Endangered and Threatened Species; Identification and Proposed Listing of Eleven Distinct Population Segments of Green Sea Turtles (Chelonia mydas) as Endangered or Threatened and Revision of Current Listings

DEPARTMENT OF THE INTERIOR Fish and Wildlife Service 50 CFR Part 17 DEPARTMENT OF COMMERCE National Oceanic and Atmospheric Administration 50 CFR Parts 223 and 224 [Docket No. 120425024-5022-02] RIN 0648-XB089 Endangered and Threatened Species; Identification and Proposed Listing of Eleven Distinct Population Segments of Green Sea Turtles (Chelonia mydas) as Endangered or Threatened and Revision of Current Listings AGENCY:

National Marine Fisheries Service (NMFS), National Oceanic and Atmospheric Administration (NOAA), Commerce; United States Fish and Wildlife Service (USFWS), Interior.

ACTION:

Proposed rule; 12-month petition finding; request for comments; notice of public hearing.

SUMMARY:

The green sea turtle ( Chelonia mydas; hereafter referred to as the green turtle) is currently listed under the Endangered Species Act (ESA) as a threatened species, with the exception of the Florida and Mexican Pacific coast breeding populations, which are listed as endangered. We, NMFS and USFWS, find that the green turtle is composed of 11 distinct population segments (DPSs) that qualify as “species” for listing under the ESA. We propose to remove the current range-wide listing and, in its place, list eight DPSs as threatened and three as endangered. We also propose to apply existing protective regulations to the DPSs. We solicit comments on these proposed actions.

Although not determinable at this time, designation of critical habitat may be prudent, and we solicit relevant information for those DPSs occurring within U.S. jurisdiction. In the interim, we propose to continue the existing critical habitat designation ( i.e. , waters surrounding Culebra Island, Puerto Rico) in effect for the North Atlantic DPS.

This proposed rule also constitutes the 12-month finding on a petition to reclassify the Hawaiian green turtle population as a DPS and to delist that DPS. Although we find the Hawaiian green turtle population to constitute a DPS (referred to in this proposed rule as the Central North Pacific DPS), we do not find delisting warranted.

A public hearing will be held in Hawai`i. Interested parties may provide oral or written comments at this hearing.

DATES:

Comments and information regarding this proposed rule must be received by close of business on June 22, 2015. A public hearing will be held on April 8, 2015 from 6 to 8 p.m., with an informational open house starting at 5:30 p.m. Requests for additional public hearings must be made in writing and received by May 7, 2015.

ADDRESSES:

You may submit comments on this document, identified by NOAA-NMFS-2012-0154, by the following methods:

• Electronic Submissions: Submit all electronic public comments via the Federal e-Rulemaking Portal.

1. Go to www.regulations.gov/#!docketDetail;D=NOAA-NMFS-2012-0154 .

2. Click the “Comment Now!” icon, complete the required fields.

3. Enter or attach your comments.

OR • Mail: Submit written comments to Green Turtle Proposed Listing Rule, Office of Protected Resources, National Marine Fisheries Service, 1315 East-West Highway, Room 13535, Silver Spring, MD 20910; or Green Turtle Proposed Listing Rule, U.S. Fish and Wildlife Service, North Florida Ecological Services Office, 7915 Baymeadows Way, Suite 200, Jacksonville, FL 32256.

OR • Public hearing: Interested parties may provide oral or written comments at the public hearing to be held at the Japanese Cultural Center, 2454 South Beretania Street, Honolulu, Hawai`i 96826. Parking is available at the Japanese Cultural Center for $5.

Instructions: Comments sent by any other method, to any other address or individual, or received after the end of the comment period, may not be considered by the Services. All comments received are a part of the public record and will generally be posted for public viewing on www.regulations.gov without change. All personal identifying information ( e.g. , name, address, etc.), confidential business information, or otherwise sensitive information submitted voluntarily by the sender will be publicly accessible. The Services will accept anonymous comments (enter “N/A” in the required fields if you wish to remain anonymous). The proposed rule is available electronically at http://www.nmfs.noaa.gov/pr/species/turtles/green.htm and http://www.fws.gov/northflorida/seaturtles/turtle%20factsheets/green-sea-turtle.htm .

FOR FURTHER INFORMATION CONTACT:

Jennifer Schultz, NMFS (ph. 301-427-8443, email jennifer.schultz@noaa.gov ), or Ann Marie Lauritsen, USFWS (ph. 904-731-3032, email annmarie_lauritsen@fws.gov ). Persons who use a Telecommunications Device for the Deaf (TDD) may call the Federal Information Relay Service (FIRS) at 1-800-877-8339, 24 hours a day, and 7 days a week.

SUPPLEMENTARY INFORMATION:

Public Comments Solicited on the Proposed Listing

We intend that any final action resulting from this proposal be as accurate and effective as possible and informed by the best available scientific and commercial information. Therefore, we request comments or information from the public, other concerned governmental agencies, the scientific community, industry, or any other interested party concerning this proposed rule. We are seeking information and comments on whether each of the 11 proposed green turtle DPSs qualify as DPSs, whether listing of each DPS is warranted, and, if so, whether they should be classified as threatened or endangered as described in the “Listing Determinations Under the ESA” section provided below. Specifically, we are soliciting information on the following subjects relative to green turtles within the 11 proposed DPSs: (1) Historical and current population status and trends, (2) historical and current distribution, (3) migratory movements and behavior, (4) genetic population structure, (5) current or planned activities that may adversely affect green turtles, (6) conservation efforts to protect green turtles, and (7) our extinction risk analysis and findings. We request that all data, information, and comments be accompanied by supporting documentation such as maps, bibliographic references, or reprints of pertinent publications. We will consider comments and new information when making final determinations.

Public Comments Solicited on Critical Habitat

Though we are not proposing to designate critical habitat at this time, we request evaluations describing the quality and extent of existing habitats within U.S. jurisdiction for the proposed North Atlantic, South Atlantic (U.S. Virgin Islands), Central South Pacific (American Samoa), Central West Pacific (Commonwealth of the Northern

Section 4(b)(2) of the ESA requires the Secretary to consider the “economic impact, impact on national security, and any other relevant impact” of designating a particular area as critical habitat. Section 4(b)(2) also authorizes the Secretary to conduct a balancing of the benefits of inclusion and the benefits of exclusion from a critical habitat designation of a particular area, and to exclude any particular area where the Secretary finds that the benefits of exclusion outweigh the benefits of designation, unless excluding that area will result in extinction of the species. Therefore, for features and areas potentially qualifying as critical habitat, we also request information describing: (1) Activities or other threats to the essential features that could be affected by designating them as critical habitat (pursuant to section 4(b)(8) of the ESA); and (2) the positive and negative economic, national security and other relevant impacts, including benefits to the recovery of the species, likely to result if these areas are designated as critical habitat. We also seek information regarding the conservation benefits of designating areas within nesting beaches and waters under U.S. jurisdiction as critical habitat. Data sought include, but are not limited to the following: (1) Scientific or commercial publications, (2) administrative reports, maps or other graphic materials, and (3) information from experts or other interested parties. Comments and data particularly are sought concerning the following: (1) Maps and specific information describing the amount, distribution, and type of use ( e.g. , foraging or migration) by green turtles, as well as any additional information on occupied and unoccupied habitat areas; (2) the reasons why any habitat should or should not be determined to be critical habitat as provided by sections 3(5)(A) and 4(b)(2) of the ESA; (3) information regarding the benefits of designating particular areas as critical habitat; (4) current or planned activities in the areas that might be proposed for designation and their possible impacts; (5) any foreseeable economic or other potential impacts resulting from designation, and in particular any impacts on small entities; and (6) whether specific unoccupied areas may be essential to provide additional habitat areas for the conservation of the proposed DPSs. We seek information regarding critical habitat for the proposed green turtle DPSs as soon as possible, but no later than June 22, 2015.

Public Hearings

The Services will hold a public hearing in Hawai‘i. Interested parties may provide oral or written comments at this hearing. A public hearing will be held on April 8, 2015 from 6 to 8 p.m., with an informational open house starting at 5:30 p.m., at the Japanese Cultural Center, 2454 South Beretania Street, Honolulu, Hawai‘i 96826. Parking is available at the Japanese Cultural Center for $5. If requested by the public by May 7, 2015, additional hearings will be held regarding the proposed listing of the green turtle DPSs. If additional hearings are requested, details regarding location(s), date(s), and time(s) will be published in a forthcoming Federal Register notice.

References

A complete list of all references cited herein is available upon request (see FOR FURTHER INFORMATION CONTACT ).

