sign in sign up
<<
Home , Extinction, Habitat, Habitat destruction, Introduced species

Quantifying Threats to Imperiled in the United States Author(s): David S. Wilcove, David Rothstein, Jason Dubow, Ali Phillips, Elizabeth Losos Reviewed work(s): Source: BioScience, Vol. 48, No. 8 (Aug., 1998), pp. 607-615 Published by: University of Press on behalf of the American Institute of Biological Sciences Stable URL: http://www.jstor.org/stable/1313420 . Accessed: 22/03/2012 16:18

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information and tools to increase and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

University of California Press and American Institute of Biological Sciences are collaborating with JSTOR to digitize, preserve and extend access to BioScience.

http://www.jstor.org QuantifyingThreats to Imperiled Species in the United States Assessing the relative importance of destruction, alien species, , ,and disease

David S. Wilcove, David Rothstein,Jason Dubow, Ali Phillips,and ElizabethLosos

iologists are nearlyunanimous fine-scale analysis of the types of in their belief that humanity is affecting US in the process of extirpating a Habitat loss is the and protected under significant portion of the 's spe- threat the federal Act cies. The ways in which we are doing single greatest (ESA). We also speculate on how so reflect the magnitude and scale of to , followed these threats have changed over time enterprise. Everything from and are likely to change in the future. highway construction to - by the spread of We conclude with a brief discussion ing to leaky bait buckets has been of the implications of our findings implicated in the demise or endan- alien species for the long-term protection of im- germent of particular species. Ac- periled species in the United States. cording to Wilson (1992), most of these activities fall into four major passed by them-is responsible for An overview of the threats categories, which he terms "the mind- endangering species. In general, sci- less horsemen of the environmental entists agree that habitat destruction To obtain an overview of the threats ": overexploitation, habi- is currently the primary lethal agent to biodiversity in the United States, tat destruction, the introduction of (Ehrlich 1988, Wilson 1992), fol- we tabulated the number of species non-native (alien) species, and the lowed by the spread of alien species threatened by five categories of spread of diseases carried by alien (Wilson 1992). However, apart from threats: habitat destruction, the species. To these categories may be several notable exceptions-includ- spread of alien species, overharvest, added a fifth, pollution, although it ing studies of North American pollution (including siltation), and can also be considered a form of by Williams et al. (1989), endan- disease (caused by either alien or habitat destruction. gered plants and animals in the native ). We restricted this Surprisingly, there have been rela- United States by Flather et al. (1994, coarse-scale analysis to imperiled tively few analyses of the extent to 1998), aquatic by Richter plants and animals occurring within which each of these factors-much et al. (1997), and imperiled by the 50 states and falling into any of less the more specific deeds encom- Collar et al. (1994)-few quantita- four categories: all full species of tive studies of threats to have David S. Wilcove is a senior at species , birds, , amphib- ecologist been conducted. More such studies and with status ranks of the EnvironmentalDefense Fund, Wash- ians, DC 20009. David Rothstein re- are needed to provide conservation- "possibly extinct," "critically imper- ington, as ceivedhis J.D. in 1997 fromNortheastern ists, stewards, and decision iled," or "imperiled," determined UniversityLaw School, Boston, MA 02115 makers with a better understanding by The Conservancy (TNC) and now works for the federal govern- of the relationships between specific in association with the Network of ment. Jason Dubow is an environmental human activities and the loss of Natural Heritage Programs and Con- analyst at the Eastern Research Group, biodiversity. servation Data Centers (Master Arlington, VA 22201; he formerly worked In this article, we quantify the 1991); all full species of freshwater for The Nature Conservancy.Ali Phillips extent to which various human ac- , butterflies and skippers, ti- is a natural resources at The specialist tivities are imperiling and ani- ger , and and - WildernessSociety, Denver, CO 80221. mal in the United States. Our selflies with status ranks of ElizabethLosos is directorof the Center species possibly for Sciences at the analysis has two parts: a coarse-scale extinct, critically imperiled, or im- Tropical examination of the numbers and as determined all Smithsonian Tropical Research Institute, periled, by TNC; of US the full of vascular with Washington, DC 20560. ? 1998 Ameri- types species imperiled by species plants can Instituteof BiologicalSciences. major categories of threats, and a status ranks of possibly extinct or

