Environmental Data Collection

& LCA Methodology

Life Cycle Assessment
NPTEL MOOC Course
Week-3 Material

Brajesh K Dubey, PhD

Environmental Engineering and Management,
Department of Civil Engineering
Indian Institute of Technology, Kharagpur, WB, India
 Chemical Analysis Procedures
• Direct chemical measurement
– Titration of sample for alkalinity or hardness

• Use of a sensor or an instrument

– pH meter, chromatograph

• In many, some form of sample “preparation” will

be necessary
 Sample Preparation & analysis

Physical Chemical
Sample Analysis
Preparation Preparation

Mixing, drying, Acid digestion, Titration,

grinding solvent extraction, electrode,
chemical oxidation instrument
 Instrumental Analysis
• Many different instruments have been developed
to measure the concentration of pollutants in
environmental samples
• These instruments give a “response” that
correlates the amount of sample present
• Responses:
– Increase in conductivity
– Absorbance of light
– Emission of light
 Instrumental Analysis

Response

Concentration
 Instrumental Analysis
• Spectrophotometers – Measure
absorption of light at specific
wavelengths

• Some chemicals are directly

proportional to absorbance

• Many methods that were

developed based on color changes
in titrations can be measured using
absorbance
 Quality Control/Quality Assurance

• QA/QC
• Designed to make sure that the data you gather
are sufficiently accurate, precise, and repeatable
• Mechanisms
– Blanks
– Spikes
– Replicates
 Quality Assurance / Quality Control (QA/QC) Plan

QA project plans are list of detailed activities • Standard Operating Procedures

performed at each stage of the dredged • Sampling strategy and
material evaluation and outline project- procedures
specific data quality objectives that should • Sample custody
be archived for field observations and • Calibration procedure and
measurements, physical analyses, laboratory frequency
chemical analyses, and biological tests.
• Analytical procedures
• Data validation, reduction and
reporting
• Internal QC checks
• Performance and system audits
• Preventive maintenance
• Calculation of data quality
indicators
• Corrective actions
 Method Detection Limit (MDL)
• “The minimum concentration of a substance that
can be measured and reported with 99%
confidence that the analyte concentration is
greater than zero and is determined from analysis
of a sample in a given matrix type containing the
analyte” (EPA, 1992)
 Method Detection Limit (MDL)
• MDL = t (n-1, α=0.99)(S)

where,
t (n-1, α=0.99) = one-sided critical t-value
at 99% level
n = number of sample
S = standard deviation
MDL can be calculated by multiplying standard deviation by one sided critical t-
value at 99% level. Different t-value should be applied by number of sample. EPA
recommends that number of sample would be greater than 7 to make MDL
confident.
 Method Detection Limit (MDL) - example

Sample No. Analysis results (mg/L)

1 12.11
2 12.02
3 12.21
4 12.06
5 12.57
6 12.42
7 12.09
8 12.32
 Method Detection Limit (MDL) - example

• N=8
• Standard deviation = 0.19
• t (7, α=0.99) = 3.00

MDL = 0.19 X 3.00 = 0.57 mg/L

MDL at given results is 0.57 mg/L. It indicates that 0.57 mg/L would be
minimum concentration you can trust with 99% confidence level.
Chain of Custody - example
End of Week-3, Module-1
Analytical Instruments
 Ion Exchange Chromatograph

Ion exchange chromatograph is used to analyze liquid samples for dissolved ionic or polar
materials by charge properties
Gas Chromatograph

Gas chromatograph is an
instrument to separate
complex gas or liquid samples
to single components. Injected
samples are separated in gas
phase (mobile phase) through
the column. Separated
components are detected by
various types of detector such
as flame-ionized detector
(FID) and thermal conductivity
detector (TCD).
 Gas Chromatograph / Mass Spectrometry

GC/MS is a GC equipped with mass spectrometry as a detector. After complex

chemicals are separated through a GC column, separated components are
fragmented by high energy source in a mass spectrometry. With size and
patterns of fragments, components can be identified.
High-Performance Liquid Chromatograph (HPLC)

