Ecosystem Ecology and Env Management
Ecosystem Ecology and Env Management
Ecosystem Ecology and Env Management
In the last 40 years, the area of global agricultural land has grown by 10%, but in per cap-
ita terms agricultural land area has been in decline. This trend is expected to continue
as land is increasingly limited and the population grows.
From WBCSD/IUCN 2008
Introduction
With the adoption of the Convention on Biological Diversity (United Nations
1992), the sustainable management and protection of biodiversity shifted from
being an option to an acknowledged necessity: sustainability is now the high-
level goal of environmental management policy. Sustainability implies the abil-
ity for processes and activities to be able to continue indefinitely. Within the
specific context of environmental management this has been taken to mean
meeting the needs of the present without compromising the ability of future
generations to meet their own needs (derived from the definition of sustainable
development developed by the Brundtland Commission (World Commission
on Environment and Development 1987)). More recently the concept has been
expanded to explicitly include three elements, sometimes referred to as the
three pillars of sustainability: environmental, social and economic (Table 7.1).
For many policy makers, sustaining social structures and hence economic sys-
tems is the imperative, while many scientists would argue that you cannot
have a sustainable economy without sustainable use of natural systems. The
key message however is that the three pillars are linked and together provide
the long-term objective for environmental management.
The challenges for those charged with managing the system are determin-
ing the limits to sustainability – i.e. what are the ways and rates of use which
Ecosystem Ecology: A New Synthesis, eds. David G. Raffaelli and Christopher L. J. Frid. Published by Cambridge
University Press. © British Ecological Society 2010.
E C O S Y S T E M E C O LO GY A N D E N V I R O N M E N TA L M A N A G E M E N T 141
can be sustained – and setting in place policies to achieve those goals. The lat-
ter is a socio-political issue while the former is very much a scientific issue and
determining those limits may be the greatest challenge facing ecologists in the
third millennium.
There is no easy, one-size-fits-all tool for sustainable ecosystem management,
although the textbooks are littered with examples of tools and approaches that
have worked well for specific ecosystems (e.g. Meffe et al. 2002 and references
therein). Ecosystem management is challenging, partly because of the complex
challenges presented by managing coupled social and biophysical systems, the
science of which remains at a very young stage, and partly because of the all-
things-to-all-men nature of an ecosystem (Raffaelli and Frid, this volume). For
these reasons, we do not offer prescriptions for ecosystem management. Instead,
we highlight some of the issues and concepts which might help those who are
involved with the day-to-day management of ecosystems to construct frameworks
within which they can operate. Being aware of these issues should increase the
chance that decisions will lead to sustainable outcomes, or at least not close
down future management options. Central to this is the need to recognise that
ecosystems comprise complex, highly connected ecological and social networks,
such that changes brought about by management of one component are likely to
affect other components in complicated and non-linear ways. As a consequence,
ecosystem managers will always find themselves managing under some degree
of uncertainty, due in part to an incomplete knowledge of system dynamics and
in part to what are at best probabilistic predictions about future states under cli-
mate and social change. Here we rehearse the basis of some of that uncertainty.
Figure 7.1 A chemical ‘spectrum’ of the Earth from the OMEGA spectrometer on board
the Mars Express space probe. The peaks indicate that water (H2O) and molecular oxy-
gen (O2) dominate, while carbon dioxide (CO2) is also identified, as well as ozone (O3),
and several other minor constituents. (Adapted from www.astrobio.net/news/modules.
php?op = modload&name = News&file = article&sid = 528)
Figure 7.2 Changes in global land use from the eighteenth to twentieth centuries (data
from Riebsame et al. 1994).