August 1998 607 Table 1. Taxonomic breakdown of species used in the coarse-scale analysis. Included possible to assign a particular hu- are species classified as imperiled by The Nature Conservancy and all species, subspe- man activity to one of the major threat cies, and that, as of January 1996, are listed as endangered or threatened categories; we excluded these activi- under the Act or have been Endangered Species formally proposed for listing. ties from our coarse-scale analysis. We were able to obtain informa- Number of of Percentage tion on threats for 1880 of the Number of imperiled species imperiled species (75%) imperiled species with threats data with threats data 2490 imperiled species, , and that met our crite- 541 494 91 populations Mammals 88 85 97 ria for inclusion in this study (Table Birds 101 98 97 1). (For 52 of the species, we could Reptiles 40 38 95 not identify any anthropogenic 69 60 87 threats.) We used the resulting data- Fishes 243 213 88 base to determine the relative signifi- 471 331 70 cance of the major threats categories Dragonflies and 33 18 54 and to investigate differences be- damselflies tween species groups in their vulner- Freshwatermussels 150 102 68 ability to particular threats. We com- Crayfish 110 67 61 the distribution of threats beetles 8 6 75 pared Butterfliesand 46 33 72 among plants and animals, among skippers and animals, Other invertebrates 124 104 84 and within vertebrate classes. We also the distribution of Plants 1478 1055 71 compared threats among terrestrial and aquatic Total 2490 1880 75 species, Hawaiian and mainland vas- cular plants, and Hawaiian and main- land birds. For all comparisons, sta- critically imperiled, as determined nated as threatened or endangered tistical significance was assessed by TNC; and all species, subspecies, under the ESA), a survey of biolo- using the chi-squared contingency or vertebrate populations listed by gists conducted by Richter et al. test (two-tailed). the US Fish and Service (1997) for aquatic species, the Natu- We emphasize at the outset that (USFWS) or the National Marine ral Heritage Central Databases man- there are some important limitations Fisheries Service as threatened or aged by TNC, and interviews with to the data we used. The attribution endangered or officially proposed for specialists in particular species of a specific threat to a species is listing under the ESA as of 1 January groups and geographical regions. We usually based on the judgment of an 1996. (The ESA permits the listing of included only known threats and ex- expert source, such as a USFWS em- species and subspecies of plants and cluded potential or hypothetical ones. ployee who prepares a listing notice animals as well as "distinct popula- We did not attempt to distinguish or a state Fish and employee tion segments" of vertebrates.) A between ongoing and historical who monitors endangered species in total of 2490 imperiled species, sub- threats, partly because such infor- a given region. Their evaluation of species, and populations fit these mation is usually lacking and partly the threats facing that species may criteria. because the distinction itself is prob- not be based on experimental evi- Information on the threats to each lematic in the case of habitat de- dence or even on quantitative data. of these species, subspecies, and struction. Nor did we try to distin- Indeed, such data often do not exist. populations was obtained from a guish between major and minor With respect to species listed under number of sources, including the threats to each species because such the ESA, Easter-Pilcher (1996) has Federal Register (i.e., the listing no- information was not consistently shown that many listing notices lack tices published for all species desig- available. In a few cases, it was im- important biological information,

Table 2. Percentages of species in different groups that are imperiled by habitat degradation and loss, alien species, pollution, overexploitation, and disease. Categories are nonexclusive and therefore do not sum to 100.

Fresh- Butter- Other All Verte- Inverte- Amphi- Tiger and inverte- species brates brates Plants Mammals Birds Reptiles bians Fishes mussels Crayfish beetles skippers brates Cause (n= 1880) (n=494) (n= 331) (n= 1055) (n=85) (n=98) (n=38) (n=60) (n=213) (n= 102) (n=67) (n=6) (n= 33) (n= 104)

Habitat 85 92 87 81 89 90 97 87 94 97 52 100 97 94 degrada- tion/loss Alien 49 47 27 57 27 69 37 27 53 17 4 0 36 52 species Pollution 24 46 45 7 19 22 53 45 66 90 28 0 24 19 Overex- 17 27 23 10 45 33 66 17 13 15 0 33 30 46 ploitation Disease 3 11 0 1 8 37 8 5 1 0 0 0 0 0

608 BioScience Vol. 48 No. 8 1. The including data on past and possible Figure major 100% threatsto future impacts of habitat destruc- biodiversity. 90% Data refer to CD tion, pesticides, and alien species. species S80% classifiedas imperiled Depending on the species in ques- >% 70% by TheNature Conser- the absence of information a) S60% tion, may vancyand to all endan- reflect a lack of an a) s.) 50% data, oversight, gered,threatened, and or a determination USFWSthat a U) 40% by sub- a) proposedspecies, 0) particular threat is not harming the species, and popula- "3 30%- species. The extent to which such tions protectedunder c limitations on the data influence our theEndangered Species 20//2%10% results is unknown. Act. Seealso Table 2. 0%

