Environment Canada Green Lane Internet site. December 13. Environment Canada (1996). Canada's Second National Report on Climate Change. Actions to meet commitments under the United Nations Framework Convention on Climate Change. Cat. no. En21-125/1997E. Ottawa. Environment Canada. Environment Canada and Natural Resources Canada (1996). "Government of Canada Announces New and Strengthened Climate Change Initiatives". Press Release. December 12. Fisher, I. (1972). The Making ofIndex Numbers. 3 Edition. Boston: Houghton Mifflin. rd  Fuss, M. and L. Waverman (1975). The Demandfor Energy in Canada. Report for the Department of Energy, Mines and Resources. February. Cited in Thirsk and Wright (1977).  Goree, Langston James, VI (1998). "COP-4 Concludes - ENB Briefing Note". Earth Negotiations Bulletin, International Institute for Sustainable Development. 177  Electronic mail message, November 14. Grubel, H.G. and S. Sydneysmith (1975). "The Taxation of Windfall Gains on Stocks of Natural Resources". Canadian Public Policy. 1(1). Cited in Thirsk and Wright (1977). Harnett, Donald L. (1982). Statistical Methods. 3 Edition. Reading, Massachusetts. Addison-Wesley Publishing Company. rd  Houghton, J.T., L G . Meira Filho, J. Bruce, Hoesung Lee, B A . Callander, E. Haites, N. Harris and K. Maskell (editors) (1995). Climate Change 1994: Radiative Forcing of Climate and An Evaluation of the IPCC IS92 Emission Scenarios. Cambridge, UK: Cambridge University Press. Howarth, Richard B., Lee Schipper, Peter A. Duerr and Steinar Strom (1991). "Manufacturing energy use in eight OECD countries". Energy Economics. April: 135-142. Informetrica Limited (1995). Energy Efficiency Indicators: Conceptual Framework and Data Sources. Report prepared for Demand Policy and Analysis Division, Energy Demand Branch, Natural Resources Canada. August. Intergovernmental Panel on Climate Change (IPCC) (1995a). Climate Change 1995, The IPCC Second Assessment Synthesis of Scientific-Technical Information Relevant to Interpreting Article 2 of the UN Framework Convention on Climate Change. Jaques, A.P (1992). "Canada's Greenhouse Gas Emissions: Estimates for 1990". Report EPS 5/AP/4. Environment Canada. Marbek Resource Consultants, Torrie Smith Associates, Diener and Associates and Statistics Canada (1989). Energy Demand in Canada, 1973-1987: A Retrospective Analysis. Prepared for Energy, Mines and Resources. Ottawa: Marbek Resource Consultants. August. Margolick, Michael (1997). The Role of Government: A Briefing Paper to the Honourable Paul Martin. September 20, 1997. Vancouver, BC: David Suzuki Foundation. October. Natural Resources Canada (1996). Energy Efficiency Trends in Canada. Demand Policy and Analysis Division, Energy Efficiency Branch, Natural Resources Canada. Ottawa: Minister of Supply and Services. April.  Natural Resources Canada (NRCan) (1997a). Energy Efficiency Trends in Canada 1990 to 1995. May. Ottawa: Minister of Public Works and Goverment Services Canada. 178  Natural Resources Canada (NRCan) (1997b). Energy Efficiency Indicators Database. Availablefromthe website: http://oee.nrcan.gc.ca/general/trends/index_e.htm Organisation for Economic Coordination and Development (OECD) (1995). OECD Environmental Data Compendium 1995. Paris: OECD. Park, Se-Hark (1992). "Decomposition of industrial energy consumption". Energy Economics. 14(4): 265-270. Rolfe, Chris (1998). Kyoto Protocol to the United Nations Framework Convention on Climate Change: A Guide to the Protocol and Analysis ofIts Effectiveness. Vancouver: West Coast Environmental Law Association. Schipper, Lee, Fridtjof Unander, Celine Marie, Michael Ting, Marta Khrushch and Tom Krackeler,(1998). Energy Use In Canada In An International Perspective: Comparison  Of Trends Through The Early 1990s. Draft manuscript. Berkeley, CA: International Energy Studies Lawrence Berkeley Laboratory. Schipper, Lee and Stephen Meyers with Richard B. Howarth and Ruth Steiner (1992). Energy Efficiency and Human Activity: Past Trends, Future Prospects. Cambridge, UK: Cambridge University Press. Statistics Canada (1994). Human Activity and the Environment. Catalogue 11-509E. Ottawa: Statistics Canada. Thirsk, Wayne R. and Robert R. Wright (1977). "The Impact of the Crude Oil Subsidy on Economic Efficiency in Canada". Canadian Public Policy. ni(3): 355-364. United Nations (1992). United Nations Framework Convention on Climate Change. Wahby, Mandy J. (1984). "Petroleum Taxation and Efficiency: The Canadian System in Question". Journal of Energy and Development. 9(1): 111-127.  179  Appendix Energy Use Decompositions: Comparison to Other Analyses In general, the results presented in Chapter 3 are consistent with similar analyses conducted for Canada. These other analyses include twofromNatural Resources Canada (1996; 1997b) and onefromThe international Energy Agency/Lawrence Berkeley Laboratories (Schipper et al, 1997). NRCan (1996) considers the period 1984-1996 while NRCan, 1997b considers the period 198x-199x, depending on the sector. The base year in the former study is 1984 and the base year in the latter study is 1990. Similarly, Schipper et al (1997) consider the period 198x-199x, depending on the sector, and use 1990 as the base year.  Comparisons of Chapter 3 results with resultsfromthese studies are broken down by sector and presented in Tables A. 1 to A. 15. In order to compare the resultsfromChapter 3 with resultsfromNRCan (1997b) and Schipper et al (1997), the decompositions in Chapter 3 were recalculated using 1990 as base year. Similarly, the decompositions were recalculated using 1984 as base year in order to compare Chapter 3 results with NRCan, 1996). Any remaining differences in the results, as well as possible reasons for these differences, are discussed below.  Industrial Sector The difference between my results and Schipper et al likely stemsfromslight differences in data. They do not appear to include construction, forestry or mining, which is why I compared my decomposition for manufacturing with their results, and their subsectoral definitions are different than those used by NRCan. NRCan uses the Standard Industrial  180  Classification (SIC) scheme of 1980 while Schipper et al use the previous SIC scheme. Table A.2 lists the subsectors used in each study, along with their SICs. In addition, Schipper et al use real value-added in 1990 dollars as the activity variable while NRCan and I use real GDP in 1986 dollars.  There is a slight difference between my results (with base year 1984) and the results from NRCan (1996), but this may be explained by data differences: NRCan may have made small changes to the industrial data set since the 1996 analysis was published.  Freight Transport Schipper et al (1997), whose analysis extends only to 1995, find a smaller change in freight transport energy use than NRCan (1997b) or I do. They also find larger activity, structure and intensity effects, but the effects are in the same direction in all three analyses. NRCan (1997b) and Ifindsimilar magnitude overall energy use changes and activity effects, but I find smaller structure and intensity effects.  The difference in results can be explained by two differences between the studies. First, Schipper et al (1997) use a different data set than NRCan and I, and, second, I conduct a more disaggregated analysis at the sectoral level than NRCan or Schipper et al (i.e., I count each weight class of trucks as a separate mode while they only count trucks as a whole). Table A. 5 presents revised decomposition results, with a recalculation of Schipper et al's results using NRCan data and a recalculation of my results counting only trucks as a whole. With these revisions, the results in all three studies are very similar. What is curious is the  181  slight difference in results between NRCan and my results and Schipper et al's results, since we are all presumably now using the same data and the same calculation.  NRCan (1996)finda larger change in energy use than I do (with Chapter 3 results recalculated using base year 1984), as well as larger activity and intensity effects, while I find a larger structure effect. The difference in our results may be explained by differences in aggregation (I separate trucks into weight classes, but NRCan may have only counted trucks as a whole) and/or data (NRCan may have updated the data set since publication of the 1994 results).  Passenger Transport Schipper etal (1997), whose analysis extends only to 1995, find a larger change in passenger transport energy use than NRCan (1997b) or I. Schipper et alfindthat changes in activity account for almost all the change in energy use (their intensity and structure effects are smaller than the residual of the decomposition), while NRCan and Ifindthat a large activity effect was partially offset by the intensity effect. NRCan (1997b) and Ifindsimilar magnitude overall energy use changes and activity effects, but Ifindsomewhat larger structure and intensity effects. The differences are less than one percent, however.  The difference in results can be explained by two differences between the studies. First, Schipper et al (1997) use a different data set than NRCan and I, and, second, I conduct a more disaggregated analysis at the sectoral level than NRCan or Schipper et al (i.e., I count each kind of light vehicle and bus as a separate mode while they only consider light vehicles  182  overall and buses overall). Table A. 8 presents revised decomposition results, with a recalculation of Schipper et al's results using NRCan data and a recalculation of my results counting only light vehicles and buses at the aggregate level. With these revisions, the results in all three studies are very similar  Comparing Chapter 3 results (recalculated using 1984 as base year) with NRCan (1996), there is a slightly smaller change in energy use in the latter, as well as slightly larger activity and intensity effects. The differences are less than one percent, however. NRCan (1996) find a positive structure effect, but Ifinda negative one. However, in both analyses the structure effect is very small, considerably lower than the residual of the decomposition. The difference in our results may be explained by differences in aggregation and/or data (NRCan may have updated the data set since publication of the 1994 results).  Service Sector Schipper et al (1997 find a smaller change in energy use than NRCan (1997b) or I. Schipper et alfindthat changes in activity account for almost all the change in energy use and are augmented by a small, but positive intensity effect. They do not calculate a structure effect, nor do they adjust the intensity effect for weather or calculate a weather effect. NRCan (1997b) and Ifindthat a large activity effect was partially offset by a much smaller but negative intensity effect. Our results are nearly identical until weather is incorporated; my weather-adjusted intensity effect is about twice as large as that calculated by NRCan and my weather effect is about three times as large as that calculated by NRCan. The signs are the  183  same, however: we both find negative weather-adjusted intensity effects and positive weather effects.  