Like other chromatograph,

an HPLC is also used to
separate complex chemicals
to single components.
Unlike a GC, HPLC uses
liquid as a mobile phase.
Separated components can
be detected by UV-Vis
absorbance detector. Since
an HPLC can handle only
liquid sample, more
complex and greater
molecular-weight chemicals
can be analyzed.
Liquid Chromatograph /Mass Spectrometry

Like GC-MS, LC-MS is a LC

equipped with mass
spectrometry as a detector. An
LC-MS can be used to analyze
large molecules such as
pharmaceutical products and
protein.
Inductively Coupled Plasma (ICP)

ICP-AES is used to analyze

elements such as heavy
metals and cations using an
inductively coupled plasma
(ICP).
Each element can be
recognized by
electromagnetic radiation
produced by excited atoms
at a wavelength of a
particular element.
Atomic Absorption Spectrometer
 Statistical Calculations are Often Needed
when Working with Concentrations

• Since concentration results are often directly used

as part of regulatory decision-making (i.e. is it a
hazardous waste, is it a safe to be land applied), it
is necessary to perform a statistical analysis.
What is the Chemical Concentration of a Waste?

• Very rarely are wastes or contaminated soils

completely homogenous with respect to chemical
concentration.
• You must collect and analysis multiple samples to
determine what some “representative”
concentration is.
• The more heterogeneous the matrix, the more
samples may be needed to accurately describe the
“representative” concentration.
What is the Representative Concentration?
• Some estimate of the central tendancy
– The mean (the average)
– The median (the middle number)
– The mode (the most common number)
– Some upper or lower confidence limit
• Must consider
– The sample distribution
– What the data will be used for
 Sample Distributions

Normal LogNormal
0.45

0.45
0.40

0.40
0.35
0.35

0.30
0.30

0.25
0.25

0.20
0.20

0.15
0.15

0.10 0.10

0.05 0.05

0.00
0.00
0 5 10 15 20 25 30
1 2 3 4 5 6 7 8 9
 The Type of Distribution Dictates How You
Calculate the Central Tendency

• Consider the following data set for some TCLP

results (in mg/L)
3, 2, 3, 5, 4, 4, 5, 7, 8, 1, 6, 2, 5, 9, 6, 5, 7, 7, 3, 4, 2, 8,
5, 4, 5, 6, 6, 3, 5
N = 29
 The resulting histogram has a similar shape as a normal distribution.
The mean appears to be a good estimate of central tendency.

7 Arithmetic Mean = 4.83 mg/L

Median = 5 mg/L
6 Mode = 5 mg/L
Number of Observances

0
1 2 3 4 5 6 7 8 9 10
Concentration (TCLP mg/L)
 The Type of Distribution Dictates How You
Calculate the Central Tendency

• Consider a different data set, again for some

TCLP results (in mg/L)
3, 5, 2, 5, 3, 3, 4, 3, 7, 1, 7, 2, 4, 9, 5, 5, 11, 6, 2, 6, 2,
8, 4, 2, 4, 6, 16, 4, 20, 3, 3
N = 31
 The resulting histogram gives the appearance that it might be a log-normal
distribution. The mean appears to be somewhat less reliable as an estimate
of central tendency.

7
Arithmetic Mean = 5.32 mg/L
6
Median = 4 mg/L
Mode = 3 mg/L
5
Number of Observances

0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
Concentration (TCLP mg/L)
 It is Important to Check for Normalcy

• A quick test may be used to assess whether the

data are normally distributed:
• Is the mean greater than the standard
deviation and the variance?
• If the mean is not greater than both, the data
set is probably not normally distributed.
End of Week-3, Module-2
Statistical analysis for Environmental
Data
 The Type of Distribution Dictates How You
Calculate the Central Tendency

