Marine conservation planning in practice: lessons learned ...€¦ · Marine conservation planning...

Marine conservation planning in practice: lessons learned from theGulf of California

JORGE G. ÁLVAREZ-ROMEROa,*, ROBERT L. PRESSEYa, NATALIE C. BANa, JORGE TORRE-COSÍOb andOCTAVIO ABURTO-OROPEZAc

aAustralian Research Council Centre of Excellence for Coral Reef Studies, James Cook University, Townsville, AustraliabComunidad y Biodiversidad, A.C., Colonia Delicias, Guaymas, México

cCenter for Marine Biodiversity and Conservation, Scripps Institution of Oceanography, University of California, San Diego, USA

ABSTRACT

1. Overfishing, pollution, coastal development and climate change threaten marine biodiversity globally andcompromise the services that marine ecosystems provide. Systematic conservation planning (SCP) provides aframework to identify areas where actions can be effective in addressing these threats, while minimizing the costsof interventions. This study investigated the application of SCP in the Gulf of California, a marine hotspot whereseven prioritization exercises have been undertaken.

2. The review of planning exercises showed that the use of SCP methods has progressed slowly (gaps includeplanning for land–sea connections and ecosystem services) and highlighted benefits and difficulties of applyingSCP principles and tools.

3. Despite some convergence, important spatial differences were found in priorities between plans. Convergence wasevident in well-studied shallow and benthic marine ecosystems. There were also important differences related to theplanning approach, methods and extent. Divergence between methodological and spatial similarities between planssuggests that additional factors (e.g. manually delineating priority areas, incorporating updated datasets, randomerror), in addition to data and objectives, play an important role in defining the distribution of conservation priorities.

4. According to expert opinion, the implementation of new marine protected areas (MPAs) in the region hasbeen influenced by some of the planning exercises. However, uptake of planning outputs has progressed slowlyfor many reasons (e.g. conflicting mandates and interests between organizations, limited technical capacities andresources, insufficient political commitment). Other benefits of planning included: developing institutional skillsand knowledge; improving collaboration and coordination between organizations (including agencies, and local,regional and national NGOs); converging on the need to assess priorities for marine conservation in regionalcontext; and building trust among organizations.

5. The existence of multiple marine conservation plans in the Gulf of California also highlighted some of thecomplexities and benefits of having multiple sets of priorities.Copyright # 2013 John Wiley & Sons, Ltd.

Received 12 April 2012; Revised 15 November 2012; Accepted 31 December 2012

KEY WORDS: systematic conservation planning; conservation priority; marine spatial planning; marine protected area network;spatial similarity; gap analysis

INTRODUCTIONBiodiversity conservation of marine ecosystemsremains a global priority (Spalding et al., 2010)

because fish stocks continue to decline (Worm et al.,2009), eutrophication of coastal areas is ongoing(Howarth, 2008), and extinctions of marine species

*Correspondence to: Jorge G. Álvarez-Romero, Australian Research Council Centre of Excellence for Coral Reef Studies, James Cook University,Townsville, Australia. E-mail: [email protected]

Copyright # 2013 John Wiley & Sons, Ltd.

AQUATIC CONSERVATION: MARINE AND FRESHWATER ECOSYSTEMS

Aquatic Conserv: Mar. Freshw. Ecosyst. (2013)

Published online in Wiley Online Library(wileyonlinelibrary.com). DOI: 10.1002/aqc.2334

continue (Sala and Knowlton, 2006). Theservices provided by marine ecosystems (e.g. foodprovisioning, water quality, coastal protection) arealso being compromised (Worm et al., 2006;Aburto-Oropeza et al., 2008). Ongoing conservationefforts, including marine protected areas (MPAs)(Lester et al., 2009; McCook et al., 2010; Spaldinget al., 2010), fisheries management (Mora et al.,2009; Worm et al., 2009), and ecosystem-basedspatial planning and management (Leslie andMcLeod, 2007; Foley et al., 2010), are providingsome benefits to protect marine biodiversity, butare not yet sufficient to halt or reverse the declines.The expansion of threats to the oceans and thelimited resources available for conservation thuscall for effective, transparent, and cost-effectivestrategies to protect biodiversity and maximizebenefits for human livelihoods.

Systematic conservation planning (SCP) providesa framework to identify areas where conservationactions can be effective at achieving conservationobjectives, while minimizing the costs ofinterventions (Pressey and Bottrill, 2009). Benefitsbeyond guiding effective investment in conservationareas include increasing socio-ecological knowledgerelevant to biodiversity conservation, buildinginstitutional and individual capacity, facilitatingcollaboration among stakeholders, and increasingawareness of the need for urgent action to preventbiodiversity loss (Pressey and Bottrill, 2009;Bottrill et al., 2012; Bottrill and Pressey, 2012).Conditions for successful implementation ofactions in conservation areas – as identified throughSCP exercises – have been shown to includeintegration of ecological and socioeconomic criteriainto designs of conservation areas (Fernandes et al.,2005; Green et al., 2009), stakeholder ownershipand support (Gleason et al., 2010), input frommanagers and experts (Fernandes et al., 2009),effective participation and engagement with localcommunities (Game et al., 2011; Baker et al., 2011),consideration of practical aspects of implementation(e.g. simple boundaries of conservation areas)(Fernandes et al., 2005; Lombard et al., 2007),adaptive management (McCook et al., 2010),transparency during planning, and effectivecommunication strategies (Day, 2002).

Given the increasing recognition of the valueof systematic planning for marine conservation,research in the field is advancing rapidly (Leslie,2005). The initial limited focus on representationof biodiversity patterns is expanding toinclude planning for ecological processes and

socioeconomic considerations. Major recentadvances include planning for several kinds ofdynamics and interactions: the persistence of dynamicoceanographic processes and pelagic biodiversity(Lombard et al., 2007; Game et al., 2009; Granthamet al., 2011), connectivity between different habitats(Beger et al., 2010; Edwards et al., 2010), planning forecosystem services (Granek et al., 2010), adaptationto climate change (Green et al., 2009; McLeod et al.,2009), land–sea integration (Tallis et al., 2008;Klein et al., 2010), and socioeconomic considerations(Ban and Klein, 2009; Baker et al., 2011). Theseadvances have guided some recent conservationstrategies (Green et al., 2009; Baker et al., 2011).

While theory in conservation planning isdeveloping quickly, there has been no assessment ofthe influence of new ideas on applications of marineconservation. Previous studies have reviewed aspectsof MPA design that led to successful conservationactions (Lundquist and Granek, 2005; Osmondet al., 2010) and there are retrospectives onparticular planning exercises, e.g. the Great BarrierReef Marine Park rezoning (Fernandes et al., 2005)and the Northern Central California MPA networkdesign (Gleason et al., 2010). However, we areaware of no previous studies that assessed multiplemarine planning initiatives within a single region,using a comprehensive SCP framework for reviewand comparison.

The Gulf of California is ideal for comparativestudies of marine conservation planning, with sevenmarine planning exercises undertaken in the past15 years. This study investigated how SCP principleshave guided marine prioritization in the region,identified missing components of the planningprocess, and discussed the benefits and difficulties ofapplying SCP principles and tools. The similaritiesand differences between planning exercises wereexamined in terms of data, methods and outputs, howidentified priorities match the existing MPA system,and whether plans have guided conservation andmanagement actions. Difficulties and opportunitiesfor conservation planners and practitioners arediscussed when multiple planning initiatives coexist.

METHODS

Study region

The Gulf of California is a marine conservationhotspot (Olson and Dinerstein, 2002; Roberts et al.,2002), globally recognized for its outstanding

J. G. ÁLVAREZ-ROMERO ET AL.

Copyright # 2013 John Wiley & Sons, Ltd. Aquatic Conserv: Mar. Freshw. Ecosyst. (2013)

marine biodiversity (Brusca et al., 2005; Lluch-Cotaet al., 2007). Oceanographic processes, includinggyres, fronts, and upwellings (Lavín and Marinone,2003), contribute to high levels of primaryproduction that sustain rich coastal and marinecommunities and important fisheries (Beman et al.,2005; Lluch-Cota et al., 2007; Aburto-Oropezaet al., 2008). The Gulf is an important feeding,breeding and nursery area for 36 marine mammals(Morgan et al., 2005; Lluch-Cota et al., 2007),including the critically endangered and endemicvaquita, Phocoena sinus (Jaramillo-Legorreta et al.,2007). The Gulf supports large portions of theworld’s populations of several marine bird species(Lluch-Cota et al., 2007) and provides criticalhabitat for migrating birds (Glenn et al., 2006). Theislands of the Gulf of California have ~655 plantspecies, 28 of which can be found nowhere else(Rebman et al., 2002). Comparable levels ofendemicity have been recorded for marine fauna,including 766 invertebrate and 87 fish species(Brusca et al., 2005).

The rapid expansion and intensification of humanactivities in the Gulf of California are threateningits biodiversity and associated ecosystem services.Human population density is relatively low in thewestern coast’s catchments draining into the Gulf ofCalifornia (i.e. the Baja California Peninsula),but rapidly increasing, with associated increasesin threats to the marine environment (Lluch-Cotaet al., 2007). While the western coast of the Gulfremains comparatively undisturbed, many easterncoastal areas and catchments draining into the Gulfare affected by aquaculture, industry, agriculture,and urban development (Beman et al., 2005;Páez-Osuna and Ruiz-Fernández, 2005; Lluch-Cotaet al., 2007). Intensive use of pesticides in agriculturalareas is also a concern (Gutierrez-Galindo et al.,1992; Galindo-Reyes et al., 1999; Niño-Torres et al.,2010). Alteration of freshwater and sediment inputsfrom the Colorado River has modified the habitatfor commercial and endangered species (Carriquiryand Sánchez, 1999; All, 2006; Rowell et al., 2008).Saltgrass marshes and mangrove forests are beingreduced and degraded by coastal development,land-based pollution, and shrimp farms (INE, 2005;Glenn et al., 2006; Aburto-Oropeza et al., 2008).Deposition of atmospheric pollution and pointsources of mercury (e.g. mining, urban and industrialsewage) are also affecting marine food webs(García-Hernández et al., 2007). Finally, thedramatic increase in fishing effort and inadequatemanagement of fisheries have led to substantial

declines of many fish stocks, changes in populationsof targeted species, and composition of fishcommunities (Sala et al., 2004), which in turn haveaffected regional and local economies (Erismanet al., 2010).

Data collection and sources

Spatial marine conservation planning exercises werethe focus of this study, excluding policy, regulation,and education initiatives. Seven planning exerciseswere identified and reviewed that prioritizedcoastal or marine conservation areas in the Gulfof California. The extents of these exercises,developed between 1998 and 2006, ranged fromsub-continental (B2B: Morgan et al., 2005), tonational (RMP: Arriaga-Cabrera et al., 1998; SPM:CONABIO, 2007), regional (i.e. Gulf-wide) (CSGC:Enríquez-Andrade et al., 2005; ERA: Ulloa et al.,2006), and sub-regional initiatives (i.e. coveringportions of the Gulf) (BCP: Enríquez-Andradeand Danemann, 1998; MRN: Sala et al., 2002).Table 1 contains full names for acronyms andbrief descriptions of the exercises. Informationsources included published and unpublished reportsand digital information. A digital map of marinepriority conservation areas (hereafter ‘priority areas’)and details of datasets were not available for oneplan (BCP). This plan included only the BajaPeninsula and a portion of the Gulf of California,and defined priority areas based on modificationsto the National Marine Priority Regions (RMP).For these reasons, spatial analyses were focusedon the other six planning exercises. However, keyelements of this seventh planning process weresummarized and discussed. A comprehensivedescription of the methods and planning stages foreach plan, as well the maps of priority areas, areprovided in Appendices 1 and 2 (Supplementarymaterial).

Review of approaches to marine conservation planning

Key aspects of a systematic conservation planningframework developed by Alvarez-Romero et al.(2011) were used to identify commonalities anddifferences in approaches and methods between theseven marine conservation plans (Table 1), and toidentify missing elements or stages (see Appendix 1,Supplementary material for details). This frameworkincludes key components in planning (e.g.conservation goals and objectives, stakeholderparticipation, threats, and costs) and emergingissues in conservation planning (e.g. planning for

MARINE CONSERVATION PLANNING IN PRACTICE


Tab

le1.

