threats to biodiversity - university of florida to review the major types of threats to...

Extinction is the most irreversible and tragic of all environmental calamities. With each plant and animal species that disappears, a precious part of creation is callously erased.

Michael Soulé, 2004

Human impacts are now a pervasive facet of life on Earth. All realms—terrestrial,marine, and freshwater—bear our imprint; our pollution spans the globe, our fish-eries extend throughout the world’s oceans, and our feet tread across almost everysurface on Earth. By many estimates, we use substantial and increasing fractions ofEarth’s primary productivity (Vitousek et al. 1986; Pauly and Christensen 1995; Pos-tel et al. 1996), and our total ecological impact may already extend beyond Earth’s ca-pacity to provide resources and absorb our wastes (Wackernagel and Rees 1996).

As humans became a widespread and numerous species, our agricultural expan-sion forever changed vast landscapes; our hunting and our transport of invasive,commensal species drove numerous aquatic and terrestrial species extinct. Whenhighly organized societies began to settle and grow throughout the globe, the pace oftransformation of terrestrial and aquatic habitats sharply increased, and our use ofnatural resources began to dramatically outstrip natural rates of replacement. Thus,humans have had enormous impacts on the form and diversity of ecosystems. Ulti-mately, we have set in motion the sixth great mass extinction event in the history ofthe Earth—and the only one caused by a living species.

Human population and consumption pressures are the root threat to biodiversity(see Chapter 1; Figure 3.1). Increasing numbers of humans, and most importantly, in-creasing levels of consumption by humans create the conditions that endanger theexistence of many species and ecosystems: habitat degradation and loss, habitat frag-mentation, overexploitation, spread of invasive species, pollution, and global climatechange. Species extinction, endangerment, and ecosystem degradation are not theaims of human societies, but are the unfortunate by-product of human activities. Be-cause our practices are unsustainable, we strongly erode the natural capital that wehave used to flourish, thus endangering our and our descendents’ future.

© 2005 Sinauer Associates, Inc. This material cannot be copied, reproduced,manufactured, or disseminated in any form without express written permis-sion from the publisher.

3Threats to Biodiversity

Martha J. Groom

Major Threats to Biodiversity and Their InteractionIn this chapter, I provide an overview of patterns of ex-tinction and species endangerment, and describe effortsemployed to slow and reverse these trends. The first stepis to review the major types of threats to biodiversity,while laying the groundwork for the more in-depth cov-erage of these topics later in the book.

Habitat degradation includes the spectrum of totalconversion from a usable to an unusable habitat type (or“habitat loss”), severe degradation and pollution thatmakes a habitat more dangerous or difficult for an or-ganism to live in, and fragmentation that can reducepopulation viability. Habitat degradation can be causedby a host of human activities including industry, agri-culture, forestry, aquaculture, fishing, mining, sedimentand groundwater extraction, infrastructure develop-ment, and habitat modification as a result of species in-troductions, changes in native species abundance, orchanges in fire or other natural disturbance regimes. Inaddition, most forms of pollution affect biodiversity viatheir degradation of ecosystems. Chapter 6 contains afull discussion of the impacts of various forms of habitatloss and degradation, while the phenomenon and effectsof habitat fragmentation are detailed in Chapter 7.

Overexploitation, including hunting, collecting, fish-eries and fisheries by-catch, and the impacts of trade inspecies and species’ parts, constitutes a major threat tobiodiversity. Most obviously, a direct impact of overex-ploitation is the global or local extinction of species orpopulations. Less obvious, the decrease in populationsizes with exploitation can lead to a cascade of effects thatmay alter the composition and functionality of entireecosystems (Estes et al. 1989; Redford 1992; Pauly et al.1998). Overexploitation is discussed in detail in Chapter 8.

The spread of invasive species, species that invade orare introduced to an area or habitat where they do notnaturally occur, is also a significant threat to biodiversity.Invasive species can compromise native species throughdirect interactions (e.g., predation, parasitism, disease,competition, or hybridization), and also through indirectpaths (e.g., disruption of mutualisms, changing abun-dances or dynamics of native species, or modifying habi-tat to reduce habitat quality). The process and impacts ofspecies invasions are described in detail in Chapter 9.

Anthropogenic climate change is perhaps the mostominous threat to biodiversity of the present era. Cli-mate change appears to have caused mass extinctionsseen in the geologic record, and because the pace of cli-mate change is predicted to be at least as fast and ex-treme as the most severe shifts in climate in the geologic

64 Chapter 3


Sinauer Associates, Inc.Groom 3ePrinciples of Conservation BiologyGro0301.eps100% of size04.20.05

Human activities

Agriculture Fisheries Industry Urbanization and sprawl International trade

Exponentially increasing humanpopulation and consumption

Habitat loss and degradationPollution (especially nitrogen)Land cover changeHabitat fragmentation

Overexploitation

Climate change

Loss of biological diversityExtinction of species and populationsDegradation of ecosystems Erosion of genetic diversity and evolutionary potentialLoss of ecosystem servicesErosion of support systems for human societies

Introduction ofinvasive species

Figure 3.1 Major forces that threatenbiological diversity. All arise from in-creases in human population and con-sumption levels, often mediatedthrough our activities on the land andsea. Extinction and severe ecosystemdegradation generally result from mul-tiple impacts and from synergistic in-teractions among these threats.

record, the effect on biodiversity is expected to be enor-mous. Coupled with the extensive transformation ofEarth’s ecosystems, widespread overexploitation of pop-ulations, and introductions of species to new areas in theglobe, we can expect the effects of future shifts in climateto usher in an extremely severe mass extinction event.The probable biological impacts of the climate changesunderway today are examined in full in Chapter 10.

As we are becoming more aware of the global impactsof climate change, we also are learning more about theglobal extent of pollution. We now recognize that in addi-tion to direct discharge of chemicals into the environment,many are circulated atmospherically. Toxic compounds,such as lead, mercury, and other heavy metals, are foundin trace amounts even in remote areas of Antarctica andthe Arctic (Bargagli 2000; Clarke and Harris 2003), and aretransported from industrial sources through the atmos-phere. Importantly, many compounds can have subtle yetprofound impacts on the endocrine systems of wild ani-mals (see Chapter 6). Since the 1960s we have been awareof the dangerous consequences from many noxious chem-icals, particularly those that bioaccumulate or magnify in

the food chain (Figure 3.2). In Essay 3.1, Peter Ross de-scribes potential impacts of one particularly noxious classof pollutants—persistent organic pollutants (POPs)—onkiller whales (Orcinus orca). POPs include the infamousDDT, which caused the decline of many raptor popula-tions via eggshell thinning.

Diseases are also becoming more widespread, andour recognition of their impacts is increasing. Among themost noticeable and worrisome are outbreaks of diseasethat decimate coral reefs; as corals die, the complex com-munity of fishes, algae, and invertebrate species is com-promised (Harvell et al. 1999). Andy Dobson describesseveral examples of how diseases threaten populationsand indirectly play an enormous role in altering ecosys-tems (Essay 3.2). Often, diseases become more danger-ous as a result of interactions with the stresses caused bypollutants.

Typically, species and ecosystems face multiplethreats. Importantly, the joint effects of several threatsmay be the ultimate cause of biodiversity losses. For ex-ample, the Dodo (Raphus cucullatus) went extinct onMauritius in 1681 due to human overexploitation com-

Threats to Biodiversity 65


Sinauer Associates, Inc.Groom 3ePrinciples of Conservation BiologyGro0303 eps

Humans

Dogfish

Pinnipeds and cetaceans

Forage fish

Pacific herring and sandlance

Tufted Puffin

Bald Eagles

PCBsCly Clx

Juvenile fishes and other necton

Piscivorous fish

Pacific salmon

Seabirds

Figure 3.2 Toxic chemicals that accumulate in fatty tissues, such as PCBs and dioxins,concentrate in the tissues of organisms at the top of the food chain. (Modified fromRoss and Birnbam 2003.)

66 Chapter 3


Killer Whales as Sentinels of Global PollutionPeter S. Ross, Institute of Ocean Sciences, Canada

n Persistent organic pollutants (POPs)comprise a large number of industrialand agricultural chemicals and by-prod-ucts. While these chemicals have widelyvarying applications, they share threekey features: They are persistent, bioac-cumulative, and toxic (PBT). Thephysico-chemical characteristics of thesechemicals dictate the degree to whichthey break down in the environment(persistence), the degree to which theyare metabolically broken down or towhich they accumulate in organisms(bioaccumulative potential), and thedegree to which they bind to certain cel-lular receptors or mimic natural (endoge-nous) hormones in vertebrates (toxicity).

Marine mammals are often consid-ered vulnerable to the accumulation ofhigh concentrations of POPs as a resultof their high position in aquatic foodchains, their long life spans, and theirrelative inability to eliminate these con-taminants. Because POPs are oily(lipophilic), they are easily incorpo-rated into organic matter and the fattycell membranes of bacteria, phyto-plankton, and invertebrates at the bot-tom of the food chain. As these compo-nents are grazed upon by small fishesand other organisms at low trophic lev-els, both the lipids and the POPs areconsumed. In turn, these small fishesare consumed by larger fishes, seabirds,and marine mammals that occupyhigher positions in aquatic food chains.However, lipids are burned off at eachtrophic level and are utilized for main-tenance, growth, and development,while the POPs are left largely intact.This leads to biomagnification, withincreasing concentrations of POPsfound at each trophic level. In this way,fish-eating mammals and birds areoften exposed to high levels of POPs,even in remote parts of the world.

The killer whale (Orcinus orca) is oneof the most widely distributed mam-mals on the planet. Although elusiveand poorly studied in many parts of theworld, these large dolphins have beenthe subject of ongoing study in thecoastal waters of British Columbia,Canada, and Washington State. A long-standing photo-identification cataloguebased on unique markings has facili-tated the study of populations in this

region (Ford et al. 2000). Several com-munities, or ecotypes, frequent thesecoastal waters, including the salmon-eating resident killer whales, the marinemammal-eating transient killer whales,and the poorly characterized offshorekiller whales. There are two communi-ties of resident killer whales: the north-ern residents that ply the waters ofnorthern British Columbia, and thesouthern residents that straddle theinternational boundary between BritishColumbia and Washington. Our discov-ery that killer whales are among themost contaminated marine mammals inthe world highlights concerns about theway in which POPs move great dis-tances around the planet with relativeease (Ross et al. 2000; Figure A).

Studying marine mammal toxicologyis not easy. Because deceased individu-als are not generally considered repre-sentative of the free-ranging population,biopsy samplings of blubber from liveindividuals have increasingly been usedto generate high-quality informationabout contaminant levels in their tis-sues. However, the resulting data aremeaningless if nothing is known aboutthe individual from which the biopsysample originates. Age, sex, condition,and dietary preferences all representimportant “confounding” or “natural”factors that affect the concentration ofcontaminants in the animal’s tissue. In

the case of killer whales, the photo-identification catalogue provides such abackdrop, with each sample originatingfrom an individual of known age, sex,and community (hence dietary prefer-ence). Consequently, we were able todocument that males became increas-ingly contaminated as they aged, whilefemales became less contaminated asthey transferred the majority of theircontaminant burden to their nursingcalves via fat-rich milk.

Many POPs, including the polychlo-rinated biphenyls or PCBs and the pes-ticide DDT, are highly toxic. Laboratoryanimal studies have conclusivelydemonstrated that such chemicals areendocrine disrupting, with effects notedon reproduction, the immune system,and normal growth and development.Studies of wildlife are more challenging,as free-ranging populations are exposedto thousands of different chemicals, anda number of other natural factors canaffect their health. Captive-feedingstudies have demonstrated that herringfrom the contaminated Baltic Sea affectthe immune and endocrine systems ofharbor seals (Ross et al. 1996). Studies ofwild populations provide clues aboutthe impact of POPs on marine mam-mals. However, as is the case withhumans, a combined “weight of evi-dence” from numerous lines of experi-mental and observational evidence indifferent species provides the mostrobust assessment of health risks in ani-mals such as killer whales (Ross 2000).This weight of evidence is based oninter-species extrapolation, anddepends upon the conserved nature ofmany organ, endocrine, and immuno-logical systems among vertebrates.

While regulations have resulted inmany improvements for certain POPs,new chemicals are designed each year.Killer whales represent a warning aboutthose chemicals that are unintentionallyreleased, that may travel great distancesthrough the air, and that end up at highconcentrations at the top of food chains.Given their tremendous habitat require-ments and that of their prey (salmon),Northeastern Pacific killer whales areserving as sentinels of global pollution,and a reminder that we indeed inhabit a“global village.” n

ESSAY 3.1

Sinauer Associates, Inc.Groom 3ePrinciples of Conservation BiologyGro0302aEssay.eps100% of size04.20.05

Tota

lPC

Bs

(mg/

kglip

id)

300350

250200150100

Northernresidents

Southernresidents

Transients

500

Figure A Transient killer whales representthe most PCB-contaminated marine mammalon the planet, reflecting their high trophic level(they consume marine mammals), long lifespan, and relative inability to eliminate thesecontaminants from their bodies. Females (ingray) are less contaminated than males (inwhite) because they transfer these contami-nants to their offspring via fat-rich milk. (Mod-ified from Ross et al. 2000.)



Infectious Disease and the Conservation of BiodiversityAndy Dobson, Princeton University

n Dissect any vertebrate, invertebrate,plant, or fungus and you will find awhole community of organisms livingwithin their tissues; the richness of thiscommunity of parasites will increase asyou examine the host’s tissues at finerscales of resolution. Unfortunately, ecol-ogists and conservation biologists oftenoverlook the huge variety of biodiver-sity that lives in, upon, and often at theexpense of free-living species. Parasitesand microorganisms are a major com-ponent of biodiversity, perhaps makingup as much as 50% of all living species(Price 1980; Toft 1991). Ironically,whereas few people worry about theirlong-term conservation (Sprent 1992), itis important not to ignore parasites andpathogens, as many have profoundeffects that fundamentally influence theevolution of populations and the func-tioning of ecosystems.

The Zen of ParasitismParasites become problems when natu-ral systems are perturbed, or whenspecies escape regulation by their para-sites; an important illustration of thisoccurs when invasive species escapefrom their natural pathogens. Greencrabs (Carcinus maenus) host a significantdiversity of parasites in their naturalrange along the Atlantic coast of Europe(Torchin et al. 2002). This parasite diver-sity is considerably reduced in the manyareas of the world where green crabshave been accidentally introduced andhave established as invading species.This absence of pathogens may make asignificant contribution to the crab’sability to invade, as it considerablyreduces the energy each crab puts intoresisting the invasion of its body by adiversity of parasitic species. Indeed, inareas where crabs have successfullyinvaded they may grow to five times thesize of the largest crabs found in theirnative range.

This effect appears to be an impor-tant general result; in detailed compar-ative studies of the most successfulinvasive animal species from a varietyof taxa, Torchin et al. (2003) showedthat parasite diversity is considerablyreduced in areas where the species hasinvaded. Similar results were found inthe fungal and viral pathogens of

plants that have invaded the UnitedStates (Mitchell and Power 2003). All ofthis suggests that the parasitic, under-observed half of biodiversity plays amajor role in regulating the abundanceof the more familiar free-living species.

Parasites can have dramatic effectsat ecosystem levels. Rinderpest virus(RPV), a morbillivirus that causes wide-spread mortality in ungulates, was firstintroduced into East Africa at the endof last century with cattle (Plowright1982). RPV spread throughout sub-Saharan Africa, producing mortality ashigh as 90% in some species. Travelersthrough the region report that in someplaces the ground was littered with car-casses and the vultures were so satiatedthey could not take off (Simon 1962).Even today, the observed geographicranges of some species are thought toreflect the impact of the great rinder-pest pandemic. Vaccination of cattleproduced a remarkable and unforseeneffect; the incidence of RPV in wilde-beest and buffalo declined rapidly andcalf survival in these and other wildungulates increased significantly (Tal-bot and Talbot 1963; Plowright 1982).This led to a rapid increase in the den-sity of these species; in the Serengeti,wildebeest numbers increased from250,000 to over a million between 1962and 1976, and buffalo numbers nearlydoubled over the same period andexpanded their range. This increase inherbivore density produced a signifi-cant increase in some carnivore species,particularly lions and hyenas.

A significant threat to endangeredspecies may be pathogens acquiredfrom species with large populationsthat sustain continued infections. Inthis case, the pathogen is present in onehost species and invades another, andtwo things can happen: The combinedpopulation densities of the potentialhost species may be insufficient to sus-tain the pathogen and it dies out, or theparasite sustains itself in the new com-munity of hosts.

Pathogens with Multiple HostsWhen pathogens infect multiple hostspecies it is likely that some hosts aremore resistant to the pathogen than oth-ers; West Nile virus provides an impor-

tant example. Crow species are highlysusceptible to the disease and die withina week of infection; in contrast, HouseSparrows (Passer domesticus) seem moreable to withstand infection (interestingly,House Sparrows are an invading speciesin the U.S.; their native range overlapsthat of West Nile virus, suggesting theghost of past natural selection for resist-ance [Campbell et al. 2002]).

At a further extreme, Nipah andHendra virus have caused deadly out-breaks of disease in humans anddomestic livestock in Australia,Malaysia, and Bangladesh. The mainhosts of these viruses are fruit bats(Pteropus hypomelanus, P. vampyrus, andseveral other species), where the highlevels of prevalence imply that theyexist as a relatively benign pathogen. Inundisturbed habitats, there is little con-tact between humans, their livestock,and bats. However, massive habitatconversion in Australia and Malaysiahas compressed the range of fruit batsso that the only trees left for them toroost in are those associated with inten-sive agricultural areas. This increasesrates of contact between bats and live-stock or humans. Pathogen transmis-sion usually occurs when fruits that thebats have been feeding on drop fromtrees and are consumed by pigs orhorses. These partially infected fruitscan trigger the first case of a diseaseoutbreak, which spreads through thelivestock into the agricultural workersand on to their families and friends.When this occurred in Malaysia, sev-eral million pigs had to be culled, andover a hundred humans were infected,more than 65% of whom died.

Pathogens that use multiple hostscreate a double-edged problem for ecol-ogists and conservation biologists.Pathogens like Nipah virus representone extreme, where habitat conversionincreases human exposure to novelpathogens. There is essentially no wayof predicting when similar novelpathogens will emerge; all we can sayis that the frequency of these eventswill increase as humans increase theirrates of contact with novel environ-ments and their potential hosts.

At the other extreme are multi-hostpathogens that are vector transmittedand use mosquitoes, ticks, and fleas for

ESSAY 3.2

bined with nest predation by introduced cats, dogs, pigs,and rats. Of great concern is the likelihood that somethreats may be synergistic, whereby the total impact oftwo or more threats is greater than what you would ex-pect from their independent impacts (Myers 1987).Corals often are stressed physiologically by increases in

temperature, but may also be more susceptible to fungalpathogens when stressed (Harvell et al. 1999). This syn-ergism suggests that the combined effects of globalwarming and increasing transport of disease organismsamong coral reefs could precipitate catastrophic de-clines. At times, synergisms develop through the inter-

68 Chapter 3


transmission between wildlife reservoirsand humans and their domestic live-stock. The specificity of these pathogensis determined by the feeding choice oftheir vectors. In some cases this choicewill be very specific and the pathogenwill only occasionally spread from thereservoir into a novel host. This seemsto be the case for West Nile virus,which is less of a problem in Europe,where it is transmitted predominantlyby mosquitoes that specialize in birds.In the U.S., West Nile virus is transmit-ted by a hybrid mosquito that feeds onboth birds and mammals. However,when vectors have a choice of hosts,the diversity of species present in ahabitat will have a major bufferingimpact on the scale of an epidemic out-break. Ostfeld and colleagues haveshown that this is the case for Lymedisease in the U.S., where the diversityof hosts available for ticks leads to asignificant reduction in the rate ofattack on hosts that are susceptible tothe disease (Ostfeld and Keesing 2000;LoGuidice et al. 2003). This bufferingeffect is even stronger for pathogenswhere the abundance of the vectors isindependent of that of the hosts (Dob-son 2004); this will be the case for mos-quito-transmitted diseases such asmalaria, Dengue fever, and West Nilevirus. This creates an important incen-tive to conserve biodiversity, particu-larly in the event of future climatewarming that will allow the classictropical pathogens to spread to the tem-perate zone.

Most discussion of the response ofvectored pathogens to climate changehas focused on examining howincreased temperature allow pathogensand mosquitoes to successfully com-plete their life cycle development inthe temperate zone. Once the pathogencan develop in a shorter time periodthan the mosquitoes’ life expectancy,the pathogen can establish. However,vectored pathogens will also be mov-ing down a biodiversity gradient asthey spread from the tropics to thetemperate zone. There will be lesschoice for the mosquitoes, so they canfocus their infective bites on the mostcommon species; in many places this

will be Homo sapiens and their com-mensal domestic species.

Role of Predators in BufferingOutbreaks and Keeping HerdsHealthyPredators and scavengers may providean unsuspected ecosystem service bypreventing infectious disease out-breaks. Work by Packer et al. (2003)suggests that when predators selec-tively remove infected individuals frompopulations of prey species they willsignificantly reduce the burden of dis-ease within the prey population; thismay even lead to increases in the abun-dance of prey in the presence of preda-tors! Evidence in support of this is pro-vided in populations of game specieswhere culling of predators has led toincreases in infectious diseases and par-asites as the host becomes more abun-dant. In studies of game birds in north-ern Britain the abundance of parasiticworms varies inversely with game-keeper abundance (Hudson et al. 1992;Dobson and Hudson 1994). One of thegamekeeper’s traditional jobs is toremove foxes and birds of prey; how-ever these predators differentiallyremove heavily parasitized birds fromthe bird populations. Removing thefoxes leads to a general increase in par-asitic worm burdens that have a majorimpact on bird numbers.

In a similar fashion, the control ofscavengers such as wolves, coyotes,and vultures may have permitted theemergence of prion diseases such asscrapie and chronic wasting disease inEurope and the U.S. (Prusiner 1994;Westaway et al. 1995). The natural fociof prion diseases seem to be areas char-acterized by very poor soils such aschalk grasslands. Ungulates that grazein these habitats are extremely nutrientstressed; they are particularly deficientin phosphorus and this leads them tochew on the carcasses of individualswho have failed to survive harsh win-ters. When scavenging canids areremoved there are considerably morecarcasses to chew on, and prion dis-eases that are present at very low levelsin the population can slowly become

endemic at higher prevalences. Oncethey get into domestic livestock, theycan produce devastating economicimpacts (Anderson et al. 1996). Theyprovide an important example of howpreviously obscure and little-knownpathogens can quickly become quitesignificant when we perturb naturalecosystems.