• Consider the following data set for some TCLP

results (in mg/L)
3, 2, 3, 5, 4, 4, 5, 7, 8, 1, 6, 2, 5, 9, 6, 5, 7, 7, 3, 4, 2, 8,
5, 4, 5, 6, 6, 3, 5
N = 29
 Previous Data Sets
• Data set I
– Arithmetic mean = 4.83 mg/L
– Standard deviation = 2.0
– Variance = 4.0
• It is likely that the data are normally
distributed
 The Type of Distribution Dictates How You
Calculate the Central Tendency

• Consider a different data set, again for some

TCLP results (in mg/L)
3, 5, 2, 5, 3, 3, 4, 3, 7, 1, 7, 2, 4, 9, 5, 5, 11, 6, 2, 6, 2,
8, 4, 2, 4, 6, 16, 4, 20, 3, 3
N = 31
 Previous Data Sets
• Data set II
– Arithmetic mean = 5.32 mg/L
– Standard deviation = 4.10
– Variance = 16.8
• It is likely that the data are not normally
distributed
How Do You Handle Data that are Not Normal?

• First step, check to see whether the data are

log-normally distributed
• Environmental data are often log-normally
distributed
• Perform a log transform on the data and check
for normality again.
 Transformed Data
• Transformed Data Set #2
– Arithmetic mean = 1.46
– Standard deviation = 0.65
– Variance = 0.42
• The mean of the transformed data is greater than the
standard deviation and variation of the transformed
data. It is thus reasonable to treat the original data
as log-normally distributed.
 Geometric Mean
• When the mean of the transformed data is
transformed back, this is known as the
geometric mean.
– Geometric mean of data set II: 4.28 mg/L
 What if the Data are Neither Normally or Log-
Normally Distributed?
• Find another distribution
• Use a more advanced statistical method (e.g. a
nonparametric method)
 Incorporating Uncertainty
• Assume that the TCLP results presented for data sets
1 and 2 were for lead. The TC limit is 5 mg/L.
• The arithmetic mean for data set I (4.83 mg/L) and
the geometric mean for data set II (4.28 mg/L) were
less than 5 mg/L.
• These were based on a limited number of samples,
and thus may be somewhat different than the true
mean.
 Incorporating Uncertainty
• We need to factor in this uncertainty.
• Calculate the upper confidence limit (UCL).
• The UCL is a conservative estimate of the
mean that takes into account uncertainties in
the data set.
• It is associated with some degree of statistical
significance, e.g., 95% UCL.
 Standard Deviation
UCL1−α = Mean + t α, n −1
number of samples

Student t statistic
 Example
• For data set I, find UCL Note:
at 95%. Higher
than TC
Limit

2.0
UCL1−0.05 = 4.83 + 1.701 = 5.46
29
T statistic at 0.05 level and 28 degrees of freedom
 Example
• For data set II, find UCL at
95%. We must used
transformed data.

0.65
UCL1−0.05 = 1.46 + 1.697 = 1.65
31

e 1.65
= 5.19
T statistic at 0.05 level and 30 degrees of freedom
 Data Below the Detection Limit

• In many cases, sample results will be below the

analytical instrument’s detection limit.
• If you have BDLs in your sample set, you must
handle them appropriately with statistics. You
can not just neglect them; they say something
valuable about the data.
Options for Handling Samples Below the
Detection Limit
• Assume to be zero
• Assume to be the detection limit
• Assume to be ½ the detection limit
• Other more rigorous statistical approaches
End of Module-3, Week-3
Overview of LCA Methodology
What does
What does it mean
it mean to beto be Green??
Green?? What is
Sustainability??
What is Sustainability??

Copyright: Prof Dr. Michael Lepech, Stanford University. Reproduced with permission
 Sustainability
• Environmentally United Nations World Commission on
Friendly Environment and Development (1987)
Sustainable Development definition:
• Sustainable
Products “… development that meets the needs of
the present without compromising the
• Green product ability of future generations to meet their
• Environmentally own needs.”
Preferable
• Biodegradable
• Recyclable
• Ozone friendly
• Eco-design
• Greenwashing

[Link]

Copyright: Prof Dr. Michael Lepech, Stanford University. Reproduced with permission
 What Can the Earth Handle?
World Population Statistics
• The estimate of Earth’s Maximum
UN Population Division
supportable population is 13.4 billion.

