CRTR_ConnectivityHandbook

8/6/2019 CRTR_ConnectivityHandbook

1/80

Preserving Ree Connectivity

A Handbook or MarinProtected Area Manager


2/80

Citation: P.F. Sale, H. Van Lavieren, M.C. Ablan Lagman, J. Atema, M. Butler, C. Fauvelot, J.D. Hogan, G.P. Jones,

K.C. Lindeman, C.B. Paris, R. Steneck and H.L. Stewart. 2010. Preserving Ree Connectivity: A Handbook or Marine

Protected Area Managers. Connectivity Working Group, Coral Ree Targeted Research & Capacity Building or

Management Program, UNU-INWEH.

Edited by:Lisa Benedetti

Cover photo: Commonwealth o Australia (GBRMPA)

ISBN: 978-1-9213-17-06-4

Product code: CRTR 004/2010

Editorial design and production: Currie Communications, Melbourne, Australia, May 2010.

Coral Ree Targeted Research & Capacity Building or Management Program, 2010.

CRIOBEEPHE-CNRS

POLYNESIE

E

P H

E

E

H

E


3/80

Contents

Preserving Reef ConnectivityA Handbook for Marine Protected Area Managers

Acknowledgements 4

The decline o the coastal ocean and why this handbook exists 5

How to use this handbook 6

1. What is connectivity? 7

2. What processes cause connectivity? 11

3. Using connectivity in management 25

4. The science o connectivity 39

5. Integrating connectivity with management today 59

Reerences 68

Appendices 73

AcronymsKey defnitionsCRTR Connectivity Working Group membersAuthors


4/80


4

AcknowledgementsThis handbook is a product o the Coral Ree Targeted Research & Capacity Building orManagement Program (CRTR) an international development project unded by the GlobalEnvironment Facility (GEF), implemented by the World Bank, and executed by the Universityo Queensland, and numerous partners including the United Nations University Institute orWater, Environment and Health (UNU-INWEH), which managed the Connectivity Working Group.I thank the many members o the CRTR Connectivity program who acted as authors, provided

images or advice, or helped in other ways to bring it to ruition.This handbook has been produced by the CRTR Connectivity Working Group, with the assistance oUNU-INWEH and CRIOBE (Le Centre de Recherches Insulaires et Observatoire de lEnvironnementde Polynsie Franaise), which hosted a workshop: Connectivity in Coral Ree Systems Lessons toDate and Goals or the Future in Moorea, French Polynesia, March 2009, during which planning othis handbook was nalized. We also beneted rom discussions, provision o detailed inormationand images by the ollowing workshop participants: Jess Ernesto Arias Gonzlez, CINVESTAV-Unidad Merida, Mexico; Paul H. Barber, University o Caliornia, USA; Michael Berumen, WoodsHole Oceanographic Institution, USA; Brian Bowen, University o Hawaii, USA; Michael L. Domeier,Marine Conservation Science Institute, USA; Ccile Fauvelot, Universit de Perpignan, France;Daniel Heath, University o Windsor, Canada; Serge Planes, Universit de Perpignan, France;Tonya Shearer, Georgia Institute o Technology, USA; and Hannah L. Stewart, Department o

Fisheries and Oceans, Canada. I would also like to thank Alina M. Szmant, University o NorthCarolina at Wilmington, USA.

I am also grateul to all members o the Connectivity Working Group who have consistentlyworked to ensure that the inormation we provided has been accurate and up-to-date. I thankGabrielle Sheehan at Currie Communications and Adam Cusack at Cusack Design or theirpatience and their creativity in getting this handbook designed and nalized, Melanie King at theUniversity o Queensland, who worked many miracles, and particularly Hanneke Van Lavieren andLisa Benedetti at UNU-INWEH, who worked tirelessly as nal editors to turn the drats into thispolished, proessional product.

Peter F. SaleUNU-INWEH


5/80

5


The decline o the coastal ocean and whythis handbook existsThe coastal ocean environment provides enormous value in shery and other products, as well asecosystem services like coastal protection, water purication, and locations or ports, harbors, urbancenters, tourist destinations, and numerous recreational pursuits. Coastal environments can alsocleanse the soul, stimulate the mind, and restore the body. But 40% o all people live within 50 km oa coast, and our enthusiasm or coastal living is creating ever more environmental damage.

Unortunately, current management practices in most coastal regions are ineective, and tocontinue them will endanger the coastal economies and ecosystems that support over one halo the worlds population. The trend or coastal ocean ecosystems over recent decades has beenone o progressive decline in the ace o growing human population, rising demand or coastalresources, and increasing use o the coastal environment. Today, climate change is adding to thepressures on the coastal environment, urther stressing ecosystems there.

The decline o coastal environments has become a particularly signicant problem or manytropical countries with coral rees. In these areas, rees oten contribute to the major componento GDP because o their importance to tourism and sheries. They also provide an importantprotein ood source and help support a traditional way o lie or coastal peoples.

This handbook tackles one specic concern when contemplating eective management o coastalmarine environments the issue o connectivity. Marine protected areas (MPAs) have become animportant management tool, particularly in tropical regions, and connectivity is an importantconsideration in the eective design o MPAs and MPA networks. Connectivity issues are alsoinvolved in most other aspects o coastal management or two reasons: rst, water moves andtransports items such as sediments, nutrients and pollutants considerable distances; and second,most marine organisms also move within the water stream, transporting themselves betweenplaces. Our goal is to assist MPA managers and others in understanding and applying the concepto connectivity in their work. In this way, we hope to help managers strengthen their ability totackle the challenging task o sustaining coastal marine environments. This would help protectsheries and other goods and services they provide.

Figure 1. The coastal ocean environment provides enormous value in shery and other products, as well as ecosystem serviceslike coastal protection, water purication, and locations or ports, harbors, urban centers, tourist destinations, and numerousrecreational pursuits. Coastal environments can also cleanse the soul, stimulate the mind, and restore the body. Photo: Hanneke

Van Lavieren

5


6/80


6

How to use this handbookThis handbook contains a summary o the science o coral ree connectivity and guidance on howto use this inormation to aid in making management decisions. Although it has been written orcoral ree managers, decision makers and others who may be involved in ree management eorts,

the science discussed is relevant to managers o coastal waters in all oceans. Much o the scienceo connectivity remains to be discovered, however, substantial scientic research eort is currentlyunderway to address knowledge gaps and translate this science into practice or improving reemanagement.

This handbook describes what we mean by connectivity, and discusses the various uses o thisterm. Most attention is given to populational connectivity the extent o connection among localpopulations o a species because it is both the most dicult type o connectivity to deal with andbecause it is the least eectively used in current management practices. Populational connectivitycomes in two orms: evolutionary (genetic) connectivity and demographic (ecological) connectivity.The rst is concerned with genetic dierences in dierent populations o the same species. This canbe inormative when considering long-term (evolutionary) and large-scale biogeographic dispersalpatterns o organisms. It can also be useul or managers wanting to assess the genetic uniqueness o

populations when making decisions concerning biodiversity preservation. In contrast, demographicconnectivity involves the extent o linkage that occursamong nearby local populations o a species due tothe exchange o individuals. This type o connectivity ismost important or marine protected areas (MPAs), andparticularly no-take shery reserves (NTRs), when makingdecisions concerning design and management, andwhen trying to determine the optimum amount o reehabitat to protect when conservation or precautionarysheries management is the objective. Other ormso connectivity relate to the transmission o nutrients,pollutants, or other items between locations, by passivetransport via water currents. These are also important or

managers, but easier to understand and apply becausetransmission is due solely to physical processes.

This document provides a summary o what is currentlyknown about the science o connectivity and thetechniques and tools used or measuring connectivityor dierent types o organisms (e.g., corals, sh andlobster). It also highlights the gaps in our knowledgeand oers suggestions and advice on how to use whatconnectivity inormation is available. A strong plea ismade or scientists and managers to establish closeworking collaborations and use management activitiesin an adaptive management context to simultaneously

advance the scientic understanding o connectivity,while also using the best available knowledge to guidecurrent management decisions.

This handbook has been written to make the science as accessible as possible to managers withvarying levels o scientic background or expertise. For those who do not have time to read the entiredocument, key points are summarised on Message boards. Also provided is a list o useul contactswithin the Coral Ree Targeted Research & Capacity Building or Management Program (CRTR) andcitations to relevant scientic literature or those who wish to delve urther into the currently activeeld o connectivity research.

Figure 2. South Water Caye, Belize. Photo: Ron Schaasberg


7/80

Section 1What is connectivity?

