Comprehensive analysis of environmental impacts of projects

Document Sample
Comprehensive analysis of environmental impacts of projects Powered By Docstoc
					Chapter 4
Comprehensive analysis of environmental impacts of projects

Integrated Assessment is an established term for comprehensive environ-
mental analysis of projects. We discuss three central issues:

– How to make an Environmental Impact Statement?
– How to select good monitoring points and performance measures on the
  impact pathway?
– What are the key concepts in Life Cycle Analysis?

                       “But what happens when you come to the beginning again” Alice
                          ventured to ask. “Suppose we change the subject”, the March
                        Hare interrupted, yawning. Lewis Carroll: Alice’s adventures in


    Smoke related cancers may develop tens of years after its causal agent
has ceased to act; and pollution may drift across continents before they affect
a target. Integrated assessment (IA) is a comprehensive approach that
includes a bundle of tools for handling project effects that occur over long
distances in time and space. A project is an activity that starts at some point
in time, and is assigned time slots and resources; in our context a project is
also assumed to have environmental impacts. An integrated assessment,
IA, of a project thus includes a study of all impacts, on the environment as
well as humans. It applies a life cycle perspective, considering production,
consumption, as well as disposal and reuse. Impacts from different stages
may be quite different for different kinds of goods or services.
    IA is structured around four main themes:

– Objectives: Who is the decision maker and what are the end impacts or
  ultimate goals?

44                                                                              Chapter 4

– Industry: Industrial, agricultural and socioeconomic activities that
  produce goods and services.
– Ecology: Dilution processes and links between industrial activities and
  impacts on humans and the environment.
– Economics: Monetary assessment that links material cycles to man’s
  short and long-term strategies and goals. This includes ecological
  economics, as well as utility analysis and psychometric methods that deal
  with valuation of goods and services.

    Environmental Impact Analysis (EIA) is more restricted than IA. It
ends up with an environmental impact statement and consists of the
following elements:

– Environmental inventory: A description of the existing environment at
  the site where the project is proposed according to a checklist of
  descriptors of physiological, chemical, biological, cultural, and socio-
  economic environments.
– Environmental impact assessment: A systematic identification and
  evaluation of potential impacts on the environmental inventory of the
  proposed project, program, or legislative action.

Figure 4-1. Integrated assessment of a paper production and paper use project. At C there are
opportunities for comparisons.
4. Integrated Assessment                                                    45

    Integrated assessment is illustrated in Figure 4-1, with the life cycle of
paper as an example. The IA project starts with a life cycle analysis, LCA.
The life cycle of paper consists of timber logging, production of paper at the
mill, transport and recovery of waste products. Usually, an environmental
impact statement will be required when a paper mill is built or upgraded, i.e.,
the second step in the LCA of Figure 4-1.
    Common forest related damages are destruction of natural habitats during
logging operations. The paper mill discharges pollutants, chronic or acci-
dentally, to water and air. As part of the IA, one develops spread and
dilution models to describe their impacts. To evaluate alternative mitigation
measures that reduce the impacts, one needs to evaluate them, either through
decision analysis or – more commonly – by using standard unit prices
attached to the pollutants. In routine assessments, LCA and development of
dilution models and dose- response models are expensive activities, and
often replaced by unit prices, which are assumed to be average damage cost
to the society of emitting one unit of pollution.
    The cost of the environmental load of a project can be put into
perspective by comparing its contribution to the gross domestic product,

2.        EIA STEPS

    Many countries require EIA by law. The details depend on the type and
size of the project and the vulnerability of the environment. In general, EIA
involves six steps – some requiring considerably more attention than others.

2.1       Step 1: Screening the Environmental Problem

   The first step is to perform a screening procedure to examine whether an
impact assessment is at all required, or if it belongs to a category that is
exempted. Renovation, maintenance, and minor extensions are usually
exempted. It is often useful to describe environmental threats along two
dimensions: severity of impact and probability that the impact will occur.
Impact severity is categorized in four classes, and the combined effect of
severity and probability is attributed a level of significance. The highest
significance category – very significant – consists of those projects that
probably will cause extremely large damage. Recurrent flood is an example.
The significance of nuclear power fallout and collapse of very large dams
depends upon the probability of failure. The highest significance category
may also include projects where legal thresholds are surpassed or where
damage is certain. In cases where a stakeholder firmly believes that both
46                                                                             Chapter 4

severity and probability are higher than the experts’ risk estimates, the
precautionary principle suggests that the higher significance category be
    An overview is shown in Table 4-1. If the probability is less than 10–6,
the combined effect is usually deemed insignificant, and the project is exem-
pted. If the impact is “significant”, an EIA should be carried out.

