, 20130423, published 17 April 2013280 2013 Proc. R. Soc. B Saatchi, Catherine H. Graham, Robert K. Wayne and Thomas B. SmithSchlunegger, Janice Chan, Ophelia Wang, Christopher J. Schneider, John P. Pollinger, SassanKieswetter, Pablo Jarrín-V, Susan E. Cameron Devitt, Eliza Mason, Rena M. Schweizer, Jasmin Trevon L. Fuller, Henri A. Thomassen, Manuel Peralvo, Wolfgang Buermann, Borja Milá, Charles M. species predicts ranges of threatened onesIntraspecific morphological and genetic variation of common

Supplementary datatml http://rspb.royalsocietypublishing.org/content/suppl/2013/04/16/rspb.2013.0423.DC1.h

"Data Supplement"

References http://rspb.royalsocietypublishing.org/content/280/1760/20130423.full.html#ref-list-1 This article cites 48 articles, 8 of which can be accessed free

Subject collections

(1436 articles)evolution (199 articles)environmental science

(1315 articles)ecology Articles on similar topics can be found in the following collections

Email alerting service hereright-hand corner of the article or click Receive free email alerts when new articles cite this article - sign up in the box at the top

http://rspb.royalsocietypublishing.org/subscriptions go to: Proc. R. Soc. BTo subscribe to

on April 30, 2013rspb.royalsocietypublishing.orgDownloaded from

rspb.royalsocietypublishing.org

ResearchCite this article: Fuller TL, Thomassen HA,Peralvo M, Buermann W, Mila B, KieswetterCM, Jarrın-V P, Devitt SEC, Mason E, SchweizerRM, Schlunegger J, Chan J, Wang O, SchneiderCJ, Pollinger JP, Saatchi S, Graham CH, WayneRK, Smith TB. 2013 Intraspecific morphologicaland genetic variation of common speciespredicts ranges of threatened ones. Proc R SocB 280: 20130423.http://dx.doi.org/10.1098/rspb.2013.0423

Received: 18 February 2013Accepted: 22 March 2013

Subject Areas:ecology, environmental science, evolution

Keywords:biodiversity, conservation planning, indicatorspecies, reserve selection, surrogacy

Author for correspondence:Trevon L. Fullere-mail: fullertl@ucla.edu

†Present address: Department of ComparativeZoology, Institute of Evolution and Ecology,University of Tubingen, Auf-der-Morgenstelle28, 72076 Tubingen, Germany.

Electronic supplementary material is availableat http://dx.doi.org/10.1098/rspb.2013.0423 orvia http://rspb.royalsocietypublishing.org.

Intraspecific morphological and geneticvariation of common species predictsranges of threatened onesTrevon L. Fuller1, Henri A. Thomassen1,†, Manuel Peralvo2,Wolfgang Buermann1,3, Borja Mila1,4, Charles M. Kieswetter5, Pablo Jarrın-V6,Susan E. Cameron Devitt7, Eliza Mason8,9, Rena M. Schweizer8,Jasmin Schlunegger8, Janice Chan8, Ophelia Wang10,11, ChristopherJ. Schneider5, John P. Pollinger1,8, Sassan Saatchi1,12, Catherine H. Graham13,Robert K. Wayne8 and Thomas B. Smith1,8

1Center for Tropical Research, Institute of the Environment and Sustainability, University of California,Los Angeles, La Kretz Hall, Suite 300, 619 Charles E. Young Dr. East, Los Angeles, CA 90095, USA2Iniciativa de Estudios Ambientales Andinos, Consorcio para el Desarrollo Sostenible de la Ecorregion Andina,Diego de Brieda E17-169 y Clemente Celi (Sector Bellavista), PO Box 17-21-1977, Quito, Ecuador3School of Earth and Environment, University of Leeds, Leeds LS2 9JT, West Yorkshire, UK4Museo Nacional de Ciencias Naturales, Consejo Superior de Investigaciones Cientıficas, Jose Gutierrez Abascal 2,Madrid 28006, Spain5Department of Biology, Boston University, Boston, MA 02215, USA6Yasuni Research Station, Escuela de Ciencias Biologicas, Pontificia Universidad Catolica del Ecuador, Quito, Ecuador7Department of Wildlife Ecology and Conservation, University of Florida, PO Box 110430, Gainesville,FL 32611-0430, USA8Department of Ecology and Evolutionary Biology, University of California, Los Angeles, 621 Charles E. YoungDr. East, Los Angeles, CA 90095, USA9Department of Microbiology and Immunology, School of Medicine, University of North Carolina,804 Mary Ellen Jones Building, Chapel Hill, NC 27599, USA10Department of Geography and the Environment, University of Texas at Austin, Mailcode A3100, Austin, TX 78712, USA11Center for Sustainable Environments, Northern Arizona University, PO Box 5767, Flagstaff, AZ 86011, USA12Jet Propulsion Laboratory, California Institute of Technology, 4800 Oak Drive, Pasadena, CA 91109, USA13Department of Ecology and Evolution, Stony Brook University, 650 Life Sciences Building, New York,NY 11794, USA

Predicting where threatened species occur is useful for making informed conser-vation decisions. However, because they are usually rare, surveying threatenedspecies is often expensive and time intensive. Here, we show how regionswhere common species exhibit high genetic and morphological divergenceamong populations can be used to predict the occurrence of species of conserva-tion concern. Intraspecific variation of common species of birds, bats and frogsfrom Ecuador were found to be a significantly better predictor for the occurrenceof threatened species than suites of environmental variables or the occurrence ofamphibians and birds. Fully 93 per cent of the threatened species analysed hadtheir range adequately represented by the geographical distribution of the mor-phological and genetic variation found in seven common species. Both highernumbers of threatened species and greater genetic and morphological variationof common species occurred along elevation gradients. Higher levels of intraspe-cific divergence may be the result of disruptive selection and/or introgressionalong gradients. We suggest that collecting data on genetic and morphologicalvariation in common species can be a cost effective tool for conservation plan-ning, and that future biodiversity inventories include surveying genetic andmorphological data of common species whenever feasible.

1. IntroductionTo conserve biodiversity, 168 signatory nations at the 2010 conference of theConvention on Biological Diversity held in Nagoya, Japan agreed to establish

& 2013 The Author(s) Published by the Royal Society. All rights reserved.

on April 30, 2013rspb.royalsocietypublishing.orgDownloaded from

protected areas comprising 17 per cent of the terrestrial area ofthe planet by 2020 [1]. Because protecting every species isinfeasible, conservation planners typically focus on threatenedspecies. However, identifying the most critical areas for pro-tecting threatened species requires accurate distributionaldata for these species, most of which are rare and difficult tosurvey. This challenge has prompted conservation plannersto search for surrogates for species of conservation concern[2]. To be effective, surrogate distributions must be easilymeasured and have a high probability of encompassing areaswhere threatened species occur [3].