The difference in results can be explained by three differences between the studies. First, Schipper et al (1997) use a different data set than NRCan and I. In the data set used by Schipper et al, service sector energy use is higher and floor area is lower than in the data set used by NRCan and I. Second, Schipper et al calculate intensity as energy use per dollar of commercial sector real GDP (converted to 1990 US dollars using Purchasing Power Parities) while NRCan and I calculate intensity as energy use per m of floor area. Third, NRCan and 2  I conduct a more disaggregated analysis at the sectoral level than Schipper et al (i.e., Schipper et al do not disaggregate by building type). Table A l l presents revised decomposition results, with a recalculation of my results counting only aggregate level energy use and floor area. My revised results are still differentfromSchipper et a/'s results, but this is to be expected since we are using different data and calculating energy intensity in different ways. The difference in calculated intensity effects is smaller, but I still find a negative intensity effect while Schipper et al find a positive intensity effect.  NRCan (1996)finda larger change in energy use than I do (using base year 1984), as well as slightly larger activity and structure effects. The differences are less than one percent, however. I alsofinda larger weather-adjusted intensity effect. NRCan (1996)finda positive weather effect, but Ifinda negative one. However, in both analyses the weather effect is very small, considerably lower than the residual of the decomposition, which is fairly large in  184  both analyses. The difference in our results may be explained by differences in data: NRCan may have updated the data set since publication of the 1994 results.  A curious byproduct of these comparisons is the calculation of intensity effects of significantly different magnitudes when different base years are used. As discussed above, the energy intensities of all building types declined over the period 1981 to 1995. However, as can be seen in Figure 3.41, most of the decline in energy in intensity occurred in the 1980's, which is why the decompositions with base year 1990 show relatively small intensity effects.  Residential Sector Schipper et al (1997)findthat a positive activity effect is more than offset by negative structure and intensity effects, resulting in a small decline in energy use. NRCan (1997b) and Ifindthat a relatively large positive activity effect is not completely offset by the negative intensity effect, leading to a small increase in energy use. The weather-adjusted intensity effect nearly offsets the activity effect, but in this case a positive weather effect results in a positive change in energy use. In addition, NRCan (1997b) calculates a small but positive structure effect.  The difference in results can be explained by two differences between the studies. First, Schipper et al (1997) use a different data set than NRCan (1997b) and I. In particular, they find a significantly greater change in energy use from each year to 1990. To compare Schipper et ats results with mine, I normalized both of our results so that the change in  185  energy compared to 1990 in any year was one percent and prorated the Laspeyres indices accordingly. Even with this normalization, our results are significantly different, as might be expected. In fact, in some years our activity, structure and intensity effects have different signs. Figures A. 1 and A.2 show these normalized results (Figure A. 1 shows the activity effects and Figure A.2 shows the other effects). Table A. 14 compares our normalized decomposition results for 1995.  Second, NRCan (1997b) and Schipper et al (1997) conduct a more disaggregated analysis at the sectoral level than I do. I do not calculate a structure effect at the aggregate level because, as mentioned above, many of the end-uses are used by most or all households. NRCan (1997b) and Schipper et al (1997) both calculate structure effects within each end-use, and add these together to determine the aggregate level structure effect. For instance, Schipper et al (1997) calculate the structure effect for lighting as the ratio of m of living space per 2  person in year t to m of living space per person in 1990. The structure effects for cooking and water heating are both calculated by Schipper et al (1997) as the ratio of the square root of household size (persons per household) in year t to the square root of household size in 1990. The method used by Schipper et al (1997) to calculate the space heating structure effect is not obviousfromtheir calculations, but appears to be the change in total square metres per personfromyear t to 1990 plus the change in centrally-heated m per person from 2  year t to 1990, all divided by the sum of total m per person plus centrally-heated m per 2  2  person in 1990. The method of calculation used by NRCan (1997b) is not known since calculations are not provided in the Excel spreadsheet (i.e., NRCan, 1997b) nor in NRCan (1997a).  186  I chose not to follow the approach used by Schipper et al (1997), instead calculating a partial decomposition at the aggregate level and a full decomposition for space heating only, showing the influence of heating equipment type on energy use. This is because their calculations of subsectoral structure and intensity effects are not directly consistent with the Laspeyres index approach I develop in Chapter 2, so if I followed Schipper et afs approach, the residential sector decomposition would not be methodologically comparable with decompositions for other sectors.  NRCan (1996)finda slightly larger change in energy use than I do (with base year 1984), as well as a larger activity effect. My (weather-adjusted) intensity effect is larger than that calculated by NRCan (1996). As discussed in Chapter 3,1 do not calculate a structure effect, but that calculated by NRCan (1996) is very small. Our weather effects are similar, which is curious since weather effects in the service sector, calculated using the same degree-day data, were considerably different. This is because NRCan (1996) did not adjust space cooling intensity for weather (which may also explain at least some of the difference in the intensity effect). It may also be due to differences in calculated energy intensities.  187  Note to Tables  The reader should recognize that the values presented in the following tables are the result of calculations. The fact that the results are presented with one or more significant digits (i.e., decimal points) should not be interpreted as an indication of the accuracy of the estimates.  188  Table 1.1: Average Annual C 0 Emissions and Sinks, 1980-89 (GtC/yr) C 0 Sources: Emissions from fossil fuel combustion and cement production 5.5 + 0.5 Net emissions from changes in tropical land use 1.6 ± 1.0 2  2  Total anthropogenic emissions  7.1 ± 1.1  C 0 Sinks: Oceanic uptake Uptake by Northern Hemisphere forest regrowth Additional terrestrial sinks (residual)  2.0 + 0.8 0.5 ± 0 . 5 1.4 ± 1.5  2  Storage in atmosphere After Houghton etal (1995): Table 1.3  3.2 + 0.2  Table 1.2: C 0 Emissions Corresponding to Stabilization at Different Atmospheric C 0 Concentrations, IPCC Estimates Stabilization at Atmospheric Concentration Corresponds to Accumulated (1990of... 2100) C 0 Emissions of... 350 ppmv (current concentration) 300 - 430 GtC 450 ppmv 640 - 800 GtC 550 ppmv (double pre-industrial) 880 - 1060 GtC 650 1000- 1240 GtC 750 1220 - 1420 GtC 1980's levels (7.1 ± 1.1 GtC/yr.) 660 - 902 GtC IPCC Scenario 1992c 770 GtC IPCC Scenario 1992e 2190 GtC After Houghton etal (1995): Table a. gigatonnes of carbon 2  2  189  2  o  CD X! c CD  •a  u u  p  i-h  CU  £ 1  O u, • « ^ T3 Tl © cd td j - „ gCD iU cc <» o? 33 n c -a  CD  =3  ^  CO  CD 1C C D  CO  CD W X!  .-3  45  00 T3  S  ~*  00  -a c S3 cd  t> .2 S X «U "3Q U O  s f &  Q  w u w c Q w u z  «  o  O .Si,  ~  o  '5 m  s «  CD  3  >  T3  > J3  CD  CD P-  u  r, CD co $3 cn CD CD i c a , cn i- O <- *S <a zr CD Si 3 (D , ^ - a o i § °, CD 8 3 CD cu £ T(03 o I •o P *~ S3 CO « x : ^ CD > _j cd O ,i=l O 3 £ X! S3 CO O CD O  cc  z z -<  ed ^ * £ 3 e  6 , 0  «  cn  la O  6  p  I- 3 cn CD cd -o o cn 1  td O  I I- \S i 00 & 6  (3  -M u  u. CD +^ 3O O oo o e ^ o c c c op,, l - CD c+5 :> CD 3 13 CD 3 ed '3 cd  mH  •a x>  CD  CU U  _ e OS  o  H H Z  SI = o w  5  X  Of)  o  e « s  X  X  Q to  s © . O SB  X  a I  M  2  PM  o S3  cd .H D *± C vi  ^  -1  pa  o E  &  S a o  »  OS  Z  a  w  CD  >•> oo \-  3  o  S3  T3 O O  C  "3  l | s? S3  w  "S? ^ .2 1/5  3  -a  " CD  cd  O tn -j CD  K Q  •e ed  I  PQ oo  cd  s.  CD CD  3  S  g. «> « a, O  a>  o -  1  e3 a <H .2 c .tJ o w  e o  CO  JD  G  u  u V IQ «*•  2 13 13 £ 1-s e ** » § « CL) 1 1 5 S S I so "u - 3 co fa £ .5 13 .S o « S "> ' oH c ^ u 2 CD (U «S S .2 o > < o 58 a. ct» Q a3 £  CD  SO  T3 > ,o "E cCD >  O  O  fa <L> Ch O J D  C  00  O  13  -  KT  fa  O  «  o  CO  O  8 13  83 CJ  O  w  c  .ii  00  >  o S3  JH  CO oCD  o s a> CU eg u- ex 5 co •*o — C D g 8 a in 3 60 ^ CD -* CD >>_2 o .£ co T3 fid ^H a> o S..3 w fc a, CD S - — CD 2 w fc 00 co e 52 e _H 2 c a « 2 to <D «4H O  CO  O  6 0  * fa to m. * «» +3 CD U.  CO  -O  C  E3  CO  IS  CO CO  c CD o ooll -c o o  & £11 op g •S o E o c C CD CD C D J3  11  Ea  -  CD  CD  s  * .2 £; *co  , 5 es  I* 5 O U ,« G c e  CN ON  "3 (Sfi  1^  e SB W  ,s  s  ,«2 o L9 c o cu  '§  '3  CO  CD to >v  03  13 fa T3 C D "CD c o cC dD>  g  IH  ^ c2  o c  G +-» O  CD "3 •n  CD g W  •  IH CC  co  W  a on 6  oo c  P  CD  •C  o a.  s  > •c  c  ^ CO  CN >^ h  cn  CO  CD  s o  CU  1  •i-> c_> oo •3 O -O ed  u  e  <u  s-  u cn  e +•» o g  cd  £  £ ° 6  i W cd til fck s I?  ,Q 3  CD X> Q  eg u- C 3O cd •5 O co «  I'SJB  S  -  u>  e V  cd CO CJ a  CD cu - t>0 & to 6  «S "°  co O  e e  2 c2  o S <2 Q «  § §  _ CD 00 00 cd  O  CD "3 g> §> o & co 5 <u 0 c> o CD  8 6  £ .§> CD  11°  T3  CO  ° §  CD  c JX c cd cod CU +-» anX> CD —  H §o J3 3 O. t- CO II U td -3 £ ed 3 CD CD 00 00  fa a. o o  E •°, a  J CU  ,« c s o  x  a  o s I "S s o " -PH  iI Ha > g  i  eu  O  CU  s |Z  o  C  gB §  c  .2 '+-»  CD  3  CD cd G -3  e  C cd >» 2  •c o  cu  PS U  S & u  CO  S o  ed cu  O  C 5.  Z "53 cd cd «3 2  is J  3  C cd  W 3 t~-' fa  MH  oi  o  CD  S3 V  HO  2 s o O  « s p. i CD In • a t •° cS * a o IT*- ^ e  u  cn v  , IP  o Ol  O  00  o  ^  ^  CO  '3 .3 S  J8  o  >>  _ „ CO 00 CD  C  *  <H  i  O 03  o CD  >•  o lZ CD c d 2 o2 o - o '3 ex j>>: c "oh CM 3 <D o cu "55 > 2 CD 3 CD "O IH X+->! ca CD CD e t: w CX]| 5? O s  p s c o 3 ex c 2 £? ca ^ 42 — CD C co 5 * - CD O CD CD o .2 U o CD CD ca CD CC o • - .2 JD s ca > CD 5 > o >-» 03 s CD C o -~ ~ g. CD IH O *§ o 2 g G s CD c3o O03'•6 ca ex <D c3 CM +  >H  e S «  > ,  03  ca ca  IS ill  «H  r  .2 CD c ca a X)  s  op jz,  e »  IH  ,CD  c C oi o O. &  *H  CD  CD  Cm  .