overexploitation - ueae436/reynolds_and_peres_chap08.pdf · and wood bison (bison bison athabascae)...

There’s enough on this planet for everyone’s needs, but not enough for everyone’sgreed.

Gandhi

Exploitation involves living off the land or seas, such that wild animals, plants, andtheir products are taken for purposes ranging from food to medicines, shelter, andfiber. The term harvesting is often used synonymously with exploitation, thoughharvesting is more appropriate for farming and aquaculture, where we reap what wesow.

In a world that seems intent on liquidating natural resources, overexploitation hasbecome the second most important threat to the survival of the world’s birds, mam-mals, and plants (see Figure 3.8). Many of these species are threatened by subsistencehunting in tropical regions, though others are also threatened in temperate and arc-tic regions by hunting, fishing, and other forms of exploitation. Exploitation is alsothe third most important driver of freshwater fish extinction events, behind the ef-fects of habitat loss and introduced species (Harrison and Stiassny 1999). Thus, whileproblems stemming from habitat loss and degradation quite rightly receive a greatdeal of attention in this book, conservationists must also contend with the specter ofthe “empty forest” and the “empty sea.”

We begin this chapter with a brief historical context of exploitation, which also pro-vides an overview of some of the diverse reasons people have for using wild popula-tions of plants and animals. This is followed by reviews of recent impacts of exploita-tion on both target and nontarget species in a variety of habitats. To understand theresponses of populations, we then review the theory behind sustainability. The chap-ter ends with a consideration of the culture clash that exists between people who areconcerned with resource management and people who worry about extinction risk.

History of, and Motivations for, ExploitationHumans have exploited wild plants and animals since the earliest times, and mostcontemporary aboriginal societies remain primarily extractive in their daily quest forfood, medicines, fiber, and other biotic sources of raw materials to produce a wide

8Overexploitation

John D. Reynolds and Carlos A. Peres

range of utilitarian and ornamental artifacts. Modernhunter-gatherers in tropical ecosystems, at varyingstages of transition to agriculture, still exploit a largenumber of plant and animal populations. By definition,these species have been able to coexist with some back-ground level of human exploitation. However, the ar-chaeological and paleontological evidence suggests thatpremodern peoples have been driving other species toextinction since long before the emergence of recordedhistory. Human colonization into previously unexploit-ed islands and continents has often coincided with arapid wave of extinction pulses resulting from overex-ploitation by novel consumers. Mass extinction events oflarge-bodied vertebrates in Europe, parts of Asia, Northand South America, Madagascar, and several archipela-gos have all been attributed to post-Pleistocene humanoverkill (Martin 1984; McKinney 1997). These are rela-tively well documented in the fossil and subfossil record,but many more obscure target species extirpated byhuman exploitation will remain undetected.

In more recent times, extinction events induced by ex-ploitation have also been common as European settlerswielding superior technology expanded their territorialfrontiers and introduced market and sport hunting. Thedeath of the last Passenger Pigeon (Ecopistes migratorius)in the Cincinnati Zoo in 1914 provided a notorious ex-ample of the impact that humans can have on habitatsand species. The Passenger Pigeon probably was themost numerous bird in the world, with estimates of 1–5billion individuals (Schorger 1995). Hunting for sportand markets, combined with clearance of their nestingforests (Bucher 1992), had significant impacts on theirnumbers by the mid-1800s, and the last known wildbirds were shot in the Great Lakes region of the U.S. atthe end of the century. After that, it was simply a ques-tion of which of the birds in captivity would survive thelongest, and “Martha” in Cincinnati “won.”

The decimation of the vast bison herds in North Amer-ica followed a similar time-line, but here hunting for meat,skins, or merely recreation, was the sole cause. In the1850s, tens of millions of these animals roamed the GreatPlains in herds exceeding those ever known for any othermega-herbivore, but by the century’s close, the bison wasall but extinct (Dary 1974). Following an expensive popu-lation recovery program, both the plains (Bison bison bison)and wood bison (Bison bison athabascae) are currently clas-sified by the 2003 IUCN Red List as Lower Risk/Conser-vation Dependent, although the wood bison is listed asendangered by the U.S. Endangered Species Act and Ap-pendix II of CITES.

An example that is lesser known is the extirpation ofmonodominant stands of Pau-Brasil trees (Caesalpinia echi-nata) from eastern Brazil, a source of red dye and hard-wood for carving that gave Brazil its name. Pau-Brasil

once formed dense clusters along 3,000 km of the Brazil-ian Atlantic forest, and the species sustained the firstmajor trade cycle between the new Portuguese colonyand European markets. It was exploited relentlessly from1500 to 1875, when it finally became economically extinct.Since the advent of synthetic dyes, the species has beenused primarily for the manufacture of high-quality violinbows, and Pau-Brasil specimens are currently largely con-fined to herbaria, arboretums, and a few small protectedareas (Dean 1996). These examples suggest that evensome of the most abundant populations can be driven toextinction in the wild by exploitation. Exploitation of bothlocally common and rare species thus needs to be ade-quately managed if populations are to remain demo-graphically viable in the long term.

People exploit wild plants and animals for a varietyof reasons, which need to be understood if managementis to be effective. There may be more than food andmoney involved. The recreational importance of huntingand fishing in developed countries is well known. Forexample, hunting creates more than 700,000 jobs in theU.S. and a nationwide economic impact of about $61 bil-lion per year, supporting nearly 1% of the entire civilianlabor force in all sectors of the U.S. economy (LaBarbera2003). Over 20 million hunters in the U.S. spend nearlyhalf a billion days afield in pursuit of game, and feeslevied to game hunters finance a vast acreage of conser-vation areas in North America (Warren 1997).

The importance of exploitation as a recreational activ-ity is not restricted to wealthy countries. For example, inreef fisheries in Fiji, capture rates are highest with spearsor nets, and while these techniques are used when fishare in short supply, the rest of the time people adopt amore leisurely pace with less efficient hand-lines, usingthe extended time for social and recreational purposes(Jennings and Polunin 1995). Cultural and religious prac-tices are often important. For example, feeding taboosswitch on or off among hunters in tropical forests accord-ing to availability of alternative prey species (Ross 1978;Hames and Vickers 1982). The fate of some endangeredspecies is closely bound to religious practices, as in thecase of the babirusa wild pig, Babyrousa babyrousa, in Su-lawesi. This species is consumed heavily in the Christian-dominated eastern tip of Sulawesi, but rarely consumedover the Muslim-dominated remainder of the island(Clayton et al. 1997).

Impacts of Exploitation on Target SpeciesMany of the best-known impacts of exploitation on pop-ulations involve cases of direct targeting, whereby hunt-ing, fishing, logging, and related activities are selective,

250 Chapter 8

aimed at a particular species. In this section we presentexamples from major ecosystems in both temperate andtropical areas.

Tropical terrestrial ecosystemsTIMBER EXTRACTION Tropical deforestation is driven pri-marily by frontier expansion of subsistence agricultureand large-scale development programs involving im-proved infrastructure and access. However, animal andplant population declines are typically preceded by hunt-ing and logging activity well before the coup de grâce ofcomplete deforestation is delivered. Approximately 5.8million ha of tropical forests are logged each year (Foodand Agriculture Organization 1999; Achard et al. 2002).Tropical forests account for about 25% of the global in-dustrial wood production, worth U.S.$400 billion orabout 2% of the global gross domestic product (WorldCommission of Forests and Sustainable Development1998). Much of this logging activity opens up new fron-tiers to wildlife and nontimber resource exploitation, andcatalyzes the transition into a landscape dominated byslash-and-burn and large-scale agriculture.

Few studies have examined the impacts of selectivelogging on commercially valuable timber species, andcomparisons among studies are limited because theyoften fail to employ comparable methods that are re-ported adequately. The best case studies come from themost valuable timber species that have already declinedin much of their natural geographic distributions. For in-stance, the highly selective logging of broadleaf ma-hogany (Swietenia macrophylla) is driven by the extraor-dinarily high value of this species on internationalmarkets. These conditions make it lucrative for loggersto pay royalty payments as well as high transportationcosts of reaching remote wilderness areas. Selective log-ging of mahogany and other prime timber species affectsthe forest by creating canopy gaps and causing severecollateral damage due to construction of roads and skidtrails, particularly in the case of mechanized operations.

One of the major obstacles to implementing a sustain-able forestry sector in tropical countries is the lack of fi-nancial incentives for producers to limit offtakes to sus-tainable levels and invest in regeneration (see Essay 8.1by Steve Ball for an example involving East Africanblackwood). Economic logic dictates that trees should befelled whenever their rate of timber volume incrementdrops below the prevailing interest rate (Pearce 1990).Postponing exploitation beyond this point would incuran opportunity cost because profits from logging couldbe reinvested at a higher rate elsewhere (see Chapter 5for a full discussion of economic discounting). This isparticularly the case where land tenure systems are un-stable, and where there are no disincentives against min-ing the resource capital at one site and moving else-

where once this is depleted. This is clearly shown in amahogany study in Bolivia where the smallest treesfelled are about 40 cm in diameter, well below the legalminimum size (Gullison 1998). At this size, trees are in-creasing in volume at about 4% per year, whereas realmahogany price increases have averaged only 1%, sothat a 40-cm mahogany tree increases in value at about5% annually, slowing down as the tree becomes larger(Figure 8.1). In contrast, real interest rates in Bolivia inthe mid-1990s averaged 17%, creating a strong econom-ic incentive to liquidate all trees with any value regard-less of resource ownership. The challenges and prospectsof sustainable forestry in the Tropics are discussed inmore detail in a case study by Pinard and colleagues inCase Study 6.2

SUBSISTENCE HUNTING Humans have been hunting wildlifein tropical forests for more than 100,000 years, but con-sumption has greatly increased over the last few decades.Exploitation of bushmeat (the meat from wild animals) bytropical forest dwellers has increased due to larger numbersof consumers, changes in hunting technology, scarcity of al-ternative sources of protein, and because it is often a pre-ferred food. Recent estimates of annual hunting rates are25,850 tons of wild meat in Sarawak (Bennett 2002),73,890–181,161 tons in the Brazilian Amazon (Peres 2000a),and 1–3.7 million tons in Central Africa (Fa et al. 2001).

Overexploitation 251

25

20

15

10

5

00 40 80 120 160

Prod

ucti

vity

(%)

DBH (cm)

Minimum harvestablesize with value

Bolivian real interest rates

Financial

Biological

Figure 8.1 Productivity of Bolivian mahogany trees as afunction of size expressed as tree diameter at breast height(DBH). Financial productivity is equivalent to size incrementscombined with a 1% real increase in price. Trees equal to orsmaller than 40 cm in DBH have no commercial value, but atthis size they increase at 5% in value per year. To maximizetheir profits, loggers should fell trees at a size when the finan-cial productivity drops below the prevailing interest rate,which in Bolivia was 17% in 1989–1994. The discrepancy be-tween the interest rates and tree growth rate provides astrong incentive to overexploit. (Modified from Gullison1998.)

fpo

Hunting rates are already unsustainably high across largetracts of tropical forests, averaging six times the maximumsustainable rate in Central Africa, for example (Figure 8.2).Consumption by both rural and urban communities isoften at the end of supply chains that are hundreds of kilo-meters long, and that extend into many previously inacces-sible areas (Milner-Gulland and Bennett 2003). The rapidacceleration in tropical forest defaunation due to unsus-

tainable hunting occurred initially in Asia, is now sweepingthrough Africa, and is likely to move to even the remotestparts of the Neotropics. This pattern reflects human demo-graphics in different continents: There are 522 people perkm2 of remaining forest in South and Southeast Asia, 99 inCentral-West Africa, and 46 in Latin America.

Subsistence game hunting affects the structure oftropical forest mammal assemblages, as revealed by vil-

252 Chapter 8

East African Blackwood ExploitationSteve M. Ball, Mpingo Conservation Project

n The Miombo woodlands, whichcover a large swath of southern Africa,are a highly species-diverse ecosystemand the mammalian fauna requireslarge ranges. Thus, large contiguousareas of land must be protected for con-servation of biodiversity to be effective.Even if the protected area system werewell maintained, fragmentation wouldhave a significant impact on theintegrity of the conserved areas. Com-munity-based conservation could fillthe gap, providing buffer zones andcorridors between reserves, but thecommunities need adequate incentive.The East African blackwood (Dalbergiamelanoxylon) has the potential to do this.

The tree gets its name from its beauti-ful dark heartwood, which is inky blackin the best-quality timber. The wood isused in the West to make woodwindinstruments, especially clarinets andoboes, and locally it is a prime choice forwood carvers. It is the most valuabletimber growing in the Miombo wood-lands of southern Tanzania (Figure A).Both the species and its habitat areunder threat from commercial exploita-tion and inward migration of people asa result of recent road improvements.The blackwood’s high economic value,its status as Tanzania’s national tree, andits cultural significance both in the Westand in Tanzania make it an ideal flag-ship species to justify conservation ofthe habitat (Ball 2004).

Current exploitation levels are prob-ably unsustainable, with at least 50% oftrees being felled illegally without alicense. Local forestry officials whodepend on logging companies fortransport into the field do not have suf-ficient resources to enforce existing reg-

ulations. Villagers generally knowwhen illegal logging is taking place andcould apprehend suspects, but theyhave no incentive to do so because allhardwood timbers remain the exclusiveproperty of the government. Intentionalfires are also thought to prevent regen-eration and facilitate heart-rot in adulttrees. Plantations are not an optionbecause of their high costs, uncertainreturn, and long rotation time (black-wood is estimated to take 70–100 yearsto reach timber size).

Community-based forest manage-ment is now a major theme of conser-

vation and rural development activitiesin Tanzania. In largely deforested areasthese can succeed by focusing on fire-wood and water-catchment issues, butthis approach is unlikely to succeed insoutheastern Tanzania where forestcover currently exceeds 70%. Howevernew laws allow for villages to takeownership of local forest resources ifthey can produce a viable managementplan for the forest. This includes tenurerights over even high-value timbertrees, such as East African blackwood,which were previously government-reserved species. License fees for fellingtimber could significantly increase vil-lage income, giving local people anincentive to look after their naturalresource assets. Once schemes are wellestablished, rural communities coulduse the promise of future income as col-lateral against micro-finance loans toincrease the up-front benefits.

Simple economic analysis suggeststhat sustainable, community-basedmanagement of East African blackwoodoffers a powerful argument to justifyconservation of the forest on publiclands and to complement the protectedareas system, providing income inbuffer zones and corridors betweenreserves. However, it will be importantto work with the suppliers of musicalinstrument manufacturers, and avoid asignificant increase to their costs. Ifthese challenges can be met, blackwoodcould be used as an economic key toconserve—as managed areas—largetracts of Miombo woodland. Thiswould not happen through a system offorest reserves, but rather with a moreinclusive strategy of sustainableexploitation of the region’s naturalresources. n

ESSAY 8.1

Figure A A mpingo, an African black-wood tree. (Photograph courtesy of SteveM. Ball.)

fpo-hi res

lage-based kill profiles in Neotropical (Jerozolimski andPeres 2003) and African forests (Fa and Peres 2001). Thiscan be seen in the composition of residual game stocks atforest sites subjected to varying degrees of hunting pres-sure where overhunting often results in faunal biomasscollapses, mainly through declines and local extinctionsof large-bodied mammals (Bodmer 1995; Peres 2000a).

NONTIMBER FOREST PRODUCTS Nontimber forest products(NTFP) are biological resources other than timber thatare taken from either natural or managed forests (Peters1994). Examples of exploited products (of whole plantsor plant parts) include fruits, nuts, oil seeds, latexes,resins, gums, medicinal plants, spices, dyes, ornamentalplants, and raw materials and fiber such as Desmoncusclimbing palms, bamboo, and rattan. Forest wildlife andbushmeat can also be considered as a prime NTFP, butfor a number of reasons they will be treated as a distinctcategory.

The socio-economic importance of NTFP extraction toindigenous peoples cannot be underestimated. Manyethnobotanical studies have catalogued the wide variety

of useful plants or plant parts used by different aborig-ine groups throughout the Tropics. For example, theWaimiri-Atroari Indians of central Brazilian Amazoniamake use of 79% of the tree species occurring in a single1-ha terra firme (upland) forest plot (Milliken et al. 1992),and out of the ≥16,000 species of angiosperms in India,6000 are used for Ayurvedic or other traditional medi-cine, and over 3000 are officially recognized by the gov-ernment for their medicinal uses.

Exploitation of NTFPs often involves the systematicremoval of reproductive units from the population, butthe level of mortality in the exploited population de-pends on the method of extraction and whether vitalparts are removed. Traditional NTFP extractive practices,for either subsistence or commercial purposes, are oftenconsidered to comprise a desirable, low-impact econom-ic activity in tropical forests compared to alternativeforms of land use involving structural disturbance, suchas selective logging and shifting agriculture (Plotkin andFamalore 1992). As such, the exploitation of NTFPs isusually assumed to be sustainable and is viewed as apromising compromise between the requirements of bio-diversity conservation and those of extractive communi-ties under varying degrees of market integration.

A recent study of Brazil nuts (Bertholletia excelsa) ques-tions the standard assumptions about sustainability ofNTFPs (Peres et al. 2003). Brazil nuts are the base of amajor extractive industry supporting over half of the trib-al and nontribal rural population in many parts of theBrazilian, Peruvian, and Bolivian Amazonia, either fortheir direct subsistence value or as a source of income.This wild seed crop is firmly established in domestic andexport markets, has a history of over 150 years of com-mercial exploitation, and is one of the most valuable non-timber extractive industries in tropical forests anywhere.Brazil nuts have been widely held as a prime example ofa sustainably extracted NTFP, yet the persistent collectionof B. excelsa seeds has severely undermined the patternsof seedling recruitment of Brazil nut trees. This has dras-tically affected the age structure of many natural popula-tions to the point where persistently overexploited standswill succumb to a process of senescence and demograph-ic collapse, threatening this cornerstone of the Amazon-ian extractive economy (Peres et al. 2003; Figure 8.3).Nevertheless, the concept of sustainable NTFP extractionis now so deeply entrenched into national resource usepolicy that extractive reserves (or their functional ana-logues) have become one of the fastest growing cate-gories of protected areas in tropical forests (Fagan et al.2005). The implicit assumption is that traditional meth-ods of NTFP exploitation have little or no impact on for-est ecosystems and tend to be sustainable because theyhave been practiced over many generations. However,virtually any type of NTFP exploitation in tropical forests


Associates, Inc.G 3e

es of Conservation BiologyG .eps1 size0

9

10

11

8

7

6

5

44 5 6 7 8 9 10 11

Log

(bus

hmea

text

ract

ed[k

g/yr

–1])

Log (p roduction [kg/yr–1])

Figure 8.2 Hunting rates are unsustainably high across largetracts of tropical forests as seen in the relationship betweentotal extraction and total production of game meat through-out the Congo and Amazon basin (solid and open symbols,respectively) by mammalian taxa. The solid line indicateswhere extraction equals production; the dashed line indicatesexploitation levels at 20% of production, considered to be sustainable for long-lived taxa. Taxon symbols are as follows:ungulates (squares), primates (triangles), carnivores (circles), rodents (inverse triangles), and other taxa (diamonds). (Modi-fied from Fa et al. 2002.)

fpo

will have an ecological impact. The exact extent and mag-nitude of this impact depends on the accessibility of theresource stock, the floristic composition of the forest, thenature and intensity of extraction, and the particularspecies or plant part under exploitation (Peres and Lake2003).