Selected

aspectsof

plan

ning

exercisesthat

have

identified

marinepriority

conservation

areasin

theGulfof

California

Plan

Extent(km

2 )Goa

ls#

Objectives

App

roach/

metho

dsStak

eholderpa

rticipation

Biodiversity:

patterns

andprocesses

Threatsan

dcosts

Bajato

Bering-B2B

(Morgan

etal.,2005)

Sub-continental

(7,165,359)

1,2,

3Qua

litative

Exp

ert-ba

sed,

considering

continentalendemism,

high

beta-diversity,

sign

ificanceto

migratory

species,an

dprod

uctivity.Priority

Con

servationAreas

selected

basedon

sign

ificant

threatsan

dconservation

oppo

rtun

ities.Areas

selected

byworking

grou

ps(benthos,p

elagic

environm

ent,plan

ning

andman

agem

ent)were

overlayedto

identify

spatialov

erlap.

Other

selectioncriteria

includ

edconn

ectivity

andminim

umarea.

Based

onaworksho

pwiththe

participationof

agencies,NGOs,

universities,an

dsomefishermen.

Participa

ntsprov

ided

inpu

ton

:conservation

objectives;da

taavailability;

andselection,

delin

eation

,an

ddo

cumentation

ofpriority

areas.

Patterns:enda

ngered,

protected,

endemic,

migratory,a

ndkeystone

species(birds

and

mam

mals);c

ritical

breeding

andfeeding

habitatformultiple

species;deep

seacorals;

seam

ounts;benthic

complexitymod

el;beta-

diversity.

Processes:

oceanfron

ts,currents,

eddies,p

rimary

prod

uctivity,an

dmigration

corridors.

Exp

ert-ba

sedassessment

ofthreatsaffecting

priority

areas:land

-an

dsea-ba

sedorganican

dinorganicpo

llution

;subsistence,

commercial

andrecreation

alfishing;

aqua

culture;

damming;

tourism;c

oastlin

ealteration

;coastal

developm

entan

durba

nization

.Exp

erts

also

rank

edrelative

intensityof

threats.

Costswereno

tincorporated.

MarinePriority

Regions

-RMP

(Arriaga-C

abrera

etal.,1998)

Nationa

l(3,290,584)

1,2,

3Not

explicitly

stated

Exp

ert-ba

sed

identification

ofpriority

areasdu

ring

two

worksho

pssubd

ivided

byexpertise(e.g.

chorda

tes,fish,benthic

invertebrates,

oceano

grap

hic

processes).S

oftw

areto

facilitatedecision

-mak

ingby

consensus

was

used.A

reas

identified

wereused

toan

alyseusepa

tterns

and

potentialc

onflicts

betw

eenusean

dconservation

ofbiod

iversity;establish

priorities;an

dmak

econservation

recommenda

tion

s.

Participa

ntsin

theworksho

pinclud

edpu

blic

andprivateresearch

institutions,agencies,NGOs,MPA

man

agers,an

dinternationa

lorganization

s.Participa

ntsidentified

anddelin

eatedareaswhere

conservation

features

occur,

identified

potentialtrade-offs,defined

thepriority

status,an

dform

ulated

generalrecommenda

tion

s.

Patterns:macroalgae,

marineinvertebrates,

fish,sea

turtles,marine

mam

mals,birdsan

dcorals;man

grov

es,

coastaldu

nes,algae,

seagrass,an

dwetland

s.Processes:prim

ary

prod

uctivity

areas(i.e.

phyto/

zoop

lank

ton,

fishingareas);ph

ysical,

chem

ical

andgeolog

icoceano

grap

hic

processes.

Threatswereno

texplicitly

andsystem

atically

used

inthepriority

region

sselectionprocess.Major

threatsto

biod

iversity

andmitigation

strategies

wereidentified

anddo

cumentedfor

priority

region

s.Costs

wereno

tincorporated.

MarinePriority

Sites-SP

M(C

ONABIO

,2007)

Nationa

l(3,152,985)

1,2

Qua

litative

Exp

ert-ba

sedselection

basedon

representativeness

and

viab

ility,fisheries

sustaina

bility,

high

diversity,

ecolog

ical

processes,species

habitat,conn

ectivity,

Exp

ert-ba

sedpa

rticipatoryworksho

pinclud

edresearchers,na

tion

alan

dregion

alenvironm

entalNGOsan

dagencies.Major

inpu

tsfrom

participan

tsinclud

edad

dition

alda

ta,selection

ofconservation

objectives,selection

anddelin

eation

ofpriority

areas,

Patterns:priority

areasfor

enda

ngered,endemic,

andprotectedspecies

(vertebrates

and

invertebrates),h

abitats

(coa

sttypes,reefs,

seagrass,kelp

forests),

andspecialelem

ents

Threatswerean

alysed

afterdelin

eating

priority

sites,an

dthereforeno

tused

asacriteria

forthe

selectionof

conservation

areas.

Costswereno

tincorporated. (C

ontinues)



Tab

le1.

(Continued)

Plan

Extent(km

2 )Goa

ls#

Objectives

App

roach/

metho

dsStak

eholderpa

rticipation

Biodiversity:

patterns

andprocesses

Threatsan

dcosts

ecosystem

services

and

cultural

orresearch

value.

Priorityareas

delin

eatedin

working

grou

ps(vertebrates,

bentho

s,plan

kton

and

fisheries,processes,an

dcoastalvegetation

)were

overlapp

ed.

Representationof

features

ofinterest

across

allbioregions

guided

prioritization

.

threat

assessment,an

ddo

cumentation

ofpriority

areasan

disland

s.(aggregation

areas,

refugees,seamou

nts,rift

zones,ba

cteria

mats,

minim

umox

ygen

zones).P

rocesses:

upwellin

gs,fron

ts,

vertical

mixing,

gyres

andeddies,c

urrents,

andriverdischa

rges.

Coa

litionforthe

Sustaina

bility

oftheGulfof

California-CSG

C(Enríquez-And

rade

etal.,2005)

Regiona

l(902,225)

1Qua

litative

Exp

ert-ba

sedselectionof

biolog

ically

impo

rtan

tareas(A

IB)to

sustain

viab

lepo

pulation

san

drepresentselected

grou

ps(lan

dflora,

land

faun

a,marinebiota,

andwetland

s).Priority

areasweredefinedby

anov

erlapof

morethan

twoor

threeAIB

s.Designcriteria

also

includ

edrepresenting

impo

rtan

tecosystems

andha

bitatformultiple

speciesan

dinclusionof

impo

rtan

tph

ysical

and

ecolog

ical

processes.

Highlypa

rticipatoryconsensus-ba

sed

processwithinternationa

l,na

tion

al,

region

alan

dlocalstak

eholders

from

NGOs,acad

emia,M

PA

man

agem

ent

bodies,an

dagencies.Participa

ntsset

goals,selected

conservation

features,

prov

ided

inform

ation,

participated

inthe

analysisan

dpriority

area

portfolio

design

,an

ddevelopedmap

sof

anthropo

genic

threatsto

biod

iversity

andpo

tentialsocial

confl

ict(betweenuses

andconservation

).

Patterns:selected

species

aggregated

and

analysed

bygrou

p(fish,

mam

mals,seaturtles,

invertebrates,

macroalgae,

birds,an

dland

floraan

dfaun

a);

selectioncriteria

includ

edpo

pulation

and

habitatcond

ition,

endemism,distribu

tion

rang

e,threatsan

dtrends.Processes:

oceano

grap

hic

(upw

ellin

gs,oceanic

fron

ts,an

dcurrents)

andecolog

ical

processes

weremap

pedby

experts.Reprodu

ction,

larvae

dispersal,

recruitm

ent,an

dmigration

werealso

considered.

Threatswereno

tused

togu

ideor

mod

ifythe

selectionof

priority

areas,bu

twereon

eof

thecriteria

considered

byexpertsto

delin

eate

viab

leconservation

areas.Current

andnear

future

(5year

scenario)

threatsweremap

ped.

Athreat

indexwas

developedto

identify

thefeasibility

ofim

plem

enting

conservation

prog

rammes

orprotectedareas,bu

twas

notused

todefine

priority

sites.Post-ho

c‘costs’assessmentba

sed

onpo

tentialsocial

confl

ict.

Ecoregion

alAssessm

ent-ERA

(Ullo

aet

al.,2006)

Regiona

l(361,375)

1,2

Qua

ntitative

Prioritizationwas

based

onrepresentation

ofspecies,commun

ities,

andsystem

swithinthe

ecoregion.

The

process

follo

wsafine-filter

and

coarse-filter

approa

ch.

(Groveset

al.,2002)

Dom

ainwas

stratified

insub-ecoregions

basedon

aSS

Tmod

el.A

draft

portfolio

ofconservation

areaswas

The

plan

ning

team

definedthegeneral

goals,ob

jectives,a

ndmetho

dology.

Supp

ortan

dinpu

twas

givenby

experts,man

agersan

dconservation

organization

s.The

major

inpu

tfrom

fishermen

was

inform

ationon

fishing

areasviainterviews.Duringaworksho

ptheplan

ning

processan

dpo

rtfolio

was

presentedforexternal

expertreview

.

Patterns:im

portan

tareas

(interpo

lated

occurrences/plan

ning

unit)an

doccurrences

ofprotected,

threatened,

endemic,m

igratory,and

keystone

species

(Sargassum

,seagrass,

fish,c

orals,seaturtles,

seamam

mals,seab

irds);

man

grov

es;coasttypes;

benthiccomplexity;

reefs;seam

ounts;

A"cost"

layerwas

constructedba

sedon

thesummed

values

ofindividu

althreats

(tou

rism

,po

pulation

margina

lityan

dsize,

aqua

culture,

different

fishingtypes,na

vigation

routes

andpo

rts).

Threatswereratedin

term

sof

theirpo

tential

impa

ctson

biod

iversity

andused

inMarxanto

(Continues)



Tab

le1.

(Continued)

Plan

Extent(km

2 )Goa

ls#

Objectives

App

roach/

metho

dsStak

eholderpa

rticipation

Biodiversity:

patterns

andprocesses

Threatsan

dcosts

constructedwith

Marxan(Balle

tal.,

2009)an

dfurther

refinedba

sedon

expert

opinion.

General

design

criteria

(con

nectivity

andviab

ility

ofconservation

features)

wereused

torefine

portfolio

.

rhod

olithbeds;species

aggregations;sardine

recruitm

entareas;

impo

rtan

tinvertebrate

areas.Processes:

prim

aryprod

uctivity

andup

wellin

gzones.

minim

izethe‘cost’for

each

plan

ning

unit(i.e.

avoidhigh

lythreatened

areasan

dminim

ize

confl

ictwithfisheries

andotheruses).

BajaCalifornia

Peninsula

Priorities-BCP

(Enríquez-And

rade

andDan

eman

n,1998

Sub-region

al(237,653)

1,2,

3,4

Qua

litative

Impo

rtan

tareasfor

coastal-marine

conservation

,as

wella

skeyprob

lems/threats

andaction

s(not

only

spatial)wereidentified

andprioritizedby

expertsba

sedon

qualitativemulti-criteria

analysis.Previou

snatio

nalp

riorities

(RMP)

served

asaguide.

Preferencewas

given

toareaswith

higher

ecological

value,larger

size,stron

gerthreats,

lower

conservatio

nop

portun

ities

(except

preexistentprotectio

n),

higher

econ

omic

impo

rtance

and

historical/culturalv

alues.

Exp

ert-ba

sedpa

rticipatoryworksho

pinclud

edresearchers,indu

stry,

trad

esmen,bu

sinessmen,a

gencies,

MPA

man

agers,an

dconservation

organization

s.Natural

resource

users

(fishermen

andtourism),wereinterviewed

regardingresource

uses.Stak

eholders

also

attend

edworksho

psto

review

conservation

andman

agem

entprob

lems,

andto

prop

osean

didentify

priority

action

sto

addressthem

.A

selected

grou

pof

researchersidentified

and

prioritizedconservation

areas.

Spatialda

taon

biod

iversity

pattersor

processeswereno

tused

toidentify

priority

conservation

areas.