1-00C0 C 5 the threats threats to aquatic Ranking biodiversityin North Table 2 presents a summary of the America(Richter et percentages of species that are im- al. 1997). these speciesindividually in its for- periled by habitat loss, alien species, mallisting notice in the FederalReg- pollution, overexploitation, and dis- A closerlook at ister.For the purposesof this partof ease. Not surprisingly, habitat de- habitatdestruction ouranalysis, we have therefore counted struction and degradation emerged the entiregenus as one "species.") as the most pervasive threat to Given the primacy of habitat de- biodiversity, contributing to the en- structionas a threatto biodiversity, Categorizinghabitat destruction. For dangerment of 85% of the species we examinedits causes in greater the fine-scaleanalysis, we divided we analyzed(Figure 1). Indeed,habi- detail. For this fine-scaleanalysis, habitatdestruction and degradation tat loss is the top-ranked threat (in we focused exclusivelyon US spe- into11 majorcategories (see box page terms of the number of species it cies,subspecies, and populations that 611). As in the coarse-scaleanalysis, affects) for all species groups. Com- have been addedto the federalen- we did not distinguishbetween cur- petition with or by alien dangeredspecies list or have been rent and historical threats or be- species is the second-ranked threat formallyproposed for suchlisting by tween majorand minor threats.In in the overall analysis, affecting49% USFWSas of 1 January1996. We manyinstances, the apparentthreat of imperiled species. focused on listed species because to a specieswas actuallyspawned by Alien species affect a higher pro- more informationis usually avail- anotherthreat. Whereverpossible, portion of imperiled plants (57%) ablefor themthan for imperiledbut we attributedthreats to their ulti- than animals (39%); this difference unlisted species. We also included mate cause, based on the informa- is statistically significant (chi square species that are federallylisted or tion in the FederalRegister. For ex- = 60.23, d.f. = 1, P<<0.001). How- proposed for listing from Puerto ample, operations near a ever, certaingroups of animals (most Rico, the US VirginIslands, and the streamcan leadto siltation,which is notably birds and fish) appear to be PacificTrust Territories. A total of harmfulto certain rare fishes and as broadly affected as plants by alien 1207 species,subspecies, and popu- mussels.Thus, logging ratherthan species. Thereis also an unsurprising lationswas includedin this phaseof siltationwould have been scoredas biogeographiccomponent to the alien the analysis(Figure 4). (USFWShas the threatto those fishes and mus- species problem:Higher proportions listed as endangeredall Hawaiian sels.For all comparisons of thepreva- of Hawaiian birds and plants than snailsof the genusAchatinella. Ap- lence of specificthreats in different continental birds and plants are proximately41 speciesin thatgenus speciesgroups, statistical significance threatened by alien species (Table 3, havebeen described to date,of which was assessedusing the chi-squared Figure 2). Similarly, a much higher at least 18 are thoughtstill to sur- contingencytest (two-tailed). proportion of Hawaiian birds is vive. However,USFWS did not treat Again,we note somecaveats with threatenedby disease than is the case for continental birds. By contrast, of Hawai- Table 3. Percentagesof imperiledbirds and plants in Hawaiiand in the continental nearly the same proportion and alien ian and continental are UnitedStates that are threatenedby habitatdegradation loss, species, plants plants pollution,overexploitation, and disease. Categories are nonexclusive and therefore do affected by disease (Table 3, Figure 3). not sumto 100. For all aquatic groups (am- phibians, fish, dragonflies and dam- Continental Hawaiian Continental Hawaiian selflies, freshwater mussels, and cray- US birds birds US plants plants = = = = fish), pollution is second only to Cause (n 56) (n 42) (n 641) (n 414) habitat loss as a cause of endanger- Habitat 88 93 90 66 ment. Our finding that a large num- degradation/loss ber of aquatic species are threatened Alien species 48 98 30 99 Pollution 38 2 12 0 by pollution may reflect the fact that Overexploitation 39 24 13 6 our definition of pollution includes Disease 4 81 1 0 siltation, which is one of the leading

August 1998 609 Ranking the causes of habitat de- 100% - - struction. The most overt and wide- 90% spread forms of habitat alteration were, as might be expected, the lead- 80%- ing threats to species that are either 70% listed or proposed for listing (hereaf- "Oa)'5 .-a 80% - 0_ g 70%- ter referredto collectivelyas "endan- ip(D gered" species), as measured by the m 50%- number of species they affect (Table atS 3n0%- 4). These forms include (affecting38% of endangeredspecies), commercialdevelopment (35%), wa- ter development (30% when agricul- lo-Q4% - tural diversion is included; 17% for 10% just , impoundments,and other barriers), and infrastructure devel- (17%). Not the 0% opment surprisingly, Continental Hawaiian Continental Hawaiian impacts of water development are U.S. birds birds U.S. plants plants felt most acutely by aquatic species. Ninety-one percent of endangered Figure2. A comparisonof the impactsof alien specieson imperiledbirds and plants in fish and 99% of endangeredmussels and in the continentalUnited States. A much higherproportion of Hawaiian are affected by water development, birds and plants than continentalbirds and plants is threatenedby alien species (chi- in contrast to 10% of mammals and square= 27.60, d.f. = 1, P <<0.001 for birds;chi-square = 484.28, d.f. = 1, P<<0.001 22% of birds. Within the of for plants). Data are taken from Table 3. category infrastructuredevelopment, af- fect a wide of (15% of 100% ------array species . all endangered species), confirming 90% their reputation as "a leading threat to and 80% biodiversity" (Noss Cooper- rider 1994). co 0) 70%-- Outdoor recreation also harms a number of (41) 60% large endangered species CL M (27%). It affectsa significantlyhigher u) 50% - proportion of plants than animals Cu >,, (33% vs. 17%; chi square = 39.03, - 40% .l a)U d.f. = 1, P<<0.001). Within the cat- S 30% egory of outdoor recreation, the use of off- vehicles is in 20%- implicated E• the demise of approximately 13% of - endangered species. 0lo%o% . . ... ; - Among extractive land uses, log- Continental Hawaiian Continental Hawaiian ging, , and grazing have con- U.S. birds birds U.S.plants plants tributed to the demise of 12%, 11%, and 22%, respectively,of the endan- 3. A of the of diseaseon birdsand in Hawaii gered species we analyzed. Both log- Figure comparison impacts imperiled plants and are seri- and in the continentalUnited States. A muchhigher proportion of Hawaiianbirds than ging mining especially continentalbirds is threatenedby disease (chi-square= 62.03, d.f. = 1, P<<0.001). By ous threats to freshwater mussels, contrast,similar proportions of Hawaiianand continental plants are affected by disease probably because they result in in- (althoughthe differenceis statisticallysignificant: chi-square = 4.02, d.f. = 1, P = .045). creased amounts of silt, in the cases of both logging and mining, and of toxic pollutants, in the case of min- respectto the data in this phase of USFWSprobably was under less pres- ing. grazing, on the other the analysis. Species added to the sure to produce detailed justifica- hand, is particularly harmful to endangeredlist prior to 1980 (238 tions for its listingdecisions. We do plants, affecting 33% of endangered species)tended to havefewer threats not knowhow thispattern may have plant species compared to 14% of delineatedin the listingnotices than influencedour results. Also, as noted endangered animals; the difference specieslisted in lateryears. Although in our coarse-scaleanalysis, assess- is highly significant (chi square = there be a basis for ments of the threats to individual may biological 51.95, d.f. = 1, P<<0.001). this difference,we stronglysuspect speciesare often based on thesubjec- Finally, 168 species (14%) are that it reflectsthe less controversial tive opinionsof knowledgeableindi- threatened by disruption of re- natureof endangeredspecies protec- viduals,rather than experimental evi- gimes in the in which tion at that time. Before 1980, denceor quantitativedata. they live. Of these, 85 (7%) are