Few studies have quantitatively assessed the demo-graphic viability of nontimber plant products. A boom inthe use of homeopathic remedies sustained by over-col-lection of therapeutic and aromatic plants is threatening atleast 150 species of European wild plants and drivingmany populations to extinction (TRAFFIC 1998). Com-mercial exploitation of the pau-rosa or rosewood tree(Aniba rosaeodora), which contains linalol, a key ingredientin fine perfumes, involves a destructive technique that al-most invariably kills the tree. This species has conse-quently been extirpated from virtually its entire geo-graphic range in Brazilian Amazonia (Mitja and Lescure2000). Chanel Nº5® and other perfumes made with pau-rosa fragrance gained an enormous international marketdecades ago after being popularized by Hollywood starslike Marilyn Monroe, but the number of processing plantsin Brazil fell from 103 in 1966 to fewer than 20 in 1986, dueto the dwindling resource base. Yet French perfume con-

noisseurs have been reluctant to replace the natural pau-rosa fragrance by synthetic substitutes, and the last unex-ploited populations of pau-rosa remain threatened. Thesame could be argued for a number of NTFPs for whichthe exploitation by destructive practices involves a lethalinsult to whole reproductive individuals, such as the ex-traction of fruits and palm-hearts in many arborescentpalms. For example, in the Iquitos region of Peru, thefruits of Mauritia flexuosa palm trees, a long-lived forestemergent, are often collected only once by felling wholereproductive adults (Vasquez and Gentry 1998).

Enthusiasm for NTFPs in community developmentand conservation partly results from unrealistic studies re-porting their high economic value. For example, Peters etal. (1989) reported that the net present value of fruit andlatex extraction in the Rio Nanay of the Peruvian Amazonwas $6,330/ha, assuming a 5% discount rate and that 25%of the crop was not taken. This is in sharp contrast with a30-month study in Honduras that measured the localvalue of foods, construction and craft materials, and med-icines extracted from the forest by 32 Indian households(Godoy et al. 2000). The combined value of consumptionand sale of forest goods ranged from U.S.$18 toU.S.$24 per hectare per year, at the lower end of previousestimates (between U.S.$49 and U.S.$1,089 per hectare peryear). NTFP extraction thus cannot be seen as a solutionfor rural development and in many studies the potentialvalue of NTFPs is exaggerated by the assumption of un-realistically high discount rates, unlimited market de-mands, availability of transportation facilities, and ab-sence of product substitution.

What, then, is the impact of NTFP extraction on thedynamics of natural populations? How does the impactvary with the life history of plants and animals? Are cur-rent extraction rates truly sustainable? These are allquestions that could steer a future research agenda butthe demographic viability of NTFP populations will de-pend on the species’ ability to recruit new adults eithercontinuously or in sporadic pulses while being subject-ed to repeated exploitation.

Temperate terrestrial ecosystemsFORESTRY According to one assessment, only 22% of theworld’s original forest cover remains in large, relativelynatural ecosystems, or so-called “frontier forests”(Bryant et al. 1997). These remaining frontier forests arepredominantly classed as either boreal (48%) or tropical(44%), with only a small fraction (3%) remaining in thetemperate zone. However, much like tropical forests,most of the frontier forests in mid to high latitude re-gions are also increasingly threatened by logging andagricultural clearing (World Resources Institute 1998).

The overall impact of different sources of structuraldisturbance generated by modern forestry in the temper-

254 Chapter 8

sociates, Inc.G

f Conservation BiologyG ps1 ze0

0

40

80

120

160

DB

H(c

m)

Seed harvest level

0

20

10

30

40

50 5

10

5

3

Unharvested Light Moderate Persistant

%of

juve

nile

s(A)

(B)

Figure 8.3 Relationships between historical levels of Brazilnut collection and mean tree size, expressed in terms of DBH,cm (A); and percentage of juvenile trees (B) in different popu-lations. Numbers above bars indicate the number of popula-tions studied throughout the Brazilian, Bolivian, and PeruvianAmazon. (Modified from Peres et al. 2003.)

fpo

ate and boreal zones may depend on the spatial scale andintensity of disturbance, the history of analogous formsof natural disturbance, the groups of organisms consid-ered, and whether forest ecosystems are left to recoverover sufficiently long intervals following a disturbanceevent such as commercial thinning or clearcutting. Thepatterns of landscape-scale human disturbance in theseforests can be widely variable in intensity, duration, andperiodicity, and are often mediated by economic incen-tives to cut timber from high-biomass old-growth forests,rather than natural regrowth or fast-growing tree planta-tions on a long rotation cycle.

The expanding frontier of commercial forestry intohitherto remote, roadless wildlands often results in highlevels of forest conversion and fragmentation of remain-ing stands, with significant impacts to forest wildlife. A re-view of 50 studies in Canadian boreal forests on the effectsof postlogging silviculture on vertebrate wildlife conclud-ed that large impacts are universal when native forests arereplaced by even-aged stands of rapidly-growing nonna-tive tree species (Thompson et al. 2003). Loss of specialstructures, such as snags containing nest cavities and largedecaying woody debris, is particularly important in thedecline of forest species depending on those structures(McComb and Lindenmeyer 1999). Unlogged borealconiferous forests within protected areas in Finland,where population trends of land birds have been excep-tionally well documented, are extremely important to thenative old-growth avifauna, such as the Siberian Jay(Perisoreus infaustus) and the hole-nesting Three-toedWoodpecker (Picoides tridactylus), which have declined inareas under timber extraction (Virkkala et al. 1994). In-deed, logging is often considered to be the most importantthreat to species in boreal forests (Hansen et al. 1991; Im-beau et al. 2001). Some 50% of the red-listed Fennoscandi-an species are threatened because of forestry (Berg et al.1994). Forests actively managed for biodiversity couldsupport 100% of the species occurring in Washingtonstate, whereas timber management on a 50-year rotationat the landscape level could support a maximum of 87%and a mode of 64% of the species potentially occurring inforests (Carey et al. 1996).

In summary, many of the detrimental impacts of for-est management on biodiversity are associated with thelarge-scale structural simplification of the ecosystem atall forest levels, age-class truncation, and other conse-quences of intensive forest management intended to in-crease the yield of the desired forest component.

A reduction of this impact often involves modernprinciples of ecological forestry (Seymour and Hunter1999), which may include a partial removal of the stand,often in a patch mosaic using relatively small clear-cutsizes so that each stage of forest development is repre-sented somewhere in the landscape. As long as most

stages are present at any one time, the requirements ofnearly all species can be met somewhere (although seeexamples in Chapter 7 on the negative effects of habitatfragmentation). Furthermore, clear-cutting may only beused when and where absolutely necessary. Other ex-ploitation methods include thinning and partial cutting,each with different effects on wildlife habitat. Land-scape-level decisions to maximize the biodiversity valueof the forest may include allocating different portions ofthe total area to successively longer rotations, rangingfrom 50 years for short-lived species up to 300 years forlate-successional habitat.

HUNTING Large mammals, small game, and waterfowlare also major targets in temperate countries, whererecreational hunting can be a popular sport across alarge section of the constituents. Annual culls of largeungulates have been rather successful in North America,as shown by population trends, in both controlling pop-ulations and stocking the freezer of the average hunter.White-tailed deer (Odocoileus virginianus), the most com-mon and widespread of wild ungulates in the U.S., in-creased from fewer than 500,000 around the turn of lastcentury, to about 30 million today. Deer are among themost heavily hunted species in the U.S., with some 5million killed by over 10 million hunters every year, gen-erating about 20 billion dollars in hunting-related rev-enue in 2001 (U.S. Fish and Wildlife Service 2002). InTexas alone, it is estimated that the white-tailed deerpopulation numbered more than 3.1 million in 1991 inspite of heavy hunting pressure, and approximately474,000 animals were shot by hunters in that year. Resi-dent birds and small to mid sized mammals frequentlyhunted or trapped for sport are often referred to as“small game.” These may include game birds such asquail, pheasant, partridge and grouse, rodents and lago-morphs such as squirrels, rabbits, and hares, and evencarnivores such as coyotes and raccoons. The effects ofdifferent hunting strategies upon populations of themost popular game bird in the U.S., the bobwhite quail,have been simulated showing that maximum yields canbe sustained from annual capture rates of about 55%(Roseberry 1979). Hunting at such high rates, however,leaves little room for error in calculated bag quotas be-cause it depresses the following spring populations by53% below unexploited levels.

Ducks, geese, and swans are gregarious and often mi-gratory species that also attract enormous attention fromgame hunters. As such, the establishment of annual wa-terfowl hunting regulations is a complex procedureshared by various governmental levels and private or-ganizations, involving thousands of wildlife biologistsand habitat managers under the jurisdiction of wildlifeagencies. For example, retrieved duck and goose har-


vests during 2001 were 19.4 million in the U.S. andCanada, down by 6.6% on the total numbers bagged inthe previous year (Martin and Padding 2002). In the U.S.alone, in 2001, this involved annual sales of 1.66 millionfederal duck stamps to 1.59 million hunters who collec-tively spent nearly 15 million hunter-days in pursuit ofwaterfowl. Needless to say, seasonal hunting license feesand leases of private hunting areas generate welcomecash used for both population management and for pro-tection of habitats against other forms of land use.

The relatively orderly use of wildlife in North Americais not necessarily the rule for all temperate regions. Illegaluse and commercialization of wildlife continue to generatea substantial clandestine traffic—even for species that arefully protected on paper—in several mid- to high-latitudecountries ranging from Chile to Russia. The problem usu-ally lies not in the regulations, which are often already ex-tensive and strict, but rather in lack of law enforcementthat is often attributed to weak institutional capacity.

Aquatic ecosystems MARINE ECOSYSTEMS The impacts of marine fisheries ontarget species are well known. Roughly three-quarters ofthe world’s fish stocks are considered to be fully fished oroverexploited (Food and Agriculture Organization 2002).Since the 1990s global catches have leveled off for the firsttime in human history, despite continuing advances incapture technology (Figure 8.4). Exploitation of aquaticanimals now seems to be following the pattern that oc-curred in many terrestrial ecosystems long ago, with re-liance on hunting of wild animals being supplementedby the captive rearing of domestic stock, through aqua-culture (see Figure 8.4). No one should delude them-selves into thinking that the move toward fish farming

will save wild stocks, as many of the fish that are reared,such as salmon and trout, are fed with meal derived fromwild fish! Indeed, stocks of many wild fishes have con-tinued to decline. Recent surveys of 232 stocks showed amedian decline of 83% over the past 25 years (Hutchings2000, 2001; Hutchings and Reynolds 2004). A stock of At-lantic cod (Gadus morhua) off eastern Canada, which onceseemed absolutely limitless, has undergone a decline of99.9%, since the 1960s, leading to a designation of “en-dangered” under the Canadian Species at Risk Act (Com-mittee on the Status of Endangered Wildlife in Canada2003). This decision is not an isolated case; several otherspecies of commercially exploited fishes have been listedrecently as threatened with extinction (Musick et al. 2000;IUCN 2003). These developments show that some fish-eries concerns are moving from the traditional focus onsustainability into the realm of extinction risk (Reynoldset al. 2002; Dulvy et al. 2003).

Concerns about extinction risk in the sea are most acutefor those targeted species that have combinations of traitsthat make them most susceptible to capture, and biologi-cally least productive (Reynolds et al. 2002; Dulvy et al.2003). For example, the marine species that are most likelyto be targeted are those that occupy shallow waters acces-sible to fishing gear, those that form dense shoals in pre-dictable places, or those that are most valuable. If a specieshas one or more of these characteristics in combinationwith long generation times, the results can be hazardousto the health of the population. The Chinese bahaba (Ba-haba taipingensis) is a fish that meets many of these criteria(Sadovy and Cheung 2003). It is a huge species of croaker(family Scienidae), which may exceed 2 m in length (Fig-ure 8.5). It has traditionally been caught along its coastalrange from Shanghai to Hong Kong. This species has been

256 Chapter 8

80

100

120

60

40

20

0

1950

1953

1956

1959

1962

1965

1968

1971

1974

1977 19

8019

8319

8619

8919

9219

9519

98

Tons

(mill

ions

)

Year

Figure 8.4 Trends in global fisheries.The gray portion of the bars indicatecapture fisheries and the black portionof the bars indicate aquaculture. (FromHart and Reynolds 2002, based on datafrom FAO 2000; corrected for misre-porting of capture fisheries of China byWatson and Pauly 2001.)

fpo

popular for the medicinal properties of its swim bladder.Its numbers have declined to 1% of its abundance in the1960s. During this time its price skyrocketed to the pointwhere swim bladders in 2001 were worth seven times theprice of gold (Sadovy and Cheung 2003). Scientists do nothave enough information to say much about its popula-tion demography, but its large size invariably implies thatit will take many years to reach maturity. This biologicalfeature, combined with such strong economic incentivesfor people to catch it, have conspired to push the speciesto the edge of extinction.

The same basic rules of vulnerability for fish speciesapply to other taxa. For example, species of abalonealong the Pacific coast of North America often occur inshallow waters, where they are readily accessible todivers (Figure 8.6). There has been serial depletion ofthese species, beginning with the most valuable andthen moving on to less valuable ones (Tegner et al. 1996).This is reminiscent of the pattern that occurred with the

great whales before the International Whaling Commis-sion’s moratorium on commercial whaling in 1986. Thewhite abalone (Haliotus sorenseni) has been hit particu-larly hard. Ranging from California to Baja California,white abalone densities in some locations during theearly 1970s were estimated at 1,000–5,000 per acre (Laf-ferty 2003). Commercial and recreational capture re-duced its numbers to less than one per acre by the 1990s.In May 2001 this species became the first marine molluscto come under the U.S. Endangered Species Act, with anestimated population in the wild of only about 3,000 in-dividuals. Most remaining animals are restricted to deepwaters beyond the reach of fishing (Lafferty et al. 2003).While the abalones’ value and accessibility provided themotive and the means for overexploitation, it has alsosuffered from an additional problem: the Allee effect.This is the phenomenon whereby per capita fitness de-clines as a population becomes smaller (see Chapter 12).The ensuing feedback can cause populations to becomemore vulnerable as they become increasingly rare, po-tentially spiraling to extinction. In the case of abalone,the Allee effect arises through the need for individuals tobe within 1 m of each other for a male’s sperm to reach afemale. Biologists have had some success with artificialfertilization in captivity, with the intention of rearingabalones for release into the wild.

FRESHWATER ECOSYSTEMS Many freshwater taxa are sub-ject to exploitation for food and a variety of additionalproducts. In 2000, the global estimate for all inland cap-ture fisheries was estimated at 8.8 million tons (FAO


Figure 8.5 Chinese bahaba (Bahaba taipingensis) caught as anincidental by-catch by a trawler west of Hong Kong. (Photocourtesy of Cheng Tai-sing.)

fpo-hi res

Figure 8.6 White abalone (Haliotis sorenseni) underside.(Photograph courtesy of K. D. Lafferty.)

fpo-lo res

2002). This figure is considerably less than the estimatefor inland aquaculture (22.4 million tons). In many tem-perate countries, recreational fishing in fresh waters is amajor past-time, yielding an estimated 2 million tons offish (Cowx 2002). In 1996 it was estimated that 35 millionpeople in the U.S. spent 514 million angler-days fishing,spending U.S. $38 billion in the process (U.S. Fish andWildlife Service 1997). Among 22 European countries,the number of anglers was estimated at 21.3 million(Cowx 1998).

The world’s salmon species illustrate the vulnerabili-ty of fish species that spend all or part of their life cycle infreshwater. Populations of four of the seven species ofeastern Pacific salmon and trout in the genus Oncorhyn-cus are currently listed under the U.S. EndangeredSpecies Act. Over-fishing has contributed to severe de-clines in many populations, often in combination withhabitat loss through construction of dams that block theirmigrations, as well as degradation of spawning streamsdue to forest clearance and water abstraction (Lynch et al.2002). The plight of salmon species is particularly sober-ing when one considers the enormous amount of atten-tion that has been paid to these populations by scientistsand conservationists, as well as their high economic andcultural significance. The most recent response of the U.S.government at the time of writing (May 2004) has beenthe announcement of the intention to count hatchery-re-leased fish as “wild” (Kaiser 2004). Thus, the hundreds ofmillions of fish that are reared in captivity each year willelevate the population counts of many of the 27 popula-tions of Pacific salmon and cutthroat trout that are en-dangered. This can lead to their removal from protectionunder the Endangered Species Act and presumably, thediscontinuation of many habitat restoration programs.The decision ignores the fact that hatchery fish becomedomesticated rapidly, showing genetic and phenotypicdivergence from wild stocks (e.g., Fleming et al. 1994;Heath et al. 2003). Furthermore, there is little evidencethat hatcheries enhance the viability of wild stocks(Myers et al. 2004). Yet the regional director of the Na-tional Marine Fisheries Service, which has jurisdictionover salmon management, has declared that “Just as nat-ural habitat provides a place for fish to spawn and to rear,also hatcheries can do that...” Presumably, zoos can alsobe considered natural habitats.

Fish are not the only taxa exploited from fresh waters.During the late 1800s and early 1900s millions of fresh-water mussels (family Unionidae and Margaritiferae)were collected from freshwaters of Canada and the U.S.,primarily to make buttons from their shells (Williams etal. 1993; Helfrich et al. 2003; Williams and Neves 2004).This exploitation was rarely sustainable, due to the slowgeneration times of these species. Mussels were savedfrom major impacts of exploitation by the switch to plas-

tics for manufacturing buttons during the 1940s. How-ever, today several million kilograms of mussels are stillexported annually to Asia, where small beads are madefrom their shells and inserted into other bivalves for theproduction of pearls. Unfortunately, mussels now facemuch more critical threats from the “usual suspects” infreshwater conservation: pollution, damming, dredging,channelization, siltation, and competition with nonna-tive bivalves such as the Asian clam (Corbicula fluminea)and the zebra mussel (Dreissina polymorpha) (Strayer etal. 2004). These problems have led to 72% of the 297species that occur in the U.S. being considered endan-gered, threatened, or of special concern, including 21species that are extinct (Williams and Neves 2004).

Impacts of Exploitation on NontargetSpecies and Ecosystems Most hunting, fishing, and collecting activities affect notonly the species that are the primary targets, but alsothose that are taken accidentally or opportunistically.Furthermore, exploitation may cause physical damageto the environment, and also may have ramifications forother species through cascading interactions, phaseshifts in the structure of the ecosystem, and changes infood webs. Here we discuss a few examples of how ex-tractive activities targeted to one or a few species candrastically affect the structure and functioning of terres-trial and aquatic ecosystems.

Tropical terrestrial ecosystemsLOGGING AND FOREST FLAMMABILITY Even logging target-ed to a single timber species can puncture the forestcanopy and increase the density of treefall gaps. This in-crease can trigger major ecological changes by increasinglight and creating a warmer and drier microclimate in theunderstory, which thereby affects the dynamics of plantregeneration, and increases forest susceptibility to firedisturbance. In fact, even highly selective logging opera-tions with modest levels of incidental damage to nontar-get trees can generate enough structural disturbance togreatly augment understory desiccation and dry fuelloads, thereby breaching the forest flammability thresh-old (Holdsworth and Uhl 1999; Nepstad et al. 1999). Anysource of ignition during subsequent severe droughts caninitiate extensive ground fires that will dramatically re-duce the functional and biodiversity value of previouslyunburned tropical forests (Barlow and Peres 2004). Sur-face wildfires that are at least partly induced by loggingdisturbance currently threaten millions of hectares ofAmazonian, Mesoamerican, and Southeast Asian forests(Cochrane 2001; Siegert et al. 2001). Despite these unde-sirable direct and indirect effects, large-scale mechanized

258 Chapter 8

logging operations continue unchecked in many season-ally dry tropical forest regions.