How

ever,biolog

ical

criteria

(ecological

integrity,

ecosystem

diversity,

endemicity,

richness,nu

mberof

enda

ngered

species)

guided

area

prioritization

.Processes

considered

byexpertsin

area

prioritization

includ

edprim

ary

prod

uctivity,

upwellin

gs,migration

san

dbiolog

icalcorridors.

Literature/expert-based

assessmentan

dprioritization

ofprob

lems/threats

relatedto

fisheries,

coastalzone

use,

ecotou

rism

anduseof

island

s,an

dprotected

areas.Identified

prob

lemsinclud

edaccess

toresources

(illegalfi

shing,

overexploitation

,confl

ict);man

agem

ent

(ina

dequ

ateplan

ning

,regu

lation

and

enforcem

ent,

jurisdiction

alfragmentation

/overlap

,po

orda

ta);market

(resou

rces

andservices

valuation,

incentives);

econ

omic

(migration

);fisheriesorganisation

andinfrastructure;an

dresearch-m

anagem

ent

gap.

Costswereno

tincorporated.

MarineReserves

Network-MRN

(Salaet

al.,2002)

Sub-region

al(132,176)

@1,

2Qua

litative,

Qua

ntitativePrioritization

focusedon

reef

fish,b

utinclud

edcoastalv

egetationan

dinvertebrate

biod

iversity.

Marinereservedesign

principles

were:protect

representative

andrare

habitats;include

areas

withhigh

species

richness;m

axim

ize

ecosystem

function

ing

The

plan

ning

exercise

was

anacad

emic

processorganizedby

tworesearch

centresan

don

eNGO.Inpu

tfrom

artisana

lfishermen

was

impo

rtan

tformap

ping

fishingareas.

Patterns:Reeffish

richness

(134

spp.);ha

bitats

(san

dybo

ttom

s,seam

ounts,rhod

olith

beds,S

argassum

beds,

andblackcoralbeds)

used

assurrog

atefor

plan

tan

dinvertebrate

biod

iversity

and

impo

rtan

treef

fish

habitat;coral

A‘cost’layerwas

developedto

beinclud

edin

analyses

usingSITESto

avoid/

minim

izesocial

confl

ict

withartisana

lfisheries.

Artisan

al(small)bo

atfishingpressure

(density

boats/km

2 )was

used

tominim

iseconfl

ict

betw

eenfishermen

and

(Continues)



ecological processes, ecosystem services and land–seainteractions).

Assessment of methodological and spatial similarity

Methodological similarity between planningexercises was calculated based on the Jaccardcoefficient (which adds up the total number ofitems that are different, excluding instances ofzero ties), focusing on conservation objectives anddata because these were assumed to be mostinfluential in the selection of priority areas. First,conservation objectives and data for features ofinterest were coded as binary attributes (1 = used;0= not used). When planning exercises useddatasets differently (e.g. extent, resolution, date)or when data were grouped or disaggregateddifferently, data types were matched with the mostequivalent elements in other exercises or groupedin the same way as in other exercises. Based onan overlay of the maps (a.k.a. portfolios) ofpriority areas (hereafter ‘priority maps’), areas ofcoincidence were identified and spatial similaritybetween pairs of priority maps were measured withthe Kappa statistic using the Map ComparisonKit (Visser and de Nijs, 2006). Kappa similarityis adequate and commonly used for assessingagreement between categorical maps (Wilson et al.,2005; Edwards et al., 2010). The Jaccard coefficientand Kappa statistic were displayed with similarityphenograms based on hierarchical clustering,also applied previously (Ban, 2009; Weeks et al.,2010b). In addition, the spatial configuration ofpriority maps (i.e. number, size, perimeter, andspacing) and the sizes of the planning domains werealso compared. A second spatial analysis focusedon how areas of convergence between the sixplanning exercises were related spatially to existingMPAs (CONANP, 2010), marine bioregions(CEC, 2006), and selected ecosystems and seascapefeatures (Lugo-Hubp and Fernandez Córdova,1990; Ulloa et al., 2006; CONABIO, 2008;UNEP-WCMC, 2010).

Evaluation of achievement of goals and benefits ofplanning

Interviews were carried out with 13 experts, selectedbased on their expertise in marine conservationin the region and their involvement in one or moreof the planning processes discussed here. Affiliationof interviewees included national agencies, regionaland international NGOs, and research centres.During these interviews, experts identified bothT

able

1.(C

ontinued)

Plan

Extent(km

2 )Goa

ls#

Objectives

App

roach/

metho

dsStak

eholderpa

rticipation

Biodiversity:

patterns

andprocesses

Threatsan

dcosts

(protect

larval

sources

andnu

rseriesfor

targeted

fish;ensure

popu

latio

nconn

ectiv

itythroug

hlarval

dispersal;

minim

umreservesize

toensure

self-recruitm

ent;

andminim

izeconfl

ict

withfisheries).A

decision

-sup

porttool

was

used

(SIT

ES;

Ball

etal.,2009).Dom

ain

was

stratified

inzoog

eograp

hicregion

sfollo

wingfish

abun

dance.

commun

itiesan

dseagrass

beds;fi

shnu

rseries(m

angrov

es);

impo

rtan

tlarval

sources

andspaw

ning

aggregations

ofcommercial

fish.

Processes:Pop

ulation

conn

ectivity

incorporated

thou

ghgeneraldesign

criteria

(max.d

istancebetw

een

priority

sites:100km

)an

dim

plem

entedvia

reserveselectionrule.

potentialmarine

reserves.

#1.

Biodiversityconservation

;2.

Natural

resource

man

agem

ent;3.

Researchpriorities;an

d4.

Livelihoo

ds.

@Primarily

anacad

emic

plan

ning

exercise.



reported and unreported planning goals (from apredefined list, supplemented with their responses;see Appendix 3 Supplementary material) andindicated the extent to which these goals had beenachieved (using a 5-point Likert scale). Further,experts assessed how planning outputs guided orinfluenced spatial implementation of conservationactions by identifying (from a list) which actionshave been implemented (e.g. establishment ofmultiple-use MPAs and no-take areas, fisherieszoning, restoration, surveillance). Experts alsodescribed the main factors perceived to facilitate orconstrain actions (open-ended questions). In addition,other perceived benefits of the planning processes(besides spatial actions) were identified from apreliminary list modified from Bottrill et al. (2012),as well as the existence of any formal assessmentsof outcomes, and intentions for revising plans.Interviews were supported by a written survey(Appendix 3 Supplementary material).

RESULTS

Approaches to marine conservation planning

All seven planning exercises included fundamentalelements of systematic conservation planning(e.g. involving stakeholders, setting conservationobjectives, assessing threats and conservation costs,and delineating new priority areas), but differedwidely in the way they addressed these elements (seeTable 1 and Appendix 1 Supplementary material).Notable differences included the way costs wereincorporated (e.g. least-cost solutions vs. prioritizationof highly threatened areas) and methods for identifyingpriority areas (e.g. expert-based vs. software). In thefollowing sections the most relevant similaritiesand differences are summarized and discussed.

Stakeholders

Stakeholder participation varied considerablybetween planning exercises (Table 1). Participationof researchers was prominent in all cases, and inputfrom agencies, conservation non-governmentalorganizations (NGOs) and MPA managers wassubstantial in most exercises. Two planningprocesses (B2B and SPM) also included somerepresentatives of the fishing and tourism sectors. Acontribution by resource users was to provideinformation on areas of economic importance, suchas fishing grounds (Ulloa et al., 2006). Stakeholderinvolvement in one planning exercise (MRN) was

restricted to academics and conservation NGOs. Inthe four expert-based processes, representativesfrom agencies, academia and NGOs activelyparticipated and contributed to different stages ofthe planning process.

Conservation objectives

Approaches to setting conservation objectiveswere diverse (Table 1). A coarse/fine-filter approachto identifying conservation features (i.e. targetingcommunities/ecosystems and individual species/features independently) (Groves et al., 2002) wastaken in most exercises, but there were also importantdifferences. Quantitative objectives were set foronly two planning exercises (MRN and ERA),both of which also defined separate objectives forconservation features occurring in more than onebioregion. Qualitative objectives were common toall planning exercises. Qualitative objectives includedselecting areas with high species richness andabundance or endemic, endangered or protectedspecies, and ensuring connectivity between areas.

Threats and costs

There were major differences between planningexercises in the treatment of threats and costs. In twoexercises (SPM and CSGC), important threats tocoastal and marine ecosystems were identifiedindependently of the prioritization process andtherefore had no influence on the selection of priorityareas. One exercise (CSGC) mapped current andnear-future (5-year projection) threats to inform thefeasibility of implementing conservation actions, aswell as to identify potential areas of conflict. Threatsand potential mitigation strategies were documentedfor the national marine priority regions (RMP)after these were delineated by experts. For the B2Bexercise, a qualitative ranking of the intensityand trend of threats was developed by experts toidentify areas with both threats and conservationopportunities (e.g. ongoing conservation ormanagement efforts). In contrast, the two planningexercises supported by conservation planningsoftware (MRN and ERA) used data on threats toachieve objectives while minimizing conflicts and/orconservation costs (Ban and Klein, 2009) by placingpriority areas away from highly threatened orheavily used (e.g. fishing) areas, where possible.

Emerging issues

Uptake of recent advances in conservation planning(e.g. planning for ecosystem services, land–sea



interactions, and climate change) was limited, despitetheir relevance to address major environmentalproblems in the Gulf of California (Table 2); anexception was the consideration of ecologicalprocesses (Appendix 1 Supplementary material).Generic design criteria such as minimum area sizeand connectivity guided the selection of priorityareas to address some processes. To maintain larvaldispersal, the MRN exercise used software toensure that distances between priority areas didnot exceed 100 km. For the B2B exercise,

experts applied generic design criteria, supportedby available information on processes such asmigration routes and areas of high productivity, todelineate priority areas. Ecological processes linkingland and sea, such as nutrient flows from rivers,were considered by experts in the national marinegap analysis (SPM), but no spatially-explicit modelswere available. Cross-system threats (e.g. land-based pollution and coastal development) wereconsidered by experts in B2B and, in the ERAexercise, simple spatial models of land-based threats

Table 2. Significance of emerging issues in marine conservation planning, with examples from the Gulf of California

Topic Significance in conservation planning Examples of importance in the region

Ecosystem services Planning for ecosystem services aims to maintain thevarious benefits provided by coastal and marineecosystems (e.g. food provision, pollution buffering,coastal protection, nutrient cycling, recreation) and theeconomic gains associated with these (e.g. fisheries,tourism); the need to consider ecosystem services inmarine planning is well recognized (Granek et al., 2010;Silvestri and Kershaw, 2010).

Mangroves increase fisheries landings by providinghabitat used by many commercial species as nurseryand/or feeding grounds, thus providing importanteconomic benefits to local communities and impactingthe regional economy. However, these ecosystemscontinue to be degraded by coastal development in theGulf of California (INE, 2005; Aburto-Oropeza et al.,2008; Ezcurra et al., 2009).

Land–sea integration Integrated land–sea planning aims to mitigate threatsoriginating beyond the boundaries of marineconservation areas (e.g. land-based threats tocoastal-marine ecosystems) and maintain the ecologicallinkages between land and sea that support species andecological processes occurring across terrestrial,freshwater, and marine realms (Tallis et al., 2008;Alvarez-Romero et al., 2011).

Many eastern coastal areas of the Gulf of California areaffected by land-based threats, including agriculture,urbanization, aquaculture, and damming (Páez-Osunaand Ruiz-Fernández, 2005; Lluch-Cota et al., 2007).Intensive use of pesticides in agricultural areas is aparticular concern (Galindo-Reyes et al., 1999;Niño-Torres et al., 2010). In some cases, the potentialimpacts of these threats can extend hundreds ofkilometers from the coast (Beman et al., 2005).

Ecological processes Incorporating ecological processes is critical to maintainconnectivity (e.g. through recruitment and spillover)between conservation and fishing areas, as well as toidentify persistent high productivity areas (e.g.associated with upwellings or ocean fronts) and tominimize the likelihood of MPAs within a region beingaffected by single disturbance events (Allison et al.,2003; Almany et al., 2009).