In this section you will fnd:

The particular importance o populational connectivity

Marine Protected Area, Apo Island, Philippines. Photo: Gidi Levi


8/80


8

1. What is connectivity?Coral rees are patchily distributed habitats ound throughout oceanic environments which providea mechanism or transport among them. So are mangroves, seagrass beds, and other coastalenvironments. Each local patch o any o these environments will support populations o particular

organisms i they are big enough to do so. Thus, the patchy distribution o habitat results in a pattern onumerous and more-or-lessisolated local populations o each species characteristic o that region, i.e.,more-or-less isolated because individual coral rees and other habitat patches are seldom so remotethat there is no movement o organisms among them. This movement is one orm o connectivity.

Connectivity is the fux o items between location types that are the same or dierent (e.g., rees and/or seagrass beds). It exists or nutrients, sediments, pollutants, and individual dispersing organisms,i.e., any item that has the potential to move among and between rees and other environments.In the context o coastal management, the eective transer o individuals (usually pelagic larvae)

between local populations is one o the most important, and certainly the most dicult orm oconnectivity to quantiy. While the transer o non-living materials, such as sediments or pollutants, islikely to be determined primarily by local and regional hydrodynamics, we know that the transer oorganisms is more complex. This is because passive transport will likely be modied by the sensoryand behavioral capabilities o individual larvae. Eective transer among populations also requiressuccessul establishment within breeding populations, so connectivity among populations cannot bemeasured by ocusing on dispersal patterns alone but must also include successul recruitment to thereceiving population.

Box 1. Types o populational connectivityPopulational connectivity comes in two orms:

1) Evolutionary (genetic) connectivity: the amount o gene fow occurring among populationsover a timescale o several generations. It determines the extent o genetic dierences amongpopulations.

2) Demographic (ecological) connectivity: an exchange o individuals among local populationsthat can infuence population demographics and dynamics. It can include:

Exchangeofoffspringbetweenpopulationsthroughlarvaldispersal;

Recruitmentofjuvenilesandsurvivalofthesejuvenilestoreproductiveage;

Anylarge-scalemovementofjuvenilesandadultsbetweenlocations.

Figure 3. Patchy ree ormation Great Barrier Ree, Australia. The patchy distribution o coral ree habitats subdivides populationson many spatial scales. Photo: Ove Hoegh-Guldberg


9/80

9

1.1 The particular importance o populational connectivityThe transer o individuals between populations allows or the transer o genes. It is thereore useulto make a distinction between the two kinds o connectivity that infuence populations, evolutionaryand demographic connectivity. Relatively low exchange o individual organisms can still allow or

a sucient level o gene transer and thus can result in genetically similar populations. Whereasat exceptionally low levels o exchange, populations tend to slowly diverge genetically throughprocesses like genetic drit, mutation, and dierential selection. Over time, these populations canbecome separate species.

The low levels o exchange that maintain genetic similarity among neighboring populations is calledevolutionary (genetic) connectivity. This exchange, perhaps one or two individuals per generation, isusually ar too low to have any measurable eect on population growth rates i.e., demographically,they are insignicant exchanges. At somewhat higher rates o exchange, populations remain quitesimilar genetically, and the rates o arrival and departure o individuals are high enough that they havea measurable impact on the rates o growth or each population. In these cases, we are reerring todemographic (ecological) connectivity.

Evolutionary and demographic connectivity are equally important considerations in coastal

management, but they are important in quite dierent ways. A manager whose primary goal isbiodiversity conservation will be particularly interested in the patterns produced by evolutionaryconnectivity. That is, conservation decisions are requently based on whether a particular populationis taxonomically unique; absence o evolutionary connectivity usually permits this. As well, patternso evolutionary connectivity among locations can help reveal underlying patterns o gene fow, whichmay reveal likely biogeographic events in the recent past or near-term uture.

When demographic connectivity exists amongst populations, they can infuence each others patternso growth or decline. This occurs when the number o individuals exchanged per generation is greatenough to have a measurable impact on the population growth rate in one or each o the exchangingpopulations. A primary concern or many managers is ensuring that sheries are sustainable orthat coral rees which are being managed or tourism can continuously support the normal rangeo species. These managers will be primarily interested in demographic connectivity. MPAs knownas no-take shery reserves should be designed with due consideration or this type o connectivity,as should networks o such reserves.

Message board

TheuseofMPAsandMPAnetworksasamanagementtoolhasbecomewidespread,particularlyintropicalregions,andconnectivityisconsideredacriticalcomponentin their design.

Evolutionaryanddemographicconnectivityareequallyimportantconsiderationsincoastalmanagement,buttheyareimportantinquitedifferentways.

Inthecontextofcoastalmanagement,theeffectivetransferofindividuals(usuallypelagiclarvae)betweenlocalpopulationsisoneofthemostimportant,andcertainlythemostdifcultformofconnectivitytoquantify.

Muchofthescienceofconnectivityremainstobediscovered,however,substantialscientifc research eort is currently underway to address knowledge gaps andtranslate this science into practice or improving ree management.


10/80


10


11/80

11

Section 2What processes cause connectivity?


Water moves, oten in mysterious ways

Most marine organisms have pelagic larvae

Many marine organisms move about ater larval lie is over

Colony oFavites halicora spawning. Photo: James Guest


12/80


12

2. What processes cause connectivity?

2.1Watermoves,ofteninmysteriouswaysThe marine environment is bathed in water, a medium which is seldom at rest. Movement o water

can transport items, such as plants and animals, rom one place to another. Organisms such as kelp,oysters or corals, which are securely astened to the substratum, may not move, but water will fowpast and provide them with ood and nutrients. Organisms that are not securely astened, such assh, jellysh, killer whales or crabs, may be transported by the masses o water within which they areswimming. Indeed, in the open ocean, i an organism does not possess the equivalent o a GlobalPositioning System (GPS) x o its location, or an external reerence such as the view o a distant island,it cannot sense that it is being transported.

Movement o oceanic water is brought about by various actors; the earths rotation, wind, tides, andriction against continental margins. It is also aected by changes in water salinity or temperature.That is, water o a particular salinity and temperature will have a specic density, and so a masso water o the same salinity and temperature tends to move as a single unit. Adjacent watermasses that dier slightly in temperature or salinity can remain as distinct layers over considerableperiods o time until mixing at edges averages out dierences and causes them to merge together.As surace waters warm because o heat rom the sun, they become less dense and rise while coolerwaters sink. At the same time, evaporation due to heat rom the sun causes surace waters to becomemore saline. Increasing salinity makes water more dense and likely to sink below deeper layers.

Putting all these actors together, we can see the ocean as a complex o adjacent patches o watermoving relative to each other, both horizontally and vertically. The scale o these patterns o movementbegins with the tiniest eddies only centimeters in size, to broad-scale, long-lasting currents or rotationalgyres (large eddies) that can be hundreds o kilometers wide and travel thousands o miles. The GulStream is an example o an enormous river o ocean water that moves rom the Caribbean throughto the Florida Strait, up along the eastern coast o North America, and ultimately to the shores onorthern Europe. It plays a major role in transporting heat rom the tropics towards the poles, as docomparable large-scale currents in other ocean basins.

Figure 4. Gyre or eddie ormation behind a ree, Bowden Ree, Australia. Oceanographic processes such as these gyres andeddies and their variability over time and space greatly determines the patterns o connectivity through larval dispersal amonglocations. Photo: James Oliver, Ree Base


13/80

13

When a moving water mass comes into contact with a continental margin, a mid-ocean island, ora coral ree, rictional orces modiy the patterns o movement resulting in upwellings, reraction owaves, and places o intense wave action or calm lee conditions. The complexity o ocean movementhas become increasingly apparent with the development o more sophisticated ocean observationinstruments. Broad-scale surace movement patterns are readily seen with various satellite imaging

systems, while vertical movements can be detected using a variety o devices that can be deployedon the ocean foor, moored in midwater or towed by vessels. Along with the improved understandingo hydrodynamic patterns and processes has come greater ability to accurately model these patterns.This is the environment in which all marine organisms spend their lives.

Ocean movement becomes most complex near coastlines as this is where the orces moving parcelso water come up against the relatively immovable substratum and shoreline (e.g., shel edges,rees, banks, islands, headlands, and beaches). This interaction creates upwellings, reraction andbreaking o waves, and transport o sediments via long-shore currents. River discharge introducesless saline water into oceanic waters. It rst foats above the more saline layers, but then slowly mixesvia eddy diusion. Discharge rom large rivers, such as the Orinoco in the Caribbean, can generatea plume o low-salinity surace water that extends thousands o kilometers out rom the river mouth,and transports sediments, nutrients and pollutants, as well as dispersive phases o some organismsacross those distances. On coasts lacking rivers, there may be extensive, but more diuse dischargeo resh water (and associated nutrients and pollutants) via surace run-o or groundwater. In manyree regions, coastal landscapes are made up o heavily eroded limestone (oten ossil rees), riddledwith underground streams that can discharge large quantities o reshwater several kilometers outrom shore. This creates upwellings that are sometimes visible rom the surace. Tidal patterns, whichare essentially slow period waves, are also distorted by interactions o the water mass with shallowbathymetry or a shoreline. This interaction results in variations in tidal height and timing rom place toplace along coastal areas. Tides alter local sea level on a regular diurnal cycle, the extent and patterno which depend upon the location, and these alterations can modiy patterns o water movementdue to currents and waves as water becomes alternately deeper or shallower. It is the integratedresult o each o these separate processes that determines the actual movement patterns o waternear a ree or coast, and the manner in which water moves will determine the patterns o connectivityamongst locations.

