Content uploaded by Pablo Campra
Author content
All content in this area was uploaded by Pablo Campra
Content may be subject to copyright.
LAND USE IN LCA
Including CO
2
-emission equivalence of changes in land
surface albedo in life cycle assessment. Methodology and case
study on greenhouse agriculture
Ivan Muñoz &Pablo Campra &
Amadeo R. Fernández-Alba
Received: 3 November 2009 / Accepted: 1 June 2010 / Published online: 17 June 2010
#Springer-Verlag 2010
Abstract
Purpose Climate change impacts in life cycle assessment
(LCA) are usually assessed as the emissions of greenhouse
gases expressed with the global warming potential (GWP).
However, changes in surface albedo caused by land use
change can also contribute to change the Earth’s energy
budget. In this paper we present a methodology for including
in LCA the climatic impacts of land surface albedo changes,
measured as CO
2
-eq. emissions or emission offsets.
Methods A review of studies calculating radiative forcings
and CO
2
-equivalence of changes in surface albedo is carried
out. A methodology is proposed, and some methodological
issues arising from its application are discussed. The
methodology is applied in a practical example dealing with
greenhouse agriculture in Southern Spain.
Results The results of the case study show that the increase
in surface albedo due to the reflective plastic cover of
greenhouses involves an important CO
2
-eq. emission offset,
which reduces the net GWP-100 of tomato production from
303 to 168 kg CO
2
-eq. per ton tomato when a 50-year
service time is considered for the agricultural activity. This
example shows that albedo effects can be very important in
a product system when land use plays an important role,
and substantial changes in surface albedo are involved.
Conclusions Although the method presented in this work
can be improved concerning the calculation of radiative
forcing, it constitutes a first operative approach which can
be used to develop regionalized characterization factors and
provide a more complete evaluation of impacts on the
climate change impact category.
Keywords Climate change .Global warming potential
(GWP) .Greenhouse agriculture .Land transformation .
Land use change .Life cycle impact assessment (LCIA) .
Radiative forcing
1 Introduction
Climate change is among the most established impact
categories in life cycle assessment (LCA) (Udo de Haes et
al. 1999). Currently, impacts of products and services on
the global climate can be measured by LCA practitioners
with several approaches, such as the global warming
potential (GWP; Forster et al. 2007), which measures the
radiative forcing per unit of emission of different green-
house gases. GWP is probably the most generally used
method, although other approaches exist which go further
in the effects chain, measuring potential consequences on
humans and ecosystems (De Schryver et al. 2009; Steen
Responsible editor: Llorenc Milà i Canals.
Electronic supplementary material The online version of this article
(doi:10.1007/s11367-010-0202-5) contains supplementary material,
which is available to authorized users.
I. Muñoz (*):P. Campra :A. R. Fernández-Alba
Department of Hydrogeology and Analytical Chemistry,
University of Almería,
Ctra. de Sacramento s/n, La Cañada de San Urbano,
04120 Almería, Spain
e-mail: ivanmuno@ual.es
A. R. Fernández-Alba
Instituto Madrileño de Estudios Avanzados, IMDEA Agua,
C/Punto Net 4, Edificio ZYE,
Parque Científico Tecnológico de la U. de Alcalá,
Alcalá de Henares 28805 Madrid, Spain
Present Address:
I. Muñoz
Safety & Environmental Assurance Centre, Unilever,
Colworth Park,
Sharnbrook, Bedfordshire MK44 1LQ, UK
Int J Life Cycle Assess (2010) 15:672–681
DOI 10.1007/s11367-010-0202-5
1999a,b). A common feature of all approaches is the fact
that they only focus on emissions of greenhouse gases.
However, anthropogenic changes to the land cover can
affect surface albedo and exert a radiative forcing by
perturbing the shortwave radiation budget (Ramaswamy et
al. 2001). According to the Intergovernmental Panel on
Climate Change, in the 1750–2005 period global land cover
changes—especially deforestation—have increased the ter-
restrial albedo, resulting in a radiative forcing (RF) of
−0.2 W m
−2
(Forster et al. 2007). Even though this
influence might appear small at the global level (radiative
forcing from long-lived greenhouse gases is +2.63 W m
−2
),
in recent years the implications of surface albedo changes
have been gaining attention, especially as a climate change
mitigation strategy (Hamwey 2007; Ridgwell et al. 2009).
For example, a global increase of albedo in urban areas, by
means of using reflective building materials, has been
estimated to have a cooling effect equivalent to offsetting
44 Gt CO
2
(Akbari et al. 2009). On the other hand, Betts
(2000) found that reforestation in high latitudes could be
detrimental in terms of climate change mitigation, since the
positive forcing induced by forest albedo can offset the
negative forcing expected from carbon sequestration.
Inclusion of land use impacts in LCA is also gaining
attention, since land as a resource can be especially
important in agricultural, forestry, and mining products.
Actually, this rising interest can be easily illustrated by the
establishment in this journal of a specific “Land use’
subject (Milà i Canals 2007). To date, research on land use
impacts has been mainly focused on biodiversity and soil
quality indicators Milà i Canals et al. (2006), whereas the
only link made between land use and climate change
corresponds to the alteration of carbon stocks, by such
processes as converting forest to agricultural land
(Cherubini et al. 2009; Silalertruksa et al. 2009), but no
attempt has been made so far to tackle the subject of albedo
change in the context of LCA. In this paper we present an
approach for LCA to include albedo changes from land
cover in the climate change impact category, measuring
them as CO
2
-eq. emissions. The method presented is tested
in a practical example on greenhouse agriculture.
2 Fundamentals and review of existing methods
2.1 Relationship between surface albedo
and top-of-atmosphere radiative forcing
The amount of shortwave energy reaching the top of the
atmosphere (TOA), averaged over the entire planet, has been
estimated as 341 W m
−2
by Trenberth et al. (2009;Fig.1). Of
this amount, 79 W m
−2
is reflected back to space due to
clouds and aerosols, and 78 W m
−2
is absorbed by the
atmosphere. The remaining 184 W m
−2
reaches the Earth’s
surface, where 23 W m
−2
is reflected to space and
161 W m
−2
are absorbed. Therefore, the average surface
albedo is 23/184=0.13, whereas the TOA albedo is 102/341=
0.3. According to Le Treut et al. (2007), changing the
atmospheric and/or the surface albedo constitutes one of the
fundamental ways to disturb the radiation balance of the Earth.
