Kaphengst 13 agriculture as provisioning ecosystem service_0

Study on role agriculture as provisioning ecosystem service
Page 1 of 103
STUDY ON THE ROLE OF
AGRICULTURE AS
PROVISIONING
ECOSYSTEM SERVICE
Framework contract nr 385309 on the
provision of expertise in the field of Agri-
Environment
FRAGARIA consortium
Final report

Page 2 of 103

Page 3 of 103
Final report
version 17 May 2012
Consortium
Alterra Wageningen UR, The Netherlands
Ecologic Institute, Germany
University of Copenhagen (Denmark)
Subcontractor
EuroCARE
Study on the role of agriculture as
provisioning ecosystem service

Page 4 of 103

Page 5 of 103
Administrative summary
Specific Contract No 3 implementing Framework Contract 385309
The specific contract Study on the role of agriculture as provisioning ecosystem service was
signed by the European Union represented by the European Commission, which is
represented for the purposes of the signature of this contract by Mr Guido Schmuck, Acting
Director of the Institute for Environment and Sustainability (JRC/IES) on 8 November 2011,
and by Mr. Kees Slingerland, Managing Director of Institute Alterra of Stichting
Landboukunding Onderzoek, representing the contractor, on 1 November 2011. Total
duration of the contract is maximum six months after the contract is signed by JRC/IES and
the consortium, thus the contract will end on 8 May 2012.
Consortium:
ALTERRA Wageningen UR, The Netherlands
Ecologic Institute, Germany
University of Copenhagen (Denmark)
Subcontractor:
EuroCARE, Germany
Institute for Environment and Sustainability, European Commission Joint Research Centre
Official Responsible:
M.L. Paracchini
Co-ordinating institution:
Alterra, Wageningen University and Research Centre
Person authorised to sign the contract on behalf of the consortium:
Ir. C.T. Slingerland, General Director of Alterra
Person authorised to manage the contract:
Dr. Marta Pérez-Soba
Persons responsible for administrative matters:
Elizabeth Rijksen and Petra van den Broek (Alterra WUR)
Contact information:
Dr. Marta Pérez-Soba
Alterra, P.O. Box 47; NL-6700 AA Wageningen, The Netherlands

Page 6 of 103
Report authors and contributors
Recommended citation:
Pérez-Soba, M., Elbersen, B., Kempen, M., Braat, L., Staristky, I., Wijngaart, R. van,
Kaphengst, T., Andersen, E., Germer, L. and Smith, L., der (2012). Study on the role of
agriculture as provisioning ecosystem service. Interim report to the Institute for
Environment and Sustainability (JRC/IES). Alterra Wageningen UR, Ecologic Institute,
University of Copenhagen and EuroCARE
Project contact:
Dr Marta Pérez-Soba
ALTERRA Wageningen University and Research Centre
P.O. Box 47
6700 AA Wageningen
The Netherlands
E-mail: marta.perezsoba@wur.nl
ACKNOWLEDGEMENTS
This study was funded by the European Commission. We thank the European Commission
Desk Officer, Maria Luisa Paracchini, and the JRC/IES post-doc Celia García Feced for their
helpful advice and guidance. We are especially grateful to Joost Wolf and Kees van Diepen
(Wageningen University and Research Centre) for their expert advise on the reference layers
for grasslands.

Page 7 of 103

Page 8 of 103
Table of contents
SUMMARY.................................................................................................................................10
1. Setting the scene of the study...........................................................................................11
2. Review of main concepts dealing with agricultural production as ecosystem service, and
specifically as provisioning ecosystem service .........................................................................15
3. Analytical framework for the quantitative assessment of the energy balance ................33
4. Results................................................................................................................................43
5. Discussion ..........................................................................................................................57
6. Conclusions and recommendations ..................................................................................60
7. References .........................................................................................................................63
Annex 1 Grouping of crop groups for presentation and analysis of final energy balance
calculations ...............................................................................................................................67
Annex 2 Calculating energy input for spreading manure.........................................................68
Annex 3 Calculating energy input through labour....................................................................69
Annex 4 Energy content of output of food, feed and other biomass ......................................71
Annex 5 Allocation of input and output variables from region to HSMU level........................74
Annex 6 Preparation of the three reference layers with the MARS-CGMS system.............76
Annex 7 Analysis of most suitable crop aggregates for presentation of results......................81
Annex 8 Land use, input and output information per country and environmental zone....82
Annex 9 Variation in input levels per crop ...........................................................................99

Page 9 of 103

Page 10 of 103
SUMMARY
Agro-ecosystems provide provisioning, regulating and cultural services to human society.
This study focuses on the agro-ecosystem provisioning services regarding food, feed, fibre
and fuel. These services strongly respond to the socio-economic demands of human beings,
but do not always consider the ecological demands of the ecosystem, i.e. the bio-physical
structure and processes that take place during the agricultural production. Therefore there
is no clear agreement within the policy and scientific communities on whether all types of
agricultural production should be seen as a provisioning ecosystem service and if so, how
the ecological-socio-economic flow linked to the provisioning service should be better
assessed. Several studies have provide qualitative assessments but no one, to our
knowledge, has done it in a quantitative way. This study makes an attempt to answer the
former questions by assessing quantitatively the degree of provisioning service by the agro-
ecosystems by considering their energy balance and their different bio-physical structures
and processes.
This Final Report presents the methodology and results obtained in this study, which was
developed from November 2011 until May 2012.
The work was divided in the following tasks:
 Task 1: Conceptual approach to define the way in which agricultural
production fits into the provisioning ecosystem services framework
 Task 2: Selection of suitable units of measure and reference for the analysis
 Task 3: Definition of analytical framework for the analysis
 Task 4: Mapping of crop production as provisioning ecosystem service
For every task a research objective was formulated as follows:
1) The conceptual development of the way in which agricultural production should fit
into the reference ecosystem services frames (Millennium Ecosystem Assessment,
TEEB), keeping into consideration that it is not a “pure” ecosystem service, but it is
originating from deeply modified habitats;
2) Identification of the most suitable unit of measure (i.e. biomass, energy, yield), and of
the reference against which analysing actual plant provision (potential productivity
of the land);
3) The definition of the analytical frame to address crop production as ecosystem
service, while taking into consideration external energy flows linked to agricultural
management (i.e. labour, machinery, fertilisers, irrigation)
4) Mapping of the ecosystem service, including reference productivity considering
different degrees of human intervention as external inputs.

Page 11 of 103
1. Setting the scene of the study
1.1 Background and outline
This report presents the final results from a study commissioned by the JRC to the agri-
environmental framework contract consortium FRAGARIA. The main aim of this study was
to design and implement a methodology to assess the provisioning service of agriculture in
the EU-27.
The work was divided in the following tasks:
 Task 1: Conceptual approach to define the way in which agricultural
production fits into the provisioning ecosystem services framework
 Task 2: Selection of suitable units of measure and reference for the analysis
 Task 3: Definition of analytical framework for the analysis
 Task 4: Mapping of crop production as provisioning ecosystem service
The work resulting from these four tasks is reported in the different chapters of this report.
1.2 Policy and thematic context of the study
The concept of ecosystems services (ES) has advanced significantly since late 1970s when it
was used primarily to explain societal dependence on nature. It currently incorporates
economic dimensions and it is starting to be used as support to decision makers for
implementing effective policies that support human wellbeing and sustainable
development.
The attention to ecosystem services, resource efficiency and natural capital in the European
Union has rapidly developed in the years 2010 and 2011, as result of the compelling
evidence that the in 2001 globally agreed target of stopping the loss of biodiversity by 2010
has not been met despite of substantial efforts in order to better protect nature. In contrast,
biodiversity, ecosystems and the services they provide continue to deteriorate. Many of the
pressures that affect habitats and species, including the conversion of ecosystems for other
purposes of land use, climate change, invasive species, fragmentation of the land, pollution
and overexploitation of biological resources, continue to impact biodiversity. The main
processes around the ecosystem service concept are:
 The EU's new biodiversity strategy (COM/2011/244 final) in May 2011, which marks
a major milestone in the operationalisation of the ES concept in EU policies. In 2010
the EC proposed a renewed vision and targets for biodiversity for the ensuing period,
building on and contributing to the international deliberations on a global vision for
biodiversity beyond 2010, which will be part of a revised and updated strategic plan
for the United Nations Convention on Biological Diversity (CBD) (European
Commission, 2010). The new target includes restoration of ecosystem services,
therefore a crucial step in its achievement is the provision of a first set of biophysical
maps of ecosystem services of key importance at the EU level. Among these
agricultural production plays a special role, since it is a provisioning service
characterised by a strong human influence. It considers measuring Europe's natural
capital and integrates for the first time in European policy the value of ecosystem

Page 12 of 103
services into policymaking: Biodiversity — the extraordinary variety of ecosystems,
species and genes that surround us is our life insurance, giving us food, fresh water
and clean air, shelter and medicine, mitigating natural disasters, pests and diseases
and contributes to regulating the climate. Biodiversity is also our natural capital,
delivering ecosystem services that underpin our economy. Its deterioration and loss
jeopardises the provision of these services: we lose species and habitats and the
wealth and employment we derive from nature, and endanger our own wellbeing.
This makes biodiversity loss the most critical global environmental threat alongside
climate change — and the two are inextricably linked;
 A resource-efficient Europe – Flagship initiative under the Europe 2020 Strategy in
January 2011 (COM/2011/21). This strategy is the seventh and last of the Europe
2020 flagship initiatives which aim at building smart, sustainable and inclusive
growth for Europe. It establishes resource efficiency as the guiding principle for EU
policies on energy, transport, climate change, industry, commodities, agriculture,
fisheries, biodiversity and regional development. European Commission President
José Manuel Barroso, who steered the launch of this initiative, said: "Continuing our
current patterns of resource use is not an option. They put too much pressure on our
planet and make our economy more dependent on external supplies. A smarter use
of scarce resources is therefore a strategic necessity, but also an economic
opportunity. Through more resource-efficiency, clearer long-term policies and joint
investments in green innovation, we are strengthening the basis for growth and jobs
for our citizens and delivering on our climate and energy objectives.”;
 Publication of The Economics of Ecosystems and Biodiversity (TEEB, 2010) reports;
 Publication of "the Atlas of ecosystem services" in September 2011 by the EU
executive's in-house research facility, the Joint Research Centre, which has started
the mapping of ecosystem services at EU level;
 The proposal for a Common International Standard for Ecosystem Services (CICES)
made to the United Nations Statistical Division (UNSD) in 2010, as part of the
revision of the System of Environmental-Economic Accounting (SEEA);
 Work carried out by the European Environment Agency (EEA) on ecosystem services
accounting in Europe;
 The centrepiece of the EU's current nature and biodiversity policy is the Natura 2000
network of protected areas. However, according to the EU executive, the future
ecosystem services cannot only be delivered only through such protected areas, and
the remaining 82% will need to be addressed as well. Therefore, investment in
natural capital in protected and non-protected ecosystems, which the Commission
refers to as "green infrastructure" is needed. The Commission is set to table a green
infrastructure initiative in late 2011.
All these policy initiatives respond to the recognition that most of the ecosystem services in
Europe are ‘degraded' — no longer able to deliver the optimal quality and quantity of basic
services, as shown in Figure 1.