144 C H R I S T O P H E R L . J. F R I D A N D D AV I D G . R A F FA E L L I
1960, McCann 2000). A logical argument can be made, often referred to as the
‘Insurance Hypothesis’, that species-rich systems are more resilient to perturb-
ations. In a species-diverse system, if the abundance of one species declines,
for whatever reason, other species increase to compensate, ensuring that eco-
system goods and services are maintained (von Bertalanffy 1960). Empirical
modelling and experimental trials using simple mesocosm systems, e.g. bac-
teria and protists, lend support to the insurance hypothesis (Haines-Young and
Potschin, this volume) and so provide an argument for the conservation of bio-
diversity and diverse systems to ensure that this insurance remains available.
While this argument is both logical and supported by some experimental tests,
it should be noted that many of the most extensive and productive ecosystems
have a naturally low diversity, including boreal forests, lakes, bogs and estuar-
ies (Grime 1997). In these systems there are few taxa to replace species which
are lost. In addition, field-scale studies of biodiversity–ecosystem functioning
relationships often contradict the results from the highly controlled simplified
systems maintained in mesocosm experiments. Nevertheless, Haines-Young
and Potschin (this volume) conclude that there is compelling evidence for posi-
tive relationships between ecosystem functioning and biodiversity, but it is
likely that many ecosystem processes could be maintained by fewer species
as long as the functional groups to which they belong remain represented
(although the fewer species within each functional group, the less resilient
that group is to further species loss). Thus, a priority for environmental man-
agement is to identify and then protect the key/irreplaceable species in the sys-
tem, and to do so requires the development of field programmes that directly
assess the functional diversity of natural functioning ecosystems (e.g. Bremner
et al. 2003, Bremner et al. 2006a, 2006b, Weithoff 2003). Whilst the whole area
of biodiversity–ecosystem functioning is relatively young and thus remains a
little contentious, the argument that there is a link between biological diver-
sity and the maintenance of a range of goods and services used by humankind
is clearly sufficiently persuasive to have led to the development of strong pol-
icy drivers (Figure 7.3).
Figure 7.3 Summary of links between pelagic biodiversity and marine ecosystem proc-
esses with an indication of the areas where these impact on to environmental policy.
The dotted arrows are trophic flows which link the three major compartments of the
system. Traits likely to affect those trophic flows are shown within each compartment.
The larger black arrows indicate how the compartments influence the delivery of eco-
system services, numbered as follows:
1. The need to ensure food supply, including protein and the health benefits of poly-
unsaturated fatty acids.
2. A sink for CO2 produced by the burning of fossil fuels.
3. Climate regulation through regulation of atmospheric gases and the production of
dimethyl-sulphide and hence cloud condensation nuclei which affect the planet’s albedo.
4. Waste/nutrient assimilation.
5. Avoidance of toxic algal blooms.
6. Ecotourism based on naturalness and/or species of interest.
(Adapted from Duffy and Stachowicz 2006).
Changes in the state of one of the nodes in the network are likely to lead to
state changes in other nodes, as perturbations spread through the system. Thus,
management that focuses exclusively on a single element of an ecosystem is
unlikely to lead to sustainable management, and more holistic approaches are
required. A good example of wholesale ecosystem change brought about by
exploitation that was focused on a single element is provided by the history
of whaling in the Antarctic. Large whales were hunted almost to extinction
146 C H R I S T O P H E R L . J. F R I D A N D D AV I D G . R A F FA E L L I
during the last century, with the result that populations of other krill feed-
ers, such as smaller whales, crab-eating seals and several species of penguins,
increased in response to the increased availability of krill, presumed to be due
to competitive release. However, cessation of whaling has not led to a recovery
of whale populations, whales finding it hard to re-establish themselves in this
system (Laws 1985), a phenomenon discussed at length below. Large whales
are clearly embedded within a complex network of species interactions in the
Antarctic, and the full consequences of their exploitation for the wider ecosys-
tem were not appreciated at all.
The appreciation of the need for holistic approaches goes back a long way
historically. For example, management in the British-controlled forests of
nineteenth-century India included fire protection and maintenance of the ‘nat-
ural house-hold’ – what we would now refer to as the supporting ecosystem.