2.2MostmarineorganismshavepelagiclarvaeWith the exception o some large predators, marine organisms o rees and other coastal habitats arerelatively sedentary throughout the majority o their lives. While larger whales and sharks, and someturtles, can travel distances on ocean basin scales, many common ree sharks and larger groupersspend their lives moving kilometers rather than hundreds o kilometers. There are also numerous smallree shes that remain within the space o an average living room or an entire liespan. For example,many small gobies that are commensals on branching acroporid corals, and some damselsh thatshelter among coral branches, spend their lives within the immediate vicinity o a single coral colony.In addition to these mobile but relatively sedentary species, ree environments eature a wide rangeo species that are sessile the corals themselves, and a wide variety o taxa including tube worms,sponges, barnacles, ascidians, and algae permanently attached to the substratum.

This sedentary or sessile liestyle is abandoned during early lie as the great majority o ree speciesexperience pelagic larval stages and produce pelagic eggs. When eggs are shed into the water column,and larvae remain in the mid-water layers or days to weeks, extensive dispersal is very likely. Indeed,a widely accepted argument or why ree organisms produce pelagic larvae is that this is essentialor dispersal. In a world which changes over time, the organism that is more capable o dispersingospring is most likely to persist because no site remains permanently suitable or occupancy by anyparticular species.


14/80


15/80

15

2.2.2 Larval behaviorThis section briefy reviews the biology o larval stages. Far rom being embryonic or developingorganisms, larvae are ully unctional, well adapted to pelagic lie, and selected or abilities whichallow them to nd suitable juvenile habitat at the end o larval lie.

Pelagic eggs behave much like small particles with a setbuoyancy. On calm days, corals eggs are positively buoyantand can orm a visible scum at the waters surace. Newlyhatched larvae are usually quite limited behaviorally, butnot incapable. For example, coral planula larvae are ableto modiy buoyancy, thereby moving higher or lower inthe water column, and perhaps able to take advantageo current patterns at specic depths. Among sh, newlyhatched larvae are both physically weak and small enoughthat water viscosity becomes a major actor in determiningsinking rates or mobility. Many possess greatly elongatedn rays or other laments that likely unction to impedesinking through this viscous medium (Leis 1991).

Even when quite young, larvae can be ound at specicdepths (and change depth according to time o day andage). This indicates that they are capable o adjustingbuoyancy and can thereby move vertically within masseso water. However, coral ree larvae do not remain smalland behaviorally limited. Although they are largely at themercy o water movement at early stages, as they growthey develop limited locomotory capacity and an ability to control buoyancy permitting verticalmovement, potentially permitting selection o water masses moving in specic directions. While mostree sh species have larval lives that last about one month, some remain in larval orm or up to threeor our months. Surgeonsh larvae, at the end o their 2-3 month larval lie, can swim at speeds o36 to 42 cm per second. When maintaining 13.5 cm per second, they can continue swimming or

over 194 hours without ood, eectively covering a distance o 94 km (Stobutzki and Bellwood 1997,Hogan et al. 2007). There is also some evidence that they can swim in specic directions, changingorientation according to particular cues.

Figure 6. A Caribbean spiny lobster (Panulirus argus)larva. These long-lived larvae spend over six months inthe plankton, during which they are potentially dispersedthousands o kilometers. Yet, recent connectivity researchshows that their vertical migratory behavior may reducethis dispersal to a ew hundred kilometers, which in turnmay double their successul settlement in coastal nurseries.Photo: Mark Butler

Figure 7. This gure summarizes the spawning and larval/postlarval dispersal and behavior o Caribbean spiny lobster (Panulirusargus), and indeed most spiny lobsters. Larvae hatch rom eggs that are carried by adult emales to ree edges at night while allingtides disperse the hatched larvae oshore (white circles). Early stage larvae are attracted to light, so remain in the surace waters(< 50 m) although they move up and down in the water column each day in response to light (diurnal or diel vertical migration).Late stage larvae avoid light, and remain in deeper waters (> 50 m), an age dependent behavior reerred to as ontogeneticvertical migration, but also engage in diurnal vertical migration with greater amplitude because they are stronger swimmers.Near continental shel edges, larvae metamorphose to the last stage puerulus postlarvae. These larvae are transported by tidesbut also swim towards coastal nurseries ollowing chemical signals. Credit: Mark Butler

0

50

100

150

Shel

Ree/Nursery

Late Stage Larvae

Early Stage Larvae

Post Larvae


16/80


16

Fish are not the only ree organisms that show remarkable changes in behavioral ability during larvallie. The Caribbean spiny lobster (Panulirus argus) passes through more than twenty moults over itslong six month larval lie, during which its preerence or depth and pattern o daily vertical migrationchanges as it develops (Goldstein et al. 2008). The nal larval stage, the puerulus postlarva, is a non-eeding, rapidly swimming phase capable o swimming or 2-4 weeks at speeds up to 15 cm per

second. It does so while negotiating a path that can be tens o kilometers long rom the open oceanto vegetated coastal nurseries, which it detects using chemical cues (Goldstein and Butler 2009).

2.2.3Whatlarvaesee,hear,smell,andtastethecuestondingreefsA dispersive phase would not be very adaptive, nor make sense, i larvae were to only drit passivelyor swim in random directions. Coral rees are not common and occupy a mere 0.1% o the worldsoceans, and we should expect that the larvae o ree and other inshore species will have well developedsensory capabilities able to detect suitable habitat by the time they complete larval lie. However,identiying these sensory capabilities is not simple because late-stage larvae are at a period wheredevelopment is very rapid, and many undergo substantial metamorphosis as soon as they reachinshore habitat. This makes studies o their physiology and behavior very dicult because they ceasebeing larvae almost as soon as they are caught! Nevertheless, scientists have been able to make some

progress in this area o research.Even towards the later stages o larval lie, coralplanula larvae only have limited locomotory ability.However, they do show discriminatory capacity andclear preerence or some substrata over others assites or settlement. This discriminatory capability,also common in other invertebrates (e.g., barnaclesand oysters), is due to their ability to respond tospecic chemical cues rom suitable substratum inorder to successully settle and attach permanently.

Among shes, there is limited but growingevidence that they can use hearing and odor in

the selection o suitable juvenile habitat. This islargely based on behavioral research where larvalsh are given a choice and subsequent responsesare observed. Some o this work has involvedplacing simple foating Y mazes in marine studysites, and later putting larvae into the structureto test whether they swim in the direction o theree. This would iner whether they are able todetect the rees presence.

Some physiological studies have conrmed that late stage larvae have ears that detect noise(chiefy breaking waves) created by rees. Recent studies by Gerlach et al. (2007) in the southernGreat Barrier Ree, have very convincingly demonstrated that larval cardinalshes (Apogonidae)

and damselshes (Pomacentridae) are able to detect the odor o ree water and actively chooseto swim in water rom their home ree rather than in water o the open ocean or neighboring rees(Atema et al. 2002, Gerlach et al. 2007). It has also been ound that this ability is used dierentlyby the two species. For example, genetic analyses o the population structure o several nearbyrees demonstrated that the Doederleins cardinalsh, Ostorhinchus doederleini, and not the neondamselsh, Pomacentrus coelestis, shows strong homing behavior to particular rees (Gerlach etal. 2007). Presumably, the damselsh uses its home odor recognition to discriminate ree waterrom non-ree water to help nd its way back to ree habitat. Both these amilies care or theireggs, and the ospring only become pelagic at hatching, the time at which they are rst exposedto environmental odor and have a unctional olactory organ. Thereore, this ability to recognize andrespond to home odor may be a orm o imprinting. Similar imprinting may cause the strong

Figure 8. The nal peurulus postlarval stage o the Carribean spinylobster (Panulirus argus). The lobster larvae uses chemical andpressure cues to locate back-ree nurseries as it swims rom the opensea to complete its complex lie cycle. Photo: William Herrnkind


17/80

17

attraction o clownsh larvae, Amphiprion percula, to the odor o tree leaves, which is thought tohelp them identiy a preerred inshore settlement habitat (Dixson et al. 2008).

It should not be surprising that cardinalsh show selective preerence or the home ree. Much earlierresearch, also done in Australia, has shown that damselsh o the genus Dascyllus, which exist as smallgroups occupying single heads o branched coral, are capable o discriminating and avoring coral

colonies containing conspecics rather than other species oDascyllus, or no sh at all, when settlingto juvenile habitat at night. Related experiments using the Y maze showed that sh could detect andrespond to the odor o conspecics (Sweatman 1988).