Genetic and morphological variation of common speciescan be easily measured and are increasingly recognized fortheir potential utility in designing protected areas [4]. The abil-ity to develop and use molecular markers to assess geneticvariation in non-model organisms is increasingly possibleowing to advancements in methods and decreases in costs[5]. In addition, measures of intraspecific morphological vari-ation for many species can be collected easily in the field orobtained from museum specimens. If genetic and morphologi-cal traits of common species are effective for identifying andprotecting areas where species of conservation concern occur,these attributes would make them excellent tools to use as indi-cators for conservation. For example, in the Atlantic Forest ofBrazil, the genetic divergence of common frogs is an effectivepredictor of endemism in other taxa [6]. Here, we investigatethe ability to use genetic and morphological divergenceamong populations of seven common species to predict 29threatened species from western Ecuador. The study region isa biodiversity hotspot, represents a transition zone betweenhumid and dry tropical forests [7] and contains high levels ofendemic plants and animals, including 650 bird and 6300 vas-cular plant species, 20 per cent of which are endemic [8]. Owingto high rates of deforestation and habitat degradation, the

number of threatened vertebrates is among the highest in theworld [9].

2. Material and methodsWe sampled genetic and morphological variation in sevencommon species in western Ecuador, developed genetic markersand constructed spatial models of genetic and morphologicaldivergence among populations of each species. We then designedreserves based on this genetic and morphological variation andmeasured the percentage of threatened species in the reserves.We also compared reserves designed based on genetic and mor-phological variation in common species to reserves designedbased on only environmental variables or occurrences of birdsand amphibians (see the electronic supplementary material,figure S1 summarizes the analysis). The analysis was restrictedto western Ecuador because this region of dry forest is particula-rly threatened by anthropogenic impacts (see the electronicsupplementary material, figure S2 shows our study region).

(a) Measurement of genetic and morphologicaltraits in common species and generalizeddissimilarity modelling

Three birds, three bats and one frog species were analysed becausethey were abundant and easily sampled and represent a range ofdifferent vagilities and life histories ([4]; table 1). The seven speciesoccupy distinct niches representing a diverse range of altitudes,vegetation communities and feeding ecologies, including insectiv-ory, frugivory and nectarivory. Genetic data were available forthe three birds, one bat and the frog (table 1). We developed anon-ymous loci (amplified fragment length polymorphisms (AFLPs),microsatellites and nuclear sequence) and targeted loci (recom-bination activating gene 1 (RAG1)) following published methods[4,10] to assess genetic variation among populations of common

Table 1. Measures of intraspecific morphological and genetic variation were developed for seven common species in western Ecuador.

speciesgenetic markers (loci, individuals,sites) morphological traits (n, sites)

birds wedge-billed woodcreeper

(Glyphorynchus spirurus)

amplified fragment length

polymorphisms (AFLPs; 136, 178, 15)

wing length, tarsus length, tail length, bill length,

bill width and bill depth (195, 15)

masked flowerpiercer

(Diglossa cyanea)

microsatellites (10, 102, 12) wing length, tarsus length, tail length, bill length,

bill width and bill depth (90, 10)

streak-necked flycatcher

(Mionectes striaticollis)

microsatellites (10, 106, 9) wing length, tarsus length, tail length, bill length,

bill width and bill depth (122, 8)

bats silky short-tailed bat

(Carollia brevicauda)

none centroid size of the skull, angle of curvature of the

zygomatic arch and forearm length (167, 43)

chestnut short-tailed bat

(Carollia castanea)

none centroid size of the skull, angle of curvature of the

zygomatic arch and forearm length (86, 25)

seba’s short-tailed bat

(Carollia perspicillata)

AFLPs (311, 83, 9) centroid size of the skull, angle of curvature of the

zygomatic arch and forearm length (160, 44)

frog zurucuchu robber frog

(Hylodes buergeri)

two anonymous loci and an 840 base-

pair region of recombination

activation gene 1 (RAG1; 3,76 – 127

per locus, 9)

snout-vent length, gape width, lengths of the

metacarpal phalanges, length of the radio-ulna,

lengths of the metatarsal phalanges, length of

the tarsus, tibio – fibula length, femur length

and lower jaw length (224, 16)

rspb.royalsocietypublishing.orgProcR

SocB280:20130423

2

on April 30, 2013rspb.royalsocietypublishing.orgDownloaded from

species. We used AFLP loci for the wedge-billed woodcreeper(Glyphorynchus spirurus) [10,11] and seba’s short-tailed bat (Carolliaperspicillata; [4]), whereas for the masked flowerpiercer (Diglossacyanea) and streak-necked flycatcher (Mionectes striaticollis), wedeveloped microsatellites [4]. Finally, for the zurucuchu robberfrog (Hylodes buergeri), we sequenced two anonymous loci and an840 base-pair region of the RAG1 [4]. We then calculated the geneticdistance between sampled populations. We used Fst as the distancemeasure for thewedge-billed woodcreeper, streak-necked flycatcherand the masked flowerpiercer because this measure requires genoty-pic data, and we had microsatellites and AFLPs for these species.We used Fst for the zurucuchu robber frog because this measureuses sequence data, which we had for the frog.

To quantify morphological variation among populations, fol-lowing published methods [4], we measured morphologicaltraits on each sampled individual (table 1): for birds, wing, tail,tarsus and bill length, bill width and bill depth; for bats, skullsize, zygomatic arch and forearm length; and for the frog,snout-vent length, gape width and the length of the metacarpalphalange, radio-ulna, metatarsal phalange, tarsus, tibio–fibulaand lower jaw. The frog morphological traits were corrected forsize as described elsewhere, but regression analysis indicatedthat body size corrections were unnecessary for the birds andbats [4]. The morphological distance between pairs of sampledpopulations was calculated as the Euclidean distance betweenthe mean phenotype in each population. The total number ofgenetic and morphological traits for the seven common specieswas 41 (36 morphological and five genetic traits; table 1).

We used generalized dissimilarity modelling (GDM), amatrix regression technique [12], to identify environmentalvariables that were significantly correlated with genetic andmorphological distance. We used the implementation of GDMfor the ArcView, and SPLUS software packages to predict valueof each of the 36 genetic and morphological traits across westernEcuador. For each trait, we assessed the importance of 14 inde-pendent variables that represent climate, vegetation phenologyderived from satellite data and elevation [4,10]. The contributionsof independent variables to explaining each trait were testedby permutation (models for 36 traits performed better thanrandom [4,10]).