Page 13 of 103
Figure 1: Status of ecosystem services in Europe (RUBICODE project 2006–2009; marine
ecosystems not included)
Agro-ecosystems provide provisioning, regulating and cultural services to human society.
The provisioning services of agriculture relate to the provision of crops and livestock, which
are considered as agricultural production.
There is significant evidence that most intensively managed agricultural systems produce
services in an unsustainable way, in which the natural capital resources are progressively
depleted at a high rate and not restored.
For example, the changes in natural habitats, which are mostly due to intensive agricultural
production systems, are one of the main causes of biodiversity loss and decrease in quality
and quantity of ecosystem services. In addition, 30 % of species are threatened by
overexploitation (IUCN).
It seems therefore urgent to assess how sustainable agricultural production systems have to
be in order to consider their productive outputs as provisioning ecosystem services.

Page 14 of 103
1.3 General approach
Based on the above given context and considerations, the overall objective of this study is
mapping the provisioning services delivered by EU agro-ecosystems at the highest possible
resolution, considering the net energy use of resources and the net economic benefits.
The approach needs to take into consideration that:
 agro-ecosystems result from strongly modified habitats;
 the focus will be on the direct use of the soil, as natural resource for plant production, and
therefore will exclude the indirect animal production;
 it will exclude as well crop production in greenhouses, which have a small share of the UAA,
use mainly artificial soil and have a very negative net energy balance. It also needs to be
assessed how and to which extent other horticultural and permanent crop activities need to
be included in the energy balance, like e.g. flower and vegetable production, olives and
vineyards. These types of agricultural production will either be included or conceptually
addressed in the report;
 the way to express the provisioning ecosystem services provided by agriculture should be as
net energy balance, and therefore the energy input and output in the production agro-
ecosystem need to be expressed in energy and biomass. The input will include labour,
machinery, fertilisers and irrigation as far as data are available. The consideration of labour
as input in terms of energy needs further discussion on how these can be included (e.g. in
person hours using different agricultural machines). The output will be measured as the
biomass related to the different crops yields;
 the comparison baseline against which actual food provision can be compared, will take
account of the full productive capacity of the soil including food and feed, but also additional
biomass production not necessarily being used by humans at this moment, such as biomass
for fibre and fuel production, as far as data are available;
 the approach targets the EU in terms of its practical implementation, but conceptually
should have a more general applicability , i.e. should be applicable in other continents as
well;
 the approach should also consider economic assumptions, which are inherent to the
valuation of the benefits provided by agricultural production to human well-being.
1.4 Outline of the report
This report consists of five chapters including this first introductory chapter. Chapter 2 deals
with Task 1 and provides a review of the main concepts dealing with ecosystem services,
specially focusing on production of agricultural ecosystems, and ends with a discussion on
how the soil energy balance can be used to describe the role of agricultural production as
provisioning ecosystem service. Chapter 3 deals with Tasks 2 and 3 and describes the
analytical framework designed in this study to measure the crop energy balance, including
an explanation of the translation of the methodology from the regional soil energy balance
towards a more spatially explicit energy model. This model calculates the energy balance at

Page 15 of 103
the level of Homogenous Spatial Mapping Unit (HSMU) taking into account the detailed soil
and other biophysical characteristics and diversity in farm management. Chapter 4 deals
with Task 4 and presents, mainly in maps, the results of applying this methodology. In
Chapter 5, the results are discussed and in Chapter 6 conclusions derived.
2. Review of main concepts dealing with agricultural production as
ecosystem service, and specifically as provisioning ecosystem service
2.1 Methodological approach
The overall questions to be tackled in the methodological approach are:
 how agricultural production fits into the existing concepts of ecosystem services?
More specifically, how the ecosystem services concepts deal with provisioning
services from agro-ecosystems, considering the fact that agricultural production
results from strongly modified ecosystems and generally needs external
anthropogenic energy inputs before a final output of biomass can be delivered;
 how the provisioning services provided by agro-ecosystems can be assessed in a
quantitative way?
The former questions are answered in the following Task 1 activities:
 a literature review dealing with the history of the concept of ecosystem services in
general, mainly highlighting how the scope and perspective on ecosystems and their
services to humans have changed over time and what implication this has for the
purpose of the study;
 an analysis and comparison of the different concepts of ecosystem services and
functions focusing on agricultural production. A specific emphasis is laid on the
relationship between provisioning and other ecosystem services such as regulating,
supporting and cultural services. This analysis should sharpen the view on the
current lack of a comprehensive framework to deal with agricultural production
within the concept of ecosystem services, which is widely applied in current
biodiversity and other EU policies (see section 1.2);
 building on this comparative analysis, a preliminary conceptual framework is
developed that considers the energy balance of the agricultural ecosystem as a
measure to describe the provisioning ecosystem services provided by agriculture. It
introduces the issue of trade-offs between provisioning and other ecosystem
services, and how this can be advanced in future analyses.

Page 16 of 103
2.2 Review of the evolution of concepts on ecosystem services and their linkage to
agriculture
Developing concepts of ecosystem services dates back to the late 1970s and has changed its
scope over time, culminating in the widely used definition by The Economics of Ecosystems
and Biodiversity (TEEB) in 2010.
The origins of the modern history of ecosystem services are to be found in the late 1970s
(see for an extensive history Gomez-Baggethun et al., 2010). It starts with the utilitarian
framing of those ecosystem functions, which were deemed beneficial to society, as services
in order to increase public interest in biodiversity conservation. It then continues in the
1990s with the mainstreaming of ecosystem services in the literature (Costanza and Daly,
1992; Daily, 1997), and with increased focus on methods to estimate their economic value
(Costanza et al., 1997). The Millennium Ecosystem Assessment (MA, 2005) did put
ecosystem services firmly on the policy agenda, and since its release the literature on
ecosystem services has grown exponentially (Fisher et al., 2009).
A series of theoretical divergences within the society of Environmental and Resource
Economics led to the formation of the society and journal of Ecological Economics (Costanza
et al., 1992). Ecological Economics conceptualizes the economic system as an open
subsystem of the ecosphere exchanging energy, materials and waste flows with the social
and ecological systems with which it co-evolves. The focus on market-driven efficiency
typical for Neoclassical economics is expanded to the issues of equity and scale in relation to
biophysical limits, and to the development of methods to account for the physical and social
costs involved in economic performance using monetary along with biophysical accounts
and other non-monetary valuation languages.
A major issue in the debate between Neo-Classical and Ecological Economists is the
sustainability concept. The so-called “weak sustainability” approach, which assumes
substitutability between natural and manufactured capital, has been mostly embraced by
Neoclassical environmental economists. Ecological Economics have generally advocated the
so-called “strong sustainability” approach which maintains that natural capital and
manufactured capital are in a relation of complementarity rather than of one of
substitutability (Costanza and Daly, 1992).
The concept of ecosystem services, introduced by Ehrlich and Ehrlich (1981) builds on earlier
literature highlighting the societal value of nature's functions. In ecology, the term
ecosystem function has traditionally been used to refer to the set of ecosystem processes
operating within an ecological system irrespective of whether or not such processes are
useful for humans. However, in the late 1960s and 1970s, a series of contributions started
referring to the way particular “functions of nature” served human societies (Helliwell,
1969; Hueting, 1970; Odum, 1971; Braat et al., 1979). In the 1970s and 1980s, a growing
number of authors started to frame ecological concerns in economic terms in order to stress
societal dependence on natural ecosystems and raise public interest on biodiversity
conservation. The rationale behind the use of the ecosystem service concept was mainly