Persistent oceanographic processes, including upwellings,contribute to the occurrence of areas of highproductivity in the Gulf of California, where majorcommercial, artisanal, and recreational fisheries occur(Lluch-Cota et al., 2007). Ocean currents contribute tothe ecological connectivity between protected andfished areas (e.g. through larval dispersal), thusemphasizing the significance of connectivity forconservation and fisheries planning in the region(Cudney-Bueno et al., 2009).

Climate change Designing networks of marine protected areas that areresilient to climate change is considered a globalpriority. Strategies to deal with predicted changesinclude: design criteria (e.g. size, spacing, shape);spreading risk through representation and replication;identifying and protecting refugia for species, habitats,and special features (e.g. species aggregations);reducing other impacts (e.g. pollutants runoff) toimprove the resilience of ecosystems; and maintainingecological connectivity through networks of MPAs(Green et al., 2009; McLeod et al., 2009).

The ecological impacts of climate change in the Gulf ofCalifornia are still poorly understood, but changes intemperature and precipitation are expected, along withfluctuations in runoff and fisheries (Lluch-Cota et al.,2007, 2010). Predicted increase in temperature forecastschanges in abundance of species with limited capacityfor acclimation due to thermal stress (Stillman, 2003).On the other hand, ENSO episodes are expected toincrease in frequency and intensity, with potentialimpacts on marine populations, communities, andfisheries of highly commercial species (Velarde et al.,2004; Aburto-Oropeza et al., 2010b).

Scheduling and zoning In addition to identifying priority conservation areas,planners need to decide in which areas to invest firstconsidering limited resources (not all can be protectedimmediately) and pervasive threats (values of marineecosystems continue to degrade due to current andfuture threats). Spatial distribution of biodiversity/fisheries values, threats, costs, investments returns anddata uncertainty play a role in scheduling conservationinterventions (Wilson et al. 2006, 2007).

Limited human and financial resources have delayed theimplementation of conservation actions in the Gulf ofCalifornia. The lack of immediate intervention in somecases could mean the extinction of threatened species,such as the vaquita (Jaramillo-Legorreta et al., 2007)or the loss of critical ecosystems, such as mangroves,that sustain important fisheries in the region(Aburto-Oropeza et al., 2008). Prioritizing areas andactions within proposed portfolios of conservationareas and identifying no-take areas in the alreadyestablished MPAs could help to mitigate these negativeimpacts (Rife et al., 2012).



(i.e. multiple-ring buffers associated with higherconservation costs) were incorporated. Only oneplanning exercise (CSGC) explicitly targetedecological linkages between terrestrial, freshwaterand marine priority areas. In the national marinegap analysis, a post hoc integration with terrestrialand freshwater priorities was recommended forthe future. None of the studies recommended asequential allocation (i.e. scheduling) of investmentsin delineated priority areas, e.g. to act first in areasfacing imminent threats (Wilson et al., 2009).However, one exercise (SPM) assigned prioritylevels, based on expert advice, to priority areasafter they were delineated and another (ERA)recommended scheduling.

Similarity and spatial coincidence between planningexercises

Methodological similarity

Clustering of exercises based on objectives anddatasets (Figure 1(a)) showed that the exerciseemphasizing protection of fish (MRN) was verydifferent to the others, which also consideredother taxonomic groups and oceanographicprocesses. Likewise, similarity values between thesub-continental exercise (B2B) and other planswere relatively low. B2B focused on species ofconservation concern throughout North America,not on those more typical of the Gulf, and alsoconsidered processes operating over largergeographic extents (e.g. migratory corridors) thanthe Gulf. Common to both MRN and B2B wastheir focus on a few species and key habitats. Alsonotable was the similarity between the tworegional-scale exercises (CSGC and ERA) which,despite differences in methodology (expert-basedvs. software-driven), shared many conservationobjectives, and used similar data. Similaritybetween the two national-scale exercises (RMPand SPM) was relatively low. The analysis placedSPM closer to the regional plans andRMP somewhere between the regional and thesub-continental exercises. This is consistent withboth B2B and RMP identifying regions, ratherthan specific areas, for marine conservation andtheir use of data appropriate for that purpose.

Spatial similarity

Overall spatial similarity (Figure 1(b)), accountingfor spatial overlap of priority areas and theirrelative proportions in the region, revealed two

subgroups (CSGC-B2B and ERA-SPM) and twoplanning exercises that were substantially differentfrom the rest (RMP and MRN). Spatial similaritybased only on the locations of priority andnon-priority areas (K-loc), and not their relativeproportions (Figure 1(c)), showed higher valuesbut also a different clustering pattern. Forinstance, MRN – isolated based on overallKappa – formed a new subgroup with CSGC

Figure 1. Similarity phenograms for marine planning exercisesconstructed through agglomerative hierarchical clustering analysis(unweighted pair-group average). B2B: Baja-to-Bering; RMP: MarinePriority Regions; SPM: Marine Priority Sites; CSGC: Coalition’sImportant Conservation Areas; ERA: Eco-regional Assessment;MRN: Marine Reserves Network. (a) Methodological similaritybased on planning objectives and datasets (Jaccard coefficient);(b) overall spatial similarity based on location of priority areas andrelative proportions (Kappa statistic); and (c) location-only similarity(K-loc statistic: Visser, 2004). Spatial similarity varied, but generalstrength of agreement between pairs of priority maps was fair basedon overall similarity (mean Kappa=0.24) and moderate for similarityin location-only (mean K-loc= 0.52). A value of 1 means identicalpriority maps, while 0 indicates no more overlap than expected bychance. A significance analysis based on random constraint matchmodels (Visser and de Nijs, 2006) indicated that Kappa valuesbetween pairs of maps were significant when compared to similarity

with random maps (0.066–0.074), P< 0.001.



based on K-loc. This cluster could result frompriority areas from MRN being almost entirelycontained within CSGC and, in both exercises, thepreferential location of priority areas in coastaland nearshore environments. On the other hand,the subgroup formed by ERA and SPM based onoverall spatial similarity (Figure 1(b)) remained inthe location-only phenogram (Figure 1(c)), butwas stronger for location-only, possibly due to theinclusion (with minor adjustments) of some of theERA priority areas into the SPM priority map.

Configuration of priority areas

There was an asymptotic increase in the meansize and perimeter of priority areas as the extentsof planning domains increased, with significantvariation within each plan (Figure 2(a), (b)). Therewas an inverse relationship between the number ofpriority areas and the extent of the planningdomain, though the strength of this relationshipwas fairly low (Figure 2(c)). Excluding the valuescorresponding to MRN – the exercise thatfocused on fish protection (formed by relativelyfew, small priority areas intended to form anetwork of no-take zones) – appreciably increased

the fit between number of priority areas and extentof domain. The tendency for larger planningdomains to be associated with larger and fewerpriority areas (Figure 2(a)–(c)) was related tocoarser data resolution, larger planning units beingassessed and compared for conservation, andmore extensive targeting of ecological processes.

Figure 2. Comparison of selected features of priority maps in relation to the full planning domain extent (values on Y axes correspond to priority areaswithin the Gulf of California only): (a) mean size and (b) mean perimeter of priority areas; (c) total number of priority areas comprising each prioritymap; and (d) median separation between nearest priority areas (use of median responds to highly skewed distributions). Points corresponding to thenational-scale exercises are marked by blue (SPM) and red/dashed (RMP) circles. Fitted trend lines correspond to power (a, b), exponential (c) andlinear (d) types; light-grey lines show an improved fit when MRN is excluded from the analysis (see text). Error bars represent the 95% C.I. for the

mean (P< 0.0001) estimated by Monte Carlo simulation (a, b) and the first and third quartiles (d).

Figure 3. Frequency of spatial coincidence between the priority maps(number of plans identifying an area as a priority) and two variables:total area covered by each frequency value of spatial coincidence(blue dots and line); and proportion of the total extent of existingMPAs identified as priority areas (green squares and line). Fitted lines

correspond to second-order polynomial regressions.



The extent of the planning domain was positivelycorrelated with the separation between priorityareas and therefore with lower potentialconnectivity (Figure 2(d)). Again, excluding MRNimproved curve-fitting because its proposedmarine reserves were considerably smaller thanall other exercises, and thus were relatively widelyspaced relative to the extent of the planningdomain. Also evident from Figure 2(c), (d) is thatRMP identified fewer and significantly moreseparated priority areas than SPM. In these aspects,RMP was more similar to the sub-continentalexercise (B2B) than to the other national plan,

consistent with the methodological similaritydiscussed above.

Overall spatial coincidence, ecological representativenessand MPA coverage

The combined (summed) priority maps covered~100 000 km2, but relatively little of the summedarea was consistently identified as having priorityfor marine conservation. About 31% of thesummed area was identified by at least four plansand only 3% was selected by all plans. This wasreflected in the negative relationship between the

Figure 4. Spatial coincidence of priority areas between planning exercises in the Gulf of California. For overall frequencies, darker areas were identifiedas priorities in more plans. Pie charts represent the percentages of each marine bioregion in three classes of frequency of spatial coincidence across

plans, from none (dark blue) to most plans (red).



frequency of spatial coincidence and the totalpriority area in each frequency (Figure 3).Conversely, the percentage of existing MPAscovered by priority areas increased with thefrequency of spatial coincidence. This means thata large proportion of the areas consistentlyidentified as priorities have also been granted sometype of legal protection (Figures 3 and 4),although, in some areas protection precededprioritization exercises. Spatial coincidence wasgenerally higher in the northern Gulf and westerncoast (Figure 4).

Representation of marine bioregions variedmarkedly among plans (Figure 4). About one-thirdof the 14 bioregions within the Gulf of Californiawere covered almost entirely by selected priorityareas of most plans, including the highly productiveUpper Cortezian Inner Neritic and Midriff IslandsStraits. In contrast, significant portions of otherbioregions, including the Slope and Basins, theGuaymean Neritic, and the East Pacific Rise wereonly partially incorporated into priority maps.

Priority areas were concentrated in coastal andshallow areas (<300 m depth), with frequency ofprioritization decreasing with depth. This is evidentin the biased coverage of geomorphological features(Figure 5(a)), with priority areas more oftenidentified on the continental shelf. Features such astrenches, submarine canyons, and valleys wereoften omitted. Ecosystems were also unevenlyrepresented (Figure 5(b)). Coral reefs, seagrasses,and rhodolith and Sargassum beds were more oftenincluded in plans, while others such as seamountswere rarely included, probably because of limitedinformation. Mangroves were included to someextent in most plans, but the portion of mangrovecovered by high coincidence of priority areas (i.e.overlaps >4 exercises) was minimal. Partialinclusion of mangroves can be related to theiroccurrence along the land–sea interface (planninghas largely focused on marine ecosystems) and totheir relatively wide distribution along the coast,making their frequently represented areas small inpercentage terms.

Achievement of goals, implementation of spatialactions, and benefits of planning

Through interviews and surveys, experts assessedthe reported and implicit goals of the planningexercises in the Gulf of California, and the extentto which these had been achieved (Figure 6). Anexplicit goal cited by all experts was identifying

generic priority areas for conservation, whileidentifying the particular conservation actionsrequired within these areas was generallyconsidered to be implicit. Other explicit goalsindicated by most experts were to document andassess biodiversity and threats and/or conflictswith human activities, as well as to guideconservation and resource use planning. Themajority of experts agreed that, overall, reportedgoals had been achieved satisfactorily. Some expertsdisagreed with this perception, mostly because theapplication of planning outputs had progressed veryslowly beyond local ongoing initiatives that lackedthe perspective potentially provided by the planningexercises. Opinions regarding implicit goals weremore varied. The use of existing priority maps bylater plans was commonly perceived as a positiveoutcome from planning, but never reported ormentioned as an implicit goal.