Ramaswamy et al. (2001) defined RF as the change in
net (down minus up) irradiance (solar plus longwave; in
watts per square meter) at the tropopause. In broad terms, it
describes any imbalance in the planet’s radiation budget
caused by human interventions. Once RF is applied, the
climate system tends to adjust to recover equilibrium,
usually by means of changes in temperature (Forster et al.
2007). For most shortwave forcing agents, the instanta-
neous RF at the TOA is linked to surface temperature
change and can be used instead of the stratospheric-
adjusted RF at the tropopause (Forster et al. 2007). The
instantaneous TOA RF (in watts per square meter) is given
by Eq. 1:
RFTOA¼RTOA
Δ
apð1Þ
where R
TOA
is downward solar radiation at the TOA and Δa
p
is a variation in planetary albedo. R
TOA
is basically a
function of latitude (see Electronic Supplementary Material).
Concerning Δa
p
, according to Lenton and Vaughan (2009),
changes in surface albedo can be linearly related to changes
in a
p
(Eq. 2):
Δ
ap¼fa
Δ
asð2Þ
where Δa
s
is a variation in surface albedo and f
a
is a
parameter accounting for absorption and reflection of solar
radiation throughout the atmosphere. Under clear-sky con-
ditions, f
a
has an approximate global mean value 0.73 (Chen
and Ohring, 1985), representative of regions with very low
cloud cover, like deserts. On the other hand, lower f
a
values
Incoming solar
radiation at the top of
the atmosphere (TOA)
341
79 78
161
23
102
Absorbed by
atmosphere
Reflected by
clouds and
atmosphere
TOA reflection
Reflected
by surface
Absorbed by
surface
Fig. 1 Global mean shortwave energy flows in watts per square
meter. Absorbed incoming shortwave radiation is balanced by
releasing the same amount of outgoing longwave radiation. Source:
Trenberth et al. (2009)
Int J Life Cycle Assess (2010) 15:672–681 673
are representative of cloudy skies. Lenton and Vaughan
(2009)estimatef
a
for cloudy skies with Eq. 3:
fa¼Rs
RTOA
Tað3Þ
where R
s
is downward solar radiation at the Earth’ssurface
(in watts per square meter) and T
a
is an atmospheric
transmittance factor expressing the fraction of the radiation
reflected from the surface that reaches the TOA. Using a
global value of T
a
=0.854 and the global incident radiation at
TOA at the Earth’ssurface,LentonandVaughanobtaina
global f
a
value of 0.48. However, Eq. 3canbeusedto
calculate site-specific f
a
values where R
s
is known. As a
consequence, combining Eqs. 1,2,and3, we can obtain the
following expressions to estimate RF
TOA
as a function of
surface albedo changes:
RFTOA¼RsTa
Δ
asð4Þ
The method we present in this paper is based on RF
TOA
estimation using Eq. 4. Nevertheless, in the Electronic
Supplementary Material, we show that RF
TOA
can also be
estimated with Eqs. 1and 2.
2.2 CO
2
-equivalence of changes in surface albedo
The concept of RF allows us to compare modifications of
the Earth’s energy budget exerted by greenhouse gases,
with those caused by alterations due to changes in albedo.
However, policies like the Kyoto Protocol address climate
change mitigation targets in terms of greenhouse gas
emission reductions. For this reason, several authors
interested in the implications of changes in surface albedo
have developed calculation methods to express those
changes as CO
2
-eq. emissions or emission offsettings.
Betts (2000) developed a methodology aiming to express
the forcings from forest sequestration and albedo. Specifically,
he aimed at determining the change in terrestrial carbon stock
that would be equivalent to a change in surface albedo
resulting from a transition from agricultural land to forest land
in several regions. He simulated changes in albedo and
associated RF and then calculated the change in atmospheric
CO
2
concentration (ΔC) which would give the same forcing,
by means of Eq. 5, taken from Myhre et al. (1998):
RF ¼5:35 ln 1 þ
Δ
CC
0
=ðÞ ð5Þ
where C
0
is the 1997 global CO
2
concentration. ΔCis
converted to a terrestrial carbon stock change ΔC
T
by means
of Eq. 6:
Δ
CT¼2McMa
=
ðÞmair
Δ
CC
0
=Þð6Þ
where M
c
and M
a
are the molecular masses of carbon and dry
air and m
air
is the mass of the atmosphere. The factor of 2
accounts for an average airborne fraction of 0.5, taken from
Schimel et al. (1995). Including the airborne fraction is
necessary in the calculations since a fraction of emitted CO
2
does not remain in the atmosphere but dissolves in ocean
water and reacts with CaCO
3
in the sea floor, among other
processes. ΔC
T
was called by Betts “emissions equivalent of
the shortwave forcing”and can be easily converted to CO
2
emissions multiplying it by the molecular weight ratio of CO
2
to Cof 44/12=3.67.
Akbari et al. (2009) carried out an assessment of the
cooling potential at the planetary scale of increasing albedo
by 0.1 in urban areas. They estimated CO
2
-emission offsets
by first calculating with Eq. 5the RF of a marginal increase
of 0.128 ppmv in atmospheric CO
2
, resulting in a forcing of
+0.91 W kg CO
2
−1
. They calculated that a 0.01 increase in
the Earth’s surface albedo exerts a mean global forcing of -
1.27 W m
−2
. With these data, they concluded that
increasing albedo by 0.01 is equivalent to offsetting 1.27/
0.91=1.4 kg CO
2
m
−2
. However, this figure refers to
changes in atmospheric CO
2
. Similarly to Betts (2000),
they consider an average CO
2
airborne fraction of 0.55
(Denman et al. 2007); thus, the emission offset is 1.4/0.55 =
2.55 kg CO
2
m
−2
when surface albedo is increased by 0.01.