Similarly, within the US, in addition to forestry management the early conser-
vation movement also considered the ‘natural rights’ agenda. With the need to
boost food and forestry production after two world wars, this holistic conserva-
tion agenda was subjugated to the provision of ‘natural parks’ and protection
of a few high-profile species. The 1960s saw the birth of modern concepts of
conservation and concern about the loss of aspects of the natural world, much
of which followed from publication of Rachel Carson’s Silent Spring (Carson
1962), and a series of species-in-crisis initiatives raised public perceptions. In
general, these concerned single species. The cause of concern (hunting or pol-
lution) was readily identified and to an extent management responses were
clear and often successful. However, during the latter part of the twentieth
century there were increasing numbers of cases of species that were declining
to the point where they were threatened with extinction, or at least extirpa-
tion, due to degradation of their habitat, often from multiple causes. There
were also many examples of single-issue species rescue plans that failed to
achieve their objectives (for examples see Linquist 2008). Such cases led to the
recognition of the need to protect habitats and also other components of the
ecosystem such as the species’ food supply. With hindsight, it seems obvious
that to conserve or to exploit species sustainably, all aspects of the support-
ing, functioning, ecological system must also be understood and protected, but
many managers today still seem to find this hard to grasp.
of the concept of natural selection (Darwin 1859). Put simply, the balance of
nature implies that if a population or other component of the environment is
exploited to a given level it will decline, but once that exploitation stops it will
recover to its original level: the balance of nature will be re-established. If that
limit is exceeded then the resource is incapable of recovery. This is the limit of
sustainable use. The concept of sustainability thus implies the existence of a
scientifically determinable limit value for exploitation.
The balance-of-nature perspective ignores a key aspect of the Malthus
‘limits to growth’ tenet: if resources are in short supply, then if harvesting
removes individuals that were using that resource, other individuals will seek
to gain benefit from the resources now available. These individuals may be of
the harvested species, in which case they too will be removed. Therefore, in
general, non-exploited taxa will ultimately benefit. In a multispecies system
these adjustments could involve many taxa as competitive and predatory inter-
actions take account of the additional ‘predation’ in the system coming from
the harvesting. If the exploitation also alters the habitat and impacts on other
species, then the system-level changes can be considerable.
The balance-of-nature model holds that once a perturbation ceases the sys-
tem will swing back to its original balance. However, in some systems one or
more other configurations are equally stable, so the system does not neces-
sarily swing back to the original state, as implied above for the exploitation
of whales in the Antarctic. For example, the removal of cod from the Grand
Banks region off the north-east coast of North America resulted in an increase
in the populations of lower-value dogfish, rays and prawns. This was prob-
ably the result of a combination of the removal of predation by large cod and
release from competition with smaller cod. In the 20 years since the cod fish-
ery was closed the stocks have failed to show a strong recovery and the system
appears to have entered a new stable state as young cod seek to compete for
food with the rays, dogfish and prawns and are also predated by dogfish (Rice
2002). There are those that advocate heavy fishing of the dogfish, rays and
prawns to allow the cod to recover but this is potentially a high-risk strategy
as there could be a third (and unknown) stable state for this system that would
yield even fewer benefits. Shifts from coral-reef-dominated to algal-dominated
systems have also been widely reported as examples of human-induced switch-
ing between multiple states, which have different dynamics and deliver differ-
ent suites of key ecosystem services (Hughes 1994, Hughes et al. 2007). Raffaelli
and Frid (this volume) provide other examples.
May’s (1977) seminal paper on breakpoints in systems with multiple stable
states used simple population models and empirical data to demonstrate not
just the existence of multiple stable states in natural systems, such as grazed
pasture, fish stocks and insect pest populations, but also the existence of rapid
transitions (breakpoints) and of non-symmetrical trajectories (May 1977). The
148 C H R I S T O P H E R L . J. F R I D A N D D AV I D G . R A F FA E L L I
latter has been termed ‘hysteresis’, a term used in physiology, which can be
defined as a system that exhibits path-dependence. That is, the current system
trajectory is not just a function of the system attributes but also depends on
how they have varied previously. To put this in a management context, redu-
cing an impact may not cause the system to change back through the same
intermediates as it did when the impact was applied. Hysteresis can apply not
just to community composition states but also to functional dynamics (Potts
et al. 2006).
