High Quality
Open the downloaded document, and select print from the file menu (PDF reader required).
Articles
www.biosciencemag.org February 2011 / Vol. 61 No. 2
• BioScience 107
Oyster Reefs at Risk andRecommendations for Conservation,Restoration, and Management
Michael W. Beck, RoBeRt D. BRuMBaugh, lauRa aiRolDi, alvaR caRRanza, loRen D. coen, chRistinecRaWfoRD, oMaR Defeo, gRahaM J. eDgaR, Boze hancock, MattheW c. kay, hunteR s. lenihan, MaRkW. luckenBach, caitlyn l. toRopova, guofan zhang, anD XiMing guo
Native oyster rees once dominated many estuaries, ecologically and economically. Centuries o resource extraction exacerbated by coastal degra-dation have pushed oyster rees to the brink o unctional extinction worldwide. We examined the condition o oyster rees across 144 bays and 44 ecoregions; our comparisons o past with present abundances indicate that more than 90% o them have been lost in bays (70%) and ecoregions (63%). In many bays, more than 99% o oyster rees have been lost and are unctionally extinct. Overall, we estimate that 85% o oyster rees have been lost globally. Most o the world’s remaining wild capture o native oysters (> 75%) comes rom just ive ecoregions in North America, yet the condition o rees in these ecoregions is poor at best, except in the Gul o Mexico. We identiy many cost-eective solutions or conservation,restoration, and the management o isheries and nonnative species that could reverse these oyster losses and restore ree ecosystem services.Keywords: shellish, oyster ree, marine conservation, isheries, habitat restoration
marine ecosystems, mainly o those that are intertidal orthat exist in clear water and can be aerially assessed. Recordso the abundance and catch o oysters and the distributiono the ecosystems that they create can span centuries andmillennia, though usually not as continuous data sets.Thecondition o oyster ecosystems has been considered in partby others (e.g., Jackson et al. 2001, Kirby 2004, NRC 2004,Ruesink et al. 2005, Lotze et al. 2006, Airoldi and Beck 2007),but these estimates o condition have used data rom only a limited number o bays. To expand on these eorts wesynthesize quantitative data on the condition o oyster reesin more than 140 bays, provide an overall estimate o oysterree condition, and use this extensive inormation to identiy areas and opportunities to improve the condition o oysterrees at a global scale.
Assessing condition
We identied native oyster ree condition primarily as aunction o oyster abundance; we calculated condition usingestimates o past and present abundances rom the litera-ture. Measures o total ree area and size were occasionally available. Fishery statistics or native oysters were the mostcommonly available inormation or assessing the changesin oyster abundance and the condition o rees, but land-ings data were rarely the only inormation used to assesscondition. Ultimately, ree size is a unction o the num-ber o living oysters, and larger rees positively infuenceoyster growth and survival (Lenihan and Peterson 1998,
O
yster rees and beds (hereater rees) were once a
dominant structural and ecological component o estuaries around the globe, ueling coastal economies orcenturies. Oysters are ecosystem engineers; one or a ewspecies produce ree habitat or entire ecosystems (Lenihanand Peterson 1998). They have supported civilizations ormillennia, rom Romans to Caliornia railroad workers(Mac-Kenzie et al. 1997a, 1997b). In 1864, 700 millionEuropean fat oysters (
Ostrea edulis
) were consumed inLondon, and nearly 120,000 workers were employed as oysterdredgers in Britain. Shell piles rom historical harvests inthe southwest o France contain more than 1 trillion shellsapiece, underscoring both the productivity o the species andthe scale o harvest (MacKenzie et al. 1997b). In the 1870s,intertidal rees o the eastern oyster
Crassostrea virginica
extended or miles along the main axis o the James Riverin the Chesapeake Bay; by the 1940s, these rees had largely disappeared (Woods et al. 2005). In many coastal areas,including the Texas coast, roads were paved with oyster shells(Doran 1965).Oyster rees are one o the ew marine ecosystems orwhich direct estimates o condition can be calculated,because most underlying ree structures are created by justone or a ew oyster species. Most estimates o the conditiono marine ecosystems are indirect and are derived rom thedistribution o threats such as trawling, sedimentation, andpollution (e.g., Halpern et al. 2008). There have been only a ew direct assessments o the condition o coastal and