Table 4-1. Defining significance.1 p is impact probability.
 Impact                                  Impact likelihood category
 severity    A: p > 0.5       B: 0.05 < p <0.5            C: 10–6 < p < 0.05   D: p < 10–6
 very        very             significant                 significant          not
 high        significant                                                       significant
 high        significant      moderately significant      moderately           no further
                                                          significant          study
 moderate moderately          not significant             no further study     no further
             significant                                                       study
  low        not              not significant             no further study     no further
             significant                                                       study

    If it is “moderately significant”, an EIA may be required and a screening
test should explore the problem. “Not significant” means that a screening
study should be undertaken, but that a full EIA is not likely to be required.
“No further study” suggests that no further consideration is required.

2.2         Step 2: The EIA document

   The EIA document is to be included in the project proposal, which is sub-
mitted to the authorities. It normally contains the following:

1. Relevant government rules and regulations.
2. Benchmark information, that is, information about the environment
   without the project.
3. Estimates of the impacts of the project, as well as an outline of elements
   to be monitored before, during, and after the construction period. By
   comparing predicted and observed data, one can assess the accuracy of
   the predictions.
4. An evaluation of the predicted impacts, for example by comparing
   them to established standards. An example is transparency in waters
   designated for swimming, which should be more than 2m – the lowest
   required transparency in certified swimming ponds.
5. Identification of possible mitigation actions.
6. Writing the document.

     Step 2 takes often 5% –15% of the budget.
4. Integrated Assessment                                                    47

2.3       Step 3: Multidisciplinary Team Management

    EIA requires several types of expertise; engineers assess emissions and
discharges; atmosphere and hydrology scientists predict how these spread in
air and water; toxicologists assess the resulting levels of toxicity, and
biologists and health workers assess the impacts on ecosystems and people.
Output from one discipline is therefore input to another. Since researchers in
all disciplines tend to stress the uncertainty of their own evaluations, the
chained reasoning typically produces a large number of scenarios. It is the
task of the team manager to reduce it to something manageable.
    Allow 5% –15% of the budget for step 3.

2.4       Step 4: Public Participation

    The public is an important stakeholder in environmental matters, and
often wields considerable decision power. The purpose of public participat-
ion is to exchange information, elicit local values, establish credibility, and
observe peoples’ “right to know”. There are several levels of public decision
    Citizen control: Citizen control means that citizens without stakes in the
project have the majority in decision-making bodies, or that a referendum
decides the outcome. If a general referendum is not possible, it is necessary
to select representative citizens.
    Delegation of power: Decision making power may be delegated. The
choice of delegates is important since many projects have failed, in particu-
lar in developing countries, because the project managers did not address the
concerns of the people that were directly affected, or used a decision making
process which was unfamiliar to the people concerned .2 There are several
ways to choose delegates:

– The proponent of the project may select members to the delegation that
  balance different interests and represent all stakeholder groups.
– The proponent may select members to a voting committee, which in turn
  selects the decision-making body. Alternatively, the voting committee
  may be elected by the people.
– The proponent may identify stakeholders and allow them to select their
  representatives. The stakeholders will often be invited to an initial
  meeting where the project is explained.3

   Allocate 2 – 5% of the budget for step 4.
48                                                                    Chapter 4

2.5        Step 5: General Study

    The general study is the topic of the next section. It consists among other
tasks of goal hierarchy construction and consequence analysis.

2.6        Step 6: EIA Review

   This step is to ensure that the EIA has been adequately completed as
agreed. An independent group should perform this important task.
   Allocate 10% of the funds for step 6.