610 BioScience Vol. 48 No. 8 threatenedby fire suppressionand 83 Figure 4. Taxonomic (7%) are threatenedby controlled or breakdownof the spe- Mammals uncontrolledfires. cies,subspecies, and pop- 67 Birds ulationsused in thefine- scaleanalysis. The 1207 Reptiles Comparisonswith species,subspecies, and 39 otherstudies populations include Amphibians those that are listed as 16 Flatheret al. (1994, 1998) catalogued endangered or threat- the threats to US endangeredspecies ened underthe Endan- Fish 116 based on information from the Fed- geredSpecies Act or are eralRegister, the USFWSEndangered proposedfor listing. Species Technical Bulletin, recovery plans for individual species, federal widespreadthreat to agency reports, and consultations imperiledfishes, af- Plants with USFWS biologists and state fecting 96% of the invertebrates Natural HeritageProgram scientists. species(versus 94% 155 Their analysis covered 667 species, in our study;Table subspecies, and populations pro- 2). Next in signifi- tected by the ESAas of August 1992; cancewas an amal- it did not include species proposed gamated category of hybridization, that the three leading threats to for listing. alien species, predation, and compe- aquatic species nationwide were ag- Althoughthe way in which Flather tition, which affected 39% of the ricultural nonpoint pollution (e.g., et al. categorized threats was not fish species (versus our tally of 53% siltation and nutrient inputs), alien identical to our approach, the major for alien species, which probably species, and altered hydrologic re- findings from the two studies can covers most of the same threats). gimes due to dams and impound- still be compared.These authors also Finally, Williams et al. found that ments. This conclusion is consistent identified habitat loss and alien spe- overharvestand disease affected 4% with our findings from the fine-scale cies as the two most widespread and 2%, respectively, of the fishes analysis, which identified pollution threats to endangered species, af- (versus 13% and 1% in our study). and impoundments(including dams) fecting more than 95% and 35% of Richter et al. (1997) concluded as significantthreats to fish and mus- listed species, respectively. (Compa- rable figuresfrom our study are 85% for habitat destruction and 49% for of alien species.) The smaller percent- The major categories habitat age of species affected by exotics in destructionused in this Flather et al.'s study probably re- analysis flects the large number of Hawaiian * Agriculture(including agricultural practices, land conversion and water species that were included in our diversion for agriculture,pesticides and ;excluding livestock study but were not on the endan- grazing) gered species list at the time Flather * Livestockgrazing (includingrange managementactivities) et al. conducted theirs. Flather et al. * Mining, oil and gas, and geothermal exploration and development (1998) also point out that the rela- (including roads constructed for and pollutants generated by these tive frequency of particular threats activities) to species varies geographically. * Logging (includingimpacts of logging roads and forest management Two previousstudies have focused practices) on threats to aquatic species. Will- * Infrastructuredevelopment (including bridges, dredging for navigation, iams et al. (1989) catalogued threats and road constructionand maintenance) to 364 species and subspecies of im- * Road constructionand maintenancespecifically (including logging and periled fish from Canada, the United miningroads) States, and Mexico; Richter et al. * Militaryactivities (1997) surveyed aquatic biologists * Outdoorrecreation (including swimming, hiking, skiing, camping,and to identify the threats to 135 imper- off-road vehicles) iled freshwater fishes, crayfishes, * Off-roadvehicles specifically dragonflies and damselflies, mussels, * Water development (including diversion for agriculture, livestock, and amphibians in the United States. residentialuse, industry,and irrigation;dams, reservoirs, impoundments, Narrowing the scope of Williams et and other barriers to water flow; control; drainage projects; al. to imperiled US and Canadian ;navigational access and maintenance) fishes (254 species), we can compare * Dams, impoundments,and other barriersto water flow specifically their results with ours. The findings * Pollutants(including siltation and mining pollutants) of the two studies are similar: Will- * Landconversion for urbanand commercial development iams et al. identified habitat destruc- * Disruptionof fire (includingfire suppression) tion and degradation as the most