—I  CD  ?  I C rrO> H  &i  ca  _r o ° «*-  -es=o  X  , 5 es  S •© u  ,« a a ©  a  e es w S  •o  w  & ©  iC .© w 55 *  Is  •  M  X  | CM  CD CJ  cu  es  o o >  c o  c o  cu = es  u  I*  CN <N  CD CZ3  "U. c3 3  CD  CD  e  O  3  >  s.2  03 CN O  e  03 O  1 IH  00  CN  o  Ph  1 in CN  ca  S3  S3 co .2 T3 cd — <U ajc o > S 3 ed a. oS « >> cu cd CJ A . C S 00 3 S3 S3  s ©  IQ  S3 cd  c  X) 3  gpl  cd cuJ3 O 2Xi cu co cd cu U E O <U CO  u  o on CD  |  i  8 •« 3 o 2 o o a cu 3 cu cu . - Cu2 G. Xcu -o Vi J3 c > T3 . 2 ^ *§ § e+H cd o 2 ' O „ cu > •g td c2u 2— sa, ed •42 " >, s C C U O o D "co " E l 00 g C S 3 Uco o -2 CD C S 3 'cd ed ed S -2 cu cEu g •If •3 .S O m3 cu .o  l  C+H  4)  CO CO  1^  CO  ©  s <U y Q <u _T ©  «. .2 I S O ed CD o p  "ed E cu CI. O 00 e CC cd 2 X E U o CD op ° S 00 8 3 ed u. 5 n 2 Z 2 cu o "3 £ a g * « o E cu cd l  CD  Xi  'X! > cu. u •-cdPQ oo Xi [T. W Ho cd . o <° o o 5 'a CD  CO  U  PH  ^ O  6 Z  ">  JC  a. sa||  cu ^  ed o  xj .& a  3 "Si  ."3 • I S ffl I I CD CD o 5 PH « X  PH  co" S3 cH T3 O CU -PS to cu cu  tf C3  O  i i " c/5  •- 2  s •5 « O CD ,o e  ON  s Ml tfS o  X  es w  ,s  "O e o IS I  cu  cu _>  "•5 cu *  c ©  CU  S  z  $3 O X O • cu ed pl, X CD .3- S U E v? Xi CD  2 '3 O ~ w  —.  CO  CD  ed • — S3 -tJ O S3  HS  ^  s  s  cu X S3  ^~  .2 ^ td 2 2 "  SU 13 So -2 E N  H  K  •-  CO  ^=3 CU &^ a. psj "3 PQ '3 PH  od  CN  'S  I  cu  — rs  co  CO  •2 £ CU co .-3 C g so 3* ^ -2  2 a c 1 'S ?  SO ON. ON  !  W  o  SB  ca S U w  Q Q U  o  sw So  c^j  V  ft  IS  PQ  s  fe  >?  iO ©  a s  ft '  « ©  es  o  I'  ei—  <H  ca MH  P  o  O ;s oa CU C MH  a N J3 CO  C O C O CO  +-» TO  3 o. ccu .2 C .2 ••E -Sr o O cdo O CN C CU rv •a cu :>, E O § 3 oo 5 > U 5 O a .s s. - o £ > O oo t. CO  CO  •e §  I I .? a *f  F t: co cd CU  -o co—.3 c . -a CU 3 ° S o o cu •j3 © o r  x ! >c 13  ll  cx £P 3 o cu cu O G •G « JU > ^ cu co a cu cd S «S ScS co ex c o a^N•O* .3 T3 CU cu > ON cu a S "co a s •a.2 -o "& ca c to & CJ cu 2 *S G G o " 3 ca gl o <d E »*> Ca U> h * G _ O — <*J MH J3 g ^ "o oe 2 a. ra co: 3 O o £ 3 g '3 o c u 3 cu « 'x '3 CU e c es a CU s S X G y *> 5JH 8 CU CU a. o oppd X! On Z r.  U.  c  J  0  M  _ H  0 3  11  cu CO  1  >h  CO  »H  1  W (U  i  s« O U w  ON  e  s e o es w  s  *  c o  .© so c— •—  X  «t Ph  cu a a o .2-  -a e  cd  MH  CU CT  o  c a S o  w o a o CJ N co  CU CO  ©  eu &  Zes  T3 <U  ecud & e CO  co  on fi <N O r  R g-g ©5* WH W  a o a cu 5 ^- ex Ph 3 eu XI a. c co cd o '3 00 e cu S s ^ 3  CQ P-l  O  -a e  cu e • ',2 W cu  ^ w  >  g §5 1 icu  CJ  . cu  ^ m W <!  c o X) cd > s co •s I S CD  B  so G\  »  CO QO  ©  co CD CD  W  u fc o  3 2 cd  '-6  CU a _* CO — CD co CJ 1  C cD o  3 O 3 ^  1  o 00 § s e2 3& 731 1 T3 cd cd +3 co .2 t-H 1 "3 a CJ S « 2 0,33 > — cd co -CD £ CD a | a 73 * IH *a 3 .2 s3 j> r cr .2 _§ IH S CJ cd Xj cd , cu cj cu 2 c3u cucd — t » HH C c* .2 co 00 ,  CO  C  c  <D "9 .S 3 o es ^2 cd  ejus*  Q Q  cd 00 3  CD •£ >g  CO  CO CO  XI <u  CD  f>  CO  S> CD  -O ed O  £ O ^ 55 3 o £ H T3 3 s £ a cd "Cu-— CU £ 00X1 s ~ 3 ed CD !> > s .s J5 +-> g > X^ o g *S -S T3••5 « I € 3 C O Cu CU > 2 cu 2 v- -2J i 0 -c CU 6 9  0 S II-I 1  3 8 £ :s  cu di IH <2 6 CD J3 CJ *c S S cu CJ . £ 5/3 CU M H  o y  CU  ° ,5" o8 ^ CC  _ e «6  O  I 2 « 1  IS *> « ow "  B  oo  ON  s  0(1  cu  CC  C  _©  cs CU S  . .2 .© "so Ph  CO  Cu XI  3 o  CO  IH  CU  s  a" 3  ©  cu  s  fc  t:  CD  £ _2 cu £ m  cd >» 00 o cj cd CD -o H ed . 3 TJ- c d  PH  PH  eCDd  T3  a  Cm  -o e cd cn  13 5  © a 2 o cn 3 Si cd O 53 \ . H cu o N ed CD £ -a a . „ -a o O CD CD cu ed X! T 3 >~> \o cu cfcj C O * J3 ?ed |*s g g cn e d CJ ^ S « a U O *> s - *T3 2 a e Q cn £ -*-» a ao a Cu cn cd C o ^ o- <*$ Cu CD P - 1CD 3 3 <u oo >> 55 m 2 Ph cn a. E cn £ td OS r vCD B • i « „ O > S^. Cu O CU Cu to ed CU Q S.S W £ PQ O o 8 - f« e cu 13 S '3  e o w cn , V IP cu  CD  s  3  CU  •mm mm  l«  CU  CU  o  s o Q cu  ed . ^  ed ^ cd a c "3 .a  3  00  ccn .2 S-3 cn d CD a e •a * ±i u t5 u 3 . O T3 c« T3 r o CD cn -a i S cn cd 1 2 Cut) 8 2 .3 ° 2 « S o CD O a ^2 3 Cu o •S 3 oo cn . a O <U a o u > t<nu „^ cn a *3 •3 cn cn a cd ed S3 cnO , r-~ XCD 3 00.52 cd  O  I  x> cd cn B ed ed •3e cu 13 c3d a •3 o cd cu X! cu •a <u > CD * 3 •a3 O Xi 3 s — T3 « cd r3 cu c £ 2 ^  ?  0  CD  31  -a e  X  ag PH tf  -=  es o . © es i|  si  s  « O ej  ON s OS 12  e  w s IW  s o i .© 52  hH  P  CD  <U  00  C cd X U cu cd cu  e "cd  _>  .2 **  "S p— «*. o cu s  , *  s  s «  cu  \>  B E  00  ed eg 00 ^U S e .2 -C fd n ~ a oo xt , . 2 CD cn Pi E-> R  •I 3 5 '§> cu o 3 a . VO  R  CD «D  £ > o as  O JC O  3 O H *o o X CD CZ3  "3 a o  «  «  z s  oo'  =3  CN  a o  m  40. Best Practices, Agriculture  Name of Initiative  Education Regulation Economic Moral R&D, Demo. Info. Projects Incentive Suasion Provision  Inventory of technology and know-how to reduce GHG emissions from agriculture  PSAs on climate change  Brief Description  TABLE 1.3: SUMMARY OF FEDERAL GOVERNMENT INITIOLTIVES ANNOUNCED DECEMBER, 1996  o  X  T A B L E 3.0 SECTORAL ENERGY USE SHARES, 1984-1995 % Change 1995 1984 3.5% 40.8% Industry 39.4% -0.4% 8.9% 9.0% Freight Transport -0.1% 18.0% 18.0% Passenger Transport -4.7% 19.3% 20.2% Residential -3.0% 13.1% 13.5% Services  T A B L E 3.1 CANADA, 1995 O V E R A L L DECOMPOSITION RESULTS (BASE Y E A R 1984) % Change in Energy Activity Effect Structure Effect Weather-Adjusted Intensity Effect Weather Effect Residual  18.33% 27.95% 1.62% -8.46% 0.24% -3.03%  T A B L E 3.2 MANUFACTURING + MINING, 1983 DECOMPOSITION OF ENERGY USE (BASE Y E A R 1973) Manufacturing Total Industrial -0.34% -0.61% % Change in Energy Use 6.50% -5.58% Activity Effect -9.09% 0.05% Structure Effect 3.33% 6.40% Intensity Effect -1.08% -1.48% Residuals  T A B L E 3.4 MINING + MANUFACTURING SECTORS CHANGE IN ENERGY INTENSITY 1973 TO 1983 5.27% Manufacturing + Mining Mining 39.00% -6.42% -3.05%  Manufacturing Food & Beverages Other Manufacturing  0.27% 9.08%  Primary Metals  -2.73%  Non-metallic Mineral Products  4.43%  Paper & Allied Products  -0.50%  Chemical Products 201  T A B L E 3.5 COMPARISON OF ENERGY INTENSITY DECOMPOSITION RESULTS Marbek et al (1989) results Herbert (1998) results % Change in % Change in % Change in % Change in Energy Intensity Energy Efficiency Energy Intensity Energy Efficiency Manufacturing + Mining 6.0% 6.8% 5.95% 1973 - 1977 6.73% 13.2% 8.8% 8.79% 1973 - 1982 13.75% -5.8% -5.2% -5.72% 1973 - 1987 -5.43% Manufacturing 1973 - 1977 1973 - 1982  -5.96% -3.54%  1973 - 1987  -15.94%  -0.40% 9.74% -7.87%  -6.0% -3.5%  -6.0% 8.6%  -15.6%  -7.2%  T A B L E 3.6 MANUFACTURING + MINING, 1973-1983 CHANGE IN SUBSECTOR ACTIVITY SHARES AND A V E R A G E ENERGY INTENSITY Average Energy % Change in Activity Intensity (GJ/$) Share 8.8 -26.43% Mining 13.9 8.71% Food & Beverages 8.2 Other Manufacturing 21.01% Primary Metals  -15.39%  Non-metallic Mineral Products  -20.72%  70.3 60.6  Paper & Allied Products Chemical Products  -2.00% 48.09%  110.1 53.4  TABLE 3.7 MANUFACTURING, 1973-1983 CHANGE IN SUBSECTOR ACTIVITY SHARES AND A V E R A G E ENERGY INTENSITY % Change in Activity Average Energy Share Intensity (MJ/$) Food & Beverages - 3.63% 1.3.9 7.28% 8.2 Other Manufacturing Primary Metals  -24.99%  70.3  Non-metallic Mineral Products  -29.71%  60.6  Paper & Allied Products  -13.12%  110.1  31.28%  53.4  Chemical Products  202  T A B L E 3.8 INDUSTRIAL SECTOR, 1996 DECOMPOSITION OF ENERGY USE (BASE Y E A R 1984) Manufacturing Total Industrial 11.79% 23.14% % Change in Energy Use 6.54% 22.88% Activity Effect 0.79% 3.81% Structure Effect 0.03% 1.36% Intensity Effect 0.08% - 0.57% Residuals  T A B L E 3.9 INDUSTRY CHANGE IN ENERGY INTENSITY 1984 TO 1996 Economic Potential Actual -30% 0.21% Total industrial 19.91% -35% Mining 25.35% Construction -24.18% Forestry Manufacturing Pulp, Paper and Sawmills  - 5.72% 20.28% -14.44%  Iron and Steel Smelting and Refining  -10% -15%  -17.61% - 6.22% -15.78%  Cement Chemicals Petroleum Refining  -20% -45%  -20.74%  Other Manufacturing  - 1.93%  As estimated by Marbek et al (1989).  203  -15 to-20%  T A B L E 3.10 INDUSTRY, 1984-1996 CHANGE IN SUBSECTOR ACTIVITY SHARES AND A V E R A G E ENERGY INTENSITY Average Energy % Change in Activity Intensity (MJ/$) Share 13.55 9.59% Mining -13.13% -17.97% - 7.85%  Construction Forestry Pulp, Paper and Sawmills  -15.34% 37.80% - 2.00% - 0.36% 5.76% 3.62%  Iron and Steel Smelting and Refining Cement Chemicals Petroleum Refining Other Manufacturing  1.33 4.06 86.13 96.87 79.68 145.41 89.58 161.82 7.00  TABLE 3.11 MANUFACTURING, 1984-1996 CHANGE IN SUBSECTOR ACTIVITY SHARES AND A V E R A G E ENERGY INTENSITY Average Energy % Change in Activity Intensity (MJ/$) Share 86 13 -10.12% Pulp, Paper and Sawmills 96.87 -17.43% Iron and Steel 34.40%  Cement  - 4.41%  79.68 145.41  Chemicals Petroleum Refining  - 2.82%  89.58  3.15%  161.82  1.06%  7.00  Smelting and Refining  Other Manufacturing  204  T A B L E 3.12 FREIGHT TRANSPORT, 1996 DECOMPOSITION OF ENERGY USE (BASE Y E A R 1984) 20.49% % Change in Energy Use 14.16% Activity Effect 20.30% Structure Effect -9.78% Intensity Effect -4.20% Residuals  Marine Rail Trucks  T A B L E 3.13 FREIGHT TRANSPORT MODE SHARES Share of Freight Share of Freight % Change in t-km in 1984 t-km in 1996 Freight t-km Share 37.72% 30.89% -18.11% 47.16% 45.62% - 3.27% 15.13% 23.50% 55.33%  Marine Rail Trucks  T A B L E 3.14 F lHEIGHT TRANSPORT ENEFtGY SHARES Share of Freight Share of Freight % Change in Freight Energy Use in 1984 Energy Use in 1996 Energy Use Share 15.82% 15.50% - 2.04% 14.75% 11.82% -19.87% 69.44% 72.69% 4.68%  T A B L E 3.15 FREIGHT TRANSPORT CHANGE IN ENERG'¥ INTENSITY 1984 TO 1996 AND AVERA<GE M O D E SHARES Change in Intensity Average Mode Shares Freight Transport 5.54% Trucks -28.87% 18.64% Trucks Less than 4546 kg -21.70% 1.76% Trucks Between 4546 and 15000 kg -14.72% 2.26% Trucks Greater than 15000 kg -16.89% 14.61% Marine 26.25% 34.07% Rail -12.57% 47.30%  205  TABLE 3.16 FREIGHT TRUCKS, 1984-1995 TRUCK STOCK, FUEL EFFICIENCY, KILOMETRES PER TRUCK AND TONNE-KM PER TRUCK 1995 % Change 1984 50.63% 3710 2463 Gasoline Light Truck Stock (000) -20.83% 13.3 16.8 Gasoline Light Truck Stock Fuel Efficiency (L/100 km) Av'g Distance per Light Gasoline Truck (00 km)  171  209  22.22%  Gasoline Med/Hvy Truck Stock (000) Gasoline Med/Hvy Truck Stock Fuel Efficiency (L/100 km) Av'g Distance per Med/Hvy Gasoline Truck (00 km)  224 30.5  84.7 27.4  -62.19% -10.16%  474  593  25.11%  Diesel Light Truck Stock (000)  164  173  5.49%  Diesel Light Truck Stock Fuel Efficiency (L/100 km) Av'g Distance per Light Diesel Truck (00 km)  13.8 182  11.2 225  -18.84%  Diesel Med/Hvy Truck Stock (000) Diesel Med/Hvy Truck