    In this section, we follow up the discussion of goal hierarchies from the
previous chapter. There we defined an end-impact as a consequence that
affects fundamental values – something we care about, or want to avoid.
End-impacts are represented by decision criteria in the goal hierarchy. At the
same time, an end-impact is an end-point in a causal chain of environmental
processes. There is for instance a causal link from a production process to
emission of pollutants to discharges through a chimney to concentrations in
the air to end-point impacts like coughing in children.
    The challenge is that we have freedom of choice when we design goal
hierarchies. They are supposed to systematize values or end-point impacts
that are at stake in connection with the project. If possible, the choice of end-
impacts should reflect fundamental values rather than instrumental values.
Suppose for example that you are concerned about water pollution from
phosphorus. You could use ton nutrient discharged as a decision criterion,
but this is only an indicator of your real concerns. You could therefore go
further down the impact pathway and perhaps measure the impact as water
surface area with above guideline concentration, and use that as a decision
criterion. The guideline could be related to water quality class. Pollution
control authorities define such classes according to whether the water is used
for consumption, industry, recreation or other purposes.
    Pollutants are emitted, spread, and do damage. The reason we recom-
mend focusing on end-point impacts – the real damage – is that damage is
easier to valuate. We feel it or see it, and are willing to pay to avoid it. On
the other hand, prediction of damage effects is hard. While it is rather easy to
predict emissions from a project, it is harder to predict ambient concentra-
tions, and hardest to predict end-point impacts. Therefore we have to
4. Integrated Assessment                                                                   49

Table 4-2. Pollution monitoring point and measurement.
 Monitoring point and measure                  Examples of decision criteria
 End-of-pipe emissions or discharges           Gap between ambition and achievement
                                               such as ton emission above threshold per day
                                               Fraction of total emission and discharges
                                               Pollution fee in euro per ton emitted
 Ambient conditions                            Damage area (km2)
 Area that is contaminated above a threshold
 End-point impacts                             Number of exposed persons
                                               Cases of increased risk of heart attack

We must therefore decide at which one of the three stages along the impact
pathway effects can be most conveniently monitored and valuated. Each
stage gives different decision criteria, as shown in Table 4-2. Each stage has
its own pollution severity measures as shown in the right column of the
table. Naturally, all methods require that the effects of the pollutant can be
isolated from other sources. Each stage has advantages and disadvantages
with respect to evaluation and implementation in an industrial context. To
minimize the damage area, for instance, is an instrumental objective, and to
evaluate it, one must assume that there is a causal relationship to the end-
point objectives. If the goal is to minimize phosphorus load in a lake, it is
because one assumes that the phosphorus causes end-point environmental
effects by being converted to algal biomass. Sometimes we are able to model
the relationship between emission and end-point damages by generally
accepted causal chains, although uncertainties may be large.
   The chain between emissions, polluted area and damage to valuable
resources is illustrated in Figure 4-2.

                           A: concentration normally about
                           30 x ambient concentration
                                                    dilution depends upon
  Recovered                                         location
  in cleaning 8
  weight units
                                               C: average target
                                               concentration :1       Below NOEL, 4
                                               tree per areal unit    areal units limit
 discharge, 9
 weight units
                                                             B: above NOEL,
                                                             2 areal units

   Figure 4-2. A typical impact pathway: emissions, polluted area and end-point impacts.
50                                                                     Chapter 4

    The left part of the figure illustrates the dilution processes; the right part
illustrates the effect of pollutant concentrations on resources. The figure
shows an example where the end-of-process pollution yields nine pollutant
units. Through recycling and purification, one unit is left and discharged
through the chimney. End-of-pipe discharges are thus one pollutant units.
The consequence is two area units with concentrations greater than NOEL
(NOEL = No observable effect level), and four area units below NOEL. The
average resource concentration is one tree per area unit, thus the expected
damage affects two trees. For a worst-case damage, one could assume that
the emission becomes contained in a pocket of air and transported to a place
where there is a maximum concentration of susceptible resources. For
example, the package reaches two area units where the maximum intrinsic
resource concentration is three trees per area unit, and thus the worst-case
damage would be six units. The probability of a reasonable worst-case
damage should be higher than 10–6. If one of the hurricanes that hit the
southeastern coast of USA hits Manhattan, a substantial disaster would
occur. 4

3.1       End-of Pipe Monitoring

     Industrial emissions are monitored by the EPA and reported in the form
of an inventory list of emissions and discharges. The so-called Toxic Release
Inventory, TRI, assesses the environmental burden, EB, imposed by major
industries through a list of 517 potentially toxic chemicals. (The 2002 TRI
list, released 2004.)
     In project evaluation, it is also desirable to give information about the
severity of the pollutants. This is done in the form of an index that summa-
rizes the total burden of the pollution. If the index is below a certain level,
the project may be carried out; if it is above, the project should be cancelled.
There are at least four indices: “Total release”, “Toxicity weighted”,
“Distance-to-target” and “Fractional emission”.