HUNTING AND LOSS OF SEED DISPERSAL SERVICES Successfulseedling recruitment in many flowering plants dependson seed dispersal services provided by large-bodied fru-givores (Howe 1984). In tropical forests, the proportionof plant species with an endozoochorous dispersalmode (bearing seeds dispersed by an animal’s digestivetract) is often more than 90% (Peres and Roosmalen2002). Undispersed seeds simply fall to the ground un-derneath the parent’s canopy and have a low survivalprobability (Augspurger 1984; Chapman and Chapman1996). For example, 99.96% of Virola surinamensis seedsthat drop under the parent are killed within 12 weeks(Howe et al. 1985). Many studies have shown lowerseed mortality rates caused by fungal attack or verte-brate and invertebrate seed predators are lower atgreater distances from parents. However, there havebeen only a few studies that have examined the effectsof removing seed dispersers are lower on the demogra-phy of gut-dispersed plants. Wright et al. (2000) ex-plored how hunting alters seed dispersal, seed preda-tion, and seedling recruitment for two palms (Attaleabutyraceae and Astrocaryum standleyanum) in Panama.They found that where hunters had not reduced mam-mal numbers, most seeds were dispersed away from theparent palms, but were subsequently eaten by rodents.Where hunters had reduced mammal abundance, fewseeds were dispersed, but these tended to escape rodentpredation. Thus, seedling density increased by 3–5-foldat heavily hunted sites compared to unhunted sites. Incontrast, Asquith et al. (1999) demonstrated that recruit-ment of Hymenaea courbaril required scatterhoarding oftheir large seeds by agoutis (Dasyprocta sp.), and re-cruitment rates of many plant species that produce verylarge seeds cached by rodents is likely to be very low inheavily hunted areas.

Studies examining seedling recruitment under differ-ent levels of hunting pressure (or abundance of large-bod-ied seed dispersers) reveal very different outcomes. At thecommunity level, seedling density in overhunted forestscan be indistinguishable, greater, or less than that in theundisturbed forests (Dirzo and Miranda 1991; Chapmanand Onderdonk 1998; Wright et al. 2000), but the conse-quences of increased hunting pressure to plant regenera-tion is likely to depend on the target species. In persist-ently hunted Amazonian forests, where large primates areeither driven to local extinction or severely reduced innumbers, the probability of effective dispersal of large-seeded plants ingested primarily by these frugivores candecline by more than 60% compared to nonhunted forests(Peres and Roosmalen 2002). However, more conclusiveevidence is required before the importance of the loss or

reduction of effective animal dispersal services can beproperly understood for different plant species.

Temperate ecosystemsHigher order interactions resulting from selective ex-tinction or severe population declines of large mammalsthat play an important role as landscapers at the ecosys-tem scale have been documented in the temperate zone.Beavers (Castor canadensis) and their Eurasian congener(Castor fiber) are prime examples of ecosystem engineers,which can be defined as organisms that have the poten-tial to dramatically alter the structure and function ofecosystems at large spatial scales. Large ponds createdby the labor-intensive stream-damming activity ofbeaver colonies create large-scale semi-permanent flooddisturbance that drastically changes the structure andpatch dynamics of wetlands (Naiman et al. 1986; Wrightet al. 2002). Beavers were once locally abundant in manyparts of North America and Eurasia but their popula-tions plummeted due to the pelt trade and habitat con-version in the nineteenth and early twentieth century,radically changing wetland ecosystems. No one knowsthe precise extent of these changes but it is clear thatbeaver damming activity has profound effects on thebiogeochemistry of wetland systems (Naiman et al.1994), the dynamics of shifting successional mosaics ofaquatic patches (Johnston and Naiman 1990), and ulti-mately the population dynamics of other wetlandspecies including waterfowl (McCall et al. 1996). Beaversare now making a gradual comeback in many parts ofNorth America (although they were exterminated by1700 and are yet to be reintroduced in many parts of Eu-rope including Britain) but they often continue to be per-ceived as a nuisance requiring controversial measures ofpopulation control.

In temperate terrestrial ecosystems, large mammalsthat once had profound large-scale effects on the struc-ture of plant communities but have been hunted to nearextinction in historic times include bison, bears, andwolves. Joel Berger and colleagues (2001) demonstrateda cascade of ecological events that were triggered by thelocal extinction of grizzly bears (Ursus arctos) and wolves(Canis lupus) from the Yellowstone ecosystem. These in-clude large increases in the population of a riparian-de-pendent herbivore, the moose (Alces alces), the subse-quent alteration of riparian vegetation structure anddensity by ungulate herbivory, and the coincident re-duction of Neotropical migrant birds in the affected wil-low communities, including riparian specialists such asGray Catbirds (Dumetella carolinensis) and MacGillivray’sWarblers (Oporornis tolmiei).

Where large predators (wolves, bears) have been re-moved through predator control programs, or otherforms of direct persecution, ungulate populations can


greatly increase in size, with subsequent impacts onplant communities. White-tailed deer in the U.S. andreindeer in Europe were once controlled by wolves andbear, and despite annual harvests, their numbers arelarger than they were when controlled by native preda-tors. The result is that their larger populations may over-browse forests, which can cause extensive damage insome places (e.g., Väre et al. 1996 and Horsley et al.2003). Particularly in areas where hunting is prohibited,deer populations can become so large that most tree andherb seedlings are consumed, which may lead to achange in patterns of forest succession (e.g., Horsley etal. 2003 and Pedersen and Wallis 2004), or to decline ofrare species (e.g., Gregg 2004). Plant populations can beslow to recover from episodes of excessive herbivory.For example, a population of the rare orchid showy ladyslipper (Cypripedium reginae) in West Virginia lost up to95% of all stems during a 3-year period of excessive deerbrowsing, from which the population took 11–12 yearsto recover (Gregg 2004). Beyond effects on plants, un-controlled herbivore populations can have indirect ef-fects on decomposer (Wardle and Bardgett 2004) andspider communities (Miyashita et al. 2004).

Aquatic ecosystemsMARINE ECOSYSTEMS Marine fisheries are estimated tohave a global by-catch of roughly 27 million tons annu-ally, which is between one-third and one-fourth of thetotal marine landings (Alverson et al. 1994). This is prob-ably a conservative estimate. Most of these by-catcheswere from trawl fisheries, followed by drift nets and gillnets. Shrimp trawls, with their small mesh sizes, accountfor roughly one-third of the total by-catch, with ratios ofweight of discarded animals per weight of shrimpcaught typically about 5:1. It has been estimated thatshrimp trawlers in the Australian northern prawn fish-ery typically discard 70,000 individual organisms duringeach night of fishing.

Some by-catches are threatening species with extinc-tion. All but two of the world’s 21 species of albatross areconsidered threatened with extinction, and most of thesehave been severely affected by longline fisheries in theSouthern Ocean (IUCN 2003). Albatrosses, as well asother seabirds such as petrels, drown when they takebaited hooks as the lines are being set in fisheries aimedat species such as the Patagonian toothfish (Dissostichuseleginoides). While the rate of accidental capture per hookis very low, the potential risks are enormous when scaledup by the total fishing effort, with individual vesselsoften setting many thousands of hooks each day, and atotal of over 250 million hooks being set each year sincethe 1990s south of 30o South (Tuck et al. 2003). Long-livedspecies such as seabirds have extremely low resilienceagainst elevated adult mortality, due to their very long

generation times and low rates of productivity. Recent re-assessments by the IUCN (2003) have led to six species ofalbatross recently being “upgraded” toward more se-verely threatened status, with fisheries by-catch playinga significant role in each case. These problems are notconfined to the Southern Ocean. For example, it has beenestimated that 10,000 Black-footed Albatrosses (Phoebas-tria nigripes) may be killed each year in the central NorthPacific (Lewison and Crowder 2003). Measures adoptedto reduce mortality on seabirds include bird-scaring de-vices, the release of baited lines from below the watersurface (where albatrosses cannot reach them), use ofheavy lines that sink immediately, and use of fish oil onthe surface, which dissuades seabirds from landing onthe water.

Sea turtles are caught incidentally by longlines andtrawlers. For example, it has been estimated that 20,000turtles have been killed each year in the Mediterraneanin longlines set for swordfish. Turtle excluder devices arenow in widespread use in trawl fisheries, and these re-duce turtle mortality by providing an exit flap that tur-tles can push through, while retaining the fish. Turtlesare not the only reptiles taken as by-catch. It has been es-timated that 120,000 sea snakes are taken annually byprawn trawlers in the Gulf of Carpentaria, Australia.

There are mounting concerns about the impacts of by-catches on populations of many fish species as well. Thebest-known cases involve sharks, skates, and rays (seeCase Study 8.1 by Julia Baum). These fishes are towardthe seabird–turtle end of the life-history continuum. Forexample, the European common skate (Raja batis) doesnot reach maturity until it is 11 years old. They used tobe taken by the thousands as by-catch in prawn trawls inthe Irish Sea (Brander 1981), but only six individualswere captured by extensive government bottom fish sur-veys between 1989 and 1997 (Dulvy et al. 2000). In thissame region, there is evidence for complete local extinc-tion of one and possibly two more species.

FRESHWATER ECOSYSTEMS In fisheries, it is easy to findfreshwater equivalents of the kinds of difficulties thatface many marine fish species. An example of nontargetfishes is the Mekong giant catfish (Pangasianodon gigas)(Figure 8.7), a species that is in a similar situation to theChinese bahaba discussed earlier. This fish is restricted tothe Mekong River Basin in Thailand (where no individu-als have been caught since 2001), Laos, and Cambodia.Studies by Zeb Hogan and colleagues (2004) have shownthat individuals can reach 3 m in length and weigh 300kg, and are therefore extremely valuable to fishers. Fur-thermore, they are migratory, thereby running a gauntletof nets set for other species. Finally, their only spawninggrounds, in whirlpools and rapids in the Chiang Khong-Chiang Saen region of the Thai–Laos border, are being

260 Chapter 8

dynamited as part of a plan to enhance river navigation.Conservation biologists have set up a scheme wherebyfishers telephone them day or night if they catch aMekong giant catfish, which the biologists then purchaseat market price, tag, and release away from the nets. Thisis a stop-gap measure, being used to save a small numberof fish while working on longer-term conservation meas-ures. The prospects for this species do not look good, andits threat status has now been raised to “critically endan-gered” by the IUCN (2003).

Overexploitation can also have unforeseen effects onbiodiversity through trophic interactions among speciesin an ecosystem. A freshwater example that is currentlyreceiving a great deal of attention involves the effects ofreduced numbers of Pacific salmon on productivity ofstreams and their riparian vegetation in northwesternNorth America (e.g., Cederholm et al. 1999). All nativePacific salmon species die after they spawn, and theirdecomposing bodies provide a large proportion of thenutrients received by streams in their range. These nu-trients have been acquired when the fish were growingduring the marine phase of their life cycle. Experimentalstudies have shown a variety of effects of nutrients from

salmon carcasses. For example, carcass enrichment of ar-tificial streams has led to an eight-fold increase in densi-ties of macroinvertebrates, which also increased twenty-five-fold in artificially enhanced natural streams (Wipfliet al. 1998). These effects have been shown to have im-portant feedback to productivity of the streams forsalmonids themselves, because juvenile salmon prey onmany of these insects (Wipfli et al. 2003). Studies havealso shown higher growth of riparian trees such as whitespruce, Picea glauca (Helfield and Naiman 2002). Indeed,there has been preliminary success in matching recordsof the sizes of salmon runs in southeastern Alaska from1924 to 1994 with tree-ring growth in Pacific coastal rain-forests (Drake et al. 2002). All of these examples implythat stream ecosystems, as well as associated terrestrialcomponents, will have been affected strongly by themany extinctions and depletions of salmon populationsthat have occurred in recent decades.

Biological Theory of Sustainable ExploitationThe examples in the preceding sections give a flavor ofthe great diversity of forms of exploitation and impactson species and ecosystems. Some of these forms of ex-ploitation, such as fisheries, hunting, and forestry in tem-perate countries, have been occurring under scientifical-ly informed management. Because this is a chapter on“overexploitation,” most of the examples that we havechosen have not been very encouraging. In fact, speciesand ecosystems show a wide variety of responses to ex-ploitation. In this section we review the theory that hasbeen advanced to explain how populations and ecosys-tems respond to exploitation, and tie this to manage-ment options.

Biological populations are by definition renewable,but why should we ever expect plant and animal popu-lations to be able to withstand the elevated mortality thatoccurs with most forms of exploitation? The key is theability of birth or death rates to compensate when we re-move individuals from the population. As populationsare reduced by exploitation, there may be reduced com-petition for food, territories, shelter, and a lower trans-mission rate of diseases. This can lead to greater birthrates or enhanced survival. The tendency for such com-ponents of fitness to limit populations at high density isknown as density dependence (Figure 8.8). This does notmean that populations do not experience episodes ofdensity-independent growth, as would occur after sud-den environmental changes, for example. But the ten-dency of populations to fluctuate around some sort ofmean value, rather than growing indefinitely, can be at-tributed to density dependence.


Figure 8.7 Mekong giant catfish, Pangasianodon gigas, fromthe Mekong River. (Photograph courtesy of Zeb Hogan.)

fpo-hi res

To understand sustainable and unsustainable levelsof exploitation, we need to ask whether removal of indi-viduals is equivalent to “thinning” the population, there-by allowing the survivors to grow more quickly or sur-vive better, or is the removal occurring too rapidly forpopulations to compensate.

The simplest way to ask about the ability of a popula-tion to compensate for elevated mortality is to start withthe logistic model, which gives the number of individu-als at time t as:

[8.1]

Here Nmax is the maximum population size (Figure 8.9A).This is often called the “carrying capacity,” or equilibri-um population size. None of these terms is meant toimply that populations remain stable, but they conveythe idea of some sort of average size over a given periodof time. N0 is the initial number of individuals. The pa-rameter r is the intrinsic rate of natural increase, that is,the difference between per capita birth and death rates atsmall population sizes where there is no density depend-ence. For many species we may want to substitute bio-mass for number of individuals, as in the case of fisheries,in which case we often see the terms B substituted for N.

Figure 8.9A shows the classical population growthrate that matches the logistic equation. This is the sort oftrajectory that we would expect, for example, if we put afew Daphnia into a tank of water with phytoplankton as

food. The population grows slowly at first, because thereare few individuals producing offspring, but growth ac-celerates up to a point where the animals start runningout of food, whereupon growth stops. In the real world,populations will be buffeted by changing environmentalconditions, interactions with their predators, and so on.These can cause population fluctuations, cycles, or crash-es. But the logistic is still a reasonable “default” option,conveying the essence of the potential for density-de-pendence. Specific details of the biology of species willdetermine whether the initial upward slope and final de-cline due to density dependence are steeper, shallower,or occur sooner or later than shown here.

The growth rate of this population before exploitationcan be considered its surplus production, g(N), and isgiven as:

[8.2]g N rN NN

( )max

= −

1

NN

NN

et

rt=

+ −

−

max

max1 10

262 Chapter 8

c.

on BiologyG

Per

capi

tara

teof

mor

talit

yor

repr

oduc

tion Reproduction

Mortality

EquilibriumPopulation density

r

Figure 8.8 Density dependence stems from relationships be-tween population density and per capita rates of birth anddeath. The difference between these at a small population sizeis the intrinsic rate of natural increase, r, and the density atwhich these rates are balanced is the equilibrium populationsize.

fpo

Sinauer Associates, Inc.Groom 3ePrinciples of Conservation BiologyGro0809.eps100% of size01.17.05

Popu

lati

onnu

mbe

r(N

)

Time0

Nmax

Sust

aina

ble

yiel

d(Y

)

Population number (N)

Maximumsustainable yield

0

YMSY

Nmax

(A)

(B)

Figure 8.9 (A) Logistic population growth of a populationup to a maximum population size, Nmax. (B) Sustainableyield, Y (surplus production) against population size for thelogistic case shown in (A). The maximum sustainable yield(YMSY) occurs at 50% of the maximum population size.

fpo

This indicates, sensibly, that as the number of individu-als, N, approaches the maximum population size, Nmax,there is no growth, and even more sensibly, that growthstops when the population size is zero.

If the population in Figure 8.9A is being exploited at asteady rate, then its rate of change per unit time, dN/dtwill be the difference between its surplus productionand the yield, Y, for a given level of exploitation:

[8.3]

The population’s surplus production is the yield that canbe removed sustainably. We can plot this propertyagainst different population sizes (Figure 8.9B). Thisgives us the classic dome-shaped yield curve that under-pins all models of exploitation. This shows that the max-imum sustainable yield (YMSY) occurs at an intermediatepopulation size, which coincides with the inflection pointin the logistic curve shown in Figure 8.9A. This makes in-tuitive sense: We can take the most from a populationwhen it is at the size where it can grow most quickly. Thedome does not have to be symmetric; a failure of the lo-gistic assumption can cause it to lean to the left or to theright. For example, density-dependent processes such ascannibalism, competition, predation, or disease mayoccur at smaller or larger population sizes than depictedin Figure 8.9A (Sutherland and Gill 2001). If the curveleans to the right, we should allow the population to re-main at higher numbers.

Stability of exploitationIn theory, we have discovered how many individuals wecan take from the population to maximize the yield. Butin practice, we will have a very difficult time taking ex-actly that number. For one thing, people rarely do whatthey are told, and common experience suggests that wehave to assume that the population will probably be ex-ploited at a higher rate than we recommend. Even thebest-controlled fisheries and hunts are subject to the va-garies of uncertainty in the population estimates, or un-foreseen circumstances such as weather conditions caus-ing populations to behave in unpredictable ways. Thequestion is, what happens when (not if) the populationis exploited at a rate that differs from the one that theorysuggests should be maximally sustainable? The answerdepends on how the exploitation rate is managed.

Constant quota exploitationOur formulation implies a system of exploitation in whichthe number of individuals removed is independent of thepopulation size. That is, in Equation 8.3 we have subtract-ed Y from the surplus yield, rather than making Y pro-portional to N. This case is typically known as constant

quota because the numbers removed are “constant” in thesense of being independent of the population size. Exam-ples include fisheries that set quotas on numbers of fishthat are caught, regardless of subsequent changes in thenumber of fish in the sea, or hunters who are given fixedlimits that do not vary as the number of target animalsgoes up or down over time. This may sound extreme, be-cause common sense suggests that quotas should be ad-justed according to such changes in the quarry. And it isextreme. And so was the collapse of Peruvian anchovy inthe 1970s, which occurred in part because people did notfully appreciate the implications of this assumption (seethe discussion later in this chapter).

The constant quota situation is depicted in Figure8.10, where each of three potential quotas runs horizon-tally across the graph, implying no relationship withpopulation size. First, consider the high-quota line. Here,the yield exceeds the population’s surplus productioncapability, perhaps because of illegal hunting activity, orbecause the population’s resilience had been over esti-mated. The arrow beneath this line shows that with con-stant exploitation at this rate, the population will becomeextinct. Year after year the quota exceeds the popula-tion’s ability to keep up with the elevated mortality. Analternative that seems more sensible is the MSY quota.However, this is a very risky target because it is impossi-ble to get precise measurements of either the surplusproduction curve, or N. Furthermore, as we have ar-gued, even if this were possible, there is still a goodchance that we will be unable to set quotas sufficientlyaccurately to score a bulls eye when we shoot for theMSY point. If the population is at or smaller than NMSY,the quota will exceed its surplus production and it willdecline to zero. On the other hand, the NMSY point willbe stable if the population is initially higher, because al-

dNdt

rN NN

Y= −

−1

max



Yiel

d(Y

)

NMSY


High quota

MSY quota

Low quota

0

YMSY

NmaxN1 N2

Figure 8.10 Equilibria and population stability under con-stant quota exploitation. The arrows indicate the directions ofchange in population size for each quota.

fpo

though the MSY quota is initially higher than surplusproduction, as the population declines it will benefitfrom reduced density dependence. Still, given the diffi-culty of estimating MSY with precision, such a quota isdangerous. Finally, consider the low quota example. Ifthe population is initially below N1 it will still crash.However, if it is between N1 and N2, then its productionwill exceed the quota, and it will grow to N2. If it is ini-tially higher than N2, it will be driven down to N2 by thequota exceeding its surplus production. So N1 is an un-stable equilibrium and exploitation in that region shouldbe avoided. In contrast, as long as we are sure the popu-lation is well above N1, exploiting it with the low quotaindicated would allow it to grow to a stable equilibriumat N2. Here, finally, we have sustainable exploitation.