Stock Fuel Efficiency (L/100 km)  97.4 22.2 671  184.80% -9.39%  Av'g Distance per Med/Hvy Diesel Truck (00 km)  34.2 24.5 511  Diesel Heavy Truck Stock (000) Diesel Heavy Truck Stock Fuel Efficiency (L/100 km) Av'g Distance per Heavy Diesel Truck (00 km)  140 47.5 789  198 42.6  41.43% -10.32%  974  23.45%  Light truck tonne-km per $ Goods Production GDP  0.056  0.061  8.72%  Med/Hvy truck tonne-km per $ Goods Production GDP  0.102  0.075  -26.27%  Heavy truck tonne-km per $ Goods Production GDP  0.433  0.655  51.08%  137,658  168,012  22.05%  Goods Production Real GDP ($ million)  206  23.63%  31.31%  T A B L E 3.17 PASSENGER TRANSPORT, 1996 DECOMPOSITION OF ENERGY USE (BASE Y E A R 1984) 20.97% % Change in Energy Use 44.51% Activity Effect 0.76% Structure Effect -17.92% Intensity Effect - 6.38% Residuals T A B L E 3.18 PASSENGER TRANSPORT, 1984-1996 M<DDE SHARES AND A V E R A G E ENERGY INTENSITY Average Energy Intensity Share of Total Passenger-km (MJ/p-km) 1996 1984 % Change 2.32 0.17% 79.12% 78.99% Light Vehicle 1.82 40.42% 9.99% 36.75% Small Car 2.58 -23.50% 32.16% 24.61% Large Car 3.31 14.09% 39.91% 10.07% Light Truck 1.41 -17.96% 9.39% 7.70% Bus 0.97 -24.78% 2.65% School bus 1.99% 2.13 -17.41% 3.86% 3.18% Urban Transit 0.82 -12.46% 2.89% 2.53% Inter-City bus 2.08 0.68% 0.24% -65.04% Rail 2.86 18.26% Air 10.94% 12.94%  207  TABLE 3.19 PASSENGER TRANSPORT, 1984-1996 ENERGY INTENSITY AND A V E R A G E M O D E SHARES Energy Intensity (MJ/p-km) Average Share of p-km 1984-1995 1995 1984 % Change -17.28% 78.65% 2.56 2.12 Light Vehicles 39.47% 1.89 1.72 - 9.17% Small Cars -26.09% 27.44% 2.99 2.21 Large Cars 11.74% 3.59 3.09 -14.14% Light Trucks 1.38 - 2.01% 9.08% 1.40 Buses 2.44% 1.00 1.01 School Buses 1.01% 3.83% 2.13 2.14 0.55% Urban Transit -13.11% 2.81% 0.81 0.71 Inter-City Buses 0.44% 2.42 1.52 -37.36% Rail Air 3.30 2.58 -21.79% 11.83%  208  T A B L E 3.20 PASSENGER TRANSPORT, 1984-1995 CAR ENERGY USE, STOCK, FUEL EFFICIENCY, DISTANCE T R A V E L L E D AND PASSENGER-KILOMETRES 1984 1995 % Change 9.01% 823.0 755.0 Car Energy Demand (PJ)  Large cars (000) Small cars (000) Small car share  10,267.4 4,815.4 5,452.0 53.10%  12,624.6 4,822.6 7,802.0 61.80%  22.96% 0.15% 43.10% 43.10%  Stock Fuel Efficiency (L/100 km) Small Cars Large Cars  13.5 10.7 16.8  10.2 9.2 11.7  -24.44% -14.02% -30.36%  161,600.0 15,700.0  234,100.0 18,500.0  44.86% 17.83%  Small cars Large cars  295,197 157,420 137,776  397,508 246,105 151,403  34.66% 56.34% 9.89%  Small cars Large cars  11,485.4 6,124.8 5,360.5  13,424.8 8,311.6 5,113.2  16.89% 35.70% -4.61%  25.7  29.6  15.21%  Car Stock (000)  Total Distance Traveled (million km) Average Distance per Car (km) Passenger-kilometres  Passenger-km per person  Population (millions)  209  T A B L E 3.21 SERVICE SECTOR, 1995 DECOMPOSITION RESULTS (BASE Y E A R 1981) 18.07% % Change in Energy Use 57.10% Activity Effect 1.39% Structure Effect -26.27% Intensity Effect -14.14% Residuals Weather-Adjusted Intensity Effect Weather Effect Weather Residuals  -29.34% 4.34% -15.42%  210  % Change in Energy Activity Effect Intensity Effect Residuals Weather-Adjusted Intensity Effect 1 Weather Effect | Weather Residuals 5.34% -13.90% -15.41%  6.87% 6.19% -7.80%  5.39% -31.77% -30.63%  3.55%  3.02% -10.95%  3.87% -10.52%  4.60% -25.69%  3.66% -G.50%|  TABLE 3.22 SERVICE SECTOR, 1995 ENERGY DECOMPOSITION BY BUILDING TYPE Recreation Warehouse Hotel & Retail Other Office Religious Health Schools Restaurant Institutional 35.43% -25.22%|| 37.67% 58.54% 35.93% 20.22% -4.14% -22.86% -15.14% 87.46% -1.96%| 63.01% 84.91% 114.12% 62.82% 40.61% 16.28% 32.72% -27.75% -23.72% -15.54% -16.52% -34.98% -25.96% -33.66% -31.82% -36.06% -24.27% 0.47% -9.79% -10.37% -5.48% -29.71% -29.62% -11.80% -12.92% -30.93% -26.42% -18.69% -18.96% -38.31% -28.49% -37.53% -39.30% -36.21%  TABLE 3.23 SERVICE SECTOR, 1981-1995 FLOOR AREA SHARES AND A V E R A G E ENERGY INTENSITY BY BUILDING TYPE Average Energy Share in Floor Area Intensity (GJ/m2) 1981 1995 % Change 2.17 17.94% 15.15% -15.52% Schools -10.49% 3.47 Health 7.91% 7.08% 1.56 1.86% -25.98% Religious 2.51% 2.05 4.49% 17.70% 3.81% Other Institutional 1.93 36.30% 19.11% 26.05% Office 2.18 22.69% 23.52% 3.64% Retail 3.05 6.59% 3.76% Hotel & Restaurant 6.36% 2.08 5.24% 6.25% 19.33% Recreation -37.60% 1.09 14.43% 9.00% Warehouse T A B L E 3.24 SERVICE SECTOR, 1981-1995 ENERGY INTENSITY AND A V E R A G E FLOOR AREA SHARES BY BUILDING TYPE Average 1981 1995 % Change Floor Area Share 2.814 1.799 -36.06% 16.11% Energy Intensity (GJ/m2) Schools -39.30% 2.814 1.708 Weather-Adjusted Intensity (GJ/m2) 7.44% 2.980 -31.82% Health Energy Intensity (GJ/m2) 4.372 -36.21% 4.372 2.789 Weather-Adjusted Intensity (GJ/m2) 2.16% -33.66% Energy Intensity (GJ/m2) 1.978 1.312 Religious -37.53% Weather-Adjusted Intensity (GJ/m2) 1.978 1.236 4.19% 1.705 -34.98% Other Energy Intensity (GJ/m2) 2.622 Institutional 2.622 1.618 -38.31% Weather-Adjusted Intensity (GJ/m2) -25.96% 22.93% Energy Intensity (GJ/m2) 2.318 1.716 Office -28.49% Weather-Adjusted Intensity (GJ/m2) 2.318 1.658 2.429 2.028 -16.52% 23.48% Retail Energy Intensity (GJ/m2) 1.969 -18.96% Weather-Adjusted Intensity (GJ/m2) 2.429 -15.54% 6.54% Energy Intensity (GJ/m2) 3.376 2.851 Hotel & Restaurant 3.376 2.745 -18.69% Weather-Adjusted Intensity (GJ/m2) -27.75% 5.84% Recreation Energy Intensity (GJ/m2) 2.534 1.831 Weather-Adjusted Intensity (GJ/m2) 2.534 1.750 -30.93% 0.980 -23.72% 11.31% Warehouse Energy Intensity (GJ/m2) 1.285 Weather-Adjusted Intensity (GJ/m2) 1.285 0.945 -26.42%  212  T A B L E 3.25 SERVICE SECTOR, 1981-1995 ENERGY USE, FLOOR AREA AND ENERGY INTENSITY BY BUILDING TYPE 1995 % Change 1981 18.07% 933.68 790.78 Total energy (PJ) All Types of Buildings 57.10% 491.58 312.92 Total floor space(million m2) -24.84% 1.90 2.53 Intensity (GJ/m2) Schools  Total energy (PJ) Total floor space(million ml) Intensity (GJ/m2)  157.93 56.13 2.81  134.01 74.49 1.80  -15.14% 32.72% -36.06%  Health  Total energy (PJ) Total floor space(million ml) Intensity (GJ/m2)  108.20 24.75 4.37  103.73 34.80 2.98  -4.14% 40.61% -31.82%  Religious  Total energy (PJ) Total floor space(million m2) Intensity (GJ/m2)  15.54 7.86 1.98  11.99 9.14 1.31  -22.86% 16.28% -33.66%  Other Institutional  Total energy (PJ) Total floor space(million ml) Intensity (GJ/m2)  31.29 11.93 2.62  37.62 22.07 1.70  20.22% 84.91% -34.98%  Office  Total energy (PJ) Totalfloorspace(million m2) Intensity (GJ/m2)  138.62 59.80 2.32  219.77 128.04 1.72  58.54% 114.12% -25.96%  Retail  Total energy (PJ) Total floor space(million ml) Intensity (GJ/m2)  172.51 71.01 2.43  234.49 115.62 2.03  35.93% 62.82% -16.52%  Hotel & Restaurant  Total energy (PJ) Total floor space(million ml) Intensity (GJ/m2)  67.13 19.89 3.38  92.42 32.42 2.85  37.67% 63.01% -15.54%  Recreation  Total energy (PJ) Totalfloorspace(million ml Intensity (GJ/m2)  41.54 16.39 2.53  56.27 30.73 1.83  35.43% 87.46% -27.75%  Warehouse  Total energy (PT Total floor space(million ml Intensity (GJ/m2)  58.01 45.15 1.28  43.38 44.26 0.98  -25.22% -1.96% -23.72%  213  T A B L E 3.26 RESIDENTIAL SECTOR, 1995 DECOMPOSITION RESULTS (BASE Y E A R 1981) 13.46% % Change in Energy Use Activity Effect 31,57% -13.26% Intensity Effect - 4.85% Residuals Weather-Adjusted Intensity Effect Weather Effect Weather Residuals  -15.62% 2.80% - 5.28%  TABLE 3.27 RESIDENTIAL SECTOR, 1981-1995 HOUSEHOLDS, POPULATION /LND HOUSEHOLD SIZE 1981 1995 % Change 8.57 11.27 31.57% Households (millions) 24.90 29.61 18.92% Population (millions) -9.62% 2.91 2.63 Persons per Household  214  T A B L E 3.28 RESIDENTIAL SECTOR, 1981-1995 ENERGY USE AND INTENSITY BY END-USE 1981 1995 % Change Average Share of Energy Use 63.89% Space Heating 840.96 Total Energy (PJ) 825.63 1.86% 74.59 -22.58% 96.35 Energy Intensity (GJ/HH) 70.72 -26.60% 96.35 Weather-Adjusted Energy Intensity (GJ/HH) 0.31% Space Cooling 296.81% Total Energy (PJ) 1.84 7.28 2.34 73.75% Energy Intensity (GJ/HH) 1.35 1.35 2.84 110.97% Weather-Adjusted Energy Intensity (GJ/HH) 3,112,091 128.38% Households with Space 1,362,664 Cooling 19.17% Water Heating 211.17 285.48 35.19% Total Energy (PJ) 24.64 25.32 2.75% Energy Intensity (GJ/HH) 12.74% Appliances Total Energy (PJ) 132.28 185.89 40.53% 15.44 16.49 6.81% Energy Intensity (GJ/HH) 3.89% Lighting 34.70% Total Energy (PJ) 42.13 56.75 4.92 5.03 2.38% Energy Intensity (GJ/HH) Number of Households Heating Degree-Days Cooling Degree-Days  8,569,000  11,274,000  31.57%  4326 182  4563 221  5.48% 21.43%  T A B L E 3.29 RESIDENTIAL SPACE HEATING, 1981-1995 SHARE OF HOUSEHOLDS AND A V E R A G E ENERGY INTENSITY BY EQUIPMENT TYPE Average Energy Percentage of Households Intensity (GJ/HH) Using Equipment Type 1981 1995 % Change 95.69 -34.34% Normal efficiency equipment 73.57% 48.30% 74.49 6.40% 0% Medium efficiency equipment 50.94 0% 4.06% High efficiency equipment 65.98 56.01% 26.43% 41.23% Other equipment  T A B L E 3.30 RESIDENTIAL SPACE HEATING, 1995 DECOMPOSITION RESULTS (BASE YEAR 1981) 1.86% % Change in Energy Use 31.57% Activity Effect - 6.67% Structure Effect -14.05% Intensity Effect - 8.99% Residuals -18.52% 5.48% -10.01%  Weather-Adjusted Intensity Effect Weather Effect Weather Residuals  T A B L E 3.31 RESIDENTIAL SECTOR, 1981-1995 HOUSEHOLDS AND HEATED FLOOR AREA 1981 1995 % Change 8.6 11.3 31.57% Households (millions) 5.80% 105.0 111.1 Heated floor area per household (m ) 39.19% Total heated floor area (million m ) 899.7 1,252.4 2  216  T A B L E 3.32 RESIDENTIAL SECT OR, 1981-1995 APPLIANCE PENETRATION AND UNIT EN ERGY CONSUMPTION Penetration Rate Unit Energy Consumption (kwh) (% of households) 1981 1995 1995 % Change % Change 1981 95.00% 118.00% 24.21% 950 1180 24.21% Refrigerators & Combis 51.00% 55.00% Freezers 7.84% 1101 824 -25.16% Clothes-washers 74.00% 78.00% 5.41% 108 96 -11.11% 58.00% 73.00% Clothes-Dryers 25.86% 1455 1162 -20.14% Dish-washers 56.67% 187 126 -32.62% 30.00% 47.00% 15.90% 27.60% Air Conditioners 73.59%  TABLE 3.33 IIESIDENTIAL SECTOR, 1981-1995 DISTRIBUTION 0IF HOUSEHOLDS BY BU ILDING TYPE 1981 1995 % Change Single Detached 56.62% 55.52% - 1.94% Single Attached 8.49% 10.22% 20.43% Apartments 32.73% 32.34% - 1.18% Mobile Homes 2.17% 1.92% -11.34%  217  TABLE 4.1 GREENHOUSE GAS EMISSIONS AND ENERGY USE, 1984-1995 Percent Change Sectoral Shares Absolute Amounts 1984-89 1990-95 1984-95 1995 1990 1995 1984 1990 1984 Carbon Dioxide CO2) Emissions (million tonnes) 14.61% 7.59% 6.52% 417.1 364.0 391.6 Canada 18.03% 7.16% 81.8 87.6 20.40% 20.88% 21.00% 10.14% Passenger 74.2 Transport 18.16% 8.63% 8.78% 37.8 41.1 44.6 10.38% 10.50% 10.70% Freight Transport 8.62% 7.02% 1.50% 20.27% 20.16% 19.21% 78.9 80.1 73.8 Residential 13.80% 3.84% 9.59% 44.2 45.9 50.3 12.13% 11.71% 12.05% Services 15.29% 7.40% 7.34% 143.9 154.5 36.82% 36.76% 37.04% Industry 134.0 1  Methane (CH4) Emissions (tonnes) 70,290.3 71,546.6 Canada Passenger 55,434.6 59,006.9 Transport