Environmental Life Cycle

Assessment
CEE 12-714 / EPP 19-714

Lecture 6: LCA Data, Part 1

February 5, 2018
 Administrivia
2

• HW2 due midnight, Thursday Feb 8

• Normal office hours: Office hours
Tuesday 1:30-2:30 PM
Wednesday 1:30-2:30 PM
Thursday 12:00-1:00 PM
CEE Classroom (PH A7F)

• TA emails:
Send any emails to
 ksears@andrew.cmu.edu ALL 3 TAs for fastest
 aarfaj@andrew.cmu.edu and most consistent
 woosukc@andrew.cmu.edu responses!

2
Administrivia
• HW1 returned o State and justify your
assumptions
 Q1: 7.8/10 o Label your graphics carefully
 Q2: 7.9/10 o Scrub your final answers for
 Q3: 16.9/20 significant figures
 Q4: 8.3/10
 Overall: 40.9/50
 Carefully review grader feedback
• HW3 released Friday

3
Administrivia
• Group project (30% of your grade)
 First deliverable: Topic Proposal
 Due Monday, Feb 26, midnight via Canvas
 ~2 pages, first cut at Goal and Scope
 See Canvas/Group Project

• Possible opportunity with International Living

Future Institute’s Pittsburgh Living Product Hub
 https://pittmoss.com/

4
Project schedule
1/31-2/5: CATME survey
2/7: Release of teams
2/26: Project proposals due (5 points). Includes group members,
topic, draft goal and scope document (including your study design
parameters), initial description of data sources and brief review of
literature for existing studies. About two pages in length.
3/21: Updated proposal due (10 points). Revised/expanded topic,
updated goal/scope document, updates on data sources and other
literature found. Team meeting with instructor to discuss.
4/16: Preliminary results due (20 points). Inventory results, draft of
report.
4/30 and 5/2: Presentations (40 points). Groups will be assigned
presentation times randomly.
5/9: Written report due (50 points). Report body should be about 20
pages. Additional materials (data, sample calculations, etc.) may be
included in appendices

5
Administrivia
• LCA job search ideas
 Franklin and Associates http://fal.com/careers.html
 Environmental Resources Management
https://www.erm.com/en/careers/
 Industrial Economics Incorporated http://indecon.com
• American Center for Life Cycle Assessment
 LCA XVIII, September 25-28, 2018, Fort Collins, CO
 http://aclca.org
• International Symposium on Sustainable Systems
and Technology
 June 25-28, 2018, Buffalo, NY
 www.conftool.net/issst2018

6
Course trajectory
1. Introductions 10. Uncertainty
2. Life cycle thinking 11. Input-output LCA
3. Quantitative methods and 12. Process-matrix LCA
life cycle cost analysis
13. Hybrid LCA
4. ISO LCA framework
14. Impact assessment
5. Critical review
15. Structural path analysis
6. LCA data sources
16. Professional responsibility
7. Life cycle inventory
17. Carbon footprinting
8. SimaPro
18. LCA for big decisions
9. Handling multifunction
19. Project presentations
systems

7
Lecture 6 & 7

LCA Data

8
 Phases of an LCA - Iterative

• Not a once through

process - all phases are
iterative!

• Adjust as you go along

• Changes after iteration

happen in ~100% of
studies

9
 So, your study has a scope.
Now what?

Source: http://read.nxtbook.com/wiley/plasticsengineering/may2014/processengineeringforrecycled.html
10
 Inventory Analysis Overview

Source: ISO 14044, Fig. 1

11
Data needs overview
• Must know what you are tracking for your
inventory
• Need data for all inputs and outputs
• Level of detail/aggregation may depend on:
 Purpose of study
 Data availability
 Amount of uncertainty/variability
• Follow data quality requirements
• Remember: data needs and goal/scope are set
iteratively

12
Common response: track everything

13
Inventory scope
• We don’t just choose “air emissions” or “water
emissions”
 We choose specific emissions of gases or substances
 e.g., CO2, SO2, ..
• Can be guided by desired impact categories
 Planned focus on fuel use, climate change, etc.