Proportional (constant effort) exploitationA much more sensible way to set exploitation targets isto tie them directly to the size of the population. In prac-tice, this is always bound to happen to some extent, be-cause as plants and animals become scarce, people tendto switch to alternative species or other activities, even ifno one is forcing them to do so. For example, local de-mand for game species in tropical forests tends to in-crease sharply when catches in the marine fishing sectorfair poorly (Brashares et al. unpublished data), and trop-ical forest hunters become heavily reliant on smaller-bodied game species once large-bodied species are de-pleted (Jerozolimski and Peres 2003). Conservationistsand resource managers amplify this kind of commonsense by encouraging or forcing people to reduce pres-sures on populations that reach low abundance. So, un-like the case of constant quota above, we have exploita-tion effort that is proportional to the population size.

In this scenario the size of the yield, Y, will be equal tothe exploitation rate, E, multiplied by the populationsize, N:

Y = EN [8.4]

From Equations 8.2 and 8.3 we know that a steadystate will occur between the yield and the surplus pro-duction when g(N) = Y. This means that g(N) = EN.Therefore,

[8.5]

We can rearrange this equation to find the equilibriumpopulation size for any rate of exploitation, as long as Eis below r:

[8.6]

These equilibria are shown in Figure 8.11. The advantageof this form of management is that as long as the ex-

ploitation rate is below the intrinsic rate of natural in-crease, r, then all equilibria are stable. For example,whereas the high quota crashed the population underthe constant quota scenario in Figure 8.10, here we findthat if the population is initially below this removal rate,it will increase to N1, and if it is above this point, it willdecrease to N1. The MSY is also stable, and it still occursat half of the maximum population size.

Figure 8.12 compares the risk of extinction from ex-ploitation based on either constant quotas or proportion-al rates for a study of the American marten (Martes amer-icana) in southern Ontario, Canada (Fryxell et al. 2001).The marten is a member of the weasel family (Mustel-idae) found primarily in coniferous forests where it preyson a wide variety of small vertebrates as well as some in-vertebrates. Marten are trapped for their fur and trappersare granted licenses for exclusive access to trappinggrounds. Fryxell et al. (2001) use a simulation model ofdata from commercial trapping from 1972 to 1991 to eval-uate effects of different types of harvest. This analysisshowed that whereas exploitation in proportion to thepopulation size has a negligible chance of causing localextinction, this risk becomes quite high for moderate lev-els of exploitation under a constant quota system thatdoes not track the population.

Threshold exploitationThe final class of exploitation strategies involves the useof population size thresholds to determine not only therate of exploitation but also whether exploitation shouldtake place at all (Lande et al. 1997, 2001). All populationsare subject to random (stochastic) variations, for exam-

N N Er

= −( )max 1

rN NN

EN1−

=

max

264 Chapter 8


Yiel

d(Y

)

NMSY


High exploitation

Very high exploitation

MSY exploitation

Low exploitation

0 NmaxN1 N2

Figure 8.11 Equilibria and population stability under pro-portional exploitation (also called “constant effort exploita-tion”). Here, a constant fraction of the population is removed.The dashed lines have slopes E corresponding to differentrates of exploitation. The arrows indicate the directions ofchange in population size for each scenario.

fpo

ple, due to environmental variation. In the absence of ex-ploitation, they will sometimes exceed their carrying ca-pacities temporarily. We can take advantage of this bytaking the entire “surplus” whenever the population isabove its carrying capacity, but otherwise ceasing ex-ploitation completely. This would maximize the cumu-lative yield while minimizing the chances of collapse. Amore precautionary variant on this is to remove only aproportion of the surplus. This is an even safer versionof the proportional exploitation method outlined abovebecause it maintains the population near its carrying ca-pacity. However, while such low rates of exploitation areexcellent for conservation, it is difficult to convince peo-ple to accept such severe restrictions.

BioeconomicsUnderstanding biological constraints is necessary toachieving a sustainable level of exploitation. But as wesaw earlier in the chapter, we also need to understand thehunters’ incentives and disincentives if we are to under-stand their impacts on the hunted, and provide manage-ment advice to mitigate such impacts. Bioeconomic mod-els incorporate the costs and benefits of exploitation. Evenin societies where exploitation is a necessity rather than asource of revenue, as in subsistence hunting, a cost–bene-fit framework can be very useful for understanding peo-ples’ behavior (see Essay 8.2 for an example).

Open access and the tragedy of the commonsThe “tragedy of the commons” (Hardin 1968) providesa powerful explanation for a lot of the damage that peo-

ple are doing to the environment and to each other.Imagine you have sole access to the trout in a lake onyour property. You will probably manage these troutquite carefully, because if you take too many, you will bethe one who suffers in future. But if the pond straddlesthe boundary with your neighbor, then each of youshould only take half the number you would take if youhad exclusive access. Will you both be so prudent? Un-less you get along very well with each other, probablynot, because each person’s self-restraint can be exploitedby the other one. Imagine the outcome if we scale thisexample up to the North Sea, bordered by many coun-tries, each with thousands of fishers, all competing forthe same fish. The sea is a “common” fishing ground,and indeed, the massive over-fishing of the past centuryhas been tragic for both fishers and their prey.

We can incorporate the tragedy of the commons intothe scheme developed in the previous sections by assum-ing that the benefits from exploitation are directly propor-tional to the yield. So the dome in Figure 8.13 representsthe benefits. Assume that the costs of exploitation are pro-portional to the effort necessary to obtain that yield. Forexample, in a fishery the more days spent at sea, thegreater the costs of fuel and labor. This is depicted by thestraight line rising from the origin. If we ignore the costs,the maximum benefits will be found at EMSY, or the effortthat provides the maximum sustainable yield. However,if the goal is to maximize profits, this would lead to alower exploitation rate corresponding to the effort atwhich the difference between benefits and costs is maxi-mized, at EP. But the most important fundamental truth of


ssociates, Inc.G

of Conservation BiologyG ps1 ze

1.0

0.4

0.2

0.0200 250 300 350 400 450

Ove

rhar

vest

ing

risk

Mean number of martens taken

Proportional quota

Constant quota

Figure 8.12 Probability of overexploitation (local extinction)in relation to mean yield of marten in commercial trapping insouthern Ontario, Canada. The proportional quota line(black) refers to yields that track annual changes in the popu-lation whereas the constant quota line (gray) is based on thesame absolute number of individuals being trapped each yearregardless of population size.

fpo

Yiel

d(b

enef

its)a

ndco

sts

EMSY

Exploitation effort

Yield

Costs

0

YMSY

YP

YC

EP EC

Figure 8.13 The economics of exploitation, represented bycosts that are proportional to effort, and benefits proportionalto yields. At EP the profits are maximized, at EMSY the maxi-mum sustainable yield is obtained, and at EC the costs areequal to the benefits. Note that unlike Figures 8.10 and 8.11,the x-axis here represents fishing effort (which increases fromleft to right, corresponding to population number decreasingfrom left to right).

fpo

266 Chapter 8

Using Economic Analysis to Bolster Conservation EffortsMarine Aquaria and Coral Reefs

Gareth-Edwards Jones, University of Wales, Bangor

n Advances in technology combinedwith increased interest in marine sys-tems have contributed to a largeincrease in American householdskeeping marine aquaria. As part ofthe large global business associatedwith establishing and maintainingthese aquaria, approximately 350million fish are harvested annuallyfrom the wild and sold worldwidewith a value of $963 million (Young1997). This continued wild harvest isnecessary as some species of aquar-ium fish do not breed well in captiv-ity. Some of the more popular aquar-ium species are associated with coralreefs, and 85% of aquarium fishes arecaught from coral reefs around thePhilippines and Indonesia. Catchingthese fishes can be an importantsource of income for some communi-ties, but it is increasingly happeningon a large scale and there are 4000aquarium fish collectors in thePhilippines alone.

A network of traders are involvedin the marketing of the fish; the fish-ermen themselves may sell to localtraders who then sell to exportersand so on. There are significantfinancial gains at each step in themarketing chain. For example, anorange and white striped clownfishcosts 10 cents when bought from aFilipino collector, but sells for $25 ormore in an American pet store (Simp-son 2001). Thus, traders in developedcountries typically make the largestgains, while collectors or even thefirst traders make pennies. Becausethe returns to collectors are so low,there is room for conservation alter-natives to be attractive; it does nottake as much return to exceed theearnings of the people supplying thefish.

The impacts of collecting thesefishes from the wild are quite varied.When using traditional netting tech-niques for catching the fish the mostdirect impact is a reduction in the population density of the harvestedspecies, as shown for the Banggai cardinalfish (Pterapoggon kauderni)in Sulawesi, Indonesia (Kolm andBerglund 2003) (Figure A).

While traditional forms of harvestingcan reduce species densities, an issue ofgreater concern is the use of cyanide tocatch the fish. Fishermen typically crushhydrogen cyanide tablets in a squeezebottle and squirt the resulting liquidinto the crevices where the reef fish

hide. The cyanide triggers asphyxiationor muscle spasm in fish, stuns them,and makes them easier to catch. Thesemethods have been used since the 1950sand are believed to have quite a largeimpact on fishes and the coral reefsthemselves. Experiments have tested

ESSAY 8.2

Sinauer Associates, Inc.Groom 3ePrinciples of Conservation BiologyGro0802ACaseStudy.eps100% of size0

14

6

8

10

2Fishing intensity

3

Den

sity

/m

2se

aur

chin

s

Indonesia

Sulawesi

Sulawesi

Luwuk

Peleng

Banggai

Bankulu 10 km

(I)

(II)Figure A (I) Map of the BanggaiArchipelago of Indonesia, showingstudy sites as black circles. The Bang-gai cardinalfish is pictured in thelower left corner. (II) Correlation be-tween Banggai cardinalfish densityand the intensity of fishing at eightseparate sites in the Banggai Archipel-ago near Sulawesi, Indonesia. Becausethe cardinalfish is associated with seaurchins, their density is expressed perm2 of sea urchins on the sea bed. In-tensity of fishing was estimated frominterviews with local people. Banggaicardinalfish are caught by nets only,so reductions in density are solely re-lated to fishing pressure, not to reefdestruction (Modified from Kolm andBerglund 2003.)


the response of 10 species of coral toconcentrations of hydrogen cyanidelower than those typically used by fish-ers. Eight of the coral species diedimmediately, and the other two diedwithin three months. Death occurredbecause the cyanide disrupted the rela-tionship between the coral and its sym-biotic zooxanthellae. Doses as low of 50mg/L cause death of the zooxanthellae,and one spray (approximately 20 cc) cankill the corals over an area of 5 m2

within 3–6 months. These impacts are ofconcern as Southeast Asia has 30% ofthe world’s coral reefs, and these havebeen declining in quality due to a vari-ety of factors, of which cyanide fishingis but one. Now only 4.7% of Philippineand 6.7% of Indonesian reefs are in per-fect condition and quite naturally theseare the reefs targeted by collectors. Thecyanide also affects fishes themselves.Half the poisoned fish die on the reef,and 40% of those caught alive die beforereaching an aquarium (Simpson 2001).

Estimates of the economics of thispractice suggest that the profitability ofcyanide fishing to the fishermen isabout $33,000 per km2 over 25 years(10% discount rate). The direct losses tofisheries total $40,000 per km2 and thetourism losses can range between$3,000 and $436,000 per km2 dependingon location (Cesar 1997). The nonusecosts of losing the coral reef are likelyto be larger than these direct values.

Solutions?This is an interesting situation wherepoor local people are seeking to harvesta local resource to sell it to richer West-erners, but a side effect of this harvestis the loss of another valuable resource.The marketed resource is nonessentialto Westerners, (i.e., we could all livewithout a marine aquarium if we hadto), but we like the aquariums, andmany conservationists may argue thatthrough keeping aquaria, people’sinterests in marine conservation couldbe heightened. So what should we do?

The solution seems to be to permit asustainable harvest of fishes caught in anondestructive manner. Attempts toachieve this can be legislative; forexample the application of internationallaw like CITES (Convention on theInternational Trade in Endangered WildFauna and Flora) to the relevant fishspecies could stop all trade in thesespecies. Alternatively, tighter importcontrols could be implemented in thedestination countries (e.g., in NorthAmerica and Europe) whereby onlyfish harvested to approved standards

would be permitted entry. While somelegislative backing of this nature couldbe beneficial, in practice it will be veryhard to police.

Alternatively, it may be possible toachieve more sustainable harvest byapportioning property rights to differ-ent sets of actors. Currently mostmarine fisheries are open access, andbecause of this are susceptible to the so-called “tragedy of the commons.” Intheory the solution to this is to allocateproperty rights to different groups ofpeople. In the presence of propertyrights, the fishermen should seek tomanage their resource with a long-termperspective, as opposed to the extremeshort-term perspective that open accessencourages (Ostrom and Schlager1996). The allocation of property rightsto fishermen is on-going around theworld; for example, the Chilean gov-ernment is offering the rights to harvesta benthic gastropod Concholepas conc-holepa, known locally as “loco” (Castillaand Fernandez 1998). This involvesallocating certain areas of seabed todefined groups of local people fromlocal communities who then managethem collectively.

While the theory of property rightsis simple, recent work on the Chileanexperience shows that the allocation ofproperty rights is not a panacea andcan lead to considerable social tensionbetween different groups in a commu-nity. One important issue in all suchefforts relates to policing the agree-ments, as it is difficult for hard-pressedofficials to spot breaches of agreementsat sea. In the absence of official polic-ing, disagreements between groupsabout who can harvest where canbecome difficult, if not violent. Thesesorts of issues highlight the need forstrong, functional institutions to helpcoordinate and manage any network ofcommon property resources. While theneed for such institutions is clear fromtheory and practice, their developmentcan be a long and complex processrequiring considerable effort from allinvolved (Ostrom 1990).

A third approach to dealing withthese issues is being tested by theMarine Aquarium Council. They aim toset up a certification scheme wherebyfish caught in traditional nets would belabeled in some way, which wouldenable consumers to buy these fish inpreference to cyanide-caught individu-als. If all consumers preferentially pur-chased certified fishes, then the advan-tage of using cyanide would disappear.To get such a certification scheme to

work several issues have to be followedin parallel:

• There needs to be accurate identifi-cation of non-cyanide caught fishesnear the point of capture. This canbe done through testing samples offishes in export warehouses forcyanide exposure.

• Fisherman need to be informedabout the certification scheme andtrained to use alternative collectingtechniques, such as hand nets.

• Fishes caught without cyanideshould be labeled so that fish buyerscan choose to support practices thatpreserve reefs.

• Consumers should be educatedabout the environmental issues asso-ciated with the use of cyanide forcatching fish.

Many such certification schemes arecurrently under development aroundthe globe, particularly for food and for-est products, and are easily linked toother management schemes such as theallocation of property rights. In mostsituations certified products will costmore than non-certified products. Thisis almost inevitable if, as in the case ofaquarium fishes, the damaging fishingtechniques are used because they areeasier and cheaper than less-damagingtechniques. But if consumers can beeducated about the certificationscheme, then the hope is that theywould chose to buy the more expen-sive, certified products rather than thecheaper alternatives. In this way thecombination of an efficient marketeconomy and an educated consumercan bring real benefits to the environ-ment. One of the most positive exam-ples of this comes from Home Depot®

stores in the U.S., patronized by mil-lions of people each year. Home Depot®

has begun selling and promoting certi-fied forest products, thus vastly increas-ing their availability and visibility, anddue to their enormous market share,reducing the costs.

Unfortunately, research done to dateon food products suggests that ulti-mately price determines consumers’purchasing patterns. Generally onlywealthier and more educated con-sumers are prepared to pay more forcertified products. Thus if economicmeans are to be used to help conserva-tion, there is a need to increase the gen-eral level of education and awarenessin society—a continuing challenge to allconservation biologists. n

exploitation revealed by this diagram is that in open ac-cess operations, we can expect people to join the exploita-tion until their individual costs become equal to theirprofits, at EC. Exploitation will proceed to this risky break-even point because people are competing with each other.

The lesson from this analysis is that open-access ex-ploitation leads to much greater rates of exploitationthan are most profitable or safest for the long-term sur-vival of the population. This is tragic both for the re-source and for consumers, because each individual iscatching less than they could if they had fewer competi-tors. Governments often respond by providing perversesubsidies (Myers 1998; Myers and Kent 2001), leading tolower apparent costs, and hence encouraging furtheroverexploitation (Repetto and Gillis 1988). The capital in-vested in many activities such as commercial fisheriesand logging operations cannot easily be converted toother uses, making it difficult for people to leave thebusiness when times are hard. Naturally, this leads to re-sistance against restrictions on exploitation rates. In fact,exploitation can have a one-way ratchet effect, with gov-ernments providing aid to encourage overexploitationwhen populations are already low, as well as supportinginvestment in the activity when times are good, to in-crease profits. The consequent over-capitalization leadsto a one-way trip toward overexploitation.

Discounting Even when people have exclusive rights to exploit a re-source, they may still be tempted to exploit it heavilynow, rather than conserve it for the future. This is be-cause we place a higher value on current than on futureworth. This is due to economic discounting (see Chapter5). Discounting can have serious implications for effortsto restrain overexploitation of renewable resources(Clark 1976). For example, suppose you could either leta forest grow for 100 years, when the larger trees willfetch twice their current value, or chop it down now.Which should you do? From a purely economic per-spective the most sensible thing might be to cut it nowand invest the money because the interest may accumu-late faster than the growth rate of the standing timber.

This carries a rather sobering message for conserva-tionists: Even with small discounting rates that are wellwithin reasonable bounds for economics, it may be eco-nomically rational to exploit populations to extinctionrather than leaving them to provide future returns. Thismay be especially true for slow-growing species such aswhales and hardwood trees, which take a long time toaccumulate value. This does not imply that conserva-tionists should roll over in the face of economics, butmerely that they will need to consider more than the eco-nomic investment value of renewable resources to for-mulate cogent argues for conservation. As many of thechapters of this book make clear, this is not difficult.

Comparison of Methods for CalculatingSustainable YieldsThere are many ways of putting the theory of sustain-able exploitation into practice, depending on what peo-ple need to know, and how much time and money theycan spend getting the information. For example, the In-ternational Whaling Commission uses some of the mostsophisticated population models in the world to calcu-late the impacts of hunting on whale populations. Theirmodels need to be robust, in order to stand up to thetremendous pressures exerted by those who are “for” or“against” whaling. Yet, to be used effectively, these mod-els often demand a lot of information about the biologyof whales that is not yet available.

Methods for calculating sustainable yields have beenreviewed by Hilborn and Walters (1992), Milner-Gullandand Mace (1998), Quinn and Deriso (1999), Jennings etal. (2001), among others. Here we give only the bareminimum needed to understand the logic of eachmethod and the pros and cons of using them.

Surplus production Surplus production models are also called surplus yieldmodels or simply “production” models. They are mostoften used in fisheries. The simplest ones require very lit-tle data, thanks to an ingenious derivation by Schaefer(1954) who pointed out that if you know how yieldshave responded to different levels of exploitation effortover time, then you could estimate the dome-shapedyield curve shown in Figure 8.9B. Effort might be meas-ured as number of fishing boats out each day of the year,but in non-fisheries contexts it could also be the numberof days spent hunting by all hunters in a given area perseason. The corresponding yields might be the thou-sands of tons of bluefin tuna (Thunnus maccoyii) caughtby the fishery each year (or number of mallards, Anasplatyrhincos, shot per year in an area).

