Handbook of Ecological Indicators for Assessment of Ecosystem Health - Chapter 8 docx

Perhaps most alarming in this development is that the global fisheries production appears to have been declining steadily since 1990,3the larger predatory fish stocks are being rapidly d

CHAPTER 8

Application of Ecological Indicators for Assessing Health of

Marine Ecosystems Villy Christensen and Philippe Cury

8.1 INTRODUCTION

‘‘Roll on, thou deep and dark blue ocean — roll! Ten thousand fleets sweep over thee in vain; Man marks the Earth with ruin — his control stops with the shore,’’ Lord Byron wrote two hundred years ago Much has happened since, and humans now impact the marine environment to an extent far greater than thought possible centuries or even decades ago

The impact comes through a variety of channels and forcing factors Eutrophication and pollution are examples, and while locally they may be important, they constitute less of a direct threat at the global scale A related issue, global warming and how it may impact marine ecosystem may be of more concern in the foreseeable future This is, however, presently being evaluated as part of the ‘‘Millennium Ecosystem Assessment,’’69to which we will refer for further information

Habitat modification, especially of coastal and shelf systems, is of growing concern for marine ecosystems Mangroves are being cleared at an alarming rate for aquaculture, removing essential habitat for juvenile fishes and invertebrates; coastal population density is exerting growing influence on coastal systems; and bottom trawls perform clear-cutting of marine habitat, drastically altering ecosystem form and functioning The looming overall threat to the health of marine ecosystems, however, is the effect of overfishing,2 and this will be the focus of the present contribution

We have in recent years witnessed a move from the perception that fisheries resources need to be developed by expanding the fishing fleet toward an understanding that the way we exploit the marine environment is bringing havoc to marine resources globally, endangering the very resources on which a large part of the human population rely for nutrition Perhaps most alarming

in this development is that the global fisheries production appears to have been declining steadily since 1990,3the larger predatory fish stocks are being rapidly depleted,4,5 while ecosystem structure and habitats are being altered through intense fishing pressure.1,6,7

In order to evaluate how fisheries impact marine ecosystem health, we have

to expand the toolbox traditionally applied by fisheries researchers Fisheries management builds on assessments of fish populations Over the years, a variety of tools for management have been developed, and a variety of population-level indicators have seen common use.8 While such indicators serve and will continue to serve an important role for evaluating best practices for management of fish populations, the scope of fisheries research has widened This is due to a growing understanding that where fish populations are exploited, their dynamics must be considered as integral components of ecosystem function, rather than as epiphenomena that operate independently

of their environment Internationally, there has been wide recognition of the need to move toward an ecosystem approach to fisheries (EAF), a development strengthened by the Food and Agricultural Organization of the United Nations (FAO) through the Reykjavik Declaration of 2001,9 and reinforced

at the 2002 World Summit of Sustainable Development in Johannesburg, which requires nations to base policies for exploitation of marine resources

on an EAF Guidelines for how this can be implemented are developed through the FAO Code of Conduct for Responsible Fisheries.10The move is widely supported by regional and national institutions as well as academia, nongovernmental organizations and the public at large, and is mandated by the U.S National Oceanic and Atmospheric Administration.11

Internationally, the first major initiative related to the use of ecosystem indicators for evaluating sustainable fisheries development was taken by the Australian government in cooperation with the FAO, through a consultation

in Sydney, January 1999, involving 26 experts from 13 countries.12 The consultation resulted in ‘‘Technical Guidelines No 8 for the FAO Code of Conduct for Responsible Fisheries: Indicators for Sustainable Development of Marine Capture Fisheries.’’13 These guidelines were produced to support the implementation of the code of conduct, and deal mainly with the development

of frameworks, setting the stage for using indicators as part of the management decision process