Page 17 of 103
educational, and it aimed to demonstrate how the loss of biodiversity directly affects
ecosystem functions that underpin critical services for human well-being.
The paper by Costanza et al. (1997) on the value of the global natural capital and ecosystem
services was a milestone in the mainstreaming of ecosystem services. The monetary figures
presented resulted in a high impact in both science and policy making, manifested both in
terms of criticism and in the further increase in the development and use of monetary
valuation studies. In the late 1990s and early 2000s the concept of ecosystems services
slowly found its way into the policy arena, e.g., through the “Ecosystem Approach” (adopted
by the UNEP-CBD, 2000).
The Millennium Ecosystem Assessment (MA, 2005) constitutes a critical landmark that
firmly placed the ecosystem services concept in the policy agenda. While emphasizing an
anthropocentric approach, the MA framework stressed human dependency not only on
ecosystem services, but also on the underlying ecosystem functioning, contributing to make
visible the role of biodiversity and ecological processes in human well-being. Since the MA,
the literature on ecosystem services and international projects working with the concept
have multiplied (Fisher et al., 2009).
In the last few years several initiatives have framed global environmental problems in
economic terms and conducted global cost-benefit analysis. Some relevant examples are
the Stern Review on the Economics of Climate Change (Stern, 2006) and the Cost of Policy
Inaction study (Braat & Ten Brink, 2008). The project Economics of Ecosystems and
Biodiversity (www.teebweb.org), stemming and building on this initiative, has brought
ecosystem services now in the policy arena with a clear economic connotation. And with
increasing research on the monetary value of ecosystem services, the interest of policy
makers has turned to the design of market-based instruments to create economic incentives
for conservation, e.g. Payments for ecosystem services (PES).
In the TEEB concept, the “flow of value” is further divided in ecosystem functions providing
ecosystem services which directly or indirectly lead to benefits for humanity. The TEEB “flow
Text Box 1: Brief overview on the “evolution of Ecosystem Services concepts” since 1997
 Ecosystem Services are the conditions and processes through which natural ecosystems,
and the species that make them up, sustain and fulfil human life - Daily (1997).
 Ecosystem Services are the benefits human populations derive, directly or indirectly,
from ecosystem functions - Costanza et al. (1997).
 Ecosystem Services are the benefits people obtain from ecosystems – MA (2005).
 Ecosystem Services are components of nature, directly enjoyed, consumed, or used to
yield human wellbeing – Boyd & Banzhaf (2007).
 Ecosystem Services are the aspects of ecosystems utilized (actively or passively) to
produce human well-being – Fisher et al. (2009).
 The concept of ecosystem services refers to the flow of value to human societies as a
result of the state and quantity of natural capital – TEEB (2010b).

Page 18 of 103
of value” diagram is adapted in this study for the agro-ecosystems, to describe the role of
crop production as provisioning ecosystem service when focusing on the direct use of the
soil as natural resource and including the external anthropogenic input in the soil system
(see Fig. 2).
Figure 2: Adaptation of the TEEB “flow of value” elements (TEEB Foundations, 2010) for
agro-ecosystems in this study, i.e. biophysical structure or processes, functions, services,
benefits and values, when focusing on the provisioning services delivered by crop
production, and including the external inputs to the soil ecosystem by humans.
2.3 Comparing different concepts dealing with functions, services and benefits of
agricultural production
Given the historical debate in the context of the Common Agricultural Policy, services from
agricultural (or agro-) ecosystems in the EU usually have been associated with public goods
and services that agriculture delivers to society, in addition to the provision of those
considered as ‘private’ goods, e.g. food or biomass. However, to date, there is an apparent
lack of conceptual models and empirical evidence allowing for the valorisation of agriculture
as a public good for the maintenance of European landscapes and their ecological and socio-
economic functions (Paracchini et al. 2012). On the other hand, agricultural production is

Page 19 of 103
considered as a provisioning ecosystem service. In the following sub-sections, three
approaches describing the functions, services and benefits derived from agricultural
ecosystems to humans will be distinguished in terms of their perspectives and implications
for the description of agricultural production as provisioning ecosystem service.
(i) Ecosystem functions as the ecological basis for agricultural activities
The principal studies1
on benefits that humans can derive from nature have generally relied
on two main concepts: ecosystem services and ecological functions2
. Despite increasing
work on this subject, there is still some debate in the literature about how to define these
main terms.3
In the 1960’s and 1970’s the term “functions” combined with nature, environment and
natural environment was used to describe the “useful” (from an anthropocentric or
utilitarian point of view) properties of ecosystems for society or economic processes (a.o.
reviewed in Braat et al. 1979). Even though in most cases the “functions of nature” term
was explicitly distinguished from “ecological functions” much confusion remained. Ehrlich
and Ehrlich (1981) attempted to solve this problem when they coined the term “ecosystem
services”, which joins the ecological concept “ecosystem” (Tansley, 1935; Lindemann, 1942),
with the economic concept of services (used as short term for goods & services). But still for
more than two decades the concepts of ecosystem functions and ecosystem services co-
existed (De Groot, 1992; Daily, 1997) and prolonged the confusion, until the Millennium
Ecosystem Assessment (MA, 2005) forced a clear swing towards ecosystem services. In the
MA a widely supported definition and classification was published, which was the input for
the TEEB (The Economics of Ecosystems and Biodiversity) work in 2010. In the TEEB reports,
a major change in the classification of the MA was introduced by taking out the group of
Supporting Services, which in the MA diagrams were correctly positioned “behind” the
other groups (provisioning, regulating and cultural), because it would cause double-counting
in economic assessments and essentially referred to dynamics in ecosystems which
ecologists had been calling “functions”. “It is helpful to distinguish ‘functions’ from the even
deeper ecological structures and processes in the sense that they are the potential that
ecosystems have to deliver a service. Services are actually conceptualizations (‘labels’) of the
“useful things” ecosystems “do” for people, directly and indirectly whereby it should be
realized that properties of ecological systems that people regard as ‘useful’ may change over
time even though the ecological system itself does not (TEEB Foundations, Chapter 1, 2010).
This TEEB introductory chapter has made a point about “clearly delineating between
ecological phenomena (functions), their direct and indirect contribution to human welfare
(services), and the welfare gains they generate (benefits)” The delineation is also considered
relevant to allow spatial analysis of where the potential service (= the ecological function)
occurs, where the actual provision of the service occurs, and where the benefits are realised.
In the same TEEB chapter the concept of dis-services is mentioned. This refers to, for
example, ecosystems, which include reproduction of species that damage crops and human
1
TEEB 2010, de Groot et al. (2002) and MA (2005)
2
‘Ecological functions´ is sometimes referred to as ´ecosystem functions´; the two terms are synonymous.
3
TEEB 1.10

Page 20 of 103
health. Of course many of these disservices are the result of inconsiderate planning and/or
management of man-made ecosystems, such as agricultural systems, and thus at least
partially the consequence of human actions. Examples mentioned in this TEEB chapter are
“normalizing” rivers (leading to floods), cutting forest on hill slopes (causing erosion and
landslides), and disturbing natural food webs (leading to outbreaks of pests).
The concept of ecosystem services is distinct in that it implies the existence of humans who
have conceptualized certain ecological processes and structures as beneficial to them (de
Groot 2002). Maple tree sap, for example, could not have been considered a provisioning
ecosystem service until humans conceptualized it as a benefit to them, namely as maple
syrup. Referring to such ecological processes and structures as the basis for “services” that
humans perceive as benefits, however, follows from an anthropocentric point of view. Aside
from providing benefits to humans, ecological processes and structures of course constitute
the building blocks of the ecosystems themselves. Sap, for example, is essential for
circulating water and nutrients in trees, in addition to being used by humans in food
production. The lessons to be learned by society from this are that if all the sap is used by
humans, the supply will stop, because the tree will die.
The term “ecological functions”, which existed in the ecological literature some time before
the term “ecosystem services (de Groot 2002), was originally an ecological term describing
the role of processes and structures in the dynamics of ecosystems, but was later (e.g. Braat
et al., 1979; De Groot, 1992) used to describe the useful roles of ecological structures and
processes for human society, e.g. in economic or social activities.
Ecological functions have later been redefined and now refer to the capacity that ecological
processes and structures have to deliver an ecosystem service (TEEB 1.11 2010a; de Groot
2002). The relationship between these concepts is best described as the translation of a
large and diverse number of ecological processes and structures into a more limited set of
ecological functions, which in turn can be re-conceptualized as ecosystem services by
human beneficiaries (TEEB 1.11 2010a; de Groot 2002). So, for example, the water and
nutrient circulation in trees (an ecological process) involves the production of sap (an
ecological function), which is harvested for maple syrup (a provisioning service).
Essentially, if a beneficiary exists to enjoy the ecological function, then it can be re-
conceptualized as an ecosystem service (TEEB 1.12 2010a). As humans change their
perceptions over time about the benefits that ecological processes and structures provide,
the overlap between ecological functions and ecosystem services may shift (TEEB 1.12
2010).
Table 1 provides a comprehensive overview of the range of services provided by agricultural
ecosystems, and shows the links to the ecological functions. With appropriate structuring
and management, agro-ecosystems can provide or contribute to provision of all of these
ecosystem services.
However, agro-ecosystems that are modified to enhance the provision of some services can
lead to trade-offs with other, important services (MA 2005). In agriculture, the issue of
“trade-offs” generally arises from the fact that managers tend to focus on optimizing
provisioning services to increase profits, making decisions that often “trade off” regulating
and cultural ecosystem services. Intensive agriculture, characterized by monoculture crops

Page 21 of 103
and heavy dependence on water, pesticides, and fertilizers, has especially deleterious
effects on these services in the form of i.e. increased soil erosion, lower soil fertility,
pesticide accumulation, and river and lake eutrophication (MA 2005).
Table 1: Ecological processes and structures, ecological functions and ecosystem services
that can be provided by agro-ecoecosystems (adapted from de Groot et al. 2002). The two
ecological processes, and linked functions and services analysed in this study are highlighted
in orange.
Ecological Processes and
Structures
Ecological Functions Ecosystem Services
1) Bio-geochemical cycles e.g. CO2/O2
balance
Gas regulation Climate change mitigation
Temperature regulation
Good air quality
2) Nutrient storage and cycling Nutrient regulation Water purification
3) Influence of ecosystem structure on
dampening disturbances
Disturbance prevention Flood and fire prevention
4) Influence of land cover on runoff and
river discharge
Water regulation Drainage and natural
irrigation
5) Retention and storage of water Water supply Drinking water
6) Influence of vegetation root matrix and
soil biota
Soil retention Erosion prevention
Arable land
7) Rock weathering and accumulation of
organic matter
Soil formation Soil productivity
8) Influence of biota and vegetation in
nutrient and compound breakdown
Waste treatment Pollution control
9) Influence of biota on movement of
floral gametes
Pollination Plant pollination*
10) Trophic-dynamic population control Biological control Pest and disease control
11) Living space and conditions Habitat provision Bio- and genetic diversity
12) Conversion of solar energy into edible
plants
Vegetation growth, feed
supply
Supply of food
*
13) Conversion of solar energy into
biomass
Vegetation growth Supply of fibre, fuel and
other raw materials