(b) Species of conservation concernOur dataset on species of conservation concern comprised a totalof 29 plants and vertebrates. Peralvo et al. [13] detail the speciesselection criteria, which we will summarize here. For vertebrates,we started with 315 species classified as conservation targets byTNC-Ecuador’s technical staff because the species are underthreat and restricted to the Equatorial Pacific ecoregion of wes-tern Ecuador and northwestern Peru. The list was filtered toretain species that were in the 2012 IUCN Red List and had atleast 10 records in our vertebrate occurrence database [13]. Therefined list comprised five birds and six mammals (table 2). Werequired 10 or more occurrences because we constructed geo-graphical distribution models for each species (see below), andthe modelling technique that we used needed 10 observationsto make accurate predictions [14]. For plants, we began withTNC-Ecuador’s list of vascular plants of conservation concern,which comprises 1050 species. Botanists specializing in Ecuador(B. Leon and D. Padilla) identified one diagnostic species thatwas representative of each vegetation community in the Equator-ial Pacific (communities were defined as per [15]). To keep theanalysis tractable, we filtered the list to retain just these diagnos-tic species. Last, we excluded diagnostic species with less than10 records in the Missouri Botanical Garden’s VAST database[13]. This resulted in a list of 18 plants (table 2), all of whichare restricted to the Equatorial Pacific and five of which are inEcuador’s national plant Red List [16]. We did not limit our list

of plants to Red-listed species because the exclusive use of theRed List may not adequately represent plants [17], especially ina hotspot of floral endemism such as Ecuador, and may notcapture the threat of land clearing in dry forest vegetation com-munities west of the Andes. Instead, we selected plants thatwere deemed to be of conservation concern based on expertknowledge and were also representative of vegetation commu-nities that have a high risk of being deforested. Hereafter, theterm ‘threatened species’ refers to the 18 plants, five birds andsix mammals selected using the aforementioned criteria.

Next, we calculated the number of threatened species occur-ring in reserves designed based on genetic and morphologicalvariation of common species. We aggregated the occurrencedata to the 100 km2 resolution (see the electronic supplementarymaterial, table S1) because the correlation between surrogatesand threatened species is typically highest at this resolution[18]. The resulting dataset consisted of 639 sites that abut eachother and occupy a continuous swath of land in western Ecuador(see the electronic supplementary material, figure S2). The samespatial scale was used for the analyses based on environmentalvariables, threatened species and genetic and morphologicaltraits. In addition to analysing raw data on the known occur-rences of threatened species, we also examined models of thegeographical distribution of threatened species, which predictspecies’ niches using occurrences and environmental variables.The models used here were constructed by Peralvo et al. [13]with the GARP software package using environmental data com-prising temperature and precipitation variables, elevation andvegetation type. We calculated the number of threatened speciesthat had 10 per cent of their predicted distributions included inreserves designed based on genetic and morphological variationof common species.

(c) Modelling the geographical range of common birdsand amphibians

Since the occurrence of common species are effective surrogates insome geographical regions [19], we also calculated the number ofthreatened species in reserves designed based on occurrencesof common birds and amphibians. This aspect of the analysisdid not incorporate genetic and morphological variation withincommon species. We computed occurrences of common birdsand amphibians in western Ecuador from maps available fromthe NatureServe INFONATURA database (www.natureserve.org/infonatura; [20–22]). The dataset comprised 1397 amphibianand bird species (Columbiformes and suboscine passerines)mapped at a 1 km resolution, which we aggregated to 100 km2.For each 100 km2 site, the dataset indicated whether each of the1397 species was present or absent in the site.

(d) Modelling environmental variablesCertain climatic, edaphic and topographic features can be effectivesurrogates for predicting the occurrence of some threatenedspecies [23]. We therefore compared the predictive performanceof genetic and morphological divergence with that of environmen-tal variables. Environmental variables (aspect, climate, elevationand slope) available at the global scale were obtained fromonline global databases and clipped to our study region (see theelectronic supplementary material, table S3). With the exceptionof soil type, the environmental variables were continuous. Wediscretized each continuous variable into several categorical vari-ables. In general, the effectiveness of an environmental variableas a surrogate for threatened species is expected to increase withthe number of categories. This is because a large number of cat-egories tend to capture fine-scale environmental variation in thesites occupied by the species. However, the computational diffi-culty of the area prioritization problem also increases with the

rspb.royalsocietypublishing.orgProcR

SocB280:20130423

3

on April 30, 2013rspb.royalsocietypublishing.orgDownloaded from

Tabl

e2.

Ecua

doria

nsp

ecies

ofco

nser

vatio

nco

ncer

ninc

luded

inth

ean

alysis

(D,d

ecrea

sing;

S,sta

ble;U

,unk

nown

;MOB

OT,M

issou

riBo

tanic

alGa

rden

).

fam

ilysc

ient

ific

nam

eco

mm

onna

me

occu

rrenc

esIU

CNRe

dLis

tIU

CNpo

pula

tion

trend

Ecua

dor’s

plan

tRe

dLis

tM

OBOT

plan

tsof

Ecua

dor

data

base

birds

Cracid

aeOr

talis

eryth

ropt

eraru

fous

-hea

ded

chac

halac

a

16X

D

Fring

illida

eCa

rdue

lissie

mira

dzkii

saffr

onsis

kin15

XD

Picida

eCa

mpep

hilus

gaya

quile

nsis

guay

aquil

wood

peck

er24

XD

Psitt

acida

eAr

ating

aery

thro

geny

sred

-mas

ked

parak

eet

27X

D

Pionu

sch

alcop

terus

bron

ze-w

inged

parro

t15

XU

mam

mals

Canid

aePs

euda

lopex

sech

urae

sech

uran

zorro

21X

U

Felid

aeLe

opar

dus

pard

alis

ocelo

t45

XD

Puma

conc

olor

pum

a28

XD

Mur

idae

Oryz

omys

xant

heolu

sye

llowi

shric

erat

35X

S

Phyll

osto

mida

eAr

tibeu

sfra

tercu

lusfra

tern

alfru

it-ea

ting

bat

40X

D

Sciur

usstr

amine

usgu

ayaq

uilsq

uirrel

51X

U

vascu

lar

plant

s

Faba

ceae

Albiz

iamu

ltiflo

rano

ne28

UX

Caes

alpini

agla

brat

ano

ne25

X

Cent

rolob

iumoc

hrox

ylum

amar

illo16

UX

X

Geoff

roea

spino

saalm

endo

n24

UX

Leuc

aena

trich

odes

capr

a45

UX

Mach

aeriu

mmi

llei

chich

e27

UX

Pithe

cello

bium

exce

lsum

chaq

uiro

28U

X

Swar

tzia

haug

htii

none

38X

DX

X

Anac

ardia

ceae

Maur

iahe

terop

hylla

fenog

reco

23U

X

Arac

eae

Anth

urium

dolic

hosta

chyu

m

none

42U

XX

Bom

baca

ceae

Eriot

heca

ruizi

ipa

sallo

19U

X

Borag

inace

aeCo

rdia

lutea

yello

wge

iger

26U

X

Capp

arace

aeCa

ppar

ishe

terop

hylla

cape

rshr

ub17

XD

XX

(Cont

inued

.)

rspb.royalsocietypublishing.orgProcR

SocB280:20130423

4

on April 30, 2013rspb.royalsocietypublishing.orgDownloaded from

number of categories. Sarkar et al. [18] analysed the effect ofthe number of categories on the effectiveness of environmental sur-rogates for threatened species on three continents, and identifiedthe optimal number of classes into which climatic and topographicvariables should be partitioned to represent threatened specieseffectively while remaining computationally tractable [18,24].Our analysis uses their partitioning scheme, which divides eachclimatic variable into 4–10 equal interval classes (see the electronicsupplementary material, table S3). Next, we designed reserves toinclude at least 10 per cent of the sites in each class of each environ-mental variable. We then compared the models of the predicteddistributions of the threatened species with these reserves. Wedetermined the percentage of threatened species that had 10 percent of their predicted distributions included in reserves designedbased on the environmental variables.