Freight Transport 11,874.3 9,723.2 837.2 876.8 Residential 552.5 448.3 Services 1,531.0 Industry 1,552.1 Nitrons Oxide (N 0) Emissions (tonnes) 36,499.6 36,775.1 Canada Passenger 5,369.9 4,574.9 Transport 24,248.6 26,020.1 Freight Transport Residential 2,922.1 2,592.3 1,015.9 713.4 Services Industry 2,943.1 2,874.3 Energy Use (PJ) 5,655.6 6,140.3 Canada Passenger 1,084.1 1,194.4 Transport 540.1 586.4 Freight Transport 1,220.0 1,325.0 Residential 768.3 805.8 Services Industry 2,043.1 2,228.9  74,370.2 64,045.0  78.87%  82.47%  86.12%  3.95% 8.54%  5.80% 15.53%  7,469.4 797.4 471.3 1,587.1  16.89% 1.25% 0.79% 2.21%  13.59% 1.17% 0.63% 2.14%  10.04% -18.12% -23.18% 1.07% -4.52% -4.75% 5.12% 0.63% -18.85% 3.67% 2.13% -1.36%  -37.10% -9.05% -14.70% 2.25%  38,016.8 3,651.6  14.71%  12.44%  0.75% 7.31%  3.38% 8.57%  4.16% 16.50%  28,250.5 2,258.4 719.3 3,137.0  66.44% 8.01% 2.78% 8.06%  70.75% 7.05% 1.94% 7.82%  74.31% -14.80% -20.18% 5.94% -11.29% -12.88% 1.89% -29.78% 0.83% 9.14% 8.25% -2.34%  -32.00% -22.72% -29.20% 6.59%  6,598.4 1,281.8  19.17%  19.45%  19.43%  8.57% 10.17%  7.46% 7.32%  16.67% 18.24%  636.7 1,376.4 886.6 2,416.9  9.55% 21.57% 13.58% 36.13%  9.55% 21.58% 13.12% 36.30%  9.65% 20.86% 13.44% 36.63%  8.57% 8.61% 4.88% 9.09%  8.58% 3.88% 10.04% 8.44%  17.89% 12.82% 15.41% 18.29%|  1.79% 6.44%  2  218  9.61%  I CANADA, 1995 TABLE G H G DECOMPOSITIONf RESULTS (BASE YEAR 1984) N 0 CH co 4.16% 5.80% % Change in Emissions 14.61% -10.43% -11.60% Fuel Mix Effect -2.33% 28.19% 28.19% 28.19% Activity Effect 1.98% 1.98% 1.98% Structure Effect -17.07% -17.07% -17.07% Intensity Effect -20.39% -19.20% -19.91% Residuals 4  2  2  T A B L E 4.:i CANADA, 1995 G H G DECOMPOSITIONI RESULTS (BASE Y E A R 1990) CELt N 0 co 6.52% 3.95% 3.38% % Change in Emissions -0.54% -3.27% -3.44% Fuel Mix Effect 11.24% 11.24% 11.24% Activity Effect 0.10% 0.10% 0.10% Structure Effect -4.17% -4.17% -4.17% Intensity Effect -0.12% -0.36% 0.04% Residuals 2  2  TABLE 4.4 CANADA, 1984-1995 FUEL SHARES 1984 1990 PJ Share PJ Share Natural gas 1554.5 25.76% 1743.1 26.50% 742.6 12.31% 640.1 9.73% Oil (DFO, LFO, HFO & kerosene Coal, coke, coke oven gas, 409.4 6.78% 404.6 6.15% petroleum coke and distilled gas Propane, LPGs and gas plant LPGs 46.7 0.77% 65.6 1.00% Wood 114.3 1.89% 99.4 1.51% 1229.5 20.38% Electricity 1499.4 22.79% Diesel 364.2 6.03% 442.3 6.72% Motor Gasoline 1039.6 17.23% 1060.7 16.12% Aviation Gasoline 5.9 0.10% 5.5 0.08% Aviation Turbo Fuel 148.9 2.47% 179.7 2.73% Other fuels 378.7 6.28% 438.3 6.66% All Fuels 6034.3 100.00% 6578.6 100.00%  219  1995 PJ Share 1981.5 27.75% 541.4 7.58% 420.0 120.7 91.7 1628.8 524.0 1105.3 4.1 181.0 541.7 7140.1  5.88% 1.69% 1.28% 22.81% 7.34% 15.48% 0.06% 2.53% 7.59% 100.00%  T A B L E 4.5 GREENHOUSE GAS EMISSION F A C T O R S C 0 (t/TJ) C H (kg/TJ) N 0 (kg/TJ) 0.62 49.68 0.70 Natural Gas 70.48 3.96 6.60 Diesel Fuel Oil, Light Fuel Oil and kerosene 6.35 74.00 1.8 Heavy Fuel Oil 88.30 n/a Coal & coke, coke oven gas n/a 100.10 0.38 Petroleum coke and distilled gas n/a Liquid Petroleum Gases (LPG's) 60.61 1.18 10.75 60.61 1.18 10.75 LPGs and gas plant LPGs 81.47 0.02 8.89 Fuel Wood 0.001296 60.84 0.000648 Electricity 6.85 Diesel Oil 70.69 3.51 64.015 27.10 Gasoline 67.98 Aviation Gasoline 69.37 60.00 6.86 Aviation Jet Fuel 70.84 2.00 6.40 64  4  2  2  65  TABLE 4.6 INDUSTRY, 1996 G H G DECOMPOSITIO*r RESULTS (BASE Y E A R 1984) N 0 co CH Industry 17.21% 7.84% 2.41% % Change in Emissions -2.30% -10.10% -14.63% Fuel Mix Effect 22.88% 22.88% 22.88% Activity Effect 2.04% 2.04% 2.04% Structure Effect -4.13% -4.13% Intensity Effect -4.13% -1.28% -2.84% -3.74% Residuals 2  4  2  TABLE 4.7 INDUSTRY, 1996 G H G DECOMPOSITKttI RESULTS (BASE Y E A R 1990) N 0 co CH Industry % Change in Emissions 4.26% -0.93% -6.14% -0.71% Fuel Mix Effect -5.65% -10.62% 11.60% 11.60% 11.60% Activity Effect Structure Effect -0.84% -0.84% -0.84% -4.63% Intensity Effect -4.63% -4.63% Residuals -1.40% -1.65% -1.16% 2  4  2  Source o f emission factors i s Table S.3 i n A P Jaques, 1992. "Canada's Greenhouse Gas E m i s s i o n s : Estimates for 1990". Report E P S 5 / A P / 4 . Environment Canada. E m i s s i o n factors for electricity were not provided i n Table S.3 o f Jaques, 1992. Therefore, they were estimated b y d i v i d i n g electricity energy use i n 1990 b y the relevant emissions reported i n Table S. 1 o f Jaques, 1992 for power generation. Electricity energy use was taken from N R C a n , Sectoral database. 6 5  220  T A B L E 4.8 INDUSTRY, 1984-1996 FUEL SHARES 1990 1984 PJ Share PJ Share 9.83% 174.9 6.68% Coal & coke, coke oven gas 233.7 576.6 24.27% 654.7 25.01% Electricity 683.1 28.75% 823.5 31.46% Natural gas 157.4 6.62% 118.9 4.54% DFO, LFO and kerosene 7.74% 201.4 8.47% 202.5 HFO 8.68% 172.2 7.25% 227.2 Petroleum coke and distilled gas 18.7 0.79% 27.2 1.04% LPGs and gas plant LPGs 333.1 14.02% 388.7 14.85% Other fuels 2376.3 100.00% 2617.6 100.00% All Fuels T A B L E 4.9 F R E KJHT TRANSPORT, 1996 G H G DECOMPOSITIONr RESULTS (BASE Y E A R 1984) N 0 CH co 20.78% % Change in Emissions -33.78% -29.26% 0.24% -45.04% -41.29% Fuel Mix Effect Activity Effect 14.16% 14.16% 14.16% 20.30% Structure Effect 20.30% 20.30% -9.78% -9.78% Intensity Effect -9.78% -4.15% -13.43% -12.66% Residuals 4  2  2  TABLE 4.10 FREI GHT TRANSPORT, 1996 G H G DECOMPOSITIONf RESULTS (BASE Y E A R 1990) CH N 0 co 11.03% -16.97% % Change in Emissions -19.13% 0.05% -27.13% -25.18% Fuel Mix Effect Activity Effect 14.69% 14.69% 14.69% 7.36% 7.36% Structure Effect 7.36% -9.71% -9.71% -9.71% Intensity Effect -1.36% -4.13% Residuals -4.34% 4  2  221  2  1996 Share PJ 175.1 5.98% 743.0 25.39% 928.7 31.74% 149.3 5.10% 159.3 5.44% 8.70% 254.7 41.0 1.40% 16.24% 475.1 2926.1 100.00%  T A B L E 4.11 FREIGHT TRANSPORT, 1984-1996 FUEL SHARES 1996 1990 1984 Share PJ Share PJ PJ Share 0.00% 0.0 0.00% 0.0 0.0 0.00% Coal 455.9 70.06% 299.8 55.50% 363.2 61.94% Diesel 120.9 18.58% 184.0 34.06% 150.0 25.57% Motor Gasoline 0.00% 0.3 0.05% 0.0 0.00% 0.0 LFO and KER 54.9 8.44% 50.1 60.1 10.26% Heavy Fuel Oil 9.28% 0.0 0.00% 0.0 0.00% 0.0 0.00% Electricity 2.41% 2.03% 15.7 Propane 6.1 1.13% 11.9 3.0 0.47% 0.1 1.2 0.20% Natural Gas 0.02% 586.4 100.00% 650.8 100.00% All Fuels 540.1 100.00%  TABLE 4.12 PASSE]NGER TRANSPORT, 1996 G H G DECOMPOSmO>[ RESULTS (BASE YEAR 1984) CH N 0 co 20.83% % Change in Emissions 16.57% 17.98% Fuel Mix Effect -0.11% -3.63% -2.47% Activity Effect 44.51% 44.51% 44.51% 0.76% 0.76% 0.76% Structure Effect -17.92% Intensity Effect -17.92% -17.92% -6.40% -7.14% -6.90% Residuals 4  2  2  TABLE 4.13 PASSE NGER TRANSPORT, 1996 G H G DECOMPOSITIO*I RESULTS (BASE Y E A R 1990) N 0 CH co 9.70% 9.95% % Change in Emissions 9.51% Fuel Mix Effect -0.08% -0.26% 0.14% 17.81% Activity Effect 17.81% 17.81% 1.65% 1.65% 1.65% Structure Effect -8.45% Intensity Effect -8.45% -8.45% -1.23% -1.25% Residuals -1.21% 4  2  222  2  TABLE 4.14 PASSENGER TRANSPORT, 1984-1996 FUEL SHARES 195*4 1996 1990 PJ Share Share PJ Share PJ 5.98% Diesel 64.4 5.94% 79.0 6.62% 78.4 Heavy Fuel Oil 0.0 0.00% 0.0 0.00% 0.0 0.00% Motor Gasoline 76.19% 855.6 78.93% 910.7 76.25% 999.1 Propane 0.58% 14.0 1.46% 6.3 1.17% 19.1 Natural Gas 4.8 0.36% 0.3 0.03% 2.3 0.19% Electricity 2.6 0.24% 3.1 0.26% 3.0 0.23% Aviation Gasoline 5.9 0.55% 5.5 0.46% 3.9 0.30% Aviation Turbo Fuel 15.48% 148.9 13.73% 179.7 15.05% 203.0 All Fuels 1084.1 100.00% 1194.4 100.00% 1311.3 100.00%  TABLE 4.15 SE1RVICE SECTOR, 1995 GHG DECOMPOSITIONf RESULTS (BASE YEAR 1981) N 0 co CH % Change in Emissions 14.26% -25.69% -44.85% Fuel M i x Effect -3.62% -37.32% -53.48% Activity Effect 57.10% 57.10% 57.10% Structure Effect 1.43% 1.43% 1.43% Weather-Adjusted -29.00% -29.00% -29.00% Intensity Effect Weather Effect 4.23% 4.23% 4.23% -15.87% Residuals -22.12% -25.12% 2  4  2  T A B L E 4.16 SE1RVICE SECTOR, 1995 G H G DECOMPOSITIONf RESULTS (BASE YEAR 1990) N 0 co CH % Change in Emissions 9.59% 5.12% 0.83% Fuel M i x Effect -0.41% -4.47% -8.37% Activity Effect 10.26% 10.26% 10.26% Structure Effect 0.36% 0.36% 0.36% Weather-Adjusted -4.47% -4.47% -4.47% Intensity Effect Weather Effect 4.13% 4.13% 4.13% -0.28% Residuals -0.69% -1.08% 2  4  223  2  Electricity Natural Gas LFO HFCOther All Fuels  TABLE 4.17 SERVICE SECTOR, 1981-1995 FUEL SHARES 1995 19!n 1990 Share PJ Share PJ PJ Share 44.01% 370.1 43.27% 410.9 259.4 32.81% 407.0 43.59% 323.0 40.84% 363.7 42.52% 59.7 6.39% 13.32% 61.3 7.16% 105.3 0.97% 10.7 1.25% 9.0 60.2 7.61% 47.0 5.04% 5.42% 49.6 5.79% 42.9 855.3 100.00% 933.7 100.00% 790.8 100.00%  T A B L E 4.18 RESH)ENTIAL SECTOR, 1995 G H G DECOMPOSITIOISr RESULTS (BASE Y E A R 1981) N 0 co CH 6.81% -29.57% -34.56% % Change in Emissions -5.86% -37.93% -42.33% Fuel M i x Effect 31.57% 31.57% 31.57% Activity Effect -16.58% -16.58% -16.58% Weather-Adjusted Intensity Effect Weather Effect 3.38% 3.38% 3.38% -5.69% -10.60% Residuals -10.01% 2  4  2  T A B L E 4.19 RESD>ENTIAL SECTOR, 1995 G H G D E C O M P O S I T EI RESULTS (BASE YEAR 1990) N 0 co CH 1.50% -4.75% -12.88% % Change in Emissions -2.29% -16.14% Fuel M i x Effect -8.31% 10.17% 10.17% 10.17% Activity Effect -9.35% -9.35% -9.35% Weather-Adjusted Intensity Effect Weather Effect 4.01% 4.01% 4.01% -1.05% -1.28% -1.58% Residuals 2  4  2  TABLE 4.20 RESIDE]NTIAL SECTOR, 1981-1995 FUEL SHARES 1995 1981 1990 PJ PJ Share PJ Share Share Natural gas 450.4 37.13% 552.5 41.70% 656.5 47.70% Oil 344.0 28.36% 186.6 14.08% 139.0 10.10% Coal 4.0 0.33% 2.5 0.19% 1.7 0.12% Propane 13.7 1.13% 12.5 0.94% 0.80% 11.0 84.8 99.4 91.7 6.66% Wood 6.99% 7.50% Electricity 316.2 26.07% 471.5 35.58% 476.5 34.62% A l l Fuels 1213.1 100.00% 1325.0 100.00% 1376.4 100.00%  224  oo o  OS 5tt. in  I  V S o 1 — 1 o O >S  OS O,  «N  G v  2  00 cn ra -a  . °*  >.  K ?  00  2  o,  tO in 00  H w  o,  2OA&Jm  Io  ":>  ©  u  o £  OS O,  CN CN  2 H o  oo m "s: oo a  OS w z w  s  1 -i  00  o,  2 00 r- O SO (-+ m oo"  P  BI  B w  S3  W  S o 35 o b  o  Calculated from Energy Use Growth Rates Calculated from Underlying Factors Activity Only Intensity Only Structure Only 2911.5  1996 Actual Energy Use (PJ)  3872.2 3721.2 2987.0 3013.1  3392.4 3307.2 3173.9 4704.7 2741.3 3055.2  3134.1 2937.1 2939.2  3052.4 3012.1 2971.2  3351.3 2865.9 2949.8  1984-96 av'g growth rates 3730.4 4113.0  1990-96 av'g growth rates 3795.0  4560.6  1984-89 av'g growth rates 4407.3  3157.5  1984-96 av'g growth rates 3136.4  3186.9  1990-96 av'g growth rates 3151.8  Energy Use in 2010 (PJ) with ...  3306.5  1984-89 av'g growth rates 3289.4  Energy Use in 2000 (PJ) with ...  