Trans-boundary Carbon Emission Footprints and A Framework for Sustainable Cities
Dr. Anu RamaswamiProfessor of Civil Engineering and Director
IGERT Program on Sustainable Urban Infrastructure Center for Sustainable Infrastructure Systems
Sustainability Outcomes
Eg: Low Carbon, Resilient and Healthy Cities
Energy & Buildings
Transport Construction Materials
Water/ Waste
Policy Actors
Firms & Businesses
Citizens & Community
UC
D C
ente
r for
Sus
tain
able
In
frast
ruct
ure
Syst
ems
Food
Trans-boundary Issues in Measuring Urban Environmental Sustainability:- City-scale Carbon Emission Footprints - City-scale Water Footprints (ongoing)
Greenhouse Gas (GHG) Inventory
• The first step in undertaking climate action planning at the city-scale is to conduct a baseline greenhouse gas (GHG) inventory for the city in a base year
• What are Greenhouse Gases (GHGs?)– CO2– CH4– N2O– Plus three CFC replacements– Aggregated in terms of global warming potential
as CO2e (carbon dioxide equivalent)
City or
Urban Region w/ Buildings, Vehicles, Industries (Scope 1)
The Urban Trans-Boundary Challenge
Commuter
Airline Travel
Freight & Goods
Key Urban Flows
-Food
- Water
- Energy Electricity
….(Scope 2)Transp Fuel
- ShelterCement
…
What to Include in a City’s GHG Account?
• GHG Accounting at the city scale is confounded by the small spatial scale of cities, due to which:– Trans-boundary commuter travel and airline travel are
artificially truncated at the city boundary– Essential infrastructures transcend city boundaries
• so electricity production GHG are emitted outside the boundary of the city that uses that electricity
– Beyond infrastructures, there is also trade of goods and services across cities
• Larger-scale National Inventories: More inclusive of all human activities (agriculture, fuel refining, electricity production, etc.)
Emerging Issue: Not All CO2 is the Same
• Although a global climate forcer, we may want to know spatially WHERE CO2 is emitted.
• CO2 “domes” over cities affect atmospheric mixing that increases concentrations of short-lived climate forcers - SLCF (e.g., ozone) locally
Illustration OnlyJacobson et al., Environ. Sci. Tech., 2010
Methodological Approaches: Community-Wide GHG Accounting
Geographic-based, Boundary-Limited• In-Boundary direct GHG emission only
(Scope 1) – good for national inventories• GHG from trans-boundary purchased
Electricity (Scope 2) – reqd. for corporations • Other Trans-boundary?(Scope 3, optional)• Relates to measures of productivity
Consumption-Based Using EIO-LCA• Household Consumption $• Government Expenditures• Business Capital Investment
Geographic-Plus Trans-boundary Infrastructure Supply Chain Footprint for Cities
• Identifies Scope 3 GHG contributions important in all cities: transboundary travel life cycle impacts of producing key urban materials relating to water, fuel, shelter, food (infrastructures)
• Benchmarking with National data
Scope 1+2 underestimates GHG impact of cities
Data challenges in down-scaling to cities Test convergence
EPA Climate Leaders & WRI Protocols for Corporate GHG Reporting
• Scope 1 + Scope 2 are required reporting
• EPA and WRI recommend reporting on a few Scope 3 emissions – Optional– But creates win-win climate actions
• Full Scope 1-2-3 Accounting yields an expanded inventory that becomes a “Carbon Footprint”
• But we needed insight/research on what Scope 3 items are most important for cities
City-Scale GHG Inventory Methods –Developments from 2006-2011
• Only about 10 cities completed inventories in 2006
• No standardization on what to include/exclude:– Airports – included by Seattle and Aspen– Transport Fuel Wells-to-Pump (production) emissions
ignored by all – Upstream energy use in producing other key urban
materials – ignored by most cities (although asphalt included in some)
• Focus on “direct end-use of energy” within a city’s polygon area - tailpipe transport emissions, emissions from electricity use in buildings and credit for material recycling.
Denver, CO – Location and Goals
Denver seeks to reduce its annual per capita Greenhouse Gas (GHG) emissions by 10% from 1990 baseline levels, by 2012.