(e) Assessing the effectiveness of surrogates forthreatened species when reserves are prioritizedwithout areal constraints

We used a complementarity-based algorithm in the RESNET v. 1.2software package [25] to select sites containing each surrogate(genetic and morphological traits of common species, occurrencesof common amphibians and birds and environmental variables).Complementarity chooses sites iteratively such that the siteselected in each iteration contains the largest number of surrogatesnot included at the targeted levels in sites chosen in previous iter-ations ([3]; The targeted level is the percentage of the surrogates’occurrences that we want to include in reserves.). Since reserveselection algorithms require discrete data [26], we divided theGDM predictions about genetic and morphological divergenceinto 50 distinct categories per trait (the use of a slightly differentnumber of classes yields similar results). Fifty were selected sothat the reserve selection algorithm would remain computationallytractable. We required the selected sites to satisfy a conservationtarget of 10 per cent for each surrogate [26], meaning that thereserves included at least 10 per cent of the occurrences of eachsurrogate. We selected 10 per cent because this target has seenwidespread use in conservation planning exercises [18,24], butthe use of higher targets gave similar results (see the electronic sup-plementary material, figure S4). However, in this stage of theanalysis, there was no constraint on the amount of land in westernEcuador that could be included in the selected sites.

After designing reserves, we compared the number of threa-tened species included in sites selected based on genetic andmorphological variation of common species, environmental vari-ables and occurrences of amphibian and bird species using a one-tailed, unequal variance t-test. If we had used a two-tailed test,the rejection of the null hypothesis would have meant that thenumber of threatened species is different in areas selected basedon genetic and morphological traits, and areas selected based onenvironmental variables or occurrences of common species withoutindicating the direction of the difference. Since we hypothesizedthat there would be more threatened species at sites selectedbased on genetic and morphological variation of common species,we used a one-tailed test. The rejection of the null hypothesis in aone-tailed test implies that the number of threatened species inareas prioritized based on genetic and morphological variation isgreater than the number of threatened species in sites selectedbased on environmental variables or occurrences of commonbirds and amphibians. The test assumed unequal variance becausethe number of sites selected based on environmental variables,species occurrences and genetic and morphological variation ofcommon species were different. Owing to the unequal samplesizes, the unequal variance test was the most appropriate becauseit adjusts the test statistic and degrees of freedom based on thenumber of samples in the two groups [16].Ta

ble

2.(Co

ntinu

ed.)

fam

ilysc

ient

ific

nam

eco

mm

onna

me

occu

rrenc

esIU

CNRe

dLis

tIU

CNpo

pula

tion

trend

Ecua

dor’s

plan

tRe

dLis

tM

OBOT

plan

tsof

Ecua

dor

data

base

Meli

acea

eCa

rapa

guian

ensis

andir

oba

36U

X

Mor

acea

eSo

roce

asa

rcoca

rpa

none

32X

DX

X

Polyg

onac

eae

Tripla

riscu

ming

iana

long

john

21U

X

Solan

acea

eSo

lanum

conf

ertise

riatu

mno

ne26

UX

Sterc

uliac

eae

Guaz

uma

ulmifo

liaba

stard

ceda

r22

UX

rspb.royalsocietypublishing.orgProcR

SocB280:20130423

5

on April 30, 2013rspb.royalsocietypublishing.orgDownloaded from

( f ) Assessing the effectiveness of surrogates forthreatened species when reserves are prioritizedsubject to areal constraints

Since land cost is an important determinant of which sites maybe appropriate for inclusion in a conservation plan [27,28], andanalysing cost makes our planning exercise more realistic, weincorporated a budgetary ceiling on the total area of reservesselected based on genetic and morphological traits of commonspecies, occurrences of common birds and amphibians andenvironmental variables (see the electronic supplementary mate-rial analyses other measures of land cost, including proximity toinfrastructure). For each surrogate, we constructed a curve calleda surrogacy curve using the SURROGACY software package [29,30].A surrogacy curve is a generalization of a species accumulationcurve that lists how the number of threatened species increasesas more land is selected based on a surrogate [23]. Each curvedescribes the properties of 17 different reserve networks selectedbased on the surrogate. We first selected reserves based on a sur-rogate subject to the constraint the reserves could only occupy 1per cent of the land in western Ecuador and calculated thepercentage of threatened and endangered species in the selectedsites. Next, we selected sites based on a surrogate subject to the con-straint that the sites could only occupy 2 per cent of westernEcuador and calculated the percentage of threatened and endan-gered species that were included in the selected sites. Theprocedure was repeated at 1 per cent increments up to 17 per centof the land in western Ecuador. The upper target of 17 per centwas used because the parties to the Convention on BiologicalDiversity agreed to establish protected areas comprising 17 percent of the terrestrial area of the planet [1]. This target is arbitraryto the extent that one should protect whatever per cent of a species’range is sufficient to ensure its survival. However, owing to the dif-ficulty of estimating this percentage, the 17 per cent target is usedby many countries [31].

To provide a comparison with the three surrogate sets (geneticand morphological traits, occurrences of common species orenvironmental variables), we also randomly selected sites 10 000times. The ‘random’ curve in the surrogacy plot shows the meannumber of threatened and endangered species included in siteschosen randomly at targets ranging from 1 to 17 per cent. In thecurves representing the other surrogates in the surrogacy plot,the x-coordinate of each point is the areal constraint (1–17%) on

the sites selected based on the surrogate. The y-coordinate ofeach point is the percentage of the 29 threatened species that areincluded in sites selected subject to the areal constraint. When acurve representing a surrogate is above the curve for randomsites, the surrogate does better than random. When a curve forone surrogate is above the curve for another surrogate, the firstsurrogate is more effective than the second surrogate.