Table of Contents

I. Background II. Policies for Delineating Species Under the ESA III. Listing Determinations Under the ESA IV. Biology and Life History of Green Turtles V. Overview of the Policies and Process Used To Identify DPSs A. Discreteness Determination 1. Atlantic Ocean/Mediterranean Sea 2. Indian Ocean 3. Pacific Ocean B. Significance Determination 1. North Atlantic 2. Mediterranean 3. South Atlantic 4. Southwest Indian 5. North Indian 6. East Indian-West Pacific 7. Central West Pacific 8. Southwest Pacific 9. Central South Pacific 10. Central North Pacific 11. East Pacific C. Summary of Discreteness and Significance Determinations VI. Listing Evaluation Process A. Discussion of Population Parameters for the Eleven Green Turtle DPSs B. Summary of Factors Affecting the Eleven Green Turtle DPSs C. Conservation Efforts D. Extinction Risk Assessments and Findings VII. North Atlantic DPS A. Discussion of Population Parameters for the North Atlantic DPS B. Summary of Factors Affecting the North Atlantic DPS 1. Factor A: The Present or Threatened Destruction, Modification, or Curtailment of Its Habitat or Range a. Terrestrial Zone b. Neritic/Oceanic Zones 2. Factor B: Overutilization for Commercial, Recreational, Scientific, or Educational Purposes 3. Factor C: Disease or Predation 4. Factor D: Inadequacy of Existing Regulatory Mechanisms 5. Factor E: Other Natural or Manmade Factors Affecting Its Continued Existence a. Incidental Bycatch in Fishing Gear i. Gill Net and Trawl Fisheries ii. Dredge Fishing b. Channel Dredging c. Vessel Strikes and Boat Traffic d. Effects of Climate Change and Natural Disasters e. Effects of Cold Stunning f. Contaminants and Marine Debris C. Conservation Efforts for the North Atlantic DPS D. Extinction Risk Assessment and Findings for the North Atlantic DPS VIII. Mediterranean DPS A. Discussion of Population Parameters for the Mediterranean DPS B. Summary of Factors Affecting the Mediterranean DPS 1. Factor A: The Present or Threatened Destruction, Modification, or Curtailment of Its Habitat or Range a. Terrestrial Zone b. Neritic/Oceanic Zones 2. Factor B: Overutilization for Commercial, Recreational, Scientific, or Educational Purposes 3. Factor C: Disease or Predation 4. Factor D: Inadequacy of Existing Regulatory Mechanisms 5. Factor E: Other Natural or Manmade Factors Affecting Its Continued Existence a. Incidental Bycatch in Fishing Gear i. Longline Fisheries ii. Set Net (Gill Net) Fishing iii. Trawl Fisheries b. Vessel Strikes and Boat Traffic c. Pollution d. Effects of Climate Change C. Conservation Efforts D. Extinction Risk Assessment and Findings IX. South Atlantic DPS A. Discussion of Population Parameters for the South Atlantic DPS B. Summary of Factors Affecting the South Atlantic DPS 1. Factor A: The Present or Threatened Destruction, Modification, or Curtailment of Its Habitat or Range a. Terrestrial Zone b. Neritic/Oceanic Zones 2. Factor B: Overutilization for Commercial, Recreational, Scientific, or Educational Purposes 3. Factor C: Disease or Predation 4. Factor D: Inadequacy of Existing Regulatory Mechanisms 5. Factor E: Other Natural or Manmade Factors Affecting Its Continued Existence a. Incidental Bycatch in Fishing Gear b. Marine Debris and Pollution c. Effects of Climate Change C. Conservation Efforts for the South Atlantic DPS D. Extinction Risk Assessment and Findings for the South Atlantic DPS X. Southwest Indian DPS A. Discussion of Population Parameters for the Southwest Indian DPS B. Summary of Factors Affecting the Southwest Indian DPS 1. Factor A: The Present or Threatened Destruction, Modification, or Curtailment of Its Habitat or Range a. Terrestrial Zone b. Neritic/Oceanic Zones 2. Factor B: Overutilization for Commercial, Recreational, Scientific, or Educational Purposes 3. Factor C: Disease or Predation 4. Factor D: Inadequacy of Existing Regulatory Mechanisms 5. Factor E: Other Natural or Manmade Factors Affecting Its Continued Existence a. Incidental Bycatch in Fishing Gear b. Effects of Climate Change and Natural Disasters C. Conservation Efforts for the Southwest Indian DPS D. Extinction Risk Assessment and Findings for the Southwest Indian DPS XI. North Indian DPS A. Discussion of Population Parameters for the North Indian DPS B. Summary of Factors Affecting the North Indian DPS 1. Factor A: The Present or Threatened Destruction, Modification, or Curtailment of Its Habitat or Range a. Terrestrial Zone b. Neritic/Oceanic Zones 2. Factor B: Overutilization for Commercial, Recreational, Scientific, or Educational Purposes 3. Factor C: Disease or Predation 4. Factor D: Inadequacy of Existing Regulatory Mechanisms 5. Factor E: Other Natural or Manmade Factors Affecting Its Continued Existence a. Incidental Bycatch in Fishing Gear i. Gill Net Fisheries ii. Trawl Fisheries b. Vessel Strikes c. Beach Driving d. Pollution e. Effects of Climate Change and Natural Disaster C. Conservation Efforts for the North Indian DPS D. Extinction Risk Assessment and Findings for the North Indian DPS XII. East Indian-West Pacific DPS A. Discussion of Population Parameters for the East Indian-West Pacific DPS B. Summary of Factors Affecting the East Indian-West Pacific DPS 1. Factor A: The Present or Threatened Destruction, Modification, or Curtailment of Its Habitat or Range a. Terrestrial Zone b. Neritic/Oceanic Zones 2. Factor B: Overutilization for Commercial, Recreational, Scientific, or Educational Purposes 3. Factor C: Disease or Predation 4. Factor D: Inadequacy of Existing Regulatory Mechanisms 5. Factor E: Other Natural or Manmade Factors Affecting Its Continued Existence a. Incidental Bycatch in Fishing Gear b. Marine Debris and Pollution c. Effects of Climate Change and Natural Disasters C. Conservation Efforts for the East Indian-West Pacific DPS D. Extinction Risk Assessment and Findings for the East Indian-West Pacific DPS XIII. Central West Pacific DPS A. Discussion of Population Parameters for the Central West Pacific DPS B. Summary of Factors Affecting the Central West Pacific DPS 1. Factor A: The Present or Threatened Destruction, Modification, or Curtailment of Its Habitat or Range a. Terrestrial Zone b. Neritic/Oceanic Zones 2. Factor B: Overutilization for Commercial, Recreational, Scientific, or Educational Purposes 3. Factor C: Disease or Predation 4. Factor D: Inadequacy of Existing Regulatory Mechanisms 5. Factor E: Other Natural or Manmade Factors Affecting Its Continued Existence a. Incidental Bycatch in Fishing Gear b. Vessel Strikes c. Pollution d. Effects of Climate Change and Natural Disasters C. Conservation Efforts for the Central West Pacific DPS D. Extinction Risk Assessment and Findings for the Central West Pacific DPS XIV. Southwest Pacific DPS A. Discussion of Population Parameters in the Southwest Pacific DPS B. Summary of Factors Affecting the Southwest Pacific DPS 1. Factor A: The Present or Threatened Destruction, Modification, or Curtailment of Its Habitat or Range a. Terrestrial Zone b. Neritic/Oceanic Zones 2. Factor B: Overutilization for Commercial, Recreational, Scientific, or Educational Purposes 3. Factor C: Disease or Predation 4. Factor D: Inadequacy of Existing Regulatory Mechanisms 5. Factor E: Other Natural or Manmade Factors Affecting Its Continued Existence a. Incidental Bycatch in Fishing Gear b. Shark Control Programs c. Boat Strikes and Port Dredging d. Pollution and Marine Debris e. Effects of Climate Change and Natural Disasters C. Conservation Efforts for the Southwest Pacific DPS D. Extinction Risk Assessment and Findings for the Southwest Pacific DPS XV. Central South Pacific DPS A. Discussion of Population Parameters for the Central South Pacific DPS B. Summary of Factors Affecting the Central South Pacific DPS 1. Factor A: The Present or Threatened Destruction, Modification, or Curtailment of Its Habitat or Range a. Terrestrial Zone b. Neritic/Oceanic Zones 2. Factor B: Overutilization for Commercial, Recreational, Scientific, or Educational Purposes 3. Factor C: Disease or Predation 4. Factor D: Inadequacy of Existing Regulatory Mechanisms 5. Factor E: Other Natural or Manmade Factors Affecting Its Continued Existence a. Incidental Bycatch in Fishing Gear b. Marine Debris and Pollution c. Effects of Climate Change and Natural Disasters C. Conservation Efforts for the Central South Pacific DPS D. Extinction Risk Assessment and Findings for the Central South Pacific DPS XVI. Central North Pacific DPS A. Discussion of Population Parameters for the Central North Pacific DPS B. Summary of Factors Affecting the Central North Pacific DPS 1. Factor A: The Present or Threatened Destruction, Modification, or Curtailment of Its Habitat or Range a. Terrestrial Zone b. Neritic/Oceanic Zones 2. Factor B: Overutilization for Commercial, Recreational, Scientific, or Educational Purposes 3. Factor C: Disease or Predation 4. Factor D: Inadequacy of Existing Regulatory Mechanisms 5. Factor E: Other Natural or Manmade Factors Affecting Its Continued Existence a. Incidental Bycatch in Fishing Gear i. Longline Fisheries ii. Gillnet Fisheries iii. Other Gear Types b. Marine Debris and Pollution c. Vessel Interactions d. Effects of Climate Change e. Effects of Spatial Structure C. Conservation Efforts for the Central North Pacific DPS D. Extinction Risk Assessment and Findings for the Central North Pacific DPS XVII. East Pacific DPS A. Discussion of Population Parameters for the East Pacific DPS B. Summary of Factors Affecting the East Pacific DPS 1. Factor A: The Present or Threatened Destruction, Modification, or Curtailment of Its Habitat or Range a. Terrestrial Zone b. Neritic/Oceanic Zones 2. Factor B: Overutilization for Commercial, Recreational, Scientific, or Educational Purposes 3. Factor C: Disease or Predation 4. Factor D: Inadequacy of Existing Regulatory Mechanisms 5. Factor E: Other Natural or Manmade Factors Affecting Its Continued Existence a. Incidental Bycatch in Fishing Gear b. Pollution c. Effects of Climate Change and Natural Disasters C. Conservation Efforts for the East Pacific DPS D. Extinction Risk Assessment and Findings for the East Pacific DPS XVIII. Proposed Determinations XIX. Significant Portion of the Range XX. Effects of Listing A. Identifying Section 7 Conference and Consultation Requirements B. Critical Habitat C. Take Prohibitions D. Identification of Those Activities That Would Constitute a Violation of Section 9 of the ESA XXI. Peer Review XXII. Classification A. National Environmental Policy Act B. Executive Order 12866, Regulatory Flexibility Act, and Paperwork Reduction Act C. Executive Order 13132, Federalism I. Background

On July 28, 1978, NMFS and USFWS, collectively referred to as the Services, listed the green turtle ( Chelonia mydas ) under the ESA (43 FR 32800). Pursuant to the authority that the statute provided, and prior to the current language in the definition of “species” regarding DPSs, the Services listed the species as threatened, except for the Florida and Mexican Pacific Coast breeding populations, which were listed as endangered. The Services published recovery plans for U.S. Atlantic ( http://www.nmfs.noaa.gov/pr/recovery/plans.htm ) and U.S. Pacific (including the East Pacific) populations of the green turtle (63 FR 28359, May 22, 1998). NMFS designated critical habitat for the species to include waters surrounding Culebra Island, Commonwealth of Puerto Rico, and its outlying keys (63 FR 46693, September 2, 1998).

On February 16, 2012, the Services received a petition from the Association of Hawaiian Civic Clubs to identify the Hawaiian green turtle population as a DPS and “delist” the DPS under the ESA. On August 1, 2012, NMFS, with USFWS concurrence, determined that the petition presented substantial information indicating that the petitioned action may be warranted (77 FR 45571). Initiating a review of new information in accordance with the DPS policy was consistent with the recommendation made in the Services' 2007 Green Sea Turtle 5-year Review. The Services initiated a status review to consider the species across its range, determine whether the petitioned action is warranted, and determine whether other DPSs could be recognized. The Services decided to review the Hawaiian population in the context of green turtles globally with regard to application of the DPS policy and in light of significant new information since the listing of the species in 1978.