Bird et al. (2008) attempted to model the climate change
effects of afforestation/reforestation projects, by comparing
the RF due to carbon sequestration and to changes in land
use from grasslands to forest in various locations and forest
types in Canada. Both effects were expressed as CO
2
-eq.
emissions. They developed a set of equations describing
changes in TOA albedo, radiative forcing, and CO
2
equivalence of albedo change. This method differs from
that by Akbari et al. (2009) in the fact that local incident
radiation is used instead of global values, thus discriminat-
ing the CO
2
equivalence of albedo change in different
locations. Another difference of this approach with regard
to Akbari et al. (2009)andBetts(2000) is the conversion of
atmospheric CO
2
to emitted CO
2
, by the so-called airborne
fraction. As we have seen, Akbari et al. (2009) and Betts
(2000) deal with this by taking into account an average
airborne fraction of 0.55 and 0.5, respectively. These figures
are based on the observed constant relationship between
global CO
2
emissions and atmospheric concentration since
1958 (Schimel et al. 1995;Denmanetal.2007). On the other
hand, Bird et al. (2008) do not take into account in their
model a fixed airborne fraction, but a time-dependent
relationship. This is justified by the fact that the airborne
fraction of an instantaneous release of CO
2
decays over time.
For relatively small perturbations, it can be approximated
from the Bern carbon cycle model (Joos et al. 2001):
fðtÞ¼0:217 þ0:259e
Δ
t172:9
=þ0:338e
Δ
t18:51
=
þ0:186e
Δ
t1:186
=ð7Þ
674 Int J Life Cycle Assess (2010) 15:672–681
According to this model, after 10 years 66% of the initial
emission remains in the atmosphere, while only 36%
remains after 100 years. As a consequence, the choice of
a time horizon affects the magnitude of the CO
2
-eq.
emissions. For a time horizon of 100 years, usually used
in the calculation of GWP, the average airborne fraction,
calculated as the integral of Eq. 7from year 0 to 99, is 0.48,
quite close to those used by Betts (2000) and Akbari et al.
(2009).
3 Determination of the radiative parameters
3.1 Downward solar radiation at the Earth’s surface
For a particular site, average annual R
s
can be either
experimentally measured with a pyranometer for a repre-
sentative period of time or calculated from available
statistics from the closest meteorological station. It is also
possible to take advantage of existing tools and databases
which have been developed to determine this parameter for
the assessment of solar energy potential in different regions.
An example of this is the Photovoltaic Geographical
Information System (EC 2008).
3.2 Surface albedo
We can distinguish between empirical and modeling
approaches for the determination of a
s
. Empirical
approaches include remote sensing and field measurements.
Concerning remote sensing, data from the Moderate
Resolution Imaging Spectrometer (MODIS) are particularly
useful. MODIS is a radiometer operated aboard the NASA
Earth Observing System Terra and Aqua spacecrafts. It
collects data over a broad spectral range from the visible to
longwave infrared (Xiong et al. 2009). MODIS provides
measurements of instantaneous land surface reflectivity,
and daily mean and annual averages must be estimated
from representative data series. However, MODIS data
have a resolution of 500 m; hence, when the focus is on
small land parcels, field measurements are preferred. The
latter can be made by means of an albedometer, which
essentially consists of a combination of two pyranometers,
one facing upward and one facing downward.
Changes in surface albedo can also be estimated by
means of modeling techniques. Yin (1998) proposed a
model for the analysis and projection of albedo in vegetated
land surfaces. Models for simulation of albedo in urban
areas have also been developed, such as that by Chimklai et
al. (2004), taking into account the building height distribu-
tion, solar positions, the effects of multiple reflections and
shading. Comprehensive overviews of albedo for various
vegetation types, land covers, and materials were published
by Kondratyev (1969,1972), Iqbal (1983), Gates (1980),
and Breuer et al. (2003).
4 A framework for considering surface albedo changes
in LCA
A special feature of LCA as an environmental assessment
tool is the fact that it focuses on product systems, the
environmental burdens of which are allocated to so-called
functional units, representing a quantitative measure of the
function delivered by the product system. As a conse-
quence, CO
2
-eq. emissions from surface albedo changes
need to be attributed to a product system and functional
unit. Figure 2shows a simplified representation of albedo
changes in land cover albedo for two product systems P
1
and P
2
with two land use types, LU
1
and LU
2
, which have
surface albedo values a
sLU1
and a
sLU2
, respectively, where
a
sLU1
<a
sLU2
. For simplicity, we assume albedo to be
constant in each type of use. It is important to highlight at
this point the difference between land occupation and land
transformation: Land occupation refers to using a land area
during a certain amount of time, assuming no transforma-
tion of the land properties during this use (Lindeijer et al.
2002; Milà i Canals et al. 2007a); land occupation is
measured as the product of surface and time (square meter
year). In Fig. 1land occupation for P
1
starts at t
1
and
finishes at t
2
. On the other hand, land transformation
implies changing the properties of a land area according to
the requirements of a given new type of use (Lindeijer et al.
2002; Milà i Canals et al. 2007a); land transformation is
measured in surface units (square meter). In Fig. 2there is a
land transformation process when P
1
starts and a new one
when P
2
starts. Radiative forcings exerted by changes in
albedo are related to land transformation rather than to land
occupation. In Fig. 2when LU
1
is changed to LU
2
, albedo
LU1 LU1
LU2
t1
t0 t2
s LU1
s LU2
P1 P2
LT1 LT2
RFLT < 0
CO2-eqLT < 0
1
1
RFLT > 0
CO2-eqLT > 0
2
2
Time
Surface albedo ( s)
Fig. 2 Conceptual representation of surface albedo change in two
product systems. LU land use, LT land transformation, RF radiative
forcing, Pproduct system, CO
2
-eq. carbon dioxide equivalent
emissions
Int J Life Cycle Assess (2010) 15:672–681 675
increases, inducing a negative RF that can be expressed as a
CO
2
-eq. emission offset. It must be stressed that this offset
is a result of changing the albedo, regardless of the duration
of LU
2
. Subsequently, when activity P
1
finishes and P
2
starts we have again a land transformation, from LU
2
to
LU
1
, inducing a positive RF that can be expressed as a
CO
2
-eq. emission. Assuming that P
1
and P
2
use the same
amount of land, then the CO
2
-eq. offset by the start of P
1
and the CO
2
-eq. emission by the start of P
2
counterbalance
each other.