It is clear from much of the foregoing that ecological systems can be clas-
sified as ‘complex systems’, the dynamics of which tend to lead to a range of
behaviours that make prediction of future states difficult for managers. Even
simple non-linear systems can show complex dynamics and recognition of this
has had a major impact on the general public’s appreciation of the limits to,
for example, weather forecasting, and the suggestion that ‘a butterfly flapping
its wings in Florida could lead to a storm in Europe’ (Gleick 1988). The poten-
tial for similarly complex dynamics in biological populations has stimulated
considerable theoretical interest, but in the real world, systems with chaotic
behaviour would be expected to go extinct over evolutionary time (May 1995,
1999). Thus, we may take some comfort from the fact that most natural popu-
lations will have dynamics that, while they may contain the seeds of chaos,
will not normally exhibit chaotic behaviours. Of course, the concern for man-
agement and policy makers is that if human activity shifts the system outside
the normal range of conditions experienced over evolutionary time, this could
send the system into a region of complex dynamics making prediction and
hence management extremely difficult as well as increasing the probability of
wholesale system collapse.
While the occurrence and frequency of chaotic dynamics remain unclear,
the widespread occurrence of multiple stable states in ecosystems is now well
recognised (Beisner et al. 2003). As May (1977) demonstrated, this is likely to
mean that those systems will also contain breakpoints or tipping-points and
possible hysteresis. These have profound implications for those attempting
to manage such systems (Lenton et al. 2008, Raffaelli and Frid, this volume,
Table 7.2).
Predicting when and where these thresholds will occur (if at all) is very dif-
ficult and some authors have questioned whether knowledge of their poten-
tial existence is therefore of any use for practical management, although
Groffman et al. (2006) offer constructive suggestions for atmospheric pollut-
ants. Even if it turns out to be effectively impossible to identify threshold
effects in advance of their manifestation, their possibility should be conveyed
to stakeholders and cautious management decisions advocated to minimise
their appearance in the system. At the same time, one can accept that some
ecological surprises are inevitable and adjust the usage strategy for the eco-
system accordingly. For instance, the exploitation of spruce forests for timber
E C O S Y S T E M E C O LO GY A N D E N V I R O N M E N TA L M A N A G E M E N T 149
Table 7.2. Climate change ‘tipping elements’ identified by Lenton et al. (2008)
Arctic sea ice Some scientists believe that the tipping point for the total
loss of summer sea ice is imminent.
Greenland ice sheet Total melting could take 300 years or more but the tipping
point that could see irreversible change might occur
within 50 years.
West Antarctic ice sheet Scientists believe it could unexpectedly collapse if it slips
into the sea at its warming edges.
Gulf Stream Few scientists believe it could be switched off completely
this century but its collapse is a possibility.
El Niño The southern Pacific current may be affected by warmer
seas, resulting in far-reaching climate change.
Indian monsoon Relies on temperature difference between land and sea,
which could be tipped off-balance by pollutants that
cause localised cooling.
West African monsoon In the past it has changed, causing the greening of the
Sahara, but in the future it could cause droughts.
Amazon rainforest A warmer world and further deforestation may cause a
collapse of the rain supporting this ecosystem.
Boreal forests Cold-adapted trees of Siberia and Canada are dying as
temperatures rise.