The Services appointed a Status Review Team (SRT) in September 2012. SRT members were affiliated with NMFS Science Centers and the Services' field, regional, and headquarters offices, and provided a diverse range of expertise, including green turtle genetics, demography, ecology, and management, as well as risk analysis and ESA policy. The SRT was charged with reviewing and evaluating all relevant scientific information relating to green turtle population structure globally to determine whether any populations may qualify as DPSs and, if so, to assess the extinction risk for each proposed DPS. Findings of the SRT are detailed in the “Green Turtle ( Chelonia mydas ) Status Review under the U.S. Endangered Species Act” (hereinafter referred to as the Status Review; NMFS and USFWS, 2014). The Status Review underwent independent peer review by 14 scientists with expertise in green turtle biology, genetics, or related fields, and endangered species listing policy. The Status Review is available electronically at http://www.nmfs.noaa.gov/pr/species/turtles/green.htm .

This Federal Register document announces the 12-month finding on the petition to identify the Hawaiian green turtle population as a DPS and remove the protections of the ESA from the DPS, and includes a proposed rule to revise the existing listings to identify 11 green turtle DPSs worldwide and list them as threatened or endangered under the ESA in place of the existing listings. Our determinations have been made only after review of the best available scientific and commercial information pertaining to the species throughout its range and within each DPS. This is similar to the action we took for loggerhead sea turtles (76 FR 58868, September 22, 2011).

The ESA gives us clear authority to make these listing determinations and to revise the lists of endangered and threatened species to reflect these determinations. Section 4(a)(1) of the ESA authorizes us to determine by regulation whether “any species,” which is expressly defined to include species, subspecies, and DPS, is an endangered species or a threatened species based on certain factors. Review of the status of a species may be commenced at any time, either on the Services' own initiative—through a status review or in connection with a 5-year review under Section 4(c)(2)—or in response to a petition. Because a DPS is not a scientifically recognized entity, but rather one that is created under the language of the ESA and effectuated through our DPS Policy (61 FR 4722, February 7, 1996), we have some discretion to determine whether the species should be reclassified into DPSs and what boundaries should be recognized for each DPS. Section 4(c)(1) gives us authority to update the lists of threatened and endangered species to reflect these determinations. This can include revising the lists to remove a species or reclassify the listed entity.

II. Policies for Delineating Species Under the ESA

Section 3 of the ESA defines “species” as including “any subspecies of fish or wildlife or plants, and any distinct population segment of any species of vertebrate fish or wildlife which interbreeds when mature.” The term “distinct population segment” is not recognized in the scientific literature. Therefore, the Services adopted a joint policy for recognizing DPSs under the ESA (DPS Policy; 61 FR 4722) on February 7, 1996. The DPS Policy requires the consideration of three elements when evaluating the status of possible DPSs: (1) The discreteness of the population segment in relation to the remainder of the species to which it belongs; (2) the significance of the population segment to the species to which it belongs; and (3) the population segment's conservation status in relation to the

III. Listing Determinations Under the ESA

The ESA defines an endangered species as one that is in danger of extinction throughout all or a significant portion of its range (section 3(6)), and a threatened species as one that is likely to become endangered in the foreseeable future throughout all or a significant portion of its range (section 3(20)). Thus, in the context of the ESA, the Services interpret an “endangered species” to be one that is presently in danger of extinction. A “threatened species,” on the other hand, is not presently in danger of extinction, but is likely to become so in the foreseeable future. In other words, the primary statutory difference between a threatened and endangered species is the timing of when a species may be in danger of extinction, either presently (endangered) or in the foreseeable future (threatened).

When we consider whether a species might qualify as threatened under the ESA, we must consider the meaning of the term “foreseeable future.” It is appropriate to interpret “foreseeable future” as the horizon over which predictions about the conservation status of the species can be reasonably relied upon. The foreseeable future considers the life history of the species, habitat characteristics, availability of data, particular threats, ability to predict threats, and the reliability to forecast the effects of these threats and future events on the status of the species under consideration. Because a species may be susceptible to a variety of threats for which different data are available, or which operate across different time scales, the foreseeable future is not necessarily reducible to a particular number of years. For the green turtle, the SRT used a horizon of 100 years to evaluate the likelihood that a DPS would reach a critical risk threshold ( i.e., quasi-extinction). In making the proposed listing determinations, we applied the horizon of 100 years in our consideration of foreseeable future under the scope of the definitions of endangered and threatened species, pursuant to section 3 of the ESA.

The statute requires us to determine whether any species is endangered or threatened as a result of any one or combination of the following 5-factors: (1) The present or threatened destruction, modification, or curtailment of its habitat or range; (2) overutilization for commercial, recreational, scientific, or educational purposes; (3) disease or predation; (4) the inadequacy of existing regulatory mechanisms; or (5) other natural or manmade factors affecting its continued existence (section 4(a)(1)(A-E) of the ESA). Section 4(b)(1)(A) of the ESA requires us to make this determination based solely on the best available scientific and commercial data available after conducting a review of the status of the species and taking into account any efforts being made by States or foreign governments to protect the species.

IV. Biology and Life History of Green Turtles

A thorough account of green turtle biology and life history may be found in the Status Review, which is incorporated here by reference. The following is a succinct summary of that information.

The green turtle, C. mydas, has a circumglobal distribution, occurring throughout tropical, subtropical, and, to a lesser extent, temperate waters. Their movements within the marine environment are not fully understood, but it is believed that green turtles inhabit coastal waters of over 140 countries (Groombridge and Luxmoore, 1989). The Status Review lists 468 known nesting sites worldwide, with 79 having nesting aggregations with greater than 500 females. The largest green turtle nesting aggregation, with an estimated number of nesting females greater than 132,000, is Tortuguero, Costa Rica (Sea Turtle Conservancy, 2013). There are 14 aggregations estimated to have 10,001-100,000 nesting females: Quintana Roo, Mexico (Julio Zurita, pers. comm., 2012); Ascension Island, UK (S. Weber, Ascension Island Government, pers. comm., 2013); Poilão, Guinea-Bissau (Catry et al., 2009); Aldabra Atoll, Seychelles (Mortimer et al. , 2011; Mortimer, 2012; J. Mortimer, unpubl. data.); Mohéli, Comoros Islands, France (Bourjea, 2012); Mayotte, Comoros Islands (Bourjea, 2012); Europa, Esparses Islands, France (Lauret-Stepler et al., 2007; Bourjea, 2012); Ras Al Hadd, Oman (AlKindi et al., 2008); Ras Sharma, Yemen (PERSGA/GEF, 2004); Wellesley Group, Australia (Unpubl. data cited in Limpus, 2009); Raine Island, Australia (Chaloupka et al., 2008a; Limpus, 2009); Moulter Cay, Australia (Limpus, 2009); Capricorn Bunker Group of Islands, Australia (Limpus et al., 2003); and Colola, Mexico (Delgado-Trejo and Alvarado-Figueroa, 2012).

Most green turtles spend the majority of their lives in coastal foraging grounds. These areas include fairly shallow waters in open coastline and protected bays and lagoons. While in these areas, green turtles rely on marine algae and seagrass as their primary diet constituents, although some populations also forage heavily on invertebrates. These marine habitats are often highly dynamic and in areas with annual fluctuations in seawater and air temperatures, which can cause the distribution and abundance of potential green turtle food items to vary substantially between seasons and years (Carballo et al., 2002).

At nesting beaches, green turtles rely on beaches characterized by intact dune structures, native vegetation, little to no artificial lighting, and 26 to 35° C beach temperatures for nesting (Limpus, 1971; Salmon et al., 1992; Ackerman, 1997; Witherington, 1997; Lorne and Salmon, 2007). Nests are typically laid at night at the base of the primary dune (Hirth, 1997; Witherington et al., 2006). Complete removal of vegetation, or coastal construction, can affect thermal regimes on beaches and thus affect the incubation and resulting sex ratio of hatchling turtles. Nests laid in these areas are at a higher risk of tidal inundation (Schroeder and Mosier, 2000).

Hatchlings emerge from their nests en masse and almost exclusively at night, presumably using decreasing sand temperature as a cue (Hendrickson, 1958; Mrosovsky, 1968). Immediately after hatchlings emerge from the nest, they begin a period of frenzied activity. During this active period, hatchlings crawl to the surf, swim, and are swept through the surf zone (Carr and Ogren, 1960; Carr, 1961; Wyneken and Salmon, 1992). They orient to waves in the nearshore area and to the magnetic field as they proceed further toward open water (Lohmann and Lohmann, 2003).

Upon leaving the nesting beach and entering the marine environment, post-hatchling green turtles begin an oceanic juvenile phase during which they are presumed to primarily inhabit areas where surface waters converge to form local downwellings that result in linear accumulations of floating material, especially Sargassum sp. This association with downwellings is well-documented for loggerhead sea turtles ( Caretta caretta ), as well as for some post-hatchling green turtles (Witherington et al., 2006; 2012). The smallest of oceanic green turtles associating with these areas are relatively active, moving both within Sargassum sp. mats and in nearby open water, which may limit the ability of et al., 2012).

Oceanic-stage juvenile green turtles originating from nesting beaches in the Northwest Atlantic appear to use oceanic developmental habitats and move with the predominant ocean gyres for several years before returning to their neritic (shallower water, generally to 200 m depth, including open coastline and protected bays and lagoons) foraging and developmental habitats (Musick and Limpus, 1997; Bolten, 2003). Larger neonate green turtles (at least 15-26 cm straight carapace length; SCL) are known to occupy Sargassum sp. habitats and surrounding epipelagic waters, where food items include Sargassum sp. and associated invertebrates, fish eggs, and insects (Witherington et al., 2012). Knowledge of the diet and behavior of oceanic stage juveniles, however, is limited.

The neritic juvenile stage begins when green turtles exit the oceanic zone and enter the neritic zone (Bolten, 2003). The age at recruitment to the neritic zone likely varies with individuals leaving the oceanic zone over a wide size range (summarized in Avens and Snover, 2013). After migrating to the neritic zone, juveniles continue maturing until they reach adulthood, and some may periodically move between the neritic and oceanic zones (NMFS and USFWS, 2007; Parker et al., 2011). The neritic zone, including both open coastline and protected bays and lagoons, provides important foraging habitat, inter-nesting habitat, breeding, and migratory habitat for adult green turtles (Plotkin, 2003; NMFS and USFWS, 2007). Some adult females may also periodically move between the neritic and oceanic zones (Plotkin, 2003; Hatase et al., 2006) and, in some instances, adult green turtles may reside in the oceanic zone for foraging (NMFS and USFWS, 2007; Seminoff et al., 2008; Parker et al., 2011). Despite these uses of the oceanic zone by green turtles, much remains unknown about how oceanography affects juvenile and adult survival, adult migration, prey availability, and reproductive output.