• Looking at the population statistics, Population Annual Growth Rate

where are we headed? (2007) (00 – 07)

Renewable Energy Source Will World 6.7 Billion 1.24%

Limit Our Growth!
LDC 5.4 Billion 1.44%
MDC 1.2 Billion 0.36%
US 301 Million 1.03%
Average per capita energy
consumption is 3 kW/person
•US 12 kW/person
•Industrialized 7.5 kW/person
•Denmark 5.1 kW/person
•Developing 1kW/person
Copyright: Prof Dr. Michael Lepech, Stanford University. Reproduced with permission
 Being Green is Trendy . . . . . . . . . . .
What Does Science Say?
• Industry is looking for ways
to green their products and
manufacturing processes.
• Individuals and Families are
looking to green their
homes and lifestyles.

• How can you tell if

something really is green??
• What is currently happening
to achieve this goal?

• Scientists perform a Life

Cycle Assessment (LCA)
[Link]

Copyright: Prof Dr. Michael Lepech, Stanford University. Reproduced with permission
Life Cycle Assessment

A Scientific Way to Look at

Going Green!

Copyright: Prof Dr. Michael Lepech, Stanford University. Reproduced with permission
 Definition:
“Compilation and evaluation of the ISO = International Organization for
inputs, outputs and the potential Standardization
environmental impacts of a product Ensures that an LCA is completed
system throughout its life cycle” in a certain way.
This establishes an environmental
WHAT CAN BE DONE
profile of the system!
WITH LCA?
[Link] or project
development and
improvement
[Link] planning
[Link] policy making
[Link] and eco-
declarations
[Link]

Copyright: Prof Dr. Michael Lepech, Stanford University. Reproduced with permission
 Product Life Cycle

M, E M, E M, E M, E M, E M, E

Ra w Mat erial Ma terial Ma nufac ture Us e & Re tirem ent T reat ment
Ac quisit ion Proc ess ing & As sem bly Serv ice & Rec overy Dis pos al

W W W W W W

re u s e
rema nufac ture

c lo sed-loop recy cle o p e n - lo o p

re cy cle

M, E = Material and Energy inputs to process and distribution

W = Waste (gas, liquid, or solid) output from product, process, or distribution

Material flow of product component

Copyright: Prof Dr. Michael Lepech, Stanford University. Reproduced with permission
 What Makes Up LCA
• Goal & Scope Definition • Impact Assessment (LCIA)
What are the environmental, social,
What is the purpose of the LCA
and economic affects?
and who is the audience?
• Inventory Analysis (LCI) • Interpretation
Ways to reduce environmental
1. What is the function & functional
impacts.
unit?
2. Where are the boundaries? What conclusions can you draw
3. What data do you need? from the study?
4. What assumptions are you What recommendations can be
making? made?
5. Are there any limitations?

Copyright: Prof Dr. Michael Lepech, Stanford University. Reproduced with permission
End of Module-4, Week-3
 Product Life Cycle

M, E M, E M, E M, E M, E M, E

Ra w Mat erial Ma terial Ma nufac ture Us e & Re tirem ent T reat ment
Ac quisit ion Proc ess ing & As sem bly Serv ice & Rec overy Dis pos al

W W W W W W

re u s e
rema nufac ture

c lo sed-loop recy cle o p e n - lo o p

re cy cle

M, E = Material and Energy inputs to process and distribution

W = Waste (gas, liquid, or solid) output from product, process, or distribution

Material flow of product component

Copyright: Prof Dr. Michael Lepech, Stanford University. Reproduced with permission
 What Makes Up LCA
• Goal & Scope Definition • Impact Assessment (LCIA)
What are the environmental, social,
What is the purpose of the LCA
and economic affects?
and who is the audience?
• Inventory Analysis (LCI) • Interpretation
Ways to reduce environmental
1. What is the function & functional
impacts.
unit?
2. Where are the boundaries? What conclusions can you draw
3. What data do you need? from the study?
4. What assumptions are you What recommendations can be
making? made?
5. Are there any limitations?