Despite differences in opinions, experts commonlyindicated that plans had guided or in some wayinfluenced the implementation of spatial actions

Figure 5. Representation of geomorphology classes and ecosystemswithin different levels of overlap of priority conservation areas.Graphs show the percentages of the total area of (a) marinegeomorphology classes and subclasses and (b) selected coastal andmarine ecosystems that correspond to three classes of frequency of

spatial coincidence: low (1–2), medium (3–4), and high (5–6).



within priority areas (Figure 7). According to morethan half of the experts interviewed, a commonoutcome of plans was the implementation ofmultiple-use protected areas (e.g. Bahia de losAngeles, Espiritu Santo National Park, SanLorenzo Archipelago National Park, San PedroMartir Biosphere Reserve). Two actions, althoughonly cited a few times, were the creation of no-takezones (e.g. Guaymas Hydrothermal Vents) andthe promotion of socio-ecological monitoring orresearch projects within priority areas (e.g.PANGAS: http://pangas.arizona.edu/es/inicio). Toa lesser extent, experts also considered that plans hadcontributed to informing coastal-marine informingand zoning plans, mainly by providing informationto agencies and industry regarding the location ofareas of conservation value.

According to experts, factors that facilitatedspatial implementation included: bringing priorityareas to the attention of agencies, organizations,and general public; having a regional or national

spatial framework that provided sound scientificsupport for ongoing marine conservation effortsand supported decisions by agencies regardingthe creation or zoning of MPAs; and participationof multiple stakeholders in planning, in particularthose directly influencing conservation decisionsin the region. On the other hand, some factorswere perceived as major constraints on theimplementation of plans or causes of the slow uptakeof plans in general, including: conflicting mandatesor priorities between conservation and naturalresource management agencies and organizations;limited technical capacities and public conservationresources (particularly at the municipal and Statelevels); insufficient political determination (e.g. fromState governments) to implement MPAs; theneed to further refine the design of conservationareas (e.g. incorporate socioeconomic constraints,biological connectivity, zones) and schedule theirimplementation; and the fact that plans have notbeen incorporated by relevant agencies intotheir strategic plans and annual work programmes,particularly agencies not directly related toconservation, e.g. tourism, water management,infrastructure, and development.

In addition to spatial actions, experts consideredthat planning brought other benefits within theregion (Figure 8). Despite differences in opinions,enhancing institutional and knowledge/skills wascommonly perceived as a benefit of planning.Institutional benefits included: standardizing orhaving more rigorous planning processes,facilitating the development/implementation ofregional or local conservation strategies, andinfluencing conservation strategies and actions by

Figure 6. Expert opinion on reported and implicit planning goals. (a) Percentages of experts (N=13) that identified the described planning goals asbeing explicitly reported (black portion of the bar) or implicitly considered (grey portion) by planners when developing the planning exercises; onlythe last goal was common to all planning exercises (either implicitly or explicitly). (b) Percentages of experts that considered the overallachievement of goals within each level of achievement, from ‘Not at all’ (experts strongly disagreed that goals were achieved) to ‘Fully’ (experts

completely agreed that overall goals were met).

Figure 7. Expert opinion on how planning influenced or led to spatialimplementation of actions. Bars denote the percentages of experts(N=13) that considered the described actions as being implemented

as a result of planning, either directly or indirectly.



http://pangas.arizona.edu/es/inicio

organizations/agencies/partners, guided by thepriority maps. Other important benefits identifiedwere developing skills/expertise of organizationsin conservation planning and building thesupporting scientific evidence for decision-makingby participating organizations, agencies or partners.Among the social benefits, improving collaborationand coordination between organizations, creating aregional conservation approach for conservation,and building trust among organizations wereapparent common outcomes. In comparison,direct financial benefits (e.g. making better use ofavailable conservation resources, improvingefficiency in the operations of organizations) wereless commonly noted.

DISCUSSION

The state of marine conservation planning in the Gulfof California

Despite the growing number of studies and marineapplications of systematic conservation planningmethods and tools, uptake has been relativelylimited in the Gulf of California. Only two planningexercises (Sala et al., 2002; Ulloa et al., 2006) usedconservation planning software to find least-costsolutions that achieved conservation objectives,even though these tools have been readily availablefor more than 10 years (Ball et al., 2009), withapplications to marine planning dating back to 1999(Leslie, 2005). Emerging issues, including targeting

Figure 8. Expert opinion on the benefits of planning apart from spatial implementation of actions. The black portion of each bar represents thepercentages of experts (N=13) that considered the described benefits were a result of planning; the grey portion depicts the percentages that

indicated that those benefits were unexpected; white represents the percentages of experts not citing those benefits as outcomes of planning.



of ecosystem services and consideration of land–seaconnections, have also received limited attentionin the Gulf. Notably, none of the planning exercisesdealt with climate change or scheduling ofconservation actions.

Conservation plans in the Gulf of Californiafollowed some best-practices regarding stakeholderinvolvement but fell short for others. Stakeholderinput was decisive in five stages of planning:identifying broad goals; setting objectives; definingselection criteria for priority areas; supplying,validating, and generating spatial data to representconservation features; and identifying priority areas(see Pomeroy and Douvere (2008) for a review onstakeholder participation in marine planning). Expertworkshops facilitated stakeholder involvement indecision-making processes in the Gulf of Californiaand contributed to filling gaps concerning ecologicaland socioeconomic data (Wheeler et al., 2008). Inaddition, highly participatory processes (e.g. SPMand CSGC) also contributed to building commonvisions of conservation problems and opportunities.In expert-driven processes, tools other thanconservation planning software helped to visualizespatial data (Enríquez-Andrade et al., 2005;CONABIO, 2007), facilitate group decision-making(Arriaga-Cabrera et al., 1998), and delineate priorityareas (Morgan et al., 2005). A notable gap inmost exercises was the limited input from naturalresource users and industry, which was mostlyrestricted to providing information on highly usedareas. Including all stakeholders, but particularlyresource users, from the outset of planning helps tobuild ownership and support (Lundquist andGranek, 2005; Gilliland and Laffoley, 2008), and hasbeen pivotal in successful regional-scale exercises inmarine planning (Fernandes et al., 2005; Gleasonet al., 2010).

Only two planning exercises set quantitativeobjectives (Sala et al., 2002; Ulloa et al., 2006),which contributed to more explicit, transparent,and justifiable decision-making processes (Margulesand Pressey, 2000). While qualitative objectives (e.g.maximize species richness) were common to allexercises, only Sala et al. (2002) quantified andtargeted species richness (for fish species).Both exercises that used quantitative objectives alsopartitioned these objectives between occurrences ofthe same features in different bioregions.Incorporating bioregional representativeness as aconservation objective is critical to ensuring thatMPA networks capture the variety of biological

assemblages, including species that are poorlyrecorded or undescribed (Roberts et al., 2003a).Bioregional representation has been fundamentalto other marine planning initiatives, e.g. theCalifornia Channel Islands Marine Reserves(Airame et al., 2003) and the Great Barrier ReefMarine Park rezoning (Fernandes et al., 2005).

Incorporating spatially explicit socioeconomicdata (e.g. human activities, opportunity costsassociated with use restrictions, and costs ofmanaging MPAs) in marine planning is critical tominimizing the costs of interventions (Klein et al.,2008; Ban and Klein, 2009). Exercises from theGulf of California illustrate contrasting approachesto incorporating threats and costs in planning. Thetwo exercises that used conservation planningsoftware explicitly avoided threatened areas andminimized conflict with fisheries (Sala et al., 2002;Ulloa et al., 2006). In both cases, field data onfishing effort informed the cost layers. Despitepotential limitations and coarse spatial resolution ofthese data, they are likely to be more accurate thansurrogates for socioeconomic variables (Weekset al., 2010b). In contrast, the B2B exercise(Morgan et al., 2005) focused on areas whereconservation features of interest and threatsintersected (i.e. prioritized vulnerable areas), anapproach that aims to reduce further loss ofconservation features. Although these approachesmight seem opposite, they are likely to converge.Application of actions following selection of areassuch as those by Sala et al. (2002) and Ulloa et al.(2006) would require scheduling, in which areas ofboth high irreplaceability and high threat wouldneed to be prioritized (Pressey and Bottrill, 2008).Surprisingly, most exercises identified priorityareas independently of their vulnerability or levelof use, either assessing threats independently(Enríquez-Andrade et al., 2005) or designingstrategies to address threats after priorities were set(Arriaga-Cabrera et al., 1998; Enríquez-Andradeand Danemann, 1998; CONABIO, 2007).

An interesting finding of the CSGC exercise(Enríquez-Andrade et al., 2005) was that priorityareas tended to be heavily used or highly threatened,indicating that conservation management in someused and threatened areas might be inevitable. Thisemphasizes the need for better data on threats andcosts and an understanding of management actionsthat are feasible and effective in contested parts ofthe Gulf. Conflict over conservation and extractiveuses can be reduced, but not altogether avoided,



by explicitly incorporating costs and threats inthe prioritization process. Combining multiple costs(e.g. to different stakeholder groups) is challengingfor marine planners (Ban and Klein, 2009), butshould be considered in the Gulf of California.

Despite spatial data limitations, ecological processes(e.g. migrations and oceanographic phenomenaassociated with high productivity) were considered bymost planning exercises. Readily-available biologicaland oceanographic data (e.g. satellite-derived dataon spatial variability of high-productivity areas)(Hidalgo-González and Alvarez-Borrego, 2011),combined with existing oceanographic-ecologicalmodels (Marinone et al., 2008; Cudney-Bueno et al.,2009), can help to improve planning for processesin the Gulf of California and elsewhere.Recent spatial planning that could be adapted tothe Gulf has incorporated extensive pelagic andoceanographic phenomena (Grantham et al., 2011)and considered the effects of climate change(Hobday, 2011). These approaches will help toensure local functionality of priority areas as wellas contributing to regional-scale conservationobjectives (Roberts et al., 2003b), including thoserelated to fisheries.

Notwithstanding the global importance anddegradation of the services provided by coastalmarine ecosystems (Worm et al., 2006; Barbieret al., 2010), particularly in the Gulf of California(Aburto-Oropeza et al., 2008; Ezcurra et al.,2009), these were not considered explicitly byany of the planning exercises. This gap might havebeen partially addressed, however, by targetingconservation features based on their potentialeconomic value, including: areas of highproductivity and plankton abundance for fisheries(Arriaga-Cabrera et al., 1998; CONABIO, 2007);coastal ecosystems (e.g. coral reefs and kelp forest)that provide protection to sandy beaches againstwave action; estuaries and marshes as exporters ofnutrients/energy and buffers against land-basedthreats (Enríquez-Andrade and Danemann, 1998;Morgan et al., 2005); and mangroves as nurseriesfor commercial fish and invertebrates (Sala et al.,2002; Enríquez-Andrade et al., 2005). The omissionof ecosystem services is common in marine spatialplanning (Leslie, 2005; Granek et al., 2010),although there have been recent advances in theirvaluation and mapping (Tallis and Polasky, 2009;Silvestri and Kershaw, 2010). Further research isneeded for marine applications (Chan andRuckelshaus, 2010).

The importance of land–sea interactions wasrecognized by most planning exercises, both interms of cross-system threats and ecologicalconnections. However, spatial data on, or modelsof, land-based threats were very limited and onlyused in a couple of plans. Prioritization ofcatchments to mitigate land-based threats was notconsidered by any plan. Considering land-basedthreats can alter marine priority areas (e.g. bylocating them away from imminent land-basedsources of pollution when possible) (Tallis et al.,2008), but can also help to integrate terrestrial andmarine planning and facilitate marine management(Klein et al., 2010). Only one exercise(Enríquez-Andrade et al., 2005) explicitly targetedfreshwater and terrestrial areas important tomaintaining ecological processes connecting landand sea. Adjustments of priority maps can also berequired to plan for land–sea processes, e.g.species migrations between disjunct terrestrial andmarine habitats (Hazlitt et al., 2010) or input ofmarine-derived nutrients into coastal and insularecosystems (Lombard et al., 2007). The lack ofconsideration of land–sea connections in theNational Protected Area System of Mexico wasemphasized by Ortiz-Lozano et al. (2009), andhighlights the need to improve data and models onspatial and temporal occurrence of land–seaprocesses and cross-system threats for marineplanning (Alvarez-Romero et al., 2011).