3.1.1     Total Release Index

    This index is based on emissions of potential toxic chemicals from a list
of chemical categories in terms of discharged tons. A Dow chemical facility
in La Porte, Texas reported in 1987 that 1730 thousand pounds of toxic
compounds were released and transferred, and this included release of
monoclorobenzene to air and water. In 1993, the release was reduced to 960
thousand pounds.5

      EB1 = ∑i X i
4. Integrated Assessment                                                              51

    Xi are emissions measured in ton per time unit. In some cases the burden
is assessed by only one discharged chemical, n = 1, presumably the one with
largest impact on the environment.

3.1.2      Toxicity Weighted Index

     The concentrations of released products, although all rated as potentially
toxic, may have different threshold values before they cause toxic effects.
Threshold Limit Values, TLV, have been reported ranging from 0.37 to 1780
µg m–3 for a limited selection of chemicals on the TRI list.6 Total phospho-
rus – the sum of all chemicals that contain phosphorus, total nitrogen and a
series of other substances affecting the environment are not included in the
list. The index EB2 is obtained by dividing the amount of toxic material (on
the list of 517 potentially toxic materials) by the threshold concentrations of
the respective products. Since TLV has a threshold value of 1 µg m–3 for
sulfuric acid and phosphoric acid, EB2 expresses the burden as a trans-
formation of the emitted products into sulfuric acid equivalent emission.
With EB2 as pollution index, the ranking of firms on the TRI list changes
considerably .7

     EB2 = ∑i X i × TLVref / TLVi

   Exercise: With reference to Table 4-3, estimate the environmental burden
caused by paper mills A and B. Which factory is the worst?
   Solution: According to the Total Release Index it is factory A, but
according to the Toxicity Weighted index, it is factory B, as shown in Table

Table 4-3. Emissions in kg per 1000 kg product from paper mills A and B. TLV is Threshold
Level Value.
 Factory                SO2     NOx Sum SO2 + NOx          Sum SO2 + NOx (weighted)
 A                       0.3    0.72        1.02                     0.0071
 B                       0.2    1.00        1.20                     0.0077
 TLV, 24 hr µg m–3      100      175

3.1.3      Distance to Target Indicator

   This indicator relates pollution emissions to political targets for emission
reductions. The idea is that it is easier for people to appreciate environmental
performance regarding a certain emission if it can be reported as a fraction of
the target, or distance-to-target.8 The gap closure principle is applied when
actual end-of-process discharges are compared to reduction goals:
52                                                                   Chapter 4

      I = ∆E /( E present − E goal )                                      (4-3)

   For example, if a national goal is to reduce nitrogen discharges from
28000 ton in 2005 to 14000 ton in 2010, a reduction of 1000 ton gives I =
1000/14000 = 0.07 over the time period.

3.1.4       Fractional Emission

    The fourth method is to calculate emissions or discharges as a fraction of
total discharges in a country or a region. This method corresponds to the
distance-to-target method when national goals are zero emission.

3.1.5       Discussion

    The Toxicity weighted index has been criticized because TLV has to
refer to one type of toxicity, for instance with regard to human health.
However, pollution damage occurs also at lower levels. For example, corro-
sion from SO2 concentrations in the air occurs at much lower values than
health effects. On the other hand, “safe” pollution levels for health effects
may at closer examination be higher than reported, because excessive safety
factors have been used.9 Another uncertainty lies in subtle, long time effects
that may not have been discovered. The distance-to-target index circumvents
this critique, because it merely expresses current political concerns.

3.2         Ambient Pollutant Concentrations

    Damage can be evaluated by reporting the environmental burden in terms
of area with ambient concentration above TLV.

      EB3 = ∑i Ai

    Ai is the area (in hectare) with concentration above TLV with regard to
chemical i.
    A version of this method is the Habitat evaluation method. In this case,
the concern is about the quality of a habitat. The quality may be degraded
because of air pollution or toxic discharges, or because of noise and human
activities that affect populations of the habitat. The areas are weighted with
respect to their habitat quality, HQI, ranging from 0 to 1.0, where 1.0 is best.
Thus, the environmental quality is measured as “prime quality equivalent
4. Integrated Assessment                                                     53

      EB4 = ∑i Ai × HQI i ≤ ∑ Ai

    An underlying assumption for reporting the damaged areas, EB3, or EB4
is that most people have an idea of the extent of an area. That is, they can
intuitively relate to 10 km2 of degraded land. The Method described in 3.2
is used to calculate the maximum area damaged. Area methods reflect the
toxicity of each pollutant, as does the TRI index.