TABLE 5.4 INDUSTRIAL ENERGY USE EXTRAPOLATIONS  TABLE 5.5 INDUSTRIAL ENERGY AND ACTIVITY ANNUAL AVERAGE RATES OF CHANGE 1984-89 1990-96 1984-96 Industry Energy Use 1.87% 1.75% 2.97% Real GDP ($86) 3.45% 1.06% 1.73% Manufacturing 3.05% 1.44% Energy Use 1.28% Real GDP ($86) 3.24% 1.85% 1.94%  TABLE 5.6 INDUSTRIAL ENERGY INTENSITY ANNUAL AVERAGE RATES OF CHANGE 1984-89 1990-96 1984-96 Total Industrial -0.47% 0.80% 0.02% Mining 0.85% 3.12% 1.52% Construction -3.13% 4.54% 1.90% Forestry -2.56% -7.17% -2.28% Manufacturing -0.18% -0.56% -0.49% Pulp, Paper and Sawmills 1.47% 1.38% 1.55% Iron and Steel -1.62% -0.20% -1.29% Smelting and Refining -1.39% -1.44% -1.60% Cement -0.54% 0.88% -0.53% Chemicals -1.90% 2.20% -1.42% Petroleum Refining -1.68% -1.49% -1.92% Other Manufacturing 1.47% -2.37% -0.16%  TABLE 5.7 INDUSTRIAL SUBSECTOR SHARES ANNUAL AVERAGE RATES OF CHANGE 1984-89 1990-96 1984-96 Mining -1.74% 2.62% 0.77% Construction 1.88% -4.20% -1.17% Forestry -0.43% -1.90% -1.64% Pulp, Paper and Sawmills -0.83% -0.20% -0.68% Iron and Steel -1.55% 1.21% -1.38% Smelting and Refining 0.07% 4.50% 2.71% Cement 3.52% -3.28% -0.17% Chemicals 1.11% -1.47% -0.03% Petroleum Refining -0.83% 0.17% 0.47% Other Manufacturing -0.14% 0.87% 0.30%  227  Tin  O -iS 0O 0  °  Ofl m ©  cn  is  SD •*  CN  00 O, 2  CN  <n  e  60 CJin  I  O  o  -H  ir, on  o  CN  I  s  I© s 0\ OS ill  00  cn  cu  CN  CN  0> P  a  Z  w  o  CJ  os S 0O 0  oo -fl 3; ^ 00 p, 2 &aJ 00 in  H -1 O  t--  CN  in 00 CN  "* CN  CN  m m  90  w  I OS  c/i  00 m  >C3 CS3  W  z  P  SO  i OS 00 N ? OS 00  S O S O -1CN CN  tS  OS O, 2 fed  o  a  00  ON  O Os I S "*O CN "*  0 0s o . O  z  00  in  W  P  Si  0\\ CN V>  ©  so  tr,  c^  o  <n so -*  &  BS  SO  Z  W W  & H U  C 5S 90 i »—1 90 OS  »N  X  o  cn  H  31 Si  CN  CN  ©  H  w O  K  © Q  00 B J  cu  OS  S  o  o <s  Id'  © , — •© II H  CN  o  os Os O  CN  Its'  SO  «3 P  no  o =  I  z  Os © o OsS  oo CN CN  I  o I  N© OS i o rOs  OS cn SO • O  ON OS  Icn'l  cn  £1 £1 O NT cn |CN  SO OI  o  o  <! oo  u 3  OS 00 I CN 00 SO  IT. W J  pa  N:  00 CN  9\  2 ~ a  1  N? cn so ro o  CN  CN  Os OS CS  a  <  ©  Os Os  cn •-t  cn  90  >>  o in  D  =  I?  CJ  S I  00  e '2  S W  |_  c cC  O  cb "e| T3 Pi <u  13 O* 3 c3  cj b cj  in |  •a o CJftI  ,^ IPh  oq 'S  l« . a>  CJ  CD CDI * -«-» T3 C/3 C cS T3  e  00  _c cd "3 C S o loo  o ccj O  s cu cu O  W CN  o CJ CJ  S w o  as  •w  Ill on  w  ON  CN CN  60)  £  •  SO OIS 00 2 bu  OS  so •n  I  60 w ©  CCS CB  ©  so ^ O £ OS o Os H,  CN  i-  oo  60 to  B W  k so SO OS 00  as ca 00 o 60 O S w, C3 •&! OS a 00 o 2 601  H  X  u w  i  I  60 to C3 OS ^ OS O Os *-<.  wo z © w©  C3  H  o  o  o  o so  rn CN  PH  z H H  60  cu c W  IS -  Os SO SO  3 1 00 o  o  O 2s t-<. 6« 13 Si  n 3  Ckl  o o C3 UH 60|  P <L>  •8  a W  S o  s  a s  T3 +a  1 5 73 O  o  u3  73  p\  Lo  I  TABLE 5.15 FREIGHT TRANSPORT ENERGY AND ACTIVITY ANNUAL A V E R A G E RATES OF CHANGE 1990-96 1984-96 1984-89 Total energy (PJ): 2.38% 1.75% 1.57% 0.46% tonne-kilometers (000,000's) 2.31% 1.11%  TABLE 5.16 FREIGHT FRANSPORT MODE SHARES ANNUAL AVERA<GE RATES OF CHANGE 1984-96 Share 1984-89 1990-96 3.08% 3.89% 3.74% Trucks 4.00% -1.46% 1.88% Trucks Less than 4546 kg Trucks Between 4546 and 15000 kg -2.08% -0.77% -1.77% Trucks Greater than 15000 kg 4.04% 4.87% 5.13% -0.28% -1.65% Marine -2.06% Rail -0.85% -0.26% -0.28% r  T A B L E 5.17 FREIGHT T RANSPORT ENERGY INTENSITY ANNUAL A VERA*GE RATES OF CHANGE 1984-89 1990-96 1984-96 Trucks -1.39% -3.03% -2.80% Trucks Less than 4546 kg -3.52% -0.82% -2.02% Trucks Between 4546 and 15000 kg -1.27% -1.38% -1.32% Trucks Greater than 15000 kg 1.20% -1.78% -1.53% Marine 4.37% -1.23% 1.96% Rail 1.57% -3.57% -1.11%  231  3  CN  OS cn  V5  VO  Os  •4  OS os  0\ so I--' cn 00  OS VO  3  T,o  —  o  I  OS VO"  IT. as  o  ©' o in  Si  >  CN CN Os  •o e H  fHI X  Tf  o  as 90 l 90 as cn Os  90  so A  CN  ©  o  r-' Os 00  MO CN  I > ii  i  8 3  o o os' SO  VO  oo"  c» VO"  as • ©  o o  CN ii s-  s | i f  so X i ON St  cn  5  CN  00 oo'  IS* Tf o  oK  es  N?  a  T OS  ir. Os as  SO Tf  m  r-" VO  3  »s  N?  r-  °s  000 00  -e s?  • OS  SO SO VO r»' OS oo cn  > lis  OS  90  Tf  VO OS Os  OS V0 Tf  r»"  VO  cn  CM  S  S  CN  N? f Tf  o  o  o  CN  N? 0CN VO  N* 0*  m O  8  m  s  s  W  CN  •  IOS OS  00  0  o  ©  SSI o  N?  U5  Tf  90 Os  8 0 Os JS 00 ~  oo ©  Tf  VO A as -*-»  s i s s o *  §i  O cn  N; V0 cn  S  OS  CM SO as  NT  00 so CN V0  ^5?  s© J3  m Tf  o © ©  S O *  cN  as  rQ  1  Os CN  SO CN  s i s  90 Os  s  s  s i s S o *  90 as  0-  Nf  0-  OS VO"  so as •  cd os  a s £ *  as  as as  i5l  r-  SO OS  H  as  s  SO  H  Os 00  a  Tf  as as  as  B  CN  o  Tf  I OI  i o as  © o o  r-  8  OS  .5 . C3  o  <1> T3  C 03  o  —I  1  o  <L>  W  03  W  6 u  a o  60 w s ©o  I o  60 cutn ia 13  ©  -1-  os o, 2 5J  OS ©  P  cn  Z s o w H  < J  O  cs ca  I © © ©  O z  P >.  H c OS W W  o z w 2  o  60 w  W P >- is  0.  so m  oo a a; ^ ON .  u  H 05  > JS ca ca  00  H X  8 w Z u  60 w  so j:  <N  00  CN  -t-  51  00 o, 2 =3J  60 s os . O i © O OSs  "fa '  60  >  C3  OS CN  i/>  S ca  00  O S 00 • i -1-  so <* CN  O S o. S 6tf 00  3 « ^ P  o o  s  ca UH  D  c 0J  0 a  s •2"  1  kH  •o  1 ^ 38 •a o "Oca 3 _CJ  u  s loo  TABLE 5.22 PASSENGER TRANSPORT ENERGY AND ANNUAL A V E R A G E RATES OF CHANGE 1984-89 1990-96 Total energy (PJ): 1.57% 2.64% 4.01% 2.77% tonne-kilometers (000,000's)  ACTIVITY 1984-96 1.60% 3.12%  TABLE 5.23 PASSENGER TRANSPORT MODE SHARES ANNUAL AVERAGE RATES OF CHANGE 1984-89 1990-96 1984-96 Share -0.48% Light Vehicles 0.44% 0.01% 0.90% 0.35% 0.80% Small Cars -3.68% Large Cars -1.05% -2.21% 3.59% 2.84% Light Trucks 3.79% Bus -0.40% -3.97% -1.64% School Bus -0.84% -3.44% -2.34% Urban Transit 0.64% -4.77% -1.58% -1.44% Inter-City Bus -3.31% -1.10% Rail -2.18% -7.57% -8.39% Air 3.62% 1.41% 0.29%  TABLE 5.24 PASSENGER TRANSPORT ENERGY INTENSITY ANNUAL A V E R A G E RATES OF CHANGE 1984-89 1990-96 1984-96 Light Vehicles -1.37% -1.57% -1.41% Small Cars -0.26% -1.09% -0.80% Large Cars -2.23% -2.44% -2.49% Light Trucks -1.23% -1.09% -1.26% Bus 2.96% -0.54% -0.17% -0.42% School Bus 0.52% 0.08% 3.58% Urban Transit 0.38% 0.05% Inter-City Bus 1.79% -3.09% -1.16% Rail -2.19% -7.26% -3.82% Air -3.13% -1.17% -2.03%  234  8  <N  a  o  SO SO__ SO"  B H  CN  NO Os  I ,  8  "*  CN  I >  00  <o  NO •rt  SO"  ICN" oo" SO CN  00  O  o  00 OS  o  © © 00  ir,  °\  NO OS  II  as  o  i o  OS ON  ON  ON  ST  o SO  O O  c«l  OS 90  OS 00  N?  3  cn  z o  1 SO •* OS •n OsJ  <W  !/)  oo"  HH  s w  00 OS  S  SD  SO 00  cn  »' O I  Os" SO  o" cn  so  s s e  00  o •a  <3s  o  H OS  O  z  NO NO  O  O  o  N?  N?  O  SO  o  N?  CN  in o  00  o  CN  ON ON  00 Os  £1 O  N;  NO  CN  00  ON  I NO I ON  N?  N?  CN  ON  OS OS  -1-  00  o  w  13  r  i  oo"  CN  ON O  O  CN  w SO o © <  CN  CN  o  « z z z  NO ON ON  cn cn SS  so o o «N © O oC SO  CN  rN  ir.  OS u w  OS  00  <n 0 0 SO CN  s« Os  i  00 Os  Os  o  SS  oo"  IS  OS  II  in cn CN  a  Tl-  CN c » |Os" so"  H  <N  OS  oo  I •* 00 Os  ON ON  SO 00  o 13  00  o  Os OS  W  iri  W J  m  OS OS  gs cn W | O m  oo in  so o I-  o  in  ON  S  O  cn 2 < ~ O o cn « H cn Z  N?  SO  j  So  Ow  os ON 8P00i , m OO rO o «  I *  Os 22  SSI CN  90 Os  cn cn >  cn  cn cn  2  1  o"|  PH  cn Z  NO ON  \< cu  z  >  in  o is o 00 13 ON  -t->  w  H PS  o os o CN O s  O o SO os O cn 00 00 I I ON o O o ON Icn CN -4  m  TfI  0 - N?  c^l  so  HH  Os  m  •n |<N oo"  CN  OS  so o ©  O CN © SO CS  PH  ? CD  O  PH CD 5  =  1  ,  W  o o  "o  CD  Q  o  CN  55  CD co  CD  >  CS CD  co  , cS  co  , cS  o  I-a  O  CD  lo 13 o u  •c  •s  o CD W  IO H G C O _o  3 -l-»  *^-» CS  o w  6 o  W O C M  Calculated from Energy Use Growth Rates Calculated from Underlying Factors Activity Only Intensity Only 1 Structure Only  933.7  1995 Actual Energy Use (PJ)  1990-85 av'g growth rates 1214.5 1233.6 1251.4 960.9 897.1  1981-89 av'g growth rates 1137.3 1271.9 1661.7 1029.7 640.5  1981-95 av'g growth rates 990.7 1005.8 1097.1 949.7 841.0  1990-85 av'g growth rates 1019.2 1021.9 1029.4 939.9 921.1  1022.5 1131.5 957.3 821.6  1192.2 1514.9 997.7 685.1  1981-95 av'g growth rates 1115.6  Energy Use in 2010 (PJ) with...  1981-89 av'g growth rates 997.1  Energy Use in 2000 (PJ) with ...  TABLE 5.28 SERVICE SECTOR ENERGY USE EXTRAPO LATIONS  TABLE 5.29 SERVICE SECTOR ENERGY AND ACTIVITY ANNUAL AVERAGE RATES OF CHANGE 1981-95 1990-95 1981-89 1.19% Energy Use 1.32% 1.7.7% 3.28% 3.92% 1.97% Floor space(m ) 2  T A B L E 5.30 SERVICE BUILDING FLOOR AREA SHARES ANNUAL AVE!RAGE RATES OF CHANGE 1981-89 1990-95 1981-95 0.27% -1.20% -1.99% Schools Health -1.19% -0.07% -0.79% -2.13% -2.38% -1.38% Religious 1.23% 1.27% 1.17% Other Institutional 2.62% 1.41% 2.24% Office Retail 0.99% -0.94% 0.26% -0.10% 0.26% Hotel & Restaurant 0.52% Recreation 1.49% 1.35% 1.27% Warehouse -3.67% -2.70% -3.31%  T A B L E 5.31 SERVICE BUILDING ENERGY INTENSITY ANNUAL AVERAGE RATES OF CHANGE 1981-89 1990-95 1981-95 -3.82% -3.14% Schools -0.89% Health -3.43% -0.46% -2.70% -3.54% -0.33% -2.89% Religious -3.80% -0.51% Other Institutional -3.03% Office -2.63% -0.31% -2.12% Retail -1.46% 0.01% -1.28% -1.35% 0.22% -1.20% Hotel & Restaurant Recreation -3.00% -2.30% -0.28% -2.43% -0.05% Warehouse -1.92%  237  '1 b H ir.  o 00 —r I  B  ©  B H  IT; I Os  3 in  VO  VO o B —  -t-  H V OS iS  Q  o o o «s  -t-  60 w  ©  VO  60 in  ©  OS OS  OS OS  ,"3 •o > e  Z  © in  OS  ©  s  B H  o oo' OS cn  B H  o s 00 I  1 — 1  a  cn CN  B H  1/5 Z  o C/2  o VO  H >r,  o H  Z W  3 O  as o  H U w w u  s w  B H  os  © in  SO  p m'  '1 IH •  SO  60 w  m  OS so in  B H ir, Os i  m m vo  OS •a  a  IT)  SO  B H  00 B H  o o  00 I 0 0 OS  |©s  cn CN m  OS  so  00 cn  O Os •n OS  CN  Os r-  CI  Os  OS  o  Ed  OS Os  J  «  < o a o  S,  s W  o  o  CN  60  CJ > C3 13  I* 00 o OS o Os 60  o o\  o  ir,  in  00  ©\  C/3  B  h  1/5 I OS  os 00  O! u  oo  00 Os I  s  w IO I  OO  m  CN  TABLE 5.35 RESIDENTIAL END-USE ENERGY INTENSITY ANNUAL A V E R A G E RATES OF CHANGE 1981-89 1990-95 1981-95 Space Heating -1.43% -1.81% -1.56% Space Cooling 11.05% 4.97% 8.20% Water Heating 0.57% -0.46% 0.19% 1.78% -1.25% 0.47% Appliances Lighting 1.07% -1.21% 0.17%  T A B L E 5.36 RESIDENTIAL SECTOR ENERGY AND ACTIVITY ANNUAL A V E R A G E RATES OF CHANGE 1981-89 1990-95 1981-95 Energy Use 1.41% 0.76% 0.91% Households 2.07% 1.96% 1.98% Heating Degree Days Cooling Degree Days  1.15% 3.01%  T.30% 1.53%  239  0.38% 1.40%  ON CN  cn  ON  '1  O  B H  in [ 90  ON  ON  ON  90 CN 00  8  cn  00  IH ir. on i  ON  o  cu  3  ©  CN  ON ON  , ° N ON  > T3 B  B  B H ON 90 l  a  z  < o  B H ON I 90  TH  o o cn  OC © N  C/2  " p NO  o  CN 00  Z  B H ir, I  o  cn C/3 I—I  ON  I—I  s B H mI  NO  cn oo oo  ON  90  w©  ON  <  — T;  cn £, cn <£;  ON  to  00  90 ON  NO  s ,IH ^  O u cn P  ON  00  00  ON  cn  NO  >Ti  ON  II | «N  ON  I  ©  o  ON ON  a cn g  o  ON  I  CN  o  ON ON  > H Z W  00  cn P  NO  t--  B H O 9N 0 i 90  | m'  o CN  la* '3 5 id w W  CN CN  r-  o ©' 00 00  ON  90  9:  ON 9  oo"  •r,  ON ON  °i  cn l l W |3 J P9 «< H  «|  cn O " Z  -1-1 oo'  ON ON  CN"  CN  CN  CN ON IT,  ON  S3  O  |3'  B  B  O  O  W  O  cj  IN  w O  W  CN  o  CJ  w CN  TABLE 6.1 CANADA ENERGY USE WITH ADJUSTMENTS FOR CLIMATE, GEOGRAPHY & INDUSTRIAL STRUCTURE 1984 1990 1995 Canada Actual Residential + Services (R+S) 2,042.4 2,201.7 2,329.1 