August 1998 611 sels (Table 4). Our coarse-scale analy- temporary threats, and they listed (1994) undoubtedly stem from our sis, which included a larger pool of only one (i.e., the primary) threat practice of tallying multiple threats imperiled species than the fine-scale per species, although they acknowl- per species. Perhaps the most notice- analysis, also highlighted the impor- edged that most species experience able difference between the two stud- tance of alien species as a threat to more than one threat. The top six ies lies in their assessments of the US fish. threats in their study (in terms of importance of alien species as a threat Richter et al. (1997) point out frequency of appearance) were de- to rare plants. Schemske et al. (1994) that there are important geographic velopment (affecting 20.4% of the considered alien species the primary differences in the nature of the threats species); grazing (10.2%); collecting threat to only 6.1% of the plants facing aquatic species. Aquatic spe- (10.2%); water control (8.2%); oil, they studied, whereas we found that cies in the eastern United States are gas, and mining (8.2%); and tram- 57% of endangered plants were af- experiencing particular harm from pling (8.2%). By contrast, our coarse- fected by alien species. Their lower agricultural nonpoint pollution; in scale analysis identified habitat de- percentage stems in part from the the West, the dominant threat is alien struction and alien species as the two small number of Hawaiian plants species, followed by habitat degra- most widespread threats to imper- that had been listed as endangered or dation and altered hydrologic re- iled plants, affecting 81% and 57% threatened at the time of their study. gimes. Richter et al. attribute these of species, respectively. Moreover, Our results do indicate that alien differences to differences in both land in our fine-scale analysis of habitat species are a frequent threat to con- use patterns in the East versus the destruction, the top five threats to tinental plants as well (Table 3), but West and in the ecological sensitivities imperiled plants protected under the they are not necessarily the primary of eastern versus western species. ESA were land conversion (i.e., de- threat, which may account for the Using information from USFWS velopment; 36%), agriculture (33%), remainder of the difference. recovery plans, Schemske et al. grazing (33%), outdoor recreation Collar et al. (1994) identified the (1994) identified the primary cause (33%), and disruption of fire ecol- primary threat to each of 1111 of endangerment for each of 98 US ogy (20%). species they regarded as imperiled. plant species protected under the The consistently higher perc6nt- Because they evaluated endangered ESA. These authors did not distin- ages for all threats in our study com- birds worldwide, focused on primary guish between historical and con- pared to that of Schemske et al. threats only, and categorized the

Table 4. Percentagesof federal endangered, threatened, or proposed species, subspecies, or populations that are harmed by various types of habitat destruction and degradation. Categories are nonexclusive and therefore do not sum to 100.

Verte- Inverte- Amphi- Arach- Crusta- Overall brates brates Plants MammalsBirds Reptiles bians Fish nids ceans Mollusks Mussels Cause (n= 1207) (n = 329) (n = 155) (n=723) (n=67) (n=91) (n=39) (n= 16) (n= 116) (n=39) (n=4) (n=20) (n=23) (n=69)

Agriculture 38 40 57 33 25 42 33 63 45 56 75 55 35 64 Livestock 22 17 10 33 19 20 8 19 16 15 0 30 9 1 grazing Mining, oil 11 12 31 11 2 3 13 13 23 10 0 0 17 58 and gas, geothermal Logging 12 16 25 7 12 18 13 19 19 5 25 5 13 46 Infrastructure 17 16 12 20 9 8 28 38 17 23 25 10 9 6 development (including roads) Road con- 15 15 10 17 8 8 23 38 16 18 25 5 9 6 struction and maintenance Military 4 2 1 5 2 3 5 0 0 0 0 5 4 0 activities Outdoor 27 16 19 33 18 15 31 25 9 41 0 30 26 4 recreation (including ORVs) Off-road 13 6 12 16 6 7 13 13 1 31 0 25 4 0 vehicles Water 30 47 66 15 10 22 28 63 91 21 0 70 48 99 development (including dams, etc.) Dams, im- 17 28 54 5 3 9 15 13 64 15 0 15 35 96 poundments, and other barriers Pollutants 20 27 66 7 5 10 21 25 55 26 75 55 48 97 Land con- 35 30 42 36 31 33 56 44 16 67 75 65 13 29 version for commercial development Disruption of 14 5 6 20 7 8 5 6 0 18 25 0 4 0