To summarize, parasites andpathogens remain an important consid-eration in the management of captiveand free-living populations of threat-ened and endangered species. Epidemi-ological theory suggests that pathogensshared among several species present alarger threat to the viability of endan-gered species than do specificpathogens. However, there is a wayparasites and pathogens may be usedto conservation advantage: Pathogenscould be effectively employed as bio-logical control agents to reduce thedensities of introduced rats, cats, andgoats that are a major threat to manyendangered island species. Obviously,caution has to be exercised when con-sidering introduction of any pathogeninto the wild, so this method of pestcontrol should be restricted to isolatedoceanic islands (Dobson 1988). How-ever, the majority of extinctionsrecorded to date in wild populationshave occurred on oceanic islands (Dia-mond 1989).

Clearly, parasites and diseases areemerging as important considerationsin conservation biology. The enormousexpansion of our ecological under-standing of parasites and their hosts inthe last fifteen years means that ecolo-gists now see a predictable structure inconditions that foster disease out-breaks. Epidemics can no longer beconsidered purely stochastic events thatoccur as random catastrophes. We nowhave a significant mathematical frame-work that delineates the general condi-tions under which a disease outbreakwill occur (Anderson and May 1986,1991; Grenfell and Dobson 1995). Amajor challenge for conservation biolo-gists is to apply and extend this frame-work so it can minimize the diseaserisk to endangered species of plantsand animals. n



Reductions inpollinators, dispersers;leads to further losses of native plants

Loss of food forpollinators, herbivores

Rise in water table

Few seeds for post firerecruitment

Death of trees and shrubs

Invasion byPhytophthora

Increase in grassand herb cover

Further reduction inassociated animals

Increase in certaininsectivores

Increase in dung beetles

Further reduction in cover, litter

Further loss of pollinators, granivores

Reduced opportunities forgermination of woodyspecies due to competition

Increased food forlarge marsupials

Increased food forcertain insect herbivores

Decreased proportion of animal-pollinated species

Loss of nectar/pollen forinsects, birds, and marsupials

Loss of seeds for granivores

Loss of nesting sites, refuges

Reduction in transpiration

Loss of canopy-stored seeds

Loss of habitat forrecruiting plants

Figure 3.3 “Snowballing” effect of invasion of the alien root pathogen Phytophthora cin-namomi into shrublands and woodlands of western Australia. (Modified from Richardsonet al. 1996.)

action of a threat and population density. In an experi-mental study, Linke-Gamenick and her colleagues (1999)showed that a toxic chemical reduced survivorship of acapitellid polycheate worm, but at high concentrationsits impacts grew more severe with increasing density.

Further, many threats can intensify as they progress,a process known as “snowballing.” Invasion of plantcommunities in western Australia by the alien rootpathogen Phytophthora causes death of woody speciesand an increase in herbaceous cover, which can suppressgermination and early growth of woody seedlings(Richardson et al. 1996; Figure 3.3). The changes in theplant community in turn decrease the suitability of thecommunity for many animal species, which may furtherreduce the capacity of the animal community to fosterthe development of woody cover. Thus, the initial inva-

sion pushes the entire system into a new balance thatmakes recovery of the original system difficult.

Finally, many species are threatened by interactions be-tween large impacts, such as direct mortality from har-vest, and more subtle impacts on their biology and popu-lation dynamics. For example, a variety of changes in thegenetic structure of populations can enhance their risk ofextinction. Species may lose functional genetic diversitydue to prolonged isolation in small populations, or theloss of entire populations, which may leave them less ableto cope with stresses of habitat degradation or climatechange. These genetic threats, as well as many genetictools for conservation are discussed in Chapter 11. Intrin-sic demographic factors, such as rarity, low reproductiverates, or low dispersal rates can further predispose a pop-ulation or species to extinction, as discussed later in this

70 Chapter 3


chapter and in Chapter 12. Importantly, species may varyin their responses to different threats based on their biolo-gy (Owens and Bennett 2000; Isaac and Cowlishaw 2004).Such demographic predispositions to risk can be exagger-ated via reduced reproduction or increased mortality dueto anthropogenic factors. Typically, both genetic and de-mographic threats interact with habitat and climatechange, overexploitation, or species invasion to causespecies loss, and ultimately changes in community andecosystem function. Thus, although rarely the primarycause of extinction, genetic and demographic factors areimportant contributors to endangerment that must beconsidered in most conservation situations.

Anthropogenic Extinctions and TheirCommunity and Ecosystem ImpactsThe most obvious and extreme unwanted effects ofhuman development are the extinction of species andpopulations, and transformation of natural ecosystemsinto degraded or even uninhabitable places. We can con-sider these the ultimate consequences of the expansionof human populations and consumption levels. Here,we will discuss patterns of extinction in some depth, andtreat habitat transformation in detail in Chapter 6.

Extinction can be either global or local. Global extinc-tion refers to the loss of a species from all of Earth, where-as a local extinction refers to the loss of a species in onlyone site or region. In addition, ecological extinction canoccur when a population is reduced to such a low densi-ty that although it is present, it no longer interacts withother species in the community to any significant extent(Redford 1992; Redford and Feinsinger 2000). All theseforms of extinction can affect remaining species, perhapscausing shifts in community composition, or ecosystemstructure and function. Global extinction is the most trag-ic loss resulting from human activities, because once aspecies is lost entirely, it cannot be recreated.

Anthropogenic extinctions are caused directly throughoverexploitation, and also indirectly via habitat transfor-mations that restrict the population size and growth ofsome species, or the introduction of species into newareas that over-consume or outcompete native species.Earliest human- caused extinctions were probably due tooverexploitation, but increasingly habitat modificationand introductions of invasive species to islands becameprimary causes. Only recently have other factors such asdisease, pollution, and anthropogenic global climatechange begun to play major roles as well. As we look tothe future, synergisms among these factors and climatechange are likely to accelerate extinction rates in the com-ing century (Myers 1987; Myers and Knoll 2001).

Because of the inherent spottiness of the fossil record,it is difficult to discern extinction events prior to record-

ed history, or to document the nature of changes to eco-logical communities. Our knowledge of prehistoric ef-fects of humans on biodiversity is thus limited to a fewcases where the fossil record can provide a clear trail,and to fairly gross-scale changes in ecosystems. Similar-ly, our incomplete knowledge of living taxa also makesthis task difficult even after ecological records were keptin detail. Yet, some patterns are traceable.

Most notably, the Pleistocene extinction of megafauna(mammals, birds, and reptiles over 44 kg in body size) andother vertebrates speaks of widespread impacts of humansthat have forever changed ecological communities. The de-mise of between 72% and 88% of the genera of large mam-mals in Australia, North America, Mesoamerica, andSouth America coincided with the arrival of humans ineach continental region (44,000–72,000 years ago in Aus-tralia and 10,000–15,000 years ago in North and SouthAmerica; Figure 3.4). Certainly the coincidence of pulses ofhuman colonization or population growth and the loss oftaxa is suggestive of a strong role of “Earth’s most ingen-ious predator” in the loss of these creatures (Steadman andMartin 2003). However, rapid climate change and con-comitant vegetation change also took place in most cases(Diamond 1989; Guthrie 2003; Barnosky et al. 2004), andloss of many taxa in Alaska or Northern Asia wherehuman populations were never large suggests that climateplayed a major, or in a few cases the only role in megafau-nal extinctions (Barnosky et al. 2004).

Careful consideration of the evidence suggests that theloss of the megafauna may have resulted from a combi-nation of range contraction due to climate change and de-creases in population sizes due to hunting and habitat al-teration by humans (see Figure 3.4). For example,mammoths often survived longest in areas withouthuman populations. Although glaciations eventuallycaused extinctions of all mammoths, fragmentation ofmammoth ranges among larger groups of Pleistocenehuman hunters may have tipped the balance for somepopulations (Barnosky et al. 2004). Both hunting andhabitat change (e.g., burning of savannahs to improvegame and forage conditions) seem the largest drivers ofthe demise of many large mammals in the conterminousUnited States (Martin 2001; Miller et al. 1999), resulting inmore pronounced extinction events among these taxathan could be accounted for by climate change alone. Fi-nally, humans were likely contributors to animal extinc-tions in Africa, although because human presence inAfrica is so ancient, it is much more difficult to establisha causal link.

Dramatic extinction events occurred among birds asPolynesians colonized Pacific Islands 1000–3000 yearsago. Over 2000 species (particularly flightless rails), andover 8000 populations were driven extinct by overex-ploitation, habitat alteration, and the introduction of rats,pigs, and other commensal mammals carried by the

Polynesians (Steadman 1995). In some cases, extinctionoccurred within the first 100 years after an island wascolonized (Steadman 1995), while in others, species mayhave persisted for a thousand years, or to the presentday (Steadman and Martin 2003).

A useful lesson emerges from study of the patterns ofextinction among birds in the Pacific islands. Certain fac-tors are correlated with rapid extinction of taxa, and oth-ers with their persistence or at least delayed extinction(Table 3.1). Most importantly, where humans introducedmany species of predators (particularly rats), consumers,and certain weeds, and began cultivation, extinction wasthe predominant outcome among certain avian groups(Steadman and Martin 2003). In other words, where in-vasive species and habitat degradation were combined,extinctions were most common. Other abiotic and biotic

factors, such as the size and shape of islands and theirspecies richness, were also predictors of extinction risk.

Somewhat surprisingly, patterns of extinction in morerecent times are often obscure because despite the exis-tence of historical records, few species were studied wellenough for the causes of extinction to be understood.Further, many species particularly sensitive to humanactivities undoubtedly were lost before they wererecorded, particularly across Europe, parts of China, andAfrica or other locations with a long history of humanoccupation (Balmford 1996). Often the cases that are wellunderstood are examples of rapid overexploitation, suchas the extinction of Stellar’s sea cow (Hydrodamalis gigas)and the Great Auk (Pinguinus impennis).

Birds are the only taxa for which causes of extinctioncan be ascribed in the majority of cases. Since 1500, we



Sinauer Associates, Inc.Groom 3ePrinciples of Conservation BiologyGro0305ADJ.eps100% of size05.05.05

9

8

33

Cause of ExtinctionHumansClimateInsufficient data

50

21

550

>30

20 14 9 50 45 20

13

9

50 13 9~160

7580 40 28

72–44

913

12.5

Figure 3.4 Mammalian megafaunal genera (species > 44 kg) went extinct soon afterhuman migrations into North America, Australia, South America, and Europe. Num-bers in mammal icons represent the total number of genera that went extinct and theshading indicates cause of extinction; see inset legend. Bars indicate the period of ex-tinction (in kya), and shading indicates the magnitude of extinctions during that time;Black = many, dark gray = some, light gray = few, white = one or none. Numbers nextto human icons indicate when humans arrived to the continent. Rapid climate warm-ing (occurring from 14–10 kya) contributed to extinction in many cases, particularly inEurasia and South America. In South America climate change and the arrival of hu-mans entirely coincided with an extinction spasm of 50 genera. (Modified fromBarnosky et al. 2004.)

know that at least 129 species went extinct (IUCN 2004).Habitat loss and degradation was a factor in most cases,particularly for species with narrow ranges. The intro-duction of alien invasive species, such as the black rat,was especially influential in the extinction of island en-demics. Finally, overexploitation for food, feathers, or thepet trade, contributed to the demise of many species, in-cluding spectacular examples such as the massive over-harvest of the Passenger Pigeon (Ectopistes migratorius)and Carolina Parakeet (Conuropsis carolinensis), both ofwhich once had populations in the millions. As dis-cussed above, extinction often was caused by a combi-nation of two or more factors.

Indirect impacts of extinctions on animal andplant communitiesAs dramatic as these prior human-mediated extinctionevents have been, an equally dramatic adjustment of ourconcepts of “undisturbed” or “pristine” communities orecosystems is now necessary. The extinction of large-bod-

ied species, as well as untold numbers of more poorly fos-silized species, is likely to have caused significant changesin the composition, character, and extent of ecological com-munities. Further, ongoing extinctions of species or popu-lations, and even the reduction of some species to low pop-ulation sizes, are changing present-day ecosystems.

Where species depend on their interactions withother species, extinction can have ripple effects as theseinteractions are disrupted. Thus, the loss of key speciescan spark a suite of indirect effects—a cascade effect ofsubsequent, or secondary extinctions, and substantivechanges to biological communities. Secondary extinc-tions, those caused indirectly by an earlier extinction, aremost likely to occur when species rely on a single or afew species as prey or as critical mutualists. For example,a plant that relies on a single bat species for pollinationwill not be able to reproduce should that bat go extinct.Similarly, a carnivore specializing on two species of in-sects may be unable to maintain itself if one of its preyspecies went extinct.

72 Chapter 3


Promotes Extinction Delays Extinction

Abiotic factors

Island size Small Large

Topography Flat, low Steep, rugged

Bedrock Sandy, noncalcareous, sedimentary Limestone, steep, volcanics

Soils Nutrient rich Nutrient poor

Isolation No near islands Many near islands

Climate Seasonally dry Reliably wet

Biotic factors

Plant diversity Depauperate Specie-rich

Animal diversity Depauperate Species-rich

Marine diversity Depauperate Species-rich

Terrestrial mammals Absent Present

Species-specific traits Ground-dwelling, flightless, large, Canopy-dwelling, volant, small, wary, tame, palatable, colorful feathers, bad-tasting, drab feathers,long and straight bones for tools short and curved bones

Cultural factors

Occupation Permanent Temporary

Settlement pattern Island-wide Restricted (coastal)

Population growth and Rapid; high Slow; lowdensity

Subsistence Includes agriculture Only hunting, fishing, gathering (especially in marine zone)

Introduced plants Many species, invasive Few species, noninvasive

Introduced animals Many species, feral populations Few or no species, no feral populations

Source: Modified from Steadman and Martin 2003.

TABLE 3.1 Factors Influencing Vertebrate Extinction Following Human Colonization of Oceanic Islands

Cascade effects are probable in any community wherestrong interactions among species occur, be they preda-tor–prey, mutualistic, or competitive in character, and thusare likely to have occurred prior to recorded history aswell as in the few recorded cases from recent times (Box3.1). One of the best-known examples of a cascade effectoccurred when local extinction of sea otters (Enhydralutris), which were aggressively hunted for their pelts, ledto a transformation of marine kelp forest communities offthe Pacific coast of North America (Estes et al. 1989). Seaotters are heavy consumers of sea urchins, and their ab-sence led to an urchin population explosion, which in turnleads to overgrazing of kelp and other algae by theurchins, creating “urchin barrens” (Estes et al. 1989). Thus,because kelp forests are a haven for a broad variety offishes and other species (Dayton et al. 1998), the local ex-tinction of sea otters leads to the local extinction of manyother species, and a radical change in the nature of thestructure and composition of the community.

Cascade effects are difficult to demonstrate in thehyper-diverse tropics, but are likely to occur there. Trop-ical forests in which species have been driven locally ex-tinct (“empty forests,” Redford 1992) or depleted (“half-empty forests,” Redford and Feinsinger 2001) now maylack effective populations of key interactors. The loss ofcritical seed dispersers could eventually result in the ex-tinction of disperser-dependent tree species (Janzen1986; see Essay 12.3). Depletion of top predators maycause the ecological release of prey species that may inturn drive down populations of their prey, as occurredwith sea otters (Terborgh et al.1999; see Case Study 7.3).Thus, many tropical communities may appear healthyon the surface, but are in fact destined to decline in di-versity due to disruption of critical interactions throughlocal extinction of pollinators, seed dispersers, and toppredators.

Overexploitation of great whales from 1700 to themid-1900s had enormous impacts on marine ecosystems



Awide variety of studies haveshown that losses of a single orgroup of critical species can have

a cascade effect on biological commu-nities, with implications for the func-tioning of ecosystems. The loss of toppredators is most commonly cited ascausing cascade effects. Reduction oftop predators typically results inincreases in prey, and thus enhancedpopulations of medium-sized preda-tors (mesopredators), which in turnleads to strong declines of their preyspecies (especially birds and mammals)(e.g., Palomares et al. 1995, Crooks andSoulé 1999, and Terborgh et al. 1999,2001). Herbivore release from the lossof top predators can reduce plant pop-ulations and plant community diver-sity, which in turn reduces diversity ofother herbivores (e.g., Estes et al. 1989,Leigh et al. 1993, and Ostfeld et al.1996). Cascade effects can lead to thescavenger community as well (Berger1999; Terborgh et al. 1999). Similar cas-cade effects have also been seen infood webs with insect top predators(Rosenheim et al. 1993; Letourneau etal. 2004), although these are much lessstudied, so little is known about thecommonness of these effects.

Studies of lake ecosystems hasrepeatedly shown that loss of piscivo-

rous fishes release fish that graze onzooplankton, leading to a reduction inzooplankton, increases in phytoplank-ton, and broad rearrangements incommunity composition (e.g., Carpen-ter et al. 1985, Vanni et al. 1990, Car-penter et al. 2001, and Lazzaro et al.2003). Effects can include changes inwater clarity and large scale shifts inmacroinvertebrate abundance anddiversity (Nicholls 1999). Importantly,cascades do not always occur whenpiscivorous fish are removed, but onlyunder specific conditions (Benndorf etal. 2002). Both ecosystem and commu-nity level effects can result from theloss of anadromous fishes. In thenorthwest of North America reductionin, or loss of, salmon populations cancause decreases in the input of nutri-ents to inland streams (Schindler et al.2003), and loss of food for grizzlybears, bald eagles, killer whales, andpredaceous fishes (e.g., Francis 1997and Willson et al. 1998).

The loss of large-bodied species canoften cause ripple effects through acommunity. Many studies in tropicalforest have documented that the lossof many large-bodied species preferredby hunters leads to release of theirprey, and loss of any services to othercommunity members (Dirzo and

Miranda 1991; Redford 1992; Redfordand Feinsinger 2001). Reduction ofplant diversity can result throughenhanced granivory and herbivory(Terborgh and Wright 1994; Ganzhornet al. 1999), or through loss of seed dis-persal, and sometimes pollination,services (e.g., Chapman and Onder-donk 1998, Andersen 1999, Hamannand Curio 1999, and Wright et al. 2000).

Many cascade effects can be causedby changes in the abundance ofecosystem engineers. Loss of ecosys-tem engineers results in large struc-tural changes in ecosystem, such asloss of pools created by beavers(Naiman et al. 1988), or shifts in plantdiversity with loss of grazing by bison(Knapp et al. 1999). Loss of detritivoresin streams can also cause large-scalechanges (Flecker 1996).

Finally, cascade effects can resultfrom species introductions. Introduc-tion of species to marine estuaries orto lake systems has been shown toresult in massive reduction in nativealgae, loss of native crayfish, molluscs,and other invertebrates, and even in areduction in waterfowl, fishes, andamphibians (e.g., Olsen et al. 1991, Hilland Lodge 1999, and Nyström et al.2001).

BOX 3.1 Cascade Effects Resulting from Loss of a Critical Species orTaxon, or from Species Introductions

(Roberts 2003). The removal of such large consumers re-leased their prey, causing changes throughout the foodchain. Industrial-scale fishing of large-bodied fishspecies that occupy higher places on the food chain alsohas initiated dramatic cascading effects throughout ma-rine ecosystems (Dayton et al. 1995; Parsons 1996). Thenet result may be marine communities that bear little re-semblance to the structure and abundances that wouldhave been typical before humans began whaling andfishing on extensive scales (Pauly et al. 1998). Presum-ably, the loss of megafauna in the Pleistocene had simi-lar community-level impacts to those that we can de-scribe from more contemporary events.

Although it would be helpful in conservation to knowwhich species are most critical to communities, in practiceit is not simple to identify which species have the largestimpacts, or are involved in the greatest number of stronginteractions (Berlow et al. 1999). Dominant species arethose that are very common and that also have strong ef-fects on other members of the community (Figure 3.5). Ex-amples include reef-building corals, forest trees, and largeherbivores, such as deer. Ecosystem engineers, thosespecies such as beavers or elephants that strongly modifyhabitat, are also ones whose absence or presence willchange communities (Naiman et al. 1986; see Case Study9.1). Generally, both community dominants and ecosys-tem engineers are relatively easy to identify.

A keystone species is a species that has a greater im-pact on its community than would be expected by thecontribution of its overall numbers or biomass (Paine1969; Power et al. 1996; see Figure 3.5). If a bat species isnecessary for pollinating many species, and no other

species can serve its role, then it would be considered akeystone species. Similarly, large carnivores frequently actas keystone species through their impacts on other pred-ators, a wide variety of prey species, and the competitorsof their prey (Crooks and Soulé 1999). Ecosystem engi-neers are usually also keystone species, as defined above.Unfortunately, unlike dominant species whose impactsare easily discernable, often a keystone species is not soeasily recognized and is only discovered after its numbershave been reduced and the impacts become obvious.

Current Patterns of Global EndangermentAs we consider the present era, the challenges to biodi-versity have intensified in many respects. Where prehis-toric impacts were dominated by overexploitation, mod-erate habitat modification, and introduction of humancommensals, in recorded history we add the problems ofpollution and human infrastructure development, andultimately human-mediated climate change. In our era,vastly larger human populations and greater consump-tive habits ensure that each primary threat has increasedin magnitude and extent. In Chapter 6 we will examinehow these threats have resulted in habitat degradationacross the globe, but here we will focus on effects onspecies. To help us direct our efforts, and to help moti-vate social and political will to act on this biodiversitycrisis, it may help to review what is known about globalpatterns of species endangerment, as well as those of se-lected countries.

74 Chapter 3


Keystone species(wolves; bats; beavers; fig trees; disease-causing organisms)

Dominant species(forest trees; deer; giantkelps; reef-building corals)

Rare species(wildflowers; butterflies; mosses)

Common species with low impact(understory herbs; shrubs)

Proportional biomass of species

Tota

lim

pact

ofsp

ecie

s

Figure 3.5 Keystone and dominant speciescan have large impacts on biological commu-nities. Keystone species by definition make uponly a small proportion of the biomass of acommunity, yet have a large impact, whereasdominant species have impacts that are moreproportional to their biomass or abundance.Reductions in the biomass or extinction ofkeystone or dominant species can be expectedto cause cascade effects in communities. Manyrare and common species have low impacts,and changes in their abundance may not havenoticeable effects on other species. (Modifiedfrom Power et al. 1996.)

Globally threatened speciesThe best data on global endangerment are collated in theIUCN Red List of Threatened Species (www.redlist.org).The Red List classifies all species reviewed into one ofnine categories (Box 3.2), with three primary categoriesof endangerment, in order of the risk of extinction: Criti-cally Endangered (CR), Endangered (EN), and Vulnera-ble (VU). Very specific rules have been adopted to stan-dardize rankings of each species, and to allow use of thelist to index changes in the status of biodiversity overtime (see Box 3.1, Table A; IUCN 2001). Taxonomic ex-perts, conservationists, and other biologists work to-gether in teams to conduct the reviews, and thus assurethat the best available information is used in each case,although large uncertainties due to incomplete knowl-edge often make judgments difficult. Thus, the Red Listis seen as a work in progress, undergoing constant revi-sion both to document true changes in status, and to re-flect updates in our knowledge. Through these efforts,the IUCN Red List has become the most complete data-base on global status of species available.