Convergence and divergence between plans

Based on a spatial overlay of priority maps,Aburto-Oropeza and Lopez-Sagastegui (2006)noted important spatial coincidences between theoutputs of different marine planning exercises inthe Gulf. These similarities have led to a generalperception among conservation practitioners in theregion that planning exercises identified the samepriority areas. The present analysis shows that,despite areas of convergence, there are importantdifferences, some of which can be related to theplanning approach, methods and extent. The smalloverlap between all exercises raises the question ofwhether the convergence of multiple independentplans is in fact pointing to areas that hold valuablebiodiversity features and processes that have beenrecognized widely by planners. In some cases,areas commonly identified as marine prioritiesseem to be located in better studied areas (e.g.Upper Gulf, Midriff Islands, Loreto, La Paz Bay),a common outcome in marine planning in general



(Ardron et al., 2008) and a plausible explanation ofwhy these areas also overlap with existing MPAs(Weeks et al., 2010a), some of which wereestablished before prioritization exercises.

The similarities and differences in plans mightbe explained by commonalities between them,including similar data, similar approaches (e.g.expert-based vs. data driven, prioritization ofthreatened areas with biodiversity value, focus onnearshore environments), the use of priority mapsfrom previous exercises, and the resolution of dataand targeted conservation features. Contrary toexpectations, clustering based on spatial similaritydid not show clear differences between prioritymaps developed at different spatial scales. Animportant result from the Kappa location analysiswas the greater similarity values among all exercisescompared with the overall Kappa. This indicatesthat the relative proportions/size of priority areas(which can be related to the extent of planningdomain) can be an important driver of overallspatial similarity between overlapping priority maps.

Interestingly, grouping based on spatial similaritydid not closely match that based on planningobjectives and datasets. Divergence between themethodological and spatial similarity phenogramssuggests that other factors, in addition to data andobjectives, played an important role in defining thedistribution of conservation priorities. These couldinclude inaccuracy in manual delineation of priorityareas from expert-based exercises, incorporation ofupdated datasets, and random error. Unfortunately,the limited and fragmented spatial data availablefrom these studies did not allow furtherinvestigation of these differences.

Size and perimeter have important designimplications, both biologically and practically(Roberts et al., 2001; Leslie et al., 2003). Remarkably,priority areas proposed by the only exercisethat explicitly included minimum size as a designcriterion (Sala et al., 2002) were significantly smallerthan those from other exercises (Figure 2(a)). Inpart, this is explained by the reduced number ofplanning units, but also reflects differences inobjectives and how these can influence priority areadesign.

Convergence of priorities was observed in wellstudied shallow and benthic marine ecosystems (e.g.coral reefs, seagrass, Sargassum beds). Coastalecosystems were also represented in establishedMPAs, some of which were established at leastpartially in response to the exercises reviewed here.

The recent addition of an MPA in the Gulf ofCalifornia to protect the benthic communities(hydrothermal vents) in the Guaymas Basin – anarea identified as a marine priority by only oneexercise – is an exception to the general trend. Theunder-representation of deep-ocean and pelagicenvironments in the Gulf of California, incomparison with nearshore/shallow ecosystems,reflects the relatively limited data for deep-oceanand pelagic environments in general (Glover andSmith, 2003; Webb et al., 2010) and the emphasison the protection of near-shore ecosystems(Wood et al., 2008; Spalding et al., 2010). Among thesolutions, vertical zoning (i.e. implementing differentuse restrictions along the water column) provides away of designing priority areas suitable for protectingpelagic and/or benthic environments as required(Grober-Dunsmore et al., 2008). New technologiesare allowing the exploration of deeper environmentsin the Gulf of California (Aburto-Oropeza et al.,2010a) and the data from these studies should beused to update marine planning exercises.

Uptake of planning: implementation of spatial actionsand other benefits

Assessing the implementation of plans was notan objective of this study. However, based onthe information provided by experts, theimplementation of new MPAs in the region hasbeen influenced by some of these planning exercises.It seems that monitoring and research activitieshave resulted from the long history of prioritizingconservation areas in the region. However, asthe experts noted, enforcement and many otherproblems (Rife et al., 2012) indicate that MPAsare not necessarily fulfilling the objectives stated intheir decrees.

An element regarded by some experts as aconstraint on the implementation of spatial actionswas the lack of regard for ongoing refinement ofpriority area designs to maximize the likelihood ofimplementation. Such refinements include definingsimple and readily recognizable boundaries tomaximize compliance and facilitate enforcement.These refinements have been applied in otherregions to maximize implementation feasibility, thusleading to the successful application of planningoutputs (Fernandes et al., 2005, 2009; Lombardet al., 2007). These exercises also illustrate the needto work with stakeholders once preliminary prioritymaps are proposed. The execution of specificactions such as surveillance, species recovery, and



invasive species eradication/control programmeswithin priority areas were seldom mentioned byexperts as being implemented following planning.This suggests that the failure to identify specificactions within priority areas (to address problemsassociated with particular areas) may also belimiting the direct application of conservationplans because further work to identify adequateinterventions is needed. Interestingly, none of theinterviewees mentioned actions related to fisheriesmanagement, which could in part reflect the limitedparticipation of the fisheries sector in planning andthe general disconnect between this sector with thenational and regional conservation agenda.

The implementation of MPAs in the Gulf ofCalifornia, as everywhere in Mexico (Bezaury-Creel,2005), has been mostly opportunistic, thus limitingthe uptake of marine planning outputs. Reasonsunderlying this approach include incomplete andbiased socioecological data, accelerated degradationof certain areas (and reactive local efforts to protectthem), and concurrence of suitable socioeconomicand political conditions. These elements can,however, be considered and incorporated intoSCP initiatives by following an adaptive approach.Adaptive conservation planning is critical tomake the best use available socioecological data,deal with ongoing loss of biodiversity, and takeadvantage of emerging conservation opportunities(Grantham et al., 2010). Some key requirements toadaptive planning are: setting explicit and preferablyquantitative conservation objectives (critical to thetransparency, traceability and revision of planningdecisions); developing spatial data infrastructures(needed to organize and share data to revise the plansor to inform other plans, as well as to continuouslyand systematically update socioecological dataderived from monitoring and research); and assessingthe effectiveness or response of socioecologicalsystems to conservation interventions (essentialto adjust objectives, priority area design, andimplementation strategies). Unfortunately, fewlong-term monitoring programmes (a pre-requisiteto adaptive management) exist in the regiontoday, and there are no explicit intentions to updatethe plans, even when information from theseprogrammes is available. The lack of these twostrategies is a major limitation of marine planningin the Gulf of California. Conservation planninginitiatives in South Africa exemplify how nationaland regional-scale plans are being documented,updated and implemented under a national

multi-scale planning framework (e.g. South AfricanNational Biodiversity GIS: http://cpu.uwc.ac.za/).Additional elements that have constrained orfacilitated implementation include institutionalcontinuity, funding availability, and political trends,so these need to be considered by planners.

CONCLUSIONS

The Gulf of California has seen seven marineplanning exercises in the past decade. Yet uptakeof marine applications of systematic conservationplanning methods and tools has been relativelylimited, and the existence of multiple plans withinthe region pointed to some difficulties as well asbenefits. On one hand, multiple plans havecontributed to uncertainty, confusion and tensionbetween stakeholders. Multiple plans may havealso contributed to the spreading and inefficientuse of financial and human resources, as well as tounnecessary redundancy in actions. Despite goodintentions, the coexistence of multiple plans(particularly when these are developed withoutconsidering pre-existing plans) can hamper thegoal to create a common approach and focus forconservation, thus resulting in uncoordinatedactions undertaken by the different stakeholders.On the other hand, the existence of multipleplanning processes and plans seems to havecontributed to improved methods, data and tools,as well as being the basis for more robust andrefined plans developed later.

Given that the National Marine Gap Analysis(CONABIO, 2007) considered the outputs ofprevious plans (incorporating some priority areasdelineated in previous exercises) and engaged withstakeholders who participated in previous exercises,this plan has now been widely accepted and isconsidered (at least by environmental organizationsand agencies) the national framework for marineconservation in Mexico, including the Gulf ofCalifornia. Thus, the logical next step is to build onthis plan and refine priority areas (which providethe required regional context) into conservationareas associated with specific actions (e.g. throughzoning) appropriate to address current threats,maintaining connectivity incorporating localsocioeconomic data, and ensuring participation oflocal stakeholders. This refinement should includeintegration with terrestrial and freshwater priorities,in particular for areas where important land–seaconnections have been identified (Enríquez-Andrade



http://www.rivm.nl/milieu/modellen

et al., 2005) or where land-based threats are aconcern. A stronger commitment of agencies fromdifferent sectors (environment, fishing, tourism,water management, infrastructure anddevelopment) is also needed to incorporate planningresults into the relevant policies, regulations andannual programs, as well as to continueparticipating in the refinement of plans. Increasedcoordination between organizations (from local tointernational) will be fundamental to implement afunctional network of MPAs that contributes toachieving conservation goals and the sustainableuse of natural resources in the Gulf of California.

ACKNOWLEDGEMENTS

We thank the following persons for providinginformation on the planning exercises: AmyHudson-Weaver, Anne Gondor, Cesar Sanchez-Ibarra,Diana Hernandez, Enric Sala, Fan Tsao, Juan E.Bezaury, Lance Morgan, Marcia Moreno-Baez,Melanie Kolb, Patricia Koleff, Raul Ulloa, RocioEsquivel, and Veronica Aguilar. We thankJohnathan Kool and Morena Mills for theirvaluable comments to improve this manuscript. Wealso thank the following organizations for provingdatasets and reports on the planning exercises:Comision Nacional para el Conocimiento y Usode la Biodiversidad, Comunidad y Biodiversidad,Pronatura Noroeste, Marine ConservationBiology Institute, North America’s Commissionfor Environmental Cooperation, The NatureConservancy-Mexico, and World WildlifeFund-Mexico. We also thank Gordon Bailey forproviding IT support and the High PerformanceComputing Unit at James Cook Universityfor computational facilities. JGAR gratefullyacknowledges support from Mexico’s ConsejoNacional de Ciencia y Tecnología (CONACYT)and Secretaría de Educación Pública (SEP), aswell as from the Australian Research CouncilCentre of Excellence for Coral Reef Studies.RLP and NCB acknowledge the support of theAustralian Research Council.

REFERENCES

Aburto-Oropeza O, Lopez-Sagastegui C. 2006. A marinereserve network for the Gulf of California the next step inconserving the world’s aquarium. Greenpeace, Mexico, D.F.

Aburto-Oropeza O, Ezcurra E, Danemann G, Valdez V,Murray J, Sala E. 2008. Mangroves in the Gulf of

California increase fishery yields. Proceedings of theNational Academy of Sciences of the United States ofAmerica 105: 10456–10459.

Aburto-Oropeza O, Erisman B, Caso M, Ezcurra E. 2010a.Log of the Deep Sea: An Expedition to the Seamounts of theSea of Cortes. Instituto Nacional de Ecologia: Mexico, DF.

Aburto-Oropeza O, Paredes G, Mascareñas-Osorio I, Sala E.2010b. Climatic influence on reef fish recruitment andfisheries. Marine Ecology Progress Series 410: 283–287.

Airame S, Dugan JE, Lafferty KD, Leslie H, McArdle DA,Warner RR. 2003. Applying ecological criteria to marinereserve design: a case study from the California ChannelIslands. Ecological Applications 13: S170–S184.

All JD. 2006. Colorado river floods, droughts, and shrimpfishing in the upper Gulf of California, Mexico.Environmental Management 37: 111–125.

Allison GW, Gaines SD, Lubchenco J, Possingham HP. 2003.Ensuring persistence of marine reserves: catastrophesrequire adopting an insurance factor. EcologicalApplications 13: 8–24.

AlmanyG, Connolly S, Heath D, Hogan J, Jones G,McCook L,Mills M, Pressey R, Williamson D. 2009. Connectivity,biodiversity conservation and the design of marine reservenetworks for coral reefs. Coral Reefs 28: 339–351.

Alvarez-Romero JG, Pressey RL, Ban NC, Vance-Borland K,Willer C, Klein CJ, Gaines SD. 2011. Integrated land–seaconservation planning: the missing links. Annual Review ofEcology, Evolution, and Systematics 42: 381–409.

Ardron JA, Possingham H, Klein C. 2008. Marxan GoodPractices Handbook. Pacific Marine Analysis and ResearchAssociation: Vancouver, BC.