This example raises at least two methodological ques-
tions concerning how to allocate CO
2
-eq. from changes in
surface albedo in LCA studies: (1) allocation to a given
product system and (2) allocation to a functional unit.
4.1 Allocation to a product system
As we have seen from the example in Fig. 2,P
1
involves
transforming the land to a more reflective type of land
cover, and this can be expressed as a CO
2
-eq. emission
saving, analogous to a carbon sequestration. However,
when the activity of P
1
finishes, land is changed again by
P
2
to its original state before P
1
so that the environmental
achievement by P
1
is canceled. The question here is if P
1
should then be allocated a CO
2
-eq. offset. If we analyze
Fig. 2with a focus on land as a whole system in the t
2
–t
0
period, the net environmental benefit is zero. However,
LCA deals with product systems; thus, there is a need to
allocate transformation interventions in Fig. 2to P
1
and P
2
.
Using causality as guiding principle, we suggest that land
transformed by a use of the land for new purposes should
be attributed to this future new use. In such a case, P
1
should receive an environmental credit due to the increase
in surface albedo from a
s LU1
to a
sLU2
, whereas P
2
should
receive an environmental burden caused by decreasing
surface albedo from a
sLU2
to a
sLU1
. This allocation
principle is in accordance with current practice in the LCI
database ecoinvent (Frischknecht and Jungbluth 2007).
4.2 Allocation to a functional unit
The second question is how to allocate a CO
2
-eq. emission
or emission offset within a product system to a functional
unit. Changes in albedo are attributed to land transforma-
tion; hence, they constitute one-time interventions, just as
manufacturing capital equipment (machinery, buildings) or
clearing a forest before an agricultural activity. The only
way of allocating one-time interventions to a functional unit
is to assume an expected lifetime for the affected activity,
which can be uncertain. In economics, for instance, this is
dealt with by means of the depreciation concept. The
subject of one-time interventions or preparatory processes
in the context of land use in LCA has already been debated
in the past (see Udo de Haes 2006; Milà i Canals et al.
2007b). The problem of allocating climate burdens from
land transformation to a functional unit is not different
from, for instance, allocating the manufacture of a tractor to
an agricultural product. As Milà i Canals et al. (2007b)
point out, there is no scientific way to predict the future of
markets, and a “clear”allocation of preparative interven-
tions to the future years of the created structure has to be
based on societal agreements to avoid arbitrariness. As they
also point out, an exclusion of preparatory processes from
LCA, due to their less direct allocability to the product or
service, would seriously jeopardize the usability of LCA as
a decision tool. The latter is supported by Frischknecht et
al. (2007) who assessed the influence of capital goods in
the environmental profile of hundreds of datasets from the
ecoinvent database. This does not necessarily mean that
changes in surface albedo have the same influence in LCA
studies than capital goods, but we want to stress the fact
that they do not involve a particularly special type of
allocation problem. Some examples of how land use change
can be dealt with are provided by the carbon footprinting
methodology according to the British PAS 2050 standard
(BSI 2008) and by the European Directive on energy from
renewable sources (European Union 2009), in which
emissions from land use change must be taken into account,
distributing them over the functional unit during the first
20 years after land was changed.
4.3 Mathematical expression for CO
2
-equivalence
of changes in surface albedo and characterization factors
The derivation of a general expression for CO
2
-eq.
emissions from surface albedo changes is based on previous
work by Bird et al. (2008). The reader is referred to that
work for further details. Based on Betts (2000), Bird et al.
(2008) express CO
2
-eq. emissions (in grams) from surface
albedo changes as
CO2eq:¼ARFTOA ln 2pCo2;ref MCO2mair
AEarh
Δ
F2X Mair AF ð8Þ
where Ais the area affected by the change in surface albedo
(in square meter), RF
TOA
is measured in watts per square
meter, pCO
2,ref
is a reference partial CO
2
pressure in the
atmosphere (383 ppmv), M
CO2
is the molecular weight of
CO
2
(44.01 g mol
−1
), m
air
is 5.148×10
15
Mg, A
Earth
is the
area of the Earth (5.1×10
14
m
2
), ΔF
2X
is the radiative
forcing resulting from a doubling of current CO
2
concen-
tration in the atmosphere (+3.7 W m
−2
), M
air
is the
molecular weight of dry air (28.95 g mol
−1
), and AF is
the average CO
2
airborne fraction. All the variables taking
constant values in Eq. 8can be grouped in a single
parameter, the value of which is 1,101 W g CO
2
−1
. The
676 Int J Life Cycle Assess (2010) 15:672–681
inverse of this constant is the marginal RF of CO
2
emissions at the current atmospheric concentration, which
we will express in kilograms (0.908 W kg CO
2
−1
). Substi-
tuting in Eq. 8, we obtain Eq. 9:
CO2eq:¼ARFTOA
RFCO2AF ð9Þ
In this equation, CO
2
-eq. emissions are expressed in
kilograms. Now we call Aas the land transformation per
functional unit (LT
FU
) and substitute RF
TOA
by means of
Eq. 4:
CO2eq:¼LTFU RsTa
Δ
as
RFCO2AF ð10Þ
Substituting Δa
s
by an initial and final surface albedo
values, a
sLU1
and a
sLU2
, we finally obtain Eq. 11:
CO2eq:¼LTFU RsTaasLU1 asLU2
ðÞ
RFCO2AF ð11Þ
The average AF is a function of time, since the airborne
fraction of an instantaneous release of CO
2
decays over
time. Given a time horizon of nyears, AF is calculated with
Eq. 12, where f(t) is the Bern carbon cycle model (Eq. 7):
AF ¼
R
n
1
fðtÞdt
nð12Þ
Equation 12 gives AF values of 0.69, 0.48, and 0.32 for
the usual time horizons considered in GWP of 20, 100, and
500 years, respectively.