Complete evaluations have been undertaken for birdand amphibian species, and are nearly complete formammals, and gymnosperms (conifers, cycads, andginkgos) among plants. However, only a small percent-age of all other taxa have been reviewed (about 6% ofreptiles and fishes; 0%–3% of invertebrates, and 1%–5%of other plant groups; Table 3.2). Overall only 2.5% of de-scribed species have been evaluated. Among thosespecies that have been evaluated, many are so poorlyknown that they cannot be categorized, and are given aData Deficient ranking (9.5%; see Box 3.1).

Of the 38,046 species evaluated as of November 2004,41% are endangered (CR, EN, or VU), 10% are NearThreatened or Conservation Dependent (meaning theymay become endangered in the coming decades), while38% are designated as Least Concern (indicating a verylow risk of extinction for the foreseeable future). The RedList also includes a conservative tally of extinction,recording a total of 784 extinctions, plus 60 extinctions inthe wild (where the only living individuals are in captiv-ity or cultivation) (Baillie et al. 2004). Globally, 317 ma-



Percent of Percent of Described Evaluated Threatened described species evaluated speciesspecies species species threatened threatened

Vertebrates

Mammals 5416 4853 1101 20 23

Birds 9917 9917 1213 12 12

Reptiles 8163 499 304 4 61

Amphibians 5743 5743 1856 32 32

Fish 28,600 1721 800 3 46

Invertebrates

Insects 950,000 771 559 0.1 73

Molluscs 70,000 2163 974 1 45

Crustaceans 40,000 498 429 1 86

Others 130,200 55 30 0.02 55

Plants

Mosses 15,000 93 80 0.5 86

Ferns 13,025 210 140 1 67

Gymnosperms 980 907 305 31 34

Dicotyledons 199,350 9473 7025 4 74

Monocotyledons 59,300 1141 771 1 68

Lichens 10,000 2 2 0.02 100

Total 1,545,594 38,046 15,503 1 41

Note: A “threatened species” includes any species designated as CR, EN, or VU by the IUCN Red List.Source: Modified from IUCN 2004.

TABLE 3.2 Number of Globally Threatened Species by Taxon

76 Chapter 3


BOX 3.2 The IUCN Red List System

T he IUCN Red List System, a sys-tematic listing of species in threatof extinction, was initiated in

1963 to be used in conservation plan-ning efforts around the globe. Overtime, hundreds of scientists haveworked to create listing criteria thathave been carefully defined to be max-imally useful as a diagnostic tool tohelp establish extinction risk over alltaxa. Species are assigned to one ofnine categories, which indicate theirthreat status or their status in thereview process (Figure A). These cate-gories are defined as follows:

Extinct (EX)A taxon is Extinct when there is no rea-sonable doubt that the last individualhas died. A taxon is presumed extinctwhen exhaustive surveys in knownand expected habitat, at appropriatetimes (diurnal, seasonal, annual) to thetaxon’s life cycle and life form,throughout its historic range havefailed to record an individual.

Extinct in the Wild (EW)A taxon is Extinct in the Wild when it isknown only to survive in cultivation, incaptivity, or as a naturalized popula-tion (or populations) well outside thepast range, and there is no reasonabledoubt that the last individual in thewild has died, as outlined under EX.

Critically Endangered (CR)A taxon is Critically Endangered whenthe best available evidence indicates

that it meets any of the criteria A–E inTable A for Critically Endangeredspecies, and is therefore facing anextremely high risk of extinction in thewild.

Endangered (EN)A taxon is Endangered when the bestavailable evidence indicates that itmeets any of the criteria A–E forEndangered (see Table A) and is there-fore facing a very high risk of extinc-tion in the wild.

Vulnerable (VU)A taxon is Vulnerable when the bestavailable evidence indicates that itmeets any of the criteria A–E for Vul-nerable (see Table A) and is thereforefacing a high risk of extinction in thewild.

Near Threatened (NT)A taxon is Near Threatened when ithas been evaluated against the criteriabut does not qualify for CriticallyEndangered, Endangered, or Vulnera-ble now, but is close to qualifying foror is likely to qualify for a threatenedcategory in the near future.

Least Concern (LC)A taxon is deemed Least Concernwhen it has been evaluated againstthe criteria and it neither qualifies forthe previously described designations(Critically Endangered, Endangered,Vulnerable, or Near Threatened), nor isit likely to qualify in the near future.

Widespread and abundant taxa areincluded in this category.

Data Deficient (DD)A taxon is Data Deficient when there isinadequate information to make adirect or indirect assessment of its riskof extinction based on its distribution,population status, or both. Every effortis made to use this category as a lastresort, as this is not a category ofthreat, but only indicates more infor-mation is needed to make a statusdetermination.

Not Evaluated (NE)A taxon is Not Evaluated if it is has notyet been evaluated against the criteria.

Assignment to one of the threethreatened categories (CR, EN, or VU)is made on the basis of a suite ofquantitative standards adopted in1994 that relate abundance or geo-graphic range indicators to extinctionrisk (see Table A). The different criteriaand their quantitative values (A–E)were chosen through extensive scien-tific review, and are aimed at detect-ing risk factors across the broad diver-sity of species that must beconsidered (IUCN 2001).Qualificationunder any of the criteria A–E is suffi-cient for listing; however, evaluationsare always made as completely aspossible for use in evaluating changesin status over time, and for conserva-tion planning purposes. Thus, the sta-tus of a taxon will be evaluatedaccording to most of these criteria, as

Extinct (EX)

Extinct in the Wild (EW)

Critically Endangered (CR)

Endangered (EN)ThreatenedAdequate data

Evaluated

Availabledata

Not Evaluated (NE)

Data Deficient (DD)

Vulnerable (VU)

Near Threatened (NT)

Least Concern (LC)

Figure A IUCN Red List categories. Every effort ismade to employ all available data to avoid placingspecies in the Data Deficient category.


rine, 2981 freshwater, and 13,657 terrestrial species areconsidered endangered, and recorded extinctions aresimilarly apportioned.

For the most complete evaluated taxa, amphibiansand gymnosperms stand out as particularly threatened(see Table 3.2). Cycads, an ancient group of gym-nosperms, are especially vulnerable, with 52% endan-gered. The true level of threat is undoubtedly higher thanthese estimates due to the large number of Data Deficientrankings: 1290 amphibians (23%), 360 mammals, 78 birdsand 77 gymnosperms all are too poorly known to beranked, but certainly some of these are endangered.

Among mammals, ungulates, carnivores, and pri-mates are particularly at risk mostly due to habitatdegradation and overexploitation (Baillie et al. 2004).Albatrosses, cranes, parrots, pheasants, and pigeons areparticularly threatened among the birds due to bycatch,habitat loss, the pet trade, and direct exploitation(Birdlife International 2004). Amphibians appear to beat greatest risk of extinction, with a high fraction ofthese species listed as critically endangered (21%; IUCN2004). A wide variety of threats affect amphibian popu-lations throughout the world, many of which exert syn-ergistic effects on these sensitive animals; these threats


far as is possible given current knowl-edge.

A major advance in risk evaluation,the Red List criteria require efforts toplace quantitative bounds on ourknowledge, and explicitly allow foruncertainty. The assignments to cate-gory are not assignments of priority,

but rather a reflection of our currentbest judgment of how great the risk ofextinction is for this species, given thebest available information at present.All species on the list must be reevalu-ated at least once every ten years.

In addition to quantifying risk ofextinction, the Red List compiles data

on the nature of the threats to thespecies. These evaluations are usefulfor initial efforts to conserve thethreatened species, and in aggregatecan guide efforts to reduce threaten-ing processes.

Critically Endangered Endangered Vulnerable

Criterion (CR) (EN) (VU) Qualifiers

A.1Reduction in >90% >70% >50% Over 10 years or 3 generations in the past where

population size causes are reversible, understood, and have ceased

A.2–4Reduction in >80% >50% >30% Over 10 years or 3 generations in the past, future,

population size or combination, where causes are not reversible, not understood, or ongoing

B.1Small range <100 km2 <5000 km2 <20,000 km2 Plus two of (a) severe fragmentation or few

(extent of occurrences (CR = 1, EN = 2–5, VU = 6–10), occurrence) (b) continuing decline, (c) extreme fluctuation

B.2Small range <10 km2 <500 km2 <2000 km2

(area of occupancy)

CSmall and <250 <2500 <10,000 Mature individuals, plus continuing decline either

declining over a specific rate in short time periods, or withpopulation specific population structure or extreme

fluctuations

D.1Very small population <50 <250 <1000 Mature individuals

D.2Very small range — — <20 km2 or Capable of becoming CR or EX within a very short

<5 locations time

EQuantitative >10% in >20% in >50% in Estimated extinction risk using quantitative

analysis 100 years or 3 20 years or 5 100 years models, e.g., population viability analysesgenerations generations

Source: IUCN 2001.

TABLE A Overview of Criteria (A–E) for Classifying Species as CR, EN, or VU in IUCN Red List

are described by Joe Pechmann and David Wake inCase Study 3.1.

While levels of endangerment among the remaininggroups reflect a tendency for worrisome cases to be putforward for evaluation ahead of general analyses (for ex-ample, see crustaceans and mosses in Table 3.2), it seemslikely that high levels of threat may occur in many othergroups. Turtles and tortoises have been more complete-ly assessed among the reptiles, and 42% are endangered(Baillie et al. 2004). Marine and freshwater fishes areplaced on the list in increasing numbers, suggesting a se-rious level of threat, whatever the exact percentage. Twomessages from these statistics are clear: First, a large frac-tion of vertebrate and gymnosperm diversity is in greatdanger of extinction over the next century, and second,we know very little about the status of most other taxa.

Globally threatened processes Not only are species at risk of extinction, but some process-es that undergird ecosystem functions, or that are gloriousin and of themselves, are put at risk from human activities.Lincoln Brower discusses the concept of an “endangeredbiological phenomenon,” in which a species is likely tosurvive, but some spectacular aspect of its life history, suchas the mass annual migration of monarch butterflies be-tween Mexico and the United States, is in jeopardy of dis-appearing (Essay 3.3). The mass migrations of springbokin southern Africa have already been eliminated but theseasonal migrations of vast herds of wildebeest and zebrain the Serengeti still exist. Not only is this mass migrationan amazing spectacle, but the Serengeti grasslands areadapted to the impacts of high densities of these grazers,as well as other ungulates. The loss of wildebeest migra-

tions would be tragic, even though wildebeest would sur-vive in many places.

What factors are most threatening to biodiversity globally?As species are evaluated, all threats faced at present or inthe past that led to endangerment are coded into the RedList database. Our knowledge about the threats faced byspecies varies tremendously as certain types of threats areeasier to document (e.g., forest conversion versus compe-tition from invasive species), and for some species it is onlypossible to say that they are declining and endangered, butnot to diagnose why. Most species face multiple threats. Fi-nally, the red list is dominated by evaluations primarily forvertebrates and some vascular plants, and for areas wheremany biologists already work, perhaps reflecting biases ininterest among biologists more than intrinsic levels ofthreat (Burgman 2002). Thus, we cannot be sure that thepatterns that may hold for birds will serve for mollusksand other species groups (and indeed, we should expectthat they will not in many cases). Nonetheless, to a limiteddegree, we can use these data to give us a sense of the per-vasiveness of different types of threats.

Habitat loss and degradation is the most pervasive andserious threat to mammals, birds, amphibians, and gym-nosperms (Figure 3.6). Overexploitation is the most perva-sive threat to fishes, and a predominant one for mammalsand birds as well, whereas invasive species pose particularrisks for birds on islands (Baillie et al. 2004). Amphibiansare more challenged than birds and mammals by pollutionthat directly kills individuals, and by changes in the abun-dance of native species, particularly diseases. Intrinsic fac-tors, such as a limited reproductive capacity or limited

78 Chapter 3


Mammals Birds GymnospermsAmphibians

.40Percent of threatened species affected

.80.40 .80 .40 .80.400 0 0 0.80

Intrinsic factors

Pollution

Disease

Invasive species

Overexploitation

Habitat loss and degredation

Figure 3.6 Habitat loss and degradation is the greatest threat to global biodiversityamong mammals, birds, amphibians, and gymnosperms. Because not all threats aredocumented, this figure underestimates threat levels among Red Listed species. Over-exploitation includes both direct mortality and by-catch. (Modified from IUCN 2004.)



An Endangered Biological Phenomenon Lincoln P. Brower, Sweet Briar College

n Much of conservation researchfocuses on describing diminishingspecies diversity and on understandingthe processes that lead species to smallpopulations, and thence to extinction.Here I discuss endangered biologicalphenomena, defined as a spectacularaspect of the life history of an animal orplant species, involving a large numberof individuals, that is threatened withimpoverishment or demise. The speciesper se need not be in peril; rather, thephenomenon it exhibits is at stake(Brower and Malcolm 1991).

Examples of endangered biologicalphenomena include the ecologicaldiversity associated with naturallyflooding rivers, the vast herds of bisonof the North American prairie ecosys-tems, the synchronous flowering cyclesof bamboo in India, the 17-year and 13-year cicada emergence events in easternNorth America, and scores of currentanimal migrations. Instances of the lat-ter include seasonal migrations of theAfrican wildebeest and North Ameri-can caribou, the wet and dry seasonmovements of Costa Rican sphingidmoths, the billion individuals of 120songbird species that migrate fromCanada to Neotropical overwinteringareas, and the highly disrupted migra-tions of numerous whale species.

There are two principal reasons whyanimal migrations are endangered byhuman activities. First, migrant speciesmove through a sequence of ecologicallydistinct areas, any one of which couldbecome an Achilles’ heel. Second, aggre-gation of the migrants can occur, makingthe animals especially vulnerable. Themajor impact on migratory species isdue to accelerating habitat modificationthroughout the world. Even when prob-lems are recognized, mitigation is diffi-cult because of varying policies andenforcement abilities in the differentcountries the animals occupy during thedifferent phases of their migration cycles.The extraordinary migration and over-wintering behaviors of the monarch but-terfly in North America well exemplifythe concept of endangered phenomena.

The monarch butterfly (Danaus plex-ippus) is a member of the tropical sub-family Danainae, which contains 157known species. It is alone in its subfam-

ily for having evolved extraordinaryspring and fall migrations (Figure A)that allow it to exploit the abundantAsclepias (milkweed) food supplyacross the North American continent,becoming one of the most abundantbutterflies in the world. Remarkably,and in contrast to vertebrate migra-tions, the monarch’s orientation andnavigation to its overwintering sites iscarried out by descendants three ormore generations removed from theirmigrant forebears. Its fall migration,therefore, is completely inherited, withno opportunity for learned behavior.This, together with the vastness of themigration and overwintering aggrega-tions, constitutes a unique biologicalphenomenon.

Two migratory populations of themonarch occur in North America. The

larger one occurs east of the RockyMountains and undoubtedly representsthe stock from which the smaller, west-ern North American migration evolved.Both migrations are threatened becausethe aggregation behavior during winterconcentrates the species into severaltiny and vulnerable geographic areas.

By late summer, the monarch popu-lation in eastern North America buildsto an estimated 0.5–3 billion individu-als over an enormous area east of theRocky Mountains. Beginning in lateAugust, the adult butterflies migrate tocentral Mexico, where they overwinterfor more than five months in high-ele-vation fir forests, about 90 km west ofMexico City (see Figure A). Here thebutterflies coalesce by the hundreds ofmillions into dense and stunninglyspectacular aggregations that festoon

ESSAY 3.3

Northern limit ofAppala

chia

nM

ount

ains

Tropic of Cancer

Sierra Madre O

ccidental

50ºN

40ºN

30ºN

20ºN

50ºN

40ºN

30ºN

20ºN

80ºW90ºW100ºW110ºW120ºW

Sierra Madre O

riental

Asclepias

RockyM

ountains

Neovolcanic Belt

Summer breeding rangeSpring breeding rangeMigration routesPossible migration routesOverwintering sites

Figure A Fall and spring migrations of the eastern and western populations of the monarchbutterfly in North America. (Modified from Brower 1995.)

80 Chapter 3


0.1 to 5 hectares of forest. The butter-flies, effectively in cold storage, remainsexually inactive until the approach ofthe vernal equinox.

Survivors from the Mexican over-wintering colonies begin migratingnorthward in late March to lay theireggs on sprouting milkweed plants(Asclepias spp.) (see Figure A). These 8-month-old remigrants then die andtheir offspring, produced in late Apriland early May, continue the migrationnorthward to Canada. Over the sum-mer, two or three more generations areproduced as each previous generationdies. Toward the end of August, butter-flies of the last summer generationenter reproductive diapause and thecycle begins anew as these monarchsmigrate instinctively southward to the

overwintering grounds in Mexico.To date, about 30 overwintering

areas have been discovered on 12 iso-lated mountain ranges in central Mex-ico at elevations ranging from 2900 to3400 m. This elevational band coincideswith a summer fog belt where boreal-like oryamel fir (Abies religiosa) forestsoccur that probably are a relict ecosys-tem from the Pleistocene. The fivelargest and least disturbed butterflyforests occur in an astoundingly smallarea of 800 km2. By clustering on thetrees in the cool and moist environ-ment, individual monarchs are able tosurvive in a state of reproductive inac-tivity until the following spring.

Until recently, human impact on thehigh-elevation fir forests has been lessthan that on other forest ecosystems in

Mexico. However, negative develop-ments began to occur in the 1970s.Commercial harvest of trees, includingthinning and clear-cutting hasincreased illegal removal of logs andfirewood, and local charcoal manufac-turing in pits dug within the fir forest.Villages have expanded, reaching loca-tions up the mountainsides, and thisexpansion has lead to an increased fre-quency of fires associated with forestclearing for planting corn and oats, andalso with killing young trees for localhome construction. Forest lepidopteranpests have invaded some areas of thefir ecosystem, probably due to stresscaused by thinning and deforestation atlower elevations. As a result of increas-ing lepidopteran pests, spraying of theorganic pesticide, Bacillus thuringiensis,

CampanarioCampanario

Chivati–HuacalChivati–Huacal

1971 1999

ChincuaChincua

AngangueoAngangueo

CampanarioCampanario

Chivati–HuacalChivati–Huacal

ChincuaChincua

AngangueoAngangueo

Campanario

Chivati–Huacal

Chincua

Angangueo

Campanario

Chivati–Huacal

Chincua

Angangueo

Figure B Forest cover has dropped dramatically within, and adjacent to, the three majormonarch butterfly reserves from 1971 to 1999. Forest is shown in dark gray, forest habitat that issomewhat degraded is shown in light gray, and highly modified areas are in white. Squares de-note any butterfly overwintering sites that have been recorded, but not all those shown persisttoday. For each reserve, the inner line denotes core areas and the outer line buffer zones. (Modifiedfrom Brower et al. 2002.)



was initiated and considered for wide-spread use without adequate knowl-edge of its effect on monarchs. Mostrecently, heavy ecotourism is tramplingthe infrastructure and generatingsevere dust precipitation on vegetationalong paths through the colonies;because of the income generated, thereis demand to open all colonies totourists.

Encroachment on existing monarchoverwintering sites, even those inreserved status, has been extensive inthe past two decades. A comparison ofphotos from 1971 and 1999 show shock-ing reductions in forests both outsideand inside some of the largest and mostimportant reserves (Figure B; Brower etal. 2002). The remaining fragments offorests experience a disrupted microcli-mate that will not support the butter-

flies once fragments become too smallor perforated via selective logging.

As a result of documenting thesedrastic habitat losses, in November2000, then President Ernesto Zedilloaccepted recommendations to reviseand increase the area protected to atotal of 56,259 ha and to redefine thearea as the “Monarch Butterfly Bios-phere Reserve” (Brower et al. 2002).Importantly, this new decree was asso-ciated with a trust fund to compensatelocal inhabitants for the profits theywould have made through logging.Notwithstanding this new decree, ille-gal logging has accelerated and isseverely degrading critical areas of thereserve (Galindo and Honey-Roses2004). Conservation of this migratoryphenomenon will not succeed withoutenforcement of logging bans together

with appropriate compensation toaffected people. Public education aboutthe values of protecting these forestsand the remarkable spectacle of naturethat the monarch butterfly migrationsrepresent is desperately needed.

If we fail to conserve these overwin-tering sites, we will not lose themonarch butterfly as a species, becausenumerous nonmigratory populationswill persist in its tropical range. How-ever, the monarch’s spectacular NorthAmerican migrations will soon bedestroyed if extensive overwinteringhabitat protection and management arenot implemented on a grand scale. Theimpending fate of this remarkableinsect is an omen, warning us that wemust incorporate the concept of endan-gered biological phenomena into ourplans for conserving biodiversity. n

dispersal ability, are critical factors in endangerment for alarge number of species as well (see Figure 3.6).

Where are species most at risk worldwide?Some biomes contain greater fractions of threatenedspecies, particularly biomes that are already species rich(Figure 3.7). Tropical and subtropical moist and dryforests, grasslands, savannas, shrublands, montanegrasslands, and xeric biomes all have substantial num-bers of threatened mammals, birds, and amphibians. Asglobal climate change intensifies, many worry that mon-tane habitats in particular will become unsuitable formany species, greatly increasing biodiversity losses.

Across the globe, there are particularly high numbersof threatened species in South America, Southeast Asia,sub-Saharan Africa, Oceania, and North America (IUCN2004). To a large degree, this distribution reflects thevastly greater numbers of species in the tropics, as wellas intensifying pressures, particularly in Southeast Asiaand sub-Saharan Africa. The greater numbers in NorthAmerica reflect both high diversity and levels of threatin the Hawaiian Islands, California, and Florida, as wellas extensive habitat modification, particularly for agri-culture, and substantial efforts to identify species at riskof extinction. Lower numbers of endangered species insome regions reflect low species diversity due to severeenvironments and low human densities (Antarctica, Sa-haran Africa, and to a lesser extent Northern Asia).

Endangered species in the United StatesAmong the first countries to address the problems ofendangered species seriously, the United States has list-ed 1264 species as threatened or endangered as of Feb-

ruary 2005 (USFWS 2005), and 1143 are on the Red List(IUCN 2004). In addition, more detailed data on the lo-cations and status of at risk species, populations, andecosystems has been collected on a state-by-state basisthrough Natural Heritage Programs (now expanded toinclude Canada and Latin America, and managed byNatureServe). The total list of at risk populations in theU.S. and Canada now includes over 15,500 occurrences(NatureServe 2005). Overall, a third of U.S. species areconsidered to be at risk of extinction (Stein et al. 2000),and the U.S. is second only to Ecuador in the number ofspecies thought to be at risk of extinction globally(IUCN 2004).

Although a greater number of threatened species inthe U.S. are plants (>5000), freshwater species (mussels,crayfish, stoneflies, and fishes) are threatened in higherproportions than any other group (Figure 3.8). Freshwa-ter mussel species appear to be particularly diverse inthe U.S., so the fact that 70% are imperiled is of globalimportance. Neither biodiversity nor the risk of extinc-tion is distributed uniformly across the U.S. California isthe state with the greatest species richness, as well as thegreatest number of endemic species, but Hawaii has thegreatest number of species at risk and the greatest num-ber of species that have become extinct (Stein et al. 2002).