3.3       End-Point Impacts

   There are many different end point impacts, and we only give some
examples below. We start with birds; their wings make them especially
challenging since they move easily.

3.3.1     Birds that Move and Communities with Long Recovery Times

    An oil spill may pollute large areas of the ocean, but there is not
necessarily a direct connection to the number of birds that die. Thus, for
natural resources that move, like birds, some kind of estimate of concen-
trations at the time of a discharge must be available. For some recipients, it
is also relevant to include the time it takes for the environment to return to
normal state. This was done in the oil spill case in Chapter 3. If the restitu-
tion time is long, the effects are worse than if it is short, other parameters
being equal. For example, if an oil spill hits a colony of kittiwakes (Rissa
tridactyla) and kills 1000 birds, this may be considered as a smaller environ-
mental damage than if the oil spill kills 1000 Guillemots (Uria alge). The
reason is that it takes much longer time to restitute the guillemot population
than it takes to restitute the kittiwake population.10

3.3.2      Box Calculation of the Polluted Area

    This method can be used to calculate the polluted area by assuming that
the pollutant is confined for a certain time period to a box with a certain area
and height. We have to know i) the amount, Q, of the pollutant emitted over
a time period, ii) the mixing height, h, in air, water or soil, depending upon
the receiving medium, and iii) the No Observable Effect Level, NOEL, of
the concentration of the pollutant. NOEL is measured in per mille concen-
tration or µg m–3. For contamination of water by nutrients, an upper thresh-
old level defining the best “Water class” can be used.
    End-of-process discharges can sometimes be obtained from Life-Cycle
Analyses, LCA. LCA typically computes discharges per “functional unit”,
54                                                                   Chapter 4

for instance 0.01 kg SO2 per 1000 kg newsprint paper produced. The total
amount of pollutant, Q, can then be found by multiplying with the
production volume over a time period corresponding to the residence time of
the pollutant in the receiving medium. Typical residence times for pollutants
in air are two hours to 24 hours, and in water one day to one week. The
height of the box corresponding to the minimum mixing height, can be set to
5 m in air and 1 m in water although 100 m and 10 m would be more typical
minimum mixing heights.
    The polluted area is calculated this way:

           10 9 Q
      A=                                                                  (4-6)

    Q is discharge (kg) during the residence time – 24 hours in air and a
week for water pollutants. A is area in km2, c is the NOEL (24 hour) concen-
tration of potential pollutant in µg m–3. The factor 109 converts from µg to
kg. h is height of equivalent air volume or water volume, 5 meters for air and
one meter for water.

3.3.3       Resource Damage Calculation

    This method estimates the damage within an area, A (km2) caused by a
certain degree of pollution. For air pollutants, we use as input the
concentration of natural resources or people susceptible to the pollutant in
the area. For health effects, we would for example use the density of people
living in, or passing through, the area considered. Calculate the resource
damage as:

      D = CR × F × A                                                      (4-7)

   Here, CR is the density of resources, such as the number of people per
km2. F is the fraction of the resources that is susceptible. The damage index
unit is then number of people. (The average density of people in 29 cities in
the Baltic drainage basin is 10,000 km – 2.)11 If there are several resources or
several areas with different levels of pollutant, they have to be divided into
areas with approximately uniform distributions.

3.4         Valuation of End-Impacts

   To assign monetary values to end-impacts is probably one of the greatest
challenges in environmental decision-making. It is a necessary step if
4. Integrated Assessment                                                  55

monetary gain from a project is to be compared directly to environmental
loss. As we discussed in Chapter 2, valuation of environment and health is a
question of subjective preference, and there are no clear answers regarding
the best way to perform valuation. Experts may do it on behalf of the public,
or it could be done by popular votes, or assessments based on answers to

Table 4-4. Estimated costs of NOx emissions from a project.12
 Impact                            Damage cost, euro kg–1 1994
 Acidification of water                   0.0025
 Acidification of forest                  0.0625
 Health impacts                          12.0250

    We will discuss methodological issues in depth in Chapter 8 on
Willingness-to-pay for environmental amenities (WTP).
    In several other chapters in the book we also include some information
on observed WTPs. It is very practical, especially in Integrated Assessment,
if some information about estimated monetary costs of emission of one unit
of a polluting agent can be included in the report. Such estimates must be
based on predicted damages, such as in a Norwegian study where the cost of
1 kg NOx emission was estimated. The result is shown in Table 4-4. NOx
emission was assumed to have three end-points, water, forest and human
health, and damage costs judgmentally assessed by the Norwegian pollution
control authority. Note that damage caused by acidification is valuated much
lower per kg NOx than damage to human health.
    If it is not known which target the pollutant hits, one has to make
assumptions about probabilities, and compute an expected damage cost
based on the target specific unit costs.