The guidelines do not discuss properties of indicators, nor how they are used and tested in practice This instead became the task of an international working group, established jointly by the Scientific Committee on Oceanic Research (SCOR) and the Intergovernmental Oceanographic Committee (IOC) of UNESCO SCOR/IOC Working Group 119 entitled ‘‘Quantitative Ecosystem Indicators for Fisheries Management’’ was established in 2001 with 32 members drawn internationally The working group’s aim was defined as to support the scientific aspects of using indicators for an ecosystem approach to fisheries, to review existing knowledge in the field, to demonstrate the utility and perspec-tives for new indicators reflecting the exploitation and state of marine ecosystems, as well as to consider frameworks for their implementation The current overview article is influenced by the work of the SCOR/IOC Working Group 119, while prepared prior to the conclusion of the working group

We see the key aspects of ecosystem health as a question of maintaining biodiversity and ecosystem integrity, in line with current definitions of the term What actually constitutes a ‘‘healthy’’ ecosystem is a debatable topic This debate includes the way we can promote reconciliation between conservation and exploitation interests It also includes the recognition and understanding of system states to minimize the risk for loss of integrity when limits are exceeded.14From a practical perspective we assume here that we can define appropriate indicators of ecosystem health and evaluate how far these are from a reference state considered representative of a healthy ecosystem We will illustrate this describing indicators in common use as well as the reference state they refer to

8.2 INDICATORS

A vast array of indicators have been described and used for characterizing aspects of marine ecosystem health; a non-exhaustive review found upwards of two hundred related indicators.15On this background it is clear that the task

we are faced with is not so much one of developing new indicators, but rather one of setting criteria for selecting indicators and evaluating the combination

of indicators that may best be used to evaluate the health of marine ecosystems Indeed, the key aspects of using indicators for management of ecosystems is centered on defining reference states and on development of indicator frameworks, as discussed above.16 However, here we will focus on a more practical aspect: What are the indicators that have actually been applied to evaluate the health status of marine ecosystems?

8.2.1 Environmental and Habitat Indicators

Human health is impacted by climate; many diseases break out during the colder winter months in higher latitudes or during the monsoon in the lower

We do not expect to see a similar, clear impact when discussing the marine environment, given that seasonal variability tends to be quite limited in the oceans We do, however, see longer-term climate trends impacting ocean systems, typically over a timescale of decades, and often referred to as regime shifts.17,18 Climate changes especially become important when ecosystem indicators signal change — is a change caused by human impact through, for example, fishing pressure, or are we merely observing the results of a change in, for example, temperature? Understanding variability in environmental indicators is thus of fundamental importance for evaluating changes in the status of marine ecosystems This conclusion is very appropriately supported

by the first recommendation of the U.S Ecosystem Principles Advisory Panel

on developing a fisheries ecosystem plan: ‘‘[T]he first step in using an ecosystem approach to management must be to identify and bound the ecosystem Hydrography, bathymetry, productivity and trophic structure must be considered; as well as how climate influences the physical, chemical and biological oceanography of the ecosystem; and how, in turn, the food web structure and dynamics are affected.’’11

A variety of environmental indicators are in common use, including atmospheric, (wind, pressure, circulation), oceanographic (chemical composi-tion, nutrients/eutrophicacomposi-tion, temperature and salinity), combined (upwelling, mixed layer depth), and indicators of the effect of environmental conditions for, for example, primary productivity, plankton patterns, and fish distribu-tion.19

Habitat impacts of fisheries have received increasing attention in recent years, focusing on biogenic habitats such as coral reefs, benthic structure, seagrass beds and kelp forests, which are particularly vulnerable to mechanical damage from bottom trawl and dredging fisheries.20The trawling impact on marine habitats has been compared to forest clear-cutting and estimated to annually impact a major part of the oceans shelfs.21While habitat destruction has direct consequences for species that rely on benthic habitats for protection (as is the case for juveniles of many fish species),22 it is less clear how even intensive trawling impact benthic productivity.20,23 A recent study found though that the productivity of the benthic megafauna increased by an order of magnitude in study sites where trawling had ceased, compared to control sites with continued trawling.24