3. Results(a) The effectiveness of genetic and morphological

traits as surrogates for threatened species whenreserves are prioritized without areal constraints

When we designed reserves without restricting the amount ofland that could be put under a conservation plan, sitesselected based on genetic and morphological divergence ofcommon species contained a higher proportion of threatenedspecies than sites selected at random or regions selected usingenvironmental variables or bird and amphibian occurrencesas a proxy (figure 1). In particular, sites selected based ongenetic and morphological traits contain 2.1 times as manythreatened species as sites selected based on environmentalvariables, and 1.7 times as many as sites based on bird andamphibian occurrences (see the electronic supplementarymaterial, figure S2). The number of threatened and endangeredis significantly greater than in sites selected based on birdand amphibian occurrences (one-tailed, unequal variancet-test, t ¼ 6.495, d.f. ¼ 116.043, p ¼ 1.076 ! 1029) or environ-mental variables (one-tailed, unequal variance t-test, t ¼ 2.102,d.f.¼ 171.355, p ¼ 0.0185). Sites chosen based on genetic andmorphological variation of common species contained steeperelevational gradients than unselected sites in the study region(one-tailed, unequal variance t-test, t ¼ 3.582, d.f.¼ 14.409,p ¼ 0.00144). These steep altitudinal gradients also containeda high number of threatened species. Finally, we found signi-ficant spatial overlap between reserves selected based ongenetic and morphological traits of common species andmodels of the geographical distributions of threatened species

(i)

20

16

12

8

4

0no. t

hrea

tene

d sp

ecie

s in

clud

ed in

the

sele

cted

site

s(m

ean

± 95

% C

I)

(iv)(ii) (iii)

Figure 1. The mean number of threatened and endangered species (+95% CI) in the sites selected based on: (i) genetic and morphological traits of sevencommon species, (ii) environmental variables, (iii) occurrences of amphibians and birds, and (iv) randomly selected sites (!10 000). The average number of threa-tened and endangered species in sites selected based on genetic and morphological traits of common species is significantly greater than the number in sitesselected based on environmental variables, occurrences of amphibians and birds or a null model consisting of sites selected at random.

rspb.royalsocietypublishing.orgProcR

SocB280:20130423

6

on April 30, 2013rspb.royalsocietypublishing.orgDownloaded from

(Cramer–Von Mises test: null hypothesis of significant overlap,p . 0.05; electronic supplementary material, table S2).

(b) The effectiveness of genetic and morphologicaltraits as surrogates for threatened species whenreserves are prioritized subject to areal constraints

When the maximum amount of land that could be put under aconservation plan ranged from 1 to 17 per cent of westernEcuador, intraspecific variation of the seven common speciesremained the most effective surrogate for species of conservationconcern. First, we compared the reserves selected based on gen-etic and morphological data of common species with models ofthe geographical distributions of threatened species. We foundthat genetic and morphological traits of the seven commonspecies were the most effective surrogate for threatened speciesbecause the curve that represents genetic and morphologicaltraits is consistently higher than the curves for random sites,environmental variables or bird and amphibian occurrences(figure 2a). Of the 29 threatened species analysed, 27 have atleast 17 per cent of their ranges represented in regions identifiedbased on intraspecific morphological, and genetic variationexhibited by the seven common species. In every case, regionsselected to represent genetic and morphological divergence in

the common species we surveyed performed better thanrandom, though the benefit of the method is limited as oneapproaches the 17 per cent target (figure 2a). Second, we com-pared reserves selected based on genetic and morphologicaltraits of common species with known occurrences of threatenedspecies. We found that reserves selected based on genetic andmorphological traits include a significantly higher percentageof threatened species than sites selected at random. In the corre-sponding surrogacy plot, the curve that represents genetic andmorphological traits is consistently higher than the curve forrandom sites (figure 2b). In addition, when we incorporatedland cost into the prioritizations, sites selected based on intra-specific genetic and morphological divergence includedsignificantly more threatened species than random sites or sitesprioritized based on environmental parameters or bird andamphibian occurrences (see the electronic supplementarymaterial, figure S3).

4. Discussion(a) Processes that generate divergence and threatened

species along gradientsOur results show that the genetic and morphological vari-ation exhibited by common species effectively predicts the

thre

aten

ed s

peci

es in

pro

tect

ed a

reas

(%)

area selected to be protected (%)area selected to be protected (%)0 2.5 5.0 7.5 10.0 12.5 15.0 2.5 5.0 7.5 10.0 12.5 15.00

20

40

60

80

100

areas priorititized to include genetic andmorphological traits of 7 common species

areas priorititized to includeenvironmental surrogates

areas priorititized to include occurrencesof amphibians and birds (1397 species)

random area prioritizations

(a) geographic distribution models (b) species occurrences

Figure 2. Effectiveness of genetic and morphological surrogates for threatened species. (a) Habitat models. Uses models of the geographical distributions of threatenedspecies. The solid black line with white circles represents the percentage of threatened species that is found in areas designed based on unique intraspecific morphologicand genetic divergence of common species (areas priorititized to include genetic and morphological traits of seven common species). The solid black line with whitetriangles depicts areas prioritized based on occurrences of birds and amphibians (areas priorititized to include environmental surrogates). The solid black line with squaresrepresents environmental surrogates (areas priorititized to include occurrences of amphibians and birds (1397 species)). The dashed grey line shows the percentage ofthreatened species in random areas (random area prioritizations). We carried out 10 000 randomizations for each point, and calculated the mean and 99% CIs (theconfidence intervals were too small to be depicted). Since the black line with white circles is to the left of the other lines, areas designed based on intraspecific variationof common species protect the highest percentage of threatened species. (b) Species occurrences. Uses raw data on the occurrences of threatened species rather thangeographical distribution models. The ‘random’ lines differ in (a) and (b) because they are based on different areas prioritized at random.

rspb.royalsocietypublishing.orgProcR

SocB280:20130423

7

on April 30, 2013rspb.royalsocietypublishing.orgDownloaded from

occurrence of threatened species in western Ecuador. Inparticular, when we prioritized reserves without areal con-straints, we found that elevational gradients supportedhigher intraspecific divergence of genetic and morphologicaltraits in common species and higher numbers of threatenedspecies than other sites, such as lowlands. Genetic and mor-phological traits of common species may predict the rangesthreatened species well simply because the two overlapspatially along gradients of elevation. Why should this bethe case? Below we examine the processes acting alonggradients that may help explain patterns.

Elevational gradients typically have substantial ecologicalturnover because some ecological communities are restrictedto a small altitudinal range [4,10,32]. The high number ofcommunities along an elevational gradient could promotediversifying selection as populations adapt to differing abio-tic and biotic conditions. This could generate high levels ofintraspecific genetic and morphological variation withincommon species. For example, since temperature and precipi-tation generally decrease with elevation, selection could leadto both phenotypes adapted to dry, cold environments at thehigher ends of gradients and hot, wet environments towardsthe lowlands [33,34]. The outcome would be high levels of vari-ation in morphological traits related to thermal and watertolerance across the gradient. Such patterns of selectionacross ecological gradients may ultimately lead to reproductiveisolation and speciation [35–38].