Using the expanded list of imperiled species existingin the natural heritage database as well as federally listedspecies as of 1996, Wilcove et al. (1998) classified threatsfor 1880 species and populations for which sufficient in-formation was available. Over 85% of these were threat-ened by habitat degradation, while 49% were affected byinvasive species, 24% by pollution, 17% by overexploita-tion and 3% by disease (Table 3.3). Vertebrates, freshwa-

82 Chapter 3

© 2005 Sinauer Associates, Inc. This material cannot be copied, reproduced,manufactured, or disseminated in any form without express written permis-


Tropical/subtropical moist broad-leaved forest

Tropical/subtropical dry broad-leaved forest

Tropical/subtropical coniferous forest

Temperate broad-leaved and mixed forest

Temperate coniferous forest

Boreal forests/Taiga

Tundra

Mammals Birds Amphibians

Montane grassland and shrubland

Mediterranean forest, woodland, and scrub

Desert and xenic shrubland

Mangroves

Flooded grassland and savanna

Temperate grassland, savanna, and shrubland

Tropical/subtropical grassland, savanna, and shrubland

4000 4000Number of species

8000 2000 400020000 0 0

Figure 3.7 Species richness of mammal, bird, and amphibian species inthe major biomes of the world. Black portions of the bars indicate num-ber of threatened species. (Modified from Baillie et al. 2004.)

Freshwater Other Mammals Birds Reptiles Amphibians Fishes mussels Crayfish invertebrates Plants

Threat (85) (98) (35) (60) (213) (102) (67) (143) (1055)

Habitat loss and 89 90 97 87 94 97 52 96 81degradation

Overexploitation 45 33 66 17 13 15 0 43 10

Invasive species 27 69 37 27 53 17 4 46 57

Pollution 19 22 53 45 66 90 28 20 7

Disease 8 37 8 5 1 0 0 0 1

Note: Numbers in parentheses are the number of threatened species with threat data available.Source: Wilcove et al. 1998.aListed by taxonomic group.

TABLE 3.3 Percent of U.S. Threatened and Endangered Species Affected by Five Types of Threata

ter mussels, butterflies, and other invertebrates werethreatened particularly by habitat destruction, while pol-lution and siltation impacts were most severe for aquaticspecies. Disease has limited distributions of many popu-lations, and may have contributed to extinction of birdsin the Hawaiian Islands (Van Riper et al. 1986). Invasivespecies pose threats to a greater proportion of plant andbird species in the Hawaiian Islands than in the conti-nental U.S., and also to fish populations more than toother taxa.

Czech et al. (2000) examined the associations among18 causes of endangerment for 877 species that werelisted in the U.S. and Puerto Rico as of 1994. Urbaniza-tion was most frequently associated with other threats,as it involves both destruction of habitat directly anddepletion of resources to support urban populations.Urbanization is also associated strongly with pollution,nonnative species, and the development of roads. Roadsare not directly threatening to many species, but facili-tate many forms of development that are threatening,such as urbanization, logging, industrial development,mining, and agriculture. Agricultural development wasthe most widespread threat, and associated moststrongly with urbanization, reservoirs and pollution.

Threatened species in other countriesAlthough many countries maintain lists of threatenedspecies, few have been analyzed in detail. Based on RedList data, Ecuador and the U.S. have the greatest numberof endangered species (Table 3.4). Ecuador has a muchhigher number of listed plants than any other country.This has both a biological basis and a fortuitous one.Ecuador is a center of plant diversity where develop-ment, particularly in the Andean slopes has been ex-tremely rapid recently, and thus a substantial fraction ofevaluated plant species are threatened. Also, due to sub-stantial efforts by concerned botanists, more than 2000species have been evaluated—about half of those knownfrom Ecuador. Such concerted efforts at reviewingthreatened status have not been made for other centersof plant diversity, nor have they been undertaken forother taxa (for example, invertebrates) within the samecountry. Similarly, high numbers of species threatened inthe U.S. reflects both true threat levels and much greateractivity by its large cadre of biologists.

Although many countries do not harbor as long a listof globally threatened species, many have particularlyhigh fractions of their species at risk. For example, over80% of all plants and 30% of all vertebrates evaluated in




Perc

ento

fspe

cies

60

70

50

40

30

20

Fresh

water m

ussels

Crayfis

hes

Stonefl

ies

Fresh

water fis

hes

Amphibian

s

Flowerin

g plants

Gymnosp

erm

s

Ferns/

fern all

ies

Tigerbee

tles

Butterfl

ies/sk

ipper

s

Reptil

es

Dragonfli

es/dam

selfl

ies

Mam

mals

Birds

10

0

Figure 3.8 Proportion of speciesthreatened with extinction by plantand animal groups in the UnitedStates. (From Stein et al. 2000.)

Madagascar are at risk of extinction. Madagascar hasbeen subject to very high deforestation rates and otherpressures, and also has a very high number of endemicorganisms. Thus, the island nation is a focus of consider-able conservation interest (Case Study 3.2).

Australia lists 1554 threatened species, the majority ofwhich are vascular plants, followed by birds and mam-mals (Australian Government Department of Environ-ment and Heritage 2005), while only 621 of these are alsoRed Listed and few of these are plants. Habitat degrada-tion and invasive species are the most common factors inendangerment. Australia’s landmass is roughly equal tothat of the United States, but it has a very high degree ofendemism, with roughly 85% of flowering plants, mam-mals, and freshwater fishes unique to this continent.Thus, most threatened species in Australia representones at risk globally.

Recently, Li and Wilcove (2005) summarized threatsto vertebrates in China. Overexploitation is overwhelm-ingly the most important threat, contributing to endan-germent of 78% of 437 species, and habitat degradationwas also a pressing concern for 70% of these imperiledvertebrates (Figure 3.9). Nearly all endangered reptilesin China are overexploited for food and medicines. Thisis not surprising given the importance of traditionalmedicine for China’s population, and strong depend-ence on wild species for protein (see Chapter 8). Pollu-tion is a cause of endangerment in 20% of the cases, butas in other parts of the world, freshwater fishes are par-ticularly susceptible with over 40% threatened by indus-trial pollutants and pesticides. Dams do not yet appearas a major problem, perhaps because they are still rela-

tively new and unstudied. Invasive species appear muchless important, although this could be because they arelittle studied, or because global trade has a very long his-tory and the species most vulnerable to invasives havealready gone extinct (Li and Wilcove 2005).

Overall, understanding patterns of global endanger-ment, or even threatened status at a national level isfraught with uncertainty. These difficulties have led tosome controversy concerning the relative rate of extinc-tion currently compared with those occurring in geolog-ic history (Box 3.3). Although, these issues are not entire-ly resolved, the consensus is that current rates ofextinction already exceed “background” rates of extinc-tion, and are at least approaching those of the mass ex-tinctions.

84 Chapter 3


Mammals Birds Reptiles Amphibians Fishes Invertebrates Plants (%)a Total

Ecuador 34 69 10 163 12 48 1815 (81) 2151

United States 40 71 27 50 154 561 240 (63) 1143

Malaysia 50 40 21 45 34 19 683 (60) 892

Indonesia 146 121 28 33 91 31 383 (60) 833

China 80 82 31 86 47 4 443 (73) 773

Mexico 72 57 21 190 106 41 261 (69) 748

Brazil 74 120 22 24b 42 34 381 (68) 697

Australia 63 60 38 47 74 283 56 (33) 621

Colombia 39 86 15 208 23 0 222 (69) 593

India 85 79 25 66 28 23 246 (71) 552

Madagascar 49 34 18 55 66 32 276 (80) 530

Source: IUCN 2004 Red List Summary tables 5, 6a, and 6b.aPercent refers to the percent of those species evaluated that are threatened.bA complete revision of the list of threatened amphibians in Brazil has not been completed.

TABLE 3.4 Numbers of Threatened Species (CR, EN, VU) in 12 Countries with the Greatest Overall Numbers of Red Listed Species

Overex

ploitatio

n0

20

40

60

80

Habita

t loss

Pollutio

n

%ve

rteb

rate

spec

ies

affe

cted

Alien sp

ecies

Diseas

e

Figure 3.9 Contribution of major threats to endangermentamong vertebrate species in China. (From Li and Wilcove2005.)



BOX 3.3 Have We Set in Motion the Sixth Mass Extinction Event?

For a number of decades conserva-tion biologists have called atten-tion to the current rate of global

extinction in extreme terms. Estimatesof extinction rates vary, with mostranging from 50–10,000 times back-ground extinction rates seen in thegeologic record (Wilson 1992; May etal. 1995; Millenium Ecosystem Assess-ment 2005). To put our influences onbiodiversity in context, many havetried to compare the current rate ofextinction to the rates discernablefrom the fossil record. It is helpful tounderstand how these estimates aremade, to provide the most defensibleestimates for public discussion.

Before humans were dominant onEarth, the “background” rate of extinc-tions is estimated to have been about0.1–10 species per year (May et al.1995). During mass extinction events(see Figure 2.7), the rate of extinctionswas 1–2 orders of magnitude higher(Raup 1994). Because neither the num-ber of species on Earth during themass extinctions is known (nor pre-cisely how rapidly these extinctionsmay have occurred during a given10,000–1,000,000 year interval), thisestimate is highly imprecise, and thusdifficult to compare with current esti-mates of extinction rates. Further, 95%of fossils are of marine invertebrates,while about 85% of extant species areterrestrial (May et al. 1995). Howextinctions of marine invertebratescompared to extinctions in terrestrialand freshwater habitats is unknown.

Estimating current extinction ratesis surprisingly difficult as well. Thiscomes partly from the fact that it ismuch harder to prove that somethingdoes not exist, than that it does. Toprove a species is extinct one needs tosearch exhaustively throughout aspecies’ range, during appropriate sea-sons, and sometimes for many years.The recent rediscovery of the Ivory-billed Woodpecker (Campephilis princi-palis; Fitzpatrick et al. 2005) longbelieved to be extinct, illustrates thisdifficulty. As this is obviously impossi-ble to accomplish for all species onEarth, conservation biologists primarilyuse two methods to try to estimatecurrent extinction rates.

First, we can derive extinction rateestimates from species-area curves. Asdescribed in Chapter 2, species rich-ness increases with area. Species rich-ness therefore also declines withdecreases in area; thus as habitats areseverely degraded, we can expect tosee a concomitant loss of species. Con-servation biologists thus use estimatesof habitat change combined withthose for species richness to projectspecies losses (Figure A). The propor-tional reduction in species is then cal-culated as:

∆S = z ∆A

where S is the number of species, z isthe slope of the species–area curve fora given location, and A is the area ofthat location (May et al. 1995). The dif-ficulty is that while increasingly we canestimate changes in forest cover withaccuracy (∆A), we do not know therelationship between extinction andreduction in area (z). We also don’tknow the total number of species thatmay inhabit a given location, and thuswe further extrapolate when we trans-late proportional reduction to anabsolute number of losses per year.For example, tropical forest cover isbeing lost at a rate of 0.5%/year

(Achard et al. 2002), and if the value ofz is 0.2 (as is commonly assumed), thenwe can expect to lose 0.125% of thesespecies each year. If there are roughly5 million species in tropical forests,then we are losing 5000 species peryear. Of course, if the total number ofspecies is smaller or larger, or z ishigher or lower, we would get a differ-ent total (e.g., if z = 0.15 and only 1million species exist in tropical forests,then the loss is 750 species per year).Despite the crudeness of these esti-mates, reasonable values of theparameters yield substantially morethan 10 species per year (a high esti-mate of background extinction rates).

Second, extinction rate estimatesare often based on numbers of directlyobserved extinctions (i.e., recorded inthe IUCN Red List, or comparablenational databases). The Red List docu-ments 844 extinctions since 1500, aloss of 2.2% of all evaluated speciesand 0.04% of all described species.Certainly, this is an underestimate ofglobal extinctions (and even more oflocal extinctions) because only a fewtaxa and locations have been docu-mented well enough to allow a defini-

Sinauer Associates, Inc.Groom 3ePrinciples of Conservation BiologyGro0302aBox.eps100% of size05.17.05

–55%

–37.5%

1000

1009080706050

40

30

20

10

Log

num

ber

ofsp

ecie

s

Log area100 1000

z = .2

z = .35

Figure A Species–area relationships based on the equation S = cAz, with c = 10 and z = 0.2 or0.35. Note that a 90% decrease in area, from 1000 to 100 ha, would result in a predicted loss of37.5% to 55% of the species, for a z value of 0.2 and 0.35 respectively. Greater z values (steeperspecies–area relationships) imply greater species losses per unit area. Most z values calculatedfrom data fall between 0.15 and 0.35.

Continued on next page

86 Chapter 3


tive determination of extinction.Because we know the Red List total istoo low, attempts are made to correctfor this by extrapolating the rates forthe best-known group to all species.

For example, roughly 100 birds andmammals have gone extinct between1900 and 2000. To compare this to thefossil record, we need to calculate theloss as a percentage of all birds andmammals (currently 15,333 species).This translates to the loss of 0.65% ofall birds and mammals each century,and 1 bird or mammal species per year.Average extinction rates of birds andmammals in the fossil record are on the

order of 0.003 species per year (McKin-ney 1997). Thus, the current extinctionrate is 333 times greater than the back-ground rate of extinction.

By these and related methods ofestimation, we can conclude that thecurrent rate of extinction alreadyexceeds background rates of extinc-tion, and either already or will soonequal or exceed those observed in theother five mass extinction events ofEarth’s history. Even losses less extremethan those seen in the mass extinctionevents are abnormal; a loss of 30% ofthe species on Earth is roughly equiva-lent to what occurs on average once in

10 million years (Raup 1994). Yet, morethan 30% of amphibians, and over 40%of all species evaluated by the IUCN arethreatened with extinction today.

Perhaps more important than thefact that species are going extinct at agreat rate, should be the reflectionthat the time needed to recover levelsof species diversity similar to thoseprior the extinction event is exceed-ingly long—e.g., 5–10 million years forcoral reefs (Jablonski 1995). Ouractions are causing biodiversity lossesthat cannot be recovered in our chil-dren’s lifetime, or even that of ourgrandchildren or great grandchildren.

Box 3.3 continued

What Types of Species Are Most Vulnerable to Extinction?Although we do not know the absolute rate of extinctionof described and undescribed species today (see Box 3.3),we can describe with greater confidence which types ofspecies are most likely to go extinct. Many attributes canmake a species more vulnerable to extinction, includingrarity, narrow habitat range, large area requirements, lowreproductive rates, extreme specialization or coevolu-tionary dependencies. Further, species may be vulnerableto different threats, depending on their traits (Owens andBennett 2000; Isaac and Cowlishaw 2004). Reviewing ge-ological and historical patterns gives us insight intowhich species are most likely to go extinct in the future.Although a wide variety of possible factors may con-tribute to vulnerability for various species (see reviewsby Ehrenfeld 1970; Terborgh 1974; McKinney 1997), mostrelate either to aspects of specialization or rarity.

Species vulnerability due to specializationMany species, particularly in tropical moist forests, haveevolved very narrow ranges of environmental toleranceand highly specialized diets or habits. Often these speciesare at particular risk because such ecological specializa-tion lowers their resilience in the face of perturbations.Further, when these species are reliant on one or fewother species, their existence will be threatened shouldthese other populations decline or go extinct, as dis-cussed earlier for the case of plant species dependent onone or few species of pollinators or seed dispersers.

Top carnivores, with low densities, large body size, hightropic position, and large area requirements are often citedas being particularly vulnerable to extinction, particularly

through habitat degradation, and also overexploitation (di-rect or indirectly of their prey) (Purvis et al. 2000). Whilethis suite of factors certainly seems associated with greaterextinction risk, large body size does not seem to have ledto added risk for marine invertebrates (McKinney 1997),and other scientists have found stronger correlations be-tween extinction risk and other life history traits, such aslow rates of reproduction.

Because so many large-bodied species went extinct inthe Pleistocene, it has long been assumed that large bodysize per se carries higher risks of extinction—in part be-cause larger species were more valuable to hunters.While this has validity, close analysis reveals that extinc-tion risk for Pleistocene megafauna was greatest forspecies with low reproductive rates, regardless of bodysize (Johnson 2002). Thus, what seems to have happenedis that those species that could not replace themselves tomeet the pace of human hunting went extinct. Further,nearly all large-bodied, slow-reproducing mammalspecies that survived to the present day were nocturnal,arboreal, alpine or deep-forest dwellers, suggesting thatonly those species that could elude human hunters sur-vived to the present day (Johnson 2002).

Species with low reproductive rates typically can relyon adults to weather minor environmental variations,and thus have adapted to follow what is often called a“K-selected” life history strategy. These species are longlived, produce few young (which in some cases arecared for extensively by their parents), have high juve-nile survivorship, and abandon reproduction when en-vironmental conditions are poor to maintain high adultsurvivorship. Overall, these life history characteristicsmay make them unable to cope with the rapid changesoccurring today. Because the evolutionary response to


stresses is to reduce reproduction, such species will de-cline in abundance, eventually to extinction.

Certainly, the decline and extinction of marine mammalspecies follows this pattern. Larger species were huntedinitially, but these were also the species with the slowestmaturity and lowest reproductive rates, and thus unable toreplace themselves under industrial-scale exploitation(Roberts 2003). Many species of sharks, rays, and bonyfishes today that have lower reproductive rates also showsigns of greater vulnerability to exploitation, and thusgreater extinction risks (Jennings et al. 1998; Dulvy et al.2000). Often the most vulnerable species have later matu-ration, which also increases their susceptibility, as harvestbefore individuals have reproduced is thus more likely.

Vulnerability of rare species A species may be rare for many reasons: because of ahighly restricted geographic range, because of high habi-tat specificity, or because of low local population densi-ties, as well as due to a combination of these characteris-tics (Rabinowitz et al. 1986; Gaston 1994; Table 3.5). Oftena species may have attributes that seem to confer abun-dance, combined with ones that create rarity, and thus itis considered to be rare overall. For example, a speciesthat occurs across an entire ocean basic (broad geograph-ic range) and in high abundance, but is only found in afew isolated habitats such as deep-sea vents (high habitatspecificity), is considered a rare marine species. Similarly,some species are never abundant anywhere, althoughtheir range may be quite extensive across broad geo-graphic areas, as well as habitat types. Higher extinctionrates are correlated particularly with species with re-

stricted ranges and low density (Jablonski 1991; Gaston1994; Rosenzweig 1995; Purvis et al. 2000).

Species with highly restricted habitat preferences,such as a plant growing only within a rare soil type orseabirds that can only nest on cliffs with the correct pre-vailing winds, are particularly vulnerable to habitatdegradation, as once their habitat is destroyed they areunable to adjust to another location. Common speciesthat aggregate only in very particular locations, such asmany bat species that roost in caves, are not only vulner-able to habitat degradation, but can be overexploitedwith ease. Often, recognizing habitat restrictions chal-lenges us to transcend our experience to understandwhat environmental factors are critical to a marine in-vertebrate, an insect, or a fern.

Many of the highest estimates of future extinctionrates are derived from our observations of extremelynarrow ranges, or extreme endemism among manyspecies in the tropics—particularly plants and insects,most of which are undescribed species (see Box 3.3). Werarely know why these species are so narrowly distrib-uted, although many may be extreme habitat specialistsor simply may not have dispersed to other places wherethey could persist, and thus have a restricted range forhistorical reasons. However, many analyses have shownthat a restricted geographic range is the strongest pre-dictor of extinction risk (e.g., carnivores and primates,Purvis et al. 2000; birds, Manne et al. 1999).

To better understand how extreme endemism can im-peril species, consider an event documented in the mid-1980s. A group of scientists were doing RAP inventoriesin the western Andes of Ecuador, finding rapid turnover


Large Small

Broad Restricted Broad Restricted

Eve

ryw

her

esm

all

Som

ewh

ere

larg

e

Common

Locally abun-dant over a large range ina specific habitat

Constantly sparse over a large range and in several habitats

Constantly sparse in aspecific habi-tat but over a large range

Constantly sparse and geographicallyrestricted in several habitats

Constantly sparse and geographicallyrestricted in a specific habitat

Locally abun-dant in several habitats but restricted geographically

Locally abun-dant in a spe-cific habitat but restricted geographically

Geographic range

Habitat specificity

Pop

ula

tion

size

TABLE 3.5 Seven Forms of Species Rarity, Based on Three Distributional Traits

of plant species, with large numbers of endemics. On asingle ridge, Centinela, they found 90 endemic plantspecies. Just after their work was completed, the entireridge was cleared for agriculture, and the 90 plant specieswere forever destroyed (Dodson and Gentry 1991). Al-though it is possible that some of these species may befound in other locations, many are likely now globally ex-tinct, as well as the numerous associated animal species(especially insects) we can presume specialized on theseendemic plants. In the words of E. O. Wilson (1992),

[Centinela] deserves to be synonymous with silenthemorrhaging of biological diversity. When the for-est on the ridge was cut, a large number of rarespecies were extinguished. They went just like that,from full healthy populations to nothing, in a fewmonths. Around the world such anonymous extinc-tions—call them “centinelan extinctions”—areoccurring, not open wounds for all to see and rushto stanch but unfelt internal events, leakages fromvital tissue out of sight.

Island communities are typically rich in endemics, butalso poorer in species than comparable mainland com-munities. Low species richness on islands is attributed toa combination of low colonization rates of new species(the functional equivalent of a low rate of speciation), andhigh extinction rates (because populations are usuallysmall and subject to decimation by local catastrophes andstochastic variation) (MacArthur and Wilson 1967). Also,island communities have experienced extremely high ex-tinction rates of species during recent centuries, primari-ly due to anthropogenic introductions of mammalianpredators (mammals other than bats disperse poorlyacross ocean barriers) and mainland diseases.

Many groups are more endangered on island systemsthan on mainlands. In the new world, amphibians aremuch more endangered among the islands of theCaribbean (84% are at risk of extinction) compared withthe adjacent continental landmasses of Mesoamerica(54%), South America (31%), and North America (21%;Young et al. 2004). As discussed earlier, avian extinctionshave been much more common on islands, particularlydue to combination of pressure from rapid habitatchange, overexploitation, and invasive species that mayact in an ecological role not represented on the island(e.g., introduced mammalian predators), and thus repre-sent a challenge outside the evolutionary experience ofmany island species. However, carnivores and primatesare no more likely to be endangered on islands than oncontinents (Purvis et al. 2000).

Small populations are generally more vulnerable toextinction than large ones. This has been seen in severalcomparisons of faunal surveys taken 50–100 years apart.Nearly 40% of small populations of birds (with less than10 breeding pairs) went locally extinct in the CaliforniaChannel Islands over an 80-year period compared with

only 10% of those populations with 10–100 pairs, only 1population with 100–1000 pairs, and none of those withgreater than 1000 pairs (Jones and Diamond 1976).