    Life cycle analysis, LCA, deals with the environmental burden caused by
a product or a service during its life span “from cradle to grave”. Nowadays,
one should perhaps say “from cradle to cradle” since materials are often
reused. LCA analysis is most often conducted with the help of commercial
LCA software, like SimaPro13 and the Carnegie-Mellon system.14 A full
description of LCA that complies with the ISO 14040 standard can be found
elsewhere.15 Here, we shall only give a brief account of the method.
56                                                                                 Chapter 4

4.1         Three Central Concepts

    Three terms are important in LCA. The first is functional unit, which is
the unit of analysis in LCA. A functional unit is a product that has certain
functions and causes certain environmental impacts. For example, assume
that you want to use LCA to compare the environmental burden of glass and
plastic used as packing material. Then you have to choose a unit of analysis
that makes comparison possible. You could for instance define the product
as material that will pack 1 liter of wine. This product is the functional unit.
    The second important term is the boundary of the system supporting the
good through its life. A good system boundary definition is important in
LCA, and Figure 4-3 shows a scheme for that purpose.
    The horizontal streams in Figure 4-3 illustrate the flow of the good as it
moves through production and usage. The upper stream is the production
stream, and the lower the reuse-stream. The vertical streams symbolize the
accumulated burdens during the lifetime of the product. (1) is pollution
associated with production; (2) is pollution associated with scrapping of
production equipment; (3) is pollution associated with product use; (4)
shows the reuses stream, (5) shows end-of-life waste depots, and (6) shows
the pollution distantly related to the production of the product
    The third important term is inventory analysis. This is the step in LCA
where energy sources, raw material, and environmental loadings for the
entire life cycle of the product or process are identified.

 (1) “horizontal”          (2) “Vertical”            (3) “Vertical”     (4)reuse
 production                production                production         stream
 stream                    equipment                 use stream
         Raw              Productio                   Use             Waste
         erials                                                                (6)outside
      Important for            Important for            Create
      nuclear                  cars, i.e.,              waste
      power plants             gasoline use             depots

                      Figure 4-3. System boundaries and material streams.
4. Integrated Assessment                                                     57

   Often, LCA stops here. The result is a list like the following for a
washing machine costing 1000 USD (the Carnegie Mellon system): 700
kWh electricity used, 12 GJ energy used, 1 ton CO2 equivalents released, 0.4
ton ores used, 4100 kg SO2 and 600 kg PM10 emitted.16,17

4.2        Impact Analysis

    A next step, impact analysis, or risk analysis, characterizes the effects on
environment and human health. To relate end-of-pipe to end-points is
difficult and laborious. There are, however, several shortcuts available,
depending upon the purpose of the study. One approach is to represent all
effects in monetary units by using unit prices. A second approach is to apply
models that are generic for a geographical region. To illustrate: The result of
an LCA could be that in a particular region, PM10 emitted from a factory that
manufactures washing machines will become diluted with resulting air con-
centration around 0.1 µg m–3. By using dose-exposure models, this finding
could translate into a probability of 0.009 per year that a person in the region
acquires respiratory symptoms. A single washing machine would then be
responsible for part of this risk.
    In some industries, such as nuclear energy, the costs of factory scrapping
and site remediation are substantial. More than 30 years ago, such costs were
seldom included in LCA; industrial sites were simply abandoned. It is
estimated that there are about 1500 toxic waste sites in the US, which is
about 0.6 × 10–5 sites per capita.

4.3        Outside System Boundary Contributions

    To save time and efforts, the life cycle analyst may specify narrow LCA
boundaries, and take into account contributions from outside the boundary in
a more cursory way. She may for instance decide to consider only the 10
presumed highest cost items,18 or 0.1% of the environmental load. It is also
important to take into consideration that many products, like petrol, now
have added a “green” tax to its sales price. The environmental burdens of a
product from outside boundary activities, Iob, can be estimated by assigning a
proportion of countrywide total discharges, Itot, as they are reported in
official statistics.19 However, there are arguments that suggest that this
procedure introduces larger errors than by not using it.