Most green turtles exhibit slow growth rates, which has been described as a consequence of their largely herbivorous ( i.e., low net energy) diet (Bjorndal, 1982). Consistent with slow growth, age-to-maturity for green turtles appears to be the longest of any sea turtle species (Chaloupka and Musick, 1997; Hirth, 1997). Published age at sexual maturity estimates are as high as 35-50 years, with lower ranges reported for known age turtles from the Cayman Islands (15-19 years; Bell et al., 2005) and Caribbean Mexico (12-20 years; Zurita et al., 2012) and some mark-recapture projects ( e.g., 15-25 years in the Eastern Pacific; Seminoff et al., 2002a). Mean adult reproductive lifespan of green turtles from Australia's southern Great Barrier Reef (GBR) has been estimated at 19 years using mark-recapture and survival data (Chaloupka and Limpus, 2005). The maximum nesting lifespan observed in a 27-year tag return dataset from Trindade Island, Brazil was 16 years; however, nesting monitoring was discontinuous over time (Almeida et al., 2011). Tag return data comprising 2,077 females (42,928 nesting events, 1968-partial 2012 season) from continuous monitoring at French Frigate Shoals (FFS), Hawai`i show maximum nesting lifespans of 37-38 years (n=2), with many individuals (n=54) documented nesting over a minimum of 25-35 years (I. Nurzia-Humburg, S. Hargrove, and G. Balazs, NMFS, unpublished data, 2013).

V. Overview of the Policies and Process Used To Identify DPSs

The SRT considered a vast array of information in assessing whether there are any green turtle population segments that satisfy the DPS criteria of being both discrete and significant. In anticipation of conducting a green turtle status review, NMFS contracted two post-doctoral associates in 2011 to collect and synthesize genetic and demographic information on green turtles worldwide. The SRT was presented with, and evaluated, this genetic and demographic information. Demographic information included green turtle nesting information; morphological and behavioral data; movements, as indicated by tagging (flipper and passive integrated transponder (PIT) tags) and satellite telemetry data; and anthropogenic impacts. Also discussed and considered as a part of this analysis were oceanographic features and geographic barriers.

A population may be considered discrete if it satisfies either one of the following conditions: (1) It is markedly separated from other populations of the same taxon as a consequence of physical, physiological, ecological, or behavioral factors; or (2) it is delimited by international governmental boundaries within which differences in control of exploitation, management of habitat, conservation status, or regulatory mechanisms exist that are significant in light of section 4(a)(1)(D) of the ESA (61 FR 4722, February 7, 1996). According to the policy, quantitative measures of genetic or morphological discontinuity can be used to provide evidence for item (1). The SRT compiled a list of attributes that suggested various population groups might be considered discrete, identified potentially discrete units, and discussed alternative scenarios for lumping or splitting these potentially discrete units. After arriving at a tentative list of units, each member of the SRT was given 100 points that could be distributed among two categories: (1) The unit under consideration is discrete, and (2) the unit under consideration is not discrete. The spread of points reflects the level of certainty of the SRT surrounding a decision to call the unit discrete. The SRT determined that there are 11 discrete regional populations of green turtles globally. Each of these was then evaluated for significance.

A population may be considered significant if it satisfies any one of the following conditions: (1) Persistence of the discrete segment in an ecological setting unusual or unique for the taxon; (2) evidence that loss of the discrete segment would result in a significant gap in the range of the taxon; (3) evidence that the discrete segment represents the only surviving natural occurrence of a taxon that may be more abundant elsewhere as an introduced population outside its historical range; and (4) evidence that the discrete segment differs markedly from other populations of the species in its genetic characteristics. Because condition (3) is not applicable to green turtles, the SRT addressed conditions (1), (2) and (4). The SRT listed the attributes that would make potential DPSs (those determined to be discrete in the previous step) significant. As in the vote for discreteness, members of the SRT were then given 100 points with which to vote for whether each unit met the significance criterion in the joint policy. All units that had been identified as discrete were also determined to be significant.

For more discussion on the process the SRT used to identify DPSs, see Section 3 of the Status Review document.

A. Discreteness Determination

In evaluating discreteness among the global green turtle population, the SRT began by focusing on the physical separation of ocean basins ( i.e., Atlantic, Pacific, and Indian Oceans). The result was an evaluation of data by major ocean basins, although it quickly became clear that the Indian and Pacific

Within each ocean basin, the SRT started by evaluating genetic information. The genetic data consisted of results from studies using maternally inherited mitochondrial DNA (mtDNA), biparentally inherited nuclear DNA (nDNA) microsatellite (a section of DNA consisting of very short nucleotide sequences repeated many times), and single nucleotide polymorphism (a DNA sequence variation occurring commonly within a population) markers. Next, the SRT reviewed tagging, telemetry and demographic data, and additional information such as potential differences in morphology. The SRT also considered whether the available information suggests that green turtle population segments are separated by vicariant barriers, such as oceanographic features ( e.g., current systems), or biogeographic boundaries.

Genetic information that was presented to the SRT resulted from a global phylogenetic analysis (analysis based on natural evolutionary relationships) based on sequence data from a total of 129 mtDNA haplotypes ( i.e., mtDNA sequences, which are inherited together) identified from approximately 4,400 individuals sampled at 105 green turtle nesting sites around the world (Jensen and Dutton, NMFS, unpublished data; M. Jensen, NRC, pers. comm., 2013). Results indicated that the mtDNA variation present in green turtles throughout the world today occurs within eight major clades ( i.e., a group consisting of an ancestor and all its descendants) that are structured geographically within ocean basins. These clades represent similarities between haplotypes on evolutionary timescales as opposed to ecological timescales. See Figure 1 for a visual representation of these clades. There is divergence among individual haplotypes within each green turtle clade (M. Jensen, NRC, pers. comm., 2013) and discrete populations can exist within these clades.

BILLING CODE 3510-22-P EP23mr15.000

BILLING CODE 3510-22-C 1. Atlantic Ocean/Mediterranean Sea

Two of the eight major mtDNA clades, Clades I and II, are found in the Atlantic/Mediterranean region. Clade I includes haplotypes primarily found in turtles from the Mediterranean and the western North Atlantic. Within Clade I, two strongly divergent groups of haplotypes are found, with one group being restricted to the Mediterranean and the other being restricted to the western North Atlantic. Mediterranean and western North Atlantic turtles share only one specific haplotype that has been found in only two individuals, indicating very strong long-term isolation of females. As such, there is strong evidence that these two geographically-separated groups of divergent haplotypes may be considered discrete.

In addition to genetic evidence for discreteness, in the Mediterranean, green turtles are spatially separated from populations in the Atlantic and Indian Oceans, with the nearest known nesting sites outside the Mediterranean being several thousand kilometers away in the Republic of Senegal (Senegal), and the North Atlantic population being more than 8,000 km away. Further, no turtles tagged in the eastern Mediterranean have been recovered farther west than the Tunisian Republic (Tunisia) inside the Mediterranean. Nesting females from Cyprus, Turkey, the Syrian Arab Republic (Syria), and the State of Israel (Israel) have been satellite tracked to the Arab Republic of Egypt (Egypt), Libya, and Turkey—with movements largely restricted to the eastern Mediterranean (Godley et al., 2002; Broderick et al., 2007). Post-nesting turtles from this region migrate primarily along the coast from their nesting beach to their foraging and et al., 2002; Broderick et al., 2007).

Demographic evidence of discreteness of Mediterranean green turtles lies in the fact that Mediterranean green turtles are the second smallest green turtles worldwide (the smallest being in the eastern Pacific), with a mean nesting size in Alagadi, Cyprus of 92 cm Curved Carapace Length (CCL; Broderick et al., 2003), compared with 95 cm to 110 cm CCL size range for most other populations.

In the North Atlantic, tag recovery and telemetry data indicate that nesting females primarily reside within the North Atlantic. Some nesting females tagged at Tortuguero, Costa Rica were recaptured in the South Atlantic (Troëng et al., 2005). There is some degree of mixing of immature turtles on foraging pastures between the North and South Atlantic; however, nesting sites in the eastern Caribbean carry mostly mtDNA haplotypes from a different clade (II), indicating strong long-term isolation. Tagging studies have identified juveniles from this population in waters off Brazil and Argentina, but we found no evidence of movement of mature individuals.

The second clade within the Atlantic Ocean basin, Clade II, includes haplotypes found in all South Atlantic nesting sites, some eastern Caribbean turtles, and some turtles in the southwest Indian Ocean. With a few exceptions, green turtles in the South Atlantic carry an mtDNA haplotype that is found nowhere else, indicating strong isolation of matrilines over evolutionary time periods. The exceptions to this pattern are: (1) One nesting site from the eastern Caribbean, which exhibits a low frequency of a haplotype from the North Atlantic/Mediterranean clade (Clade I); (2) nesting sites from the Gulf of Mexico/Central America, which have a low frequency of Clade II haplotypes; and (3) two nesting sites from southeast Africa, which have high frequencies of Clade II haplotypes. The presence of a shared haplotype in South Atlantic and southwest Indian Ocean rookeries demonstrates for the first time a recent matrilineal link between Atlantic and Indian Ocean green turtle populations (Bourjea et al., 2007b). However, the SRT believes all these exceptions reflect historical events rather than contemporary connectivity. This interpretation is supported by satellite telemetry, which reveals extensive movements of turtles within the South Atlantic region but no evidence for migrations into other areas, other than rare instances of movement into foraging areas in the North Atlantic. Long stretches of cold water along the coasts of Patagonia and southwest Africa serve to isolate South Atlantic turtles from populations in the Indian and Pacific Oceans.

Foraging ground studies in the Atlantic have generally shown regional structuring with strong stock contribution from nearby regional nesting sites, but little mixing over long distances (Bolker et al., 2007). Overall, the distribution of the two genetic haplotype lineages (Clade I and Clade II) is very similar to what is seen for the nesting sites and indicates a strong regional structuring with little overlap (Bolker et al., 2007). However, a recent study showed that a large proportion of juvenile green turtles in the Cape Verde Islands in the eastern Atlantic originated from distant nesting sites across the Atlantic, namely Suriname (38 percent), Ascension Island (12 percent) and Guinea Bissau (19 percent), suggesting that, like loggerheads, green turtles in the Atlantic undertake transoceanic developmental migrations (Monzón-Argüello et al., 2010). The fact that long distance dispersal is only seen for juvenile turtles suggests that larger adult-sized turtles return to forage within the region of their natal nesting sites, thereby limiting the potential for gene-flow across larger scales (Monzón-Argüello et al., 2010).

In the South Atlantic, flipper tag recoveries have established movement between feeding grounds and nesting sites in the Caribbean and Brazil (Lima et al., 2003; Lima et al., 2008; Lima et al., 2012), and telemetry data indicate that juvenile green turtles move from Argentina to Uruguay and Brazil, from Uruguay to Brazil, and from the Guianas to Brazil. Telemetry studies indicate that nesting females from the eastern South Atlantic (west coast of Africa) are confined to the eastern South Atlantic, and nesting females from the western South Atlantic are confined to the western South Atlantic. In the eastern South Atlantic, all tracked turtles remained in the general vicinity of their release location. Nesting females from Ascension Island were tracked to foraging grounds along the coast of Brazil.