Page 22 of 103
14) Landscape structure Landscape with the potential
to be:
aesthetically attractive
recreational
culturally, spiritually,
artistically, historically or
educationally valuable
Scenic roads and housing
Outdoor sporting, eco-
tourism
Use of landscape for cultural,
spiritual, artistic, historical or
educational purposes
*
has been adjusted from de Groot 2002 to fit the context of agriculture.
(ii) Positive and negative externalities of agricultural production
Agriculture affects the natural environment in a diverse set of ways. For example, like most
economic activities, agricultural production uses natural resources and environmentally
harmful substances as inputs and can exploit the environment as a sink for pollution and
waste (Pretty et al. 2001). Such activities have side-effects for a diverse set of third parties
uninvolved in agricultural production or consumption processes themselves such as
communities and ecosystems located downstream from agricultural areas; these effects
constitute the “externalities of agriculture”.
“Externalities” is an economic term generally understood as the unintended, non-monetary
impacts of a production or consumption process on third parties (RISE 2009; Exiopol 2009).
So to qualify as an externality, an impact must be ‘’unintended’’, or external to the
production or consumption process’s main rational (RISE 2009) and ´´non-monetary´´, or not
transmitted through price (Kahn). And, it must affect a ´´third party´´, or an entity not
involved in the production or consumption process (RISE 2009). Pollution is a frequently
cited example of an externality because it generally both occurs incidentally and affects
other entities than the polluters. Though people often consider externalities to be negative,
as it is the case with pollution, there are positive externalities of economic activities as well.
For example, when landowners plant gardens to increase the aesthetic value of their
property, they inadvertently change the environment by increasing the biodiversity of the
area. This benefit can be considered as a positive externality of gardening.
The externalities of agricultural production are, accordingly, the unintentional, non-
monetary effects of agricultural production on third parties. These third parties could be the
agricultural habitat itself, which might lose biodiversity at the hand of crop species
restriction, or an area downstream, which could experience fertilizer contamination in its
water source. The third party could also be society in general, which might experience
climate change as a result of greenhouse gas emissions from manure. On the other hand,
agricultural practice has the capacity to increase resilience to floods, regulate water and
nutrient supplies and support farmland habitats; these are all examples of positive
externalities of agriculture.

Page 23 of 103
Table 2: Externalities of agricultural activities
Resource
Negative
Externalities
Positive
Externalities
Water
Contamination
Eutrophication
Transfer of zoonoses
Water accumulation and supply
Flood protection
Recycling and fixation of nutrients
Air and climate change
Carbon dioxide emissions (CO2)
Ammonia emissions (NH3)
Nitrous oxide emissions (N2O)
Methane emissions (CH4)
Carbon sequestration (soil, biomass)
Soil
Erosion due to missing soil cover
Salinisation caused by improper
irrigation practices
Contamination by dangerous
substances (organic pollutants,
pesticides)
Compaction due to use of heavy
machinery
Decline in soil organic matter due to
missing crop rotation or arable stubble
management
Increase in soil fertility by sustainable
land use practices
Biodiversity and
Landscape
Habitat destruction
Loss of landscape elements
Loss of genetic diversity among
agricultural crops
Support of wildlife dwelling
Conservation of agricultural landscape
and aesthetic value
Recreation and amenity
Source: Ecologic Institute, Exiopol 2009
The highly managed and manipulated character of agro-ecosystems provides ample
opportunity for humans both to conserve and enhance the environment, and to damage it.
For example, “carbon dioxide emissions” and “carbon sequestration” are included in Table 2
as a negative and positive externality, respectively, i.e. depending on the practices that are
employed, agricultural production systems can both increase or mitigate carbon in the
atmosphere. Porter et al. (2009) even offers the concept of a combined food and energy
agro-ecosystem (CFE), a fully-functioning agro-ecosystem that is also a net energy producer
(Porter et al. 2009). Therefore, balancing positive and negative externalities of agricultural
production requires consideration of the specific agricultural practices involved in each
system.
When agricultural producers prioritize the maximization of agricultural production, they
tend to create too many negative externalities and too few positive ones (RISE 2009). This
often occurs because agricultural producers respond to market signals for their saleable

Page 24 of 103
outputs and disregard the effects of production for which there are no functioning markets
(RISE 2009). In essence, producers are not paid for providing positive externalities, nor are
they required to pay the full social or environmental costs of providing negative externalities
(RISE 2009). The resulting undersupply of positive externalities and oversupply of negative
externalities constitutes a classic case of market failure, or the inability of the market to
allocate resources efficiently (Kahn). In the context of externalities, this means that the
market fails to provide the socially optimal level of positive externalities. A core focus of
environmental economics is how governments can best correct such market failures by
introducing subsidies, taxes or regulation to bring provision of these goods and services up
or down to the desired level (RISE 2009).
(iii) The public goods provided by agriculture
The debate surrounding the CAP has unveiled that agriculture provides other goods than the
mere provision of food, biomass and livestock, which were summarised as the public goods
of agriculture (Cooper et al. 2009). This follows from the fact that historically, agricultural
production in the EU has received continuous public support, while other economic sectors
with similar characteristics have not. The public goods provided by agriculture were, in fact,
commonly utilised as a justification for subsidies because they were not rewarded by
markets (RISE 2009). In the current debate about the CAP reform, the provision of public
goods through agriculture is gaining an even higher significance due to its relevance for the
payments provided to farmers.
“Public goods´” is a well-established economics concept defined by the following criteria:
(i) non-excludability, i.e. the good is available to everyone;
(ii) non-rivalry, i.e. one person’s consuming the good does not reduce the amount of it
available to others.
A good must fulfill at least one of these criteria to a reasonable extent in order to be
considered public. Pure public goods, which fulfill both criteria completely, are rare, and
therefore most public goods lie along the continuum between private and public (RISE
2009). For example, though theoretically no one can be excluded from enjoying a scenic
landscape open to the public, there might be competition for physical occupation of the
landscape, should it become congested. In this case, the public good exhibits complete non-
excludability but some degree of rivalry due to congestion. Air, on the other hand, is a pure
public good because air is available to everyone, and one person´s consuming it does not
reduce the amount of it available to others.
The public goods of agriculture are those goods provided by agriculture that qualify as
´´public´´. They include both tangible structures and dynamic processes or flows. Besides
socio-cultural public goods (e.g. employment for people working on farms), which are not in
the focus of this study, the main environmental public goods of agriculture for the EU are:
 agricultural landscapes,

Page 25 of 103
 farmland biodiversity,
 water quality and water availability,
 soil functionality,
 climate stability (carbon storage and greenhouse gas emissions),
 air quality,
 resilience to flooding and resilience to fire (Cooper et al. 2009).
All of these goods show non-excludability or non-rivalry at least to some degree (Cooper et
al. 2009). Farmland biodiversity, for example, exhibits a high degree of non-excludability as
enjoying the sight of and benefiting from the ecological services of species and habitats is
available to everyone. Farmland biodiversity also exhibits a high degree of non-rivalry; one
important exception, however, is hunting, whereby excessive hunting of a targeted species
can reduce the number of animals available to others. Water quality and availability, which
can be both depleted and enhanced by agriculture, exhibit both public and private
characteristics. Whereas water usage and extraction is subject to private control, the longer-
term benefits associated with sufficient amounts of high quality water constitute public
goods. Similarly, soil functionality is rival and excludable due to private ownership and
control of land, but also essential to public goods such as food security, climate stability,
biodiversity and landscape. In contrast, climate stability is a pure public good.
In the context of this study, it is important to note that food, which is strongly associated
with agriculture, does not appear on the list of public goods. This is because food products
are generally provided via a market that excludes consumers who do not have the means to
purchase them, and creates competition between consumers who do. As a result, food
products do not generally fulfill either of the public goods criteria.
Precisely due to their non-excludable and non-rival properties, the provision of public goods
from agriculture generally cannot be secured through markets (Cooper et al. 2009) (i)
because the producers of goods have little incentive to provide them, and (ii) the consumers
of such goods have little incentive to pay for them (Cooper et al. 2009). These conditions
have created a significant undersupply of public goods from agriculture in the EU (RISE
2009).
In addition to showing varying degrees of publicness, the environmental public goods of
agriculture exhibit varying degrees of dependence on agricultural production. Some
farmland habitats and species, for example, have co-evolved with certain (mostly
extensively used) agricultural systems to the extent that they are now fundamentally bound
to them, and are unlikely to adapt to other forms of land use (Cooper et al. 2009). These
habitats and species exhibit a significant and direct dependence on agricultural land use. In
contrast, other public goods such as soil functionality can be maintained by various land use
types, and so are not directly dependent on agriculture. Nevertheless, the increasing
demand for food worldwide will likely increase the demand for agricultural land as well,

Page 26 of 103
making the soil functionality of this land ever more important. For this reason, it is
important to find ways for agriculture to provide those public goods that are not dependent
on agricultural land per se, as well as for the public goods that are dependent on it directly
(Cooper et al. 2009).
This comparative analysis provides a broad context for the development of the conceptual
approach in the next section.
2.4 Dealing with the trade-offs between different ecosystem services
Regarding trade-offs between different ecosystem services, past research mainly restricted
the view on highlighting the disservices or externalities incurred from agro-ecosystems as a
result of optimizing production or provisioning services at the expense of other (regulatory
or cultural) eco-system services (see de Groot et al. 2002; Zhang et al. 2007). Recent
research, however, emphasizes more strongly the importance of informed management to
mitigate the trade-offs between provisioning and other ecosystem services and to enhance
the often overlooked regulating and cultural services of agro-ecosystems (see Power 2010).
The COPI study added an important visualisation of the trade-offs between provisioning and
other ecosystem services with an increase in intensity of land use. Figure 3 illustrates the
relationship between the provisioning and other ecosystem services and biodiversity (Mean
Species abundance indicator) in different land uses.
Figure 3: Relationship between Ecosystem service provision and land use types (The Cost of
Policy Inaction (COPI), Braat & Ten Brink, 2008)