From Eq. 11 we can derive characterization factors (ChF,
in kilograms CO
2
-eq. per square meter) for the initial and
final land uses, respectively:
ChFLU1 ¼þRsTaasLU1
RFCO2AF ð13Þ
ChFLU2 ¼RsTaasLU2
RFCO2AF ð14Þ
Positive and negative signs in Eqs. 11,13, and 14 are
used in such a way that negative CO
2
-eq. emissions are
obtained when albedo is increased, and vice versa. As it can
be seen, characterization factors, besides being albedo-
dependent, allow the user to define specific values for
locations receiving different solar radiation levels, as well
as for different time horizons. It must also be highlighted
that this approach can be used either for land trans-
formations taking place in the foreground system as well
as for those related to the background system, provided
that the background data include land transformation
interventions.
4.4 Uncertainty
The uncertainty in these calculations depend on the respective
uncertainties of the user-defined data (LT
FU
,a
sLU1
,a
sLU2
,and
R
s
) and of the default values for T
a
,RF
CO2
,andAF.An
approximate error of ±30% is associated with T
a
, based on the
range of values suggested by Lenton and Vaughan (2009),
whereas for RF
CO2
Akbari et al. (2009) suggest a ±10% error.
Concerning AF, the error is less than ±15% (Forster et al.
2007,p.211).Thus,anoverallerrorforCO
2
-eq. emissions
around ±35% should be expected, excluding the contribution
from user-defined parameters. The uncertainty of LT
FU
will
be associated with that of the inventory data used but also
with the definition of a service life for the activity under
study, as discussed in Section 4.2. On the other hand, R
s
can
be quantified with a low level of error, either with measure-
ments or with models. With regard to initial and final albedo,
substantial uncertainty can be expected if the values used do
not come from measurements; otherwise, they should be
substantially lower. In the following section, the overall
uncertainty of the CO
2
-eq. emissions calculation is provided
for plastic greenhouses in Almería.
5 Case study: horticultural production in Almería
As an example of application, a cradle-to-gate LCA of
intensive tomato production in the province of Almería
(southeastern Spain) is carried out. This region has
experienced from the 1970s a rapid development of
greenhouse horticulture. According to Sanjuan (2007),
almost 26,000 ha of land were covered by plastic green-
houses in 2007, and they currently increase at an average
rate of 500 ha year
−1
. The study focuses only on the climate
change impact category, using GWP-100 as characteriza-
tion model, taking into account carbon from both biogenic
and fossil sources. Nevertheless, values for GWP-20 and
GWP-500 are also calculated as a sensitivity analysis.
5.1 Inventory for the farming activity
Unfortunately, there are no published LCA studies for tomato
production in Almería. As a consequence, most of the data
used corresponds to the same process in Barcelona (Antón et
al. 2005), with a similar climate to that of Almería. It is
assumed for this example that the amount of inputs to the
farm in Barcelona is comparable to those in Almería.
Nevertheless, some processes related to soil preparation,
greenhouse maintenance, and water pumping are taken from
a study in this region (Muñoz et al. 2009). The following
processes are included in the inventory: change in biomass
carbon stocks due to clearing of land prior to the agricultural
activity, greenhouse infrastructure production, maintenance
Int J Life Cycle Assess (2010) 15:672–681 677
and disposal, soil preparation, carbon fixation by the crop,
fertilizers production and N
2
O emissions from their applica-
tion, water pumping, transport, and treatment of green waste.
Soil CO
2
emissions due to changes in soil organic matter
during the farming period are not included due to lack of
data. This can be considered as an important limitation of
this case study, since these emissions have been found to be
of the outmost importance in agricultural systems (Koerber
et al. 2009; Brandão et al. 2010). As background data, the
ecoinvent 2.0 database has been used (Swiss Centre for Life
Cycle Inventories 2008). Production of pesticides has been
excluded from the study, since their contribution outside
toxicity-related impact categories is very low (Antón et al.
2005). The detailed inventory data for the farming activity
are shown in the Electronic Supplementary Material.
5.2 Calculation of CO
2
-eq. emissions from surface albedo
change
The calculation is made using the following data: The
tomato yield considered is 12 kg m
−2
year
−1
(Antón 2004);
therefore, land occupation is 83 m
2
year t
−1
. As already
discussed, the CO
2
equivalence of albedo change is related
to land transformation. As a consequence, a time span for
the farming activity has to be defined. For this example we
choose a period of 50 years, resulting in a LT
FU
of
1.67 m
2
t
−1
. The annual mean incident solar radiation (R
s
)
in this area is 196 W m
−2
for the 2001–2005 period,
according to Campra et al. (2008). Concerning the surface
albedo values, a mean annual value of 0.19±0.02 was
observed for the replaced grassland and 0.4±0.06 for an
area fully covered by greenhouses (Campra et al. 2008). If
we use Eq. 15 with AF for 100 years, the resulting GWP-
100 is −134 kg CO
2
-eq. per ton tomato. The corresponding
results for GWP-20 and GWP-500 can be also calculated as
in Eq. 11, replacing 0.48 by 0.69 and 0.32, respectively.
CO2eq:¼1:67 196 0:854 0:19 –0:40ðÞ
0:908 0:48
¼134 kg CO2eq:ton1ð15Þ
The uncertainty involved in this calculation, excluding
the contribution from LT
FU
, is up to ±45%. The contribu-
tion of R
s
to this uncertainty is not included either, although
it is considered to be very small, given that the value used
comes from field measurements with a pyranometer
(Campra et al. 2008).