Finally, demographic evidence for discreteness of South Atlantic green turtles lies in the fact that the South Atlantic is home to the largest green turtles in the world, with a mean nesting size of green turtles at Atol das Rocas, Brazil of 118.6 cm CCL (n=738), compared with 95 cm to 110 cm CCL size range for most other populations.

Based on the information presented above, the SRT concluded, and we concur, that three discrete populations exist in the Atlantic Ocean/Mediterranean: (1) North Atlantic, (2) Mediterranean, and (3) South Atlantic. These three populations are markedly separated from each other and from populations within the Pacific Ocean and Indian Ocean basins as a consequence of physical (including both oceanographic basins and currents), ecological, and behavioral factors. Information supporting this conclusion includes genetic analysis, flipper tag recoveries, and satellite telemetry.

2. Indian Ocean

Green turtles from the Indian Ocean exhibit haplotypes from Clades II, III, IV, VI, and VII. In the southwest Indian Ocean, Bourjea et al. (2007b) genetically assessed the population structure among 288 nesting green turtles from 10 nesting sites. Overall, the southwest Indian Ocean appears to have at least two genetic stocks: (1) The South Mozambique Channel (Juan de Nova and Europa); and (2) the North Mozambique Channel. As stated earlier, the authors recorded a high presence of a common and widespread South Atlantic Ocean haplotype (CM-A8) in the South Mozambique Channel. However, the observation that only a single Atlantic haplotype has been observed and that it occurs in high frequency among South Mozambique Channel rookeries suggests that gene flow is not ongoing (Bourjea et al., 2007b). Nesting sites in the North Mozambique Channel share several haplotypes (including CmP47 and CmP49) with nesting sites in the eastern Indian Ocean, Southeast Asia and the Western Pacific, indicating strong-connectivity with the eastern Indian Ocean population. However, tagging and tracking data document movements within the Southwest Indian Ocean but not between it and the eastern Indian and western Pacific Oceans. Although there is some evidence of trans-boundary movement between the southwest Indian Ocean and the population in the North Indian Ocean, evidence from tag returns indicates that most remain in the southwest Indian Ocean. Indeed, some green turtles in Tanzania are probably resident, and others are highly migratory, moving to and from nesting and feeding grounds within the southwest Indian Ocean in Kenya, Seychelles, Comoros, Mayotte, Europa Island and South Africa (Muir, 2005). From 2009 to 2011, 90 satellite transmitters deployed on nesting green turtles at five nesting sites in the southwest Indian Ocean showed that nearly 20 percent of the tracked turtles used Madagascar coastal foraging grounds while more than 80 percent

In the North Indian Ocean, limited information from only a single nesting site (Jana Island, Saudi Arabia, n=27) exists on the genetic structure (M. Jensen, NRC, pers. comm., 2013). Nonetheless, four mtDNA haplotypes never reported from any other nesting site were identified from Jana Island, and are highly divergent from other haplotypes in the Indian Ocean. This population also appears to be isolated from other Indian populations by substantial breaks in nesting habitat along the Horn of Africa and along the entire eastern side of the Indian subcontinent.

Tagging of turtles on nesting beaches of the North Indian Ocean started in the late 1970s and indicates that some turtles in the North Indian Ocean migrate long distances from distant feeding grounds to nesting beaches while others are quite sedentary, but all stay within the North Indian Ocean. Tagging studies have revealed that some turtles nesting on Ras Al Hadd and Masirah, Oman can be found as far away as Somalia, Ethiopia, Yemen, Saudi Arabia, the upper Gulf, and Pakistan (Ross, 1987; Salm, 1991), and a green turtle tagged in Oman was found in the Maldives (Al-Saady et al., 2005). No tagging has been carried out on feeding grounds (Al-Saady et al., 2005).

A few green turtles in the North Indian Ocean have been fitted with satellite transmitters and reported at www.seaturtle.org , but no data have been published. One telemetered female green turtle remained in the coastal areas of the Persian Gulf for 49 days (N. Pilcher, Marine Research Foundation, pers. comm., 2013), and two nesting turtles were telemetered at Masirah Island, Oman, both of which moved southward along the Arabian Peninsula and were found in the Red Sea when the transmissions ceased (Rees et al. 2012). Telemetry data for captive-hatched and reared green turtles at Republic of Maldives (Vabbinfaru Island, Male Atoll) have indicated wide movement patterns within the Indian Ocean (N. Pilcher, Marine Research Foundation, pers. comm., 2013).

In the eastern Indian Ocean, turtles mix readily with those in the western Pacific. Genetic sampling in the eastern Indian and western Pacific Ocean regions has been fairly extensive with more than 22 nesting sites sampled although, because there are a high number of nesting sites in this region and there is complex structure, there remain gaps in sampling relative to distribution ( e.g., Thailand, Vietnam, parts of Indonesia, and the Philippines). Most nesting sites are dominated by haplotypes from Clade VII, but with some overlap of Clades III and IV throughout the Indian Ocean—evidence of a complex colonization history in this region. While one common haplotype is shared across the Indian Ocean, substantial gaps in nesting sites along the east coast of India and in the southern Indian Ocean serve to isolate the eastern Indian-western Pacific population from those in the north and southwest Indian Ocean. The Wallace Line (a boundary drawn in 1859 by the British naturalist Alfred Russel Wallace that separates the highly distinctive faunas of the Asian and Australian biogeographic regions) and its northern extension separate this population from populations to the east, which carry haplotypes primarily from Clade IV. Nesting sites to the northern extreme (Taiwan and Japan) show more complex patterns of higher mixing of divergent haplotypes, and the placement of individual nesting sites within this area is somewhat uncertain and may become better resolved when additional genetic data are available.

Significant population substructuring occurs among nesting sites in this area. Mixed-stock analysis of foraging grounds shows that green turtles from multiple nesting beaches commonly mix at feeding areas across northern Australia (Dethmers et al., 2006) and Malaysia (Jensen, 2010), with higher contributions from nearby large nesting sites. Satellite tracking also shows green turtle movement throughout the eastern Indian and western Pacific (Cheng, 2000; Dermawan, 2002; Charuchinda et al., 2003; Wang, 2006).

Given the information presented above, the SRT concluded, and we concur, that three discrete populations exist in the Indian Ocean, with the third overlapping with the Pacific: (1) Southwest Indian, (2) North Indian, and (3) East Indian-West Pacific. These three populations are markedly separated from each other and from populations within the Atlantic Ocean as a consequence of physical, ecological, and behavioral factors. Information supporting this conclusion includes genetic analysis, flipper tag recoveries, and satellite telemetry.

3. Pacific Ocean

The central west Pacific encompasses most of the area commonly referred to as Micronesia as well as parts of Melanesia. Genetic sampling in the central west Pacific has recently improved, but remains challenging, given the large number of small island and atoll nesting sites. At least five management units have been identified in the region (Palau, Independent State of Papua New Guinea (PNG), Yap, CNMI/Guam, and the Republic of the Marshall Islands (Marshall Islands); Dethmers et al., 2006; M. Jensen, NRC, pers. comm., 2013; Dutton et al., 2014). The central west Pacific carries haplotypes from Clade IV, while the populations to the west carry haplotypes predominantly from Clade VII, so any mixing presumably reflects foraging migrations rather than interbreeding. The boundary between the central west Pacific and the East Indian-West Pacific populations is congruent with the northern portion of the Wallace Line. Wide expanses of open ocean separate the central west Pacific from the central north Pacific, and genetic data provide no evidence of gene flow between the central west Pacific and the central north Pacific over evolutionary time scales. Tagging studies also have not found evidence for migration of breeding adults to or from adjacent populations.

In the southwest Pacific, genetic sampling has been extensive for larger nesting sites along the GBR, the Coral Sea and New Caledonia (Dethmers et al., 2006; Jensen, 2010; Dutton et al., 2014). However, several smaller nesting sites in this region have not been sampled ( e.g., Solomon Islands, Republic of Vanuatu (Vanuatu), Tuvalu, PNG, etc.). The southwest Pacific population is characterized by haplotypes from Clade V, which have been found only at nesting sites in this population. It also has a high frequency of haplotypes from Clades III and IV, as well as low frequency of haplotypes from Clades VI and VII, making this area highly diverse (haplotypes from the widespread Clade IV differ from those found in the central west and central south Pacific).

Traditional capture-mark-recapture studies (Limpus, 2009) and genetic mixed-stock analysis (Jensen, 2010) show that turtles from several different southwest Pacific nesting sites overlap on feeding grounds along the east coast of Australia. This mixing in foraging areas might provide mating opportunities between turtles from different stocks as evidenced by the lack of differentiation found between the northern and southern GBR nesting sites et al., 1997). However, tagging, telemetry, and genetic studies show movement of breeding adults occurs mainly within the southwest Pacific.

In the central South Pacific, genetic sampling has been limited to two nesting sites (American Samoa and French Polynesia) among the many small isolated nesting sites that characterize this region, but they both contain relatively high frequencies of Clade III haplotypes, which are not found in the central west and southwest Pacific populations. Nesting sites from this area share some haplotypes with surrounding nesting sites, but at low frequency. There are also limited data on mixed-stock foraging areas from this region. Flipper tag returns and satellite tracking studies demonstrate that post-nesting females travel the complete geographic breadth of this population, from French Polynesia in the east to Fiji in the west, and sometimes even slightly beyond (Tuato'o-Bartley et al., 1993; Craig et al., 2004; Maison et al., 2010; White, 2012), as far as the Philippines (Trevor, 2009). The complete extent of migratory movements is unknown. The central South Pacific is isolated by vast expanses of open ocean from turtle populations to the north (Hawai`i) and east (Galapagos), and in both of these areas all turtle haplotypes are from an entirely different clade (Clade VIII), indicating lack of genetic exchange across these barriers.

The central North Pacific, which includes the Hawaiian Archipelago and Johnston Atoll, is inhabited by green turtles that are geographically discrete in their genetic characteristics, range, and movements, as evidenced by genetic studies and mark-recapture studies using flipper tags, microchip tags, and satellite telemetry. The key nesting aggregations within the Hawaiian Archipelago have all been genetically sampled. Mitochondrial DNA studies show no significant differentiation (based on haplotype frequency) between FFS and Laysan Island (P. Dutton, NMFS, pers. comm., 2013). While the Hawaiian Islands do share haplotypes with Revillagigedos Islands (CmP1.1 and CmP3.1) at low frequency, the populations remain highly differentiated, and there is little evidence of significant ongoing gene flow. The Frey et al. (2013) analysis of mtDNA and nDNA in scattered nesting sites on the main Hawaiian Islands (MHI; Molokai, Maui, Oahu, Lanai, and Kauai) showed that nesting in the MHI might be attributed to a relatively small number of females that appear to be related to each other and demographically isolated from FFS.