While this is a helpful simplification, it has severalpitfalls that have led to extremely dangerous outcomeswhen this method was first applied. First, it treats eachyear as an independent replicate, which it will not be,due to lags in the ability of the population to respond todifferent levels of mortality. For example, cod in theNorth Sea take about four years to reach maturity. There-fore, no matter how much you reduce density depend-ence by reducing the number of adults in any given year,it will take four years for the survivors to produce off-spring that become old enough to be caught. So if youdouble the fishing effort from one year to the next, youmay catch twice as many fish. But obviously this cannotbe due to the survivors producing twice as many young,because in that second year, those young fish will onlybe one year old, and not big enough to be caught. Whatis really happening is that more fish are caught with a

268 Chapter 8

doubling of effort simply because you are scooping moreadults out of the population. You are not learning muchabout the ability of populations to respond to lower den-sities, you are simply cutting more deeply into your orig-inal stock of fish. The population is decidedly not inequilibrium with the fishery. This dangerous assumptionwas the downfall of the original production models(Hilborn and Walters 1992). The biggest collapse in thehistory of fisheries happened to the Peruvian anchovy(Engraulis ringens) in the early 1970s when the fisherywas operating at a level that a surplus production modelsuggested to be safe (Boerema and Gulland 1973).

A number of much more sophisticated surplus pro-duction models have been developed since the rise andfall in popularity of the original Schaefer version (Po-lacheck et al. 1993; Schnute and Richards 2001). Whilethese require more information than the elegantly simpleearly version, they can get around the dangerous equilib-rium assumption. In fact, in fisheries they have beenshown to sometimes outperform more complex models.

Yield per recruit Yield-per-recruit models were developed originally aspart of the “dynamic pool” concept in the landmark fish-eries book written by Beverton and Holt (1957). The dy-namic pool refers to models that keep track of separateprocesses that add to a population, such as recruitmentand growth, or that subtract from it, such as natural mor-tality and fishing mortality. The basic principle can beapplied to many other forms of exploitation.

Imagine a species that becomes more valuable as itages, due to increasing size. Older fish provide moremeat and older trees provide more wood. If exploitationis weak, most of the fish or trees will be big when theyare taken, which is good. But if we wait too long, naturalmortality will have taken its toll and there won’t be asmany individuals available to us. Yield-per-recruit mod-els search for the level of mortality that maximizes theyield under this tradeoff between numbers and value.Here, a “recruit” is defined as an individual that has be-come big enough to be captured, not in the demograph-ic sense of an individual that has reached maturity. Oncethe level of mortality has been found that maximizes theyield per recruit, we can calculate the total yield that willbe obtained from a given level of mortality, if we knowhow many recruits are coming each year. That is howfishing quotas are set in many countries today (Jenningset al. 2001). Yield in this context is measured in biomass(number of fish caught in each age cohort multiplied bythat cohort’s mean weight).

This method requires information about how the value(often size) of individuals increases with age, as well as anestimate of natural mortality rates, because natural mor-tality determines the rate at which individuals die beforewe have a chance to get them. While growth data are

often available, natural mortality can be difficult to esti-mate because scientists are usually late on the scene, withexploitation well underway by the time they start collect-ing data. This makes it difficult to disentangle naturalfrom human-induced mortality. Yield-per-recruit modelsare the standard approach used in many temperate fish-eries, but developing countries in the Tropics rarely havethe resources to collect the growth and recruitment dataneeded to use this technique as a management tool.

Full demographyFor some species there has been sufficient economicvalue or conservation concern to lead to the productionof full-blown population models. These models combineinformation on vital rates such as births, juvenile sur-vival, age at maturity, and adult survival. These param-eters are often analyzed with matrix models in whichpopulations are divided into discrete classes based ontheir age or stage in the life cycle (reviewed by Caswell2001, and Lande et al. 2003). The most comprehensivemodels incorporate stochastic variation in parameters,and quantify uncertainty in the outputs.

Assessment of population growth rates is a commonpractice to deduce whether populations are declining asa result of direct exploitation, or exploitation of theirhabitats. For example, Lande (1988) used a model tomake a preliminary determination of whether theNorthern Spotted Owl (Strix occidentalis caurina) was indecline in the northwestern United States. This specieswas suffering from overexploitation of old-growthforests, prompting potential listing under the U.S. En-dangered Species Act. A significant population declinewould be indicated by a population growth rate, λ, thatis significantly less than 1.0. Lande’s model was basedon estimates of clutch size, fledgling survival probabili-ty, the probability of successful dispersal, and survival ofsubadults and adults. The estimated population growthrate was 0.961. However, this estimate had a high degreeof uncertainty, due to uncertainty in the underlying pa-rameter estimates. When these uncertainties were scaledup through the model, the estimate of λ had a confi-dence interval of ±0.0562. Therefore, λ could not be dis-tinguished statistically from a value of 1. Further field re-search ensued, yielding better parameter estimates, andworse news for the owls, with significant population de-clines detected in 10 of 11 sites (Forsman et al. 1996).

Other uses of demographic analyses are to diagnosethe reasons for population declines, and also to suggestthe likelihood of success of various management proce-dures (Caswell 2001). For example, many populations ofsea turtles suffer from egg collecting, hunting of adults,and bycatch of juveniles and adults in fisheries. Shouldconservationists concentrate on protecting eggs, juvenilesor adults? Crouse et al. (1987) examined the effects onpopulation growth rates of enhancing survival of each of


270 Chapter 8

these and other life stages in loggerhead turtles (Carettacaretta). They found that protection of eggs was far lessimportant to population demography than protection ofolder life stages. This and subsequent research (e.g.,Crowder et al. 1994, Heppell et al. 1996) contributedstrongly to changes in conservation tactics, most notablythrough the adoption of mandatory turtle excluder de-vices in trawlers in the United States. Spatial analysesalso can improve understanding of what managementinterventions may be most effective. For example, analy-ses of the coincidence of turtle sightings and shrimptrawling activities allows specific recommendations forarea closures that can reduce bycatch mortality by avoid-ing larger concentrations of sea turtles in areas of lowereconomic importance (McDaniel et al. 2000). Finally, con-tinued monitoring coupled with population modeling al-lows evaluation of how well management efforts aredoing, and suggestions for improvements (e.g., increasedcompliance; Lewison et al. 2003) (Figure 8.14).

The simplest structured population models often ignoredensity dependence. That is, the probability of survivinglong enough to move from one age class (or stage) in the lifecycle and eventually reproducing, is kept constant as thepopulation changes in size. As we argued earlier, this as-sumption may be defendable in some situations, but be-cause the concept of sustainability is founded on density-de-pendent compensation, some people would argue that itmakes little sense to ignore density dependence in exploita-tion models. A study of the edible palm, Euterpe edulis,showed the value of incorporating density dependence ex-plicitly (Freckleton et al. 2003). This edible palm is found inthe Brazilian Atlantic forests. Edible “palm hearts” corre-spond to the apical meristem of this species. Extraction ofthe apical meristem kills the plant. Freckleton et al. (2003)modeled a population of edible palms divided into sevenstages according to size of the plant and its reproductive

state. Probabilities of surviving and growing from one stageto the next, as well as annual reproductive output, were cal-culated from field studies, which showed strong density de-pendence in these parameters. For example, seedlings (thefirst stage in the model), compete with each other for lightand resources, and are also shaded out by high densities ofadults (Silva Matos et al. 1999). Analyses showed that whenthe correct amount of density dependence in survival andgrowth from one stage to the next were incorporated, pop-ulations were predicted to be able to sustain considerablyhigher levels of exploitation before they become eradicated.

Adjustments based on recent resultsIf we can monitor either the population itself, or the num-bers of individuals that are taken as a result of variousmanagement measures, then we can adjust the quotaseach year to take account of new information. For exam-ple, in North America hunting licenses for waterfowl andmammals are allocated each year according to the healthof the populations. So, if goose populations increase fromone year to the next, the hunting season may be extend-ed. Box 8.1 provides an example of this method appliedto the setting of quotas for trapping marten in Canada.

Demographic rules of thumbParameter-hungry models are useless for the vast major-ity of the world’s exploited species, because we usuallylack even the most fundamental information about thebiology of the organism and the activities of the peoplewho exploit them (Johannes 1998). The cruel irony is thatit is often the species that we know least about that are inmost trouble from intentional or accidental exploitation.Most of the tropical vertebrates that are hunted for theirmeat fall into this category, as do most whales, sea turtles,sharks, and rays. Various demographic rules of thumbhave been developed that seek to overcome this problem.

200

250

150

100

50

0

1990

1991

1992

1993

1994

1995

1996

1997

1998 19

9920

0020

0120

0220

0320

0420

0620

05

Kem

p’s

rid

ley

stra

ndin

gs

Year

b

Figure 8.14 Projected annual strand-ings for Kemp’s ridley sea turtlesbased on three potential levels ofcompliance by shrimpers in using tur-tle excluder devices in their nets.Black squares = compliance at 2000levels, gray squares = 50% improvedcompliance over 2000 levels, andopen squares = full compliance. Fluc-tuations in strandings reflects varia-tion in compliance and sea turtleabundance in areas where shrimpfisheries are most active. (Modifiedfrom Lewison et al. 2003.)

fpo


BOX 8.1 Adjustments of Quotas for the Marten (Martes americana)According to Previous ResultsJohn Reynolds, University of East Anglia

Commercial trapping of martens inOntario is regulated by trap-linequotas, which are issued by the

Ontario Ministry of Natural Resources.Fryxell et al. (2001) analyzed data froma district in the southern part of theprovince where trappers were com-pelled to report how many animalsthey caught each year. Trappers hadalso been asked to submit carcassesvoluntarily so that biologists coulddetermine the age structure of the ani-mals. Over the period of 1972–1991,53% of the trapped animals werebrought in for age determination. Thenumbers and ages of animals caughtwere the only information available tolocal managers, who used this infor-mation to adjust quotas upward ordownward each year.

How successful were the local man-agers in guessing the right quotasaccording to recent trapping results?To answer that question, Fryxell et al.(2001) used an age-structured modelto calculate population sizes and opti-mal quotas over the time period. Theyfound that the marten population hadvaried three-fold over the 20 years,and that although the local managersdid not have this clear information,they set quotas that matched this vari-ation quite well (Figure A). The propor-tion of young animals in each year wascorrelated with the quotas that themanagers set the following year (Fig-ure B). This proved to be a very sensi-ble strategy, because the proportion ofyoung trapped was closely correlatedwith estimates of the annual rate ofincrease of the population (Figure C).Simulations by Fryxell et al. (2001) sug-gested that the highest average yieldduring this period would have beenobtained if 36% of the animals hadbeen trapped each year. This wasremarkably similar to the averagevalue of 34% that had been set by thelocal managers!

How did the local managers do sowell? Close cooperation between man-agers and the trappers was essential,as were the strong links between theproportion of young animals trapped

each year, the population growth rate,and the population size. This exampleshows that good cooperation com-bined with an ability to predict popu-

lation growth rates can lead to effec-tive conservation management, evenin the absence of detailed information.

Sinauer Associates, Inc.Groom 3ePrinciples of Conservation BiologyGro0801ABox.eps100% of size01.17.05

1250

1000

500

2501970 1975 1980 1985 1990

Num

ber

ofm

arte

ns

Year

AbundanceQuotaHarvest

0 .3

–300

0

150

0 .4 0 .5 0 .6 0 .7

Cha

nge

inqu

ota

num

ber

Pro p ortion of young in p revious year

300

–150

0.8

0.6

0.4

0.2

0.3 0.4 0.5 0.6 0.7

Rat

eof

incr

ease

(r)

Proportion of young animals in harvest

Figure C The proportion ofyoung trapped was a goodpredictor of estimates of annu-al population growth rates, cal-culated as ln(Nt / Ns), where Ntis the abundance at the start ofthe trapping season and Ns isthe number of survivors fromthe previous year’s trappingseason.

Figure A Data from commer-cial trapping of marten, Martesamericana, in a district in south-ern Ontario, Canada. Estimatedpopulation sizes (open circles)were mirrored by quotas (graycircles) and number of martensharvested (black circles).

Figure B Changes made bylocal managers to quotas wererelated to the proportion ofyoung that were trapped inthe previous year.

fpo

fpo

fpo

One of the best-known rules of thumb has come to beknown as the Robinson and Redford model, which wasdeveloped as a quick shortcut to assess whether ex-ploitation of tropical mammals can be defined as sus-tainable (Robinson and Redford 1991). This model sim-ply calculates the total annual sustainable production(population increments through births and immigra-tion), assuming that the maximum potential production(Pmax) occurs at about 60% of carrying capacity (K). Thisfigure accounts for any density-dependent effects onpopulation growth and is intermediate between the(MSY) at 50% of K, as used in classic logistic models ofpopulation growth, and models where MSY is reachedat about 70% of K (McCullough 1982). The model is pos-sibly more realistic for tropical forest vertebrates and isexpressed as:

Pmax = (0.6D × λmax) – 0.6D [8.7]

In this equation, D is an equilibrium population den-sity estimate near K (or in a nonhunted area similar instructure and composition to the hunted area); and λmaxis the maximum finite rate of increase for a given speciesfrom time t to time t + 1 (measured in years), which de-pends primarily on the number of breeding females andthe annual number of offspring produced per female. Inaddition, this method calculates maximum quotas for aspecies by assuming that the proportion of the maxi-mum production that can be taken is 60% for very short-lived species, 40% for short-lived species, and only 20%for long-lived species. Estimates of maximum sustain-able offtakes are then compared with hunting data (ob-tained from household interviews, counts of animal car-casses consumed at forest dwellings, and surveys ofwildlife meat sold in urban markets) to determinewhether the population production exceeds or is lessthan the demand within a given hunting catchment.

These data, however, cannot be easily translated intoactual cull rates because the collateral mortality inducedby hunters may be considerable or even exceed the num-ber of kills that are actually retrieved and consumed. Inthe Brazilian Amazon, for example, the number of wool-ly monkeys (Lagothrix lagotricha) that are lethally wound-ed by shotgun pellets, but subsequently fail to be re-trieved by hunters, can be far greater than those thatactually reach consumer households (Peres 1991). More-over, many studies are likely to overestimate production,and therefore sustainable hunting quotas, because reliablepopulation density estimates are obtained from sites thatare not comparable ecologically to the sites where huntingis taking place. Moreover, density estimates are often ob-tained from high-density sites, where field studies aremost feasible, and then extrapolated to sites with muchlower densities (Peres 2000b). Finally, λmax can be highly

variable, is rarely known for any given target species atthe sites being evaluated, and can easily be overestimatedif the only available data come from populations in cap-tivity or long-term studies in high-quality habitats. Nev-ertheless, this model provides a preliminary measurementof hunting sustainability in poorly studied systems andhas the advantage of simplicity and side-stepping moredata-hungry approaches that could not be applied to mostgame hunting scenarios in tropical forests. There havebeen a number of applications (e.g., Fitzgibbon et al. 1995,Muchaal and Ngandjui 1999) and criticisms and exten-sions of the Robinson and Redford model (e.g., Slade et al.1998, Peres 2000b, Milner-Gulland and Akcakaya 2001,Salas and Kim 2002), but it continues to be popular intropical hunting studies.

Spatial and temporal comparisons Most of the time, we have to settle for inferences aboutsustainability from simple evidence such as comparisonsof densities of target species in locations that vary in theintensity of hunting, or interviews with hunters aboutchanges in the abundance of their prey. Often these typesof evidence can be combined to yield powerful conclu-sions about whether extractive activities are sustainable,but without being able to make precise quantifications ofbenchmarks such as maximum sustainable yields. Forexample, a study of hunting of Madagascar radiated tor-toises, Geochelone radiata, used a combination of inter-views with hunters, a comparison of changes in thespecies’ range size over time, and comparisons of thetortoise’s abundance in sites that differed in huntingpressure (O’Brien et al. 2003). The tortoises are collectedfor food and the pet trade. This study estimated that45,000 tortoises are collected each year from southernMadagascar in the catchment of two cities. Hunters aretraveling increasingly far to obtain their animals, anddensities were highest in the most remote, unexploitedsites. This activity is taking place in spite of the speciesbeing protected by law. This information, combined withdocumentation of range contraction, led the authors toconclude that the species is headed for extinction unlesscorrective measures are taken. Hunting rates could be re-duced through education and awareness programsaimed at reducing demand, alternative income-generat-ing schemes, and enhanced legal enforcement.

Sustainable Use Meets Biodiversity This chapter has repeatedly run into contrasts betweentheory and practice, goals and implementation. Acompar-ison between “north” and “south” illustrates the diversityof issues surrounding exploitation. In most temperatecountries hunting, fishing, and forestry management pro-

272 Chapter 8

tocols have been developed through a long history of trialand error based on biological principles. In most tropicalcountries, exploitation regulations are typically nonexist-ent or unenforceable. The concepts of game wardens, fish-eries officers, bag limits, no-take areas, and hunting licens-es are completely unfamiliar to the vast majority of tropicalsubsistence hunters and fishers who often rely heavily onwild animals as a critical protein component of their diet.In many tropical countries, wildlife is an “invisible” com-modity and local offtakes are often not restrained until thestock is depleted. This is reflected in the contrast betweencarefully regulated and unregulated systems where largenumbers of hunters may operate. For example, Minnesotahunters sustainably kill over 700,000 wild white-taileddeer every year, whereas Costa Rica can hardly sustain anannual hunt of a few thousand individuals without push-ing the same cervid species, albeit in a very different foodenvironment, to local extinction.

But even well-meaning management prescriptions canbe completely misguided, bringing once highly abun-dant target species to the brink of extinction. The 97% de-cline of saiga antelopes (from >1 million to <30,000) in thesteppes of Russia and Kazakhstan over a 10-year periodhas been partly attributed to conservationists activelypromoting exports of saiga horn to the Chinese tradition-al medicine market as a substitute for the horn of endan-gered rhinos (Milner-Gulland et al. 2001). Saiga antelopeswere finally placed on the Red List of Threatened Speciesin October 2002 following this population crash.

Even within cultures, there may be a diversity of per-ceptions about the objectives of management. For exam-ple, in countries that can afford to engage in fisheriesmanagement, the standard goal has been to achieve max-imum sustainable yields of the target species. In the past15 years, however, collapses of fish stocks and concernsabout by-catches and damage to marine habitats have ledto new objectives, such as obtaining sustainable yieldswhile minimizing impacts on ecosystems (Reynolds et al.2002). Not everyone agrees with these objectives, espe-cially those who perceive that their traditional livelihoodsare at risk from reduced quotas. Furthermore, there isalso debate within the scientific community about whatthese objectives mean in practice, and how to implementthem. In some sectors, such as forestry, managers areused to dealing with highly polarized debates betweenthose who seek to preserve forests and those who seek tolog them. Today, fisheries managers are facing similarscrutiny from many conservationists, who are askingnew questions about impacts on habitats and ecosys-tems. Indeed, the past decade has seen the first listing bythe IUCN of commercially exploited fishes as threatenedwith extinction (e.g., Atlantic cod), and the first cases ofexploited marine fishes and mollusks being listed under

the U.S. Endangered Species Act (e.g., smalltooth saw-fish, Pristis pectinata, and white abalone).

In many instances failure to protect species from over-exploitation has more to do with institutional short-com-ings than lack of scientific knowledge. European andNorth American fisheries scientists know a tremendousamount about the demography of the major commer-cially exploited fish species, yet this information has notprevented some species from reaching historically lowpopulation levels in the past decade. Of course it wouldhelp to know even more about the animals that we aretrying to manage, such as impacts of low densities on re-productive behavior, trophic interactions in relation tohabitat alterations, and links between environmentalvariability and recruitment. But translating this informa-tion into management action for the long-term benefit offishers and fishes would still run into the age-old prob-lem of lack of political strength in the face of criticismfrom those affected negatively by restrictions. In manydeveloping countries, such problems are compoundedby the dire lack of alternatives for food and employment.When combined with a lack of institutional structures toimplement any policies that may be acceptable, it is easyto understand why exploitation often runs a course thatis slowed only by accessibility of the resource.