612 BioScience Vol. 48 No. 8 threatsdifferently than we did, their probably an underestimate). What is were accidentally introduced into the results are not directly comparable indisputably clear, however, is that United States in this dry ballast, in- to ours. Nonetheless, it is worth not- the cumulative number of alien spe- cluding fire (Solenopsis invicta ing that both studies identified habi- cies in the United States has skyrock- and Solenopsis richteri) and purple tat loss as the most widespread threat. eted since the late 18th century (Sailer loosestrife (). To- In Collar et al.'s study, the next most 1978, OTA 1993); this pattern holds day, ships use water for ballast in- important threats, in order of de- for all types of species, from plants, stead of dry material, thus ending creasing frequency, were small range to insects, to vertebrates. Given that the spread of alien species via dry or , overhunting, and alien the cumulative number of alien spe- ballast. However, the release of bal- species. In our study, the next most cies is increasing over time, one may last water into US waterways has important threats, also in order of confidently predict that alien species been implicated in the introduction decreasing frequency, were alien spe- will pose an ever-increasing threat to of at least eight alien species since cies, disease,overhunting, and pol- native and . 1980, including the , lution. The higher rankings accorded A somewhat more complicated Dreissena polymorpha (OTA 1993). alien species and diseases in our question is whether the rate of alien Finally, the public's growing infatu- analysis are probably due to the introductions has increased over ation with ornamental plants, tropi- Hawaiian avifauna, which consti- time, which would indicate a rapidly cal fish, and tropical birds has led to tutes a large fraction of endangered worsening situation for imperiled numerous unintentional releases of birds in the United States and is pro- species. The data from published alien species, including over 300 plants foundly affected by these threats. In studies are ambiguous on this point. in California alone (McClintock our study, we did not classify small Reviewing the numbers of alien ter- 1985). range per se as a threat. restrial vertebrates, fishes, mollusks, Looking ahead, as the human and plant pathogens added to the population of the United States con- Changesin threatsover time United States per decade over the tinues to grow, one might predict an past 50 years, OTA (1993) found no increase in the frequency of bio- As human activities and customs consistent increase for any of the diversity threats associated with ur- change over time, one would expect groups. The greatest numbers of ter- banization, such as infrastructure to see corresponding changes in the restrial vertebrates and fishes were development, water development, threats to biodiversity. Because our added during the 1950s and 1960s, and land conversion. Comparable study does not distinguish between whereas the 1970s saw the greatest increases in the proportion of spe- historical and contemporary threats, increase in the numbers of mollusks cies affected by agriculture are also a it is not well suited to test this hy- and plant pathogens. On the other possibility. There is, in fact, good pothesis. For example, the relatively hand, a detailed study of alien species reason to suspect that a growing large percentage of species affected in the San Francisco shows human population in the United by overexploitation (17%) includes that there have been more introduc- States will disproportionately affect a variety of animals that were once tions in recent years than in earlier this nation's imperiled species. Dob- hunted but are now reasonably well years (Cohen and Carlton 1995). son et al. (1997) have shown that protected from this threat (e.g., the Many factors influence the rate at most endangered species in the United whooping [Grus americana] which alien species are introduced States are clustered in a relatively and the [Gymno- into the United States, so the lack of small number of areas, particularly gyps californianus]). Similarly, pes- a consistent increase in that rate over in Hawaii, Southern California, and ticide pollution is listed as the pri- time should not be surprising. Spe- Florida. The human populations in mary threat to the bald cies can be brought into the country all three states are projected to in- (Haliaeetus leucocephalus) and to and released intentionally, or their crease at rates well beyond the na- North American populations of the release can occur as an unintentional tional average. Thus, whereas the (Falco peregrinus), byproduct of cultivation, commerce, population of the United States as a but the primary pollutant harming tourism, or travel. Each new devel- whole is expected to grow by 14% both species-DDT-has been opment in the field of transportation between 1995 an4 2010, the popula- banned in the United States since creates new opportunities for the tions of Hawaii, California, and 1972 (although it continues to be transport of alien species, from the Florida are projected to increase by used in other countries where per- first sailing ships to reach US shores, 27%, 27%, and 22%, respectively egrines spend the winter). Thus, our to the building of the nation's road (US Bureau of the Census 1995). study may overestimate the number and highway system, to the advent of Although change was not of animals that are currently harmed jet airplanes. As transporation tech- listed as a current threat to any spe- by overexploitation and pollutants. nology changes, so do the opportu- cies in our databases, it is almost There are no accurate figures on nities for alien stowaways. Empty certain to become one in the foresee- the total number of alien species now cargo ships arriving in the United able future due to increasing concen- established in the United States, al- States, for example, used to carry trations of greenhouse gases from and though the Office of Technology dry ballast in the form of rocks and -fuel use, land-use changes, Assessment [OTA] (1993) has esti- , which was then off-loaded agriculture. Climate models devel- mated that there are at least 4500 (a around wharves to provide cargo oped by the Intergovernmental Panel number that OTA acknowledges is space. Numerous insects and plants on predict a 0.9-