Elevational gradients may also be hotbeds of genetic diver-gence within common species because they undergo strongcyclical fluctuations in the environment. Quantitative geneticmodels predict that when the environment varies in a cyclicalmanner, having high additive genetic variance will signifi-cantly increase a population’s average fitness [39]. Overdecadal time scales, there is significant variation in tempera-ture along elevational gradients in Ecuador that is triggeredby the El Nino–Southern Oscillation, and this variation ismore pronounced along gradients than at lower altitudes[40]. In the light of this, populations that diverge geneticallyas a means of adapting to cyclically varying climate alongelevational gradients may have higher fitness, which couldresult in high intraspecific genetic variation along gradients.

Moreover, introgression, hybridization between species fol-lowed by backcrossing with parentals, might also result ingenetic divergence along elevational gradients because hybridscarry alleles from both parental lineages [41]. Since introgres-sion has been shown to occur along elevational gradientswhen high- and low-altitude species hybridize at intermediatealtitudes [42–44], elevational gradients in western Ecuadormay have high genetic divergence among populations ofcommon species because these gradients are hybrid zones.

Finally, because elevational gradients have high ecologi-cal turnover, species along the gradient may exhibit highlevels of specialization [45]. These specialist species wouldbe confined to relatively small ranges, because only anarrow elevational band is ecologically suitable for them,and consequently they may be more easily threatened sinceextinction risk decreases with range size [46]. Anthropogenicdisturbance would further increase the risk of extinction ofrare, specialized species along gradients. An example ofsuch disturbance is deforestation. Significant clearing of for-ests in western Ecuador began in the 1950s [8]. The maindrivers of deforestation in this region have been the expan-sion of cattle pasture, cropland and timber extraction [47].

We hypothesize that elevational gradients, because they areharder to reach than lowlands, contain some of the lastforest remnants west of the Andes but that deforestation isexpanding to steeper areas. When we analysed a satellite-derivedmap of western Ecuador’s land cover [48], we found that theproportion of land recently converted to cropland and artificialareas is greater along steep elevational gradients than in flatareas (one-tailed t¼ 5.6109, d.f.¼ 19.953, p¼ 1.044! 1025).This could be due to the fact that flat areas were largely cleareddecades ago, or owing to the expansion of infrastructure insteep areas such road development that results in deforestationof previously inaccessible sites. For example, in the AtlanticForest of Brazil, a disproportional amount of road constructionin steep areas from the 1960s to the 1980s contributed signifi-cantly to forest fragmentation [49]. If roads have expanded insteep areas in Ecuador in a similar manner, threatened spe-cies may concentrate along gradients, because gradients areundergoing high rates of road-associated land clearing.

Although testing this hypothesis remains an importantarea for future research, findings from elevational gradientsin other regions may provide insight into whether deforesta-tion could have led to a high number of threatened speciesalong elevational gradients in western Ecuador. For example,deforestation along elevational gradients in the ColumbianAndes during the first half of the twentieth century resultedin a significant decrease in insect diversity [50]. Furthermore,in the Sierra Nevada mountain range of California during thepast century, animal populations occurring along altitudinalgradients have experienced range contraction owing toanthropogenic effects such as land clearing and climatechange [51,52]. A similar process may have occurred insteep areas in western Ecuador, leading to a high numberof threatened species along altitudinal gradients.

(b) Implications for biodiversity monitoringSurveys of biodiversity in remote areas often have the goalof inventorying all species in a region. However, suchapproaches are costly because they require that experts bedispatched to survey multiple taxonomical groups in thefield and funding for subsequent analyses. Our analysesshow that genetic and morphological traits of commonspecies may be used as effective surrogates for the occurrenceof species of conservation concern. Given that they are effec-tive surrogates, future rapid assessments of biodiversitymight be improved or supplemented by also samplingamphibian, avian and chiropteran genetic and morphologicaldivergence of selected common species. This has the potentialto significantly decrease the time and cost of biodiversityassessments, making it possible to carry out a greaternumber of assessments each year. Lastly, surveying geneticand morphological variation could lead to direct tests aboutthe processes resulting in biodiversity and how it mayrespond to climate change [53,54].

Since our conclusion that intraspecific variation of birds,bats and a frog is an effective surrogate for threatened speciesis based on analysis of only seven species, this raises the ques-tion of how generalizable our conclusions are. A search of theliterature reveals that intraspecific diversity has also beenshown to be correlated with diversity and endemism at thespecies level in a number of other taxa and geographicalregions, including birds in the West Indies [55], butterflies inIndonesia [56], stream fishes in the midwestern US [57] and

rspb.royalsocietypublishing.orgProcR

SocB280:20130423

8

on April 30, 2013rspb.royalsocietypublishing.orgDownloaded from

beetles in the Aegean archipelago [58]. The association betweenspecies diversity and intraspecific variation in diverse taxasuggests that our results may be useful for other taxa. However,future work should test whether the association betweengenetic and morphological traits of common species and threa-tened species is limited to areas where there are steep altitudinalgradients. Given the relative ease associated with assayinggenetic variation in non-model organisms, we believe theapproach presented here represents a potentially importantnew tool for the conservation of biodiversity. Whether thistool can be implemented in real-world conservation planningwill depend largely on economic cost: if collecting genetic

samples from common species and analysing genetic and mor-phological data is less expensive than surveys to detectthreatened species, then our method will have practical utilityfor governmental agencies and conservation NGOs. Compar-ing the cost-efficiency of these two methods remains animportant area for future research.

We thank R. Calsbeek, R. Harrigan, B. Larison, R. E. Ricklefs, threereviewers and Associate Editor Daniel Rabosky for comments thatimproved the manuscript. This work was supported by NSF grantno. IRCEB9977072 to T.B.S, R.K.W. and C.J.S.; and NASA grant nosIDS/03–0169–0347 to T.B.S., R.K.W., and C.J.S.; and NNG05GB37Gto C.H.G.

References

1. Convention on Biological Diversity. 2010 Strategicplan for biodiversity, 2011 – 2020. Montreal,Canada: Secretariat of the CBD.