Page 27 of 103
There is a gradual fall of regulating services (water, air, climate) services with further
degradation of the ecosystem (Fig. 3). For recreation and tourism, values of ecosystems for
humans are high if a certain degree of accessibility and infrastructure is provided, whereas
they also fall with the degradation of the ecosystem leading to humans seeking for a
substitution of the service. Hence, according to the figure, the optimum can mostly be
found in agro-ecosystem of light use/extensive agriculture. However, exceptions exist as for
instance wine areas that are intensively managed but highly appreciated in their aesthetic
value by many people. Provisioning services are maximised with further conversion of
ecosystems, with the maximum value depending on the soil quality and how vulnerable the
ecosystem is for degradation, as well as on the revenue generated for the product on the
market. The maximum level is dependent on the additional inputs from humans into the
provisioning process.
In sum, all ecosystem services tend to decrease in value with further degradation of the
ecosystem. The key problem of finding the right balance between those services is the
difference between the maximum value of the services which vary depending on the degree
of land use intensity. Regulating, cultural and other services reach their peak in value in
agro-ecosystems that are managed with low input. On the contrary, provisioning services
are usually maximised through an increase of inputs leading automatically to a decline of
the other services. Finding a compromise between provisioning on the one hand and other
ecosystem services on the other is site and context specific and cannot be limited to a
(mathematical) maximisation function alone as divergence from the “classical” relation
shown in figure 3 might occur and preferences of humans can vary.
‘The European assessment of the provision of ecosystem services’ (Maes et al., 2011)
provides an interesting visualisation of the trade-offs between ecosystem services based on
a Principal Component Analysis that assesses the statistical correlation of thirteen spatial
indicators that map the capacity of ecosystems to provide services. (see Figure 4).
This analysis shows clearly the secluded position of crop production capacity (provisioning
service), which is either uncorrelated or more importantly negatively correlated to other
ecosystem services, particularly the regulating services directly linked to agricultural
production, i.e. water regulation, erosion control and soil quality regulation. The analysis
also shows that NUTS X regions rich in agro-ecosystems are essentially producing crops and
are relatively poor in delivering other ecosystem services. They cover large portions of
Spain, France, Italy, Lithuania, Bulgaria and Poland.

Page 28 of 103
Figure 4: Trade off analysis based on Principal Component Analysis (PCA) on 13 ecosystem
services (EU scale, NUTS X resolution). Vectors or arrows closely to each other represent
correlated services and vectors pointing in perpendicular directions represent services that
are not spatially related (Maes et al., 2011).
Moreover, the trade-off framework has to take into account that services has to be
expressed in the same units to be compared. Provisioning services can be measured in
economic terms (yield or revenue) through market prices, next to physical units such as tons
of grain, while regulating services can readily be measured in physical or biological units but
require alternative monetization methods to be expressed into monetary values. Similarly,
to quantify cultural services bio-physical measures are generally available, but monetization
requires a mix of market and non-market methods (Source: see TEEB 2010b, chapter 5).
After monetization of each of the services an optimization routine can be applied to find the
local optimal mix.
However, another guiding measure for finding the right balance between ecosystem
services could for instance be the level of degradation, because the value of all ecosystem
services are declining with further degradation of the ecosystem – even though the gradient
might be different. A core question is therefore, to what extent degradation can or should
be tolerated in order to maximise the value across all ecosystem services.
In the first decades since the term was coined (Ehrlich & Ehrlich, 1981) ecologists provided
most of the research on ecosystem services and the trade-offs between them (culminating
in the Millennium Ecosystem Assessment, 2005), but in the last decade, economists added
useful insights in valuation techniques and incentive programmes, much of which is brought
together in the TEEB reports (2009, 2010a,b )
A possible approach for better integration of ecological fundaments into economic decisions
when addressing trade-offs between ecosystem services, specifically, could be the

Page 29 of 103
“ecological production functions” mostly developed by Polasky (2008). Ecological
production functions rely on extensive research of a specific ecosystem service in a small or
localized area and typically equate the output of goods (e.g. bushels of grain) to the inputs
(e.g. irrigation, fertilizer) used. Insufficient on its own, this equation seeks to develop an
“eco-economic” production function that weighs the output of a range of ecosystems
services in relation to the impact on ecosystem structure and function.
There are two important components of these production functions: firstly, there has to be
advanced knowledge on the ecosystem functions; and secondly, proper valuation
techniques must be developed. In terms of the first task, research has come far and greatly
improved our understanding of ecosystem functions, but there will always be unknowns.
For the second task, Polasky (2008) suggests distinct physical units of measurement for
ecosystem services (e.g. bushels of crops, tons of carbon sequestered, concentrations of
nitrate in water).
Using an ecological production function that takes land-use management as the input,
Polasky (2008) explores joint biological and economic impacts of management decisions. His
study finds that different land use patterns can substantially increase both biodiversity
conservation and the value of economic activity (commodity production). This methodology
employs localized knowledge and small-scale focus and provides a more accurate valuation
method for specific services, which can then be implemented into local or regional policy.
The major shortcoming of this approach is the fact that it analyses a “steady state” only. In
other words it is a very localized approach that does not consider the externalities of land
use and it does not consider the spatial and temporary dimension of trade-offs. The
challenge is to extend the meticulousness of the “local approach” to broad scale
assessments that can be applied on a global level.
The analysis has shown that trade-offs are complicated by different scales, which can have
at least three dimensions (see Rodriguez et al. 2006): that of space, time and irreversibility.
Like Polasky (2008), Rodriguez et al. (2006) takes land-use as the input and studies the
effects of different decisions based on their spatial repercussions. Rodriguez, therefore,
adds the spatial and temporary dimension to ecological production functions and focuses on
whether the effects of the trade-off are felt locally or at a distant location, whether the
trade-off is felt rapidly or slowly, or irreversibly, whether the eco-system service may return
to its original state if the perturbation ceases.
This approach addresses the complexity of trade-offs that are often unpredictable and
whose impact occurs over space and time. Such an approach provides managers with the
ability to monitor the short-term provisions of services along with the long-term evolution
of slowly changing variables. Policies can then be developed to take account of ecosystem
service trade-offs at multiple spatial and temporal scales (See Rodriguez et al. 2006).
Taking land-use management as the input factor, Rodriguez develops a variety of “future
assessments,” based on the Millennium Ecosystem Assessment scenarios. He experiments
with different land-use management decisions that prioritize different ecosystem services,

Page 30 of 103
and charts the potential effects of these decisions over space, time and irreversibility. By
highlighting the impacts of trade-offs through future scenarios, Rodriguez offers a critical
contribution to making better decisions about trade-offs themselves. These predictions are
useful indicators of the cumulative and synergistic effects of certain decisions. Drawing on
historical examples, he provides further support to Polasky´s argument that the
enhancement of “other” ecosystem services does not come at the elimination of meeting
provisioning agricultural needs.
Difficulties again arise in applying the methodology globally, however, experimenting with
different management decisions and using existing knowledge can help predict different
scenarios on a global scale (see Rodriguez et al. 2006).
2.5 Towards a conceptual framework: the energy balance as a possible measure to
describe provisioning services provided by agro-ecosystems
This last section off Chapter 2 studies the possibility to use the energy balance to assess
quantitatively the provisioning services provided by agro-ecosystems.
Ecosystem services can be viewed as the flows of energy from ecological systems to human
or social-economic systems (H.T. Odum, 1984 a.o.). In the agro-ecosystems there may be
energy embodied in biomass (e.g. food, fibre) or in water streams; i.e. provisioning services),
in the work by ecosystems influencing environmental conditions (e.g. climate, water levels;
i.e. regulating services), or in generating information (e.g. the diversity of genes and species
in ecosystems and landscapes; i.e. cultural services).
The energy flows involved in (food/feed) biomass production from agro-ecosystems are very
complex as it is shown in Figure 5. The energy flows include:
1. R&D energy: many if not all crops result from human intervention in genetic structure of
crop and grass plant species, either through selection, crossbreeding or more recently gen-
modification. In addition a lot of energy is spent to define optimal growing conditions. The
resulting seed quality (potential to produce desired type of biomass) can thus be expressed
as ratio between energy content and energy invested per seed;
2. Before starting the agricultural process seeds need to be delivered to place of application.
3. Farmers plant or sow the seeds, by manual labour or aided by mechanical tools and (fossil)
fuels.
4. Soils may be prepared for growing the crops or fodder, again by human and mechanical
energy. Often highly concentrated chemical products are added (fertilisers, fungicides,
nematocides), which again add energy cost to the process.
5. Some crops require weeding and or above ground pest control, again with manual,
mechanical labour and energy intense chemical products.
6. Crops must be harvested (and transported, on their way to food processing, distribution and
retail, which is not included in the present analysis). During harvest, desired biomass (usually
the sugar and protein rich parts are separated from the so called residual biomass (cellulose
fibres, minerals), which can be used as source of fuel or fibre products, or fed back to the
soil ecosystem, potentially saving on fertilisers.