Habitat indicators for ecosystem health are in other ecosystems typically focused on describing communities and community change over time As marine ecosystems are generally less accessible for direct studies, habitats descriptions are mostly lacking Indeed, for many ecosystems the only informative source may be charts, which traditionally include descriptions of bottom type as an aid to navigation In recent years critical habitats has, however, received increased focus, and aided by improved capabilities for linking geopositioning and underwater video surveys, habitat mapping projects are now becoming widespread activities, providing data material that in a foreseeable future will be useful for deriving indicators of ecosystem health

As indicators for human impact on marine habitats proxies such as, for example, proportion of the seabed trawled annually, the ratio of bottom-dwelling and demersal fish abundance, and proportion of seabed area set aside for marine protected areas have been used.21

8.2.2 Species-Based Indicators

Indicators of the level of exploitation is central to management of fisheries, focusing on estimating population size and exploitation level of target species.25 Such applications of indicators are, however, of limited use for describing fisheries’ impact on ecosystem health if they only consider target species Instead the aim for this is to identify species that may serve as indicators of ecosystem-level trends For example, the breeding success and feeding conditions of marine mammals and birds may as serve as indicators of ecosystem conditions.26

Another approach is to examine community-level effects of fishing, and indications are that indicators for which the direction of change brought about

by fishing can be predicted may serve as useful indicators of ecosystem status.27 Examples of potential indicators may be the average length of fishes or proportion of high-trophic-level species in the catch

Most studies dealing with community-aspects related to species in an ecosystem describes species diversity, be it as richness or evenness measures.28

A variety of diversity indices have been proposed, with selection of appropriate indices very much related to the type of forcing function that is influencing ecosystem health However, it is often a challenge when interpreting such indices to describe the reference states for ‘‘healthy’’ ecosystems.29,30

Using indicators to monitor individual species is of special interest where there are legal or other obligations; for example, for threatened species From

an ecological perspective, special interest has focused on keystone species due

to their capability to strengthen ecosystem resilience and thus positively impact ecosystem health.31Keystone species are defined as strongly interacting species that have a large impact on their ecosystems relative to their abundance Who are they, and what are their roles in the ecosystem? The classical example from the marine realm is one of sea otters keeping a favorite prey, sea urchins in check, allowing kelp forests to abound.32 Eradication of sea otters has a cascading effect on sea urchin, which in turn deplete the kelp forests Identification of keystone species is currently the focus of considerable research efforts, reflecting that protection of such species is especially crucial for ecosystem health Surprisingly, few examples of keystone species in marine systems have been published so far

8.2.3 Size-Based Indicators

It was demonstrated more than thirty years ago that the size distribution of pelagic communities could be described as a linear relationship between (log) abundance and size.33It is commonly observed that there will be a decreasing

relationship between the log abundance and size The intercept of the size distribution curve will be a function of ecosystem productivity, while the slope

is due to differential productivity with size Forcing functions, such as fisheries, are expected to impact notably the slope of the size distribution curves, with increasing pressure associated with increased slopes as larger-sized organisms will be relatively scarce in an exploited system (Figure 8.1) The properties of size distribution curves and how they are impacted by fishing are well understood,15,29,34,35while there is some controversy around the possibility of detecting signals from changes in exploitation patterns based on empirical data sets.30 Still, size distribution curves have been widely used to describe ecosystem effects of fishing, and studies have indeed shown promising results,

as demonstrated in one of the main contributions to the 1999 International Symposium on Ecosystem Effects of Fishing.36

Fisheries impact fish populations by selectively removing larger individuals (see also section 8.5 below), and thus by removing the faster-growing, large size-reaching part of the populations It is widely assumed that if such phenotypic variability has a genetic basic, then exploitation will result in a selective loss in the gene pool with potentially drastic consequences.37There is, however, limited empirical evidence of such loss of genetic diversity and genetic drift, but this may well be because the area so far hasn’t been the subject of much research New studies indicate that it may be a real phenomenon.38