An ecosystem-based approach
The IUCN’s Commission on Ecosystem Management defines the Ecosystem
Approach as a strategy for the integrated management of land, water and liv-
ing resources that promotes conservation and sustainable use in an equitable
way. The Ecosystem Approach places human needs at the centre of environ-
mental management. It aims to manage human impacts on the ecosystem,
150 C H R I S T O P H E R L . J. F R I D A N D D AV I D G . R A F FA E L L I
based on the multiple functions that ecosystems perform and the multiple
uses that are made of these functions. The Ecosystem Approach does not aim
for short-term economic gains, but aims to optimise the use of an ecosystem
without damaging it. It was endorsed at the fifth Conference of the Parties to
the Convention on Biological Diversity (CoP 5 in Nairobi, Kenya, May 2000/
Decision V/6) as the primary framework for action under the Convention, and
led to the so-called Malawi Principles (Table 7.3).
Embedded firmly within the Ecosystem Approach and the Malawi Principles
is the idea that ecosystems provide services (Haines-Young and Potschin, this
volume) and that there will inevitably have to be trade-offs made between dif-
ferent services when considering management interventions. For instance, the
decision to use the environment for one activity, such as road building or the
siting of complexes of offshore wind turbines, will impact on ecosystem ser-
vices that were provided by the biodiversity now lost as a result of that inter-
vention (see White et al., Haines-Young and Potschin, both this volume). The
full environmental costs of any management decisions therefore need to be
evaluated with respect to the lost services and the net gain (or loss) of benefits
presented to those in society who are the ultimate beneficiaries. This process
requires knowledge and understanding of how that ecosystem works, what
services it currently delivers and might potentially deliver, and how changes to
one dimension (e.g. by constructing a road) affect the delivery of other services
(Haines-Young and Potschin, this volume).
There are no off-the-shelf, empirical models available which will allow the
manager to explore these trade-offs and their consequences, yet such infor-
mation will certainly be demanded by those involved (often in conflict) in the
decision-making process. Our present knowledge and understanding of the con-
sequences of trade-offs across different spatial and temporal scales are currently
poor although the subject of much research effort worldwide. For instance we
are particularly ignorant of the fundamental relationships between specific bio-
diversity elements and specific ecosystem processes within service-providing
units, such as habitats, nested within a landscape. Also, we do not understand
how these different service-providing units interact with each other across the
different spatial scales over which management takes place. Because of this
ignorance, managers often have to make decisions on anecdotal information,
logical arguments or expert judgement, running a significant risk of poor man-
agement and perhaps closing down future use options for that system.
A better evidence base is clearly desirable and the most persuasive evidence
comes from large-scale, controlled experiments. For instance, if one wishes to
know what the impact of forest logging is on other services such as water qual-
ity, flood protection and recreation, then one should ideally clear-fell a patch
of trees and measure changes in service provision in experimental (felled) and
control (intact) areas. Such experiments need to be done at ecologically relevant
landscape scales, since the outcomes are likely to be scale dependent. Carrying
E C O S Y S T E M E C O LO GY A N D E N V I R O N M E N TA L M A N A G E M E N T 151
1 The objectives of management of land, water and living resources are a matter of
societal choice
2 Management should be decentralised to the lowest appropriate level
3 Ecosystem managers should consider the effects (actual or potential) of their
activities on adjacent and other ecosystems
4 Recognising potential gains from management, there is usually a need to
understand and manage the ecosystem in an economic context. Any such ecosys-
tem-management programme should:
• Reduce those market distortions that adversely affect biological diversity
• Align incentives to promote biodiversity conservation and sustainable use and
• Internalise costs and benefits in the given ecosystem to the extent feasible
5 Conservation of ecosystem structure and functioning, to maintain ecosystem
services, should be a priority target of the Ecosystem Approach
6 Ecosystems must be managed within the limits of their functioning
7 The Ecosystem Approach should be undertaken at the appropriate spatial and
temporal scales
8 Recognising the varying temporal scales and lag-effects that characterise
ecosystem processes, objectives for ecosystem management should be set for the
long term
9 Management must recognise that change is inevitable
10 The Ecosystem Approach should seek the appropriate balance between, and
integration of, conservation and use of biological diversity
11 The Ecosystem Approach should consider all forms of relevant information,
including scientific and indigenous and local knowledge, innovation and practices
12 The Ecosystem Approach should involve all relevant sectors of society and scientific
disciplines
out such experiments is costly and often requires long timescales to ensure that
all ramifications are manifest and thus accounted for in the decision-making
process. Managers can rarely afford to wait that long and it is therefore import-
ant to estimate the required duration of the experiment before embarking on
an expensive venture that may not be able to be seen through to its conclusion.