      I OB = I Total × FOB × AV / TAV

    Here, AV is added value from the industry in the study – it can be found
in standard governmental statistics, FOB is the fraction of the industry’s added
58                                                                    Chapter 4

value from outside system boundaries, and TAV the total value added from
the main segments of the industry division.


    Having completed an IA, we perhaps include an LCA, and an
accompanying decision analysis to evaluate alternatives; the task now is to
convince public or private decision makers about the merits of the work. If
the decision makers accept the results, the work is finished.
    Very often they do have objections, however, and for various reasons.
They may find the consequence model inadequate; they may find the study
incomplete, or disagree about the choice of decision criteria. And that does
not complete the list. There are also at least five less tangible reasons the
decision makers may have for being reluctant.

1. Legitimacy: The legitimacy and the quality of the study may be
   questioned.20 Did those that judged the criteria weights have sufficient
   authority to do so?
2. Risk: The decision-maker goes for a project with less expected return
   than the one you recommend. This may be because risk may not have
   been dealt with adequately. If it is possible that things could turn out
   much worse than anticipated in the models, the risk should be made
   explicit. The standard way is to use risk-averse utility functions, which
   translates into a risk premium, meaning that the decision-maker is willing
   to sacrifice some the expected returns from a risky project and go for a
   less risky project with smaller expected returns. The size of the risk
   premium is a subjective matter, however, and depends on the feelings of
   the decision-maker.
3. Stakeholder conflict: The receiver of the benefits may not be the same as
   the one carrying the cost. Examples are long-range transport of
   pollutants, or a government that only looks to its own monetary income.21
4. Discount interest rate: The benefits may have been discounted according
   to an official interest rate, but the decision maker, in rejecting the project
   may unconsciously have used a higher rate. For example, many people
   would use a low discount rate to calculate the present value of future
   lives saved.22
5. The “dumb farmer” syndrome: This syndrome refers to a situation in
   which the environmental assessment presupposes that everything is
   constant except the direct consequences of the project. Decision-makers
   understand, however, that people normally will do things to improve the
   situation. Dumb farmers do nothing, but real farmers shift to other crops.
4. Integrated Assessment                                                     59


    Where should paper mills be sited to minimize adverse environmental
and socio-economic impacts of production, use, and disposal of pulp and
paper? Figure 4-4 shows a typical paper mill. The emissions in this case may
be water vapor, but for many mills there are significant chronic emission and
discharges, and for all mills accidental pollution incidents. A third concern is
the possible detrimental effects of unidentified chemicals.
    Below, we describe a study where two projects were compared, a paper
mill at a polluted site, and a paper mill in a pristine environment. It was
assumed that benefits are the same regardless of location; thus only the
externalities of the paper mill will affect the decision. 23

                              Figure 4-4. Paper mill.

6.1       Decision Analysis

   The impacts of paper production and use are not restricted to the site of
the paper mill, which may have a lifetime of 100 years. The forest where
wood is cut for pulp production and the waste dump or incineration plant,
may be somewhere else, are affected as well – and it also matters where the
paper finally ends up.
   Decision-maker and stakeholders: The decision maker is the paper
company’s board of directors, and that the Local Ministry of Environment,
NGOs and forest owners are important stakeholders (Chapter 3, section 2.3).
60                                                                               Chapter 4

                                  Minimize ecological
                                  and socio-economic

                                                           Land-use           Socio-
  Energy    Climate     Trans-         Air      Water
                                                           & ecology          economy
                                                          flora & fauna

              Figure 4-5. Generic goal hierarchy for an industrial product.

    If the paper mill were built in a developing country, forms of ownership –
tenure – may differ from that of developing countries, and discussion on how
to manage the site or the forest where the timber is cut may be conducted in
a discursive format where the weakest part is empowered (Chapter 9, section
    Decision objectives: Figure 4-5 shows a generic goal hierarchy, which
had previously been used in studies that involve LCA of paper products.24
End-point criteria were selected according to the main transportation
medium or “vehicle” bringing the pollutant to the affected target. In addition,
three categories not directly related to any vehicle, and one category that
summarizes transient effects related to the construction period, was added.
Thus, there were seven subgoals as shown in Figure 4-5.
    Adding decision criteria that measure the achievement of the subgoals
completes the goal hierarchy. Here is a list of the seven subgoals and 22 end
impact categories that need to be further developed into decision-criteria; the
details will depend on the particular setting:

1. Energy: Fraction of the world’s non-renewable energy resources used.
   Other effects of non-renewable energy use, like emissions of toxic
   substances, can be categorized according to the vehicle that dilutes and
   transports them, e.g., water and air.
2. Climate: Global warming potential, Ozone depletion potential.
3. Transient effects during the construction period can be aggregated and
   described with one criterion.
4. Air: Human mortality, Human sub-lethal effects, Noise, Odor, Corrosion
   on buildings and monuments, Winter fog.
5. Water: Eutrophication, Water use.
4. Integrated Assessment                                                   61

6. Land use and ecology: Vegetation, Wildlife, Endangered species, Storage
   of waste.
7. Socio-economic damage: Recreation.