6 Results and discussion
In Table 1the CO
2
-eq. emissions associated to the cradle-
to-gate farming activities are summarized. As it can be
seen, the GWP-100 is 303 kg CO
2
-eq. per ton tomato,
which is reduced to 168 kg CO
2
-eq. per ton tomato if the
change in surface albedo is taken into account. The choice
of time horizon in the GWP affects the magnitude of the
albedo effect, being increased with longer time horizons
such as 500 years. These results show that the local
radiative forcing caused by the land cover change has a
remarkable offset effect on the overall greenhouse gas
balance of this particular product system, equivalent to 44%
of its emissions when GWP-100 is considered. Campra
et al. (2008) showed the first empirical evidence to support
that changes in surface albedo caused by the highly
reflective plastic cover in this area have led to a cooling
trend in surface temperature. However, the magnitude of
this effect, when measured as CO
2
-eq. emissions per unit
product, depends on the choice of a service lifetime, which
in this example was taken as 50 years. The emission offset
increases for shorter lifetimes, for example it increases from
134 to 269 kg CO
2
-eq. per ton tomato when a 25-year
lifetime is considered but decreases to 67 kg CO
2
-eq. per
ton tomato when it is expanded to 100 years. Therefore,
when the implications of changes in albedo are of such high
magnitude as in the system under study, the choice of this
lifetime is of the utmost importance. Nevertheless, it should
not always be expected that changes in albedo have such a
high influence in LCA studies. The example shown is a
very particular case in which a sharp increase in surface
albedo is caused by white greenhouses. Another particular
case where albedo could have an important influence in the
CO
2
-eq. emission balance is in the context of forestry or
any other system involving land use in high latitudes,
where long-lasting snow cover is affected in some way
(Betts 2000) or anywhere where the reflectance of land
cover materials is changed on purpose, as is the case of
buildings and urban areas (Akbari et al. 2009). These are
examples of product systems where land use plays a
significant role; in product systems where land use is only
a background issue, the influence of changes in surface
albedo is expected to be less important than that from
emissions of greenhouse gases.
Nonetheless, the presented method can be used to assess
both interventions in the foreground and the background
systems. In the latter case, a set of characterization factors
should be developed for different land use classes, although
a certain level of regionalization is needed due to several
reasons: The first is that characterization factors depend on
the local intensity of solar radiation, and the second is the
effect of snow: Even though a coniferous forest may have a
similar summer albedo in different locations, the presence
or absence of snow in winter would make a substantial
difference in the annual mean albedo for locations in, for
instance, southern and northern Europe. Regionalized
impact assessment methods have already been developed
678 Int J Life Cycle Assess (2010) 15:672–681
for other well-established impact categories like acidifica-
tion and eutrophication (Huijbregts et al. 2000), but this is
the first time that such a need is identified for climate
change, an impact category with site-independent charac-
terization factors. While greenhouse gases are assumed to
be well mixed and distributed in the atmosphere, regardless
of where they are emitted, changes in albedo involve effects
on the climate at a regional–local scale, in the area where
solar energy budget is changed.
This method constitutes a simple analytical approach to
assess the climate burdens from changes in land surface
albedo. Besides the allocation problems discussed in
Section 3, it has other limitations, such as the uncertainty
involved. In the presented case study, the calculations are
estimated to have an uncertainty of up to ±45%. Among the
most important factors contributing to this uncertainty is the
simple modeling of atmospheric transmittance (T
a
). Further
refinement of this parameter would require a more
sophisticated modeling, which should include local data
on cloud cover, as provided for example by the Interna-
tional Satellite Cloud Climatology Project (ISCCPD2).
Another limitation is the need to obtain local data on
surface albedos, either by means of field measurements or
remote sensing. Although it might be tempting to use
literature albedo values, it must be stressed that the CO
2
-eq.
emissions are very sensitive to small changes in albedo, and
general albedo values in the literature are sometimes given
as ranges with rather broad limits. For instance, according
to Taha et al. (1988) albedo for crops is in the 0.15–0.25
range, whereas for urban areas, it can be anything from 0.1
to 0.35. Finally, it is also important to consider that RF and
hence the GWP metric itself have their limitations (see
Forster et al. 2007, pp. 210–211), especially when the focus
is on climate impacts from land use change. For example,
deforestation in the tropics decreases evapotranspiration
rates and increases sensible heat fluxes, resulting in
regionally decreased precipitation and increased surface
temperature (Bala et al. 2007). These kinds of effects
cannot be quantified in terms of radiative forcing nor,
therefore, as GWP either. Some authors have proposed new
metrics to quantify land use disturbances on the climate,
such as the Regional Climate Change Potential by Pielke et
al. (2002). Nevertheless, direct comparison of land cover
change effects with greenhouse gas emissions remains a
challenge.
7 Conclusions
A method has been introduced to include in LCA studies
the radiative forcing exerted by changes in land surface
albedo, expressed as CO
2
-eq. emissions. This method uses
a simple analytical approach, based on previous work in the
field of climate geoengineering. Besides enabling the
assessment of foreground interventions on land, it can also
be used to assess background interventions, although this
will require the development of regionalized characteriza-
tion factors. A practical example of intensive tomato
cultivation under reflective plastic greenhouses in southern
Spain has shown that the effect of surface albedo changes
can have a very important influence in the climate change
impact category, provided that (1) land use plays an
important role in the system (such as in agriculture, forestry
and mining, buildings, and urban areas) and (2) substantial
changes in surface albedo are expected in the product
system.
This method raises some methodological problems in the
context of LCA, which have also been discussed. They are
related to the fact that the CO
2
-eq. emissions associated to
changes in surface albedo are a consequence of land
Process GWP-20 GWP-100 GWP-500
Change in biomass carbon stock
a
22 2
Carbon fixation by crop
a
−190 −190 −190
Greenhouse infrastructure 283 226 204
Soil preparation 6 6 6
Greenhouse maintenance <1 <1 <1
Fertilizers 94 93 65
N
2
O emissions 55 57 29
Greenhouse disposal 6 3 2
Water pumping 24 22 21
Green waste treatment
b
83 84 83
Overall emissions (a) 365 303 223
Change in surface albedo (b)−93 −134 −202
Net with albedo change (c=a+b) 272 168 21
Ratio (c)to(a) 75% 56% 9%
Table 1 GWP (kilogram
CO
2
-eq.) for growing 1,000 kg
tomatoes
a
Biogenic CO
2
b
Mostly biogenic CO
2
Int J Life Cycle Assess (2010) 15:672–681 679
transformation, not of land occupation. As a consequence,
these emissions are one-time interventions which have to be
allocated to the functional unit by means of an expected
service lifetime. Another problem arising from the land
transformation dependency is the fact that impacts are
reversible: If in the same product system albedo is changed
but returned to its final state at the end of the service
lifetime, the net albedo change, and thus, the CO
2
-eq.
emissions are zero. These methodological problems are
analogous to those from accounting changes in carbon
stocks in LCA.