Page 31 of 103
A. Natural ecosystems have so far provided the genetic capital, with which crops and fodder
have been produced in agricultural systems.
B. Sunlight, rain, wind are often grouped as abiotic or environmental services, to distinguish
them from the ecosystem services which are derived from biotic processes. Nutrients are
cycled with the solar based energies in the hydrological cycle and with the complex soil
ecosystem processes involving microbial chemistry such as mycorrhiza providing essential
nutrients to the root systems of the crops (C).
Figure 5: The energy flows involved in food / feed biomass production. Solid lines indicate
energy flows. Red lines = human activity; yellow lines = environmental / ecosystem
processes; other colours: energy flows resulting from interaction of (natural) ecosystem &
human flows.
As previously mentioned in section 1.3, this study focuses only on the direct use of the soil,
as natural resource for plant production, and therefore excludes all the indirect input (see
production) and output processes (e.g. animal production). Therefore the energy input will
include labour, machinery, fertilisers and irrigation. The approach should also consider
economic values, which are inherent to the valuation of the benefits provided by
agricultural production to human well-being.
Consequently, not all the energy flows shown in Figure 5 are considered. Figure 6 shows the
major energy flows from agro-ecosystems to society in the provisioning ecosystem services,
which are the objective of this study.

Page 32 of 103
Figure 6: Energy flows in provisioning services
The energy balance approach can easily be applied to provisioning ecosystem services in a
broad sense, as biomass produced for energy consumption is directly captured by humans
building on the support of biological processes (photosynthesis and other). The harvest of
the amount of biomass planned, can only be ensured by manipulation of the ecosystem, e.g.
in the selection of particular (crop) species, minimisation of nutrient shortages, optimisation
of water availability etc. All of these activities can be expressed in energy units as well as the
output gained in biomass.
This study focuses on the assessment of the degree of human intervention in the agro-
ecosystem for provisioning services compared to natural ecosystems. Therefore it does not
quantify the also required (natural) external energy inputs, (e.g. sun, rain and wind), nor
internal inputs within the ecosystem, (e.g. nutrient and water flows, microbial activity in
root systems). In addition to the desired types of biomass (for food, feed, fibre, fuels),
production processes also deliver plant components that are not always used in
consumptive processes, but nonetheless contain energy that goes somewhere and has to be
included in an overall energy balance system.
It has to be kept in mind that the caloric (Joule) content of the agricultural product is not
reflecting the whole level of embodied energy within the product, as heat energy has been
lost in processing activities in relation to the output energy.

Page 33 of 103
3. Analytical framework for the quantitative assessment of the energy
balance
3.1 Introduction
The central objective of the project is to analyse the role of agriculture as provisioning
ecosystem service. Since in the former it was shown that ecosystem services can be viewed
as the flows of energy from ecological systems to human or social-economic systems the
energy balance approach seems a useful concept to analyse the provisioning service of
agriculture. Another advantage of using an energy balance is that it enables a quantified
assessment of many different human influenced and natural ecosystems which are all
characterised by flows of energy. In addition to the flows of energy and the net energy gains
it will also be assessed to which extent the net energy production of an agricultural system
is valued in terms of economic value.
The focus of the analysis will therefore be on the input and output relation of agriculture,
the extent to which there is a willingness to pay for the output and not on the way
agricultural production may impact other ecosystem services. Production in agriculture
relates to feed, food, fibre and fuel. These products can be expressed in terms of biomass
and also in terms of energy. Such energy input-output relations can be assessed for the
actual agricultural production systems, but also for more ‘natural’ systems in which there is
no human interference, or systems with very low and very high human interference. By
comparing these 4 situations a better understanding can be derived of the role of
agriculture in provisioning ecosystem service.
An energy balance approach has many advantages as it enables to:
 express in a comparable way the size of the output even though these have a
very different nature (e.g. wheat, grass, wood, corn);
 assess the net energy production by plants, comparing the energy input in the
soil with the energy output in the yields;
 estimate the efficiency of the production systems which may range strongly
within and between regions by management;
 express the production function of agricultural ecosystems in a range of
indicators (see Table 3), which can be used further to develop a statistically
robust analysis of the level of provisioning services delivered by agricultural
ecosystems in the whole EU.
 Analyse the provisioning service of agricultural ecosystems at different levels of
human interference ranging from no interference, low interference to very high
interference.
3.2 CAPRI energy balance model
The CAPRI energy balance model was designed for evaluating energy use and energy
reduction policies in EU agriculture. In the CAPRI energy module several energy indicators
are calculated incorporating the energy requirements for the input quantities of mineral
fertilizer, direct energy sources, machinery, buildings, plant protection, seeds, production

Page 34 of 103
support systems (such as irrigation) and others. The CAPRI energy module data and
methodology enables to calculate various indicators in relation to energy (Kempen and
Kranzlein, 2008). An overview of the type of indicators and units is given in Table 3.
Table 3: Overview of parameters produced in the CAPRI energy module and the related units
For the assessment of the energy balance in this study it was first necessary to convert the
calculation approaches in the CAPRI module which were applicable to the farm level to the
level of the soil. This meant that among the available energy indicators at farm level, only
those that affect the energy balance at soil level were selected, and aggregated to the
energy input and output at soil level. This implies that energy input and output included in
the balance has to be directly linked to crops and to the land management activities of
establishment of a crop, management during cultivation (e.g. weeding, spreading plant
protection products and fertilisers and irrigating) and harvesting. The processing of the
harvest in further end-products for human consumption is excluded. The same applies to
the production of meat or milk is excluded from the balance or at least stops after the
cutting of grass, even though this grass may in fact be fed to animals to produce the milk
and meat. The latter however need further inputs not linked directly to the soil (e.g.

Page 35 of 103
external feed, labour, machinery). The abiotic or environmental input such as solar energy
water and nutrients from the soil are considered both in natural and farming systems as a
constant factor and are therefore not included in the soil balance calculation. This also
applies to the reference layers against which the soil-energy balance of the actual
agricultural systems are compared.
For the soil energy balance calculations produced in this study we focused on two main
indicators to analyse the provisioning service:
1) MJout/MJin per ha.
2) Net MJ per output per ha=MJout-MJin
The two units are calculated per crop type and per crop group type. This will be discussed in
further detail in the next section.
The calculation of the energy balance are done at regional level (Capri regions) and to take
account of the diversity in agro-environmental diversity also at the level of Homogenous
Spatial mapping units (HSMUs) as will be further explained in the next Section.
The following factors are considered in the energy balance calculation:
 On the input side we consider energy input in relation to machinery, seeds, fertilisers
(including nitrogen from manure), irrigation and labour.
 On the output side biomass production and related energy output is taken into
account in produced food, feed and other biomass potentially used for fibre, fuel
and other products. To determine the total biomass output, the starting point is the
biomass which can be removed sustainably. How this is further defined is discussed
in detail in next Section. When presenting the final calculations in chapter 4 it will
become clear that there are agricultural crop activities which produce relatively
small amounts of energy, e.g. vineyards, olives and fallow and will therefore
generally show a negative energy balance result. In the interpretation of the results
as categorized according to crop type and grouped crop types account needs to be
taken of the composition of the crop activities on the final results.
3.3 Approach to calculating a soil energy balance
The first step is to convert the farm energy balance to a soil energy balance. In order to do
this all cropping activities in a region are considered, and for these activities energy input
and output factors are linked as far as these are directly linked to the soil on which these
crops are cultivated.
The crops included in the assessment are given in Table 4. The energy balance is calculated
per crop, but then aggregated to different clusters of crops to produce final results of the
analysis.

Page 36 of 103
Table 4: Overview of crops included in CAPRI
Crop accronyms Crops In/excluded
SWHE softwheat in
DWHE durumwheat in
RYEM rye in
BARL barley in
OATS oats in
MAIZ sugarmaize in
OCER other cereals in
RAPE oil seed rape in
SUNF sunflower in
SOYA soya in
OOIL other oil crops in
OIND other industrial crops ex
NURS nursery crops ex
FLOW flowers ex
OCRO Other crops ex
MAIF fodder maize in
ROOF fodder root crops in
OFAR fodder other on arable land in
GRAE extensive grassland in
GRAI intensive grassland in
PARI paddy rice in
OLIV olives in
PULS pulses in
POTA potatoes in
SUGB sugarbeet in
TEXT flax and hemp in
TOBA tobacco in
TOMA tomatoes in
OVEG other vegetables in
APPL apples in
OFRU other fruits in
CITR citrus in
TAGR table grapes in
TABO table olives in
TWIN wine in
FALL fallow in
ISET Set aside obligatory - idling in
GSET Set aside obligatory used as grass land in
TSET Set aside obligatory - fast growing trees in
VSET Set aside voluntary in
The final energy balance results are calculated per crop as specified in Table 4. However for
the presentation of the crops different crop groups were made. The grouping of the crops is
presented in Annex 1.

Page 37 of 103
On the input side there are two dimensions of energy inputs:
1) Input per resource (e.g. fertiliser, machinery, fuel)
2) Input per activity/process (e.g. cultivation, irrigation)
The difference between these dimensions can be illustrated with the following example.
Ploughing a field requires 4000 MJ for fuel and 3000 MJ for energy used to produce the
machinery (tractor and trailed machinery). The latter is allocated to the crop according to
the hours of machinery use in the crop and the depreciation of it. Irrigating the plot requires
2000MJ for fuel and 1000 MJ for energy used to produce the pump in the factory which is
again allocated to the crop according to the hours of irrigation and the depreciation of the
pump. In total the energy input is 10000 MJ, which can be allocated to the crop and
aggregated in two ways:
1) in 6000 MJ fuel and 4000 MJ machinery (resource dimension) , or
2) 7000MJ for cultivation and 3000 MJ for irrigation (activity dimension).
An overview of all energy input indicators per crop per resource and per activity is given in
Table 5. The energy input per resource refers to all the energy that is used to produce the
resource that is further used in the establishment, cultivation and harvesting of a crop.
Table 5: Input indicators included in the soil energy balance
Indicator Unit Description
Plant protection products MJ/ha
Energy that is needed to produce the plant protection products that
are needed per hectare per crop
Electricity MJ/ha Energy input as electricity
Diesel MJ/ha
Energy input as diesel fuel (energy content of diesel + energy used
in processing)
Other fuels MJ/ha
Energy input as other fuel (energy content + energy used in
processing)
Machinery MJ/ha
Energy that is needed to produce the machinery that is used during
the planting, cultivation and harvesting of the crop.
Seed MJ/ha Energy used during production of the seed
Mineral fertiliser (Nitrogen, Phosphates and
potassium) MJ/ha Energy used during production of the mineral fertiliser
Seeding/planting MJ/ha Energy used for planting/seeding the crop.
Cultivation management MJ/ha
Energy used in mechanisation (tractor use) and fuel for managing
the crop once established (e.g. weeding)
Application of fertiliser MJ/ha Energy used for applying the fertilisers
Application of manure MJ/ha Energy used for applying manure
Application of plant protection products MJ/ha Energy for plant protection products
Application/pumping of irrigation water MJ/ha
Energy used in mechanisation (e.g. pump) and fuel for applying
irrigation water
Processing harvested goods MJ/ha Energy used to conserve harvested good, mainly drying of cereals
Labour MJ/ha
Energy needed by humans to perform all the crop production
related activities
Plant protection products, seeds and mineral fertilisers all need energy when produced. The
input of this energy can directly be linked to the crop as it is known how much of these
inputs are used per crop. So these can also be linked easily to the land on which these crops
are grown and therefore expressed in an input per hectare.
For the energy input used in the production of machinery this is more complicated as the
machinery is not only used for a single crop, furthermore some crops need more, while