August 1998 613 3.5 'C increase in global mean tem- gists have long suspected: Habitat program of prescribed burning, or perature over the course of the next loss is the single greatest threat to do any of the other things that may century (Houghton et al. 1995). That biodiversity, followed by the spread be absolutely necessary for the long- increase will cause a rise in levels of alien species. However, the dis- term survival of many imperiled spe- of 15-95 cm and significant changes covery that nearly half of the imper- cies. In fact, it may be possible for a in the frequencies of severe iled species in the United States are landowner to rid himself of an en- and . threatened by alien species-com- dangered species "problem" by liter- These changes are likely to affect bined with the growing numbers of ally doing nothing and waiting until a broad array of imperiled species. alien species-suggests that this par- the habitat is no longer suitable for For example, Morse et al. (1993) ticular threat may be far more seri- the species in question. Even those estimate that 7-11% of North ous than many people have hereto- landowners who care deeply about America's species fore believed. The impact of alien endangered species and wish to pro- would no longer encounter a suit- species is most acute in the Hawaiian tect them face a daunting burden: able climatic regime ("climate enve- , as demonstrated by the fact The costs of undertaking these man- lope") within their present ranges in that nearly 100% of the archipelago's agement actions can be considerable the event of a 3 'C increase in tem- imperiled plants and birds are threat- and, at present, are usually not tax perature. Due to their small ranges ened by alien species, compared with deductible. and weak dispersal abilities, imper- 30% and 48%, respectively, for With a growing list of species in iled plants would be disproportion- mainland plants and birds (Table 3). need of attention and less money to ately affected. Morse et al. (1993) This finding is also consistent with spend per species (Wilcove et al. also estimate that 10-18% of North numerous other studies that have 1996), the USFWS cannot hope to America's rare plants could be ex- highlighted the unique vulnerability cover the necessary management cluded from their climate envelope of communities to alien species costs for most of the plants and ani- due to climate change. (Culliney 1988, Simberloff 1995). mals it aspires to protect. Nor can it In another well-publicized study, Pollution (including siltation) count on the goodwill of landowners Britten et al. (1994) noted that ranks well below alien species as a to contribute their own money or relictual populations of the critically threat to imperiled species in gen- labor for actions they are not obli- endangered Uncompahgre fritillary eral, but it exceeds alien species as a gated to perform and that ultimately butterfly (Boloria acrocnema) living threat to aquatic taxa. As Richter et may result in restrictions on the use atop a few peaks in the San Juan al. (1997) point out, the pollutants of their property. As a nation, there- Mountains of southwestern Colo- affecting the largest number of fore, we are incurring a growing rado were extremely vulnerable to aquatic species are agricultural pol- "management debt" associated with unusual weather events. They fur- lutants, such as silt and nutrients, efforts to protect imperiled species. ther hypothesized that a regional that enter and as runoff To address this problem, it will be warming trend (as might occur due from farming operations. These necessary to supplement the regula- to global climate change) could elimi- nonpoint source pollutants have tory controls of the ESA and other nate all of the butterfly's habitat, proved to be exceedingly difficult to wildlife protection laws with a wide essentially pushing it off of the moun- regulate and control (Young and array of incentives to reward land- tains and into . Indirect Congdon 1994). owners who wish to manage their support for this hypothesis comes Finally, this study and one by property to benefit endangered spe- from a recent study of another but- Wilcove and Chen (in press) raise cies (Wilcove et al. 1996). Without terfly. Parmesan (1996) censused troubling questions about the future such incentives, the United States populations of the Edith's checker- of imperiled species in the United stands to lose a large proportion of spot (Euphydryas editha) through- States. Both studies found that a high its imperiled plants and animals. out its known range (Baja Califor- proportion of imperiled species is nia, the western United States, and threatened by either fire suppression Acknowledgments western Canada) and found signifi- within their fire-maintained cant latitudinal and altitudinal dif- or alien species. Both types of threats This paper is part of ongoing col- ferences in the proportion of popula- must be addressed through active, laboration between The Nature Con- tions (in suitable habitat) that had "hands-on" management of the habi- servancy, the Association for Bio- become extinct. Populations in tat, such as pulling up alien plants diversity Information, and the Mexico were four times more likely and trapping alien animals or using Network of State Natural Heritage to have vanished than those in prescribed fire to regenerate early Programs as part of the forthcoming Canada, a North-South gradient in successional habitats. Although the book Precious Heritage. We thank survival that is consistent with the ESA prohibits actions that directly the Natural Heritage Programs for predicted impacts of global warming harm listed animals and, to a lesser sharing their data, without which on species' ranges. extent, listed plants, it does not re- this study would not have been pos- quire landowners to take affirmative sible. We also thank Lynn Kutner for the data. Conservation actions to maintain or restore habi- her assistance in analyzing implications tats for listed species. Thus, a land- Valuable information and ideas on The major findings of this study con- owner is under no obligation to con- threats to imperiled species were sup- firm what most conservation biolo- trol exotic , undertake a plied by Larry Master, Geoff Ham-