In the Baltic Sea, intense harvests of skates havecaused the smallest populations to disappear, while larg-er ones are still extant (Dulvy et al. 2000). Many specieswith small populations may belong to species with largebody size, which is often correlated with slower repro-ductive rates and other life history attributes that in-crease vulnerability to extinction.

Widely dispersed species that are always rare may bevulnerable to local extinction events, although they areless likely to become globally extinct. Some marine in-vertebrates may fall into this category, as well as popula-tions of top carnivores. Although such species may notbecome globally extinct, local extinctions are potentiallyvery damaging to communities and ecosystems, as dis-cussed earlier.

We may need to worry most about artificial rarity,where a once widespread or abundant species has beenreduced to a very small population size through humanactivities. Species that are artificially rare seem more likelyto go extinct than species that are evolutionarily adaptedto rarity (Kunin and Gaston 1997). Thus, small populationsize alone may be less predictive than a sudden reductionto small population size (which is one rationale for the cri-teria for listing species in the Red List; see Box 3.2).

“Bad luck”: Extrinsic causes of extinction due tohuman activitiesIn addition to these categories, we might also add one forspecies who simply have “bad luck” (Raup 1991). Thesespecies are not intrinsically vulnerable due to their traits,but rather have the misfortune to be in the wrong place atthe wrong time, or of being particularly palatable to hu-mans. That is, living in areas of greatest human impact,such as on arable soils, in river systems that are heavilyused for commerce and subject to substantial pollution,or along coasts where human populations congregate,will thrust species that share those locations directly intoharm’s way. For example, freshwater fishes in the mostextensively altered ecosystems may all share an increasedlikelihood of extinction, despite having a wide variety oftraits (Duncan and Lockwood 2001). Purvis et al. (2000)estimated that roughly half of the variation in extinctionrisk among carnivores and primates related to their ex-posure to anthropogenic disturbances, particularly tohigh rates of habitat loss or intense commercial or bush-meat exploitation.

Environmental changes may occur very rapidly, asthey appear to have in the mass extinction events of ge-ologic history, and then vulnerability to extinction maybe widely shared among species (Raup 1991). The inten-sification of human pressures on ecosystems may havethe same effect as the environmental changes of the past.

88 Chapter 3


Our need to grow enough food to feed the increasinghuman population, particularly if large proportions eatmeat, requires greater conversion of land for agriculture(see Chapter 6). Increasing rates of exploitation may hitthose species with low reproductive rates hardest, but tosome degree all species that are marketable may be en-dangered as local demand for bushmeat, for example,explodes with local human population density (see CaseStudy 8.3), or as commercial fisheries strive to meetworld demand for protein (see Chapter 8). Global tradeis increasing transport of species (see Chapter 9), andtherefore increasing the chance that invasive species willtake hold in some new place. Finally, our consumptionof fossil fuels and other practices that drastically increasegreenhouse gases has begun to change our climate (seeChapter 10)—which may be the worst luck of all.

An awareness of factors that enhance vulnerability toextinction can help us formulate plans to counteractthese factors to some degree. Understanding which traitsincrease vulnerability also may help us predict whichspecies will most need protections as human pressuresincrease in areas where they have been slight. However,it is also possible that increasingly species will best becategorized as simply having the bad luck to be in thepath of human development.

Economic and social contexts of endangerment Underlying threats caused by human development arepowerful economic and social drivers. Perhaps most ofall, economic growth and rising affluence do the most toincrease the pace of habitat conversion, overexploitationof marine species for food, pollution of waterways, andtransformation of our atmosphere that is changing ourclimate. In the U.S., causes of endangerment are thoseprimarily associated with rapid economic growth—ur-banization and agricultural expansion, infrastructure de-velopment (roads, reservoirs, wetland modification),and the byproducts of these activities (pollution, habitatdegradation) (Czech et al. 2000). Moreover, geographiccenters of endangerment currently occur where high di-versity meets rapidly expanding economic activity insouthern California, east-central Texas, and southernFlorida (Dobson et al. 1997; Czech et al. 2000). As Czechet al. (2000) put it, “The list of endangered species isgrowing because the scale of the integrated economy,and therefore the causal network of species endanger-ment, is increasing.”

At the other end of the economic continuum, the ma-jority of people struggle in poverty (UN Millennium Pro-ject 2005). Extreme poverty afflicts over 1 billion people,who live on less than $1/day, are chronically hungry, lackclean water and sanitation, are burdened with preventa-ble diseases, and often lack shelter and sufficient clothes.An additional 2.7 billion live on scarcely $2/day, and arejust able to meet their basic needs, but no more. Thus, it is

no wonder that unsustainable levels of burning, small-scale agriculture, grazing, and bushmeat hunting occurwherever these practices help the poorest people to sur-vive. Biodiversity losses are clearly associated with thishuman tragedy, but even more serious are the signs thaterosion of ecosystem services increasingly undercut theability of human populations to escape extreme and mod-erate poverty (Millennium Ecosystem Assessment 2005).

Further, globalization spreads the dangers of econom-ic expansion as countries prioritize international trade innatural resources to enhance their economies. This canexacerbate poverty where trade is not tied to sustainablepractices that improve the lives of local people, but ratherprovide cheap goods that fuel the economies of devel-oped nations. Often, pressures to overexploit and trans-form land come principally from decisions of large multi-national corporations, and consumer demand in far-offcountries. In addition, globalization also has radically ac-celerated the spread of nonnative species. Rapid eco-nomic expansion in China and other densely populatedcountries has raised concern that pressures on develop-ing countries will intensify sharply in the near term (Mil-lennium Ecosystem Assessment 2005).

Responses to the Biodiversity CrisisClearly, we live in a time of crisis for biodiversity. Whilemany fine points can be debated, and some uncertaintiesremain, it is certain that humans have wrought radicalchange throughout the globe. Moreover, as our popula-tions grow and consume more, ecosystems will continueto degrade, more species will go extinct, and human suf-fering on a staggering scale will continue. How can werespond to crises of this magnitude? Conservation biol-ogists and promoters of human development agree thatsolutions will need to come from a mixture of (1) scien-tific analysis of and communication about the drivers ofchange in biodiversity and human welfare, (2) techno-logical improvements, (3) legal and institutional instru-ments, (4) economic incentives and plans, and (5) socialinterventions. The mechanisms for achieving solutionswill include preventative measures, such as establish-ment of protected areas (see Chapter 14), targeted inter-ventions at genetic, species, and ecosystem levels that in-tegrate ecological understanding with community-basedproblem solving (see Chapters 11–13), restoration ofdamaged ecosystems or endangered populations (seeChapter 15), and creation of truly sustainable forms ofdevelopment (see Chapter 16).

Most approaches to conservation focus on species,ecological communities or ecosystems, or landscapes, al-though some also focus more broadly (and vaguely) onbiodiversity overall (Redford et al. 2003). Because biodi-versity conceptually spans the range from genes toecosystems, and spatial scales from single ecological



communities to landscapes and larger, effective conser-vation is most likely to emerge from multiple approach-es undertaken at a range of spatial scales, directed to-ward different conservation targets (species, ecosystemsor landscapes). Conservation approaches focus from sin-gle species in a single site to all species across the globe,including species or ecosystem targets, and often explic-itly including human communities and development(Redford et al. 2003).

Conservationists recognize that species approachesalone will not be sufficient to conserve biodiversity, andthat actions on larger spatial scales targeting ecosystemsand landscapes may be more efficient and effective asecological processes will be more likely to remain intactwhen conservation is undertaken at these scales (Redfordet al. 2003). However, conservation is usually achieved bymany smaller actions, undertaken at smaller spatialscales. Overall, to address the numerous threats to biodi-versity, conservationists need methods to prioritize con-servation activities, and plans for how to achieve conser-vation at various scales. As Redford et al. (2003) put it,conservationists’ generally are asking “where” questionsto set geographical priorities, and “how” questions aboutdeveloping and implementing strategies to conserve con-servation targets at priority places.

The remainder of this section focuses on just two is-sues in promoting solutions—the use of national and in-ternational agreements to address our biodiversity crisis,and the issues involved in providing clear indicators ofproblems and of progress. The focus here will remainmore on actions that are centered on species, whereas adiscussion of approaches aimed more at habitat protec-tion is included in Chapter 6, and longer treatments ofapproaches to solving the biodiversity crisis are includ-ed in Chapters 11–18.

Laws and international agreements that address biodiversity lossForemost among efforts to conserve biodiversity are na-tional laws and international agreements that limithuman impacts to threatened species and ecosystems,and establish a mandate for governmental and citizenaction. Major international agreements, the landmarkU.S. Endangered Species Act (ESA), and other key U.S.laws that protect biodiversity are described by DanielRohlf in Case Study 3.3. These institutional policy in-struments are most effective when they reflect ecologicaland evolutionary understandings (as in the ESA, whereprotecting distinct population segments was given im-portance for future evolution and local community in-teraction), are well linked to human activities that can beamended, and have adequate enforcement

Many of the laws to protect biodiversity do so via listsof threatened species, as described earlier. In a broad

sense, these lists are used to set priorities for species pro-tection, although of course many other factors—includ-ing economic and political ones—determine whichspecies are targeted for conservation actions (Czech andKaufman 2001). Laws and international agreementsmore effectively achieve conservation goals when theyreduce the tendency for economic and political pressuresto be paramount in decision-making, instead requiring afair examination of potential harms and benefits to bio-diversity. This may be accomplished via effective en-forcement (e.g., the provision for citizen lawsuits in theESA, or for severe trade restrictions to accompany failureto abide by animal trade prohibitions under CITES; seeCase Study 3.3), and by mechanisms that promote eco-nomic incentives and disincentives (as are being devel-oped under the Convention on Biological Diversity), orthat increase public involvement in solution creation(e.g., Safe Harbor agreements under ESA).

Some conservationists have cautioned against usingthreatened species lists as too literal a guide for priori-ties. Because so many groups are poorly known, there isunder-representation for many taxonomic groups (e.g.,nearly all groups of invertebrates and nonvascularplants) and in areas of the world (most of the develop-ing world, and most marine habitats). Prudence sug-gests we should be very cognizant that the biases in ourknowledge could lead us to ignore taxa, or indeed otherlevels of biodiversity besides that of species that may becritical conservation targets (Possingham et al. 2000;Burgman 2002). Thus, often policies that are directed ex-plicitly toward broader protections for habitat ratherthan species, or toward enhancing sustainable practicesmay be preferred to avoid such biases. However, usual-ly conservation is not an either-or venture, but rather re-quires a suite of complementary approaches, some ofwhich include admittedly imperfect policy instrumentsthat motivate and guide action at many levels.

International agreements are meant to serve as moti-vators to action. At the 2002 Johannesburg World Sum-mit on Sustainable Development, 190 countries commit-ted to the reduction of biodiversity losses significantlyby 2010. This in turn motivated efforts to develop morespecific policies for conservation on global, national, andlocal levels. Further, this has motivated an examinationof how we can track both problems and progress.

Identifying driving factors and trends in species endangermentTo create an effective plan to recover declining species, orto address biodiversity loss more broadly in some regionof the world, a necessary step is to use rigorous scientificmethods to determine which factors most influence popu-lation decline, and address our conservation actions to re-move or ameliorate these threats (Caughley and Gunn

90 Chapter 3


1996). While to some degree this is obvious, Caughley andGunn found numerous examples where assumptionshave led well-meaning conservationists astray. Deducingwhich causal factors drive declines should involve clearlydelineating a plan of analysis to separate potential threats;this plan should include setting up testable hypothesesand predictions that follow from these hypotheses, and re-viewing all evidence to evaluate each hypothesis carefully.To do this it is necessary to learn enough about the species’natural history to develop reasonable hypotheses. Com-monly, the agent or agents of decline are not obvious, soconservation scientists have to become good detectives.

Beyond assessing the drivers of endangerment in par-ticular cases, it is critical to track status trends to set priori-ties for intervention, and also to know when we are suc-ceeding in protecting species. The IUCN has kept data onspecies endangerment for many years, and can now trackthe progress of the best-studied groups. The Red List Index(RLI) tallies changes in status due to either a deteriorationor improvement of all threatened and near-threatenedspecies since 1988 (Butchart et al. 2004). As applied tobirds, RLI analyses show a portrait of steady deteriorationof status (e.g., moving from vulnerable to endangered)(Figure 3.10). The RLI for birds has decreased by nearly 7%since 1988, roughly equivalent to 10% of all species in NT,VU, EN, and CR moving up into the next higher categoryof threat. For albatrosses and petrels the decline has beenparticularly steep, equivalent to a 25% drop in RLI since1995, indicating a very perilous situation. A retrospectiveanalysis for amphibians also indicates steep declines ofroughly 15% in RLI (Baillie et al. 2004).

Most species groups are too poorly known to ade-quately evaluate trends, but a variety of indices helpbridge this gap. These range from the index of biotic in-

tegrity that is used to assess stream health (Karr 1991), tocomposite indices that attempt to summarize globalchanges for the general public, business community, andgovernment leaders. One of these composite indices, theLiving Planet Index (LPI), summarizes change over timein populations of over 1100 terrestrial, freshwater, andmarine vertebrate species (Loh et al. 2005). Because theyare much better known, most of the data are drawn frombird and mammal species, on land in temperate climes.At present, these are the best data available, but it maybe possible to weight the data to compensate for thesebiases as more data are collected on currently under-rep-resented and little known groups (Loh et al. 2005). TheLPI shows that terrestrial vertebrates have declined by25% between 1970 and 2000 (Figure 3.11A). Further,freshwater species have declined more sharply than ei-ther terrestrial or marine vertebrates (see Figure 3.11B).Both these trends are disturbing. Importantly, the resultsof the LPI are discussed internationally, and are succeed-ing in conveying a straightforward and compelling mes-sage about overall trends.

Increasingly it is helpful to examine indicators of futurechange based on economic and social parameters (see also



1988

100

Red

listi

ndex

(set

to10

0in

1988

)

95

90

85

80

751992 1996

Albatrosses and petrels

All birds

2000 2004

Worse

Better

Year

Figure 3.10 The Red List Index (RLI) shows a worsening ofthreatened status between 1988 and 2004 for all Red Listedbird species and a precipitous decline for albatrosses and pe-trels. (Modified from Butchart et al. 2004.)

0.4

0.2

1970 1975 1980 1985Year

1990 1995

T

MFW

2000

0.6

0.8

1.0

1.2

1.4

(A)

Liv

ing

plan

etin

dex

0.4

0.2

0.6

0.8

1.0

1.2

1.4

(B)

Liv

ing

plan

etin

dex

Figure 3.11 The Living Planet Index (LPI) tracks populationtrends for over 1100 terrestrial, freshwater, and marine verte-brates. (A) Terrestrial vertebrates have declined 25% from1970 to 2000. (B) Freshwater (FW) vertebrates have declinedmore than either terrestrial (T) or marine (M) vertebrates.(Modified from Loh et al. 2005.)

Chapter 18). McKee et al. (2003) developed a stepwise re-gression model that predicted 87.9% of the variance in thedensity of threatened mammal and bird species based onhuman population densities and species richness in 114countries. Because the model showed no bias, they used itto project potential impacts of increases in human popula-tion density on future levels of threat. If human popula-tions grow as predicted by the United Nations and the cor-relations between human density and density ofthreatened mammals and birds do not change, then thenumber of threatened mammals and birds should increaseby 7% by 2020 and by 14.7% by 2050 (McKee et al. 2003).While it does not direct specific action, the model can beused to help convince decision makers that interventionsare needed now, and that delay has serious consequences.

The most comprehensive survey of the current statusof biodiversity as it supports human life was made by aconsortium of thousands of scientists in the MillenniumEcosystem Assessment (MA), completed in March 2005.Their conclusions are perhaps the most sobering of all.Although we have substantially increased food produc-tion worldwide, nearly all other indicators of ecosystemservices have declined sharply since 1950, as humanpopulation grew from 2.5 to nearly 6.5 billion, and eco-nomic growth worldwide increased nine-fold (and in theU.S., twenty-five-fold). Threatening processes are ex-

pected to intensify, particularly habitat loss and degra-dation via pollution, species invasion, and climatechange (Figure 3.12). Inland and coastal waters andgrasslands have already born the greatest impacts, andare expected to continue to be subject to intense pres-sures. The results of the MA should serve as a particu-larly strong wakeup call that, unless we wish to leave asubstantially impoverished and dangerous world to ourchildren, we need to make far-reaching changes in ourglobal, national, and local behavior now.

To achieve our goals for conserving biodiversity, wewill be well served by working to understand better thefactors that drive human behavior, and the means thatcan effectively change destructive behaviors. JeffreySachs, leader of the UN Millennium Development Pro-ject recently noted that the world’s poorest people donot lack the will to follow more sustainable practices,but rather they and their governments lack the means(Sachs 2005). Human values, needs, and desires arepowerfully, and quite imperfectly, translated into eco-nomic constructs that guide public policy and individ-ual actions. One key to a long-term solution to the bio-diversity and human poverty crises is to find ways totranslate our ethics and the importance of biodiversityinto our economics, and thus more integrally into ourdecision making.

92 Chapter 3


Forest

Boreal

Habitatchange

Pollution(nitrogen,phosphorus)

Over-exploitation

Invasivespecies

Climatechange

Temperate

Tropical

Dryland

Inland water

Coastal

Marine

Island

Mountain

Polar

Temperate grassland

Mediterranean

Tropical grasslandand savanna

Desert

Figure 3.12 Projected trends inthreatening processes in differenthabitat types in the comingdecades. Shading indicates the in-tensity of each process’ impactson biodiversity over the past cen-tury, with dark gray indicatingthe most impact and white theleast. Straight upward arrows in-dicate very rapid increase in in-tensity of this threat at present; di-agonally upward facing arrowsindicate increases in the threat;sideways arrows indicate contin-uing levels of the threat; anddownward facing arrows indicatedeclines in threat intensity. (Modi-fied from Millennium EcosystemAssessment 2005.)


CASE STUDY 3.1Enigmatic Declines and Disappearances of Amphibian Populations

Joseph H. K. Pechmann, University of New Orleans, and David B. Wake, University of California, Berkeley

The First World Congress of Herpetology in 1989 and a Na-tional Research Council workshop in 1990 brought the acceler-ating losses of amphibian biodiversity to the forefront of con-servation concerns. It long had been obvious that habitatdestruction and alteration, introduction of nonnative species,pollution, and other human activities exact an increasinglyheavy toll upon amphibians as well as on other taxa. By thetime of these meetings, however, herpetologists realized thatsome of the declines and disappearances of amphibian popu-lations and species were unusual in one respect: The causes oftheir loss were undetermined. These unexplained losses oc-curred in isolated areas relatively protected from most humanimpacts, particularly in montane regions in tropical/subtropi-cal Australia, the western United States, Costa Rica, and otherparts of Latin America. Especially noteworthy, detailed retro-spective studies of declines in amphibians in large nationalparks in California (Drost and Fellers 1996; Fellers and Drost1993) and in the Monteverde Cloud Forest Preserve in CostaRica (Pounds et al., 1997) presented the first conclusive evi-dence of community-wide declines of amphibians.

Population declines and range contractions have continued.For example, during the latter 1990’s 35 of 55 species disap-peared from a study area in the Fortuna Forest Reserve inPanama (Lips 1999; Young et al. 2001). However, many of these

cases are complicated by the fact that not all sympatric am-phibian species appeared to be affected.

A general problem with these reports was the lack of per-spective, with a focus on what may be special cases. To rectifythis situation, about 500 amphibian biologists were enlisted toparticipate in a Global Amphibian Assessment (GAA) con-ducted from 2000 to 2004 (Stuart et al. 2004). The assessmentattempted to evaluate every species of amphibian across theplanet, a challenging task when one considers the fact that thevast majority of species are tropical and most of these havebeen little studied. The GAA documented that the reportswere representative of a widespread phenomenon, and thatthe problem is more acute than generally had been thought;32.5% of the known species of amphibians are “globallythreatened” (Figure A). The assessment also showed that com-munity-wide amphibian declines were more likely to be en-countered in the tropics. Many species showed no evidence ofdeclines, however.

The causes of many documented declines and disappear-ances remain poorly understood despite a burgeoning literatureon the subject. Nearly half of the 435 amphibian species classi-fied as rapidly declining by the GAA are threatened primarilyby “enigmatic” (unidentified) processes (Stuart et al. 2004). Inthis case study we briefly review the taxonomic and spatial pat-

terns of these enigmatic losses andtheir major hypothesized causes. Al-though the enigmatic declines are ourfocus, well-documented threats such ashabitat loss and overharvesting that arethe primary cause of most losses of am-phibian biodiversity (including overhalf of the rapid declines, Stuart et al.2004) require attention as well.

Although amphibians are morethreatened than some other groups,their plight is not unique (Gibbons etal. 2000). The case study of amphibiandeclines illustrates several themes ap-plicable to studies of biodiversity loss-es in all taxa:

1. Biodiversity loss has many causesthat are not mutually exclusive;several factors may act simultane-ously on one population.

Extinct/Extinct in the Wild0.6% (35 species)

Critically Endangered7.4% (427 species)

Endangered13.3% (761 species)

Vulnerable11.6% (668 species)Near Threatened

6.3% (359 species)

Least Concern38.4% (2203 species)

Data Deficient22.5% (1290 species)

Figure A Over one-third of amphibian species have a threatened designation under the IUCNRed List. (From Global Amphibian Assessment 2004.)


2. Species, and populations within species, vary withrespect to the importance of different threats.

3. The combined effect of several stressors can be worsethan the sum of their individual effects. These syner-gistic interactions are especially difficult to study.

4. Experimental demonstration that a factor may affect apopulation is not proof that it is responsible forobserved population declines.

5. No area on Earth is protected from human activities.

Species declining for enigmatic reasons are concentrated inthe tropics, especially in Mesoamerica, the northern Andes,Puerto Rico, southern Brazil, and eastern Australia, althoughthe phenomenon may be underestimated in poorly-monitoredregions such as Africa (Stuart et al. 2004). Most species (82%)having unexplained declines are frogs and toads in the fami-lies Bufonidae, Hylidae, and Leptodactylidae. Ecologically,enigmatic declines are associated with species found in forests,streams, and tropical montane habitats (Stuart et al. 2004). Avariety of hypotheses as to the underlying causes of these enig-matic declines have been proposed and studied by a host ofconservation scientists.