Arriaga-Cabrera L, Vázquez-Domínguez E, González-Cano J,Jiménez-Rosenberg R, Muñoz-López E, Aguilar-Sierra V.1998. Regiones marinas prioritarias de México. ComisiónNacional para el Conocimiento y uso de la Biodiversidad(CONABIO): Mexico, DF.

Baker N, Beger M, McClennen C, Ishoda A, Edwards F. 2011.Reimaanlok: a national framework for conservation areaplanning in the Marshall Islands. Journal of Marine Biology2011: 1–11.

Ball I, Possingham H, Watts DJ. 2009. Marxan and relatives:software for spatial conservation prioritization. In SpatialConservation Prioritization Quantitative Methods andComputational Tools, Moilanen A, Wilson K, PossinghamH (eds). Oxford University Press: Oxford; 185–195.

Ban NC. 2009. Minimum data requirements for designing aset of marine protected areas, using commonly availableabiotic and biotic datasets. Biodiversity and Conservation18: 1829–1845.

Ban NC, Klein CJ. 2009. Spatial socioeconomic data as a costin systematic marine conservation planning. ConservationLetters 2: 206–215.

Barbier EB, Hacker SD, Kennedy C, Koch EW, Stier AC,Silliman BR. 2010. The value of estuarine and coastalecosystem services. Ecological Monographs 81: 169–193.

Beger M, Linke S, Watts M, Game E, Treml E, Ball I,Possingham HP. 2010. Incorporating asymmetricconnectivity into spatial decision making for conservation.Conservation Letters 3: 359–368.

Beman JM, Arrigo KR, Matson PA. 2005. Agricultural runofffuels large phytoplankton blooms in vulnerable areas of theocean. Nature 434: 211–214.

Bezaury-Creel JE. 2005. Protected areas and coastal and oceanmanagement in México. Ocean & Coastal Management 48:1016–1046.

Bottrill MC, Pressey RL. 2012. The effectiveness andevaluation of conservation planning. Conservation Letters 5:407–420.



Bottrill MC, Mills M, Pressey RL, Game ET, Groves C. 2012.Evaluating perceived benefits of ecoregional assessments.Conservation Biology 26: 851–861.

Brusca R, Findley LT, Hastings PA, Hendrickx ME, Cosio JT,van der Heiden AM, Cartron J, Ceballos G, Felger R. 2005.Macrofaunal diversity in the Gulf of California. InBiodiversity, Ecosystems, and Conservation in NorthernMexico, Cartron J, Ceballos G, Felger R (eds). OxfordUniversity Press: New York; 179–203.

Carriquiry JD, Sánchez A. 1999. Sedimentation in theColorado River delta and Upper Gulf of California afternearly a century of discharge loss. Marine Geology 158:125–145.

CEC. 2006. Ecological Regions of North America: Toward aCommon Perspective - Update. Commission forEnvironmental Cooperation: Montreal, Quebec, Canada.

Chan KMA, Ruckelshaus M. 2010. Characterizing changes inmarine ecosystem services. F1000 Biology Reports 2.

CONABIO. 2007. Analisis de vacios y omisiones enconservacion de la biodiversidad marina de Mexico: oceanos,costas e islas. Comision Nacional para el Conocimiento yUso de la Biodiversidad (CONABIO); Comision Nacionalde Areas Naturales Protegidas; The Nature Conservancy;Pronatura A.C.: Mexico, DF.

CONABIO. 2008. Distribución de los manglares de México.Escala 1:50 000. Extraído del proyecto DQ056: Programa demonitoreo de los manglares de México a largo plazo, primeraetapa. Comisión Nacional para el Conocimiento y Uso dela Biodiversidad: Mexico, DF.

CONANP. 2010. Cobertura de las Áreas Naturales ProtegidasFederales y Estatales de México. Comisión Nacional deÁreas Naturales Protegidas, Dirección de Evaluación ySeguimiento, Subdirección Encargada de la Coordinaciónde Geomática: Mexico, DF.

Cudney-Bueno R, Lavín MF, Marinone SG, Raimondi PT,Shaw WW. 2009. Rapid effects of marine reserves via larvaldispersal. PloS One 4: e4140.

Day JC. 2002. Zoning - lessons from the Great Barrier ReefMarine Park. Ocean & Coastal Management 45: 139–156.

Edwards HJ, Elliott IA, Pressey RL, Mumby PJ. 2010.Incorporating ontogenetic dispersal, ecological processesand conservation zoning into reserve design. BiologicalConservation 143: 457–470.

Enríquez-Andrade R, Danemann G. 1998. Identificación yestablecimiento de prioridades para las acciones deconservación y oportunidades de uso sustentable de los recursosmarinos y costeros de la Península de Baja California. Reportetécnico de proyecto. Pronatura Península de BajaCalifornia, Ensenada, Baja California.

Enríquez-Andrade R, Anaya-Reyna G, Barrera-Guevara JC,C-M MdlÁ, Martínez-Delgado ME, Vaca-Rodríguez J,Valdés-Casillas C. 2005. An analysis of critical areas forbiodiversity conservation in the Gulf of California Region.Ocean & Coastal Management 48: 31–50.

Erisman B, Mascarenas I, Paredes G, de Mitcheson YS,Aburto-Oropeza O, Hastings P. 2010. Seasonal, annual,and long-term trends in commercial fisheries for aggregatingreef fishes in the Gulf of California, Mexico. FisheriesResearch 106: 279–288.

Ezcurra E, Aburto-Oropeza O, C-M MdlÁ, Cudney-Bueno R,Torre J. 2009. Gulf of California, Mexico. In Ecosystem-basedManagement for the Oceans, McLeod K, Leslie H (eds). IslandPress: Washington, DC; 227–252.

Fernandes L, Day J, Lewis A, Slegers S, Kerrigan B, Breen D,Cameron D, Jago B, Hall J, Lowe D, et al. 2005. Establishingrepresentative no-take areas in the Great Barrier Reef:Large-scale implementation of theory on marine protectedareas. Conservation Biology 19: 1733–1744.

Fernandes L, Day J, Kerrigan B, Breen D, De’ath G,Mapstone B, Coles R, Done T, Marsh H, Poiner I, et al.2009. A process to design a network of marine no-takeareas: Lessons from the Great Barrier Reef. Ocean andCoastal Management 52: 439–447.

FoleyM,Halpern B,Micheli F, ArmsbyM,CaldwellM,CrainC,Prahler E, Rohr N, Sivas D, Beck M. 2010. Guidingecological principles for marine spatial planning. MarinePolicy 34: 955–966.

Galindo-Reyes JG, Villagrana LC, Lazcano-Alvarez G. 1999.Environmental conditions and pesticide pollution of twocoastal ecosystems in the Gulf of California, Mexico.Ecotoxicology and Environmental Safety 44: 280–286.

Game ET, Grantham HS, Hobday AJ, Pressey RL, LombardAT, Beckley LE, Gjerde K, Bustamante R, PossinghamHP, Richardson AJ. 2009. Pelagic protected areas: themissing dimension in ocean conservation. Trends in Ecology& Evolution 24: 360–369.

Game ET, Lipsett-Moore G, Hamilton R, Peterson N,Kereseka J, Atu W, Watts M, Possingham H. 2011.LETTER: Informed opportunism for conservation planningin the Solomon Islands. Conservation Letters 4: 38–46.

García-Hernández J, Cadena-Cárdenas L, Betancourt-LozanoM,García-De-La-Parra LM, García-Rico L, Márquez-Farías F.2007. Total mercury content found in edible tissues oftop predator fish from the Gulf of California, Mexico.Toxicological and Environmental Chemistry 89: 507–522.

Gilliland PM, Laffoley D. 2008. Key elements and steps in theprocess of developing ecosystem-based marine spatialplanning. Marine Policy 32: 787–796.

Gleason M, McCreary S, Miller-Henson M, Ugoretz J, Fox E,Merrifield M, McClintock W, Serpa P, Hoffman K. 2010.Science-based and stakeholder-driven marine protectedarea network planning: a successful case study fromnorth central California. Ocean and Coastal Management53: 52–68.

Glenn EP, Nagler PL, Brusca RC, Hinojosa-Huerta O. 2006.Coastal wetlands of the northern Gulf of California:inventory and conservation status. Aquatic Conservation:Marine and Freshwater Ecosystems 16: 5–28.

Glover AG, Smith CR. 2003. The deep-sea floor ecosystem:current status and prospects of anthropogenic change by theyear 2025. Environmental Conservation 30: 219–241.

Granek EF, Polasky S, Kappel CV, Reed DJ, Stoms DM,Koch EW, Kennedy CJ, Cramer LA, Kacker SD, BarbierEB, et al. 2010. Ecosystem services as a common languagefor coastal ecosystem-Based Management. ConservationBiology 24: 207–216.

Grantham HS, Bode M, McDonald-Madden E, Game ET,Knight AT, Possingham H. 2010. Effective conservationplanning requires learning and adaptation. Frontiers inEcology and the Environment 8: 431–437.

Grantham HS, Game ET, Lombard AT, Hobday AJ,Richardson AJ, Beckley LE, Pressey RL, Huggett JA,Coetzee JC, van der Lingen CD, et al. 2011. Accommodatingdynamic oceanographic processes and pelagic biodiversity inmarine conservation planning. PloS One 6: e16552.

Green A, Smith SE, Lipsett-Moore G, Groves C, Peterson N,Sheppard S, Lokani P, Hamilton R, Almany J, Aitsi J,Bualia L. 2009. Designing a resilient network of marineprotected areas for Kimbe Bay, Papua New Guinea. Oryx43: 488–498.

Grober-Dunsmore R, Wooninck L, Field J, Ainsworth C,Beets J, Berkeley S, Bohnsack J, Boulon R, Brodeur R,Brodziak J, et al. 2008. Vertical zoning in marine protectedareas: ecological considerations for balancing pelagic fishingwith conservation of benthic communities. Fisheries 33:598–610.



Groves CR, Jensen DB, Valutis LL, Redford KH, Shaffer ML,Scott JM, Baumgartner JV, Higgins JV, Beck MW,Anderson MG. 2002. Planning for biodiversityconservation: Putting conservation science into practice.Bioscience 52: 499–512.

Gutierrez-Galindo EA, Flores Munoz G, Ortega Garcia ML,Villaescusa Celaya JA. 1992. Pesticides in coastal waters ofthe Gulf of California: Mussel Watch Program 1987–1988.Pesticidas en las aguas costeras del Golfo de California:Programa de Vigilancia con Mejillon 1987–1988 18: 77–99.

Hazlitt SL, Martin TG, Sampson L, Arcese P. 2010. Theeffects of including marine ecological values in terrestrialreserve planning for a forest-nesting seabird. BiologicalConservation 143: 1299–1303.

Hidalgo-González RM, Alvarez-Borrego S. 2011. Total andnew production in the Gulf of California estimated fromocean color data from the satellite sensor SeaWIFS. DeepSea Research Part II: Topical Studies in Oceanography 51:739–752.

Hobday AJ. 2011. Sliding baselines and shuffling species:implications of climate change for marine conservation.Marine Ecology 32: 392–403.

Howarth RW. 2008. Coastal nitrogen pollution: a review ofsources and trends globally and regionally. Harmful Algae8: 14–20.

INE. 2005. Evaluacion preliminar de las tasas de perdida de lasuperficie de manglar en Mexico. Instituto Nacional deEcologıa, SEMARNAT: Mexico, DF.

Jaramillo-Legorreta A, Rojas-Bracho L, Brownell Jr RL,Read AJ, Reeves RR, Ralls K, Taylor BL. 2007. Saving thevaquita: immediate action, not more data. ConservationBiology 21: 1653–1655.

Klein C, Steinback C, Scholz A, Possingham H. 2008.Effectiveness of marine reserve networks in representingbiodiversity and minimizing impact to fishermen: acomparison of two approaches used in California.Conservation Letters 1: 44–51.

Klein CJ, Ban NC, Halpern BS, Beger M, Game ET,Grantham HS, Green A, Klein TJ, Kininmonth S, Treml E,et al. 2010. Prioritizing land and sea conservationinvestments to protect coral reefs. PloS One 5: e12431.

Lavín M, Marinone S. 2003. An overview of the physicaloceanography of the Gulf of California. In NonlinearProcesses in Geophysical Fluid Dynamics, Velasco OU,Sheinbaum J, Ochoa J (eds). Kluwer Academia Publishers:Dordrecht, Holland; 173–204.