This is well illustrated by the proposed use of a pathogen, calicivirus, to control
rabbit populations (an alien species) in New Zealand (Figure 7.4). Like all eco-
systems, the New Zealand pastureland system is complex, with rabbits embed-
ded in a network of interactions with other species, including alien predators
(cats, stoats and ferrets) and the giant skink, an endangered endemic lizard. A
key uncertainty for those wishing to increase agricultural production by using
the virus to remove the rabbits is the implication for skink populations: will
152 C H R I S T O P H E R L . J. F R I D A N D D AV I D G . R A F FA E L L I
Figure 7.4 New Zealand pastureland system linkages in a putative field experiment
designed to detect the effects of rabbit removal by a viral agent on the abundance of
giant skink (modified from Raffaelli and Moller 2000).
Government_Policy Crop_Price
production 0 profit 0
stewardship 100 breakeven 0
loss 100
Land_Use
arable90 0
arable50 19.2
arable10 80.8
River_Nutrients
low 98.1
elevated 1.91
Algal_Biomass
w100 92.4
w300 6.71
w1000 0.91
Mudflat_Invertebrates
Sea_Trout_Fishery
killed 0
good 86.6
normal 100
bad 13.4
enriched 0
Shorebirds Breeding_Success_Shorebirds
reduced 16.7 low 33.3
good 33.3
more 33.3
natural 50.0 high 33.3
Figure 7.5 A simplified Bayesian Network for the Ythan catchment, north-east
Scotland. Each box (node) presents an important factor linking land use practice in the
catchment with shorebirds of conservation interest in the estuary. The probabilities
shown for the factors are determined by combinations of the parent nodes and reflect
expert opinion, stakeholder suggestions or empirical data. For instance, the different
farming practices are determined by a combination of high-level government policy
and farmers’ expectations of financial returns expected from crops. Thus, when there
are financial incentives for stewardship and there is a low grain price, this will encour-
age farmers to grow fewer hectares of arable crops. As a result, nitrogen fertiliser
run-off is low, there is little growth of algal mats on mudflats, invertebrate abundance
is good and shorebirds and sea trout will thrive. Altering the parent nodes to ‘produc-
tion’ and ‘profit’ will lead to cascading changes in crop cover, nutrient levels, extent
of algal mats, invertebrate abundance and shorebird numbers. Note that the number
of shorebirds on the estuary is also driven by success on the breeding grounds many
hundreds of kilometres distant (data from Raffaelli et al. 1999).
154 C H R I S T O P H E R L . J. F R I D A N D D AV I D G . R A F FA E L L I
those provisioning services which have a market value, the real worth of
other services for which there is no market are either traditionally under-
valued using monetary-based approaches or cannot be captured at all using
standard economic valuation methods. Combining non-commensurate meas-
ures of worth can be achieved using formal techniques, such as Multi Criteria
Analysis, whilst informal approaches, such as Rural Participatory Appraisal,
may be more appropriate in other contexts (Edwards-Jones et al. 2000). The
task of accommodating a heady mix of biophysical, social and economic fac-
tors, together with sets of non-commensurate values within a decision-making
framework will be one of the real challenges for an Ecosystem Approach to
environmental management.
Given the increasing focus on the maintenance of the flows of services from
stocks of natural and other capital which underpin the Ecosystem Approach,
it seems likely that systems-based approaches, including those based on
energy, will see a resurgence (Stoy, this volume, Raffaelli and Frid, this vol-
ume). Systems-based approaches have been applied to the management of
many marine fisheries systems, freshwater lakes, species populations such as
black bear, wetland swamps, estuaries and lagoons (reviewed in Jorgensen et al.