Although the method presented can be improved
concerning the calculation of radiative forcing, it constitutes
a first operative approach for LCA to go beyond green-
house gas accounting and provide a more complete
evaluation of human contributions to climate change.
Acknowledgments The authors acknowledge the useful comments
made during the preparation of this paper by Neil Bird, from
Joanneum Research and Assumpció Antón from Institut de Recerca
i Tecnologia Agroalimentària (IRTA).
References
Akbari H, Menon S, Rosenfeld A (2009) Global cooling: increasing
world-wide urban albedos to offset CO
2
. Climatic Change 94(3–
4):275–286
Antón A (2004) Utilización del Análisis de Ciclo de Vida en la
evaluación del impacto ambiental del cultivo bajo invernadero
mediterráneo. Universitat Politècnica de Catalunya, Barcelona,
Doctoral Thesis
Antón A, Montero JI, Muñoz P, Castells F (2005) LCA and tomato
production in Mediterranean greenhouses. Int J Agric Resources,
Governance Ecol 4(2):102–112
Bala G, Caldeira K, Wickett M, Phillips TJ, Lobell DB, Delire C, Mirin
A (2007) Combined climate and carbon-cycle effects of large-
scale deforestation. P Natl Acad Sci USA 104(16):6550–6555
Betts RA (2000) Offset of the potential carbon sink from boreal
forestation by decreases in surface albedo. Nature 408(6809):
187–190
Bird DN, Kunda M, Mayer A, Schlamadinger B, Canella L, Johnston
M (2008) Incorporating changes in albedo in estimating the
climate mitigation benefits of land use change projects. Biogeosci
Disc 5(2):1511–1543
Brandão M, Milà i Canals Ll, Clift R (2010) Soil organic carbon changes
in the cultivation of energy crops: implications for GHG balances
and soil quality for use in LCA. Biomass Bioenerg (in press)
Breuer L, Eckhardt K, Frede HG (2003) Plant parameter values for
models in temperate climates. Ecol Model 169(2–3):237–293
BSI (2008) PAS 2050:2008. Specification for the assessment of the
life cycle greenhouse gas emissions of goods and services. ISBN
978 0 580 50978 0
Campra P, García M, Cantón Y, Palacios-Orueta A (2008) Surface
temperature cooling trends and negative radiative forcing due to
land use change toward greenhouse farming in southeastern
Spain. J Geophys Res 113:D18109
Chen TS, Ohring G (1985) On the relationship between clear-sky
planetary and surface albedos: a parameterization for simple
energy balance climate models. Adv Space Res 5(6):141–144
Cherubini F, Bird ND, Cowie A, Jungmeier G, Schlamadinger B,
Woess-Gallasch S (2009) Energy- and greenhouse gas-based
LCA of biofuel and bioenergy systems: key issues, ranges and
recommendations. Resour Conserv Recy 53(8):434–447
Chimklai P, Hagishima A, Tanimoto J (2004) A computer system to
support Albedo calculation in urban areas. Build Environ 39
(10):1213–1221
De Schryver AM, Brakkee KW, Goedkoop MJ, Huijbregts MAJ
(2009) Characterization factors for global warming in life cycle
assessment based on damages to humans and ecosystems.
Environ Sci Technol 43(6):1689–1695
Denman KL, Brasseur G, Chidthaisong A, Ciais P, Cox PM et al
(2007) Couplings between changes in the climate system and
biogeochemistry. In: Solomon S et al (eds) Climate change 2007:
the physical science basis. Contribution of working group I to the
fourth assessment report of the intergovernmental panel on
climate change. Cambridge University Press, Cambridge, pp
499–588
EC (2008) Photovoltaic Geographical Information System (PVGIS).
http://re.jrc.ec.europa.eu/pvgis/index.htm. Accessed 22 Oct 2009
European Union (2009) Directive 2009/28/EC of the European
Parliament and of the Council of 23 April 2009 on the promotion
of the use of energy from renewable sources and amending and
subsequently repealing Directives 2001/77/EC and 2003/30/EC.
Official Journal of the European Union L-140: 16–62
Forster P, Ramaswamy V, Artaxo P, Berntsen T, Betts R et al (2007)
Changes in atmospheric constituents and in radiative forcing. In:
Solomon S et al (eds) Climate change 2007: the physical science
basis. Contribution of working group I to the fourth assessment
report of the intergovernmental panel on climate change. Cam-
bridge University Press, Cambridge, pp 129–234
Frischknecht R, Jungbluth N (eds.) (2007) Overview and methodol-
ogy. Final report ecoinvent, v2.0 no. 1. Swiss Centre for Life
Cycle Inventories, Duebendorf.