While there is clearly an enormous amount to learn about how pelagic larvae nd home rees,the results o investigations to date are clear. Dispersive pelagic larvae do not drit aimlessly in theocean. They use their varying behavioral and sensory capabilities to minimize the extent o dispersal,and in many species, are active agents in ensuring successul return to ree habitat, and to specicmicrohabitats that will be suitable or juvenile lie.

2.2.4 Connectivity through larval dispersalThe act that most ree species experience a pelagic larval phase means that the majority o adultree organisms exist as local breeding groups (local populations) that occupy suitable habitat andare predominantly inter-connected by larval dispersal. On scales o tens o kilometers or less, there isconsiderable mixing as larvae disperse rom one population to another. However, on scales o hundredso kilometers, populations are largely isolated demographically (though still linked genetically).The details o the patterns o dispersal and larval exchange vary among species, so that some taxadisperse only over quite limited distances, while others disperse more widely. At present we only havelimited inormation detailing these dierences, yet it is already clear that dierent species breeding atthe same time, and in the same location, can show markedly dierent dispersal patterns during larvallie. This can be attributed to variations in larval phase duration, behavior, and sensory capabilities(Gerlach et al. 2007).

Figure 9. The larval phase o coral ree shes was once considered a black box. Larvae come in all shapes and sizes and seemto be designed or going places, but what they do and where they go in the 3-dimensional oceanic environment has beenshrouded in mystery. Through the use o new technologies, these mysteries are rapidly being solved. Credit: C.M. Guigand &R.K. Cowen, Rosenstiel School o Marine and Atmospheric Science, University o Miami


18/80


18

The details o the patterns o dispersal and larval exchangevary among species, so that some taxa disperse onlyover quite limited distances, while others disperse morewidely.

Some important conclusions can be drawn with respect to larval dispersal. The rst is thatconnectivity amongst populations o ree species is primarily, or (or sessile species) exclusively,due to dispersal during larval lie. Secondly, or the majority o ree species that have been studied,demographic connectivity has been shown to act on scales o up to tens o kilometers, rather thanon scales o hundreds o kilometers or more. Thereore, the concept o a demographically well-connected population across the Caribbean, or along the length o the Great Barrier Ree, doesnot apply. Genetic (evolutionary) connectivity operates at larger spatial scales because the rareindividual larva will occasionally get transported ar beyond its usual dispersal range. I MPAs areintended to play a role in sheries management, the smaller scale o demographic connectivityshould be taken into account in the design o MPA networks. This type o connectivity can alsobe inormative when considering extensive ree destruction caused by bleaching, crown-o-thornoutbreaks, and major hurricanes, because it denes the distance over which natural re-seeding o

ree habitat is likely to occur.

Message board Thefactthatsometimesarebetterthanotherstoplaceeggsorlarvaeintothe

watercolumn,andtheassociatedfactthatreproductiveeffort isnormallymoresuccessfulwhenmembersofthesamespeciesreproduceatthesametime,hasresulted in many species exhibiting precise timing o spawning activities.

Manyreeforganismsshowremarkablechangesinbehaviorandappearanceduringlarval lie.

Amongshes,thereislimitedbutgrowingevidencethattheycanusehearingandodorintheselectionofsuitablejuvenilehabitat.

Manyreeforganismsshowdiscriminatorycapacityandclearpreferenceforsomesubstrata over others as sites or settlement. They oten do this by responding tothe chemical characteristics o a surace.

Connectivity amongst populations of reef species is primarily, or sometimesexclusively,duetodispersalduringlarvallife.

Formostreefspecies,demographicconnectivityhasbeenshowntoactonscalesofup totensofkilometers, ratherthanonscalesofhundredsofkilometersormore.IfMPAsareintendedtoplayaroleinsheriesmanagement,thesmallerscaleofdemographicconnectivityshouldbetakenintoaccountinthedesignofMPAnetworks.

2.3 ManymarineorganismsmoveaboutafterlarvallifeisoverThe largest part o the lives o most marine organisms is not spent as larvae at sea, but as juvenileand adult orms, associated with the bottom in various manners throughout lie. Lie ater settlementrom the larval phase varies developmentally and ecologically among dierent species. I not eatenby a marine predator or man, individuals o many species can survive or decades and occupy manydierent habitats throughout lie.


19/80

19

2.3.1 Settlement and recruitmentThe transition rom a pelagic oceanic environment to a benthic ree habitat, during which therelationship between the organism and its environment changes radically, is a particularly dangerousphase in any marine organisms lie. Settlement o larvae to ree habitats occurs in many dierent waysamong shes and invertebrates and is typically sporadic, nocturnal and/or cryptic. This parameter

is dicult to measure so ecologists tend to sample recruitment (animals which settle and survive)quite soon ater settlement. The term recruitment, in the broadest sense, means the addition o newindividuals to populations or to successive lie-cycle stages within populations. In more specic terms,recruitment can have several dierent meanings:

Larval Recruitment: New individuals being added to a population by arrival o incoming larvae tobottom habitats. Figure 10a. Newly settled larvae (8-12 mm). Photo: D.B. Snyder

Inter-habitatRecruitment:Individuals arriving at a later-stage habitat - not the rst larval settlementevent, but a later habitat shit. Figure 10b. Photo: Gerald Nowak/WaterFrame/Specialist Stock

Fishery Recruitment: Individuals reaching a size at which they are rst retained by specied shinggears (i.e., when they enter the shery). This can occur many years and habitats ater larval recruitment.Figure 10c. Photo: Photoshot/VISUM/Specialist Stock

Following the settlement stage, movement between habitats may happen one or more times during thedevelopment o some species, including grunts, and sometimes never in others like damselshes.

Credit or gure: Kenyon C. Lindeman


20/80


20

2.3.2MovementbetweenhabitatsandcoastaldevelopmentMany coastal areas contain a wide array o habitats, including vegetation, hardbottom, or rees, whichoccur along shallow to deep gradients along coastal shelves. Some remarkable connections existamong marine animals and habitats, especially when considering the lie cycles o the shery specieswe humans eed upon. For some species, a single habitat within a complex seascape is sucient

to complete an entire lie cycle. Many other species however, move between habitats at dierenttemporal and spatial scales. Some habitats are critical to the early developmental stages, survivaland growth, o many species o sh, lobster, and shrimp, while others serve as spawning and eedinggrounds. Marine organisms may also make repeated migrations between habitats on various timescales, especially daily and seasonal. Daily shits typically involve nightly eeding migrations betweeneeding and resting habitat every 12 hours. For example, daily movement and habitat use patternshave been shown or goatsh (Mullidae) and grunts (Haemulidae), which undertake crepuscularoraging migrations between daytime ree and nocturnal sand fat habitats (Meyer et al. 2000).In some sh species, these daily shits can lead to direct transer o nutrients between seagrass eedinghabitats and mangrove and ree resting habitats.

Some remarkable connections exist among marine animals

and habitats, especially when considering the lie cycleso the fshery species we humans eed upon.

Adult population size depends upon the successul survival o developing, bottom-associated earlylie stages. Even under the best natural conditions, individuals at these stages are oten subject toextremely high mortality rates. Predators requently eed on nocturnally migrating prey, and anyhuman-caused disruption to pathways between habitats can increase mortality rates. Access to shelterand ood provided by critical inshore habitats is essential or survival. Unortunately, important habitats(e.g., nurseries or those visited or daily eeding) used by the youngest shes and other ree organismsare oten in shallow areas that are vulnerable to human impact.

Figure 11. Spiny lobster (Panulirus argus), the most important sheryresource in the Greater Caribbean, changes habitats several timesduring growth. Larvae settle in shallow vegetation while juvenilesmigrate to hard bottom habitats, and eventually to deeper rees.Other important species in the Caribbean that move among habitats,oten into deeper waters, over the course o post-settlement lieinclude groupers, snappers, conch, sea turtles, porgies, parrotshes,grunts, and jacks. Photo: Mark Butler

Figure 12. The Queen conch (Strombus gigas) is an importantshery species in the Caribbean. The positive eect o reserves isnot conned within the borders o a reserve because conchlarvae produced within reserves have been ound to drit outside oreserve boundaries and seed surrounding areas (Stoner et al. 1996).Photo: Ron Schaasberg


21/80

21

Many coastal areas in coral ree regions are beingdeveloped or tourism with a ocus on rapid andspeculative coastal growth. The Caribbean orexample, represents some o the worlds mostconcentrated coastal tourism, with places like

Cancun, Mexico, at the doorway to the southeastU.S., processing 5 million tourists annually.