2007). A less-well-appreciated application is their potential for estimating the
economic value of ecosystem components, through the concept of embodied
energy or emergy. Many aspects of the environment have been historically
undervalued by economists, partly due to a limited appreciation of the com-
plex interrelationships between different ecosystem components so that the
full costs of management interventions are not captured, and partly because
many benefits of ecosystems are ‘imperfectly owned’. A good example is the
recreational and cultural services that landscape managers provide for the gen-
eral public. The owner of an iconic, heather-covered grouse moor landscape
in the Scottish Highlands does not charge walkers, writers, poets, artists or
birdwatchers for the benefits they derive from his landscape assets, nor can
he effectively deny those recreationists access to those services, since they
may be perceived from far outside the owner’s legal boundaries. Such services
therefore generally have no market and they are consequently undervalued.
Methods do exist for estimating the value of many non-market aspects of the
environment, the most familiar of which are Contingent Valuation approaches,
such as a person’s Willingness-To-Pay. Many ecologists feel uncomfortable with
such approaches because they depend heavily on consumer preferences that
may change over time as knowledge and understanding and the individual’s
financial circumstances change. Also, those being questioned may not always
be entirely honest in their answers, the process not involving an actual trans-
action of money.
Energy-based approaches offer an intriguing solution to these difficulties.
Emergy is the energy required by nature to produce the goods or services of
E C O S Y S T E M E C O LO GY A N D E N V I R O N M E N TA L M A N A G E M E N T 155
interest: the amount of work required to make, for example, a forest or a blue
whale (Odum 1988). The more work required, the higher the value of that
asset. By comparing the work done by the environment with the work of the
human economy, a monetary value can be obtained for emergy, in units of
emergy dollars (Em$) (Odum 1996, Odum and Odum 2000). The use of emergy
in environmental accounting and decision making is described in detail in
Odum (1996), but has not been widely applied, with the relatively few applica-
tions restricted to North America. Turner et al. (1988) applied this approach
to natural landscapes in Georgia, USA. They first estimated the Gross Primary
Production (GPP) of the system, a measure of how much solar energy is used
to fix the carbon that, in the case of forests, provides the services of flood
protection through water-flow moderation, recreation, water purification, ero-
sion moderation, timber production and non-timber products. GPP was then
converted to a relevant input into the economy, in this case fossil fuel equiva-
lents (FFE) by considering the fuel efficiency of the natural forest resource.
Finally, they converted the FFE into dollars using the ratio of Gross National
Product (GNP) to total energy use in the US economy. They concluded that
estimates of value based on energy analysis more properly captured the mag-
nitude and importance of non-market aspects of biodiversity. Whilst it is clear
that emergy analysis can generate far higher valuations of aspects of biodiver-
sity than monetary-based approaches (Odum and Odum 2000), the approach
does not reflect social preferences, something which is seen as an advantage by
some, a disadvantage by others, and which presents a very real philosophical
tension between economists and systems-analysis ecologists.
Concluding remarks
The considerable challenges in terms of both science delivery and societal
understanding that accompany the adoption of an ecosystem approach mean
that at present there is no simple mechanism for its delivery (Maltby et al.
1999). Rather a host of organisations and individuals are engaged in various
aspects of translating the aspiration into practical tools for implementation
of policy. This has to be done with an incomplete knowledge about the eco-
system’s dynamics and the distinct possibility of major changes, like thresh-
old effects. Armed with this knowledge, ecosystem management might seem
daunting, but for most managers there is no choice. The alternative is to stand
by wringing one’s hands and deny a living to those whose livelihoods depend
on that ecosystem, a political unreality. Probably the best that can be done is
to take a pragmatic, but cautious, approach that is based on what we under-
stand at present, which will not close down options for the future and which
can be monitored so that changes in trends or approaches to limits can be
identified and responded to (see White et al., this volume). This is the Adaptive
Management approach, where management is continually adjusted according
156 C H R I S T O P H E R L . J. F R I D A N D D AV I D G . R A F FA E L L I
References
Beisner, B. E., Haydon, D. T. and Cuddington, K. Antarctic penguin populations: reduced
(2003) Alternative stable states in ecology. competition with whales or a loss of sea
Frontiers in Ecology and the Environment, 1(7), ice due to environmental warming? Polar
376–82. Biology, 11, 525–31.