Ultraviolet-B RadiationIncreased ultraviolet-B (UV-B) radiation is a possible explana-tion for declines and losses of amphibian populations in com-paratively undisturbed areas. A large body of experimental ev-idence indicates that ambient levels of UV-B are harmful tosome amphibian species at some locations, although effectsmay vary among populations, species, and geographic regionsand with ecological factors such as elevation, water chemistry,and the presence of other stressors (reviewed by Blaustein andKiesecker 2002; Blaustein et al. 2003a). For example, egg mor-tality in the western toad and Cascades frog was higher wheneggs were exposed to ambient levels of sunlight in naturalponds in Oregon than when 100% of UV-B was blocked(Blaustein et al. 1994a; Kiesecker and Blaustein 1995). UV-B hadno effect on the survival of Pacific treefrog eggs in the same ex-periments. The Pacific treefrog was found to have higher levelsof an enzyme, photolyase, that facilitates the repair of DNAdamaged by UV-B (Blaustein et al. 1994a). Ambient UV-B didnot affect survival of western toad eggs in a field experimentconducted in Colorado (Corn 1998). Differences between theOregon and Colorado results could be due to differences ingenes, ecological conditions, or experimental design. UV-B hada greater effect on survival of long-toed salamander larvae fromlow-elevation populations than those from high-elevation pop-ulations under identical conditions in the lab (Belden andBlaustein 2002).

There is also much experimental evidence that UV-B has agreater effect on amphibians when combined with other stres-sors (Blaustein and Kiesecker 2002; Blaustein et al. 2003a). Forexample, UV-B decreased the survival of Rana pipiens eggs inacidic water but not at near-neutral pH (Long et al. 1995). In

some experiments western toad and Cascades frog eggs diedas a result of UV-B increasing their susceptibility to a patho-genic fungus, Saprolegnia (Kiesecker and Blaustein 1995;Kiesecker et al. 2001a).

This body of experiments helps us understand some of theimpacts of UV-B on amphibians, but does not show that UV-Bis responsible for population declines and disappearances. Itremains unknown whether the affected populations were ex-posed to increased UV-B, or that the effects of UV-B extendfrom individuals to the population level.

Ozone depletion is one factor that may have increased ex-posure of amphibians to UV-B, even in some locations at lowlatitudes (Middleton et al. 2001). Many other factors also affectthe UV-B dose experienced by amphibians, including the atten-uation of UV-B by water and by dissolved organic matter inwater, breeding phenology, and behavior (e.g., Corn and Muths2002, Palen et al. 2002). Climate warming and anthropogenicacidification may reduce concentrations of dissolved organicmatter and thereby increase the depth to which UV-B pene-trates aquatic habitats (Schindler et al. 1996; Yan et al. 1996).

Ecologists disagree on interpretation of research on the ex-tent to which natural levels of dissolved organic matter cur-rently protect amphibians from harmful levels of UV-B(Blaustein et al. 2004; Palen et al. 2004). Global climate changecan affect precipitation, and consequently water depths andUV-B exposure (Kiesecker et al. 2001a). In dry years eggs willbe closer to the surface of the water and thus more exposed toUV-B. A complicating factor is that in some regions, such as theRocky Mountains, montane snow melt and amphibian breed-ing occur earlier in the season in dry years, when UV-B radia-tion is lower (Corn and Muths 2002). These issues illustratewhy it is difficult to evaluate possible relationships betweenglobal climate change and UV-B exposure (Blaustein et al.2004; Corn and Muths 2004). Integration of information onlethal and sublethal effects of UV-B throughout an amphibian’slife cycle with demographic models will bring the phenome-non into a population dynamics framework (Biek et al. 2002;Vonesh and De la Cruz 2002).

Disease PathogensPathogens may cause declines and disappearances of amphib-ian populations. There was little evidence for this hypothesisuntil a pathogen new to science, Batrachochytrium dendrobatidis(a chytrid fungus), was found in dead and dying frogs collect-ed in the mid-1990’s from declining populations in Australiaand Panama (Berger et al. 1998; Longcore et al. 1999). B. den-drobatidis attacks only tissues that contain keratin, which in-clude the mouthparts in anuran (frog and toad) tadpoles andthe skin in salamanders and metamorphosed anurans (Bergeret al. 1998; Bradley et al. 2002; Davidson et al. 2003). Researchto date suggests that chytrid infections cause little or no mor-tality of anuran tadpoles or salamanders, although it can de-crease the growth rates of the former, making them more sus-ceptible to predators or other stressors (e.g., Davidson et al.

94 Chapter 3


2003; Parris and Beaudoin 2004). B. dendrobatidis can kill meta-morphosed frogs, either by producing a toxin or by interferingwith skin functions such as respiration and osmoregulation(e.g., Berger et al. 1998; Rachowicz and Vredenburg 2004).

Batrachochytrium dendrobatidis has been isolated from otherfrog populations undergoing mass mortality events associat-ed with population declines and disappearances, includingRana yavapaiensis, R. chiricahuensis, and H. arenicolor in Arizona(Bradley et al. 2002), the midwife toad (Alytes obstetricans) inPeñalara Natural Park, Spain (Bosch et al. 2001; Martinez-Solano et al. 2003b), and (retrospectively, using preservedspecimens) several species in Las Tablas, Costa Rica (Lips etal. 2003a). It has also been isolated from populations notknown to be in decline, including tiger salamanders in Ari-zona (Davidson et al. 2003), Litoria wilcoxii or jungguy (taxon-omy uncertain) in Australia (Retallick et al. 2004), and Xenopuslaevis in southern Africa (Weldon et al. 2004). Many individu-als in these populations had light infections and appearedhealthy. Antimicrobial peptides located in the skin may pro-vide some species with natural defenses against B. dendroba-tidis (Rollins-Smith et al. 2002a, b). Environmental conditionsin some locations may not be favorable for the fungus. For ex-ample, B. dendrobatidis can grow and reproduce at tempera-tures of 4ºC–25ºC, whereas growth ceases at 28ºC and 50%mortality occurs at 30ºC (Piotrowski et al. 2004). This may ex-plain why amphibian population declines and disappear-ances associated with chytrids have been observed in cooltropical montane areas (Berger et al. 1998; Lips et al. 2003a),but not in tropical lowlands where temperatures are oftenabove 30ºC.

Several scenarios for the association between chytrid fun-gus infection and population declines and disappearances arepossible: (1) amphibians have long coexisted with B. dendroba-tidis, and observed population changes are cyclical phenome-non that previous researchers may not have noticed; (2) am-phibians have long coexisted with B. dendrobatidis, butenvironmental change or stress has made them more suscepti-ble to the fungus or the fungus more pathogenic to the am-phibian; and (3) B. dendrobatidis is a novel disease recently in-troduced to susceptible populations around the world throughhuman activities such as the pet trade.

Scenario 1 would follow that of many wildlife diseases, in-cluding ranaviruses in North American tiger salamanders andin United Kingdom common frogs (Daszak et al. 2003). Thisscenario is considered unlikely for amphibian chytridiomyco-sis because of the large magnitude of the population changesand the lack of recovery in many cases (Daszak et al. 2003).

There are many ways in which an established coexistence be-tween chytrids and amphibians may have changed over time, asin scenario 2. For example, increases in UV-B (Kiesecker andBlaustein 1995), exposure to pesticides (Taylor et al. 1999;Gilbertson et al. 2003), or other stresses can result in immuno-suppression and disease emergence. Climate change can inducedroughts, causing amphibians to aggregate around water bod-

ies and increasing their exposure to waterborne diseases such asB. dendrobatidis (Pounds et al. 1999; Burrowes et al. 2004). Im-munosuppression would be likely to increase the prevalence ofmany diseases, however, not just chytridiomycosis.

Several pieces of data are consistent with the hypothesis thatB. dendrobatidis is a novel pathogen that has recently beenspread around the world with the assistance of humans. Forfour cases where B. dendrobatidis was associated with popula-tion declines and disappearances, chytrids could not be detect-ed in museum samples collected prior to the population crash(Berger et al. 1998; Fellers et al. 2001; Lips et al. 2003a). Diseasewould have to have been very prevalent to allow detectionfrom the small number of samples that were available, howev-er (Lips et al. 2003a). Little genetic variation has been found sofar in B. dendrobatidis collected around the world, although ad-ditional work is needed (Morehouse et al. 2003). Two amphib-ians that have been transported all over the world, the Ameri-can bullfrog (Rana catesbeiana) and the African clawed frog(Xenopus laevis), are carriers of B. dendrobatidis (Daszak et al.2004; Hanselmann et al. 2004; Weldon et al. 2004). The sudden,catastrophic nature of some declines and disappearances alsosuggests an introduced pathogen (Daszak et al. 2003).

If the chytrid is novel to most areas, where might it haveoriginated? Some workers suggest it is endemic in populationsof Xenopus in Africa (Weldon et al. 2004). The earliest docu-mented case of chytridiomycosis was in a X. laevis collected inSouth Africa in 1938 (Weldon et al. 2004). The chytrid seems tohave a benign relationship to this species, and X. laevis is apopular laboratory animal exported worldwide beginningwith its use in pregnancy tests in the 1930s.

Introduced SpeciesPredation and competition from introduced species other thanpathogens may have caused declines of some amphibian pop-ulations in isolated, seemingly protected areas. For example,the introduction of trout for sport-fishing is thought to havebeen an important factor in some disappearances of frogs in theSierra Nevada of California. All but 20 of the 4131 mountainlakes of the state were fishless in the 1830s, as were most high-elevation streams in the Sierra (Knapp 1996). Stocking inYosemite National Park reached a peak of a million fish eachyear in the 1930s and 1940s (Drost and Fellers 1996). Stockinghas recently been reduced in the Sierra Nevada and discontin-ued in the region’s national parks, in part because of concernabout its effects on amphibians (Carey et al. 2003).

The best-documented effects of introduced fishes are for themountain yellow-legged frog (Rana muscosa). It was once acommon frog in high-elevation lakes, ponds, and streams ofthe Sierra Nevada (Grinnell and Storer 1924), but disappearedfrom over 85% of historical sites in the Sierra (Bradford et al.1994b; Drost and Fellers 1996; Vredenburg 2004). Rana muscosahas a larval period of 1–4 years in the Sierra, and therefore re-quires permanent bodies of water, as do fishes (Wright andWright 1949; Knapp and Matthews 2000). There is a strong



negative association between the presence of fish and of R.muscosa, even when habitat type and isolation are taken intoaccount (e.g., Bradford 1989; Knapp and Matthews 2000). Den-sities of R. muscosa are low in the few sites where it does co-occur with fishes (Knapp and Matthews 2000; Knapp et al.2001; Vredenburg 2004). Rana muscosa recolonizes lakes fromwhich fishes have disappeared (Knapp et al. 2001) or havebeen removed experimentally (Vredenburg 2004; Figure B).

Although introduced fishes can account for the disappear-ance of R. muscosa from some sites, R. muscosa also disappearedfrom lakes without fishes in Sequoia, Kings Canyon, andYosemite National Parks (Bradford 1991; Bradford et al. 1994b;Drost and Fellers 1996). Furthermore, some disappearances oc-curred during or after the late 1970s, after fish stocking was onthe wane. One school of thought is that these other disappear-ances are an indirect result of fish introductions (Bradford1991; Bradford et al. 1993; Knapp and Matthews 2000; Vreden-burg 2004). According to this hypothesis, R. muscosa exists inmetapopulations in which populations sometimes go extinctdue to natural stochastic processes such as winterkill, drought,disease outbreaks, and predation. Sites are then recolonized byindividuals migrating along streams. Stocking of fishes in lakesand streams has reduced the number, size, connectivity, andaverage habitat quality of R. muscosa populations, however.Thus, populations are now more likely to go extinct, and lesslikely to be recolonized if they do. The time lag between fishstocking and the disappearances of some populations is the

“extinction debt” predicted by metapopulation theory (Hans-ki 1998; Hanski and Ovaskainen 2002).

Introduced fishes and bullfrogs (Rana catesbeiana) arethought to have contributed to widespread range reductions oflowland yellow-legged frogs, Rana boylii (Hayes and Jennings1986; Kupferberg 1997) and red-legged frogs, Rana aurora(Adams 2000; Hayes and Jennings 1986; Kiesecker et al. 2001b)in the western U.S. Their role is difficult to evaluate, however,because fish introductions, bullfrog introductions, and habitatalterations have occurred concomitantly across the landscape(Hayes and Jennings 1986). Several experiments have soughtto disentangle these factors. Kupferberg (1997) found that bull-frog tadpoles outcompeted R. boylii tadpoles, and documentedthat breeding populations of R. boylii were greatly reduced ina stretch of stream that was unaltered except for the presenceof R. catesbeiana. Experiments with R. aurora and native Pacifictreefrog (Pseudacris regilla) tadpoles concluded that direct ef-fects of the introduced species were less important than thewidespread conversion of temporary ponds to permanentponds, which provide better habitat for the introduced thanthe native species (Adams 2000). Kiesecker et al. (2001b) foundthat these habitat alterations intensified competition betweenR. aurora and R. catesbeiana tadpoles.

Negative relationships between the distribution of intro-duced fishes and the distribution and abundance of other am-phibians have also been found, including the Pacific treefrog(Pseudacris regilla) in the Sierra Nevada (Matthews et al. 2001)

and most native species in the mountains ofnorthern Spain (Brana et al. 1996; Martinez-Solano et al. 2003a). The effects of fishes on P.regilla and other species are apparently local-ized, perhaps because they use some habitatsthat fish cannot, such as ephemeral ponds andterrestrial areas, which may ameliorate land-scape-level effects of fish stocking (Bradford1989; Matthews et al. 2001). Bufo boreas and B.canorus frequently breed in fishless temporaryponds and produce toxins that fish avoid (Brad-ford 1989; Drost and Fellers 1996) and thesefishes may have little affected their declines. Be-cause fishes have been introduced into nearlyall montane systems on the planet, the possibil-ity of impacts exists for many amphibianspecies, most of which are as yet unstudied.

Chemical PollutantsPesticides, herbicides, heavy metals, and otherchemical pollutants can have lethal, sublethal,and indirect effects on all organisms, includingamphibians (Sparling et al. 2000; Blaustein et al.2003a; Linder et al. 2003). Chemical pollutionmight seem to be a minor threat in isolated, pro-tected areas; however, chemical contaminantsmay be transported long distances through at-

96 Chapter 3


Num

ber

ofta

dpo

les/

10m

45

40

35

30

25

20

15

1

1

2

2

3

3

4

4

5

2003200120001999199819971996

0

10

5

Figure B Density of larval Rana mucosa in 21 lakes in the Sierra Nevada, California,from 1996 to 2003. Filled triangles represent trout removal lakes (n = 5), and numberscorrespond to individual lakes. Open circles indicate fishless control lakes (n = 8). Thehorizontal bars beneath the graph indicate the time period over which trout were re-moved. R. mucosa had not yet recolonized lake 5 after trout were removed in 2001. Control lakes with trout (n = 8) never had any frogs present during the duration of the study. No tadpole counts were made in 2002. (Modifed from Vredenburg 2004.)

mospheric processes. Further, extremely low, supposedly safedoses of chemicals can harm biota (e.g., Marco et al. 1999; Re-lyea and Mills 2001). For example, exposure to 0.1 ppb of theherbicide atrazine causes gonadal deformities in northernleopard frogs (Rana pipiens) and African clawed frogs (Xenopuslaevis), whereas the U.S. drinking water standard for atrazineis 3 ppb (Hayes et al. 2003, Hayes 2004).

The best-documented connection between long-distancetransport of chemical pollutants and enigmatic changes in am-phibian populations is in California. Prevailing winds transportpesticides and other contaminants from areas of intensive agri-culture in the Central Valley to national parks, forests, andwilderness areas in the Sierra Nevada (Fellers et al. 2004; LeNoiret al. 1999; Sparling et al. 2001). Disappearances of Rana auroradraytonii, R. boylii, R. cascadae, and R. muscosa populations in Cal-ifornia are correlated with the amount of agricultural land useand pesticide use upwind (Davidson et al. 2001; Davidson 2004).Rana muscosa that were translocated to an area of Sequoia Na-tional Park from which they had disappeared (despite the ab-sence of fish) developed higher tissue concentrations of chlor-danes and a DDT metabolite than were found in persisting R.muscosa populations 30 km away (Fellers et al. 2004).

These studies provide correlative evidence that pesticidescontributed to declines and disappearances of amphibian pop-ulations in the Sierra Nevada. Acceptance of this hypothesiswill require additional evidence connecting the presence ofpesticides and their effects on individuals to effects on popula-tions. This connection cannot be assumed. For example, leop-ard frogs remain abundant in many areas highly contaminat-ed with atrazine in spite of this herbicide’s negative effects onreproductive organs (Hayes et al. 2003). Interactions betweenpesticides transported long distances and other factors havebeen suggested as a cause of amphibian declines and disap-pearances in other protected areas, including Monteverde andLas Tablas, Costa Rica (Pounds and Crump 1994; Lips 1998)and The Reserva Forestal Fortuna, Panama (Lips 1999), but re-main little investigated at these sites.

Acid rain is another type of pollution that could affect am-phibian populations in isolated areas (Harte and Hoffman

1989). Chemical analyses of water samples in several regions,however, suggested that acid precipitation is an unlikely causalfactor (e. g., Richards et al. 1993, Bradford et al. 1994a, Vertucciand Corn 1996).

Natural Population FluctuationsPopulation sizes of amphibians, like those of many other taxa,may fluctuate widely due to natural causes. Drought, preda-tion, and other natural factors may even cause local extinc-tions, necessitating recolonization from other sites. Some de-clines and disappearances of amphibian populations in areaslittle affected by humans may be natural occurrences fromwhich the populations may eventually recover, provided thatsource populations exist elsewhere in the general area(Blaustein et al. 1994b; Pechmann and Wilbur 1994). Naturalprocesses may account for declines and extinctions in the At-lantic forest of Brazil (severe frost and drought, Heyer et al.1988; Weygoldt 1989), and the loss of some montane popula-tions of the northern leopard frog in Colorado (drought anddemographic stochasticity, Corn and Fogleman 1984). Naturalfluctuations may also interact with human impacts, resultingin losses from which recovery is unlikely.

Pechmann et al. (1991) used 12 years of census data for am-phibians breeding at a pond in South Carolina to illustrate howextreme natural fluctuations may be, and how difficult it canbe to distinguish them from declines due to human activities(Figure C). Even “long-term” ecological studies rarely capturethe full range of variability in population sizes (Blaustein et al.1994b; Pechmann and Wilbur 1994). Formulating “null mod-els” of the expected distribution of trends in amphibian popu-lations around the world, against which recent losses may becompared, is a challenging task (see Pechmann 2003 for a re-view). For example, populations in which juvenile recruitmentis more variable than adult survival may decrease more oftenthan they increase, because the average increase is larger thanthe average decrease (Alford and Richards 1999). The expecta-tion for these populations is that more than half will exhibit adecline over any given time interval even if there is no trueoverall trend.



Ambystoma talpoideum3000

2000

1000

0

12,000

10,000

8000

6000

4000

2000

079 81 83 85 87 89

3

(I)

Ambystoma tigrinum300

200

100

0

1200

1000

800

600

400

200

0

(II)

79 81 83 85 87 89 2 1

Figure C Natural fluctuations in popu-lation size of two species of salamandersin a temporary pond, Rainbow Bay,South Carolina from 1979 to 1990. Num-bers of both breeding females (blackbars, left axis) and metamorphosing ju-veniles (gray bars, right axis) vary great-ly annually, and some species disappearfrom and reappear in the system. Num-bers in the figure refer to extremely lowcounts of 3 or fewer individuals. Suchspecies may be poor choices as indica-tors for the larger system since their nat-ural population fluctuations are so large.(From Pechmann et al. 1991.)

Climate ChangeChanges in amphibian population sizes and distributions areoften related to changes in environmental temperatures andprecipitation (e.g., Bannikov 1948; Bragg 1960; Semlitsch et al.1996). This variation has traditionally been viewed as natural(Pechmann and Wilbur 1994). The effects of anthropogenicgreenhouse gas increases on global climates suggest alterna-tive interpretations in some cases, however.

Some of these alternatives are mediated by the possible ef-fects of greenhouse warming on the El Niño/Southern Oscil-lation (ENSO), a cyclical warming (El Niño) and cooling (LaNiña) of the eastern tropical Pacific. Kiesecker et al. (2001a)found that during El Niño events, precipitation in the OregonCascade Mountains was low, resulting in low water levels atBufo boreas breeding sites. The shallow water afforded toadeggs little protection from UV-B radiation, which increasedtheir vulnerability to a pathogenic fungus (Saprolegnia ferax),causing high mortality. Kiesecker et al. (2001a) hypothesizedthat if global warming is increasing the intensity and frequen-cy of El Niño conditions, this may result in amphibian popula-tion declines via scenarios analogous to that elucidated bytheir research with western toads. The effects of global warm-ing on El Niño conditions remain unresolved, however (Cane2005; Cobb et al. 2003, see Chapter 10).

The disappearance of the golden toad (Bufo periglenes) andother frogs from the Monteverde Cloud Forest Preserve inCosta Rica in the late 1980s was associated with an El Niñoevent, as were subsequent cyclical reductions of some remain-ing amphibian populations (Pounds et al. 1999). During ElNiños, the height at which clouds form increases at Mon-teverde, reducing the deposition of mist and cloud water thatis critical to the cloud forest during the dry season. Pounds etal. (1999) suggested that greenhouse warming has exacerbatedthis El Niño effect (see Case Study 10.3), citing a global climatemodel simulation that predicted warming will increase cloudheights at Monteverde during the dry season (Still et al. 1999).This model represents only a crude proxy, because its spatialresolution is too coarse to explicitly model cloud formation ona particular mountain (Lawton et al. 2001; Still et al. 1999). Re-gional atmospheric simulations and satellite imagery suggestthat lowland deforestation upwind also may have increasedcloud base heights at Monteverde (Lawton et al. 2001).

Atelopus ignescens and several other Atelopus species arethought to have disappeared from the Andes of Ecuador during1987–1988, which included the most extreme combination ofdry and warm conditions in 90 years (Ron et al. 2003). Temper-atures in this region increased 2ºC over the last century, proba-bly largely due to greenhouse warming (Ron et al. 2003). Otherstudies also have detected associations between enigmatic de-clines and disappearances of amphibian populations and tem-perature and precipitation anomalies (Laurance 1996; Alexanderand Eischeid 2001; Burrowes et al. 2004). In these cases theanomalies were within the range of natural variation, thus it isunlikely that they were the direct cause of the amphibian losses,although they may have been a contributing factor. Global

warming has been associated with earlier breeding of some am-phibian species (e.g., Beebee 1995; Gibbs and Breisch 2001). Al-though these phenological changes could potentially result indemographic or distributional changes, there is no evidence thatthey have to date, except when earlier breeding is associatedwith dry conditions as for western toads in Oregon (Blaustein etal. 2003b, see also Corn 2003; Kiesecker et al. 2001a).

Subtle Habitat ChangesSubtle habitat changes are an understudied potential cause ofenigmatic losses of amphibians. For example, fire suppressionin Lassen Volcanic National Park, California, has allowed en-croachment of trees and shrubs in and around open meadowponds, streams, and marshes, rendering these sites unsuitablefor Rana cascadae breeding (Fellers and Drost 1993). Pondcanopy closure is known to have local effects on amphibian bio-diversity elsewhere (Halverson et al. 2003; Skelly et al. 1999).Construction of dams in the Sierra Nevada has altered temper-atures and hydrological regimes downstream, making the habi-tat unacceptable for Rana boylii breeding (Jennings 1996).