Denver, within the Denver Regional County of Governments (DRCOG) Region
Denver’s GHG Inventory Method –Geographic Plus
• Spatially-allocates travel-related GHG across cities
• Key Material Flows: Views the city as a demand-center for energy, transport and key materials– Functionality of cities led to choice of key urban material
flows in cities :water, food, fuel and shelter (concrete).– Supported by major home-scale expenditures and MSA-
level material flows
• Applies Material flow Analysis (MFA) and Life Cycle Assessment (LCA) to quantify upstream indirect GHG emissions of key urban materials– Analogous to electricity indirect GHG
Transportation Energy Use: Separating Trips to and from Denver
Denver's Resident
Workforce41%Workers
Commuting into Denver59%
Demand Centered Approach - DRCOG regional transportation model was analyzed to isolate trips to and from Denver, and, to ignore pass-through trips. [Dr. Bruce Janson]
Transportation Energy Use: Airport Fuel Use Allocated to Denver• Trips from Denver to the
Airport were isolated from all trips to the Airport to allocate airline fuel use to Denver – Road Trip Ratio = 0.22
• Tracked well with population ratio of Denver versus DRCOG region– Denver/DRCOG
Population Ratio = 0.22
Road Trip Ratio To Airport vs. Population Ratios for 25 U.S. Cities
y = 0.99x
R2 = 0.96
0.00
0.10
0.20
0.30
0.40
0.50
0.60
0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70
Regional Population Ratio
Airp
ort T
rip R
atio Portland
Austin
Denver
Hillman, Janson, & Ramaswami, ASCE J. Transportation Engineering, 2010
SeattleMinneapolis
Hybrid LCA of Key Urban Materials
• Department of Energy’s GREET model was used for upstream wells-to-pump GHG emissions associated with fuel processing for all transport fuels
• Regional cement production and transport data was used to model cement LCA [Reiner & Ramaswami]
• Denver Water data showed upstream energy use for trans-boundary water supply to be small (this is not the case for all cities)
• Economic Input-Output LCA used for food
Denver GHG Inventory Results
Trans-Boundary GHG Emissions Denver, CO
• A Trans-boundary GHG emissions footprint facilitates holistic bio-physical infrastructure systems design.– Cross sector strategies like
Tele-presence vs Airline Travel– Cross-scale Supply Chain
strategies such as “green concrete”
– Shifts in nature of demand in cities (e.g., healthy foods)
– Connects consumption-production
• Prevents shifting of GHG burden from within boundary to outside boundary– Hydrogen-cars are “zero”
emitting in-boundary, but with significant CO2 from fuel production outside boundaries
Comparison w/ National, State and Other City Data
DenverPer Capita
GHG Emissions (mtCO2e/person
)
National, State & Other CitiesPer Capita
GHG Emissions (mtCO2e/person)
Direct energy use plus airline travel and key urban materials
Denver:25.3
National; State:25; 25.5
Direct energy use (no airline travel, no fuel refining, noproduction of concrete, food and food packaging)
Denver:18.9
Other Colorado Cities
18.4 – 19.6
Key Question: Do the trans-boundary inclusions lead to scale consistency in inclusions from the city-scale to national-scale?
Measuring Sustainability: Greenhouse Gas Footprints of 8 US Cities
Hillman and Ramaswami, Environ. Sci. Technol., 2010
UCD Scope 3 Inclusions Increased Per Capita Emissions by 46% on Average
0
5
10
15
20
25
Denve
r, CO
Boulde
r, CO
Fort C
ollins
, CO
Arvada
, CO
Portlan
d, OR
Seattle
, WA
Minnea
polis,
MN
Austin
, TX
Averag
eGH
G E
mis
sion
s pe
r Cap
ita (m
t-CO
2e/c
apita
)
Long Distance FreightWater / WW / WasteCementFoodFuel Processing (W2P)Airline (P2W)Surface Transport (P2W)Buildings Energy Use
Hillman and Ramaswami, Environ. Sci. Technol. 2010
Scale-Convergence w/ National Data?
0.0
5.0
10.0
15.0
20.0
25.0
30.0
Per
Cap
ita G
HG
Em
issi
ons
(mt-C
O2e
/cap
ita)
City per Capita GHG Emissions (with Scope 3 Inclusions and U.S. Average Electricity Emissions Factor)
U.S. GHG Emissions = 24.5 mt-CO2e/cap
Hillman & Ramaswami, Environ. Sci. Technol., 2010
Consistency Across Scale
• Convergence between city-scale and national per capita demonstrate that the UCD expanded inventory method may be approaching a footprint with these inclusions.– Deviations from national per capita could be explained by unique
characteristic, e.g., low commercial-industrial activity in Arvada
• Cities need to track and report benchmark metrics in individual sectors to better understand their footprint and assess progress. (See examples next page)
• GHG per-capita is not be the best metric for inter-city comparisons in the geographic-plus method– Consider outliers with very low or very high industrial-commercial
activity– Consider easy metrics to indentify highly producing cities,
consuming cities and roughly carbon trade-balanced cities.