BioScience
61: 107–116. ISSN 0006-3568, electronic ISSN 1525-3244. © 2011 by American Institute o Biological Sciences. All rights reserved. Requestpermission to photocopy or reproduce article content at the University o Caliornia Press’s Rights and Permissions Web site at
www.ucpressjournals.com/ reprintino.asp
. doi:10.1525/bio.2011.61.2.5
108 BioScience •
February 2011 / Vol. 61 No. 2
www.biosciencemag.org
Articles
Lenihan 1999). Water quality also aects oysters but is lessreliable as an indicator o oyster condition, as oysters canthrive well past the point at which human health concernsbecome an issue.We considered the condition o native oyster rees at twodierent spatial scales: bays and ecoregions. For our pur-poses, “bays” reers to bays, estuaries, embayments, coastalcounties, and portions o coastlines (e.g., Mobile Bay, Wad-den Sea, Venice Lagoon). “Bays” was the most consistently used term to describe this common ecological unit inreports o oyster science, management, and conservation.Ecoregions are regional, biogeographic units with coherencein their species and ecosystems; we used the boundariesidentied by Spalding and colleagues (2007).We established our categories o condition on the basiso comparisons o current with historical oyster abundanceindicators (e.g., surveys, landings, catch per unit eort)or aerially measured ree extents: (a) less than 50% lost(good), (b) 50% to 89 % lost (air), (c) 90% to 99% lost(poor), and (d) more than 99% lost (unctionally extinct).We looked at records rom between 20 and 130 years beorepresent to estimate historical abundances and extents. Webased the date range on the availability and reliability o the data. Interestingly, surveys rom a century ago wererequently better than records rom decades ago or eventhe present.We used practical and conservative rules or assigningcondition. When sources indicated that it was dicult tond rees, or that no rees remained in bays where annualcatch records were high (usually > 10,000 metric tons) buthistorical observations indicated that rees had once beenextensive, we estimated that more than 99% o the habitatwas lost, and classied the condition as unctionally extinct(Jackson 2001). Such was the case, or example, in theWadden Sea (European Union), Narragansett Bay (UnitedStates), Southport (Australia), and Ciénaga Grande deSanta Marta (Colombia). The condition was classied aspoor (90% to 99% habitat lost) when evidence indicatedthat sheries were collapsing (or collapsed) but there wasevidence that rees remained, even i long-term viability was questionable (e.g., Chesapeake Bay [United States],Bohai Bay [China]). There is abundant evidence thatshellsheries continue well past the point at which 90%o the habitat has been lost (MacKenzie et al. 1997b, Kirby 2004). We classied the ree condition as air (50% to 89% habitat lost) when abundance indicators were below50% o historical gures or records indicated greater than50% loss in rees and there was evidence o signicantremaining rees (e.g., Apalachicola Bay [United States],Golo San Matías [Argentina]). We considered the con-dition good (< 50% lost) i sheries were only lightly tomoderately exploited (or not at all exploited) and i many areas o rees remained relative to earlier abundances (e.g.,Mobile Bay [United States], Nootka Sound [Canada]).I there was any question, we discussed the data-drivenestimates with shery and habitat managers and scientistsrom each location. The only common debate in publica-tions was not about the condition o the ecosystems but thecause o the decline (e.g., Kirby 2004, Ogburn et al. 2007).We identied the condition o oyster rees across coastalecoregions using inormation rom multiple bays withinecoregions and national and regional publications on thestatus o oyster populations (e.g., red lists) and sheries (e.g.,MacKenzie et al. 1997b, Gillespie 2009). We identied thecondition o oyster rees in an ecoregion i there were one ormore reerences that characterized regional condition or i the condition was rmly documented in three or more bayswithin the ecoregion. When there were several bays in anecoregion and no other regional sources o status inorma-tion, the condition estimates were averaged or all bays in theecoregion and rounded to the nearest integer.