Bremner, J., Frid, C. L. J. and Rogers, S. I. (2003) Frid, C. L. J., Paramor, O. A. L. and Scott, C. L.
Assessing marine ecosystem health: The (2006) Ecosystem-based management of
long-term effects of fishing on functional fisheries: is science limiting? ICES Journal of
biodiversity in North Sea benthos. Aquatic Marine Science, 63(9), 1567–72.
Ecosystem Health & Management, 6, 131–7. Gleick, J. (1988) Chaos. Sphere, London.
Bremner, J., Rogers, S. I. and Frid, C. L. J. Grime, J. P. (1997) Biodiversity and Ecosystem
(2006a) Matching biological traits to Function: The Debate Deepens. Science,
environmental conditions in marine 277(5330), 1260–1.
benthic ecosystems. Journal of Marine Groffman, P. M., Baron, J. S., Blett, T., Gold, A. J.,
Systems, 60, 302–16. Goodman, I., Gunderson, J. H., Levinson,
Bremner, J., Rogers, S. I. and Frid, C. L. J. B. M., Palmer, M. A., Paerl, H. W., Peterson,
(2006b) Methods for describing ecological G. D., Poff, N. L., Rejeski, D. W., Reynolds,
functioning of marine benthic assemblages J. F., Turner, M. G., Weathers, K. C. and
using biological traits analysis (BTA). Weins J. (2006) Ecological thresholds: the
Ecological Indicators, 6, 609–22. key to successful environmental
Cain, J. (2001) Planning improvements in natural management or an important concept with
resource management. Guidelines for using no practical application? Ecosystems, 9,
Bayesian networks to support the planning and 1–13.
management of development programmes in the Holmes, R. (2006) Imagine Earth without
water sector and beyond. CEH, Wallingford. people. New Scientist, 192(2573), 36–41.
Carson, R. (1962) Silent Spring. Houghton Hughes, T. P. (1994) Catastrophes, phase-shifts,
Mifflin, Boston. and large-scale degradation of a Caribbean
Darwin, C. (1859) On the Origin of Species by means coral-reef. Science, 265(5178), 1547–51.
of natural selection. John Maurray, London. Hughes, T. P., Rodrigues, M. J., Bellwood,
Duffy, J. E. and Stachowicz, J. J. (2006) D. R., Ceccarelli, D., Hoegh-Guldberg,
Why biodiversity is important to O., McCook, L., Moltschaniwskyj, N.,
oceanography: potential roles of genetic, Pratchett, M. S., Steneck, R. S. and Willis,
species, and trophic diversity in pelagic B. (2007) Phase shifts, herbivory, and the
ecosystem processes. Marine Ecology Progress resilience of coral reefs to climate change.
Series, 311, 179–89. Current Biology, 17(4), 360–5.
Edwards-Jones, G., Davies B. and Hussain, S. Jorgensen, S. E., Fath, B. D., Bastianoni, S.,
(2000). Ecological Economics. An introduction. Marques, J. C., Muller, F., Nielson, S. N.,
Blackell Science Ltd, Oxford. Patten, B. C., Tiezzi, E. and Ulanowicz, R. E.
Fraser, W. R., Trivelpiece, W. Z., Ainley, D. G. (2007) A New Ecology. Systems Perspective.
and Trivelpiece, S. G. (1992) Increases in Elsevier, Amsterdam.
E C O S Y S T E M E C O LO GY A N D E N V I R O N M E N TA L M A N A G E M E N T 157