14
Data needs overview
• Must know what you are tracking for your
inventory
• Need data for all inputs and outputs
• Level of detail/aggregation may depend on:
 Purpose of study
 Data availability
 Amount of uncertainty/variability
• Remember: data needs and goal/scope are set
iteratively

15
 Elementary and product flows

• From ISO 14044:

 Energy inputs, raw material inputs, ancillary inputs, other
physical inputs
 Products, co-products and waste
 Releases to air, water and soil
 Other environmental aspects (noise, odor, land, resource
depletion …)
16
Data needs overview
• Must know what you are tracking for your
inventory
• Need data for all inputs and outputs
• Level of detail/aggregation may depend on:
 Purpose of study
 Data availability
 Amount of uncertainty
• Remember: data needs and goal/scope are set
iteratively

17
Aggregation: simplest models

Natural gas
Pre-production
Wet Natural
Natural gas Gas Natural gas
Extraction Processing
Co-products
Natural Gas
Dry Natural
(life-cycle)
Gas
Dry Natural
Natural gas Gas
Transmission
1 year of energy
services
Natural gas
Distribution Emissions

Boiler/CHP
Water withdrawals
plant

1 year of
energy services

18
 Aggregation or more detail

Level of detail/aggregation
may depend on:
Purpose of study
Data availability
Amount of uncertainty
19
Data needs overview
• Must know what you are tracking for your
inventory
• Need data for all inputs and outputs
• Level of detail/aggregation may depend on:
 Purpose of study
 Data availability
 Amount of uncertainty/variability
• Remember: data needs and goal/scope are set
iteratively

20
Data Collection

21
 ISO 14044: 4.2.3.6

SDP: Data Quality Requirements

• Fundamental expectations of needed data
 Temporal
 Geographical
 Technology
 Sources
 Precision
 Uncertainty
 Completeness
 See ISO 14044 for more

22
Data Quality Requirements: Temporal
• Why does this matter?

23
 Data Quality Requirements: Temporal
• What is my year of analysis?
• Has the process changed recently?
• Is it likely to change in the near future?
• Have upstream/downstream processes
changed?
• Has new information arisen recently?
• From what year is the data?
 NOT the same as the year of publication!!!
• …
24
Data Quality Requirements:
Geographical
• Are you looking for a specific location? A general
region? ...
• Why does geography matter?

25
Data Quality Requirements:
Geographical
• Geography could affect:
 Regional emissions factors
 Technology choice
 Operational requirements
 Transportation
 Potential emissions / impact of emissions
 Transparancy
 …

26
Data Quality: Technology
• Different pathways to same product
 Main technology and supporting technologies

• Possible criteria:
 Specific company
 Specific technology/technologies
 Average technology mix
 Representative technology

27
Data Quality: Data Sources
• Trustworthy source?
• Peer reviewed?
• Underlying assumptions?
• Type of data?
 Measured?
 Estimated?
 Calculated?
 Aggregated?
• Selecting multiple sources (why?)
 Careful! What is the original data source? (may not provide
independent data points!)

28
Common Data-driven Mistakes
• Using data that is inconsistent with targeted
process
• Developing internally inconsistent models
• Treating missing data as zero
• Treating highly uncertain data as zero
• Treating analytical detection limits as detections

29
Data sources

30
Data sources
• Primary Data
• LCI Databases
• Government Sources
• Industry Reports
• Academic Literature
• Estimated

31
Primary Data

32
Data sources
• Primary Data
• LCI Databases
• Government Sources
• Industry Reports
• Academic Literature
• Estimated

33
LCA Databases
• Don’t despair, you do not need to collect all of
your own data for LCAs:
 US NREL LCI Database (broad focus, using
extensively, free)
 Simapro databases (broad focus, using extensively
under CMU license, $$, includes ecoinvent)
 BEES (construction materials, free)
 Athena (building materials, etc. $$)