As with most of the other conservation issues coveredin this book, we feel that while the future holds someenormous challenges, it is by no means entirely bleak.For example, the adoption of precautionary principles inexploitation by many countries, combined with greaterconcern for impacts on ecosystems, is certainly a movein the right direction. We are also encouraged by thelarge number of imaginative programs in many coun-tries that are aimed at giving local people incentive tomake sustainable use of plant and animal populationsand their habitats; Elizabeth Bennett et al. discuss someof these programs in Case Study 8.2 and Michelle Pinardet al. discuss them in Case Study 8.3 (see examples inChapter 16, also see Case Studies 6.1 and 6.2). Indeed,there is a growing appreciation for the enormous eco-nomic and social value of ecosystem services from wildnature that far exceed the economic benefits of conver-sion to agriculture or other uses (Costanza et al. 1997;Balmford et al. 2003). For example, a forest can providefar more than just timber, including flood defense, car-bon storage and sequestration, and pollination of adja-cent crops. Many researchers are now working to findnovel ways of rewarding local people for exploiting re-sources sustainably in return for benefits received byothers. These are exciting times for integrating sciencewith the policy of exploitation, and it is imperative thatthese efforts are successful if we are to meet the chal-lenge of sustaining wild populations and their habitats.


274 Chapter 8

CASE STUDY 8.1Overexploitation of Highly Vulnerable SpeciesRational Management and Restoration of Sharks

Julia Baum, Dalhousie University

“Sharks … may be considered as one of the richest ‘strikes’ in theonly inexhaustible mine—the sea.”

Barrett, 1928, Scientific Monthly

Anthropogenic impacts on natural ecosystems have almost al-ways begun with the overexploitation of large vertebrates.Consequent species collapses or extinctions had markedlytransformed terrestrial, freshwater, and coastal ecosystems bythe early twentieth century, while oceanic species were stilllargely protected by their distant location and large geograph-ic ranges. Oceans, it seemed, were too vast and their creaturestoo plentiful and widespread to be unduly impacted by hu-mans. But in the past few decades, humans have come to dom-inate these ecosystems as well. Today, industrialized fishingfleets are ubiquitous throughout the world’s oceans, reachingeven the remotest locations and leaving few refuges for marinespecies. As a result, overexploitation now threatens the futureof many large oceanic vertebrates, including whales, tunas,billfishes, and sea turtles. And sharks, rather than being one ofthe richest “strikes” in the ocean are one of the most vulnera-ble (Figure A).

Sharks are among Earth’s oldest inhabitants, having evolvedover 400 million years ago. Along with skates and rays, sharkscomprise the subclass Elasmobranchii within the class Chon-drichthyes, distinguished from bony fishes by their cartilagi-nous skeletons. The world’s 350 shark species are a diversegroup, ranging in size from dwarf sharks that reach less than 20

cm to the 12 m long whale shark, occupying habitats fromfreshwater lakes to entire ocean basins, and occurring in arcticand tropical waters from the surface to depths of a thousandmeters. Our perception of sharks as ferocious predators beliesthe fragility of their populations. Sharks tend to grow slowly,mature at a late age, and give birth to few pups following along gestation period—characteristics more reminiscent of ma-rine mammals than fishes. These traits result in low intrinsicpopulation growth rates, leaving sharks unable to withstandheavy fishing pressure, and with little compensatory ability torecover from over fishing.

Localized shark fisheries that developed in the few decadesfollowing Barrett’s description were characterized by “boomand bust” patterns. Fisheries for the soupfin shark in Califor-nia and Australia, dogfish in Scotland, and porbeagle shark inthe Northwest Atlantic all collapsed in under a decade.

Despite this history of population collapses and sharks’known vulnerability, their exploitation has intensified globally.Shark fisheries underwent a resurgence in the 1980s, driven bothby the decline of traditional food fish species and by increaseddemand for shark products. Shark meat is increasingly con-sumed in Western societies, and the high demand for shark-finsoup in Asia has made shark fins one of the most highly valuedmarine products (Rose 1996). The practice of finning—removingthe fins and returning the remaining carcass to the water—hasnow been made illegal in the Atlantic, but a worldwide ban isneeded. Besides these directed fisheries, many sharks are caughtincidentally, in fisheries targeting other species. In particular,pelagic longline fisheries, which target swordfish and tunasthroughout temperate and tropical oceans, are the world’slargest source of shark by-catch (Bonfil 1994).

Quantifying the impact of current exploitation levels onshark species has proven difficult. Sharks have typically been alow research and management priority because of their histori-cally low economic value and because their catches are often in-cidental. As a result, basic population parameters such as age atmaturity and longevity are unknown for many species, andshark catches have been poorly monitored. Fisheries observersusually monitor only a fraction of the overall fishing effort—per-haps 5% in domestic waters, and even less in international wa-ters—and until recently most fisheries recorded shark by-catchin a generic “shark” category rather than identifying individualspecies. Thus, many shark assessments are done on aggregatedspecies groups, obfuscating any variation in species’ responses

Figure A Scalloped hammerhead sharks (Sphyrna lewini) swim-ming at Cocos Island, Costa Rica. (Photograph © Philip Collawww.oceanlight.com.)

fpo-hi-res-replace?


to exploitation. Compounding these problems are the wideranges of many sharks, which commonly cross internationalboundaries, and even entire oceans, imposing serious con-straints on our ability to monitor their populations.

New research provides strong evidence that coastal andoceanic shark populations have undergone rapid, large de-clines in the Northwest Atlantic (Baum et al. 2003). To demon-strate this, researchers analyzed the largest dataset samplingthe Northwest Atlantic pelagic ecosystem, that of the U.S.pelagic longline fishery. This type of gear consists of mainlinestens of miles long suspended horizontally in pelagic waters(upper water column) by floatlines and buoys, with baitedhooks attached on branchlines at set intervals. Thus it resem-bles a transect through the pelagic ecosystem. The fishery,which operates from Newfoundland into the Gulf of Mexicoand south to Brazil, and includes both coastal and distant off-shore waters, covers the entire range of Northwest Atlanticcoastal shark populations and a substantial proportion ofoceanic shark populations. Because fishers were required torecord shark catches from each longline set and because theirfishing effort was intense, researchers could estimate changesin abundance even for rare shark species: Between 1986, whenfishers first started recording some sharks to species, and 2000,the dataset comprises over 200,000 longline sets and 117 mil-lion hooks! Researchers used statistical models to standardizecatch rates for variations in the gear, location, and timing ofsets. Estimated mean catch rates can then be compared acrossyears as an index of a species’ relative abundance.

All shark populations examined in the study are estimatedto have declined by 40%–89% since the mid-1980s (Baum et al.2003). Researchers estimate that since 1986 hammerhead sharkshave declined by 89%, great white sharks by 79%, threshersharks by 80%, and tiger sharks by 65% (Figure B). The remain-ing coastal sharks (CST), examined together because of the po-tential for misidentification in the logbooks of these similarlooking congenerics, declined by 61% just since 1992. Blue andmako sharks are estimated to have declined by 60% and 40%respectively since 1986; oceanic whitetip sharks by 70% since1992 (see Figure B). These latter species range across the entireAtlantic. Thus, while the data do not cover their entire popula-tions, because other longline fisheries exert intense fishing pres-sure across the Atlantic, it is quite plausible that the patternfound in the Northwest Atlantic is representative of the entireregion. Despite the enormity of documented declines, they arelikely underestimates because they do not account for changesthat occurred during the first three decades of offshore sharkexploitation in the Northwest Atlantic (i.e., from 1957 to 1985).

Halting shark population declines is necessary if we are toavoid species extinctions; reversing them is essential if we areto restore populations to their former levels. Restoration re-quires that we understand the composition and abundance ofnatural shark assemblages, that is, that we have some knowl-edge of what it is we want to restore. Jackson (2001) notes thatbecause most ecological studies began years, and sometimes

centuries, after anthropogenic influences began, the formernatural abundance of many large vertebrates in coastal ecosys-tems was fantastically greater than today. Assessments basedon recent data alone obscure the original abundance of species,such that we may become complacent about species that arenow rare. This lack of historical perspective on what is naturalis coined the “shifting baseline” (Pauly 1995).

Estimating the baseline abundance of pelagic sharks in theNorthwest Atlantic is possible because of the short history ofdisturbance in the open ocean compared to most other ecosys-tems: We need only look back to the early development of in-dustrial offshore fisheries in the late 1950s. In the Gulf of Mexi-co, research surveys were conducted in the mid-1950s usingpelagic longlines, to determine the potential for a commercialtuna fishery (Wathne 1959). These surveys provided some ofthe first biological information available on pelagic sharks, andon the region’s offshore resources. Sharks were so abundant onthese surveys that they damaged a high proportion of tuna onthe longlines and were considered a serious problem. Re-searchers in the 1950s documented that between 2 and 25oceanic whitetip sharks were usually seen around their vesselduring gear retrieval (Wathne 1959). Oceanic whitetip and silkysharks were the second and fourth most commonly caught fish-es overall on the pelagic longline sets, comprising almost 15%of the total catch, and 85% of all shark catches. Matching thesedata to comparable information from the current pelagic long-line fishery reveals a very different picture. By the late 1990sthese two species accounted for less than 0.5% of the total catch,and only 16% of shark catches, suggesting that substantial

Sinauer Associates, Inc.Groom 3ePrinciples of Conservation BiologyGro08EssayB.eps100% of size01.17.05

0

–60

–40

–20

–8 0

–100

HHD WSH TIG CST OCS

Est

imat

edch

ange

inab

und

ance

Sp eciesB SH MAK THR

Figure B Total estimated change in relative abundance (± 95% con-fidence interval) for scalloped, great and smooth hammerhead(HHD), great white (WSH), tiger (TIG), blue (BSH), shortfin andlongfin mako (MAK), common and bigeye thresher (THR) sharks be-tween 1986 and 1999, and for bignose, blacktip, dusky, night, silky,and spinner (CST) and oceanic whitetip (OCS) sharks between 1992and 1999.

fpo

changes have occurred in the Gulf of Mexico’s shark assem-blage in less than half a century. Recent research papers onsharks have either not mentioned the oceanic whitetip shark orhave dismissed it as a rare exception in the Gulf of Mexico, withno recognition of its former prevalence in the ecosystem. In con-trast, a new analysis that incorporates these historic data esti-mates that oceanic whitetip sharks have declined by over 99%in 50 years, suggesting that this species is ecologically extinct inthe Gulf of Mexico (Baum and Myers 2004).

Beyond the risk of species extirpations in the Northwest At-lantic, estimated shark declines may indicate broad ecosystemchanges. Consumers such as sharks usually exert importantcontrols on food web structure, diversity, and ecosystem func-tioning. The generality of the loss of pelagic consumers in theNorthwest Atlantic is particularly worrisome: Most pelagicshark populations appear to be remnants of their natural abun-dance, and many other apex predators, including the sword-fish and marlin, are drastically reduced. Although the ecosys-tem impacts of overexploitation in the open ocean remainlargely unexplored, any changes are likely to be massive inscale given the vastness of these ecosystems, and are likely tobe difficult to reverse given the life history of sharks.

The estimated loss of pelagic sharks in the Northwest Atlanticmay also reflect a common global pattern of decline in sharkspecies. Longlines and other pelagic fisheries are pervasive in alloceans, catching many of the same shark species as in the North-west Atlantic, turning the risk of extirpation in that region intothe risk of global extinction. Apart from the species discussedhere, most other sharks face similar exploitation pressures and

monitoring constraints. Around the world, most sharks arecaught in multispecies fisheries, be it pelagic longlines, bottomtrawls targeting other fishes, or directed shark fisheries that catchseveral species. This is problematic because the pursuit of pro-ductive target species will continue to drive the fishery, even afterless-productive shark populations have collapsed. Finally, be-cause of the difficulty in adequately sampling the large geo-graphic ranges of shark populations, shark collapses and localextinctions may occur unnoticed. Local extinctions of two otherelasmobranch species, the common skate (Raja batis) and thebarndoor skate (Dipturus laevis), for example, were only docu-mented after the fact (Brander 1981; Casey and Myers 1998).

Restoration of shark populations will require fundamentalchanges to their management. If we are to avoid extinctions ofshark populations, fisheries need to be managed for the viabil-ity of the most vulnerable, rather than the most productive,species. This means that substantial reductions in fishing pres-sure in most fisheries that catch sharks are needed. Recoverytargets that incorporate a historical perspective are needed,and management must work toward these. Effective interna-tional management plans need to be implemented and en-forced to account for sharks’ wide geographic ranges. Finally,as individuals we need to make responsible choices about thefish we consume by becoming informed about their status andthe effect of fisheries on incidentally caught species.

Sharks have existed for over 400 million years, but havebeen threatened with extinction in less than 50. We may nowbe a critical juncture—where the decisions made today will de-cide the ultimate fate of these species.

276 Chapter 8

CASE STUDY 8.2The Bushmeat CrisisApproaches for Conservation

Elizabeth L. Bennett, Wildlife Conservation Society, Richard G. Ruggiero, U.S. Fish and Wildlife Service, Heather E. Eves, Natalie D. Bailey, and Andrew Tobiason, Bushmeat Crisis Task Force

In Africa, forest is often referred to as “the bush,” thus wildlifeand the meat derived from it is referred to as “bushmeat.”Bushmeat is eaten for subsistence, and is also traded in localand regional markets to supply protein to people in cities andtowns.The bushmeat trade in Africa is a multi-billion dollar ayear industry involving millions of animals from cane rats toelephants (Wilkie and Carpenter 1999; Fa et al. 2002), and mil-lions of hunters, traders, market sellers, and consumers (FigureA, Mendelson et al. 2003). Scientific studies show that the ma-jority of this unregulated, commercial hunting for wild meat isunsustainable and threatens biodiversity conservation goals

(Robinson and Bennett 2000). Recent estimates of hunting ratesin the Congo Basin show that a majority of species are huntedunsustainably (Fa et al. 2002). The expansion of bushmeathunting poses a risk not only to the viability of wildlife species,but also to the livelihoods of the people who share the landwith them (Figure B; Milner-Gulland et al. 2003), as well as tohuman, wildlife, and livestock health (Hardin and Auzel 2001;Wolfe et al. 2004).

Recommendations for promoting and regulating the bush-meat trade as a means to support rural livelihoods (Brown2003) are generally not supported by current management ca-


pacity, policy, ecological, or economic realities (Fa et al. 2003;Wilkie et al. in press). While some suggest that a sustainabletrade in wildlife for bushmeat can exist (Cowlishaw et al.2005), these claims are generally based on species with high re-productive potential, such as rodents, in degraded habitats(Robinson and Bennett 2004). Such scenarios generally emergeafter all the large-bodied mammals have already been huntedout (Bennett et al. 2002). Long-distance commercial trade inwildlife from tropical forests is almost inevitably unsustain-able, and under the very limited circumstances in which itmight be biologically feasible, strong management capacitymust be in place. Such capacity is lacking across central Africa.

Robinson and Bennett (2000) document numerous root caus-es of the bushmeat crisis that must be addressed in efforts to con-serve wildlife. Expanding human populations increase demandfor bushmeat, while poverty and food insecurity create increasedreliance on natural resources. Private companies in extractive in-dustries usually do not engage in wildlife management plan-

ning, thereby facilitating increased access and unsustainable ex-ploitation of wildlife resources. Governments and other landmanagers often have low capacity for monitoring and enforce-ment, or have poorly designed or implemented wildlife policies.Rural and urban human populations frequently have inaccurateperceptions of the limitlessness of wildlife and a general lack ofawareness of wildlife use impacts. The poor, in particular, lackprotein and income alternatives to bushmeat. These drivingcauses, compounded with the unsustainable commercial wildlifehunting levels impacting populations all across Africa have re-sulted in a crisis which threatens massive population declines;particularly vulnerable are the forest species which are an orderof magnitude less productive than grassland species.

Solutions to the bushmeat crisis must simultaneously ad-dress these causes while supporting both undisturbed coreareas and some regulated-use areas. Such plans may assurethat wildlife and biodiversity are conserved for future genera-tions and that the livelihoods of marginalized rural peoples aresupported (Milner-Gulland et al. 2003). Protected areas are anessential component to any landscape management plan. Bio-diversity conservation goals and alternatives for income mustbe coupled with appropriate wildlife regulations and enforce-ment to limit hunting to local areas and to nonendangeredspecies with high reproductive rates (Eves and Ruggiero 2000).

Solutions include policy development, capacity building foreducation and enforcement personnel, development of bothprotein and income-generating alternatives, incorporatingwildlife management planning into economic developmentand extractive industry activities (especially logging, mining,and road-building), and collaborative information sharing andmanagement for development of best practices, adaptive man-agement, and long-term monitoring. Such solutions necessitatethe commitment and collaboration of government, non-

governmental organizations (NGOs), the pri-vate sector, and communities to support essen-tial wildlife management planning activitiesthat include governance, enforcement, and com-munity participation to support biodiversityprotection (Ruggiero 1998).

Progress in many of these areas is now beingmade. While the bushmeat issue was not well-known or understood before the late 1990s, ithas since become a central issue for many con-servation projects and international efforts(Bushmeat Crisis Task Force 2004). The Bush-meat Crisis Task Force (BCTF) was created in1999 as the first coordinated effort by NorthAmerican wildlife conservation groups and pro-fessionals from a range of disciplines to focus ef-forts on the growing unsustainable, illegal, andcommercial trade in bushmeat in Africa andaround the world. BCTF has prioritized fourareas of engagement for this collaborative action:information management; engaging with key

Bushmeat

Wholesaler Farmerhunter

Commercialhunter

Customer

Chopbars Market traders

Figure A The cycle of the bushmeat trade.

Figure B Duikers on the back of a logging truck in Gabon. (Photograph by R. Ruggiero.)

fpo

fpo

decision-makers; formal education/training program develop-ment; and public awareness. Professionals from conservation,government, industry and development fields work with theBCTF staff and steering committee to identify and analyzeavailable information, develop consensus on solutions, produceresources, and identify areas of collaboration potential. MajorBCTF initiatives include working with key decision makers(KDM) in Africa and the U.S. toward policy development andmobilization of resources; the Bushmeat Education ResourceGuide (BERG), which supports public education by zoologicalparks and other institutions; the Bushmeat Promise, a cam-paign designed to engage the public toward action; and theBushmeat Information Management and Analysis Project(IMAP), which provides geo-referenced BCTF project and pub-lication databases online, assembles bushmeat-relevant datasetsfrom around the world, and gathers new data on relevant in-formation in the Congo Basin. For more information about thebushmeat crisis and what is being done to address it, visit theBushmeat Crisis Task Force website: www.bushmeat.org.

Since the establishment of BCTF and other bushmeat efforts,public awareness in the U.S. and Europe has been raised con-siderably. Only a handful of media articles were produced priorto 2000, but more than 750 were produced in the five years thatfollowed (Bushmeat Crisis Task Force 2004). A number ofawareness programs in urban and rural Africa communicatethe impacts of the bushmeat trade on wildlife and human com-munities that enable consumers to make better-informed deci-sions (Obadia et al. 2002). Formal education and training toolsare also developed in Africa and the U.S. to further educate keydecision makers, field personnel, and the public (Bushmeat Cri-sis Task Force 2004; Bailey and Groff 2003).

At the international policy level, the World ConservationUnion (IUCN) adopted an official bushmeat resolution in 2000,the Convention on International Trade in Endangered Speciesof Wild Fauna and Flora (CITES) created a bushmeat workinggroup in 2000, the Africa Law Enforcement and Governance(AFLEG) treaty signed in 2003 by both timber producer andconsumer nations identified bushmeat as a priority concern,and the Convention on Biological Diversity (CBD) establisheda working group to review the issues related to nontimber for-est products, including bushmeat.