Coastal development, pollution and natural eventscan work together to alter or damage importantinshore habitat used by developing shes, lobsterand other organisms, e.g., making inshore habitatno longer suitable or juveniles and disrupting vitalpathways between these and oshore habitats.Moreover, any negative impact during an organismsearly lie stages could indirectly aect the abundanceo adults and the ood webs they are embeddedin. In addition, although not well studied, modestalterations to coastal environments may disruptdaily or seasonal migratory patterns. This couldlead to reduced populations or local extirpation oshery species which could in turn impact sheriesthat operate in deeper waters where environmentalconditions appear unchanged.

Protection o crucial habitat required by developingsh species can be a very cost-eective managementapproach or enhancing shery production. Whenan MPA is designed to protect even just oneor a ew species, it is critical to have inormationconcerning the specic migration patterns andhabitat requirements o that species. In order to be

eective, the MPA or MPA network must be largeenough to encompass all these habitats as wellas the daily and seasonal migration routes o thespecies they aim to protect.

In nearly all regions, the use o the EnvironmentalImpact Assessment (EIA) procedure to assess thepotential threat o coastal development has beeninadequate and poorly managed. Causes or thisinclude:

Conclusionsarenotalwaysbasedonsoundscienticinformation;

Absenceofindependent,third-partypeer-reviewofdocuments;

Nocontrolovernaturalspatialortemporalvariations;

Lackofcommunityparticipation;

Poor coordination.

Given the oten poor quality o impact assessment, it is not surprising that many projects proceeddespite having severe deleterious impacts on connectivity, and ultimately the ecological unctioningo coastal marine systems. Political pressure that encourages development and corruption in theapprovals process is simply making this situation worse.

Figure 13. Dubai, UAE. Most marine shes and invertebrates usemore than one habitat throughout their lives. Coastal developmentcan make inshore habitats no longer suitable and disrupt vitalpathways between these and oshore habitats. Photo: iStockphoto


22/80


22

Box 2. TurbinariainFrenchPolynesiaconnectivityand colonization via dispersal o adults

While larval dispersal is o primary importance to ree species, dispersal o detached adult organismsthrough passive drit on ocean and wind-driven currents or by hitching rides on other driting organisms

can also play an important role in maintaining connectivity and colonization o new locations. Individualswhich have already reached adult stage have higher survival rates than larval orms. Also, since theyare reproductively mature, they can establish new populations immediately upon arrival in a newlocation. Adults are also generally larger and more evident to ree managers than minute larval orms,and their arrival into a new area may be more obvious. It is important to remember that invasive ornuisance species that disperse in this manner can also pose a serious threat to ree ecosystems.

For example, the current spread othe alga Turbinaria ornata acrossFrench Polynesia, shows the potentialor connectivity achieved by adultsand the challenges this can pose toree managers. This large widespread

Indo-Pacic macroalga traditionallyoccurred in only a ew areas withinthe French Polynesia. However, sincethe early 1980s it has been spreadingand becoming so abundant that it isnow considered an invasive speciesand is displacing coral in many reesthroughout this region (Stewart 2008).

Thalli o Turbinaria grow attachedto rees, but as they reach sexualmaturity, they become buoyant and

their attachment to the substratum

weakens (Stewart 2006). Followingstorms, large rats o detached thalli are blown away and drit rom island to island (Martinez et al.2006). Detached thalli are able to maintain ertility and viability even ater foating or 3 months(perhaps even longer) (Stiger and Payri 1999). During this time, ertilization events occur (motilemale gametes nd eggs in emale thalli) at least once a month. Young germlings are releasedrom the parent plant, and then become established successully across the ree, creating newpopulations o the alga. Examination o the genetics oTurbinaria reveals that there is very littlegenetic dierentiation across the French Polynesia, perhaps a result o the high connectivitybetween populations maintained by this driting dispersal mechanism.

It has been observed that a diverse assemblage o invertebrates and algae drit passively alongin association with rats oTurbinaria (Stewart and Meyer, unpublished data). As large foatingrats oTurbinaria are relatively recent in French Polynesia, it may present a new mechanism oconnectivity in the region. Researchers have only begun investigating the potential impact on theconnectivity o these associated species.

Increasing numbers o this alga are causing severe problems. In addition to shading, abrading andoutcompeting coral or ree space, foating thalli damage shing nets and sh harvest, clog motors,rot on beaches, and are negatively aecting the tourism industry. The impact on ree nutrient dynamicsdue to the increase in algal biomass has yet to be determined. Researchers have been searchingin vain or economic incentives to harvest this alga (e.g., cosmetic or pharmaceutical), and localshing groups are beginning to organize Turbinaria removals in an attempt to abate its spread.As rees increasingly ace shits rom coral to algal dominated systems, this type o whole adultconnectivity, that is characteristic o many algae, could become increasingly important to considerin coastal management.

Figure 14. A detached, foating rond o the tropical brown alga Turbinariaornata in the lagoon o Moorea, French Polynesia. Photo: Hannah L. Stewart


23/80

23

2.3.3 Spawning migrationsDuring yearly spawning migrations, the adults o many grouper, snapper, and other species, undergolarge scale oceanic movements. Some undergo migrations o hundreds o kilometers, although mosttravel much shorter distances; this can involve weeks o travel between diering habitats until suitablespawning sites are reached. These annual events utilize important bottom habitats, and specic sites

and routes, in order to broadcast egg and larval stages that are largely independent o the ree duringpelagic stages. The resulting spawning aggregations represent some o the most concentratednumbers o adult ree sh that can be seen around the world. Not surprisingly, these groups are verysusceptible to shing pressure. The ate o eggs and larvae generated rom these migrations cansubstantially determine the level o connectivity o shes among diering habitat systems. The relativedegree o such connectivity is a critical determinant in the population structure o a target species, anda key actor when developing coherent spatial management policies.

Box 3. Spawning aggregations and connectivityWith respect to spawning aggregations, connectivity occurs through two distinct mechanisms:

1) The movement o sh as eggs and larvae rom a spawning aggregation site to settlement sitesvia dispersal;

2) The movement o adults rom normal residence sites (= catchment area) to spawning sites.

Both must be studied to determine the relationship o a particular spawning aggregation or siteto its surrounding area.

Figure 15. Some sh species such as these Nassau groupers, Epinephelus striatus, gather at specic spawning grounds eachyear where they are extremely vulnerable to overshing. These spawning aggregation sites should be incorporated in no-takereserves to protect these sh at this vulnerable stage. Photo: Enric Sala


24/80


24

Message board Settlementoflarvaetoreefhabitatsoccursinmanydifferentwaysamongshes

andinvertebratesandistypicallysporadic,nocturnaland/orcryptic.

Remarkableconnectionsexistbetweenanimalsandhabitats.Theseconnections

are central to the ecological unctioning o coastal habitats and to the productiono their environmental goods and services.

Coastaldevelopment,pollutionandnaturaleventscanworktogethertoalterordamageimportantinshorehabitatsusedbydevelopingshes,lobsterandotherorganisms, e.g., making inshore habitats no longer suitable for juveniles anddisrupting vital pathways between these and oshore habitats.

WhenanMPAisdesignedtoprotectevenjustoneorafewspecies,itiscriticalto have inormation concerning the specifc migration patterns and habitatrequirementsforthosespecies.

Figure 16. This mature Sweetlips (Plectorhinchus albovittatus) has a ully ripe ovary that almost lls its body cavity, but wascaptured at a spawning aggregation site in Palau. Larger, older sh are notably ecund because their larger body cavities permita great expansion o ovary size as eggs mature. Photo: Patrick L. Colin


25/80

25

Section 3Using connectivity in management


Marine protected areas

MPA networks

What MPA networks cannot do

The value o coastal marine ecosystems

Marine park managers, Akumal, Mexico.Photo: Miguel Angel Maldonado, Centro Ecolgico Akumal


26/80


26

3. Using connectivity in management

3.1MarineprotectedareasFaced with widespread decline in ocean health, many nations are turning to marine protected areas

(MPAs) as a tool to manage the most important marine habitats and species. Many types o MPAshave been developed to serve dierent purposes in diverse ways. They can range rom no-takereserves (NTRs), which are small areas where all extractive activities (e.g., shing) are prohibited inorder to conserve target species or sensitive habitat, to extensive marine management areas (MMAs),which have a single comprehensive management plan that oten includes spatial zoning to permitdierent management tools, including NTRs, in dierent locations. MMAs are an attempt to integratethe management o many species, habitats, and uses within a specic region.

MPAs ll some or all o the ollowing roles:

Sustainsheriesbyprovidinginsuranceagainststockcollapse;actasabufferagainstrecruitmentfailure,andpossiblyalsoprovidecentresforpropaguleandadultdispersal tosurroundingshedareas(recruitmentsubsidyandspilloverrespectively);

Conservemarineecosystemsandbiodiversity;Protectattractivehabitatsandspeciesonwhichsustainabletourismcanbebased;

Contributetothescienticknowledgeofmarinespecies,communitiesand ecosystems,byprovidingrelativelyundisturbedsitesforresearch,and ecologicalbenchmarksagainstwhichtomeasurehumanimpacts;

Preservegeneticdiversity;

Protectculturaldiversity(e.g.,sacredplaces,shipwrecksandlighthouses).