Turtles foraging in the MHI originate from Hawaiian nesting sites, with very rare records of turtles from outside the central North Pacific (Dutton et al., 2008), and there is a general absence of turtles from the Hawaiian breeding population at foraging areas outside the central North Pacific. From 1965-2013, 17,536 green turtles (juvenile through adult stages) were tagged. With only three exceptions, the 7,360 recaptures of these tagged turtles have been within the Hawaiian Archipelago. The three outliers involved recoveries in Japan, the Marshall Islands, and the Philippines (G. Balazs, NMFS, pers. comm., 2013).

Information from tagging at FFS, areas in the MHI, the Northwest Hawaiian Islands (NWHI) to the northwest of FFS, and at Johnston Atoll shows that reproductive females and males periodically migrate to FFS for seasonal breeding from the other locations. At the end of the season they return to their respective foraging areas. The reproductive migrations of 19 satellite tracked green turtles (16 females and 3 males) all involved movements between FFS and the MHI. Conventional tagging using microchips and metal flipper tags has resulted in the documentation of 164 turtles making reproductive movements from or to FFS and foraging pastures in the MHI, and 58 turtles from or to FFS and the foraging pastures in the NWHI (G. Balazs, NMFS, unpubl. data).

Hawaiian green turtles also exhibit morphological features that may make them discrete from other populations, possibly reflecting genetic as well as ecological adaptations. In the Hawai`i population, and in Australian populations, green turtles have a well-developed crop, which has not been found in Caribbean or eastern Pacific populations of green turtles (Balazs et al., 1998; J. Seminoff, NMFS, unpubl. data). In addition, juvenile green turtles in Hawai`i have proportionally larger rear flippers than those in the western Caribbean (Wyneken and Balazs, 1996; Balazs et al., 1998). These anatomical differences may reflect adaptive variation to different environmental conditions. A crop that holds food material in the esophagus would permit more food to be ingested during each foraging event in a more dynamic feeding environment, which is helpful along wind-swept rugged coastlines where large waves crash ashore. Larger flippers would also aid in making them stronger swimmers in this feeding environment, and during reproductive migrations across rough pelagic waters, as opposed to calmer coastal waters (Balazs et al., 1998).

The central North Pacific population and those in the central South Pacific and central west Pacific appear to be separated by large oceanic areas, and the central North Pacific and the eastern Pacific populations are separated by the East Pacific Barrier, an oceanographic barrier that greatly restricts or eliminates gene flow for most marine species from a wide range of taxa (Briggs, 1974).

In the eastern Pacific, genetic sampling has been extensive and the coverage in this region is substantial, considering the relatively small population sizes of most eastern Pacific nesting sites, which include both mainland and insular nesting. This sampling indicates complete isolation of nesting females between the eastern and western Pacific nesting sites. Recent efforts to determine the nesting stock origins of green turtles assembled in foraging areas have found that green turtles from several eastern Pacific nesting stocks commonly mix at feeding areas in the Gulf of California and along the Pacific coast in San Diego Bay, U.S. (Nichols, 2003; P. Dutton, NMFS, unpubl. data). In addition, green turtles of eastern Pacific origin have been found, albeit very rarely, in waters off Hawai‘i (LeRoux et al., 2003; Dutton et al., 2008), Japan (Kuroyanagi et al., 1999; Hamabata et al., 2009), and New Zealand (Godoy et al., 2012). A recent study of juvenile green turtles foraging at Gorgona Island in the Republic of Colombia indicated a small number (5 percent) of turtles with the haplotype CmP22, which was recently discovered to be common in nesting green turtles from the Marshall Islands and American Samoa (Dutton et al., 2014). This shows that, despite the isolation of nesting females between the eastern and western Pacific, a small number of immature turtles successfully cross the Pacific during developmental migrations in both directions. However, it is important to point out that there is no evidence of mature turtles inhabiting foraging or nesting habitat across the Pacific from their region of origin.

Recent nDNA studies provide insights that are consistent with patterns of differentiation found with mtDNA in the eastern Pacific. Roden et al. (2013) found significant differentiation between FFS and two eastern Pacific populations (the Galápagos Islands, Ecuador and Michoacán, Mexico) and greater connectivity between Galapagos and Michoacán than between FFS and either of the eastern Pacific nesting sites.

Flipper tagging and satellite telemetry data show that dispersal and reproductive migratory movements of e.g., Seminoff et al., 2002b). There were two apparent groupings, with tags attached to turtles nesting in the Galápagos largely recovered along the shores from Costa Rica to Chile in the southeastern Pacific, and long-distance tag returns from the Michoacán nesting site primarily from foraging areas in Mexico to Nicaragua. However, there was a small degree of overlap between these two regions, as at least one Michoacán tag was recovered as far south as Colombia (Alvarado-Díaz and Figueroa, 1992).

Satellite telemetry efforts with green turtles in the region have shown similar results to those for flipper tag recoveries. A total of 23 long-distance satellite tracks were considered for the Status Review (Seminoff, 2000; Nichols, 2003; Seminoff et al., 2008). Satellite data show that turtles tracked in northeastern Mexico (Nichols, 2003; J. Nichols, California Academy of Sciences, unpubl. data) and California (P. Dutton, NMFS, pers. comm., 2010) all stayed within the region, whereas turtles tracked from nesting beaches in the Galápagos Islands all remained in waters off Central America and the broader southeastern Pacific Ocean (Seminoff et al., 2008).

Demographic evidence of discreteness is also found in morphological differences between green turtles in the eastern Pacific and those found elsewhere. The smallest green turtles worldwide are found in the eastern Pacific, where mean nesting size is 82.0 cm CCL in Michoacán, Mexico (n=718, (Alvarado-Díaz and Figueroa, 1992) and 86.7 cm CCL in the Galápagos (n=2708; (Zárate et al., 2003), compared to the 95 cm to 110 cm CCL size range for most green turtles. In addition, Kamezaki and Matsui (1995) found differences in skull morphology among green turtle populations on a broad global scale when analyzing specimens representing west and east Pacific (Japan and Galápagos), Indian Ocean (Comoros and Seychelles), and Caribbean (Costa Rica and Guyana) populations. The eastern Pacific was different from others based on discriminant function analysis (used to discriminate between two or more naturally occurring groups).

Given the information presented above, the SRT concluded, and we concur, that there are five discrete populations entirely within the Pacific Ocean: (1) Central West Pacific, (2) Southwest Pacific, (3) Central South Pacific, (4) Central North Pacific, and (5) East Pacific. These five populations are markedly separated from each other and from populations within the Atlantic Ocean and Indian Oceans as a consequence of physical, ecological, behavioral, and oceanographic factors. Information supporting this conclusion includes genetic analysis, flipper tag recoveries, and satellite telemetry.

Collectively, all observations above led the SRT to propose that green turtles from the following geographic areas might be considered “discrete” according to criteria in the joint DPS policy:

(1) North Atlantic Ocean (2) Mediterranean Sea (3) South Atlantic Ocean (4) Southwest Indian Ocean (5) North Indian Ocean (6) East Indian Ocean-West Pacific Ocean (7) Central West Pacific Ocean (8) Southwest Pacific Ocean (9) Central South Pacific Ocean (10) Central North Pacific Ocean (11) East Pacific Ocean B. Significance Determination

In accordance with the DPS Policy, the SRT next reviewed whether the population segments identified in the discreteness analysis were biologically and ecologically significant to the taxon to which they belong, which is the taxonomic species C. mydas. Data relevant to the significance question include ecological, behavioral, genetic and morphological data. The SRT considered the following factors, listed in the DPS Policy, in determining whether the discrete population segments were significant: (1) Evidence that loss of the discrete segment would result in a significant gap in the range of the taxon; (2) evidence that the discrete segment differs markedly from other populations of the species in its genetic characteristics; and (3) persistence of the discrete segment in an unusual or unique ecological setting. The DPS policy also allows for consideration of other factors if they are appropriate to the biology or ecology of the species, such as unique morphological or demographic characteristics, and unique movement patterns.

1. North Atlantic

Green turtles in the North Atlantic differ markedly in their genetic characteristics from other regional populations. They are strongly divergent from the Mediterranean population (the only other population within Clade I), and turtles from adjacent populations in the eastern Caribbean carry haplotypes from a different clade. The North Atlantic population has globally unique haplotypes. Therefore, the loss of the population would result in significant genetic loss to the species as a whole.

The green turtles within the North Atlantic population occupy a large portion of one of the major ocean basins in the world; therefore, the loss of this segment would represent a significant gap in the global range of green turtles. Green turtles take advantage of the warm waters of the Gulf Stream to nest in North Carolina at 34° N., which is farther from the equator than any other nesting sites outside the Mediterranean Sea. Tagging and telemetry studies show that the North Atlantic green turtle population has minimal mixing with populations in the South Atlantic and Mediterranean regions. The mean size of nesting females in the North Atlantic, which could reflect the ecological setting and/or be genetically based, is larger (average 101.7-109.3 cm CCL; (Guzmán-Hernández, 2001, 2006) than those in the adjacent Mediterranean Sea (average 88-96 cm CCL), and smaller than those at varying locations in the South Atlantic, such as those at Isla Trindade, Brazil (average 115.2 cm CCL; Hirth, 1997; Almeida et al., 2011), Atol das Rocas, Brazil (112.9-118.6 cm CCL; Hirth, 1997; Bellini et al., 2013), and Ascension Island (average 116.8 cm CCL; Hirth, 1997).

Another factor indicating uniqueness of the North Atlantic population is a typical 2-year remigration interval, as compared to 3-year or longer intervals that are more common elsewhere (Witherington et al., 2006).

2. Mediterranean

Mediterranean turtles differ markedly in their genetic characteristics from other regional populations, with globally unique haplotypes and strong divergence from the other population within Clade I (the North Atlantic population). Therefore, the loss of the population would result in significant genetic loss to the species as a whole. Given this genetic distinctiveness and the distinctive environmental conditions, it is likely that turtles from the eastern Mediterranean have developed local adaptations that help them persist in this area. Mediterranean females are smaller than those in any other regional population except the Eastern Pacific, averaging 92.0 cm CCL (Broderick et al., 2003) compared to the global average of 95 cm-110 cm CCL.

The loss of the population would result in a significant gap in the range

Finally, the Mediterranean Sea appears to be a unique ecological setting for the species. It is the most saline marine water basin in the world (38 parts per thousand (ppt) or higher), is nearly enclosed, and is outside the normal latitudinal range for the species, being the farthest from the equator of any green turtle population. Although similar information is not available for green turtles, it has been postulated that the high salinity of sea water in the Mediterranean acts as a “barrier” preventing loggerhead sea turtles from moving among the areas of the Western Mediterranean, explaining why they do not mix between the north and south Mediterranean as juveniles (Revelles et al., 2008). All nesting sites within the Mediterranean are between latitudes 31-40° N., which not only affects temperature but results in more seasonal variation in day length and environmental conditions, which may have fostered local adaptations in green turtles living there.