Page 38 of 103
others less of machinery input. For this CAPRI uses the (average) operation time of
machinery per crop as a distribution factor which are based on data derived from national
machinery inventories. In case of data gaps, values of countries are used which have most
similar farming characteristics.
To calculate the energy contents of fertilizers both artificial and manure fertilisers need to
be included and allocated to a crop. The incorporation of manure fertiliser required
additional processing as in the CAPRI farm energy balance calculation all manure fertilizer
was (indirectly) allocated to animal production, while for the soil energy balance this needs
to be allocated to the cropping activities (including grasslands).
Since CAPRI calculates input of nitrogen (N), phosphate (P) and potassium (K) in kg per crop,
the energy input used for spreading the manure also needs to be allocated to the nitrogen,
phosphate and potassium contents of the manure. How this is calculated is explained in
Annex 2. The reason why the energy input only includes the fuel consumption of the tractor
and other machinery use, and not the energy used in the production of the machinery, is
because according to the logic of the CAPRI energy model this part of the energy input is
completely allocated to the ‘cultivation’ part of the cropping activities.
For irrigation figures from different sources were used to get a most up to date and spatially
detailed overview of irrigation share per crop and total irrigation water consumption per
crop. Several of these sources were already included in the in the CAPRI model. They are
based on various national sources providing information on irrigated crop area and/or water
use combined with crop specific expert information. However as part of this project these
CAPRI irrigation data were further up-dated with more spatially detailed irrigation data
based on Wriedt et al. (2008) in which irrigation shares per crop area and total irrigation
water consumption are provided at 10*10 km grid. For further details see Annex 5.
Labour was not included in the CAPRI energy balance at first. Within the scope of this
project a first simple estimate was made of the energy contents of one hour of labour
input. How this was done and included in the up-dated CAPRI energy balance calculation is
explained in Annex 3.
On the output side we distinguish between
 output of harvested products used for food and feed and
 output of biomass that can be used for production of non-food products including
bioenergy.
The latter category includes all biomass that can be harvested sustainably and which is
already partly harvested as part of regular crop management activities such as pruning and
cutting activities.
The CAPRI model calculates crop yield in kg fresh weight. The CAPRI energy module was fed
with data on energy content of the output products (food, feed and other biomass) which
were collected from literature. In Annex 4 an overview of the energy content of all output

Page 39 of 103
included in the assessment is given. As a starting point, coefficients are estimated from the
energy of forage ( as defined in animal science literature) and heating value of biomass. As
values are typically given per kg dry matter, all the coefficients had to be converted to fresh
weight.
Calculation of energy balance at regional and HSMU level3.3.1
The calculation of the energy balances is made at the scale of regions (CAPRI regions) and at
a more detailed scale of Homogenous Spatial Mapping Units (HSMUs) (see Box 1). The
reason to use these units is explained underneath. The approach to converting all the input
and output factors to this detailed spatial level is explained in Annex 5.
Since most administrative regions (e.g. NUTS regions) are very diverse from an agro-physical
perspective there is a need to split these regions up into smaller entities to take better
account of the diversity in farming conditions and farm management practices. These
conditions and practices are very influential in the energy input and output relations of
cropping activities. It therefore makes sense to also establish a soil energy balance at the
level of HSMUs enabling a better approach to the diversity within regions. In order to do this
there is also a need that HSMU specific energy input and output factors are available which
will reflect the diversity in HSMU characteristics in the final energy balance. In Annex 5 a
description is given of the characteristics of the HSMUs and the approach applied to add
farm specific input and output factors to the individual HSMUs. In Chapter 4, results are
presented at regional and HSMU level.
Box 1: Homogeneous Spatial Mapping Units (HSMUs)
Within the Dynaspat project, the Homogeneous Spatial Mapping Units (HSMUs) have been created
and land use information has been assigned to these units in a statistical allocation procedure. In the
SEAMLESS project the allocation of Farm Accountancy Data Network (FADN) farms to HSMU then
followed a similar statistical and econometric procedure as the land use allocation and the results
were then aggregated into dimensions of a farm typology.
HSMUs are an intersection of land cover (Corine LC 2000), relief (slope in five classes), Soil Mapping
Units (so-called soil landscapes from the European soil map) and the NUTS 2/3 boundaries
(depending on the size of the NUTS regions) (see Figure 10). Each HSMU has identical values for land
cover class, slope class and Soil SET. Other parameters (such as annual rainfall) may differ inside the
HSMU. These HSMUs can be multiple polygons (open) which implies that one HSMU can be spread
over different locations within a NUTS area. Attributes belonging to every HSMU are calculated
(characteristics in terms of soil, climate, land cover, yielding capacity). These attributes were used to
allocate the land uses to the HSMUs, but also the farms. Further details on the allocation of land
uses and farm types can be derived from Kempen et al. (2011) and Elbersen et al. (2006 and 2010).
An HSMU is an intersection of land cover, slope, soil mapping units and Nuts boundaries

Page 40 of 103
In order to calculate the soil energy balance at the HSMU level it was necessary to first allocate all energy input
and output factors to the HSMU level. In the CAPRI-Dynaspat module all crop areas are already distributed
over HSMUs. The same applies to some input and output factors which have already been allocated in a
statistical allocation procedure to the crops per HSMU (see Table 12 for overview) or will still be allocated to
HSMU within the scope of this project. The allocation already done took yield level as the distribution factor
(so everything is proportional to yield). The yield was derived from the MARS-CGMS
4
, a crop growth model
providing yield predictions for all major crops in the EU taking account of detailed soil and meteo data
integrated with statistical yield information. Details on the spatial allocation procedure of already allocated
farm management factors can be derived from Leip et al. (2008).
Calculation of the reference layers3.3.2
In order to compare the different land use intensity levels of agricultural ecosystems in
terms of energy balance, different reference situations that reflect various human
interference into the provisioning function of natural systems have been created.
Given data availability and the logic discussed in Chapter 3 the following reference
situations are available:
 Only natural grassland (all present agricultural land use per region/HSMU is covered
by natural grassland, extensively grazed by wild animals)
 Low-input farming
 High input farming (intensive crop production)
The natural grassland layer assumes a situation in which the present agricultural land area
of the EU is covered by grassland that is maintained under grass by grazing with wild
animals. No inputs are assumed. The climate and soil conditions determine the biomass
4
See: http://mars.jrc.ec.europa.eu/mars/About-us/AGRI4CAST/Crop-yield-forecast/The-Crop-Growth-
Monitoring-System-CGMS
Slope
CORINE Land Cover
NUTS3 Regions
Soil Mapping Unit
Soil mapping units

Page 41 of 103
yield in combination with the extensive grazing of wild animals. The yield is both water
limited and nutrient limited. This layer was prepared using the MARS system (for details see
Annex 6.
Map 1 Dry biomass yield (kg/ha) for low input layer
The low input layer assumes a similar land use pattern as the actual land use, but at a 50%
lower input level for nitrogen and no irrigation (see Map 1). Also this layer was prepared
using the MARS-CGMS system (see Annex 6).
The high input layer assumes a similar land use pattern as the actual land use pattern but a
maximum yield. This implies that crop growth is simulated with the MARS CGMS assuming
no water, nor nitrogen limitation. More details on the preparation of this layer is given in
Annex 6.
For the three reference layers the energy balance and the net energy balance are
calculated. The results are compared against the energy balance of the actual farming
situation to understand the relative position of actual farming in provisioning services.
Calculation of the economic value3.3.3
The economic value per crop and ha is defined as product of yield and national market
prices (yield*market price).
For non-marketable feed a shadow value is estimated based on marketable commodities.
For example, the value of roughage is determined by taking the value of the replacement,
e.g. the value of oil cake and cereals to be bought to replace the roughage in terms of crude

Page 42 of 103
protein and energy. The amount of replacement and the related price determines the
shadow value of the roughage.
For the calculation of the output values the subsidies paid under pillar 1 and 2 of the
common agricultural policy of the EU are excluded, but the market price may be affected by
market intervention policies (e.g. export subsidies, intervention).