614 BioScience Vol. 48 No. 8 merson, Bruce Stein, David Braun, dangeredSpecies Act. BioScience46: 355- ParmesanC. 1996. Climateand species'range. and Brian Richter, for which we are 363. Nature 382: 765-766. We thank EhrlichPR. 1988. The loss of diversity:Causes RichterBD, BraunDP, MendelsonMA, Master grateful. Jane Lubchenco, and consequences.Pages 21-27 in Wilson LL. 1997. Threatsto imperiledfreshwater Margaret McMillan, and three EO, ed. Biodiversity.Washington (DC): fauna.Conservation 11:1081-1093. anonymous reviewers for their help- National AcademyPress. Sailer RI. 1978. Our immigrantinsect fauna. ful reviews of earlier versions of this FlatherCH, Joyce LA, Bloomgarden CA. 1994. EntomologicalSociety of AmericaBulletin E. Losos and A. would Speciesendangerment patterns in the United 24: 3-11. paper. Phillips States.Ft. Collins (CO):US Departmentof SchemskeDW, HusbandBC, Ruckelshaus MH, like to thank The WildernessSociety AgricultureForest Service, Rocky Moun- GoodwillieC, ParkerIM, BishopJG. 1994. for its support of their work on this tain Forestand RangeExperiment Station. Evaluatingapproaches to the conservation project. GeneralTechnical Report no. RM-241. of rareand endangeredplants. Ecology75: FlatherCH, Knowles MS, KendallIA. 1998. 584-606. Threatenedand endangeredspecies geogra- SimberloffD. 1995. Whydo introducedspecies Referencescited phy. BioScience48: 365-376. appearto devastateislands more than main- HoughtonJT, Meira Filho LG, CallanderBA, land areas?Pacific 49: 87-97. BrittenHB, BrussardPF, Murphy DD. 1994. Harris N, KattenbergA, Maskell K, eds. US Bureauof the Census. 1995. Statisticalab- The pendingextinction of the Uncompah- 1995. Climate Change 1995. Cambridge stractof the UnitedStates: 1995. Washing- gre fritillarybutterfly. Conservation Biol- (UK):Cambridge University Press. ton (DC):US GovernmentPrinting Office. ogy 8: 86-94. MasterLL 1991. Assessingthreats and setting Wilcove DS, Chen LY. In press. Management CohenAN, CarltonJT. 1995. Nonindigenous priorities for conservation. Conservation costs for endangeredspecies. Conservation aquaticspecies in a UnitedStates estuary: A Biology 5: 559-563. Biology. case studyof the biologicalinvasions of the McClintockE. 1985. Escapedexotic weeds in WilcoveDS, BeanMJ, BonnieR, McMillanM. San FranciscoBay and Delta. Washington California.Freemontia 12(4): 3-6. 1996. Rebuildingthe Ark:Toward a More (DC): US Fish and Wildlife Service and MorseLE, Kutner LS, Maddox GD, KarteszJT, Effective EndangeredSpecies Act for Pri- Groton(CT): The National Sea GrantCol- HoneyLL, Thurman CM, ChaplinSJ. 1993. vate Land.Washington (DC): Environmen- lege Program,Connecticut Sea Grant. The potential effects of climate change on tal Defense Fund. CollarNJ, CrosbyMJ, StattersfieldAJ. 1994. the nativevascular flora of : Williams JE, Johnson JE, Hendrickson DA, Birdsto Watch2.The WorldList of Threat- A preliminaryclimate envelopes analysis. Contreras-Balderas S, Williams JD, enedBirds. Cambridge (UK): BirdLife Inter- Palo Alto (CA): Electric Power Research Navarro-MendozaM, McAllisterDE, Dea- national. Institute.Report no. EPRITR-103330. con JE. 1989. Fishes of North America CullineyJL. 1988. Islandsin a FarSea: Nature Noss RF, Cooperrider AY. 1994. Saving endangered,threatened, or of special con- and Man in Hawaii. San Francisco:Sierra Nature's Legacy:Protecting and Restoring cern: 1989. Fisheries14(6): 2-20. Club Books. Biodiversity.Washington (DC): Island Press. Wilson EO. 1992. The Diversityof . Cam- Dobson AP, Rodriguez JP, Roberts WM, [OTA]Office of TechnologyAssessment. 1993. bridge(MA): Belknap Press. WilcoveDS. 1997. Geographicdistribution Harmful non-indigenous species in the Young TF, CongdonCH. 1994. PlowingNew of endangeredspecies in the United States. United States.Washington (DC): US Gov- Ground:Using Economic Incentives to Con- Science 275: 550-553. ernment Printing Office. Publication no. trolWater Pollution from Agriculture. Oak- Easter-PilcherA. 1996. Implementingthe En- OTA-F-565. land (CA):Environmental Defense Fund. AMAS

t.S . . . , . . . :..

...,...... ,=:,..=.:•..-.•....,,,.....,

Pleasejoin usfor the SACNAS Silver Anniversary Conference!

Washington,D.C., October8-11, 1998

p it Scheduled Events Include: S " Two Nobel Laureate presentations 1997 ,- IobemIppztrIt..a.Smalley , t s " Workshops & symposia S e jw at National. 1wn'll Makepeace pr~esellftIon n -C " The K-12 Teacher ers Dr.Richard Workshops .\(:NAS :seNA winerBrendi " Exhibits& student presentations

Call 408-459-0170 to join our growing membership, or email us at: [email protected].

To learn more about SACNAS, please visit our website ot:wwwosacnas*org August 1998 615