• Look at these for ideas before finalizing ideas

and scope for Course Project

34
US NREL LCI Database

35
SimaPro

36
 Data Sources - Heart of Tools
37

• All ‘tools’ such as SimaPro are ‘front ends’ to

databases
 Aggregate and calculate inventories
 Do impact assessment
• The data is the important part
• The interface is just there to help
• Examine data documentation and metadata
to get what you need

37
Metadata in NREL USLCI

38
Metadata in NREL USLCI (continued)

39
Metadata in SimaPro

40
Metadata in SimaPro (continued)

41
 What you can get from these tools
42

• Key inputs / outputs

• Unit process emissions and other inventory data
• Life-cycle emissions and other inventory data
• Data sources
• Impact Assessment

42
Not all data is measured

43
Data sources
• Primary Data
• LCI Databases
• Government Sources
• Industry Reports
• Academic Literature
• Estimated

44
Government Sources
• Government databases are an excellent source
of up-to-date information
• There is no central/comprehensive list of where
to look.
 Let’s look at some examples

45
U.S. EIA: Fossil Fuel Data

http://www.eia.gov/naturalgas/data.cfm
46
U.S. EIA: Fossil Fuel Data (continued)

EIA has an
enormous amount
of data on a large
range energy-
related topics

http://www.eia.gov/dnav/ng/ng_prod
_sum_dcu_NUS_a.htm

47
U.S. EPA: eGRID

https://www.epa.gov/energy/emis
sions-generation-resource-
integrated-database-egrid
48
U.S. EPA: Toxics Release Inventory

http://www2.epa.gov/toxics-
release-inventory-tri-
program

49
U.S. EPA: Greenhouse Gas Reporting
Program

Note: EPA has an

enormous amount
of data spread
across different
programs

http://www.epa.gov/gh
greporting/

50
USDA: Quick Stats

http://quickstats.
nass.usda.gov/
51
Government Sources

• Lots of other sources

• Be creative!

52
Data sources
• Primary Data
• LCI Databases
• Government Sources
• Industry Reports (Christmas tree study)
• Academic Literature
• Estimated

53
Data sources
• Primary Data
• LCI Databases
• Government Sources
• Industry Reports
• Academic Literature
• Estimated

54
Academic Literature
• Academic literature is an important source of LCA
information
 Complete inventories from existing LCA studies
 Inventories from unit processes
 Can be useful even if study relates to a different system
 Reporting of primary data not available elsewhere
 Estimation of difficult parameters
 May not necessarily be an LCA study
 See how similar studies were scoped out
• Find using databases, Google Scholar, Web of
Science, etc.
• CMU Library – Xiaoju (Julie) Chen is our resource!

55
Data sources
• Primary Data
• LCI Databases
• Government Sources
• Industry Reports
• Academic Literature
• Estimated

56
Data “Collection”
• LCA rarely is primary data only
 Usually have to use at least some secondary sources!
 Many studies are 100% based on secondary sources

• This is usually the longest, most labor intensive

part of an LCA study.

57
Example: Greenhouse Gas Emissions
From Polyethylene Plastic Production
• Natural gas pre-production
• literature value
• Natural gas extraction
• direct from government data
• Natural gas processing
• analysis of government data
• Steam cracking
• estimated
• Polymerization
• from LCA databases and industry reports
• End of life?

Posen et al. (2015). Environ. Sci. Technol. 49(1): 93-102 58

Data Validation

59
Data Validation
• After collecting data, pause, assess
 Meeting data quality requirements?
 Within range expected?
 Mass balance maintained?
 Comparable to other processes?
 Modification of scope needed?
• Results go in your report

60
Wrap-up
• LCA is very data hungry
 Often the exact data you need will not exist
• Data can come from lots of places
 Almost all studies have to ‘mix and match’
• It’s ok to redefine your study as you go
• Think before you use a data source!

61
Questions?

62
Next up
• Complete HW 2, due Thursday, midnight
• Finish reading Chapter 5
• Read the NREL User Guide in more detail
• Skim the paper on Physical versus Digital Music

63