   How to structure objectives and goals are discussed further in Chapter 3,
secton 3.
   Consequence analysis: To assess goal attainments for the two alternative
projects, one has to consider expected effects, how probable they are, and the
potential for natural or artificial restoration. The case study was based on a
study of paper mills in Norway and central Europe, and scores were assessed
for “typical” mills at the respective sites. LCA was used to describe
processes such as logging of wood, transportation, production of pulp and
paper, distribution of paper products and finally their disposal. Simple
models such as the box model described in this chapter, section 3.3, were
used to describe discharges into water and air. Gaussian dispersion models
described in Chapter 21, section 4.3 could be used to estimate dilution
caused by large chimneys. Emission of carcinogenic substances must be dealt
with through toxicity slope factor calculations, whereas non-carcinogenic
emissions can be dealt with through the hazard quotient method, both in
Chapter 14, section 4. Noise impacts could be described by estimating the
number of people exposed to noise above 55 dB(A), or one may use the
noise impact index in Chapter 10, section 4.4. Logging in forests may cause
5% to 10% of a natural forest to be in non-attainable state, (Chapter 19,
section 3). In Chapter 19, section 2.5 one will also find data to estimate
forest area required to support the paper mill with raw materials. Water
quality criteria are supplied in the chapters for the lake and river
environments respectively (Chapters 17 and 18).
   Preferences: The decision-maker determined the criteria weights, within
constraints imposed by environmental authorities, and based on legislation.
It may also be valuable to solicit the preferences of stakeholders with
methods described in Chapters 7 and 8.

6.2       What Happened?

    No single criterion appeared to dominate in the current case, in the sense
that it was judged much more important than the others, although previous
decisions often have emphasized closeness to timber resources and possi-
bilities for discharging process water. The consequence analysis showed that
end-point damage ranged from 3% to 23% of “worst case” damages for 9 out
of 12 decision criteria if sited in a pristine region. However, to compromise
the area of pristine environments of a region is a political decision and
62                                                                 Chapter 4

requires methods to solicit political judgments, such as the utility method or
the pair-wise comparison method (Chapter 7, section 2.2.).
    The external cost of paper production was estimated to range from 7% to
17% of the contribution to the gross domestic product of paper production,
this chapter Figure 4-1. Methane production at the dumpsites for paper
contributed the most.

7.         CONCLUSION

“Integrated assessment” is a term that describes how a decision problem is
embedded in a framework that includes all potential effects of the project.
However, the term “potential” indicates that not all effects are included in
the analysis, and good judgment has to be used when deciding what to
include and what to exclude. LCA is a major contribution to the science of
integrated assessment, with useful concepts like “functional” unit, and
“system boundary”. “Monetary unit price” for pollutants are only rough
estimates of the damage the pollutants cause in a given region, but they are
convenient when more detailed information is lacking.

  Based on Canter (1996) p. 24
  Davis and Wali (1994)
  Keeney (1992) p. 96
  Michaelis, Malmquist et al. (1997)
  Greer and Sels (1997)
  Horvath, Hendrickson et al. (1995)
  Horvath, Hendrickson et al. (1995)
  This method was used by Seip and coworkers (2000)
  Power and McCarty (1997)
   Seip, E. Sandersen et al. (1991)
   Folke, Aa. Janson et al. (1997)
   Data Renskaug (1998)
   Sima-Pro (2003)
   Carneigie-Mellon (2003) CM
   Guinée, Gorrée et al. (2002)
   Carnegie-Mellon (2003)
   Matthews, Lave et al. (2002)
   Matthews, Lave et al. (2002: 854)
   See Seip, Betele et al. (2000)
   Wenstøp and Seip (2001)
   Cifuentes (2000)
   Poulos and Whittington (2000)
   Seip et al. (2000)
   Seip, Betele et al. (2000)

Shared By:
Description: Comprehensive analysis of environmental impacts of projects