Conusion over MPA terminology complicates the dialogue about whether, when, and how thesemanagement tools should be used. Likewise, MPAs having similar names can sometimes dierundamentally in their eectiveness in protecting habitats and resources. For example, there is awidespread misperception that all MPAs are no-take because most are not. Box 4 lists and denesthe most requently used terms that dene the various types o MPA used today. The remainder o thissection ocuses on no-take reserves.

Figure 17. School o yellow goatsh (Mulloidichthys martinicus). NTRsgreatly reduce shing pressure on animals living within their bordersand tend to maintain higher population levels o some species.

Photo: Robert Steneck

Figure 18. The once abundant Elkhorn coral (Acropra palmata) isconsidered one o the most important ree-building corals in theCaribbean and Florida Keys. It is now listed as an endangered species on

the IUCN Red List and in Appendix II o CITES. Since 1980, an estimated90-95% have been lost due to disease, storms and bleaching. WhileNTRs can act to protect and enhance habitat and ecosystem recoverythey cannot solve all management problems or species such as these.Photo: Robert Steneck


27/80

27

Box 4.MPADenitionsA bewildering array o names are used or marine protected areas (MPAs). Below, the most commondenitions are given with specic ocus on no-take reserves and MPA networks.

Marineprotectedarea(MPA)Any area o intertidal or subtidal terrain, together with its overlying water and associated fora, auna,historical and cultural eatures, which has been reserved by law or other eective means to protectpart or all o the enclosed environment (IUCN/WCPA 1994).

A geographical dened area, which is designated or regulated and managed to achieve specicconservation objectives (UN CBD 1992).

According to the IUCN, an MPA also has to ollow the IUCN denition or a Protected Area (PA):A clearly dened geographical space, recognised, dedicated and managed, through legal or othereective means, to achieve the long-term conservation o nature with associated ecosystem servicesand cultural values (IUCN/WCPA 2008).

Marinemanagementarea(MMA)These are usually relatively large, legally delineated locations in the (coastal) ocean that are intendedto be under active management or purposes o conservation or resource management. They arerequently subdivided, or zoned, to provide or dierent patterns o management at dierent locations.Some o their zones are usually no-take shery reserves.

No-takesheryreserve(NTR)No-take shery reserves (NTRs) are also reerred to as marine reserves, no-take areas, or ecologicalreserves. NTRs are a special category o MPA within which extractive shing activities are regulated(usually not permitted). Within some NTRs, all biological resources are protected through prohibitionso shing and removal, disturbance, or harm to any living or non-living marine resource, except whennecessary or monitoring or research (Lubchenco et al. 2003).

Some NTRs restrict access and/or other activities (e.g., pollution, construction, research, boating and

diving) that may adversely impact resources, processes or the ecological and cultural services theyprovide. Others restrict only extractive activities.

Marineandcoastalprotectedarea(MCPA)Any dened area within or adjacent to the marine environment, together with its overlying waters andassociated fora, auna, and historical and cultural eatures, which has been preserved by legislationor other eective means, including custom, with the eect that its marine and/or coastal biodiversityenjoys a higher level o protection than its surroundings (UNEP-WCMC 2008). An MMA adjacent to acoast, and associated terrestrial protected areas, would comprise an MCPA.

MPAnetworkorsystemThe use o network and system can be conusing as neither term has a globally accepted denition,and because they are oten used interchangeably with the same meaning in the same document.

The commonly used denition or an MPAnetwork is: a collection o individual MPAs or reservesoperating cooperatively and synergistically, at various spatial scales and with a range o protectionlevels that are designed to meet objectives that a single reserve cannot achieve alone (IUCN/WCPA2008). Although there are exceptions, the word system tends to be used most requently orterrestrial protected areas, whereas the term network is more prevalent when discussing MPAs(UNEP-WCMC 2008).


28/80


28

3.1.1Theno-takesheryreserveClosing parts o the ocean to shing, so that sh stocksare preserved, holds great intuitive appeal. In manyregions o the world, one o the main arguments used to justiy new MPAs is the claim that they help maintain or

replenish depleted sheries stocks in surrounding waters.By prohibiting (or sometimes severely restricting) shingactivities, no-take shery reserves (NTRs) serve as theonly type o MPA which can assist sheries management,whether by operating singly or as several reserves in anMPA network. Some NTRs may also restrict access and/or other activities, such as development, construction,research, boating and diving.

Since shing pressure on animals living within NTR bordersis greatly reduced, these areas help promote sh survival andreproduction even i the surrounding area is severely over-shed. They also tend to maintain higher population levels

o site-attached species and help protect site-attachedecological unctions such as spawning aggregations.Furthermore, by serving as reuges or heavily shedspecies, NTRs can protect overshed species rom localextinction. However, none o these eects directly impactsthe populations o the shed species in the surroundingarea.

Demographic connectivity in marine populations is keyto the sheries-management role o no-take reservesbecause it provides a mechanism or reserves to enhancesh production outside borders (Kritzer and Sale 2004).Because o connectivity, reserves may supplement a

shery population in the surrounding shed area isome o the production within is exported as spillover or recruitment subsidy. This argument isoten used to convince shing communities to support the introduction o NTRs, yet supportingevidence remains limited. Unortunately, it is technically very challenging to demonstraterecruitment subsidy, and or slow-growing late-maturing shes and invertebrates, any positiveeects o NTRs may not be evident until many years ater establishment. Hence there is a needor long term protection and monitoring, coupled with well-designed experiments, to quantiyspillover and recruitment subsidy i the ull benets o a reserve are to be revealed.

Empirical studies have shown, to varying degrees, our changes inside (Mumby et al. 2006) and/or outside (Roberts et al. 2001, Russ et al. 2003) NTRs that may benet shed populations outsidereserves. These changes include:

1) Increased reproductive output within the NTR because o increases in sh abundance, spawning

biomass, mean age, and body size;2) Higher net export o juveniles and adults to surrounding shed areas (spillover);

3) Higher net export o eggs and larvae to surrounding shed areas (recruitment subsidy); and

4) Protection and recovery within the reserve o both the habitat and entire ecosystems on whichshed species depend.

Figure 19. The underside o a emale Caribbean spiny lobster(Panulirus argus) showing the black, tar-like spermatophoredeposited by a male which will ertilize the bright orangeeggs attached to her abdomen. In shed areas, small spinylobsters may produce a ew hundred thousand eggs, whilelarge emales protected in MPAs can produce millions o eggsin each o several clutches. Photo: Mark Butler


29/80

29

3.1.2 Protection o a portion o the populationThere is some compelling evidence that NTRs help protect animals within borders rom the eects oshing. Any well-managed NTR that is large enough to cover the majority o an individual organismsmovements will come to hold denser populations o older and larger individuals. This can beattributed to increased survivorship resulting rom reduced shing impacts. In planning such reserves,

consideration should be given to the specic habitat requirements o individuals at dierent lie stages,and the extent o daily or seasonal movements. Ideally, an NTR should also be large enough that areasonable proportion o target species larvae will complete pelagic lie stages and settle within theNTR borders. Even i the NTR coners a good level o protection or individuals inside, i a sucientarea is not covered, the target population inside the NTR may end up depending on reproductionoutside NTR borders or replenishment. Overshing would then lead to declines in abundance bothwithin and outside the reserve.

Our relative lack o scientifc inormation on matters such asthe correct size, spacing or placement o no-take reserveslimits our ability to predict the eects that a proposed

no-take reserve will have on surrounding fsheries orbiodiversity conservation.

For this reason, the appropriate size (and sometimes shape) o an NTR should depend upon thegeography o a region (presence o required habitat), the hydrodynamics, and the habits o thetarget species. It ollows that an NTR cannot simultaneously be optimal in size and placement or abroad suite o species, unless habits and habitat requirements are similar. However, there is currentlyinsucient inormation about this or most ree species so it is not yet possible to dictate minimumsize requirements by species. Nor do we yet have enough inormation on the precise benets ocreating networks o neighboring NTRs, rather than stand-alone reserves. Still, available evidenceshows that organisms within NTRs attain greater longevity and larger sizes, which indicates that evensmall reserves a ew hectares in area provide protection or many site-attached ree species. Whatremains to be discovered is whether these small reserves can continue to sustain viable populationsin the ace o continued over-shing beyond borders.

3.1.3 Spillover and recruitment subsidyProtection o a portion o the shery population, as insurance against shery collapse or speciesextinction, is one benet o no-take reserves. More important, is i the protected population is ableto signicantly enhance productivity o the shed populations beyond reserve borders. Connectivityshould lead to this enhancement through sustained net export o target species biomass rom thereserve to surrounding areas. The act that this net export should be sucient to also compensate orthe loss o shing area is also important, and requently orgotten. However, precise assessment osuch export unctions is technically and logistically dicult, and recruitment subsidy has rarely beendemonstrated (Russ 2002, Sale et al. 2005).