ConclusionsThe current thinking of the majority of researchers is that thereare many interacting causes for enigmatic amphibian losses.Great challenges face those studying declining amphibianpopulations in scaling individual effects to the populationlevel. Further challenges include directly testing hypothesesformulated from correlative studies with well-designed fieldexperiments to elucidate those mechanisms that are drivingthe perceived patterns of decline. Synergistic studies can startby dealing with individual phenomena such as the effects ofintroduced species, and from this foundation scale up to in-clude multiple factors.

Meanwhile, is there nothing we can do to combat these de-clines? Not at all. A good example is the rapid rebound ofRana muscosa populations following trout removal in the highSierra Nevada (Vredenburg 2004; see Figure B). Managementchanges are most likely to be put into effect when direct evi-dence of the phenomenon has been demonstrated, as has beenthe case in large national parks in California where fish re-moval is being actively pursued. Even though fish do not ex-plain all of the declines detected, removal of fishes nonethelesshas a salutary effect.

A greater challenge is the infectious disease problem, espe-cially chytridiomycosis. Whereas adult Rana muscosa withchytridiomycosis in the laboratory invariably die, field studieshave shown that some infected adults can survive the summer(at a minimum) under certain circumstances in the field (Brig-gs et al. 2005). Models indicate that survival of at least somefraction of infected post-metamorphic individuals is crucial tothe persistence of populations (Briggs et al. 2005), and studiesin progress should help us understand what factors are associ-ated with survival of infected individuals in this and otherspecies. Achieving a balance between scientific understandingof problems and conservation action is a continuous challenge.

98 Chapter 3




CASE STUDY 3.2Hope for a Hotspot Preventing an Extinction Crisis in Madagascar

Carly Vynne, University of Washington and Frank Hawkins, Conservation International

Madagascar as a Biodiversity HotspotThe island nation of Madagascar is world-renowned for itshigh levels of endemism and biodiversity richness. Despite itsproximity to the African mainland, Madagascar does not shareany of the typical African animal groups such as elephants, an-telopes, or monkeys. Its 160 million year separation has insteadresulted in a unique assemblage of plants and animals. Mada-gascar is characterized by tropical rainforest in the east, dry de-ciduous forests in the west, and a spiny desert in the far south.

The Madagascar and Indian Ocean Islands Hotspot, whichincludes the neighboring islands of the Mascarenes, Comoros,and Seychelles, has an impressive 24 endemic families, thusmaking it one of the highest-priority biodiversity hotspots onearth. At the species level, 93% of mammals, 58% of birds, 96%of reptiles, and 99% of the amphibians are endemic (Figure A).All 651 terrestrial snails and 90% of the estimated 13,000–15,000plant species are endemic (Mittermeier et al. 2004). In the last 15

years, thousands of new plant species have been identified, aswell as many new animal species (22 mammal, 40 fish, 40 but-terfly, 150 reptile and amphibian, and 800 ant species). Certain-ly, many new species remain to be discovered and described.

Hotspot under SiegeMadagascar stands out as one of the highest global prioritiesfor conservation not only because of its high biodiversity val-ues, but also because of the grave threats facing the country. Itis estimated that 83% of the primary forest cover of Madagas-car has already been lost. The percentage is even higher for thehotspot’s smaller islands, and only 10% of the original 600,461square kilometers of natural habitat remain throughout the en-tire hotspot. Should current rates of forest destruction contin-ue, all of Madagascar’s forests would be lost within 40 years.

The current human population of between 17 and 18 mil-lion is growing at about 2.8% per year, doubling every 20 to 25years. This population is mostly rural, and extremely poor; theper capita annual income in 2002 was U.S.$268. Currentlyabout 2.4 million people live within 10 km of protected areas.

Due to its geographic isolation, humans only arrived onMadagascar about 2000 years ago. The Malagasy people, whoare of both African and Asian descent, brought with them agri-cultural methods such as slash and burn agriculture and exten-sive cattle grazing on pasture that needs to be burned regularly.These land uses have been particularly detrimental because ofinfertile soils and low-life resistance of native vegetation. Tavy,the traditional slash and burn method of clearing rainforest forrain-fed rice production, continues to be a primary threat today,with around 1%–2% of forest being lost annually to this cause.

The disappearance of forest cover has reduced soil fertilityand increased erosion, and poses a dire threat to biodiversity.Estimates show that Madagascar loses the equivalent of 5% to15% of its gross domestic product annually through declines insoil productivity, loss of forests, and damage to infrastructurelinked to forest destruction. The high central plateau of Mada-gascar has been essentially deforested, and soils washed intothe sea are visible from space, giving Madagascar the appear-ance of having a “red ring” around the island.

Subsistence fuel wood collection, charcoal production, andhunting, also affect Madagascar’s forests. Wildlife trade has aserious impact on some endemic amphibians, reptiles, andsucculent plants. The proliferation of invasive species affectssome ecosystems and species diversity, particularly in aquaticareas. For example, Bedotia tricolor, a native endemic fish, is

Figure A Madagascar is home to an enormous number of endemicspecies, including this panther chameleon (Furcifer pardalis). (Photo-graph by F. Hawkins.)

now listed as Critically Endangered due in part to competitionfrom an introduced predator, the spotted snakehead fish(Loiselle et al. 2004).

Impacts of commercial timber exploitation are greater thanin some parts of the tropics, as trees are generally smaller andcompete poorly with introduced species. Regeneration of pri-mary rain forest after logging and slash-and-burn consistsmostly of introduced species. Presence of invasive species isnot a temporary state; rather, invasive species dramaticallyalter the trajectory of forest succession (Brown and Gurevitch2004).

With such human pressures, the resultant threats to biodi-versity are not surprising. Currently, 283 birds, mammals, andamphibians are globally threatened with extinction. Of these, allbut two are endemic. To help thwart a major extinction spasmfrom one of the most biologically rich places on Earth, a sub-stantial effort in conservation is currently underway.

Promise for Protected AreaExpansionAt the 2003 World Parks Congress held in Dur-ban, South Africa, the President of Madagascar,Marc Ravalomanana, announced the Durban Vi-sion—the Government of Madagascar’s commit-ment to increase protected areas. In his words,“We can no longer afford to let one single hectareof forest go up in smoke or let our many lakes,marshes and wetlands dry up, nor can we incon-siderately exhaust our marine resources. I wouldlike to inform you of our resolve to bring the pro-tected areas from 1.7 million hectares to 6 millionhectares over the next five years. This expansionwill take place through strengthening of the pres-ent national network and implementation of newmechanisms for establishment of new conserva-tion areas.”

Prior to this announcement, only 2.4% of thehotspot is designated under the most highly pro-tected categories I–IV of IUCN protected areas;overall only 3% of the land area has any protec-tion. The President of Madagascar pledged toseek $50 million from the international commu-nity to make the protected area expansion a real-ity, and within six months $18 million had beenpledged to the trust fund.

Maximizing Protected Area Designfor Biodiversity Conservation To ensure that investments in Madagascar’s pro-tected areas provide the highest possible cover-age of species diversity, many conservation or-ganizations and independent researchers areworking to assemble scientific data that will high-light areas most urgently in need of protection.

New databases make systematic planning for a network of pro-tected areas possible. First, geographically associated, digitalGIS files of Madagascar’s protected areas were developed inpreparation for the World Parks Congress. From these maps itis possible to identify “gaps” where important areas remainwithout park protection (see Essay 14.2). Second, distributionmaps now exist for many of Madagascar’s threatened species.A third important dataset is a forest cover change map. A mapof deforestation between 1950 and 2000 is shown in Figure B.From this dataset and map, planners may see areas that are un-dergoing rapid habitat conversion and areas where suitablehabitat patches remain for protection in reserves. In addition itcan be used to evaluate specific threats to species that occupysmall ranges or limited habitats.

The Durban Vision has spurred conservation biologists andplanners to take a countrywide view and use emerging infor-mation both to develop tools that can help guide decision mak-

100 Chapter 3


Forest cover2000

Madagascarforest covercirca 1950

ForestNon-forest

Figure B Loss of forest cover on Madagascar between 1950 and 2000. (Cartography by M. Denil © Conservation International.)

ing and to define key biodiversity areas. Not only will theseprocesses be used to optimize expansion of the protected areanetwork, but the information can help inform assignment ofmanagement categories so that the highest priority areas areafforded the greatest protection.

From Planning to Park Establishment: The Case of MasoalaEven with the emerging tools now available for reserve se-lection, seldom are protected areas implemented based onscientific design principles. A notable exception to this is Ma-soala National Park, established in Madagascar in 1997. Theestablishment of this park created Madagascar’s largest pro-tected area. The planning process was based on biologicaland socioeconomic principles of conservation design. Thereis much to learn from the Masoala design phase to ensure cre-ation of successful reserves in the future.

The Masoala Peninsula of northeast Madagascar is one ofthe largest forest blocks in the country. Beginning in 1990, ef-forts were underway to establish an integrated conservationand development project (ICDP), as well as a protected area, onthe peninsula. As local use of the ICDP was part of the strategy,an important component of park design was to demarcate coreforest areas for biodiversity conservation. Thus, design criteriaemphasized that the park be of sufficient size to buffer againstdisturbances and protect its resources. The second componentof the design aimed to enable socioeconomic sustainability andprevent conflicts with local communities (Kremen et al. 1999).

To design the park, maps of different forest types were com-piled. Next, a biological inventory was completed, with surveysfor birds, primates, small mammals, butterflies, and cicindelidand scarab beetles along five environmental gradients. Thesedata were used to ensure as many species as possible were rep-resented in the core area of the park. To incorporate socioeco-nomic data, all villages and settlements on the peninsula weremapped and focus-group interviews were held to obtain socioe-conomic and agricultural data. Kremen and colleagues subse-quently conducted an economic use analysis to determine whereto zone for selective logging. Within the use zone, a forest inven-tory assessed timber resources. Finally, a threats analysis identi-fied and ranked the major threats to biodiversity and identifiedareas likely to be lost without increased levels of protection.

The full range of elevation gradients present on the penin-sula were required if the park was to meet the objective of pro-tecting all known biodiversity. The peripheral, threatened zonein the park’s eastern limit had a high edge-to-area ratio andwould have been extremely difficult to protect. Thus, this areawas left outside of the park to serve as an ecological buffer andan economic support zone for local people. The economicanalyses suggested that sustainable forest management in thiszone would provide better returns to local communities thanslash-and-burn practices.

The recommended design led to the establishment of thepresent day Masoala National Park, which is 2300 km2 and is

surrounded by a 1000 km2 buffer of community managed for-est (Kremen et al. 2000). Slash-and-burn farming for subsis-tence rice production is the principal threat to unprotectedforests. Thus, the park management is working with local com-munities to create economic incentives for maintaining theboundary forest. Continued assistance to the local communi-ties will be required over a long time period.

That Masoala exists as a national park today should becredited not only to sound design, but also to persistent andpersuasive pressure from NGOs. As described by Kremen etal. (2000), “Several timber companies were prospecting forconcessions on the Masoala Peninsula during the time that theNational Park was being established, and the governmentnearly abandoned the park project in favor of a logging com-pany. The conservation and diplomatic community played alarge role in persuading the government to reject the loggingcompanies’ proposals, using both political and economic ar-guments. Without this pressure, the Masoala Peninsula, one ofMadagascar’s most important reservoirs of biodiversity,would perhaps have become a forestry concession instead of anational park.”

Continuing the Legacy: Sustainability andFinancing of a New Reserve SystemWhile local incentives are important to protecting resources,incentives at national and global scales also are essential to thesuccess of conservation. To assess benefits and costs, Kremenet al. (2000) compared estimates of benefits from the MasoalaICDP at local, national, and global scales. The results from theiropportunity–cost scenario showed how sustainable harvestversus large-scale timber clearing would benefit local commu-nities, the national government, and the global community.These results have strong implications for long term sustain-ability of not only the Masoala site, but for protected areasthroughout Madagascar.

Kremen et al. (2000) found that, at the national level, thehighest short-term return would come from large-scale indus-trial forestry concessions. The scenario is different at the localand global scales, however. Under the assumption that theforestry concessions would be foreign-owned, the local com-munities would benefit more from long-term sustainable har-vest than from commercial timber harvest. At the global scale,the greatest benefit is also from protection of forests. Loss ofthe forest would cost the international community an estimat-ed U.S. $68 to U.S. $645 million, based solely on the area’s con-tribution to reducing greenhouse gases. Thus, both the localand global communities benefit from forest protection, where-as, at the national scale, the greatest economic payoff wouldbe to sell off the forest for timber concessions. This so-called“split-incentive” situation suggests that continued pressure onnational government through financial and political incen-tives may be critical to long-term park protection.

Work is underway to estimate the costs of managing the en-tire protected area network (national parks and conservation



sites). Between 1997 and 2003, U.S.$50 million, mostly from in-ternational donors, was spent on Madagascar’s protected areasystem. However, benefits at the national level exceeded thisfigure by 25%, through water delivery, erosion avoidance, andecotourism revenue (Carret and Loyer 2003). This studychanged the government view on the economic value of pro-tected areas, and it was decided that a trust fund was needed toensure that the planned reserve network would receive the re-quired operational and management funds. To plan for thistrust fund and determine the core costs for the forthcoming ex-panded network, efforts are underway to model expected costs.

The cost model being developed is innovative in that ittakes into account priorities for management by consideringoverall biodiversity values as well as threat levels facing the re-serve. Specifically, management costs allotted to a protectedarea depend on (1) the importance of the area for biodiversity,(2) the strategic function of each area (i.e., potential for devel-oping research, ecotourism, or educational opportunities), (3)the size of the area, and (4) the threat level. Table A shows thenumber of park wardens required based on these parameters.The model incorporates supervisory, administrative, andequipment costs based on the number of wardens. This adap-tive and predictive cost tool allows a cost estimate for eachnew site proposed to the system. It provides the global costs ofmanaging the entire reserve system as well as cost by site overyears. This tool will not only provide the trust fund and donorswith a transparent analysis of anticipated costs of managingthe network, but will serve as the basis for a cost accountingand monitoring system for the national governing agency.These estimates also help conservationists and the Madagascargovernment work together to leverage the required funds.

Protected Areas: Providing Promise for MadagascarThere are many reasons to believe that a strengthened and ex-panded protected area network and expanded conservationmeasures will make a difference in stopping extinction onMadagascar. Compared with an average annual rate of lossduring the 1990s of 0.9% per year, deforestation from withinprotected areas was only 0.04% to 0.02%. Species also appearto be faring better in protected areas, as reef fish biomass isgreater within the relatively new Masoala National Park

(about 810g/m2) than outside (525 g/m2). The annual burningof the Alaotra marshes declined from 32% to 2 % from 2000 to2002. At the landscape scale, the Zahamena-Mantadia priorityconservation corridor experienced one-third the forest loss(2.2%) than the neighboring area (6.7%). Furthermore, Mada-gascar has invested heavily in its National Environmental Ac-tion Plan over the last 10 years. A new Malagasy Parks Servicehas been established that is helping to professionalize the serv-ice and implement management plans.

From Priorities to Projects Within the planning process for increasing Madagascar’s pro-tected area network, there are continued efforts to identify thegaps in protected areas. Once the gaps are identified and newareas for protection proposed, projects will be initiated to helpconserve biodiversity values as intended. The site-conserva-tion process requires three elements: an international donor in-terested in investing in biodiversity, a “conservation broker”that has contact with both the donor and local groups workingin the area (usually an international NGO or similar body), anda self-motivated and self-sustaining, field-based NGO networkthat manages biodiversity based on a general recognition ofthe economic and environmental benefits that this brings. Im-plementation of conservation programs is increasingly done atthe landscape or biodiversity scale. Rarely can site conserva-tion be achieved by focusing solely on the site of interest; com-peting interests and threats are too strong for narrow conser-vation solutions.

The Menabe region of western Madagascar provides an ex-ample in which many conservation interest groups are comingtogether to realize on-the-ground conservation projects that aredesperately needed. The Menabe is a genuine hotspot within ahotspot and one of the most important conservation prioritiesin Madagascar and Africa. The forest is one of the few sites onEarth where baobab trees grow in dense colonies. Indeed,Grandidier’s baobab (Adansonia grandidieri), one of the largestand most beautiful, is probably functionally extinct outside thisarea. Menabe’s rich biodiversity is under enormous humanpressures that have led to a 25% loss in area over the last 40years. This rate of loss has accelerated in the last 10 years.

Illegal slash-and-burn agriculture to cultivate maize has beenthe most important cause of forest loss in recent years. The soils

of the region are sandy, dry, and poor, andonce the organic matter in standing vegetationhas been used up, crop yields diminish expo-nentially and the farmers are forced to moveon. The second key threat is logging. While thedirect impacts of tree felling seem not to havea great short-term impact on most forest bio-diversity, additional hunting and increasedrisk of fire post-logging often means thatwood-cutting ultimately leads to destructionof the forest. Therefore, long-term conserva-tion is difficult to reconcile with logging with-

102 Chapter 3


Protected areaAttributes Size in hectares Hectares/warden

High threat with 5000–20,000 1500exceptional biodiversity 20,000–60,000 3000

High threat with 5000–10,000 4000high Biodiversity More than 10,000 6000

Low threat with higher Independent of size 7000exceptional biodiversity

TABLE A Amount of Management Allotted to a Protected Area Based onThreat, Biodiversity Values, and Size



out significant investment in monitoring, oversight, and en-forcement, all of which have been absent up to now in the weakpolicy and government context.

To implement conservation goals in this region, the Region-al Development Committee, helped by a range of national andinternational NGOs, is working to engage local communitiesand regional authorities in conservation, and support their ca-pacity to manage the biodiversity resources in the region. Ad-ditionally, they work to help develop economic conditions thatsupport conservation. Improved local capacity and a more fa-vorable environment for sustainable economic activities willaddress systemic conditions that enable these threats to persist.Current conditions and expected change due to conservationprojects being implemented in the region are listed in Table B.

More favorable conditions will promote access to viablelivelihood options compatible with the long-term conservationof the corridor’s biodiversity. This will help to address proxi-mate threats that cause loss of high-priority habitats and will im-prove community, local, and national government capacity tomonitor, manage, and control forest use in the following ways:

• Dramatically increasing and widening the economicbenefits from biodiversity-friendly activities such asecotourism

• Making sustainable agricultural practices more attractiveand economically viable than slash-and-burn practices

• Providing long-term sources of wood for charcoal,building materials, and firewood through communitymanagement and private sector involvement

• Creating a commercially- and ecologically-viable timberproduction system

Meeting these objectives is only possible because of revived in-terest of local NGOs in collaborative processes to address bio-diversity concerns in the region. Fanamby and Durrell WildlifeConservation Trust are two NGOs that have been leading thiseffort. They work with the National Waters and Forests Au-thority, the Center for Professional Forestry Training, the Ger-man Primate Center, ANGAP (National Protected Area Man-agement Agency), local community groups and other local

NGOs, and mayors. This work is chiefly aimed to build localcapacity to design and implement forest management plansand to enforce regulations. Durrell Wildlife Conservation Trustalso monitors the conditions of wildlife and runs an educationand awareness campaign to decrease bushmeat hunting,wildlife trade, and illegal logging. Activities of the collabora-tion include (1) increasing participation in rural resource man-agement planning, (2) identifying the highest biodiversity pri-ority areas, (3) fostering broad support for conservationenforcement, and (4) promoting the development of privatesector forest plantations to meet local demand for wood, char-coal and firewood.

A collaborative approach has much to offer. For example, inone area the local population has long been hostile to conven-tional development action and the area was considered a lostcause. However, recent negotiations with village elders for atraditional Malagasy law, or dina, has resulted in a substantialreduction of forest cutting and an openness to new income-generating activities. The key to project sustainability is devel-oping the capacity of regional and local stakeholders to bene-fit from the sound management of environmental resources,especially biodiversity. While some investment in biodiversityconservation will always be required, the overall strategy is todevelop the technical capacity of the region so that stakehold-ers in Menabe are independently capable of accessing invest-ment in biodiversity conservation. The economic opportunitiesoffered by endemic biodiversity, especially in the context ofecotourism, are likely to contribute substantially over the longterm to the very depressed economy of the region. Currentagricultural and logging processes are totally unsustainable,but opportunities exist to move these sectors away from re-source mining and into genuinely sustainable practices.

One important aspect of sustainability is social sustainabili-ty—the capacity of the region to absorb the changes that arenecessary for the development envisaged and to establish theskills and institutional capacity and mechanisms to sustain thechanges. The current constraint is the combination of very poormanagement oversight and a very depressed local economythat has little or no scope or motivation for investment. Withthis program, the partners hope to overcome both of these con-

Current systemic conditions Expected change

Low local institutional (community, government, and NGO) Improved local and regional capacity for natural resource management and technical capacity management

Inadequate protection of priority biodiversity areas within Targeted investment in biodiversity conservation in most importantthe corridor areas

Government ownership of natural resources leading to Transfer to local communities of management rights and economic land-grabs in forested areas benefits derived from sustainable natural resources

Unregulated forestry sector, with local forestry administration Improved compliance with national forestry legislationoften heavily involved

Inadequate local benefit from biodiversity conservation Increased local knowledge and participation of the local populationand low community support for conservation conservation in sustainable livelihood options resulting in support

for biodiversity

TABLE B Current Problems Facing the Menabe Corridor and Improvements Expected from Conservation Collaboration

104 Chapter 3


straints and encourage those initiatives that are already clearlymanifest in the region.

The Malagasy as Stewards

“Tsy misy ala, tsy misy rano, tsy misy vary”

The Malagasy recognize that the fate of the forest is their own.As the above saying goes, without forest, there is no water,there is no rice. Many efforts underway are channeling interna-tional interest in biodiversity conservation to the local commu-nities who are neighbors to critical habitats. For example, half

of all entrance fees to parks now go back to the local communi-ties. Conserving the country’s unique heritage into the futurewill certainly pose many challenges, but the actions taken bythe country’s leaders are providing great hope for the future ofthis hotspot. While all acknowledge the importance of Mada-gascar for global biodiversity, there have been many who havequestioned whether saving the hotspot would be worthwhile,or even feasible. Fortunately, Madagascar has recognized itsown position as unique home to biodiversity, and is making op-timists out of many who once doubted that anything could bedone to save the country from an extinction crisis.

CASE STUDY 3.3Key International and U.S. Laws Governing Management and Conservation of Biodiversity

Daniel J. Rohlf, Lewis and Clark Law School

A primary focus of conservation biology is on understandinghow one species—Homo sapiens—has affected life forms andecosystems across the planet. If conservation biologists wish tobuild knowledge that is not only relevant to making choicesabout the future of Earth’s biological resources, but is indeed in-fluential in shaping this future, it is imperative that they devel-op at least a good working knowledge of social decision-makingprocesses and regulatory systems in addition to honing their sci-entific expertise. Here, I summarize some of the major Interna-tional and U.S. laws that influence biodiversity conservation.See Box 3.3 for a short description of other national biodiversitylaws.