Oyster ree condition
The overall condition o native oyster rees is poor in mosto the 144 bays in 40 ecoregions we evaluated (gure 1; seesupplementary material table S1 at
http://conserveonline.org/ workspaces/Shellish%20Rees%20At%20Risk/documents/ oyster-rees-at-risk-supplementary-table
). Although indi-vidual oysters are still present in most places, records o historical (past 20 to approximately 130 years) and recentabundances show that many rees that were once commonare now rare or extinct as ecosystems. Oyster rees are atless than 10% o their prior abundance in most bays (70%)and ecoregions (63%). They are unctionally extinct—inthat they lack any signicant ecosystem role and remainat less than 1% o prior abundances in many bays (37%)and ecoregions (28%)—particularly in North America,Australia, and Europe. Very ew bays and ecoregions arerated as being in good condition (> 50% o rees remain-ing). Our results most likely underestimate losses because o the lack o historical abundance records, which particularly aects assessments in South America, temperate Asia, andSouth Arica.Our estimates o ree conditions are conservative because(a) where there was question o status, we applied the higherranking; (b) or most bays and ecoregions it was clear thatabundances were usually at the lower end o their conditionranking; and (c) the estimates were usually based on only part o the historical loss, as rees were probably more abun-dant beore the recording o shery catches began.Overall, we estimate an 85% loss o oyster ree ecosys-tems globally (gure 1). We calculated this by using themidpoint value or each condition category o oyster reeslost in ecoregions (e.g., 95% o habitat lost or ecoregionsin poor condition), and then averaged the loss among allecoregions.Prior records rom many bays indicated that oysterrees were abundant and supported large sheries—up tohundreds o thousands o metric tons o recorded catch—but those rees and sheries are now greatly reduced orgone (MacKenzie et al. 1997a, 1997b, Kirby 2004, NRC2004, Ruesink et al. 2005, Lotze et al. 2006). We ound
www.biosciencemag.org February 2011 / Vol. 61 No. 2
• BioScience 109
Articles
requent evidence that restrictions had been placed onharvests and o concerns about the incidence o diseaseand environmental degradation, but in the great majority o cases, harvests continued until oysters could no longerbe shed commercially (MacKenzie et al. 1997a, 1997b,NRC 2004).The decline o oyster isheries ollows a commonsequence o events in many places globally (MacKenzieet al. 1997a, 1997b, NRC 2004). Typically, the extensiveharvest o wild oyster populations results in the loss o ree structure. Most declines start with the loss o verti-cal relie and complexity, oten as a result o dredgingand trawling, which exacerbates the impact o additionalstresses rom anoxia, sedimentation, disease, and nonna-tive species (Lenihan and Peterson 1998, 2004, Lenihan1999, Lenihan et al. 1999). In many cases, years o declin-ing harvest are ollowed by introductions o nonnativeoysters that are released directly into the wild or thatescape rom nearby aquaculture (Ruesink et al. 2005).Overharvest and disease oten lead to a population crash.Although oyster diseases occur in native populations, inmany places the incidence o disease is associated withtransers o nonnative oysters or aquaculture and romballast waters (NRC 2004). Other anthropogenic actorssuch as alterations o shorelines; changes in reshwaterinlows; and increased loadings o sediments, nutrients,and toxins also contribute to declines (NRC 2004). Thereare ew i any bays where only one stressor has aectedoyster rees.
Wild fsheries and remaining rees
To compare present-day levels o wild oyster harvest amongecoregions, we used global commercial catch data developedby the Sea Around Us Project (Watson et al. 2004). Thesecatch data are primarily based on the national catch statis-tics compiled by the Food and Agriculture Organization,allocated to hal-degree cells o latitude and longitude, andthen summed by ecoregion. To account or annual variationin catches, we used the average catch in metric tons o nativeoysters per ecoregion rom 1995 to 2004. The biomass esti-mates are based on numbers o oysters in their shells.Most o the world’s remaining wild capture o native oysterscomes rom just ve ecoregions on the East and Gul coasts o North America, which together account or more than 75%o the global catch (gure 2). Only 10 ecoregions in the worldreported wild oyster capture rates o more than 1000 metrictons per year rom 1995 to 2004; only six ecoregions haveaverage captures above 5500 metric tons, and ve o theseare in eastern North America (Virginian to Southern Gul o Mexico ecoregions). Although there is catch remaining inthese six ecoregions, it is much lower than in the past. Indeed,in most o the bays (20 o 34) in these six ecoregions, therehas been at least a 90% loss in oyster rees; in some cases theloss has been more than 99%. Thereore, the condition o theoyster rees is poor or unctionally extinct in these bays andregions, but oysters continue to be harvested. Contemporary native oyster catches in the Gul o Mexico are the highest inthe world, despite signicant declines in abundance and reesin numerous bays (e.g., gure 1, table S1).
Figure 1. The global condition o oyster rees in bays and ecoregions. The condition ratings o good, air, poor, and unctionally extinct are based on the percentage o current to historical abundance o oyster rees remaining: less than 50% lost (good), 50% to 89% lost (air), 90% to 99% lost (poor), more than 99% lost (unctionally extinct). Ecoregion boundaries are rom Spalding and colleagues (2007). Not all regions with oysters could be assessed because o a lack o data (see text).
Add a Comment