Although evidence or spillover is increasing, the mechanisms that encourage adult sh movementrom reserves to shed areas remains poorly understood (Abesamis and Russ 2005). Spillover is otenassumed to be driven by density-dependent processes. Density-dependent movement occurs whenthe rate and directionality o individual movement changes with population density (Sutherland etal. 2002). This is oten thought to be driven by high rates o aggressive interactions within denserpopulations. To cause spillover by this mechanism, measurable density dierences between thereserve and surrounding area are necessary, and it is also likely that the density within the reserveneeds to approach the carrying capacity o the local environment beore spillover can occur. Spilloveris generally assumed to be a very local process that can enhance shing success close to no-takereserve boundaries, but not ar rom them.

A special situation somewhat analogous to spillover occurs i the reserve is established over thenursery habitat o a species. In such a case, enhanced survival o organisms would be expected as


30/80


30

a result o habitat protection (e.g., rom destructive shing gears, such as traps or trawls) than romlack o shing pressure. Only in instances where a shery targets juveniles, or takes these as by-catch,would direct protection rom shing lead to enhanced survival. Still, so long as enhanced survivalresults, a protected nursery habitat would likely yield a greater number o sh that reach an age wherethey are able to leave nursery grounds or adult habitat. This would lead to enhanced production

out o the reserve to support a shery that targets mature sh, thereby having a more widespreadpositive impact on shing success when compared to the eect o spillover between protected andunprotected patches o the same (adult) habitat.

Recruitment subsidy should aect shing yield at urther distances rom reserve boundaries. Denserpopulations o larger (and thereore more ecund) individuals inside a reserve can be expected toproduce larger numbers o larvae, many o which will disperse beyond the boundaries o all but thelargest reserves. There have been ew experimental demonstrations o dispersal kernel shapes to date,but in theory, it should be possible to use such data on a target species, together with data on localgeography and oceanography, to calculate the optimal size and spacing o individual and networkso no-take reserves. Until the science progresses to this point, we are limited to making estimates,such as that or typical ree sh species where demographically important recruitment subsidy mightextend rom 10-30 km beyond the borders o a no-take reserve.

The spatial scale o connectivity and its resolution is a critically important issue or management o reesheries using NTRs or NTR networks. The resolution o connectivity also has important implicationswhen trying to gain a undamental understanding o the structure and dynamics o these communities,and o the appropriate scales at which to mount management interventions. For example, i larvae arepredominantly retained at local (kilometer) spatial scales, local scale management may be eective, buti larvae disperse urther, management will need to be similarly scaled up i it is to be eective. The tasko dening dispersal patterns or important shery species will demand careully designed experimentsthat include spatially large-scale sampling o organisms. Such experiments will be most easible i theyare implemented jointly by managers and scientists in the context o adaptive management.

To conserve biodiversity, regardless o the particular historyo establishment, an eective management program

should be put in place across the ull network, includingthe space between NTRs, and should encompass thecoastal marine ecosystem and land areas that aect it.

3.2MPAnetworksBecause o connectivity, a set o MPAs (usually NTRs) in a given region may operate ecologically asa network with individual organisms dispersing rom one NTR to the other as well as rom one NTRto the surrounding area. Critical or success, the details o dispersal patterns will help determinethe appropriate scale o an NTR network. That is, NTRs within a network must be close enough sothat there is some exchange o individuals. Moreover, it can be expected that an NTR network willbe unctionally more eective than an equivalent number (and area) o NTRs operating in isolation.This expectation exists because demographic connectivity among NTRs within a network presumablyconers resilience to individual populations, in the same way that dispersal processes within ametapopulation coner greater resilience to local subpopulations.

MPA networks have been widely advocated or the conservation o marine biodiversity, protectionagainst natural and human disturbances including overshing, and as a tool to increase resilience ocoastal ecosystems and their ability to adapt to climate change. However, the theory supporting theirbenets is still incomplete, and in any event, the lack o comprehensive data on the connectivity otarget species would preclude ormal application o theory into network design. At present, it can onlybe anticipated that benets occur, and that networks should be planned so that neighboring NTRs areseparated rom 10-30 km apart; an ideal scale or most target ree species (this is essentially the sameapproach to be taken when establishing single NTRs or sheries management).


31/80

31

In practice, MPA networks develop in one o two ways: as separately established and managed NTRsaugmented with additional NTRs interspersed as needed, or as regional-scale marine managementareas zoned to include an appropriate number and spacing o NTRs. To conserve biodiversity,regardless o the particular history o establishment, an eective management program should beput in place across the ull network, including the space between NTRs, and should encompass the

coastal marine ecosystem and land areas that aect it.

3.2.1Metapopulationdynamics,sourcesandsinksCoral rees are inherently patchy and ragmented habitats, and many ree organisms exist as spatiallydistinct local populations connected by an unknown degree and distance (Kritzer and Sale 2004).The level o connectivity among local ree populations will essentially determine whether theyunction as isolated almost closed populations, as metapopulations where the dynamics oseparate populations are buered by recruitment subsidy rom nearby populations, or as spatiallydiscontinuous but otherwise unitary populations with no particularly interesting demographicsubstructure. O the three, metapopulations are the least understood, but in theory possess increasedresilience at the single local population level because o the exchange o individuals. Much o thedeveloping theory o MPA networks hinges on the expectation that connectivity among MPAs withina network coners resilience comparable to that which exists within a metapopulation. This resilienceshould make the MPAs within such networks less susceptible to decline i overshing, or other actors,impact populations outside MPA borders. In general, connectivity between subpopulations shouldincrease in a species-specic way as distance decreases. Knowing the level o connectivity among a

Figure 20. The size and spacing o no-take reserves with respect to dispersal distance o the species o interest.

The white circles represent reserve boundaries while the dome shape represents the pattern o larval dispersal (highernumbers o larvae occur at the birthplace, i.e., within the reserve, and gradually decrease in number with distance).Reserves intended or:

1) Conservation: should be large enough to retain a substantial portion o larval dispersal to ensure adequatesel-recruitment;

2) Fisheries enhancement: should be sized and spaced so that a signicant proportion o larvae can disperse tosurrounding shed areas. I reserves are to unction as a network they must be spaced close enough to ensureconnectivity via larval dispersal.

Credits: Photo, Commonwealth o Australia (GBRMPA); Graphics, Zeke Pesut


32/80


32

set o nearby populations is important or understanding the demographics o each population. Fordispersing organisms, the spatial arrangement o populations and/or prevailing patterns o watermovement may make certain populations consistent sources or sinks. Sink populations are thosethat ail to replenish themselves and are only saved rom extinction by the dispersing surplus o otherpopulations (sources). It is not uncommon in discussions o metapopulation theory to assume the

presence o both population types. It is also likely, however, especially when hydrodynamic patternsare variable, that ew i any populations can be permanently labelled as sources or sinks. Still, sourcepopulations should be considered intrinsically more important to the unctioning o a metapopulationbecause they are sel-sustaining and are best able to subsidize recruitment to other populations.

Determining the actors that determine whether an NTR unctions as a source or sink population or aparticular species is directly relevant to the science and design o marine reserve networks. Currently,these actors have not yet been clearly identied, and veriying whether an NTR unctions as a sourceor sink will require sampling o species production and dispersal over several years. At this stage it canonly be stated that certain preconditions may avor source or sink status. For example, a consistentphysical oceanography, which might change seasonally but consistently through time, is essentialor permanent source or sink status, i.e., uniorm oceanography leads to consistent patterns o larvaldispersal. In a variable oceanographic setting, most populations likely spend some time as sources.Also, or most habitats, an up-stream location with a consistent oceanography should ensure sourcepopulation status. However, this will not guarantee the strength and viability o a source. On the otherhand, although not always the case, a down-stream location permits the existence o sink populations.Also, a population occupying marginal habitat, or consistently experiencing higher than usual shery-independent mortality, is more likely to be a sink.

Despite our current inability to speciy status, there is a general consensus on how best to maximizethe eectiveness o an MPA network in a region where source and sink populations are present. Thisconsensus has resulted in a set o principles that are reasonable, but have not been validated, andare virtually impossible to apply:

1) A network in which reserves are placed in source habitats will be superior to one that placesreserves at random locations or in sink habitats;

2) The importance o source-sink population structure is increased i the MPA network displacesrather than reduces shing eort;

3) Appropriate siting o MPAs becomes increasingly important as the proportion o the environmentconsisting o poor-quality (sink) habitat increases;

4) I the environment contains directional currents, the spatial location o reserves will be critical topopulation enhancement.