2. Caro TM. 2010 Conservation by proxy. Washington,DC: Island Press.

3. Sarkar S et al. 2006 Biodiversity conservationplanning tools: present status and challenges forthe future. Ann. Rev. Environ. Res. 31, 123 – 159.(doi:10.1146/annurev.energy.31.042606.085844)

4. Thomassen HA et al. 2011 Mapping evolutionaryprocess: a multi-taxa approach to conservationprioritization. Evol. Appl. 4, 397 – 413. (doi:10.1111/j.1752-4571.2010.00172.x)

5. Hayden S, Bekaert M, Crider TA, Mariani S, MurphyWJ, Teeling EC. 2010 Ecological adaptationdetermines functional mammalian olfactorysubgenomes. Genome Res. 20, 1 – 9. (doi:10.1101/gr.099416.109)

6. Carnaval AC, Hickerson MJ, Haddad CFB, RodriguesMT, Moritz C. 2009 Stability predicts geneticdiversity in the Brazilian Atlantic Forest hotspot.Science 323, 785 – 789. (doi:10.1126/science.1166955)

7. Myers N, Mittermeier RA, Mittermeier CG, daFonseca GAB, Kent J. 2000 Biodiversity hotspots forconservation priorities. Nature 403, 853 – 858.(doi:10.1038/35002501)

8. Dodson CH, Gentry AH. 1991 Biological extinction inwestern Ecuador. Ann. Mo. Botanical Garden 78,273 – 295. (doi:10.2307/2399563)

9. Hoffmann M et al. 2010 The impact of conservationon the status of the world’s vertebrates. Science330, 1503 – 1509. (doi:10.1126/science.1194442)

10. Thomassen HA et al. 2010 Modelingenvironmentally associated morphological andgenetic variation in a rainforest bird, and itsapplication to conservation prioritization. Evol. Appl.3, 1 – 16. (doi:10.1111/j.1752-4571.2009.00093.x)

11. Mila B, Wayne RK, Fitze P, Smith TB. 2009Divergence with gene flow and fine-scalephylogeographical structure in the wedge-billedwoodcreeper, Glyphorynchus spirurus, a Neotropicalrainforest bird. Mol. Ecol. 18, 2979 – 2995. (doi:10.1111/j.1365-294X.2009.04251.x)

12. Ferrier S, Manion G, Elith J, Richardson K. 2007Using generalized dissimilarity modelling to analyse

and predict patterns of beta diversity in regionalbiodiversity assessment. Divers. Distrib. 13,252 – 264. (doi:10.1111/j.1472-4642.2007.00341.x)

13. Peralvo M, Sierra R, Young KR, Ulloa-Ulloa C. 2007Identification of biodiversity conservation prioritiesusing predictive modeling: an application for theEquatorial Pacific region of South America. Biodivers.Conserv. 16, 2649 – 2675. (doi:10.1007/s10531-006-9077-y)

14. Peterson AT. 2001 Predicting species’ geographicdistributions based on ecological niche modeling.Condor 103, 599 – 605. (doi:10.1650/0010-5422(2001)103[0599:PSGDBO]2.0.CO;2)

15. Ceron C, Palacios W, Valencia R, Sierra R. 1999 Lasformaciones naturales de la costa del Ecuador. InPropuesta preliminar de un sistema de clasificacionde vegetacion para el Ecuador Continental(ed. R Sierra), pp. 55 – 78. Quito, Ecuador: INEFAN/GEF-BIRF and EcoCiencia.

16. Valencia R, Pitman N, Leon-Yanez S, Jorgensen PM.2000 Libro Rojo de las Plantas Endemicas delEcuador. Quito, Ecuador: Pontificia UniversidadCatolica del Ecuador.

17. Butchart SHM, Stattersfield AJ, Baillie J, Bennun LA,Stuart SN, Akcakaya HR, Hilton-Taylor C, Mace GM. 2005Using Red List Indices to measure progress towards the2010 target and beyond. Phil. Trans. R. Soc. B 360,255– 268. (doi:10.1098/rstb.2004.1583)

18. Sarkar S, Justus J, Fuller T, Kelley C, Garson J,Mayfield M. 2005 Effectiveness of environmentalsurrogates for the selection of conservation areanetworks. Conserv. Biol. 19, 815 – 825. (doi:10.1111/j.1523-1739.2005.00236.x)

19. Ricketts TH. 2001 Aligning conservation goals: arepatterns of richness and endemism concordantat regional scales? Anim. Biodivers. Conserv. 24,91 – 99.

20. IUCN CI, NatureServe. 2008 An analysis ofamphibians on the 2008 IUCN Red List. Gland,Switzerland: IUCN.

21. NatureServe. 2004 InfoNatura: birds, mammals, andamphibians of Latin America. Arlington, VA:NatureServe.

22. Ridgely RS, Greenfield PJ. 2001 The birds ofEcuador. Volume II: a field guide. London, UK:Christopher Helm.

23. Faith DP, Walker PA. 1996 Environmental diversity:on the best possible use of surrogate data forassessing the relative biodiversity of sets of areas.Biodivers. Conserv. 5, 399 – 415. (doi:10.1007/BF00056387)

24. Sarkar S, Mayfield M, Cameron S, Fuller T, Garson J.2007 Conservation area networks for the Indianregion: systematic methods and future prospects.Himalayan J. Sci. 4, 27 – 40. (doi:10.3126/hjs.v4i6.980)

25. Sarkar S, Fuller T, Aggarwal A, Moffett A, Kelley C.2009 The ConsNet software platform for systematicconservation planning. In Spatial conservationprioritization: quantitative methods andcomputational tools (eds A Moilanen, KA Wilson,HP Possingham), pp. 235 – 248. New York, NY:Oxford University Press.

26. Margules CR, Sarkar S. 2007 Systematic conservationplanning. Cambridge, UK: Cambridge UniversityPress.

27. Ando A, Camm J, Polasky S, Solow A. 1998 Speciesdistributions, land values, and efficientconservation. Science 279, 2126 – 2128. (doi:10.1126/science.279.5359.2126)

28. Bode M, Wilson KA, Brooks TM, Turner WR,Mittermeier RA, McBride MF, Underwood EC,Possingham HP. 2008 Cost-effective globalconservation spending is robust to taxonomic group.Proc. Natl Acad. Sci. USA 105, 6498 – 6501. (doi:10.1073/pnas.0710705105)

29. Garson J, Sarkar S. 2002 Surrogacy manual ver 1.1.Austin, TX: Biodiversity and Biocultural ConservationLaboratory, Section of Integrative Biology, Universityof Texas at Austin.