3. South Atlantic

The South Atlantic population has globally unique haplotypes. Therefore, the loss of the population would result in significant genetic loss to the species as a whole. The South Atlantic population contains the only nesting site in the world associated with a mid-ocean ridge. This unique ecological setting at Ascension Island, one of the largest nesting sites within this population, ensures diverse nesting habitats and promotes resilience for the species. This population spans an entire hemispheric ocean basin, and its loss would result in a gap of at least 12,000 km between populations off southeast Africa and those in Florida, clearly a significant gap in the range of the taxon. Brazil and Guinea Bissau may have acted as a refuge for Atlantic green turtles during the Pleistocene period (Reece et al., 2005). The average size of nesting females is larger here than in any other populations, ranging from 112.9-118.6 cm CCL (Hirth, 1997; Almeida et al., 2011) compared to 95-110 cm CCL worldwide, which could reflect an adaptation to local environmental conditions such as habitat, availability of food, water temperature, and population dynamics.

4. Southwest Indian

Within the Southwest Indian Ocean, strong upwelling in the Mozambique Channel produces distinctive areas of high productivity that support a robust turtle population, and complex current patterns in the area create a distinctive ecological setting for green turtles. Madagascar is one of the largest islands in the world and its proximity to the African coast, along with a proliferation of nearby islands, creates a complex series of habitats suitable for green turtles. Loss of this population would leave a gap of over 10,000 km between populations in southern India and those in west-central Africa. Nesting turtles from this population are the largest within the Indian Ocean, ranging from 103 cm (SCL)-112.3 cm (CCL) (Frazier, 1971; 1985) which could reflect growth due to presence of a network of foraging areas and localize migratory movements.

5. North Indian

The ecological setting for this region is unique for green turtles in that it contains some of the warmest and highly saline waters in the world, indicative of the partially enclosed marine habitats within this system. The salinity in the North Indian Ocean varies from 32 to 37 ppt comparable only to the Mediterranean Sea. Salinity in this region varies with local and seasonal differences particularly in the Arabian Sea (dense, high-salinity) and the Bay of Bengal (low-salinity). Although genetic data are very limited for this population, with the only sample being from the Persian Gulf, it has two groups of highly divergent haplotypes that are not found anywhere else in the world ( i.e., markedly different genetic characteristics). The loss of this population, and its globally unique haplotypes, which are not found in any other population, would result in significant genetic loss to the species as a whole. This population is isolated from other Indian Ocean populations which would render its loss a significant gap in the range of the species. Nesting turtles are smaller here than in other Indian Ocean regions, possibly reflecting genetic adaptations to local environmental conditions.

6. East Indian-West Pacific

This area of complex habitats at the confluence of the tropical Indian and Pacific Oceans is a well-known hotspot for speciation and diversification of both terrestrial and marine taxa. It is unique in that it contains the most extensive continental shelf globally, and particularly low salinity waters in the northeastern Indian Ocean. Loss of green turtles from this vast area would create a substantial gap in the global distribution and, because this population is located at the center of the species' range, would strongly affect connectivity within the species as a whole. Connectivity is important for the maintenance of genetic diversity and resilience of the species. Genetic data indicate the presence of ancestral haplotypes with significant mtDNA diversity. The loss of this population, and its ancestral haplotypes, would represent a significant genetic loss to the species. The wide size range of nesting females within this population (82.1 cm-105.6 cm; Charuchinda and Monanunsap, 1998; Cheng, 2000) is also an indication of the high level of diversity within this population.

7. Central West Pacific

The Central West Pacific population is genetically significant in that it has both globally unique haplotypes and ancestral haplotypes. The Central West Pacific has no continental shelf habitats, with all nesting occurring on small islands or atolls that are volcanic or coralline limestone. There is an apparent oceanic boundary between the Central West Pacific and the Central North Pacific population and an apparent biogeographic boundary between the Central West Pacific and the East Indian-West Pacific population. Loss of turtles from this population would create a large gap near the center of the geographic range of the species.

8. Southwest Pacific

Clade V haplotypes have only been found at nesting sites in the Southwest Pacific population. In addition to these globally unique haplotypes, the presence of the ancestral haplotypes and significant mtDNA diversity make this population genetically significant.

Unlike most other populations in the Pacific Ocean, this population includes island nesting sites in close proximity to coastal foraging areas. The Great Barrier Reef (GBR) is the largest coral reef system in the world and was periodically isolated over geological time. It provides expansive, year-round foraging habitat for green turtles and supports one of the largest nesting sites in the world.

9. Central South Pacific

This population has globally unique haplotypes. Therefore, the loss of the population would result in significant genetic loss to the species as a whole. To a greater extent than in any other regional population, nesting sites are widely dispersed among a large number of small habitats on islands and atolls. Foraging areas are mostly coral reef ecosystems, with seagrass beds in Tonga and Fiji being a notable exception.

10. Central North Pacific

Mitochondrial DNA in this extensively sampled region includes globally unique haplotypes. Although two haplotypes are shared with individuals in the Revillagigedos Islands in the East Pacific, there is little evidence of significant ongoing gene flow. The loss of this population would result in significant genetic loss to the species as a whole.

This population has no continental-shelf habitat and all nesting occurs on mid-basin pinnacles. Turtles in this population are known to bask, a rare behavior for modern-day sea turtles, and have unique morphological traits such as unusually large flippers, possibly reflecting adaptations to their ecological setting. This is the most isolated of all populations, with an apparent biogeographic boundary with the Eastern Pacific population and oceanic boundaries with the Central West and Central South Pacific populations. If all turtles were lost from this vast geographic area, it would create a significant gap in the global range of the species.

11. East Pacific

The two cold-water currents on the east side of the Pacific Ocean (the Humboldt Current in the south and the California Current in the north) leave a distinctive region of tropical ocean along the west coasts of Mexico, Central America, and northern South America that is known as the Eastern Pacific Zoogeographic Region (Briggs, 1974). Perhaps as a result, some turtles in this area exhibit a unique overwintering behavior similar to hibernation. This area also has a very narrow continental shelf and low levels of seagrass, resulting in a unique diet for green turtles ( e.g., tunicates and red mangrove fruits; Amorocho and Reina, 2007). This population has globally unique haplotypes. Therefore, the loss of the population would result in significant genetic loss to the species as a whole. Mean size of nesting turtles in the East Pacific is smaller, at approximately 82 cm CCL (Pritchard, 1971) than in any other population, which could reflect an adaptation to local ecological conditions, as could the distinctive “black” phenotype. The Galapagos Island chain is one of the few areas where green turtles bask (Hawai`i being the other). Loss of all turtles from this population would leave a significant gap in the range of the species as it occurs along much of the eastern boundary of the world's largest ocean.

C. Summary of Discreteness and Significance Determinations

In summary, the 11 discrete populations identified in the Discreteness Determination section were also determined to be significant to the species, C. mydas. Each is genetically unique, and many are identified by unique mtDNA haplotypes which could represent adaptive differences. Some populations exist in unique or unusual ecological settings influenced by local ecological and physical factors which may also lead to adaptive differences and represent adaptive potential. Some also possess unique morphological or other demographic characteristics that render them significant. Most populations represent a large portion of the species' range, and their loss would result in a significant gap in the range of the species.

Based on the information provided in the Discreteness Determination and Significance Determination sections above, the SRT identified the following 11 potential green turtle DPSs (Figure 2): (1) North Atlantic, (2) Mediterranean, (3) South Atlantic, (4) Southwest Indian, (5) North Indian, (6) East Indian-West Pacific, (7) Central West Pacific, (8) Southwest Pacific, (9) Central South Pacific, (10) Central North Pacific, and (11) East Pacific. We concur with the findings of the SRT and conclude that the 11 potential DPSs identified by the SRT warrant delineation as DPSs.

EP23MR15.001

VI. Listing Evaluation Process

A. Discussion of Population Parameters for the Eleven Green Turtle DPSs

In these sections, we describe the geographic range of each DPS. We discuss its population parameters, which are derived from population data and influence the persistence of the DPS. These population parameters include: Abundance, growth rates or trends, spatial structure, and diversity or resilience (McElhany et al., 2000). NMFS has used this approach in numerous status reviews. USFWS uses a similar approach, based on Shaffer and Stein (2000), to evaluate a species' status in terms of its representation, resiliency, and redundancy; this methodology has also been a widely accepted approach (Tear et al., 2005). Though expressed differently, these two approaches rely on the same conservation biology principles. Though this information is presented separately from the assessment of threats under section 4(a)(1) of the ESA, population dynamics represent one aspect of the other natural or manmade factors affecting the continued existence of the species that we consider under Factor E.

Complete population abundance and trend estimates do not exist for any of the 11 DPSs. The data used in the Status Review and summarized here represent the best scientific information available. The data are more robust for some areas than for others. For each DPS, the primary data available are collected on nesting beaches, either as counts of nests or counts of nesting females, or a combination of both (either direct or extrapolated). Information on abundance and trends away from the nesting beaches is limited and often non-existent, primarily because these data are, relative to nesting beach studies, logistically difficult and expensive to obtain. Therefore, the primary and best available information source for directly evaluating status and trends of the DPSs is nesting data.

Nesting female abundance estimates for each nesting site or nesting beach are presented in the Status Review for each potential DPS. Accompanying this information is trend information in the form of bar plots and Population Viability Analysis (PVA) models extending 100 years into the future for the 33 sites that met the criteria for depicting the data this way, i.e., recent (<10 year old) data over a given period of time (10 years for bar plots, 15 years for PVA) with consistent protocols and effort during that time.

With regard to spatial structure, the SRT used information from genetic, tagging, telemetry, and demographic data to identify structuring and substructuring within each DPS. This informed the SRT of metapopulation dynamics in order that it might consider these dynamics in considerations about the future of the species, including whether source populations and genetic diversity are being maintained.

With regard to diversity and resilience, the SRT considered the extent of ecological variation, including the overall nesting spatial range, diversity in nesting season, and diversity of nesting site structure and orientation, e.g., whether nesting sites are insular or continental, have a high or low beach face, and whether there are a variety of types of sites. The SRT also considered demographic and genetic diversity of the DPS which may indicate its ability to adapt and thus its resilience. One of the considerations when looking at diversity was the DPS's ability to adapt to climate change including, but not limited to, sea level rise and warming of nesting beaches.