Further evidence of an increased focus on addressing the un-sustainable bushmeat trade is the Congo Basin Forest Partner-ship (CBFP). Officially established at the World Summit on Sus-tainable Development in September 2002, CBFP is an associationof 29 governments and international NGOs collaborating to pro-mote biodiversity conservation and improved livelihoods. CBFPpartner organizations and their projects employ aspects of thebushmeat trade as indicators of the success of these landscape-level management efforts. Three leading success indicators aredirectly related to bushmeat: wildlife management planningand enforcement, protein and income-generating alternatives,and training and awareness to enable behavior change.

While numerous projects focus on one or two focal aspectsof the bushmeat trade, there are a few that stand out as exam-

ples of the multi-stakeholder, multi-level effort that is essentialto address the unsustainable bushmeat trade effectively, amidsta landscape of human development activities.

Wildlife Management in Logging ConcessionsOne example of such collaboration is the Project for EcosystemManagement of the Peripheral Zone of the Nouabalé-NdokiNational Park (PROGEPP), a wildlife management programbased in the Kabo and Pokola logging concessions in thenorthern Republic of Congo. This project, initiated in 1999, is ajoint effort of the Government of the Republic of Congo, theCongolaise Industrielle des Bois (CIB) logging company, theWildlife Conservation Society (WCS), and the communities as-sociated with the concessions. Aside from the government, theCIB logging company is the single largest employer in Congoand their logging concessions have acted as a magnet for im-migration. The population of Pokola, the largest town near theconcession, has more than doubled since 1999,from 7,200 to16,000. Many would hunt bushmeat if not controlled and ifother food options were unavailable. Instead, as a result of theproject, wildlife populations are thriving, including gorillas,chimpanzees, elephants, and bongo, and people are maintain-ing a healthy mixed, protein-rich diet. The key to this successis conducting a complex of activities that include wildlife man-agement planning, training, monitoring, education, and en-forcement while simultaneously developing alternativesources of income and protein-rich foods.

Elkan and Elkan (2002) provide an overview of PROGEPPincluding three key rules adopted with regard to hunting inthe concession—no snare hunting, no hunting of legally pro-tected species, and no export of bushmeat outside the immedi-ate community—a model established earlier with local com-munities associated with the adjacent Nouabalé-NdokiNational Park. The rules are made known to all workers andtheir families, especially truck drivers who often serve as in-termediaries for illegal trade, as well as communities not affil-iated with CIB, and are enforced by “eco-guards” at numerouscheckpoints. In addition, conservation and land-use zoningwas created according to traditional community zones andnatural resource use. No-hunting areas, community huntingzones, and buffer zones around the Nouabalé-Ndoki NationalPark were adopted and established in the logging concessions.

CIB, like any large company working in an area of limitedgovernment capacity is in a position to dramatically influencehow land and resources are being managed beyond their man-dated exploitation, and fulfills many public services for its em-ployees and nearby communities. As a complement to the so-cial, health, and education programs CIB already provides, thewildlife management plan includes company provision of af-fordable livestock and protein-rich food alternatives.

Since the project’s inception, snare rates are significantly re-duced, endangered species are found even in areas where reg-ulated hunting is taking place, and protein alternatives arebeing incorporated into the culture of the logging concession(Elkan and Elkan 2002). In addition, national policies regard-

278 Chapter 8


ing the provision and training of eco-guards within all loggingconcessions in Congo have been adopted.

Conservation is Good BusinessThe information emerging from PROGEPP identifies wildlifemanagement planning and enforcement, partnership develop-ment, and provision of protein and income-generating alterna-tives as keys to bushmeat management success. Similarly, a proj-ect in Zambia involving an agricultural cooperative is showingsigns of conservation success based on these same priority ac-tions (Lewis 2004). The basic premise for this conservation pro-gram is that “food secure, farm-based communities with alter-native sources of income to illegal use of wildlife can contributepositively to wildlife production” (Lewis 2004).. The Communi-ty Markets for Conservation and Rural Livelihoods (COMACO)efforts developed by WCS in collaboration with the ZambiaWildlife Authority (ZAWA) and the Ministry of Tourism, Envi-ronment and Natural Resources (MTENR) are designed to di-rectly address the causes that drive overexploitation of wildlifefor bushmeat in this region—food insecurity, need for income,poor land-use practices, and access to markets.

The COMACO project is innovative among conservationprograms as it actively engages individuals and communitiesin a business cooperative and has been very effective in coor-dinating the support of the private sector. Producer depotssupply a central trading post and households and communi-ties can belong to the cooperative as long as they make a com-mitment to improving land-use practices and supporting con-servation. That is, families must agree to a specific land-usemanagement plan and producer-group conservation by-lawsbefore being accepted in the program.

The COMACO project in 2003–2004 had over 750 conserva-tion farming groups supplying depots with rice, chickens,groundnuts, and honey. The cooperative negotiates higherprices for these products, and in turn farming groups turn inhunting materials; over 30,000 snares and nearly 500 guns were

collected during one three-year period. This model is most ap-plicable to subsistence and cash-crop farmers who have enoughfood but not enough income, live at a medium density, and pro-duce conservation-friendly products (Lewis 2004).

Critical to the entire initiative is the linked focus on wildlifeprotection. Experience across the entire tropical globe showsthat providing protein and income alternatives alone do not re-duce hunting. COMACO overcomes this by linking the coop-erative with increased protection of wildlife; all cooperativemembers are also involved in wildlife enforcement, and theirdoing so is a prerequisite of their eligibility for the program.Hence, they are both benefiting from improved livelihoodswhile also protecting their local wildlife.

While these examples are promising it is important to notethat such systems are site specific, require trained and com-mitted experts in wildlife, community development, enforce-ment, and education and demand long-term commitment ofpersonnel, financial resources, monitoring, and evaluation. Ap-plication of similar programs without such resources and ca-pacity in place are unlikely to succeed. Caution should be usedin responding to plans that support broad-scale policies of reg-ulated trade in bushmeat as they are unlikely to lead to achiev-ing either biodiversity conservation or development goals inthe long, or even in the short-term. Quite the opposite, they aremore likely to result in further impoverishment of rural com-munities and degradation of the natural resources upon whichthey depend. What is required is a global commitment and re-sponse that simultaneously and sufficiently supports nationsthat have identified the bushmeat problem as a threat to theirnatural and cultural heritage so that locally-designed effortssuch as those described have the opportunity to succeed. Thiswill include sufficient resources to enable training, enforce-ment, development of protein and economic alternatives, land-scape-level management that includes protected area as wellas multiple-use zones, and awareness programs for the public.

CASE STUDY 8.3Managing Natural Tropical Forests for TimberExperiences, Challenges, and Opportunities

Michelle A. Pinard, University of Aberdeen, Manuel R. Guariguata, Environmental Affairs, Convention on Biological Diversity, Francis E. Putz, University of Florida, and Diego Pérez-Salicrup, Universidad NacionalAutónoma de México

Applied forest ecologists, or silviculturalists, find managing theimmense diversity of natural tropical forests for various goodsand services a challenge, not only because of the complexity ofthese forests but because the requirements for sustainability arealso diverse and complex. While the diversity may often seem

overwhelming, natural forests in the moist tropics have manyof the attributes that one might consider conducive to manage-ment. For example, they are productive in terms of the diversi-ty of goods and services, and are known to be highly resilient;many areas that currently support old-growth tropical forest

280 Chapter 8

are the result of secondary forest suc-cession after agricultural fields wereabandoned hundreds of years ago (e.g.,Darien, Panama; Bush and Colinvaux1994). Despite the opportunities repre-sented by these attributes, examples ofgood management are few. There ishuge variation in forestry practicesacross the tropics, with some countriesand forest owners moving towardmore sustainable practices, while oth-ers continue to convert or degrade for-est, often under the pretense of timbermanagement.

Our aim is to explore some of thevariation among forest managementpractices and to identify what seems tobe working to improve management.We also discuss what we see as themain challenges and opportunities forsustainable management of naturaltropical forests. Although we focus onthe ecological, economic, and socialbases of sustainable forest manage-ment for timber, our motivation isprincipally forest conservation. It mayseem strange to be trying to save tropical forests with chain-saws and bulldozers, but the majority of the tropical foreststhat remain are only likely to be saved if they can be managedprofitably for timber. In many cases, if people cannot makemoney from forests, then the forests will likely be destroyed infavor of some more lucrative land use, such as agriculture.

Background and Context for Tropical Forest ManagementBecause of the high diversity of tropical trees and the con-comitant low population densities of marketable species, log-ging in natural forests in the tropics is typically selective; a rel-atively small number of trees above a threshold size (theminimum felling diameter) are felled and extracted. The logsare usually skidded out of the forest using bulldozers, rubber-tired skidders, farm tractors, or draft animals (Figure A).Where machines and draft animals are not available, large logsare often sawn into thick boards that are manually carried outof the forest.

Logging impacts on the residual forest vary with harvest in-tensity, level of pre-harvest planning and control, soil type andterrain, and forest stature and structure, among other factors(Figure B). Due to the selective nature of harvesting, forests arerepeatedly logged after intervals of a few years to a few decades.Where logging is scheduled so as to achieve a sustained yield oftimber, this approach to forest management is referred to as apolycyclic silvicultural system. Where properly designed, sev-eral harvesting episodes (or cuts) are scheduled within a rota-

tion (nominally, the time it takes for atree to grow to maturity). In theory, cut-ting cycle lengths and harvesting inten-sities are determined on the basis of therate of commercial timber productivity.Harvesting also can be done morestrategically, as a silvicultural treatmentto promote the regeneration of ecologi-cally similar groups of trees. For exam-ple, light-demanding tree species are fa-vored by intensive stand interventions,whereas if less timber is harvested perunit area, more shade-tolerant speciesbenefit. Unfortunately, the temptationof marketable standing timber oftenoverwhelms efforts to secure sustainedyields. As a result, timber-mining oper-ations favor both slow-growing, shade-tolerant species, and fast-growing pio-neers, not the species with marketabletimber. Moreover, given the generallack of reliable data on forest productiv-ity, managers establish cutting cycles bymaking educated guesses at how fasttrees grow.

The deleterious environmental im-pacts of logging can be minimized if well trained and closelysupervised crews fell and extract timber following a detailedharvest plan. In forests from Borneo to Brazil, loggers partici-pating in research projects on “reduced-impact logging” havedemonstrated that good logging practices are technically fea-

Figure A Selective logging typical involves cut-ting a few valuable species, and extracting themvia skidders or draft animals. (Photograph by M.Pinard).

0 20 40 60 80 100 140120 1600

10

20

30

40

50

60

Harvest intensity (m3/ha–1)

Res

idua

lsta

ndd

amag

e(%

ofun

harv

este

dst

ems)

fpo-hi res

Figure B Variability in harvest intensities (m3/ha–1) and incidentaldamage to the residual stand (percentage of unharvested trees) acrossthe tropics. Open symbols are areas in which harvesting was plannedand controlled; closed symbols are areas in which harvesting was car-ried out in a conventional way (typical for the region, generally withlittle planning and control over operations). (Modified from Pinard etal. 2005.)

fpo

sible and that their application is financially viable (e.g.,Holmes et al. 2002). Logging damage to the residual stand canbe reduced by up to 50% when trees are felled in predeter-mined directions and logs are skidded out of the forest alongpre-planned pathways (e.g., Pinard et al. 2000). Unfortunately,due to unsupportive forest policies, industry resistance tochange, corruption, and the massive profits from exploitativelogging, reduced-impact logging methods are seldom fullyused (Putz et al. 2000).

The socioeconomic and political contexts in which forestsare exploitatively logged or managed for the sustained yield oftimber vary tremendously across the tropics. Most tropical for-est-rich countries are poor in terms of most economic meas-ures, many stagger under huge international debts to develop-ment banks, and some are poorly governed. The expertiseneeded to manage forests is often scarce even in countries withgovernments committed to maintaining a permanent forest es-tate. Furthermore, corruption can seriously impede the best-in-tended efforts at management. Making matters more compli-cated is the fact that forests may be owned by individuals,communities, corporations, or governments that grant conces-sions to industrial corporations and, all too often, claims ofownership overlap or are otherwise contested. Equally worri-some is the fact that although most of the large forest tracts thatremain in the tropics are currently in remote locations far frommarkets, accessibility is rapidly increasing as major road-build-ing efforts push back forest frontiers. It also should be notedthat many forests planned for forest management are actuallyinhabited by people who are not involved in managementplanning and other decision-making processes. These social,economic, and political challenges are coupled with the bio-physical challenges of managing little-known forests underwhat are often adverse environmental conditions.

High Financial Opportunity Costs of Maintaining ForestsAnyone promoting sustainable forest management must grap-ple with the high financial opportunity costs of managingforests relative to the profits from extracting all of the mar-ketable trees and converting the forest to some more lucrativeland use. By “opportunity costs,” economists refer to the costassociated with giving up or postponing an opportunity, suchas the chance to invest in alternative income-generating activi-ties. This challenge is particularly powerful in many tropicalcountries where industrial agriculture, cattle ranching, andpredatory logging pay so much better than maintaining forestunder sustainable management (e.g., Pearce et al. 1999; South-gate 1998). Due to the “time value of money” (future profitsand future costs are “discounted” in terms of current value, seeChapter 5), even unsustainable and eventually very costlypractices can be financially very attractive. Market failures,particularly the serious undervaluing of forest goods and serv-ices, and policy failures, such as the provision of subsidies forunsustainable agricultural practices, further undermine efforts

to promote sustainable forest management (Richards 1999). Examples of how high opportunity costs interact with pol-

icy failures to promote deforestation in the tropics could becited from almost anywhere, but the liquidation of forests inthe 1980s in Costa Rica is especially illustrative. Costa Ricanforest management is conducted on a strictly private basis.There are no government concessions, nor does the govern-ment directly manage production forests. Up until recently, theopportunity costs of managing forests for sustained timberproduction were apparently too high to make the option at-tractive (Kishor and Constantino 1993). In addition, CostaRica’s relatively young, volcanically-derived soils can supportpasture for many more years than those established in less fer-tile locations such as in eastern Amazonia. Up until 1969, whenthe first modern forestry law was passed in Costa Rica, forestswere legally considered impediments to land conversion forpasture, agriculture, and development activities. During the1970s and 1980s, forest conversion was fuelled by governmentsubsidies for cattle production, coupled with excessive regula-tion of private landowners that caused them to lose interest inmanaging forests on a sustainable basis. As a result, annual de-forestation rates and consequent rates of forest fragmentationwere very high (Sanchez-Azofeifa et al. 2001). The drivingforces for forest conversion in Malaysia and Mexico were dif-ferent from those in Costa Rica, but ended up with the samemassive scale of deforestation.

Almost all Malaysian forests were claimed by the govern-ment and let out to logging concessionaires, but after logging,many were converted into oil palm plantations. Societal finan-cial benefits from logging, while a great deal less than theymight have been due to corruption and mismanagement, fu-elled national development. In Malaysia today, slash-and-burnagriculture is rare and many people live in cities and work inhigh-tech industries but the biological costs of this develop-ment were immense.

In Mexico, the extensive deforestation that occurred in the1960s and 1970s was an indirect result of the “green revolu-tion,” a global initiative to terminate hunger. The governmentimplemented a “national deforestation plan” in which small-scale farmers were given the right to lands they converted toagriculture. The objectives of the program were to producemore agricultural goods and to relieve political pressure fromlandless farmers. Of course, the cost of such program was paidby society at large, as the value of tropical forests was simplyignored (Dirzo and García 1991; Masera et al. 1997; Turner etal. 2001).

Pressures to OverexploitAs countries rich in forests develop economically, the growthof capacity in wood-processing industries often exceeds thesustainable rate of supply of raw materials. Typically, whereindustries initially thrive on a large supply of timber from for-est conversion and first cuts in heavily stocked, old-growthforests, with time, forests available for conversion decline and


the industries need to rely on the more modestharvests from the permanent forest estate (South-gate 1998). This type of “boom and bust” cyclehas occurred in many different parts of the trop-ics. For example, the government of Ghana isstruggling to impose an annual allowable cut ofonly 1 million m3, which is based on the bestavailable growth and yield data but is far belowthe industry’s capacity (Adam 2003). As in manyother places, Ghana’s timber-processing industryhas great political power and is applying pressureon the government to increase the allowable an-nual cut beyond the maximum sustainable yield(Adam 2003). Any further increases in harvestingcould only be achieved by reducing the lengths ofcutting cycles or by increasing the intensity ofeach harvest by selecting more species or moretrees of the favored species by reducing mini-mum felling diameters. Available data on stand-ing volume, forest productivity, and ecologicalimpacts suggest that these options are not ecolog-ically viable in even the short term and are noteconomically viable in the long term.

Insecurity of Land TenureAmong the problems plaguing tropical countries, propertyrights issues figure prominently. Because forest managementis a long-term endeavor, insecurity of land tenure, or at leastguaranteed access to the resources the forest provides, is a nec-essary prerequisite for making investments in sustainableforestry. The failure of many tropical governments to assurelandowners and forest concession holders that they will con-tinue to control their property forever, or at least receive ap-propriate compensation if it is taken away from them, is apowerful justification for treating forests as if their natural re-sources are not renewable. But while security of tenure seemslike an important prerequisite for sustainable management, itis not in itself sufficient to assure that forests will be treated asif the future matters. Under some conditions, providing securetenure can promote deforestation if forest owners use theirproperty as collateral for bank loans to buy cattle or chainsaws.

Marketing and Managing for More SpeciesAlthough technical challenges for tropical silviculture (e.g.,timber harvesting and stand tending) tend to be secondary tosocioeconomic and political problems, applied forest ecologistsstill have important roles to play. These roles are expanding asconservation and environmental service provisions are includ-ed more explicitly in management priorities. Furthermore, asmarkets begin to accept formerly noncommercial timberspecies, studies on the population biology and physiologicalecology of these species become critical.

Tropical silviculture has its roots in the management of afew high value Asian and African species such as teak (Tectonagrandis) and the African mahoganies (Khaya spp., Entan-

dophragma spp.). In contrast to conditions in the first half of thetwentieth century, forest managers in the moist tropics nowcan market dozens of species (Figure C). In Amazônas, for ex-ample, in the early 1990s, 90% of the state’s timber productionwas based on four species, Copaifera multijuga, Ceiba pentandra,Nacleosis caloneura, and Virola surinamensis. With the exhaustionof the economically accessible stocks of these species, increaseddemand for sawn timber, and improvement of law enforce-ment, new industries have been established and the list ofspecies used by industry has expanded considerably (e.g., to70 species in one 40,000-ha project; Freitas 2004).

Long lists of marketable timber-producing tree species pres-ent managers with both opportunities and risks. Greater num-bers of commercial species increases a manager’s flexibility inselecting trees to be harvested, trees to be retained as seedsources, and future crop trees to be favored by judicious thin-ning around their crowns. This flexibility is important giventhe uncertainties in timber markets and the wide range of con-ditions appropriate for the regeneration and growth of valu-able tree species. On the other hand, there is a risk of over-har-vesting, particularly where there is little control of forestharvesting operations or where harvesting decisions are basedsolely on minimum felling diameters without concern for theregeneration requirements of the harvested species.