Energy Use (PJ) Canada R+S Energy Use with: OECD-12 H D D 1,559.2 1,656.5 1,681.6 EU-9 H D D 1,648.6 1,747.9 1,740.0 Scand-4 H D D 1,835.6 1,956.9 1,844.6 PacRim-3 H D D 1,291.0 1,382.2 1,506.6 US H D D 1,472.0 1,584.5 1,725.9 Canada Actual Transportation Energy Use (PJ) Canada Transport Energy Use with: OECD-12 Transport Activity EU-9 Transport Activity Scand-4 Transport Activity PacRim-3 Transport Activity US Transport Activity  1,676.9  1,839.3  1,985.8  951.3 654.0 1,069.1 1,384.9 1,709.6  1,098.0 767.3 1,283.9 1,589.7 1,948.8  1,146.6 790.8 1,225.5 1,637.9 1,993.4  Canada Actual Industrial Energy Use (PJ) Canada Industrial Energy Use with... OECD-12 Industrial Subsector EU-9 Industrial Subsector Scand-4 Industrial Subsector PacRim-3 Industrial Subsector  1,928.0  2,144.4  2,281.5  Shares Shares Shares Shares  1,877.3 2,130.0 1,739.2 1,710.7  1,992.5 2,210.8 2,141.8 1,854.4  2,103.1 2,362.2 2,321.0 1,949.3  US Industrial Subsector Shares  1,713.5  1,928.3  2,004.3  5,647.3  6,185.5  6,596.4  4,387.9 4,432.6 4,643.9 4,386.6 4,895.2  4,746.9 4,726.0 5,382.6 4,826.3 5,461.6  4,818.1 4,782.2 5,358.4 4,978.0 5,622.0  Canada Total Energy Use (Actual) (PJ) Total Energy Use With... OECD-12 H D D , Trans. Activity EU-9 H D D , Trans. Activity Scand-4 H D D , Trans. Activity PacRim-3 H D D , Trans. Activity US H D D , Trans. Activity  & & & & &  Shares Shares Shares Shares Shares  241  T A B L E 6.2 COMPARISON OF PER CAPITA ENERGY 1USE (GJ/PERSON) 1984 1990 1994 Actual Canada Actual Canada Actual Canada Adjusted Adjusted Adjusted All Sectors 221.06 219.72 222.57 Canada OECD-12 170.72 136.12 170.81 142.21 164.72 146.73 EU-9 99.74 170.05 163.49 102.65 172.46 104.14 Scand-4 180.68 137.88 193.68 144.64 183.19 140.01 PacRim-3 170.67 162.08 173.66 168.71 170.19 177.00 US 190.46 205.93 196.52 208.71 192.21 217.02 Residential and Service Sectors Canada 79.47 79.42 79.22 OECD-12 60.67 47.49 59.60 56.83 49.37 47.32 EU-9 64.14 40.11 62.89 39.66 58.82 40.61 Scand-4 71.42 51.35 70.41 53.17 63.07 49.11 PacRim-3 50.23 52.76 49.74 52.64 50.86 55.39 US 57.27 70.41 57.02 58.37 69.64 67.39 Transportation Canada 65.24 66.19 66.65 OECD-12 37.01 42.48 39.51 47.10 39.20 48.24 EU-9 25.45 22.56 27.61 26.97 27.03 25.50 Scand-4 41.60 27.81 46.20 30.89 41.90 29.16 PacRim-3 53.88 56.70 57.20 61.12 56.00 63.86 US 66.52 76.07 70.12 80.19 68.15 82.06 Industry Canada 75.01 77.16 74.99 OECD-12 73.04 46.15 71.69 47.79 68.69 49.11 EU-9 82.87 37.08 79.55 37.51 77.64 36.54 Scand-4 67.67 58.72 77.07 60.58 78.23 61.75 PacRim-3 66.56 52.62 66.72 54.94 63.34 57.75 US 66.67 59.44 69.38 61.14 65.69 65.32  242  RED CTIO  N?  £^  Canada' nmitmen  c« *; CN C N ON O CN  P  Z  o  o HH  U  </>  SE  atio  MIS  c  t cu es  P  § S CO  o  T3 Ph  in VO o CN 00 o CN  •a  w -M  *s  NTS OF 1990 Pe tual  u  s-  Ph  O  u o  -  AB  Ed H  Can  o  tm es C T3 CN N es CN  in C CN N CN  rin C CN N CN  CN Q  rm C CN N CN  CO I  T3 c C ccj so o O w PL, o w crj OV  To  oten  c© es o t-. L cu CN  VO ON VO  r00 o CN  cn CN cn CN  33.4  o o o o ON o o o o o o o o CN VO CN On t> in oo On in i—i CN 00 00 oV O m m CN CN Tl-  Icu  w  •o .« "a  £  i—i o  e S i ©^ U  CM  • mm  -tt:  Ph  _7 S3 >n C CN N CN  N? i N? ^ B .ts o ON N S 5s Tf O(NN C Ph E On o © E  CN i  N?  TJON  cn nd-  Ph  co CN cn VO 00 00 ON  "«  CD  pu  O  vo  isted by J imitment 34.53 95.81 33.07 59.26 94.10  CN  Ph  NH  Con  ON  Cal culat  TT o co O N o 00 in O N Tf CN vo VO CN O N CN r- m 00 O N -W: m CN 00 CN oo s m CU Tj- rIT) *H CN ^f' V O od Tf o TT VO — r1  ER CAPITA ENER YU! Compon neirgy Use A Can ad a Ad juste 70.81 70.05 93.68 73.66 96.52  <<  s  "es E 4.95  1PL IED  »  0 CN © Tf Tf* CN VO* 1  plie  W  w  CN CN CN CN C NC NC NC N ON ON ON ON O o O O CN CN CN CN  C  s CO o w o W on o PH 03 ON  N?  cn On  TABLE A.1 INDUSTRIAL SECTOR COMPARISON OF DECOMPOSITION RESULTS (BASE YEAR 1990) Manufacturing, 1995 Schipper et al (1997) Herbert (1998) % Change in Energy Use 9.39% 8.82% 10.35% 10.19% Activity Effect -1.64% -0.59% Structure Effect Intensity Effect -0.49% 1.38% -0.44% -0.55% Residuals Industry, 1996 NRCan (1997) Herbert (1998) % Change in Energy Use 11.79% 11.79% Activity Effect 6.54% 6.54% Structure Effect 3.81% 3.81% 1.36% Intensity Effect 1.35% 0.08% 0.08% Residuals TABLE A.2 INDUSTRIAL SUBSECTORS AND ASSOCIATED STANDARD INDUSTRIAL CLASSIFICATIONS (SICs) NRCan (1997) Schipper et al (1997) Mining (SIC 06 less 617, 07, 08, 09) Construction (SIC 4010-4490) Forestry (SIC 04, 05) Paper and Sawmills (SIC 217, 2512) Paper & Allied Products (SIC 26) Iron and Steel (SIC 291) Ferrous Metals (SIC 331,2) Cement (SIC 352) Stone, Clay & Glass (SIC 32) Chemicals (SIC 371, 3721) Chemicals & Allied Products (SIC 28) Petroleum Refining (SIC 361, 369) Smelting and Refining (SIC 295) Nonferrous Metals (SIC 333,5) Other Manufacturing (SIC 101-399 less 2512, Other Manufacturing (SIC not given) 271, 291, 295, 352, 361, 369, 371 and 3721) Food & Kindred Products (SIC 20)  TABLE A.3 INDUSTRIAL SECTOR COMPARISON OF DECOMPOSITION RESULTS (BASE Y E A R 1984) Industry, 1994 NRCan (1996) Herbert (1998) % Change in Energy Use 16.45% 16.15% Activity Effect 20.60% 19.15% Structure Effect -2.26% -0.19% Intensity Effect -1.23% -2.13% -0.67% Residuals -0.68%  244  JD  JO  r-  os Os  N;  Os  c  CN  NT o s CN  O  ©  O S  t--  oS  r-  O cn  so OS OS  Os OS os  *<r "sr  o  x  CN  Os O  o  os  os ©  CN  O  cn  SO OS OS  00  00  OS  os  O S  OS  OS Os  s °  r-  O S  os SO SO t-- cn cn os  CD  Os  OS SO  SP  N ?  O S  O S  o  SO  CN  os  '-a  r-  Os OS  Os Os o  cS  •s  x  O S  SO  u.  SO  CD  SO  ~s? o s  "sr  O S  cn ros cn CN  ~s?  "sr  SO  O S  O S  00  &  OH  1-3 o  1-3 o  a.  in oo  O S  SO  CN CN  N?  N? O S  oS-  CN  CN CN  OS Os  CN CN ©  s ?  r- p00 in c ^r  ©  0  n  m  ^r  c«  CN  1  I GO. Os OS OS OS  oo os SO in CN CN  OS os OS OS CN  O CN  c  s °  00 in  os OS CN CN  SO  ^r  CN  O  oo Os os  CN  03  lo  oS-  CD  00  oo os os  s ? O S  oo in  00  oS  SO  CN CN  os NT  O S  SO OS  OS  00 m  o  CD  glinso  N? os  00 m 00  O S  SO  CN  in  >n  "sr O S  CN  CN  CN CN  OS  00  ^r  CD  <D  OS OS w CD  CD  I  3  o  tu  w  00  c CCS  U s « o s  00  a  CO  c  CD  ccj JZ\  U ~<  CD  >N  i3 ts +a-»  co  \~ CD  .S  w  CD  t3  00  c IW e a  o  | £ _C  CD  n  o  ccS  3  CD  '55 PH  w, opl  'Si i  CD  <-> i  s ? C|S  o  w 3  CC  U  US  T3  i<:  ON ON  c  o Tf  x  OO  r~  CN  ON  N: O  ON ON  N;  CN  00  o vd  in  00  c  Nr 00  t t cn  o  ON  crj  VO ON  NT  O OO  N? CN T  m  VO  VO  vd  00  v© Ov ON  00  00  ON ON  ON  r-  ON  o  x  ON ON  N*  in 00 m VO 00  CN CN  VO  o  ON  CD  o  m >n  , CD  u w  r-  <L> l_  O CO  PH  o  ON CO  ON  00 00 r-»' o  CN CN O  C--  00 00 GO  o  d,  Q o H tf MH O >M H  ON ON cd  PH  C/J  «  CD  r-  ON ON  O N  CN  co  O  m in  O  C  OH OH  N :  ON VO  m  o  ccj  N ° O x  ON CO VO  VO TT CN  IO  03  ON ON  tON ON  C  N :  O CO  00  o  N? o  N?  x  VO  03  CN O VO  ON ON  tON--  NT  ON  00 O  CN VO  o  00  vb  ON ON  IO  t:  00  i-e  00  ON ON  CN CO  m m  ON ON ON  CN CO  CN  00  o  CD  in m  N :  CO  in  VO  vd  o  CD  00  c  ccj J3  o  N? o x  CD  s CO  CD  cd  l | co  o  VO  x  o  TT ON ON  _C  o  cS W  o  c  X} (J N?  UP  oc 1CD  e e o  O  ccj  W ccj If cu ]>  cd  CD  00  <L» co >s  l-H o leg W  x  in 00  CO  001  \&  o  CL> NT  O CO  r-'  CD  001  CO  s C/3  C  CD  3  CO  'c«  Ltf_  co ,  X3  o  l l  leg  w  CO  3  -C3  CD l_ CD  2  in  3  TABLE A.10 COMPARISON OF SERVICE SECTOR DECOMPOSITION RESULTS, 1995 (BASE YEAR 1990) Herbert (1998) NRCan (1997b) Schipper et al (1997) 9.16% % Change in Energy Use 9.15% 8.97% Activity Effect 10.26% 10.26% 8.31% Structure Effect  0.39%  0.39%  0.00%  Intensity Effect  -1.34% -0.14%  -1.34% -0.16%  0.62% 0.05%  Weather-Adj usted Intensity Effect Weather Effect  -5.14%  -2.65%  4.01%  Weather Residuals  -0.35%  1.35% -0.19%  Residuals  T A B L E A.11 COMPARISON OF SERVICE SECTOR DECOMPOSITION RESULTS, 1995 (BASE Y E A R 1990) AT SECTORAL L E V E L OF AGGREGATION Herbert (1998) Schipper et al (1997) % Change in Energy Use 9.16% 8.97% Activity Effect 10.26% 8.31% Structure Effect 0.00% 0.00% Intensity Effect -0.99% 0.62% Residuals  -0.10%  0.05%  TABLE A.12 COMPARISON OF SERVICE SECTOR DECOMPOSITION RESULTS (BASE YEAR 1984) Service Sector, 1994 Herbert (1998) NRCan (1997b) % Change in Energy Use 10.82% 13.95% Activity Effect 40.87% 41.40% Weather Effect - 0.17% 0.04% Structure Effect 0.89% 0.93% Weather-Adj usted -22.20% -20.37% Intensity Effect Weather Residuals - 8.58% - 8.05%  247  248  249  ssiiuojeSaul o 0  <  o  0  o 0  >  o  n  o O  «  o  O  o  O  SDUUOJEoOlU  w  o  -  o )  0  0  u  s  OD  s?  251  0  s  USA  Luxembourg  ; Canada  Australia Germany Belgium Finland Denmark Netherlands I France Sweden Iceland Ireland Japan Austria I Norway Italy Switzerland New Zealand Spain Greece Mexico J Portugal [Turkey 10  Figure 5: Per Capita C02 Emissions from Energy Use, 1980  tonnes/capita  15  20  25  35  30  Luxembourg  USA Australia Canada  Denmark Belgium Netherlands Germany Finland UK Ireland Japan New Zealand Iceland Norway Italy Greece Austria | France Switzerland Sweden Spain  Figure 6: Per Capita C02 Emissions from Energy Use, 1993  Portugal Mexico Turkey  tonnes/capita  10  15  252  20  25  30  35  Luxembourg ! Iceland inland USA Norway Sweden Netherlands Belgium Ireland Australia Denmark Mexico UK New Zealand Austria  Figure 8: Final Energy Consumption per Unit of GDP - 1980  >ranee Turkey Japan Switzerland Greece Spain Italy Portugal  0.1  0.2  TOE/S1000 US 0.3  0.4  0.5  0.6  0.7  Luxembourg Iceland  Canada Finland Norway Sweden USA J Mexico |Netherlands New Zealand Australia Belgium Ireland Germany UK Denmark Greece Austria Turkey ranee Switzerland Portugal  Figure 9: Final Energy Consumption per Unit GDP - 1993  TOE/S1000 US 0.1  0.2  0.3  254  0.4  0.5  6SZ  CN CN  >^ tf tf  9  s o H  es  u  PH  / ~  > ^  OX)  c  • mm  as OH  P  •43 — <u S  to  §  3  j  s  g to  ta  5j B£d o  es  e CU  U  cu  e U  cu  -1  en  s  o  ee  MH -  c en es eS  "°  £  ^. £ CN O a 2  00 "O  MH MH'  CN c  CD CD  CD  tf  X>  3  CD  »-H  CD CO CCJ PH  ^  a 3  PH  C cu  s  3  CD  s  cn cn  CD PH  CD  tf  OS OS CO  %u PH  cn  PH  « " > \-H  CN £  CU  "H*  td  PH  — cu  HS  C/3  £ o  -1  o U  ee  1^ coo  - Oo  -u s1 »  s «  U m  <*$  cu  500  «I  VO CO  6  Ml  MH  h-H  5  en es  PH  U  CO  to  s  O  Os  CN  HH  s o.2ore  s*  e  O  K>  n 2. 3 5"  •o a CB  ^3  ha  •3a as  f B O  pr  *^4» a—  sO  3  *  so a  260  261  1—4  M  1= r  r. ^  c?  O  P  i  a  GO  3 2 CD o CD 5C o  5. «  O- GO  CD 3  o  s 5  z  0  0Q  »  i  •4 £  sr o o  c  -  1 2. O P ^ P 'is *  *0  £  o < CD CD sr H.d5 O CD o  o  1 c CO  CD o o CD  c  CO  is  CO  4  eg  i  *a 2.  C9  m  c •6' 3 CD 3  K>  3 * hb oo 2  r K  t » I-*  Q 3*  CD CD co Q3  3  3 S. »•  £&•  a§  I  262  263  264  267  269  2 9s  9s  oc 9s  OS Os  3C  R OS  Os  SO 9v  r9s  (59 s  9>  270 £  ON  O O  ON  00  o^  ON ON  R as  fr-> 8  Nfl r-  ON  IT,  r-  ON  ON  271  53  oo ON  oo  ON *H l  00 OS  m ON  OS  O  OS  CM  «s  '43  U  CM  cY> cu  ON  •#•1  o«  OD ON  os sc ON.  