Copyright: Prof Dr. Michael Lepech, Stanford University. Reproduced with permission
 Goal and Scope
Wooden Pencil vs. Mechanical Pencil
Goal = Compare 2 writing utensils for classroom use.

Scope: Wooden Pencil (T = Transportation)

Process Flow Diagram
T T T T T
Lumber Lumber End of
Forest Mill
Manufacture Retailer Use Life

T
Rubber (Landfill)
T Sharpening
Graphite
T (Electricity)
Packaging
T
Brass
Copyright: Prof Dr. Michael Lepech, Stanford University. Reproduced with permission
 Scope: Mechanical Pencil
PE = Polyethylene
PP = Polypropylene
Both materials are plastic polymers (large
molecules) used to make many products. [Link]

T T T T T
End of
Oil PE / PP Manufacture Retailer Use Life

T
Rubber (Landfill)
T
Graphite T = Transportation
T
Packaging
T
Spring
Copyright: Prof Dr. Michael Lepech, Stanford University. Reproduced with permission
 Function & Functional Unit
Function Example
• Service provided by a Wooden Pencil vs. Mechanical
system Pencil
• What it does! • Function = “Writing”
Functional Unit • Functional Unit = “1 meter
• Gives the function a of writing”
number value
• Allows comparison between
products
• Reference point

Copyright: Prof Dr. Michael Lepech, Stanford University. Reproduced with permission
 Items To Consider??
Inputs Outputs
What is needed to make What comes out of the
the substance! system!
1. Energy 1. Products (electricity,
2. Materials materials, goods,
3. Labor services)
2. Waste
3. Emissions
4. Co-products

Copyright: Prof Dr. Michael Lepech, Stanford University. Reproduced with permission
 Data Collection
Life Cycle Inventory Analysis
1. Time-sensitive = past 5 years Never Forget . . . . . . .
2. Geographical = does it match Precision:
the location from the goal The consistent reproducibility
of a measurement
3. Technology = best available
technology for process Completeness:
Covers all the areas outlined
4. Representativeness = reflects in the scope
population of interest
5. Consistency = matches the
procedure
6. Reproducibility = another
person could find it

Copyright: Prof Dr. Michael Lepech, Stanford University. Reproduced with permission
 LCA in Action: Think About It!
Paper Plate vs. China (Plate You Wash & Reuse)

 What is the function?  What are the impacts

 What is the functional to the environment?
unit?  Is there waste?
 What materials &  Does washing the
resources are used? China produce waste?
 What does it take to  What types of data do
produce both? you need?
 How do you know
which is better?

Copyright: Prof Dr. Michael Lepech, Stanford University. Reproduced with permission
 Data Analysis
Environmental Impact Categories
Global Warming Potential Abiotic Depletion
• Gases in the • Consumption of non-
atmosphere that absorb living resources
and emit radiation Human Toxicity Potential
• Trap heat from the sun • Value that shows harms
• Water vapor, CO2 , CH4 , to humans from
ozone, NO2 chemicals
Land Use
• How much land is
needed

Copyright: Prof Dr. Michael Lepech, Stanford University. Reproduced with permission
 Environmental Impact Categories
Continued . . . . .
Eutrophication Acidification
• Increase in chemical • caused by pollution from
nutrients containing fuels & acid rain
nitrogen or phosphorus • low pH
• land or water Smog (Winter or Summer)
• overgrowth of plants Energy Use
• killing organisms at Solid Waste
bottom of water Oil
Water Use . . . . . . . AND MANY MORE!!
Mercury

Copyright: Prof Dr. Michael Lepech, Stanford University. Reproduced with permission
 Where Do We Go From Here?
• What is the purpose of • Industry & Individuals
all this data?? can take a closer look at
• Scientists can make how they can make a
recommendations of difference
choices that are less
impactful
• Scientists can analyze a
particular impact and
focus on a solution

Copyright: Prof Dr. Michael Lepech, Stanford University. Reproduced with permission
End of Module-5, Week-3