Variation among ForestsTropical forests are justly famous for their overall species diver-sity, but ecologists have only recently recognized how much di-versity there is among forests in the tropics. Similarly, forest man-agers are becoming increasingly aware that the silviculturaltreatments they apply must be appropriate for the particularcharacteristics of the species and forests they are managing. Forexample, retaining canopy cover by carefully harvesting spa-

282 Chapter 8

GroomFigure Gro0803bCaseStudy.eps

0

10

20

30

40

50

60

70

1939

1935

1931

1927

1923

1919

1915

1911

1981

1977

1973

1969

1965

1962

1947

1943

1999

1985

Year

Num

ber

ofex

ploi

ted

spec

ies

Figure C The number of tree species exploited for timber in Ghana between 1911and 1989. Data compiled from annual reports of the Ghana Forestry Department. Forthose years without bars, no data were available. (Modified from Adam 2003.)

fpo

tially isolated trees is critical for managing forests forshade-tolerant trees that are abundant in the sub-canopy and understory. The same approach to har-vesting light-demanding species, in contrast, wouldnot contribute to sustainability because these speciesonly regenerate in large clearings. Foresters in theDipterocarpaceae-dominated forests of southeast Asiaas well as those working in the forests on the GuyanaShield of South America, where many of the commer-cial timber tree species are at least moderately shade-tolerant, are justifiably adamant about reducing theimpacts of logging on the residual stand. Conversely,silvicultural intensification, including exposure of min-eral soil in large canopy openings, is being called for inseasonal forests on the rim of the Amazon where mostof the commercially important timber tree species arelight-demanding (Fredericksen and Putz 2003). Whilemany forest management practices are widely applica-ble, such as the need to avoid sediment loading ofstreams with runoff from logging roads, no single sil-vicultural system is appropriate for all tropical forests.

Variation among forests also means that some forests maybe easier to manage than others. For example, some of greatestsuccess in management of tropical forests has been in low di-versity forests with a high density of commercial timber trees.Mangrove and swamp forests are particularly notable, al-though low diversity forests can also be found in uplands. Forexample, silviculture in oak-dominated montane forests inCosta Rica has proven to be straightforward partially becausethe canopy trees produce very large crops of acorns that satiatetheir seed predators resulting in high seedling densities in theunderstory (Saenz and Guariguata 2001; Guariguata andSaenz 2002). The lowland forests of Peninsular Malaysia areextremely diverse, but their management is also relatively easybecause the canopy is dominated byspecies of Dipterocarpaceae thatshare silvicultural requirements. Un-fortunately, the successful regenera-tion of an extensive area by good for-est management has been erased asmost of these areas have been cut tomake way for oil palm plantations(Figure D).

Working With Less Valuableand Fragmented ForestsAfter widespread deforestation forcattle ranching, industrial farmingand other agricultural uses, as wellas oil palm and wood fiber planta-tions, the land left for natural forestmanagement for timber is mostly re-mote, steep, rocky, or swampy. Theconditions that render land unsuit-able for more intensive uses also add

to the challenge of forest management. Natural forest man-agers have to be comfortable working in remote areas andneed to be very sensitive to environmental issues whereslopes are steep and soils are easily damaged by heavy equip-ment. Additionally, the costs of applying even the simplest ofstand improvement treatments, such as cutting vines on fu-ture crop trees, can become excessively expensive under theseadverse conditions.

Managers should also become aware that timber-producingforests are getting smaller. In the neotropics, while the annualcutting area of a single logging company in the Bolivian Ama-zon might average 2000 ha, the total area of production forest ofindividual concessionaires or forest owners in Nicaragua, CostaRica, and Honduras rarely exceed this same figure (Table A; one


Figure D Oil palm plantations are becoming more common, replacing selectivelogging practices in much of Southeast Asia. (Photograph by M. Pinard).

fpo-hi res

Country Mean (range)a Nb Observations

Honduras 1342 (196–4149) 45 Includes community and privately-owned forest in the northern lowlands

Nicaragua 2814 (5–42887) 39 Includes the entire country; management plans granted during 1999 only

Costa Rica 62 (2–481) 404 Privately-owned forests in the northern lowlands(excludes Carapa-Pentaclethra forests on poorly-drained soils)

Perú 17460 (2697–49620) 31 State and privately owned forest, Amazonian lowlands

Bolivia 62420 (1171–365847) 95 Concessions and community forests, Amazonian lowlands

Source: Data from the following governmental institutions: Corporación Hondureña de Desarrollo Forestal(COHDEFOR) and Instituto Nacional Forestal (INAFOR). Data also from the non-governmental organizationFundación para el Desarrollo de la Cordillera Volcánica Central (FUNDECOR), and from the bilateral/multi-lateral research initiative BOLFOR and Center for International Forestry Research (CIFOR).aThe mean (range) of the forest area is measured in hectares.b Number of sites.

TABLE A A Regional (Central versus South American) Contrast in the EstimatedAreas of Mixed-Species, Lowland Broad-Leaved Forest Currently Managedfor Timber

notable exception to this pattern is the forest community con-cessions in Petén, Guatemala that reach thousands of hectares[ha] [Ortiz and Ormeño. 2002]). Although generalizing howfragmentation affects the biological sustainability of timbermanagement is currently difficult, penetration of light and windinto fragmented forests from their edges results in increasedflammability, a huge concern given the frequency of human in-duced ignitions of wildfires in many areas of the tropics (Lau-rance and Cochrane 2001). Also, vines and other forest weedsproliferate on forest edges, and can greatly increase the mortali-ty of edge-exposed trees (Laurance et al. 2001). Seed dispersal byvertebrates can also vary between fragmented and extensivelyforested areas (Guariguata et al. 2002).

Biological considerations aside, a potential constraint ontimber production in small fragments of forest relates to land-scape-level planning of management operations. Forest frag-mentation may limit the viability of selective logging schemesthat might work well in extensive forests because of the lowprobability of securing an adequate volume of a given timberspecies. Even well intentioned policies can exacerbate thisproblem. By law in Costa Rica, for example, harvesting quotasfor a given timber species should not exceed 60% of the totalnumber of harvestable individuals (Comisión Nacional deCertificación Forestal 1999). Given that most Costa Ricanforests destined for timber production are only a few tens of hain extent, harvesting quotas guided by the “60%” rule may befinancially unrealistic in many forest patches due to the smallabsolute number of trees.

How the Challenges of Tropical Forestry Can Be ConfrontedMany approaches may be helpful in reducing unsustainableforestry practices. Among these, demand from consumers, ad-justed markets, policy changes, and technological advances allmay play a part, often together.

Pressure from Socially and Environmentally Concerned Consumers

Frustrated by continued unsustainable forestry practices in thetropics, environmentally concerned consumers of forest prod-ucts are using their market power to promote better manage-ment. Initially, boycotts of tropical forest products were staged,but they back-fired—driving down the market price of tropi-cal timber only served to increase the rate of logging as com-panies strove to maintain their profits. Much more successfulhas been an international program of voluntary third-partycertification of good forest management practices coordinatedby the Forest Stewardship Council (FSC).

While every year the area of tropical forest certified as wellmanaged according to FSC Principles (Table B) increases byabout 1 million ha, global demand for certified forest productscontinues to exceed the supply. Unfortunately, consumerawareness of the beneficial influences of selecting certified for-est products in the U.S. still lags far behind Europe. As more

and more Americans move from the countryside to cities andsuburban areas where they have few opportunities to learn thedifference between sound and unsound forest managementpractices, efforts to promote forest management as a conserva-tion strategy will become increasingly challenging. Counter-balancing this trend, country of origin labeling, organic foodcertification, “fair-trade” programs, and marketing of “bird-friendly, shade-grown” coffee, are all helping to make con-sumers in the U.S. aware of their potential power to increasesocial welfare and protect the environment.

Paying the true value for tropical timberOne way to reduce the opportunity costs inherent in managingforests for timber is to capture the full economic value of trop-ical forests. Many of the values of forests are not included intraditional market transactions for timber and even most non-timber forest resources (e.g., rattan canes, Brazil nuts, and palmhearts). Although we recognize the roles of tropical forests asstorehouses of fantastic biodiversity and for their effects onlocal and global climates, forest owners are generally not fi-nancially compensated for these values. The costs or benefitsof forests that accrue to people other than the property ownersare known as externalities. Economists are working with ecol-ogists around the world to capture these externalities in eco-nomic analyses by creating markets for sequestered carbon,transpired water, and protected biodiversity. In these efforts,Costa Rica has been a world leader.

An innovative mechanism for forest conservation, Paymentfor Environmental Services, was included in Costa Rica’s mostrecent forestry law and implemented through a decentralizedbody (National Forestry Financing Fund [FONAFIFO]). Thesepayments represent an attempt to financially compensate smalllandholders for the services their forests provide including car-bon sequestration, watershed functions, maintenance of biolog-ical diversity, and protection of scenic beauty, going some wayto reduce the opportunity costs of maintaining forest cover.

Markets for sequestered carbon are developing all over theworld, but use of carbon credits to promote forest conservationthrough improved management faces a number of seriouschallenges. Unfortunately, reforestation of deforested areas, af-forestation of naturally nonforested areas (e.g., grasslands andsavannas), and, to a lesser extent, strict forest protection haveall received more support from international policymakersthan reduced-impact logging and other changes in forest man-agement that promote carbon sequestration (e.g., Putz andPinard 1993). Part of the problem is that many policymakerscannot recognize the difference between well and poorly man-aged forests, and doubt that anyone else can either.

Policy and legal reformPolicy and legal reform are often prerequisite to the removal ofdisincentives to sustainable forest management. For example,ejidatarios (communal owners) or indigenous communities con-trol 80% of the forest resources of Mexico. Initially they were

284 Chapter 8

obliged by law to sell their forest products to companies andthus had limited real control over the fates of their forests. Thislegal arrangement discouraged sustainable use and fuelled con-flicts between the companies and the ejidatorios. The companieswere not interested in sustainable management because theirconcessions were short term and they had no prospects of be-coming forest owners. The ejidatarios, in turn, had little interestin the forests because they had such little control over their use.This conflict was resolved in the 1980s when a new law allowedejidatarios to manage their forests directly. Today, the best exam-ples of good forest management in Mexico come from ejidatar-ios and indigenous communities that recognize the long-termvalue of keeping their forests (Flachsenberg and Galletti 1998;Bray 2004). Moreover, some of the indigenous communities ac-

tively promote reforestation of landsformerly cleared for agriculture(Bocco et al. 2000).

Policy reforms are only effective inimproving forest management prac-tices when supported by the appro-priate regulations and the institutionsneeded to implement and enforce thenew policies. In Bolivia, for example,the passage of new forestry legislationin 1996 was followed by the creationof a new institution that was assignedthe task of assuring compliance withthe legislation. In the first year of ac-tivity, the new national forestry au-thority resolved millions of hectaresof conflicting land claims, issuedlong-term cutting contracts, and in-creased tax revenue to the Boliviangovernment four-fold (Nittler andNash 1999).

Technological strategiesTropical silviculturalists have ap-proached the problem of managinghigh diversity in tropical forests invarious ways. In the 1960s, uniformsilvicultural systems (e.g., shelter-wood systems or patch clearcuts)were promoted as a means of domes-ticating the forest and increasing pro-ductivity by favoring a few light-de-manding species at the expense ofother more slow-growing, shade tol-erant species. Polycyclic managementsystems (those more commonly usedtoday) were considered less appro-priate because of the high losses ofvolume caused by incidental damageassociated with selective timber har-

vesting, the risk of a shift in forest composition to more shade-tolerant species, and the risk of over-harvesting if cutting cy-cles are too short (Dawkins and Philip 1998).

Domestication of tropical forests through intensive silvicul-tural interventions is generally not practiced today despite thefact that the silvicultural objectives of increasing stocking andgrowth rates appear to have been met in many forests treated asearly as the 1950s (e.g., Alder 1993; Manokaran 1998; Osafo 1968).Two factors appear related to the lack of support for radical do-mestication—one is the relatively high risk of forest clearing foragriculture during the first decades in the management cyclewhen the forest holds little value, and the other is an assumed in-compatibility between intensive silviculture and biodiversityconservation (Dawkins and Philip 1998; but see de Graaf 2000).


Principle 1: Compliance with laws and FSC principles Forest management shall respect all applicable laws of the country in which they occur,and international treaties and agreements to which the country is a signatory, and complywith all FSC principles and criteria.

Principle 2: Tenure and use rights and responsibilities Long-term tenure and use rights to the land and forest resources shall be clearly defined,documented and legally established.

Principle 3: Indigenous peoples’ rights The legal and customary rights of indigenous people to own, use and manage their lands,territories, and resources shall be recognized and respected.

Principle 4: Community relations and worker’s rights Forest management operations shall maintain or enhance the long-term social and eco-nomic well-being of forest workers and local communities.

Principle 5: Benefits from the forest Forest management operations shall encourage the efficient use of the forest’s multipleproducts and services to ensure economic viability and a wide range of environmental and social benefits.

Principle 6: Environmental impact Forest management shall conserve biological diversity and its associated values, waterresources, soils, and unique and fragile ecosystems and landscapes, and, by so doing,maintain the ecological functions and the integrity of the forest.

Principle 7: Management plan A management plan—appropriate to the scale and intensity of the operations—shall bewritten, implemented, and kept up to date. The long term objectives of management, andthe means of achieving them, shall be clearly stated.

Principle 8: Monitoring and assessmentMonitoring shall be conducted—appropriate to the scale and intensity of forest manage-ment —to assess the condition of the forest, yields of forest products, chain of custody,management activities and their social and environmental impacts.

Principle 9: Maintenance of high conservation value forests Management activities in high conservation value forests shall maintain or enhance theattributes which define such forests. Decisions regarding high conservation value forestsshall always be considered in the context of a precautionary approach.

Principle 10: Plantations Plantations shall be planned and managed in accordance with principles and criteria 1–9,and principle 10 and its criteria. While plantations can provide an array of social and eco-nomic benefits, and can contribute to satisfying the world’s needs for forest products, theyshould complement the management of, reduce pressures on, and promote the restorationand conservation of natural forests.

Source: Modified from Forest Stewardship Council website, http://www.fscoax.org/.

TABLE B Ten Principles of Forest Stewardship According to the Forest Stewardship Council

Typical approaches to timber management in natural trop-ical forests currently are conservative, aimed at reducing theimpacts of interventions (Fredericksen and Putz 2003). To agreat extent, reduced-impact logging has become synonymouswith sustainable forest management, even though the formeris only one component of the latter. Where light-demandingspecies are being harvested yet failing to regenerate, more in-tensive, locally imposed interventions are being promoted(Dickinson and Whigham 1999; Fredericksen and Mostacedo2000; Snook 1996). For example, thinning interventions thatliberate individual crop trees and maintain overall diversity instructure are sometimes recommended (see Wadsworth 1997for review). In forests in which the commercial timber is pro-duced by both light-demanding and shade-tolerant species, amixed system would seem most appropriate. For example, sin-gle-tree selection might be applied where there is advance re-generation of shade-tolerant species, while in other areas largegaps might be created near potential seed trees to promote re-generation of light-demanding species (Pinard et al. 2000).

Looking Forward:Where Will the Next Decades Lead? Whilst uncertainty in timber prices, politics, and prioritiesmake it difficult to predict where the next decades will leadtropical forestry, a few pathways seem more likely than others.Overall, while deforestation rates are likely to remain high, for-est management practices should continue to improve underthe influence of the FSC’s timber certification processes (seeTable B). These improvements will be a consequence of moreforest owners working towards certification, certified produc-ers improving their practices through the monitoring and feed-back programs built into the certification process, and in re-sponse to indirect influences of certification such as theexchange of information and experience among forest owners,managers, and regulators.

It is also clear that the focus of the timber industry will in-creasingly move to marginal lands where opportunity costs arelow and natural forest management is relatively attractive as aland use. With widespread devolution of control of forestlands to indigenous groups and other rural communities, tim-ber buyers will be motivated to develop socially acceptable ap-proaches to company–community partnerships (Mayers and

Vermeulen 2002). Farm forestry will also continue to develop,contributing to a shift in milling capacity to process smallerstems. As the big profit margins associated with exploiting oldgrowth forest give way to the more modest profits associatedwith sustainable harvests, smaller operators are likely to be at-tracted to the business. Provision of technical support to manysmall operators may be more difficult than to fewer large op-erators, but levels of motivation and innovation may also behigher. Where managers are committed to sustainable prac-tices, more silviculture will be introduced into management.Meanwhile, integration of inventory data, harvest planning,and monitoring operations into geographic information sys-tems (GIS) will facilitate the implementation of more intensiveand directed operations.

The trends in policy towards broadening societal participa-tion in forestry are likely to continue with more efforts to de-volve forest management to rural communities. But early ex-perience gained across a broad range of socio-political andeconomic conditions emphasizes the need for strategies thatcombine both short- and long-term objectives (du Toit et al.2004), institutional development, conflict resolution (Castroand Nielson 2001), and a recognition that in some cases, it willtake a lot of time and effort for communities of subsistencefarmers to become sustainable forest managers.

As pressure to place remaining tracts of natural forest underprotection, the divide between those arguing for sustainableuse and those arguing for complete protection may grow wider(see Dickinson et al. 1996 and Pearce et al. 2003 for summariesand primary literature therein), one consequence being in-creased competition for sources of funds for development. Allprogress towards sustainable forest management depends ongood governance. Where corruption and illegal logging contin-ue, it will be difficult to improve forest management practices(Ravenal et al. 2004).

The challenges facing tropical forest management are many.The field of natural forest management in the tropics needs en-ergetic and dedicated people who are willing to put up withthe rigors of tropical forestry work. Input from biologists isneeded to ensure that the development of good practices, andthe criteria and indicators used to assess and monitor thesepractices, are based on relevant biological and ecological infor-mation (Putz and Viana 1996).

286 Chapter 8

Summary1. Overexploitation is ranked second only to habitat

loss as a cause of extinction risk in species whosestatus has been assessed globally. It involves the un-sustainable use of wild animals, plants, and theirproducts for purposes such as food, medicines, shel-ter, and fiber. The motivations for such exploitationare as varied as the plants and animals that are

taken, ranging from subsistence hunting and fish-ing, to recreational and economic pursuits carriedout by wealthy individuals and corporations. Whilemost environmentalists are well aware that fisheriesand large-bodied birds and mammals are oftenoverexploited, recent studies have shown that thesechanges can be far more dramatic than previouslyrecognized. Fisheries can reduce the biomass of tar-geted and nontargetted species by 80% or more in


10–20 years, and subsistence hunters in the Neotrop-ics can create a vacuum of large-bodied specieswithin 10–20 miles of their villages.

2. The biological theory of sustainable exploitation isfirmly rooted in the field of population ecology,which seeks to understand the responses of popula-tions to increased mortality of individuals throughdensity-dependent compensation. This theory hasproduced a range of methods for estimating sus-tainable limits of exploitation from simple rules ofthumb based on life histories to highly sophisticatedmodels that are appropriate for only a small fractionof cases where the necessary input data are avail-able. This theory cannot be put into practice to makeexploitation sustainable without a clear understand-ing of the motivations of people who use wild ani-mals and plants, the alternatives available to them,and the effects of management recommendations ontheir livelihoods. The “tragedy of the commons”often looms large in overexploitation in both terres-trial and aquatic ecosystems, exacerbated by “dis-counting,” whereby we give wild populations ahigher value for what we can get from them todaythan in the future. Yet there is a growing realizationof the enormous economic and social benefits thatwe receive from ecosystem goods and services. Withencouragement from processes such as the WorldSummit on Sustainable Development in 2002 andthe Millennium Ecosystem Assessment, many peo-ple are now rising to the challenge of finding inno-vative ways helping people to reap the rewards ofsustainable exploitation.

Questions for Further Discussion1. How can rich countries help encourage poor coun-

tries to exploit their plants and animals sustainably?

2. Actions taken to protect vulnerable species may af-fect some people’s livelihoods. Because the decisionto protect biodiversity is an expression of publicwill, do you think that governments have an obliga-tion to assist families and businesses whose liveli-hoods are harmed by a conservation plan?

3. Many cases of overexploitation involve not onlypeople’s livelihoods, but their subsistence. How canwe prioritize conservation versus human subsis-tence? What mechanisms might we foster to reducethis conflict?

4. In what ways can models help us exploit wild pop-ulations more sustainably? What are some of thelimitations of the models typically used to estimatesafe harvest levels?

5. Why is overexploitation so large a problem in themarine realm, and relatively less important in mostterrestrial biomes?

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

overexploitation - ueae436/reynolds_and_peres_chap08.pdf · and wood bison (bison bison athabascae)...

Documents