OD OS  ON  ON SC CN  272  CN  O  CN  oc o^  9  -O  o  00 3  PJ oc  oo a a o  5 •a 5 o u 9 CQ § ca fcs  .5 , 3 .9 s o I£ gz  HIM 00  00  00 as  r-  ON I ON  00 os  «S  01) C  § 5 + eg WD 5  OS  SO o\  c  rOs  ro\  ox  OS  eg  sC3 CU  OD  «  Xr  CA  5  C/5  •e n  ss 1  PsS  cu  O  OD  2  fa S  s o  cu cu  S c3 .S oo  o  I  I  f £Sf J O  w  \ C o  3 ±3  t> 4S  « i>  < 55 5 oi  1111 o  s?  sr  274 7  o  s?  277  I  ON ON  01)  SO  cct  «» *§ c - ^  c  CB  s ao • S3  g «5  c c § U  -—  CO —i  lU  CU  O  3  T3  T3 ao  ON  g 8  T3  o I  a  u u PHu  i  5  43  5  Os ON  ON  o  NO ON ON  ON ON  00  ON  o»  ON  00 00 ON  r00 ON  11 S3  NO  00 ON  c  IT)  Tl 00 ON  s-  23 OX  4- * 278  s  279  © ©  ©  © ©  -r  © ©  © © ©  CN •—  © © © ©  ©  ©  1  © ©  >—  ©  © © ©  ©  ©  ©  © CN  © ©  ©  ©  —  OA  CU  S  2  ° 4  3 oo ON  I CU  a fc = © Q 5©  CU GA  c  cu  S3  es O  IN  H  cu u  fc 00  CO  OD «  cu ©  U fc CA  © © a  c C5  fc  © © o © ©  © © © © ©  281  © © © © ©  © © © © ©  ©  o  N© CN ON  SO O  o 0  >/->  M  1  "«  es  ao  NO lO  ^  s  s  °  1s  — g SO rs s C3  $  -J CA  3  H  ON  00  |  to  SO CB  5  > 2 u < a> sPQ O 3 % as 3  3  H  H  2  H H H ON  ON  ON ON  © ON ON  I»  I  00  ON  ac  ON  ON  o .s C  S3  H  £  00 00 ON  00  ON  £ >>  o X  W> gf U,  ON  S3  ON oc ON  ox  ON  CN)  282  ON ON 00 ON  ft  08  ox £  g  «  08  ip OX  • P")  98 CS  §w IN OX  • l—l  fc  283  r i  o  o  NO  o  o  I ON  ON ON.  ON ON  ON  00 Ov  m ON ON  V O T f  S3 03  >  ^ ON ON  H  ON.  ON  p ^ > ox  ON OC ON  oc  gt S3  00  ON.  30 ON  to  *» 33  OX  IT,  00  ON  fa  x ON  o o  o o o  o  NO  CN  284  o © o  CNI  ©  o  o os  -T Os  ID Os OS  S esONON  SO  i OS Ov  00 C  CU  ON ^ 0. s  ON  ©  cu  JS  cu  < a  s  C3  IN  CU  H  OS Os  OS GO OS  CA 30  ox  • pN  Ss  • »  cu DO u cu fc C  <s fc sC 00  CU -  on  • PN  00 OS  fc  00  0> © © © SO  ©  SO  285  © © ©  so  © © ©  © © ©  CN  m  o  CN  O CN  O  f>  <—<  CN  o oo  286  o VO  a rj-  °  w CN  \  S  o o  ©  ©  ©  ©  ©  o r-i  ON  ON ON  NO ON ON  ON ON  rn ON ON  00  ON  o 08  - a  <  CJ  ON  ON  4>  ON  so  SM  fc  "S  ON  DO JM  o  30 ON  d fc  r-  c*5  ON  s  NO  00 ON  OD fc  IT, ON  ON  © © © ©  © © ©  o ©  287  © © © o  © © © © ©  © 5  ID  o  IT)  -+-  as as  VJ  ON  ON ON  VO Os 0\ • 1 #  ON ON  co  00  as CN  o  ON ON  ao ON ON  &  Tr  03  OX  S3 03 CA  CU CU  ON  a  OC ON  c s o cu CA CA  « •2 i  oc 00 ON  CU  -  so  OX  •  C U CA  cu  00 ON  CA  03  o  Fig  NO 00 ON  oc ON  00  o o o o o  o o  o o  o o © o o m  289  o o  o p  ON  3:  00 O  ON ON  • • O OH OA  CS i-  H  5CU  5A  OD es CS  cu OA  PH  ©  *i  en  cs ox U OX  E 05  M  u S os  290  ON  in ON ON  ON  ON  so  8  ON  ON  ON  00  ON fH  ON ON  CU  ON ON  s1  S  ON ON  *- s  cu cu OX) - 4 -  e  CU  ON SO  c  ON  HH 0©  fc  SC  &  ON  •  r SO  cu  ON  u 3  OX  N*  SC ON  00 ON  so ON  291  u  03 cu .  b O  CU  OA 03  13 «  CU CU ON CU  00  cu -tr  CZ3  3  as  ON «U  ^  fa  a O  00  O,  £ ©  cu CU  W1  292  M  PH ID Q I  00  CU CU »PP PP  "  c O  cmu  cu cu  PP  ©  -T  cu o *J3 cu 3  C/5  g S  «  > ©  <V5  S PH 3  CU  cu  • pp u  CU PP  3  OX  cu C3  cu  J--  • PN  MH  ^ PN © ©  293  00  O N *—<  ON ON  m  11!  © CU  cu b  CU 93  cu j3  cu CU <«  < ©  cu O s3  OX  294  *a  ON. ON ^ *? *•« oo  £  ON  § 8 es "g « c O CD OD 3M  S 5 C/5  <3-  ©  q d  ©  295  to o  O  o  OS  Os  ON  os  IT)  ON  o>  ON ON  00 08 Os QJ  ss  55*  MH  PH  O  CU  -<M»  ON OO  ON  00 00  CU  ON  cu  CZ)  «l  cu cu  00 ON  cu  CU  fi  MH  cu  '/-, 00  ON  IN C3  J3 m C Z) CU  S3 00  00 ON  00 ON OO  o  oo  c  d  so ro  T  rv)  o  o  o  oo  SO  so  NO  o  o  so  o CO  296  ON  ON  es oo  H  ON  <3  OX ©  .S fa 2  ."5  mm  O  PQ cu cu •#•< cu SS O  OX  >  cu  "S cu a  en  cu '»  GO  o  ox  o  o  O  o  e  d  297  o d  u-l p O  in ON  CN  o-. ON  in ON ON  I  00 ON  »H  "O  o  SM O  o  ON  CC  =N  CC CC  cu  S3 cu  CC ON  55 C  ^ ^  NH  £  U  ™  cu  =  ^  cu  °  es  I ON  o© ON  -r x  ON  cu  =  OD • PM co ON  fa  00 ON  298  ON ON  ON  ON  IT)  ON ON  00  ON  ON  S3 «U  ON  ^  ?M  - o ON  oc  OX fa  .9  ON  u  *a © P-5  3  <U  pq  CU CU  OS  ON  NO oc  so £ .3  ON  cu  &  IT, OC ON  93  a *  ox^  T  00 ON  •p»«  fa  2 ON  O l  00 ON.  00 CI  in  CN  299  >/-> o  ON  300  m ON ON  ON ON  ON ON  ON ON  ON ON  ON  oo es ON ^ OA  +£  S3 53  ON  ON ON  .  is ©  ON  oo ON  im  33  CA  o  3C  00  cu  ON  CU CU  S3  GO cu cu  30 ON  P>  cu cu CA t3 GO  NO  00  ON  x.s cu  00 ON  oo £ rn es ff$ J3  00  cu *-  ON  33 OX  8 ON  00  ON  00 ON  301  ON  ON ON  ID ON ON  ON ON  00 ON  ON ON  •> CU  o ON ON  2  •o  o  ON  is  00 00  00 ON  ON  93 CU CZ) I CU  13  00 ON  S3 •>  ° C  L cu  NO  00  ® pp pfip CS CZJ  ON  cu cu cu , p N  00  ON  00 ON.  fD CU PP  =1  00  •pp DJ)  ON  MH  00 ON  00  o  NO  o  o d  302  en o  ON  CS  o  © d  %s *• <  S b 2  ©  CU *~ c« O 3  is ® 93  CU  rn . 5 cu  5-  Q> S-  3 93 Of JS  so d  2  d  oo o  ©  303  so O  o  O  ON  ON ON  s ON  ON. ON  ON ON  ON ON  if) 00 00 ON  00 ON  ^ .2 C 2 e -a •4^  cu cu  CC  NO 00 ON  3  ^  00  ^  ON  cu  -a  cu  s  00  CC  0>  ON  ^  OX  ON  00 ON  cu u cu 3  ox MH  00 ON  3  a  304  5M  CSS  CU >H  Se  o +-* PJ  CU SA  PQ TH OO ON  CU TS CA(  CU  tf  CA  CU  fN  •  ur  cu  ox  fa  CA  tf S3 O • PP  CA  ©  S © cu cu  Q  ON  OS  OS  Os  OS OS  Os OS  OS OS  ON GC OS  SC GC OS  sc ON  so GC OS  if; GC  ON  -flGC ON  oc ON  OC ON  GC ON  306  M M  33UFi|dtlF  qjjw spioiiasnoq jo anSiua.u.xI  312  -p o cu cu  3 T3  c TT  O  *  .2 2 g cu  -iON  •  go W  E  o  ON ON —*  ON ON — . 1  II  -I  c  o  o  CO  SA  3 O  •8 s  © o o  ©  o c o  © o ©  <N  313  © © © ©  © © o  </"j  sCU >H  2 ©> « in  T f  CU  00  00  C \  «  *w  ca ptS •o s  S  ^  51 2 2 WD S E  S  o o cu  316  mm  C3  CU  > cu  ON  PQ  7 © T f  00  ON  ss  ON ON ^  3  C3 OS  U ca oo T f  cu  ox  fa  a  -IS  o S §  cu  o  317  318  319  320  321  cu  Ia 00  1-  on  s  r- .2 ^  i  s S  (2  b£ o  MH  CU  "SI  32 U  326  327  sc ON —H  •  o  ON ON  l—1  11  <r, ON ON i-H  5 fa  328  SO  **•  I  •* 1  <N  O  00  SO  •*  •*  •*  m  cn  m  1  1  1  L.  1  333  334  u  SO 0>  53 cu  1  cu  00 ON  C3 OH  : o >N  ON  O  TH  2  «>  ca  2 WD cu J-  _ so  2 ^  i* 3  I I I  g  w g >  u  j-  <C  C/D  i5  s g § ^ 1  . -  o a,  I I I ! I  s©  1  •a  I  cu cu fa  WD  ^  ^  fa 33 U  NT  c  335  o S •* i N  NO ON ON  «  >H  •  cu  cu  oo « CQ  ON  ~  i-» SO  CU  s  ON ON  wH ^  sa ±s  2 * <M  PS  ox o g — £  as 6/3  o  T f  o  cu cu s- cu  E O  336  o ©  <rpi  •  o  n  Tjr !  o  o :  v  **  -  o  i  in  o  o  v  o  -  i  ISJ  337  o o  -p*  v  o  -  i  *-*  o  o  »-«  w  o  i  o  338  ON ON  o  a  2  c*j S3  S3 .©  ?3  "c*3  H  §  £ fa  2P < * > S3 93 a a  o ON ON  fa = o 2 — :  S3 (3D  O  30  ON  o © © ©  o o © ©  © © © ©  339  © © © ©  © © © o  344  o  9 <c m  5 S  W  1O S £ Bi  I I I  oi  345  I  fN  346  347  oc -1-  -r  ND  -I-  -t-t  -1-  CN  OA  c  t  o  IT, ON ON  *OA OA • II  s  u  OA  a  OA  3  o  e cu cu  -  ©  ON ON  o  cu cu CC cu cu cu CC TIP  -  CU  3 S5JD  90 ON  n i  « l  <E , UJ ^ >> •5 & & S m  2 3  u-  1  s l  m  5  >; §  <  cw  C3  1>  :>  cd  u  4;  -  ^ Q4  O  T  356  5A  S  CN OS  m B  •  fa  a CA  cu  ^ ^ s  QJ  cu -  a is  Oxi  -a  c  •  3  g  z  H M  Dio  Si  Mel  3 O S  .8  Car  CA  o 0\  •  ON  a  cu CA  CU tf  00  -  CU  3  -r 30  ON  01)  1  •1  o o  o es ©  o o  •n CM  es cs o CM  357  es o ©  o o  IT)  u CU  2?  > cu  0 u  5 IT  CU O C/3 ON  1a  E  CJ  UJ  eg  i) so c  ^  I  co  ur  u  -  s  a  o  <  11111 u.  >a  8 cd  •§ "3 (ft  cu  (2 * io .2  ^ -s cu S3 OX  O  ^  « ©  cu  o U fs  o CN  C  a o  •s  53  cu  •s {a PJ eo  CU  9 53  w  <  o  LU -=  CJ  cu CO " cu o  9 PH  I I I I I  CZ) ON P-N ON  55 CU  1—I S  3  T5 • PM CA  Qft  •  -P  3? </> .2 ^ *S5 cu o  S ft OD S  fa  cu cu  p  X  u  o  O  in  o cs  OS  u m U  CU  to  •Z CO cu " cu o  CZ) P-I  93  cu -O  ON ON H  w  5  3  P-I  t/j  CM  til  tf  5 tf «r> .a  ^ -a cu  o  OD  S  fa  CU  o cu  o  O *3-  6 •  o  %63  o  c-  o  O  364  e w5  Oil  Natu ral  •  60 Coal  Woo  Eleci  •c  Propane  C3  II  •  •  a.  •5  o HP*  cu CU  1 fc CU  .1  m cu  SO 1/5  ^  •  M C o  pQ  H  CU  •ai  ^ !S£ S-  cu  a  cu UH  SJC  3  © ©  WD  u o  a  ;-  :  *  •  ... - :  •  i  tee Hea  MM  cl i  o s  ON  O  o  oc  i  o o  t--  i  i  i  © ©  o  © -1-  365  i  ©  i  i  C5 ©  c ©  T  ©  to  s3uuoi uoinim  366  O o  ©  367  ra CN  c  o  ON ON  •r, ON ON —1  • •  o o ©  O  o  •  fa  "R  O m  3  ON  O i-  d CN  -•^  C3  -o  53  3  u  ID  cu -  3  o ra i-  CN  fa  oo  368  369  370  o o o •  /  o o o  "  )  o o o  O  f  o o o  >  O  o o o  W  o o o >  371  .  o o o  0  «  n  o o o  O  w  o o o "  i  372  373  OS  -t-  C  00  JO o  NO CN  *—*  '—1  •  m  CN ON  fa CU  Q o cx  CN  o o  o 05  3  O NO ON  ©  3M  o  cu cu  is 3 T3 3  ON ID  GJ ra  CU  CN  s-  00  3 WD  374  375  NO  ON 1  o  SN  ON  00 I -floe  376  377  >  378  GJ <N  CJ ©  o  CJ  00  \0  379  o  •*  o  CS  380  381  9  o  382  383  384  385  r  5 NC  o © o  o o o  386  o o o  o o o  CM  © a ©  387  01  c  t  o m OA  1  fa cu  s o  NO ON 00  o  NO ON ON  ON ON  ' 1  © © ©  CNl  ©  i-H  ©  II  •  s o  u  o ra  u  NO ON  u  ©  ©  ON  s OA  03  H SM  cu  c cu OA OA  C3  fa  2 ON 00  ID  cu ~  oc  3  ©  © oo  ©  NO  388  © -T  ©  NO ON  on  S. *6ft CM  fa  -I00  ©  NO ON ON  ON ON  -H  •  1!  © © ©  cs|  ©  i—i  ©  CM  c C3 cu  o cu  NC ON  Oft  2>  s  C8 -  H  sQJ  bJD  c cu 5ft 5ft  S3  fa IT! i-  cu -  ON OC I N*  S  00  © © © ©  ON  © © © ©  OC  © © © ©  © © © ©  NO  © © © o  389  © © © ©  -I-  o © s o -t-  © ©  a  o o o o en  © © o  O O  o o  IT) rN  fN  390  © © O «/-.  o o © o  © © o Wl  391  399  C3  C m  o CA  s  • mm  o  as as  W •  in Os OS —<  II  o Q  O  <N  CN  O  o  CU  O  c  o u o  ON  o OS  t2 CU  CZ) cu cu  it cu  CZ)  ON fS QJ  C3  mm  OS  a  OC  o  o  SO  o  o -i-  394  o  CI  o CN  395  396  397  o  rM  O o  ©  © oc  NO  399  Q  -t  © CN  400  ©  CN CN  HH  •  i n CN CN <  ©  ©  ©  rH  CN)  CS  ©  ©  •  401  m T5cu cu  • PP  © H  >  cu  -J  ON Os  °s 2  PH  U i,,  cu bD  O  'n O  OS  g  C3  * u — S3  cu  to  33 p3 % cu tf fa  00  cu  c  s  o  Os  cu PP  3  OA  O 00  OS  r-  oo  402  o  >r->  -1-  403  a  O  405  C W  CU C3  CZ) 2-  CO 73  09 00 =3  I  cu cu  C W  cw  j3  CN Q  o CU  CO  -£  T  S3 :3  3  cu  *  as S  fa  CO  fa C3  2  -5  X3 &  .ts *3 7 3  C3 fi  fO  (J  N O  cw  s £ cu  53  Ml  S3  >  cu C3  •a C3 C S3  u  C C  o r".  fd  407  o o ©  0)  1-1  CO  ca o, 4=  ON  i P  409  ON  CA  S3 CA  ID ON NH P H  00  ON «H  cy W  S3 °  2  m O  ^ a  °s  eu eu cu cu «M P  cu  ON ON  ^  03  CU  cu  1e fc cs *a CA  g  a eg  <  w  S3  cu o S-  03  a CJ  CJ  CJ  CJ  pj  M  IH  CA  .2f * S o  C3  3  I  tj aj  a  5  a  H "2  I i 30 ON ON  30 ON ON  U  ffi  S  55  trt  r-  i^-  r--  ON  ON  ON ON  ON ON  CS  03  <3  I cj  C/3  U  g  cj  C/3  s C/3  )U33J3(I  410  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0089005/manifest

Comment

Related Items