Leslie H, RuckelshausM, Ball IR, Andelman S, PossinghamHP.2003. Using siting algorithms in the design of marine reservenetworks. Ecological Applications 13: S185–S198.

Leslie HM. 2005. Synthesis of marine conservation planningapproaches. Conservation Biology 19: 1701–1713.

Leslie HM, McLeod KL. 2007. Confronting the challengesof implementing marine ecosystem-based management.Frontiers in Ecology and the Environment 5: 540–548.

Lester SE, Halpern BS, Grorud-Colvert K, Lubchenco J,Ruttenberg BI, Gaines SD, Airame S, Warner RR. 2009.Biological effects within no-take marine reserves: a globalsynthesis. Marine Ecology Progress Series 384: 33–46.

Lluch-Cota SE, Aragon-Noriega EA, Arreguin-Sanchez F,Aurioles-Gamboa D, Bautista-Romero JJ, Brusca RC,Cervantes-Duarte R, Cortes-Altamirano R, Del-Monte-LunaP, Esquivel-Herrera A, et al. 2007. The Gulf of California:review of ecosystem status and sustainability challenges.Progress in Oceanography 73: 1–26.

Lluch-Cota SE, Parés-Sierra A, Magaña-Rueda VO,Arreguín-Sánchez F, Bazzino G, Herrera-Cervantes H,Lluch-Belda D. 2010. Changing climate in the Gulf ofCalifornia. Progress in Oceanography 87: 114–126.

Lombard AT, Reyers B, Schonegevel LY, Coopers J,Smith-Adao LB, Nel DC, Froneman PW, Ansorge IJ,Bester MN, Tosh CA, et al. 2007. Conserving pattern andprocess in the Southern Ocean: Designing a marineprotected area for the Prince Edward Islands. AntarcticScience 19: 39–54.

Lugo-Hubp J, Fernandez Córdova C. 1990. GeomorfologíaMarina. Obtenido de Geomorfología 1. IV.3.3., AtlasNacional de México, Vol. II, Escala 1:4000000. Instituto deGeografía, UNAM: Mexico, DF.

Lundquist CJ, Granek EF. 2005. Strategies for successfulmarine conservation: integrating socioeconomic, political,and scientific factors. Conservation Biology 19: 1771–1778.

Margules CR, Pressey RL. 2000. Systematic conservationplanning. Nature 405: 243–253.

Marinone SG, Ulloa MJ, Parés-Sierra A, Lavín MF,Cudney-Bueno R. 2008. Connectivity in the northern Gulfof California from particle tracking in a three-dimensionalnumerical model. Journal of Marine Systems 71: 149–158.

McCook L, Ayling T, Cappo M, Choat J, Evans R,De Freitas D, Heupel M, Hughes T, Jones G, Mapstone B,et al. 2010. Adaptive management of the Great BarrierReef: a globally significant demonstration of the benefitsof networks of marine reserves. Proceedings of theNational Academy of Sciences of the United States ofAmerica 107: 18278–18285.

McLeod E, Salm R, Green A, Almany J. 2009. Designingmarine protected area networks to address the impacts ofclimate change. Frontiers in Ecology and the Environment 7:362–370.

Mora C, Myers RA, Coll M, Libralato S, Pitcher TJ, SumailaRU, Zeller D, Watson R, Gaston KJ, Worm B. 2009.Management effectiveness of the world’s marine fisheries.PLoS Biology 7: e1000131.

Morgan L, Maxwell S, Tsao F, Wilkinson TAC, Etnoyer P.2005. Marine priority conservation areas: Baja California tothe Bering Sea. Commission for Environmental Cooperationof North America and the Marine Conservation BiologyInstitute: Montreal, Quebec, Canada.

Niño-Torres CA, Zenteno-Savín T, Gardner SC, Urbán RJ.2010. Organochlorine pesticides and polychlorinatedbiphenyls in fin whales (Balaenoptera physalus) from theGulf of California. Environmental Toxicology 25: 381–390.

Olson DM, Dinerstein E. 2002. The Global 200: priorityecoregions for global conservation. Annals of the MissouriBotanical Garden 89: 199–224.

Ortiz-Lozano L, Gutiérrez-Velázquez AL, Granados-Barba A.2009. Marine and terrestrial protected areas in Mexico:Importance of their functional connectivity in conservationmanagement. Ocean and Coastal Management 52: 620–627.

Osmond M, Airame S, Caldwell M, Day J. 2010. Lessonsfor marine conservation planning: A comparison of threemarine protected area planning processes. Ocean andCoastal Management 53: 41–51.

Páez-Osuna F, Ruiz-Fernández AC. 2005. Environmental loadof nitrogen and phosphorus from extensive, semiintensive,and intensive shrimp farms in the Gulf of Californiaecoregion. Bulletin of Environmental Contamination andToxicology 74: 681–688.

Pomeroy R, Douvere F. 2008. The engagement of stakeholdersin the marine spatial planning process. Marine Policy 32:816–822.

Pressey RL, Bottrill MC. 2008. Opportunism, threats, and theevolution of systematic conservation planning. ConservationBiology 22: 1340–1345.

Pressey RL, Bottrill MC. 2009. Approaches to landscape- andseascape-scale conservation planning: convergence, contrastsand challenges. Oryx 43: 464–475.



Rebman JP, De La Luz JLL, Moran RV. 2002. Vascular plantsof the Gulf islands. In A New Island Biogeography of the Seaof Cortes, Case TJ, Cody ML, Ezcurra E (eds). OxfordUniversity Press: New York; 465–511.

Rife AN, Erisman B, Sanchez A, Aburto-Oropeza O. 2012.When good intentions are not enough . . . insights onnetworks of “paper park” marine protected areas.Conservation Letters doi: 10.1111/j.1755-263X.2012.00303.x.

Roberts CM, Halpern BS, Palumbi SR, Warner RR. 2001.Designing marine reserve networks: why small, isolatedprotected areas are not enough. Conservation Biology inPractice 2: 11–17.

Roberts CM, McClean CJ, Veron JEN, Hawkins JP, AllenGR, McAllister DE, Mittermeier CG, Schueler FW,Spalding M, Wells F, et al. 2002. Marine biodiversityhotspots and conservation priorities for tropical reefs.Science 295: 1280–1284.

Roberts CM, Andelman S, Branch G, Bustamante RH,Castilla JC, Dugan J, Halpern BS, Lafferty KD, Leslie H,Lubchenco J, et al. 2003a. Ecological criteria for evaluatingcandidate sites for marine reserves. Ecological Applications 13:S199–S214.

Roberts CM, Branch G, Bustamante RH, Castilla JC,Dugan J, Halpern BS, Lafferty KD, Leslie H, Lubchenco J,McArdle D, et al. 2003b. Application of ecological criteriain selecting marine reserves and developing reservenetworks. Ecological Applications 13: 215–228.

Rowell K, Flessa KW, Dettman DL, Román MJ, Gerber LR,Findley LT. 2008. Diverting the Colorado River leads to adramatic life history shift in an endangered marine fish.Biological Conservation 141: 1138–1148.

Sala E, Knowlton N. 2006. Global marine biodiversitytrends. Annual Review of Environment and Resources 31:93–122.

Sala E, Aburto-Oropeza O, Paredes G, Parra I, Barrera JC,Dayton PK. 2002. A general model for designing networksof marine reserves. Science 298: 1991–1993.

Sala E, Aburto-Oropeza O, Reza M, Paredes G, López-LemusLG. 2004. Fishing down coastal food webs in the Gulf ofCalifornia. Fisheries 29: 19–25.

Silvestri S, Kershaw F. 2010. Framing the flow: innovativeapproaches to understand, protect and value ecosystemservices across linked habitats. UNEP World ConservationMonitoring Centre: Cambridge, UK.

Spalding M, Wood LJ, Fitzgerald C, Gjerde K. 2010. The 10%target: where do we stand? In Global Ocean Protection:Present Status and Future Possibilities. Toropova C,Meliane I, Laffoley D, Matthews E, Spalding M (eds). Agencedes aires marines protégées. Brest, France:; IUCN WCPA;UNEP-WCMC; TNC; UNU; and WCS: Gland, Switzerland;Washington, DC and New York, USA; Cambridge, UK;Arlington, USA; Tokyo, Japan; and New York; 25–41.

Stillman JH. 2003. Acclimation capacity underliessusceptibility to climate change. Science 301: 65.

Tallis H, Polasky S. 2009. Mapping and valuing ecosystemservices as an approach for conservation and natural-resourcemanagement. Annals of the New York Academy of Sciences1162: 265–283.

Tallis H, Ferdana Z, Gray E. 2008. Linking terrestrialand marine conservation planning and threats analysis.Conservation Biology 22: 120–130.

Ulloa R, Torre J, Bourillón L, Gondor A, Alcantar N. 2006.Planeación ecorregional para la conservación marina: Golfode California y costa occidental de Baja California Sur.Informe final a The Nature Conservancy. Comunidad yBiodiversidad: A.C., Guaymas, Mexico.

UNEP-WCMC. 2010. Coral Reefs of the World. WorldConservation Monitoring Centre, United NationsEnvironmental Program (UNEP-WCMC): Cambridge, UK.

Velarde E, Ezcurra E, Cisneros-Mata MA, LavÍn MF. 2004.Seabird ecology, El Niño anomalies, and prediction ofsardine fisheries in the Gulf of California. EcologicalApplications 14: 607–615.

Visser H, de Nijs T. 2006. The Map Comparison Kit.Environmental Modelling and Software 21: 346–358.

Visser H. (Ed.), 2004. The MAP COMPARISON KIT:methods, software and applications. RIVM report550002005. Available from http://www.rivm.nl/milieu/modellen.

Webb TJ, Vanden Berghe E, O’Dor R. 2010. Biodiversity’s bigwet secret: the global distribution of marine biologicalrecords reveals chronic under-exploration of the deeppelagic ocean. PloS One 5: e10223.

Weeks R, Russ GR, Alcala AC, White AT. 2010a.Effectiveness of marine protected areas in the Philippinesfor biodiversity conservation. Conservation Biology 24:531–540.

Weeks R, Russ GR, Bucol AA, Alcala AC. 2010b. Shortcutsfor marine conservation planning: The effectiveness ofsocioeconomic data surrogates. Biological Conservation 143:1236–1244.

Wheeler M, Chambers FMJ, Sims-Castley R, Cowling RM,Schoeman D. 2008. From beans to breams: howparticipatory workshops can contribute to marineconservation planning. African Journal of Marine Science30: 475–487.

Wilson KA, Westphal MI, Possingham HP, Elith J. 2005.Sensitivity of conservation planning to different approachesto using predicted species distribution data. BiologicalConservation 122: 99–112.

Wilson KA, McBride MF, Bode M, Possingham HP.2006. Prioritizing global conservation efforts. Nature440: 337–340.

Wilson KA, Underwood EC, Morrison SA, Klausmeyer KR,Murdoch WW, Reyers B, Wardell-Johnson G, Marquet PA,Rundel PW, McBride MF, et al. 2007. Conservingbiodiversity efficiently: what to do, where, and when. PLoSBiology 5: e223.

Wilson KA, Carwardine J, Possingham HP. 2009. Settingconservation priorities. In Year in Ecology and ConservationBiology 2009. Blackwell Publishing: Oxford; 237–264.

Wood LJ, Fish L, Laughren J, Pauly D. 2008. Assessingprogress towards global marine protection targets: shortfallsin information and action. Oryx 42: 340–351.

Worm B, Barbier EB, Beaumont N, Duffy JE, Folke C,Halpern BS, Jackson JBC, Lotze HK, Micheli F, PalumbiSR, et al. 2006. Impacts of biodiversity loss on oceanecosystem services. Science 314: 787–790.

Worm B, Hilborn R, Baum JK, Branch TA, Collie JS,Costello C, Fogarty MJ, Fulton EA, Hutchings JA,Jennings S, et al. 2009. Rebuilding global fisheries. Science325: 578–585.





Marine conservation planning in practice: lessons learned ...€¦ · Marine conservation planning...

Documents

Transcript of Marine conservation planning in practice: lessons learned ...€¦ · Marine conservation planning...