8.2.4 Trophodynamic Indicators

Fish eat fish, and the main interaction between fish may well be through such means,39 indeed a large proportion of the world’s catches are of

Figure 8.1 Particle size distribution curves for an ecosystem in unexploited and exploited

states Data are binned in size classes and logarithmic abundance (usually of numbers, occasionally of biomass) is presented Exploitation is assumed to mainly reduce abundance of larger-sized organisms, while cascading may cause increase

of intermediate sized (not shown here).

piscivorous fishes There has, for this reason, been considerable attention for development of trophic models of marine ecosystems over the past decades,41,42 and this has led to such modeling reaching a state of maturity where it is both widely applied and of use for ecosystem-based fisheries management.43,44 When extracting and examining results from ecosystem models it becomes a key issue to select indicators to describe ecosystem status and health, we describe aspects of this in the next sections

8.3 NETWORK ANALYSIS

One consequence of the current move toward ecosystem approaches to management of marine resources is that representations of key parameters and processes easily get really messy When working with a single species it is fairly straightforward to present information in a simple fashion But what do you do

at the ecosystem level when dealing with a multitude of functional groups? One favored approach for addressing this question is network analysis, which has identification of ecosystem-level indicators at its root

Network analysis is widely used in ecology (as discussed in several other contributions in this volume), and also in marine ecology.45 In marine ecosystem applications, interest has focused on using network analysis to describe ecosystem development, notably through the work of R.E Ulanowicz, centered around the concept of ecosystem ascendancy.46,47 Related analyses have seen widespread application in fisheries-related ecosystem modeling where it is of interest to describe how humans impact the state of ecosystems.48,49 Focus for many of the fisheries-related modeling has been

on ranking ecosystems after maturity sensu Odum.50 The key aspect of these approaches is linked to quantification of a selection of the 24 attributes of ecosystem maturity described by E.P Odum, using rank correlation to derive

an overall measure of ecosystem maturity.51

8.4 PRIMARY PRODUCTION REQUIRED TO

SUSTAIN FISHERIES

How much do we impact marine ecosystems? This may be difficult to quantify, but the probable first global quantification that went beyond summing up catches, and incorporated an ecological perspective estimated that human appropriation of primary production through fisheries around 1990 globally amounted to around 6% of the total aquatic primary production, while the appropriation where human impact was the biggest reached much higher levels: for upwelling ecosystems, 22%; for tropical shelves, 20%; for nontropical shelves, 26%; and for rivers and lakes, 23%.52 These coastal system levels are thus comparable to those estimated for terrestrial systems, where humans appropriate 35 to 40% of the global primary production, be it directly, indirectly or foregone.53

In order to estimate the primary production required (PPR) to sustain fisheries, we use an updated version of the approach used for the global estimates reported above Global, spatial estimates of fisheries catches are now available for any period from 1950, along with estimates of trophic levels for all catch categories.54,55 We estimate the PPR for any catch category as follows,

PPR ¼ Cy 1

TE

ð8:1Þ

where Cyis the catch in year y for a given category with trophic level TL, while

TE is the trophic transfer efficiency for the ecosystem We use a trophic transfer efficiency of 10% per trophic level throughout based on a meta-analysis,52and sum over all catch categories to obtain system-level PPR

We obtained estimates of total primary production from Nicolas Hoepffner from the Institute for Environment and Sustainability, based on SeaWiFS chlorophyll data for 1998 and the model of Platt and Sathyendranath.56