Examples of ConsumptionMetrics & Benchmarks
Sample Consumption Metrics for Tracking Over Time
National or (CO State)
BenchmarkDenver, CO
Commercial-Industrial Energy Use Intensity
(kBtu/sf)138 122
Residential Electricity Use(kWh/hh/mo)
(667) 545
Jet Fuel per passenger (gallons/enplaned passenger)
22 19
Road Transport(VMT/capita/day)
(28) 24
* (Hillman & Ramaswami, 2010)
Policy Relevance: Geographic-Plus vs Consumption-Based Carbon Footprints for Cities
Geographic-Plus: Policy Relevance
• Pro: Well suited to show impacts of local infrastructure policies and trans-boundary changes/policies across scale
• Pro: Well suited to link local energy use to local climate change impacts like urban heat island effect, local public health
• Pro: Able to link energy use to economic measures (GDP, local jobs created)
• Con: Needs better blended metrics for inter-city comparability, e.g., GHG/[resident plus jobs] – to be studied
Consumption-Based: Policy Relevance
• Pro: Most rigorous method of comparing per capita GHG emissions
• Pro: Facilitate shifting government and household expenditures toward “cleaner” products/regions of the world
• Con: City is split up – with household consumption emphasized Ignores local commercial-industrial activity that is exported elsewhere even though these create local jobs and can be governed locally
• Con: IO data unavailable at sub-county level for smaller county/towns, therefore needing approximations
Denver Final Demand Consumption GHG’s, by Scope: Geographic Plus Captures ~81% in this IMPLAN Run
0.0 1.0 2.0 3.0 4.0
Energy/UtilitiesPersonal Transport: Cars & …
ServicesFood Manufacturing
Industry/ManufacturingWholesale and Retail
Transport: Air and otherMiningHealth
Restaurants/HotelsGovernment Services
CommunicationsEducation
Furniture, Misc GoodsRecreationElectronics
Alc Bevs/TobaccoApparel
AgricultureOther Textiles
mt-CO2e/capita
Scope 1: Natural Gas Combustion
Scope 1: Petroleum Combustion
Scope 2: Electricity Generation
Scope 3: Geographic Plus
Remainder
Meta Analysis of 45 US Counties
• Goal: Identify improved Carbon-emission metrics for Intercity Comparisons and Public Communication.
• Consider various typologies of cities – highly producing cities (e.g., resort towns); highly-consuming cities (suburbian towns) and balanced (metro) communities
Example: Denver vs Routt County
26.424.2
26.1
39.8
0
5
10
15
20
25
30
35
40
45
Denver: C-B Denver: G-P Routt: C-B Routt: G-P
mt-
CO2e
/cap
ita
C-B: Consumption-BasedG-P: Geographic-Plus
Denver GHGTrade Balance = 3% of G-P total
Routt GHGTrade Balance = 39% of G-P total
Ramaswami et al., 2011, submitted ES&T
Impact on Developing a Common Community-wide GHG Protocol
• ICLEI-USA is releasing a draft community-wide protocol for cities to report energy use and GHG emissions, based on some of these studies:
• Community-wide distinguishes from municipal government energy use
• Recommends a Basic and Comprehensive community impact report based on Geogrphic Plus method (this presentation)
• Separate Consumption-based report is recommended where resources and data are available.
Conclusions• Transboundary challenges can be overcome
by developing two different types of energy use or carbon emission footprints for cities:
• Infrastructure Based expanded Geographic Footprint– Detailed data available by infrastructure sectors
• Consumption Based Footprint (Input-output) – data challenges
• Conduct societal learning experiments – how do people use footprint information?
Our Work in U.S. Cities
*Hillman, T; Ramaswami, A. . ES&T 2010
Our Work in Global Cities
Cities in India: Hyderabad, Rajkot and Coimbatore
Our Global City Network*:
• Bangkok (city)• Barcelona (city)• Cape Town (city)• Denver (city and county)• Geneva (Canton)• London (GLA)• Los Angeles (county)• New York City• Prague (GPR)• Toronto (GTA)
*Kennedy, C et al. “Greenhouse Gas Emissions from Global Cities.” Environmental Science & Technology 2009, 43, 7297-7302, August 20, 2009.
Sponsored Program:National Science Foundation
• Integrative Graduate Education & Research Traineeship– $3.2 Million grant– 26 total PhD students
• Engineering, Public Affairs, Planning, Health & Behavioral Sciences
– ~ 20 Masters students from many disciplines (MS Associates)
– Certificate in Sustainable Infrastructure
Other Program Sponsors and Partners: Public Sector
• Colorado Cities:
• Governor’s Energy Office• Colorado Municipal League (CML)• National Civic League • Urban Drainage and Flood Control District• Colorado Department of Public Health &
Environment• Office of Naval Research• US EPA – P3 Program
- Denver, Arvada, Central City, Broomfield, Aurora, Durango, Lafayette, Thornton
• AT&T Foundation: Faculty Fellowship in Industrial Ecology
• Wal-Mart Foundation: Outreach to Colorado Cities in partnership with CML
• MWH: Sponsored project with funded student
• Clinton Global Initiative University Outstanding Commitment Award: funding student outreach
Other Program Sponsors and Partners: Private Sector