Page 43 of 103
4. Results
The soil energy balance calculations based on the approach described in the former section
are presented in this chapter for the actual situation. With the actual situation we refer to
the situation which was calculated with the detailed statistically based farm information for
the years 2003-2005 contained in the Coco and Capreg database belonging to the CAPRI
system.
For the presentation of the results different crop groups have been produced (see Annex 1).
A further analysis of which groups of crops are most suitable for the presentation of the
final results is further investigated in Annex 6. From this analysis it became clear that the
best coverage of HSMUs is reached by including the category CropsAll and
ArablePermGrassFallow.
In the following first an overview is given of the input and output levels specific per crop
group. This provides an understanding of the differences in crop types in input and output
mixes and how these also differ between EU regions, environmental zones and within
regions. This is then followed by a presentation of the final energy balance results and an
analysis of how the input levels relate to the output levels and the net energy output relates
to the economic value. The economic value represents a proxy for the willingness to pay for
the energy output of agriculture. Finally the net energy balance of the actual farming
situation is compared against situations with more and less human interference.
4.1 Energy balance results
The soil energy balance calculations were made per crop and were then aggregated to total
area averages and total crop group averages to make them presentable and analyse the
overall patterns and trends. Overall, we see that there are very large differences in input
and output levels between crops, but also within crop groups between EU regions.
In Figure 10 an overview is given of the average per hectare input per category. Overall it
becomes clear that input levels are generally lower in EU-10 then in EU-15 countries. This
particularly applies to Romania, Bulgaria, Estonia, Lithuania and Latvia. In the EU-15 group
the UK jumps out as a country with a relatively low input level per hectare.
High average input levels per hectare in the EU-15 are particularly found in the Netherlands
and Belgium, Italy, Spain and Germany. In the EU-10 Slovenia jumps out with a very high
input per hectare.
The categories taking the largest part of the input are mostly energy for cultivation and
fertilisers. In the Mediterranean countries irrigation also adds significantly to the input side.

Page 44 of 103
Figure 10 Composition of input (MJ/ha) for all crops
Figure 11 Average output (MJ/ha) for all crops in terms of food, feed and other biomass
0 5000 10000 15000 20000 25000 30000
Austria
Bulgaria
Belgium-Luxemburg
Czech Rep
Germany
Denmark
Estonia
Greece
Spain
Finland
France
Hungary
Ireland
Italy
Lithuania
Latvia
Netherlands
Poland
Portugal
Romania
Sweden
Slovenia
Slovakia
UK
EU-10
EU-15
Cultivation
Irrigation
Labour
Manure Fertilizer
Mineral Fertilizer
Plant Protection
Process Harvest
SEED
0.00
20000.00
40000.00
60000.00
80000.00
100000.00
120000.00
Ireland
Portugal
Spain
Slovenia
UK
Sweden
Austria
Bulgaria
Latvia
Netherlands
EU-15
Hungary
Lithuania
Slovakia
Estonia
Romania
Greece
Italy
France
EU-10
Belgium-Luxemburg
CzechRep
Germany
Finland
Poland
Denmark
Other biomass
Feed
Food

Page 45 of 103
On the output side a distinction was made between output in food, feed and other biomass.
The latter category includes biomass such as straw and cuttings not necessarily harvested
from the field at this moment. The results in Figure 11 show that highest output levels in
food are found in countries like Denmark, Poland, Finland, Germany, Czech Republic and
Belgium. The EU-10 who have generally a lower input level have an higher output level then
the EU-15, at last when looking at the food output. The comparison of the input and the
output already shows that high input levels often go together with high output levels and
vice versa especially in relation to total biomass output, but not necessarily in relation to
food output. This is also confirmed when looking at the net energy balance in Figure 12.
Figure 12 Average net energy balance per hectare (MJ/ha) for all crops
The highest net output in terms of total biomass is reached in Ireland, Belgium, Denmark,
Netherlands, UK and Germany which were countries showing both relatively high and low
input levels. Explanatory factors should clearly be sought in a combination of factors, but
location in the Atlantic zone having a temperate climate could be one of them. The other
explanatory factors are of course the land use composition and the farming management
practices.
Land use patterns in the EU countries differ significantly (see Table 6 and Annex 9, Table 1
and 2).
-10000
0
10000
20000
30000
40000
50000
60000
70000
80000
90000
100000
Portugal
Ireland
Slovenia
Spain
Netherlands
Italy
Austria
EU-15
Sweden
UK
Hungary
Belgium-Luxemburg
France
Bulgaria
Greece
Latvia
Slovakia
Lithuania
Estonia
Germany
EU-10
Romania
Finland
CzechRep
Poland
Denmark
All crops net energy balance (MJout-MJin)
Food Total

Page 46 of 103
Countries with a very high share in the arable land category are Denmark, Finland, Hungary,
Czech Republic and Sweden. These high arable land shares often go together with high to
very high output levels, particularly in relation to food output. The countries with the
highest grassland area shares, e.g. Ireland, UK, Slovenia and Netherlands, are also among
the countries with the higher output levels particularly in total biomass. Countries with a
more mixed land use pattern, like is seen in most southern EU countries, generally show a
lower output level, particularly when it goes together with high fallow land areas shares, like
is the case for Portugal, Spain and Bulgaria.
Table 6 Relative area shares per crop (%/total UAA)
Country
total
arable Fallow Vegetables Grass Fruits Olives Vineyards
Austria 39% 5% 0% 54% 0% 0% 1%
Bulgaria 51% 10% 1% 34% 1% 0% 3%
Belgium-
Luxemburg 55% 2% 3% 38% 1% 0% 0%
Czech Rep 72% 2% 1% 24% 1% 0% 0%
Germany 63% 7% 1% 28% 0% 0% 1%
Denmark 84% 8% 0% 7% 0% 0% 0%
Estonia 66% 4% 0% 30% 0% 0% 0%
Greece 37% 9% 3% 29% 4% 16% 2%
Spain 35% 16% 1% 31% 4% 9% 4%
Finland 81% 15% 0% 3% 0% 0% 0%
France 58% 6% 1% 32% 1% 0% 3%
Hungary 74% 4% 2% 17% 2% 0% 2%
Ireland 27% 1% 0% 72% 0% 0% 0%
Italy 50% 4% 3% 26% 4% 7% 5%
Lithuania 58% 7% 1% 32% 1% 0% 0%
Latvia 56% 6% 1% 36% 1% 0% 0%
Netherlands 50% 2% 4% 43% 1% 0% 0%
Poland 67% 9% 1% 21% 2% 0% 0%
Portugal 36% 10% 1% 32% 4% 10% 6%
Romania 60% 3% 2% 33% 2% 0% 2%
Sweden 71% 15% 0% 14% 0% 0% 0%
Slovenia 35% 0% 1% 59% 1% 0% 4%
Slovakia 65% 0% 1% 33% 0% 0% 1%
UK 35% 3% 1% 61% 0% 0% 0%
Although part of the energy ratios per country can be explained from the composition of the
agricultural land use, diversity in management of the crops is also an important factor for
regional differences which also become clear when comparing input and output levels per
country and environmental zone level per crop (see Annex 8).

Page 47 of 103
For all crops the energy for fertilisers and cultivation on average make up the largest part of
the input, but the largest variation in input levels per region is found for irrigation as is
shown in Annex 9. Clearly average input levels are lower in the Alpine, Boreal-Nemoral and
Continental-Pannonian zones. In the Atlantic-Lusitanian zone and the Mediterranean the
input levels are higher. In the Atlantic this is cause by an overall high input level as
compared to other zones, but in the Mediterranean the extreme are larger with very low
and very high input levels occurring at the same time. An important factor of influence on
the final energy balance in the latter zone is irrigation which can be extremely high in
certain regions in certain crops. In the Atlantic high input levels are particularly caused by
high level of energy input in cultivation and through mineral fertiliser application.
For cereals the mineral inputs generally make up the largest share of the input, followed by
energy input for harvesting, but the largest regional variation is found in the irrigation level
and the harvesting. The permanent crops show by far the highest average inputs and also
the largest regional variation in input levels. This variation is caused by large variation in
both cultivation and irrigation inputs. For grassland the variation is also enormous, with
irrigation and fertiliser inputs as most regionally diverse.
Overall one can conclude that variations in input levels are very wide within crops, both for
the whole EU as within environmental zones particularly in relation to irrigation, process
harvesting and mineral fertiliser gift.
Map 2 Energy balance per hectare (MJout/MJin) calculated for food at HSMU level

Page 48 of 103
Map 3 Energy balance per hectare (MJout/MJin) calculated for total biomass at HSMU level
Map 4 Net energy per hectare (MJout-MJin) calculated for food at HSMU level

Page 49 of 103
Map 5 Net energy per hectare (MJout-MJin) calculated for total biomass at HSMU level
The variation in inputs and output shows strong differences in net energy balance results as
can be seen from Maps 2 to 5, certainly when total biomass output is taken into account.
Overall the largest energy gains per hectare when only food output is taken is found mostly
in the regions in North, west and central Europe and Italy and it concerns mainly arable land
dominated regions. The highest energy gains per hectare when total biomass is taken as
output are mostly in grassland areas in the Atlantic UK, Ireland, Sweden, western and
central France, North-western Spain and Germany.
4.2 Relation between input and output
From the former it became clear that there is a large diversity in energy input and output
levels between crops, between regions and even between similar crops in the same region.
The energy gain that can be reached per crop differs therefore strongly but overall it is clear
that the biggest energy gains are in arable and grassland systems in which generally high
outputs are reached with generally lower input levels. This is confirmed in Figure 13 in
which we see that grasslands show low input levels while their output levels vary from very
low to very high. Arable crops, like cereals and oilseed, also cluster in the lower input levels.
The output for oil crops is however also rather low, because their crop residues are not
assumed to be used as biomass. For cereals the output ranges strongly from low to high,
whereas the straw of cereals contributes significantly to the total output.
In fruits and olives the relation between input and output levels are relatively weak and
show a large diversity. But overall input levels are clearly higher than in grassland and

Kaphengst 13 agriculture as provisioning ecosystem service_0

Kaphengst 13 agriculture as provisioning ecosystem service_0

Recommended

Recommended

More Related Content

Viewers also liked

Viewers also liked (13)

Similar to Kaphengst 13 agriculture as provisioning ecosystem service_0

Similar to Kaphengst 13 agriculture as provisioning ecosystem service_0 (20)

More from Michael Newbold

More from Michael Newbold (20)

Recently uploaded

Recently uploaded (20)

Kaphengst 13 agriculture as provisioning ecosystem service_0