8.5 FISHING DOWN THE FOOD WEB

Fishing tales form part of local folklore throughout the world I caught a big fish What a big fish is, is however a moving target as we all tend to judge based on our own experience, making us part of a shifting-baseline syndrome.57 As fishing impact intensifies, the largest species on top of the food web become scarcer, and fishing will gradually shift toward more abundant, smaller-prey species This form part of a process, termed ‘‘fishing down the food web’’7 in which successive depletion results in initially increasing catches as the fishery expands spatially and starts targeting low-trophic-level prey species rather than high-low-trophic-level predatory species, followed by a steady phase, and often a decreasing phase caused by overexploitation, possible combined with shift in the ecological functioning

of the ecosystems (seeFigure 8.2).7

A series of publications based on detailed catch statistics and trophic-level estimates typically from FishBase have demonstrated that ‘‘fishing down the food web’’ is a globally occurring phenomenon.58–60Indeed, there seems to be

a general trend that the more detailed catch statistics that are available for the analysis, the more pronounced the phenomenon.60

8.6 FISHING IN BALANCE

An important aspect of ‘‘fishing down the food web’’ is that we would expect to get higher catches of the more productive, lower-trophic-level catches

of prey fishes in return for the loss of less productive, higher-trophic-level

catches of predatory fishes With average trophic transfer efficiencies of 10% between trophic levels in marine systems,52we should indeed expect, at least theoretically, a ten-fold increase in catches if we could fully eliminate predatory species and replace them with catches of their prey species

To quantify this aspect of ‘‘fishing down the food web’’ an index, termed

‘‘fishing in balance’’ (FiB) has been introduced.61The index is calculated based

on the calculation of the PPR index (see Equation 8.1):

FiB ¼ log Cy
1

TE

TL y

C1
1 TE

TL 1

ð8:2Þ

where, Cyand C1are the catches in year y and the first year of a time series, respectively, and TLy and TL1 are the corresponding trophic levels of the catches; TE is the trophic transfer efficiency (10%) The index will start at unity for the first year of a time series, and typically increase as fishing increases (due

to a combination of spatial expansion and ‘‘fishing down the food web’’), and then often show a stagnant phase followed by a decreasing trend During the stagnant phase where the FiB index is constant, the effect of lower-trophic-level of catches will be balanced by a corresponding increase in catches lower-trophic-level A decrease of 0.1 in the trophic level of the catches will as an example be balanced

by a 100.1(25%) increase in catch level There has so far been few applications

of the FiB index,62 but indications are that the index has some potential by virtue of being dimensionless, sensitive, and easy to interpret

Figure 8.2 Illustration of ‘‘fishing down the food web’’ in which fisheries initially target

high-trophic-level species with low catch rates As fishing intensity increases catches shift toward lower-trophic-level species At high fishing intensity it has often been observed that catches will tend to decrease along with the trophic level of the catch (backward-bending part of curve, starting where ‘‘crisis’’ is indicated).

8.7 APPLICATION OF INDICATORS

We illustrate the application of indicators by presenting accessible information for the North Atlantic Ocean, defined as comprising FAO Statistical Areas 21 and 27 The North Atlantic was the initial focus area for the Sea Around Us project through which information about ecosystem exploitation and resource status has been derived for the period since 1950.4,63– 65

During the second half of the twentieth century, the catches increased from

an already substantial level of 7 million metric tonnes per year to reach double this level by the 1970s, but it has since declined gradually (Figure 8.3) Catch composition changed over the period from being dominated by herring and large demersals to lower-trophic-level groups, with high landings of fish for fish meat and oil The biomass of higher-trophic-level fish in the North Atlantic has been estimated to have decreased by two-thirds over the past half century.4

8.7.1 Environmental and Habitat Indicators

There are indications, notably from the continuous plankton recorder surveys, of decadal changes linked to the atmospheric North Atlantic Oscillation Index, causing marked changes in productivity patterns as well as zooplankton composition.66Overall, the changes do not have consequences for ecosystem health, but they change the background at which to evaluate health, and as such should be considered

Figure 8.3 Total catches and catch composition for the North Atlantic (FAO Areas 21 and 27)

estimated based on information from FAO, ICES, NAFO and national sources Source: http://www.seaaroundus.org