30. Garson J, Aggarwal A, Sarkar S. 2002 Birds assurrogates for biodiversity: an analysis of a data setfrom southern Quebec. J. Biosci. 27, 347 – 360.(doi:10.1007/BF02704965)

31. Rodriguez-Rodriguez D, Bomhard B, Butchart SHM,Foster MN. 2011 Progress towards internationaltargets for protected area coverage in mountains:a multi-scale assessment. Biol. Conserv. 144,2978 – 2983. (doi:10.1016/j.biocon.2011.08.023)

32. Hall J, Burgess ND, Lovett J, Mbilinyi B, Gereau RE.2009 Conservation implications of deforestationacross an elevational gradient in the Eastern

rspb.royalsocietypublishing.orgProcR

SocB280:20130423

9

on April 30, 2013rspb.royalsocietypublishing.orgDownloaded from

Arc Mountains, Tanzania. Biol. Conserv. 142,2510 – 2521. (doi:10.1016/j.biocon.2009.05.028)

33. Luoto M. 2000 Modelling of rare plant speciesrichness by landscape variables in an agriculturearea in Finland. Plant Ecol. 149, 157 – 168.(doi:10.1023/A:1026531400356)

34. Lomolino MV. 2001 Elevation gradients of species-density: historical and prospective views. Glob. Ecol.Biogeogr. 10, 3 – 12. (doi:10.1046/j.1466-822x.2001.00229.x)

35. Smith TB, Wayne RK, Girman DJ, Bruford MW. 1997A role for ecotones in generating rainforestbiodiversity. Science 276, 1855 – 1857. (doi:10.1126/science.276.5320.1855)

36. Freedman AH, Thomassen HA, Buermann W, SmithTB. 2010 Genomic signals of diversification alongecological gradients in a tropical lizard. Mol. Ecol.19, 3773 – 3788. (doi:10.1111/j.1365-294X.2010.04684.x)

37. Smith TB, Calsbeek R, Wayne RK, Holder KH,Pires D, Bardeleben C. 2005 Testing alternativemechanisms of evolutionary divergence in anAfrican rain forest passerine bird. J. Evol.Biol. 18, 257 – 268. (doi:10.1111/j.1420-9101.2004.00825.x)

38. Schneider CJ, Smith TB, Larison B, Moritz C. 1999 Atest of alternative models of diversification intropical rainforests: ecological gradients vs.rainforest refugia. Proc. Natl Acad. Sci. USA 96,13 869 – 13 873. (doi:10.1073/pnas.96.24.13869)

39. Lande R, Shannon S. 1996 The role of geneticvariation in adaptation and population persistencein a changing environment. Evolution 50, 434 – 437.(doi:10.2307/2410812)

40. Mora DE, Willems P. 2012 Decadal oscillations inrainfall and air temperature in the Paute RiverBasin – Southern Andes of Ecuador. Theor. Appl.Climatol. 108, 267 – 282. (doi:10.1007/s00704-011-0527-4)

41. Smith PF, Konings A, Kornfield I. 2003 Hybrid originof a cichlid population in Lake Malawi: implicationsfor genetic variation and species diversity. Mol. Ecol.12, 2497 – 2504. (doi:10.1046/j.1365-294X.2003.01905.x)

42. Culumber ZW, Fisher HS, Tobler M, Mateos M,Barber PH, Sorenson MD, Rosenthal GG. 2011Replicated hybrid zones of Xiphophorus swordtailsalong an elevational gradient. Mol. Ecol. 20,342 – 356. (doi:10.1111/j.1365-294X.2010.04949.x)

43. Floate KD. 2004 Extent and patterns ofhybridization among the three species of Populusthat constitute the riparian forest of southernAlberta, Canada. Can. J. Bot. 82, 253 – 264. (doi:10.1139/b03-135)

44. Paige KN, Capman WC. 1993 The effects of host-plant genotype, hybridization, and environment ongall-aphid attack and survival in cottonwood: theimportance of genetic studies and the utility ofRFLPs. Evolution 47, 36 – 45. (doi:10.2307/2410116)

45. Fjeldsa J. 1994 Geographic patterns for relict andyoung species of birds in Africa and South Americaand implications for conservation priorities. Biodivers.Conserv. 3, 207 – 226. (doi:10.1007/BF00055939)

46. Manne LL, Brooks TM, Pimm SL. 1999 Relative riskof extinction of passerine birds on continents andislands. Nature 399, 258 – 261. (doi:10.1038/20436)

47. Sierra R, Stallings J. 1998 The dynamics and socialorganization of tropical deforestation in NorthwestEcuador, 1983 – 1995. Hum. Ecol. 26, 135 – 161.(doi:10.1023/A:1018753018631)

48. Arino O, Bicheron P, Achard F, Latham J, Witt R,Weber J-L. 2008 GlobCover: the most detailedportrait of Earth. ESA Bull. 136, 25 – 31.

49. Freitas SR, Hawbaker TJ, Metzger JP. 2010 Effects ofroads, topography, and land use on forest coverdynamics in the Brazilian Atlantic Forest. For.Ecol. Manage. 259, 410 – 417. (doi:10.1016/j.foreco.2009.10.036)

50. Medina CA, Escobar F, Kattan GH. 2002 Diversity andhabitat use of dung beetles in a restored Andeanlandscape. Biotropica 34, 181 – 187.

51. Rubidge EM, Patton JL, Lim M, Burton AC, BrasharesJS, Moritz C. 2012 Climate-induced rangecontraction drives genetic erosion in an alpinemammal. Nat. Clim. Change 2, 285 – 288. (doi:10.1038/nclimate1415)

52. Tingley MW, Koo MS, Moritz C, Rush AC, BeissingerSR. 2012 The push and pull of climate changecauses heterogeneous shifts in avian elevationalranges. Glob. Change Biol. 18, 3279 – 3290. (doi:10.1111/j.1365-2486.2012.02784.x)

53. Etterson JR, Shaw RG. 2001 Constraint to adaptiveevolution in response to global warming. Science294, 151 – 154. (doi:10.1126/science.1063656)

54. Gienapp P, Teplitsky C, Alho JS, Mills JA, Merila J.2008 Climate change and evolution: disentanglingenvironmental and genetic responses. Mol. Ecol. 17,167 – 178. (doi:10.1111/j.1365-294X.2007.03413.x)

55. Vellend M. 2003 Island biogeography of genesand species. Am. Nat. 162, 358 – 365. (doi:10.1086/377189)

56. Cleary DFR, Fauvelot C, Genner MJ, Menken SBJ,Mooers AO. 2006 Parallel responses of species andgenetic diversity to El Nino Southern Oscillation-induced environmental destruction. Ecol. Lett. 9,301 – 307. (doi:10.1111/j.1461-0248.2005.00876.x)

57. Blum MJ, Bagley MJ, Walters DM, Jackson SA,Daniel FB, Chaloud DJ, Cade BS. 2012 Geneticdiversity and species diversity of stream fishescovary across a land-use gradient. Oecologia 168,83 – 95. (doi:10.1007/s00442-011-2078-x)

58. Papadopoulou A, Anastasiou I, Spagopoulou F,Stalimerou M, Terzopoulou S, Legakis A, Vogler AP.2011 Testing the species-genetic diversitycorrelation in the Aegean Archipelago: toward ahaplotype-based macroecology? Am. Nat. 178,241 – 255. (doi:10.1086/660828)

rspb.royalsocietypublishing.orgProcR

SocB280:20130423

10

on April 30, 2013rspb.royalsocietypublishing.orgDownloaded from