I ree species are distributed as consistent source and sink populations, this arrangement shouldbe recognized when setting up an MPA network. Just as in instances where lack o knowledge odispersal patterns or target species precludes a precise and objective set o decisions on NTR size, sotoo the lack o inormation on source-sink dynamics, or even whether consistent source-sink dynamics

Figure 21. Buoys can be used to delineate the borders and zones o a marine reserve. Photo: Miguel Angel Maldonado,Centro Ecolgico Akumal


33/80

33

exist, precludes objective decisions on NTR placement in a network. Nevertheless, even thoughthese science gaps need to be lled, an MPA network should not be designed without reerence tothese important demographic issues. Overall, eorts to advance science in the context o designing

and implementing new networks should be encouraged. Again, this will require that scientists andmanagers work together in a long-term adaptive management process in which setting up a networkbecomes a way o testing ideas about the eectiveness o the choices made. It is regrettable that the

use o MPAs as shery management tools has proceeded as ar as it has with so little concern aboutthe lack o sound demographic science to underpin it, but there are excellent opportunities to worktowards redressing this problem in the course o improving ree shery management.

It is regrettable that the use o MPAs as fshery managementtools has proceeded as ar as it has with so little concernabout the lack o sound demographic science to underpinit, but there are excellent opportunities to work towardsredressing this problem in the course o improving reefshery management.

Beyond metapopulations, coral ree communities are now increasingly being viewed asmetassemblages (or metacommunities) in which each species exists in its own metapopulation. Thus,a single location represents one node within each metapopulation, and the metassemblage comprisesa spatial overlay o individual metapopulations. In a metassemblage, metapopulations are organizedon dierent spatial scales depending on the dispersal properties o a species. Here, species interactionscan be particularly rich. The theory o metassemblages is not well developed, and any attempt toapply it to MPA network design, or management o ree communities, relies on a scanty and eebleset o rules o thumb. While managers operate in this theory-ree state, there is considerable scope orpopulation ecologists to build the needed theory, and introduce it into management regimes.

Figure 22. Environmental education and awareness programmes aimed at stakeholders and users o marine areas are animportant part o coastal management. Photo: Miguel Angel Maldonado, Centro Ecolgico Akumal


34/80


34

3.3WhatMPAnetworkscannotdoMarine protected areas, and particularly NTRs and NTR networks, are valuable tools or reemanagers. NTRs protect target species rom shing mortality, and also protect habitat romshery-related degradation. Other types o MPAs provide limited to solid protection against

other orms o site-specic human impact. However, the strong advocacy or the use o MPAshas obscured the act that they are not the only tool that ree managers need, nor even the besttool or every task. Fisheries management in particular, is a complex task that cannot be solvedby the application o a single tool.

Furthermore, the establishment o MPAs is rarely ollowed by good management andenorcement. Many MPAs exist on maps and in legislation but oer little real protection. Otenreerred to as paper parks, these sites represent a ailure to protect resources and ecosystems.Adding more o these paper parks does nothing or conservation or sheries management. Inact, the ineective deployment o minimal unds and ew personnel to provide the semblanceo management, represents a depletion o resources that could have been used to properlymanage MPAs in other areas.

Box 5. No-takereservescanbeineffectiveSome circumstances where NTRs will be ineective:1) Highly mobile species will be poorly served by any orm o MPA unless a substantial portion o their

habitat lies within MPA boundaries. Thus, sheries or ree-associated pelagics, and more widely-ranging demersal species, are unlikely to be sustained by the creation o an NTR network unlessthe NTRs are exceptionally large or numerous. Otherwise, individuals will spend most o their timeoutside reserve borders and be subject to shing.

2) Fishery species which experience a critical lie stage within nursery habitats degraded by pollutionor coastal development will not be sustained by an NTR that only protects adult habitat. The lacko suitable nursery habitat becomes a bottleneck that restricts production and limits replenishment

to the adult sh population.

3) NTRs are also an ineective shery management tool in regions where pollution or other generalactivities are degrading habitat and reducing the NTRs capacity to support the target species.Habitat quality will continue to degrade inside as well as outside the NTR and the shery willlikely decline.

4) To be optimally eective in sustaining a particular shery, the spacing and sizing o NTRs in anetwork must refect the dispersal characteristics o the target species. It also ollows that an NTRnetwork cannot be simultaneously optimal or several target species, especially i they each havevery dierent patterns o dispersal as larvae. That is, one size does not t all.

5) In most cases, the introduction o NTRs results in redistribution rather than reduction o shing eortrom the now protected to remaining unprotected locations. In circumstances where shing eort

is greater than that which is sustainable and overshing continues unchecked, the target specieswill likely continue to decline in size, age and abundance. While NTRs provide some insuranceand mitigation against overshing, poor shing practices will eventually lead to shery collapses. Isheries are to be sustainable, imposition o other controls on shing eort and catch must occurin combination with NTRs.

Finally, and perhaps most importantly,

6) NTRs which are not managed to ensure that compliance with regulations is enorced, will not ullltheir intended role. Under these circumstances, NTRs will not be ree rom shing mortality, targetspecies will not survive longer within borders, increase in production o ospring will not occur, andthere will be no net spillover or recruitment subsidy to surrounding areas.


35/80


36/80


36

3.4 The value o coastal marine ecosystemsHealthy marine resources require healthy, intact ecosystems. Marine and coastal ecosystems arehighly productive and support communities and economies by delivering various goods and services(e.g., ood security, clean water, recreational opportunities and other benets).

It is estimated that by 2050, 91% o the worlds coastlines will be aected by development. Manycoastal areas o developing countries are dominated by sun and beach tourism with a ocus onrapid coastal growth. Development oten proceeds because it seemingly brings jobs and revenue inthe short-term. But the long-term costs o inappropriate development in lost ecosystem goods andservices, degraded local cultures, and other neglected impacts are estimated to be ar greater.

A number o major economic activities are by denition coastal:

Recreationalandcommercialsheries;

Portsandshipping;

Sunandbeachtourism;

Communityrecreationalservices;

Natureandadventuretourism;Onshoreconstruction,includingseawalls,groins,andotherstructurestoprotectshores.

Many coastal businesses and recreational activities rely heavily on the natural and non-market servicesthat healthy coastal habitats provide. These services include shoreline protection, sh nursery grounds,and destinations or valuable tourist industries.

Figure 23. Ocean habitat types are connected by the movements ojuvenile and adult organisms and through the transer o materials andnutrients. These connections should be considered in the design oMPAs and MPA networks. Photo: Stillpictures

Figure 24. Many coastal businesses and recreational activities relyheavily on the natural, non-market services that healthy coastalhabitats provide. Sustainably managed coastal eco-systems servemany recreational, amily and cultural purposes. Photo: Stillpictures


37/80

37

It is necessary or all coastal communities, local businesses, coastal managers and governments torecognize that it is business-smart to conserve coastal habitats and the wide network o industryand ecosystems they support, because this will ensure compounded returns on investmentsthrough time.

3.4.1MaintaininghabitatcorridorsandeconomicservicesThe habitats which link the various lie stages o species across continental shelves are vital orhealthy shery and ecosystem unctioning. Unortunately, although maintaining ecosystem unctionis important or economic production in coastal areas, it is poorly understood. A mangrove, seagrassor ree habitat can be seriously damaged due to coastal developments that block, divert, slow, orenhance water fow (and transer o substances) rom one habitat to another, even i constructionoccurs some distance away.

The last century has witnessed extensive modication o our coastal ecosystems. Individuals,communities, business entities, environmental scientists, management and regulatory agencies,and governments need to work together so that these impacts can be successully managed.We need to apply the best science based on the best inormation available to ensure thateective policy decisions are made, and that all groups accept resulting management decisions.This requires thinking on time-scales which last longer than an election cycle.

Box 6.MajorcoastaleconomicactivitiesTo achieve more sustainable management o coastlines, communities, governments and managersshould insist on taking the ollowing actions:

Anticipateandplanforchangesincoastalhabitatson5to20yeartimescales;

Anticipate cumulative impacts, i.e., coastal development is a continuous process andnegativeimpactscanbuildupovertime;

Provideincentivessothatcoastalenterprisesadoptsustainablebusinesspractices;Ensurethatallcoastalstakeholdersarepubliclyinvolvedindecisionmaking;

Avoidurbansprawlbyapplyingstrictzoningrulestolanduseplans;

Adoptbestpracticesinwastemanagementtoreducecoastalpollution;

Acquireobjectiveandcomprehensiveenvironmentalassessmentsforcoastaldevelopmentproposals;

Useindependentenvironmentalexpertstoevaluateproposalsforcoastaldevelopment.

Sustainably managed coastal communities serve many recreational, amily and cultural purposes andare wise investments or uture generations.


38/80


38

3.4.2 Alleviating poor connectivityThere are many reasons one population o a ree species may not be conn

CRTR_ConnectivityHandbook

Documents