Frischknecht R, Althaus HJ, Bauer C et al (2007) The environmental
relevance of capital goods in life cycle assessments of products
and services. Int J Life Cycle Asses 12(1):7–17
Gates DM (1980) Biophysical ecology. Springer, New York
Hamwey RM (2007) Active amplification of the terrestrial albedo to
mitigate climate change: an exploratory study. Mitig Adapt
Strategies Glob Chang 12(4):419–439
Huijbregts MAJ, Schopp W, Verkuijlen E, Heijungs R, Reijnders L
(2000) Spatially explicit characterization of acidifying and
eutrophying air pollution in life-cycle assessment. J Ind Ecol
4:75–92
Iqbal M (1983) An introduction to solar radiation. Academia Press,
Toronto
Joos F, Colin Prentice I, Sitch S, Meyer R, Hooss G, Plattner GK,
Gerber S, Hasselmann K (2001) Global warming feedbacks on
terrestrial carbon uptake under the Intergovernmental Panel on
Climate Change (IPCC) emission scenarios. Glob Biogeochem
Cycles 15:891–908
Koerber GR, Edwards-Jones G, Hill PW, Milà i Canals Ll, Nyeko P,
York EH, Jones DL (2009) Geographical variation in carbon
dioxide fluxes from soils in agro-ecosystems and its implications
for life-cycle assessment. J Appl Ecol 46(2):306–314
Kondratyev KY (1969) Radiation in the atmosphere. Academia Press,
New York
Kondratyev KY (1972) Radiation processes in the atmosphere. World
Meteorological Organization, Geneva
Le Treut H, Somerville R, Cubasch U, Ding Y, Mauritzen C et al
(2007) Historical overview of climate change. In: Solomon S et
al (eds) Climate change 2007: the physical science basis.
Contribution of working group I to the fourth assessment report
of the intergovernmental panel on climate change. Cambridge
University Press, Cambridge
680 Int J Life Cycle Assess (2010) 15:672–681
Lenton TM, Vaughan NE (2009) The radiative forcing potential of
different climate geoengineering options. Atmos Chem Phys
9:5539–5561
Lindeijer E, Müller-Wenk R, Steen B (2002) Impact assessment of
resources and land use. In: Udo de Haes HA, Finnveden G,
Goedkoop M, Hauschild M, Hertwich EG, Hofstetter P, Jolliet O,
Klöpffer W, Krewitt W, Lindeijer EW, Müller-Wenk R, Olsen SI,
Pennington DW, Potting J, Steen B (eds). Life cycle impact
assessment: striving towards best practice. SETAC, Pensacola, pp
11–64
Milà i Canals Ll, Clift R, Basson L, Hansen Y, Brandão M (2006)
Expert workshop on land use impacts in life cycle assessment
(LCA). Int J Life Cycle Assess 11(5):363–368
Milà i Canals Ll (2007) Land use in LCA: a new subject area and call
for papers. Int J Life Cycle Assess 12(1):1
Milà i Canals Ll, Bauer C, Depestele J, Dubreuil A, Freiermuth KRF
et al (2007a) Key elements in a framework for land use impact
assessment in LCA. Int J Life Cycle Assess 12(1):5–15
Milà i Canals Ll, Bauer C, Depestele J, Dubreuil A, Freiermuth KRF
et al (2007b) Key elements in a framework for land use impact
assessment in LCA. Int J Life Cycle Assess 12(1):2–4
Muñoz I, Gómez MM, Fernández-Alba AR (2009) Life Cycle
Assessment of biomass production in a Mediterranean green-
house using different water sources: groundwater, treated
wastewater and desalinated seawater. Agr Syst 103(1):1–9
Myhre G, Highwood EJ, Shine KP, Stordal F (1998) New estimates of
radiative forcing due to well mixed greenhouse gases. Geophys
Res Lett 25(14):2715–2718
Pielke RA, Marland G, Betts RA, Chase TN, Eastman JL, Niles JO,
Niyogi DDS, Running SW (2002) The influence of land-use
change and landscape dynamics on the climate system: relevance
to climate-change policy beyond the radiative effect of green-
house gases. Phil Trans R Soc A 360(1797):1705–1719
Ramaswamy V et al (2001) Radiative forcing of climate change. In:
Houghton JT et al (eds) Climate change 2001: the scientific basis.
Contribution of working group I to the third assessment report of
the intergovernmental panel on climate change. Cambridge
University Press, Cambridge, pp 349–416
Ridgwell A, Singarayer J, Hetherington A, Valdes P (2009) Tackling
regional climate change by leaf albedo bio-geoengineering. Curr
Biol 19(2):146–150
Sanjuan JF (2007) Deteccion de la superficie invernada en la
provincia de Almeria a traves de imágenes Aster. Fundacion
para la Investigacion Agraria de la Provincia de Almeria, Almeria
Schimel D, Alves D, Enting I, Heimann M et al (1995) CO
2
and the
carbon cycle. In: Houghton JT (ed) Climate change 1995. The
science of climate change, chapter 2. Cambridge University
Press, Cambridge, pp 65–131
Silalertruksa T, Gheewala SH, Sagisaka M (2009) Impacts of Thai
bio-ethanol policy target on land use and greenhouse gas
emissions. Appl Energ 86(suppl 1):S170–S177
Steen B (1999a) A systematic approach to Environmental Priority
Strategies in product development (EPS). Version 2000-general
system characteristics. Chalmers University, Göteborg, CPM report
Steen B (1999b) A systematic approach to Environmental Priority
Strategies in product development (EPS), Version 2000—models
and data of the default method. Chalmers University, Göteborg,
CPM report
Swiss Centre for Life Cycle Inventories (2008) The ecoinvent
database. http://www.ecoinvent.ch/. Accessed 6 Oct 2009
Taha H, Akbari H, Rosenfeld A, Huang J (1988) Residential cooling
loads and the urban heat island—the effects of albedo. Build
Environ 23(4):271–283
Trenberth KE, Fasullo JT, Kiehl JT (2009) Earth’s global energy
budget. B Am Meteorol Soc 90:311–323
Udo de Haes H (2006) How to approach land use in LCIA or, how to
avoid the Cinderella effect? Int J Life Cycle Assess 11(4):219–221
Udo de Haes HA, Jolliet O, Finnveden G, Hauschild M, Krewitt W,
Mueller-Wenk R (1999) Best available practice regarding impact
categories and category indicators in life cycle impact assess-
ment: Part 1. Int J Life Cycle Assess 4:66–74
Xiong X, Chiang K, Sun J, Barnes WL, Guenther B, Salomonson VV
(2009) NASA EOS Terra and Aqua MODIS on-orbit perfor-
mance. Adv Space Res 43(3):413–422
Yin X (1998) The albedo of vegetated land surfaces: systems analysis
and mathematical modeling. Theor Appl Climatol 60:121–140
Int J Life Cycle Assess (2010) 15:672–681 681