International Legal Regimes that AffectBiodiversity International law is unique in that, unlike individual nations’systems of governance, there is no global sovereign with rec-ognized authority to impose rules for conduct. This means thatmuch of international law actually consists of agreements bycountries that choose to cooperate with one another toward aspecific goal; these agreements typically take the form oftreaties, conventions, and voluntary participation in interna-tional organizations. The following discussion summarizessome of the agreements most important to managing and pro-tecting biodiversity.

The Convention on International Trade inEndangered Species of Wild Fauna and FloraThe Convention on International Trade in Endangered Speciesof Wild Fauna and Flora (CITES)—first opened for signature in1973—attempts to regulate international trade in species

whose existence is or may be imperiled as a result of commerceand other trafficking. The agreement regulates only interna-tional trade in designated species and their parts; it does not at-tempt to govern habitat management or other species man-agement decisions made by individual countries.

CITES’ trade restrictions apply only to species (and parts orproducts made from a particular species) that member coun-tries vote to add to one of the convention’s three appendices.Species that appear in Appendix I—the most imperiled classifi-cation—are subject to the most stringent trade restrictions, in-cluding a ban on international trade for “primarily commercialpurposes.” Facing lesser threats, Appendix II species may movein international commerce, but like those species in AppendixI, any shipment must be accompanied by an export permit.CITES does not set up any sort of international authority to im-plement and enforce the convention; accordingly, each individ-ual country bears responsibility for making necessary findingsand issuing required permits. The agency that serves as the“Scientific Authority” for the government of an exporting coun-try must certify that the transaction “will not be detrimental tothe survival of that species” in the wild. However, nothing inCITES or its implementing documents defines what constitutesa “detrimental” impact to Appendix I and II species. As a result,each country’s Scientific Authority must formulate and applyits own meaning of this term.

Finally, any CITES member country may unilaterally add aspecies to Appendix III. Appendix III species are subject to re-strictions only when the trade originates from the country thatlisted the species; this provision thus allows individual coun-tries to regulate international commerce in a species that is introuble only within that country.



Convention on Biological DiversityThe Convention on Biological Diversity (CBD) was opened forsignature in 1992 at the Earth Summit in Rio de Janeiro, andnow includes 188 countries as parties (but not the United States).The agreement is the centerpiece of the international communi-ty’s efforts to craft a comprehensive approach to conserving bi-ological diversity. Administratively, a Secretariat headquarteredin Montreal serves as the permanent body for day-to-day busi-ness, and a Conference of the Parties takes place every twoyears.

The CBD requires countries to develop national strategies“for the conservation and sustainable use of biological diversi-ty.” In meeting this goal, parties pledge to carry out a litany ofconservation actions, including creating or maintaining a sys-tem of protected areas, making efforts to restore degradedecosystems and recover threatened species, eradicating nonna-tive species, pursuing programs for ex situ conservation, andusing knowledge derived from “traditional lifestyles” of in-digenous communities to foster biodiversity protection as wellas sustainable use of biological resources. However, the CBDspells out these obligations only in very general terms, leavingthe task of deciding precisely the sort and extent of biodiversi-ty protections and programs to adopt to individual countrieswith help from advisory committees. The agreement also spec-ifies that developed countries have a responsibility to provide“new and additional financial resources” to their less-devel-oped counterparts; and in a telling nod to the link between eco-nomic status and ability to pursue conservation objectives, itnotes that developing countries’ adherence to their commit-ments under the convention will depend largely on the extentto which developed countries follow through with transfers offunds and technology to their less well-off neighbors.

The Convention on Biological Diversity also attempts totackle the sensitive issues of access to and use of biological andgenetic resources. It holds that each country has a right to con-trol access to its biological resources, and that use of such re-sources requires prior informed consent of the nation of origin;a country’s consent can be contingent on reaching “mutuallyagreed terms,” (i.e., payment). The agreement also providesthat developed states have a responsibility to share technolo-gies and biotechnologies with countries that are not as well off;concern about how this provision affects intellectual propertyrights has proven to be one of the most significant obstacles toU.S. ratification of the convention.

International Habitat and EcosystemProtectionsThere are a number of international efforts to identify and pro-tect either ecosystems in general or specific types of ecosys-tems. The following paragraphs summarize three of the mostprominent.

The United Nations’ Educational, Social and Cultural Or-ganization’s (UNESCO) Man and the Biosphere Programmeprovides an institutional umbrella for a series of international

Biosphere Reserves. UN member states, acting on a purely vol-untary basis, nominate for inclusion within the system areaswithin their jurisdictions that already receive protectionsunder domestic legislation. Nominated areas must meet a listof UNESCO criteria, including a requirement that a proposedreserve “contribute to the conservation of landscapes, ecosys-tems, species and genetic variation,” and encompass “a mosa-ic of ecological systems representative of major biogeographicregions.” While formal designation as a Biosphere Reservebrings a measure of international recognition, there are no in-ternational standards for management of these reserves; na-tional laws supply the sole directives for protecting biodiversi-ty and managing human activities in these areas. More than400 Biosphere Reserves have been designated in 94 countries.

The Convention on Wetlands, also known as the RamsarConvention after the city in Iran where it was originally signedin 1971, sets forth a process for identifying important wetlandsanalogous to that of UNESCO’s program for Biosphere Re-serves. The 138 countries that have signed the conventionpledge to designate at least one wetland within their bordersfor inclusion of the convention’s “List of Wetlands of Interna-tional Importance,” as well as to include wetland protectionsin their national land use planning. Again, national and locallaws provide the exclusive management standards for wet-lands included on the list, which now includes over 1300 wet-land areas covering over 100 million hectares.

Lastly, 13 marine or large freshwater areas are the focus ofcoordinated management and protection efforts under theUnited Nation Environment Program’s Regional Seas initia-tive. Working with UNEP, neighboring countries develop an“Action Plan” for coordinated action to conserve and managea regional sea. These plans typically set out broad “frame-work” strategies that rely on more specific international con-ventions and protocols among the countries surrounding a re-gional sea to initiate steps toward implementing the ActionPlans’ broad outlines. Such agreements also usually rely on na-tional laws to actually carry out identified actions. While someRegional Seas programs, such as that covering the Mediter-ranean, have developed a number of conventions and proto-cols that countries have integrated into their national laws,other programs exist mainly on paper.

Federal Laws in the United States that AffectBiodiversityThe federal government in the United States has regulated var-ious elements of biodiversity for over a century, but many ofthe most prominent laws date back only to the 1970s. The U.S.has a number of important statutes that deal with biodiversity,a fact which itself provides insights into this complex arena:On one hand, the many laws dealing with biological resourcesserve as evidence of the country’s recognition of the impor-tance of biodiversity and the United States’ leadership role incrafting laws to manage and protect it; on the other hand, thismultitude of not-always-consistent laws and policies illustrate

106 Chapter 3


the federal government’s fragmented approach to the naturalworld.

Endangered Species ActSince 1973, the Endangered Species Act (ESA) has served asthe United States’ principal wildlife conservation law, and itshigh-profile controversies coupled with a public fascinationwith endangered species have made this statute the country’s(and one of the world’s) best-known biodiversity protectionschemes. It provides a number of substantive protections forspecies classified as either endangered or threatened, includ-ing mechanisms for conserving not only individual organisms,but their habitat as well. Two federal agencies, the U.S. Fishand Wildlife Service and National Oceanic and AtmosphericAssociation (NOAA) Fisheries (collectively “the Services”), areresponsible for day-to-day implementation of the law.

Particularly in more recent years, determining what group-ings of organisms are eligible for protection under the ESA hasgenerated substantial controversy. The law defines “species” el-igible for protection to include, in addition to full species, “sub-species of fish or wildlife or plants, and any distinct populationsegment of any species of vertebrate fish or wildlife which in-terbreeds when mature.” It is important to note that in formu-lating its legal definition of species, U.S. Congress drew a dis-tinction between vertebrates, plants, and all other species.Populations of vertebrates can thus be listed in some placeseven if they are abundant in other parts of their ranges; insects,plants, and other invertebrates must face wholesale eliminationat the species or subspecies level before they can receive legalprotections under the ESA. However, protracted and some-times rancorous debate has marked efforts to determine pre-cisely the meaning of a “distinct population segment” of verte-brate species; a policy adopted by the implementing agencies in1996 considers two primary factors—whether a given popula-tion is (1) “discrete” from other populations of the same species,and (2) “significant” to the species as a whole.

Rather than providing biologically specific definitions of“threatened” and “endangered,” to guide the Services in mak-ing listing decisions for eligible species, the ESA contains a listof factors that the agencies must consider in making listing (aswell as delisting) calls. This includes a litany of biologicalthreats such as habitat destruction, disease, and over-harvest,as well as a catchall category that encompasses any “other nat-ural or manmade factors.” Significantly, the statute also speci-fies that a species can receive the law’s protections as a resultof “the inadequacy of existing regulatory mechanisms” to en-sure the species’ future. After listing a species, the appropriateservice must also designate its “critical habitat,” which the lawdefines as the habitat essential for the species’ recovery. Unlikewhen making listing decisions, an agency may consider eco-nomic and other nonbiological factors in designating criticalhabitat.

The ESA contains two key protections for species listed asthreatened or endangered. First, actions carried out or author-

ized by federal agencies cannot result in impacts that would“jeopardize the continued existence” of listed species or “de-stroy or adversely modify” designated critical habitat. The Ser-vices have generally interpreted these two standards to pro-hibit similar levels of impact; an agency action must put indoubt the future of the entire listed species to run afoul of theserestrictions. Second, the law prohibits anyone—privatelandowners included—from “taking,” (i.e., killing or injuring)a listed species. Regulations define the term “take” broadly toinclude habitat impacts that result in death or injury to mem-bers of a listed species. This provision of the law has generatedconsiderable opposition because it effectively gives the feder-al government wide authority over land-use decisions acrossthe country, a power traditionally exercised by state and localgovernments. However, two factors have helped to keep con-troversies over the ban on take below the boiling point. First,Congress amended the law in 1982 to add the ESA’s so-calledHabitat Conservation Plan (HCP) provisions, which allow theServices to grant a permit that allows some take of listedspecies in exchange for an agreement on the landowner’s partto adopt at least some land-use limitations for the benefit ofthese species. Additionally, the federal government has (unof-ficially) adopted a rather relaxed stance on enforcing the ESA’stake prohibition.

Finally, the ESA requires the Services to formulate “recov-ery plans” for listed species. These plans spell out specificmeasures to improve the status of the species they cover, andmust include “objective, measurable criteria” that, whenachieved, will trigger a process to delist the species as “recov-ered.” Agency statistics reveal that while the populations ofmany species listed as threatened or endangered have stabi-lized, few have actually recovered. The fact that there is noclear legal mechanism requiring implementation of recoveryplans probably plays a role in the ESA’s mixed record of con-servation successes.

National Forest Management ActEnacted by Congress in 1976 after a series of controversies overlogging practices on federal land, the National Forest Manage-ment Act (NFMA) sets management standards for the 192 mil-lion acre national forest system. This vast area encompassessome of the most valuable—as well as most intact—habitat re-maining in the contiguous states. These lands also harbor sub-stantial economic and recreational resources as well, makingthe trade-offs involved in national forest management amongthe most difficult and controversial of those facing federal landmanagers. To a large extent, the NFMA set up the U.S. ForestService to face endless conflict, both internally as well as withoutside interests, by making “multiple use” the touchstone offorest management. The statute encompasses this “do it all”approach by setting out only very general substantive stan-dards—essentially giving agency officials a vaguely-wordedset of marching orders to please all constituencies by both pro-ducing and protecting resources.



The NFMA’s provisions addressing biodiversity reflect thelaw’s broad and imprecise mandates. Each national forestmust prepare a land-management plan setting forth the over-all blueprint for forest management, but the plans’ specific pro-visions for biodiversity are largely influenced by interpreta-tions of vague statutory requirements. The law directs theForest Service to “provide for diversity of plant and animalcommunities based on the suitability and capability of the spe-cific land area in order to meet overall multiple use objectives(set forth in an individual national forest’s managementplan).” The statute contains no definition of “diversity” orother key terms in this mandate, however, thus leaving opento the agency’s own interpretation the outlines of the ForestService’s duties in this area.

For nearly a quarter century, regulations interpreting thestatute’s diversity mandate called on the Forest Service to “pre-serve and enhance the diversity of plant and animal commu-nities, including endemic and desirable naturalized plant andanimal species, so that it is at least as great as that which wouldbe expected in a natural forest.” The regulations specified thatforest managers must “maintain viable populations of existingnative and desired nonnative vertebrate species” found in eachnational forest. A “viable population” was defined as “onewhich has the estimated numbers and distribution of repro-ductive individuals to ensure its existence is well distributed inthe planning area.” The regulations further mandated that“habitat must be provided to support, at least, a minimumnumber of reproductive individuals and that habitat must bewell distributed so that those individuals can interact with oth-ers in the planning area.” To assist agency officials in carryingout these duties, the regulations also prescribed designationand monitoring of “management indicator species,” specieswhose population trends the Forest Service believed would in-dicate the ecological effects of its management activities.

The Forest Service’s interpretation of its duties under theESA grew murky in 2000. Shortly before leaving office, theClinton administration adopted sweeping changes to theNFMA regulations that established “ecological sustainability”as the management aim for biodiversity on national forests;this term was defined as “the maintenance or restoration of thecomposition, structure, and processes of ecosystems includingthe diversity of plant and animal communities and the pro-ductive capacity of ecological systems.” However, the regula-tions retained a complimentary focus on species; managerswere required to provide a “high likelihood” of maintainingviable populations of native and desired nonnative species,with a “viable” species defined as “consisting of self-sustain-ing and interacting populations that are well-distributedthrough the species’ range.”

Within weeks after they were adopted, however, the in-coming Bush administration put the new NFMA regulationson hold and later published its own proposal to replace them.This interpretation of the law’s diversity mandate also empha-sized ecosystem-level management, but proposed to give For-

est Service officials significant latitude to depart from ecologi-cally sustainable practices if necessary to meet multiple-usegoals. Additionally, in the most significant change from previ-ous interpretations of the NFMA’s diversity provision, the newproposal dropped the requirement to maintain viable popula-tions of forest species. Instead, the proposal called for manage-ment that, “to the extent feasible, should foster the mainte-nance or restoration of biological diversity in the plan area, atecosystem and species levels, within the range of biological di-versity characteristic of native ecosystems within the largerlandscape in which the plan area is embedded.” Whatever theshape of the new regulations, their changing and widely var-ied forms in recent years illustrate the broad and ill-defined na-ture of the NFMA’s underlying legal mandate for managingbiodiversity.

The Magnusun Act The Magnusun Fishery Conservation and Management Act of1976 (amended substantially in 1996) controls marine resources,particularly commercially valuable fish species, within theUnited States’ Exclusive Economic Zone, the area within 200nautical miles of the country’s coastline. The statute’s dual—and often conflicting—purposes are to promote a domestic fish-ing industry while establishing a federal program for “conser-vation and management” of the country’s fishery resources.Eight regional fishery management councils, composed of bothfishing industry officials and other interests, develop and im-plement “Fishery Management Plans” for each “stock” of com-mercially valuable fish; these plans set an allowable catch andother needed “conservation and management” requirementsdesigned to result in “optimum yield” of a fishery.

The Magnusun Act has not prevented the collapse or near-collapse of many of the nation’s fisheries. Consequently, theamended law sets out several requirements applicable tostocks that are “over-fished,” which it defines as a rate of fish-ing mortality that jeopardizes a fishery’s capacity to produceits maximum sustained yield on a “continuing basis.” Withinone year of a determination by NOAA Fisheries or one of thecouncils that a fishery is over-fished, or is likely to be over-fished within two years, the relevant council must revise itsmanagement plan to “rebuild” the stock, though this processmay take up to a decade to accommodate the “needs of fishingcommunities.”

Revisions to the Magnusun Act in 1996 added tentativesteps toward protecting marine habitat and ecosystems. Theregional councils must identify “essential fish habitat,” definedas “those waters and substrate necessary to fish for spawning,breeding, feeding or growth to maturity.” NOAA Fisheries andthe Department of Commerce must ensure that any of their“relevant programs” further the “conservation and enhance-ment” of this habitat. Additionally, the law sets up a processwhereby other federal agencies whose actions “may adverselyaffect” essential fish habitat must consult with NOAA Fish-eries, which in turn receives comments on the proposed action

108 Chapter 3


Summary1. Human impacts are a pervasive facet of life on

Earth. Increasing numbers of humans and increas-ing levels of consumption by humans create condi-tions that endanger the existence of many speciesand ecosystems: Habitat degradation and loss, habi-tat fragmentation, overexploitation, spread of inva-sive species, pollution, and global climate change.

2. Many threats are synergistic, where the total impactof two or more threats is greater than what youwould expect from their independent impacts.Also, many threats intensify as they progress. Fur-ther, populations often suffer from genetic and de-mographic problems once they are pushed to lowdensity or small sizes. All these factors make it moredifficult to design effective conservation strategies.

3. Extinction can be either global or local, where aspecies is lost from all of the Earth or from only onesite or region, respectively. In addition, ecological ex-tinction can occur when a population is reduced tosuch a low density that although it is present, it nolonger interacts with other species in the communi-

ty to any significant extent. All these forms of ex-tinction can compromise community functioning,perhaps leading to further losses.

4. Paleontological evidence reveals a long history ofhuman-caused extinctions, although many mayhave resulted from an interaction between humanhunting or species introductions and climate orhabitat changes. Extinctions can cause cascade ef-fects, where a first extinction indirectly causes sec-ondary extinctions of species strongly interactingwith the first. Extinction of dominant species, key-stone species, or ecosystem engineers, as well as theintroduction of such species, all can cause cascadeeffects.

5. The IUCN Red List of Threatened Species is themost comprehensive source of information on pat-terns of global endangerment. Overall, 41% of allevaluated species are threatened. Only a few taxahave been comprehensively surveyed (mammals,birds, amphibians, and gymnosperms), and allshow high fractions of threatened species. Habitatloss and degradation is the leading cause of speciesendangerment.

from the relevant council. If these fisheries experts determinethat an adverse impact to identified habitat is in fact likely,NOAA Fisheries provides the federal agency with recommen-dations for measures that would “conserve” the affected es-sential fish habitat. The federal agency must issue a written re-sponse that either details how the recommendations wereimplemented or explains why they were not.

Marine Mammal Protection ActThe Marine Mammal Protection Act (MMPA) takes a some-what similar management approach as that of the MagnusunAct. The MMPA targets for conservation “stocks” of marinemammals, which the statute defines as “a group of marinemammals of the same species or smaller taxa in a commonspatial arrangement, that interbreed when mature.” NOAAFisheries, the agency primarily responsible for implementingthe act, must seek to maintain these stocks at their “optimumsustainable populations.” In addition to enforcing a generalban on intentionally killing or injuring marine mammals, theagency must take steps to reduce the incidental impacts of fish-ing operations on marine mammal stocks suffering levels ofmortality that exceed their “potential biological removal lev-els.” The latter is calculated according to a detailed statutoryformula that provides the number of animals that can be “re-moved” from the population by human-caused mortality. Ifthis level is exceeded by incidental mortalities from commer-cial fishing operations, NOAA Fisheries must develop “take

reduction plans” that place mandatory conditions on thesefisheries to reduce marine mammal deaths.

The MMPA obliquely encompasses ecosystem conservationin its conservation scheme. The law emphasizes that “stocksshould not be permitted to diminish beyond the point at whichthey cease to be a significant functioning element of the ecosys-tem of which they are a part,” and thus includes ecosystemhealth as a factor in determining stocks’ optimum sustainablepopulations. The statute also specifies that “efforts should bemade to protect essential habitat” of marine mammals, but es-tablishes no substantive standards or binding requirements toactually carry out such actions

ConclusionFuture human efforts to better manage and protect the planet’sbiological riches will depend on innovations in both the sci-ences and the world of policies and regulation. A broad arrayof international and domestic regulatory regimes already playa significant role in shaping the status of biodiversity, butthoughtful input from the scientific community could maketheir implementation and evolution vastly more effective. Suchcontributions, however, can only come from conservation bi-ologists and other scientists that have developed a soundworking knowledge of these laws and policies. Better integra-tion of science into law and policy through the work of newgenerations of professionals is one of the keys to sustainingbiodiversity.



6. Roughly one third of the species in the United Statesare threatened, the majority by habitat loss anddegradation. Freshwater mussels are particularlythreatened, as are other freshwater taxa, showingthat human impacts have been particularly large onaquatic systems.

7. Attributes of species that confer rarity or specializa-tion seem to increase vulnerability to extinctionamong many creatures, on land and at sea. Specieswith low reproductive rates are vulnerable due to therate of extreme changes, and are particularly vulner-able to overexploitation. Island endemics and specieswith small population sizes are highly vulnerable toperturbations. Island endemics are particularly vul-nerable to introduced species, especially when thosespecies are from guilds not represented on the island.

8. Economic and social changes over the next centurywill have a large impact on the degree to which bio-diversity is compromised over the next century. Theextremely large number of people in abject povertycompels us to actions that will help these popula-tions, but this need not be at the expense of biodi-versity. Changes in consumptive habits among peo-ple in China and India will have enormous impactson global biodiversity.

9. International agreements, laws and regulations arecritical tools for slowing biodiversity losses. Thesepolicy instruments range from very comprehensiveand powerful to more limited in their power and use.At the 2002 Johannesburg World Summit on Sustain-able Development, 190 countries committed to re-duce biodiversity losses significantly by 2010, whichhas become a rallying point for biodiversity conser-vation during this first decade of the 21st century.

10. Strategies to reduce biodiversity losses focus on pri-oritizing places to work, understanding the causesof declines, and creating strategies that will be effec-

tive in reducing threats. Although efforts focused onecosystems and landscapes are more likely to be ef-fective in conserving multiple layers of biodiversity,strategic approaches often act on the site or specieslevel.

Questions for Discussion1. How have anthropogenic threats to biodiversity

changed over human history? Which do you expectto be the most important threats in 2050?

2. Despite considerable efforts, we know very littleabout the status of most of the world’s species. Howcan we improve our knowledge on species endan-germent? What kinds of information might be mosthelpful to garner, and why?

3. How can we use information on endangerment tomotivate more conservation-oriented policies? Whatkinds of information are particularly persuasive toyou? How might they serve to persuade national orlocal government officials?

4. Are any species “expendable”? Given the limited re-sources we have for conservation, how should weprioritize conservation efforts among species?Should we give up on those most likely to go ex-tinct?

5. In two or three decades’ time, your children or otheryoungsters may ask you a question along the linesof, “When the biodiversity crisis became apparent inits full scope during the 1990s, what did you doabout it?” What will your answer be?

Please refer to the website www.sinauer.com/groomfor Suggested Readings, Web links, additional questions,and supplementary resources.

threats to biodiversity - university of florida to review the major types of threats to...

Documents