B. Summary of Factors Affecting the Eleven Green Turtle DPSs

Section 4 of the ESA (16 U.S.C. 1533) and implementing regulations at 50 CFR part 424 set forth procedures for adding species to the Federal List of Endangered and Threatened Wildlife Species. Under section 4(a) of the ESA, the Services must determine whether a species is threatened or endangered because of any of the following 5 factors: (A) The present or threatened destruction, modification, or curtailment of its habitat or range; (B) overutilization for commercial, recreational, scientific, or educational purposes; (C) disease or predation; (D) the inadequacy of existing regulatory mechanisms; or (E) other natural or manmade factors affecting its continued existence.

In this rulemaking, information regarding the status of each of the 11 green turtle DPSs is considered in relation to the five factors provided in section 4(a)(1) of the ESA. That information presented here is a summary of the information in the Status Review. The reader is directed to the subsection within each DPS section of the Status Review titled “Analysis of Factors Listed Under ESA Section 4(a)(1)” for a more detailed discussion of the factors.

C. Conservation Efforts

In evaluating the efficacy of protective efforts not yet implemented or not yet proven to be effective, we rely on the Policy on Evaluation of Conservation Efforts When Making Listing Decisions (“PECE”; 68 FR 15100, March 28, 2003), issued jointly by the Services. Information on conservation efforts for each DPS is summarized from the Status Review. For a more detailed description of conservation efforts, please see that document. When assessing conservation efforts, the SRT assumed that all conservation efforts would remain in place at their current levels. In our final determinations, we considered the conservation benefits of continued protections under the ESA.

D. Extinction Risk Assessments and Findings

To analyze the extinction risk of each DPS, the SRT collected and presented information on the six critical assessment elements: (1) Abundance, (2) growth rates/trends, (3) spatial structure, (4) diversity/resilience, (5) five factor analysis/threats, and (6) conservation efforts. Shortly after each presentation, the SRT voted twice: A vote on the contribution of each critical assessment element to extinction risk, and a vote on the overall risk of extinction to the DPS (see section 3.3.4 of the Status Review for a more detailed discussion of this process).

In the first vote, SRT members ranked the importance of each of the four population parameters (Abundance, Trends, Spatial Structure, Diversity/Resilience) by assigning them a value from 1 to 5 for each DPS, with 1 indicating a very low risk and 5 indicating a very high risk. SRT members then ranked the influence of the section 4(a)(1) factors (threats) on the status of each DPS by assigning a value of 0 (neutral effect on status—this could mean that threats are not sufficient to appreciably affect the status of the DPS, or that threats are already reflected in the population parameters), -1 (threats described in the 5-factor analysis suggest that the DPS will experience some decline (<5 percent decline) in abundance within 100 years), or -2 (threats described in the 5-factor analysis suggest that the DPS will experience significant decline (≥5 percent decline) in abundance within 100 years). They then ranked the influence of conservation efforts on the status of each DPS by assigning a value of 0 (neutral effect on status—this could mean that conservation efforts are not sufficient to appreciably affect the status of the DPS, or that conservation efforts are already reflected in the population parameters), +1 (activities described in Conservation Efforts suggest that the DPS will experience <5 percent increase in abundance within 100 years), or +2 (activities described in Conservation Efforts suggest that the DPS will experience ≥5 percent increase in

In the second vote, SRT members provided their expert opinion (via vote) on the likelihood that each DPS would reach a critical risk threshold (quasi-extinction) within 100 years. In the Status Review, the SRT defined the critical risk threshold (quasi-extinction) as follows: “A DPS that has reached a critical risk threshold has such low abundance, declining trends, limited distribution or diversity, and/or significant threats (untempered by significant conservation efforts) that the DPS would be at very high risk of extinction with little chance for recovery.” Generally, DPSs were considered to have higher viability if they were composed of a number of relatively large populations, distributed throughout the geographic range of the DPS, and exhibited stable or increasing growth rates. DPSs were considered to be at higher risk if they were composed of fewer robust populations or with robust populations all concentrated in a small geographic area, where they might be susceptible to correlated catastrophes. Any DPS with low phenotypic and/or habitat diversity were also considered to be at higher risk because the entire DPS could be vulnerable to persistent environmental conditions (Limpus and Nicholls, 2000; Saba et al., 2008; Van Houtan and Halley, 2011) or stochastic catastrophic events (Hawkes et al., 2007; Van Houtan and Bass, 2007; Fuentes et al., 2011).

Each member was given 100 points to spread across risk categories, reflecting their interpretation of the information for that DPS; the voting results are available in the Status Review. The spread of points is meant to reflect the amount of uncertainty in the risk threshold bins. Risk categories were <1 percent, 1-5 percent, 6-10 percent, 11-20 percent, 21-50 percent, and >50 percent. We note that, presumably because this species is such a long-lived species and, as such, it is unlikely that it would go extinct within 100 years even if it was lost in many places, every DPS received numerous points in the <1 percent category, including those with the most depressed numbers and that face the highest threats.

As noted above, the SRT estimated the likelihood that a population would fall below a critical risk threshold within 100 years. The SRT did not define the critical risk threshold quantitatively but instead provided the following definition: “A DPS that has reached a critical risk threshold has such low abundance, declining trends, limited distribution or diversity, and/or significant threats (untempered by significant conservation efforts) that the DPS would be at very high risk of extinction with little chance for recovery.”

While the SRT's review of the DPSs' statuses was rigorous and extensive, the framework used does not allow us to easily or clearly translate a particular critical risk category to an ESA listing status. Structured expert opinion is a valid and commonly used method of evaluating extinction risk and forms a useful starting point for our analysis. However, in our judgment, the critical risk threshold approach used for this status review does not directly correlate with the ESA's definitions of endangered and threatened. The ESA defines an “endangered species” as “any species which is in danger of extinction throughout all or a significant portion of its range.” The critical risk threshold, as defined by the SRT, is a condition worse than endangered, because it essentially precludes recovery. Thus, while the SRT votes informed our listing determinations, we did not equate a particular critical risk category with an ESA listing status, and therefore the votes were not the basis for those determinations. However, to make our proposed listing determinations, we applied the best available science that was compiled by the SRT in examining the definitions of endangered and threatened species under section 3 of the ESA.

After considering the extinction risk, the Services then reviewed the present threats and threats anticipated in the foreseeable future for each DPS. We examined the significant threats to each DPS, how these threats affected that DPS, and how they were predicted to affect the DPS in the foreseeable future. Our analysis weighed each factor within the scope of the ESA's definitions of threatened and endangered for each DPS.

Among other things, the Services also carefully considered where current conditions or protections are present specifically because green turtles are listed under the ESA, and whether those conditions would likely exist absent such a listing. We note that the latter was not considered by the SRT, meaning the SRT conducted all risk analyses assuming all protections would remain in place.

VII. North Atlantic DPS

A. Discussion of Population Parameters for the North Atlantic DPS

The range of the North Atlantic DPS extends from the boundary of South and Central America north along the coast to the northern extent of the green turtle's range to include Panama, Costa Rica, Nicaragua, Honduras, Belize, Mexico, and the United States. It then extends due east across the Atlantic Ocean at 48° N.; follows the coast south to include the northern portion of the Islamic Republic of Mauritania (Mauritania; to 19° N.) on the African continent; and west along the 19° N. latitude to the Caribbean basin, turning south and west at 63.5° W., 19° N., and due south at 7.5° N., 77° W. to the boundary of South and Central to include Puerto Rico, the Bahamas, Cuba, Turks and Caicos Islands, Republic of Haiti (Haiti), Dominican Republic, Cayman Islands, and Jamaica. The North Atlantic DPS includes the Florida breeding population, which was originally listed as endangered (43 FR 32800, July 28, 1978). Critical habitat was previously designated for areas within the range of this DPS ( i.e., coastal waters surrounding Culebra Island, Puerto Rico; 63 FR 46693, September 2, 1998).

Green turtle nesting sites in the North Atlantic are some of the most studied in the world, with time series exceeding 40 years in Costa Rica and 35 years in Florida. Seventy-three nesting sites were identified within the North Atlantic DPS, although some represent numerous individual beaches. For instance, Florida nesting beaches were listed by county with the numerous beaches in each county representing one site and, for other U.S. beaches (from Texas to North Carolina), each state's nesting beaches were represented as one site. There are four regions that support high density nesting concentrations for which data were available: Tortuguero, Costa Rica; Mexico (Campeche, Yucatan, and Quintana Roo); Florida, United States; and Cuba. There is one nesting site with >100,000 nesting females (Tortuguero at 131,751; Chaloupka et al., 2008a; Sea Turtle Conservancy, 2013), one with 10,001-100,000 (Quintana Roo, Mexico at 18,257; Julio Zurita, pers. comm. 2012) and six with 1,001-5,000: Cayo Largo, Cuba; Campeche, Yucatan, and Veracruz, Mexico; and Brevard and Palm Beach Counties, FL, United States. There are four with 501-1,000; Tamaulipas, Mexico; Vieques, Puerto Rico; Martin and Indian River Counties,

Of the nesting sites with long-term data sets, both Tortuguero and the index beaches in Florida exhibit a strong positive trend in the PVAs that were conducted on them, as does Isla Aguada, Mexico (one beach in the Campeche group). Three beaches in Cuba (total of 489 nesting females) either showed no trend or a modest positive trend. One beach in Mexico (El Cuyo, Yucatan) exhibited no trend.

Genetic sampling in the North Atlantic DPS has been generally extensive with good coverage of large populations in this region; however, some smaller Caribbean nesting sites are absent and coastal nesting sites in the Gulf of Mexico are under-represented. Genetic differentiation based on mtDNA indicated that there are at least four independent nesting subpopulations in the North Atlantic DPS characterized by shallow regional substructuring: (1) Florida (Hutchinson Island; Lahanas et al., 1994), (2) Cuba (Guanahacabibes Península and Cayería San Felipe; Ruiz-Urquiola et al. , 2010), (3) Mexico (Quintana Roo; Encalada et al., 1996), and (4) Costa Rica (Tortuguero; Lahanas et al., 1994). These nesting sites are characterized by common and widespread haplotypes dominated by CM-A1 and/or CM-A3. A relatively low level of spatial structure is detected due to shared common haplotypes, although there are some rare/unique haplotypes at some nesting sites. Connectivity may indicate recent shared common ancestry.

Green turtles nest on both continental and island beaches throughout the range of the DPS (Witherington et al., 2006). Major nesting sites are primarily continental with hundreds of lower density sites scattered throughout the Caribbean. Green turtles nesting in Florida seem to prefer barrier island beaches that receive high wave energy and that have coarse sands, steep slopes, and prominent foredunes. The greatest nesting is on sparsely developed beaches that have minimal levels of artificial lighting. A high-low nesting pattern for Florida and Mexico occurs during the same years; however, nesting in Tortuguero, Costa Rica is not always in sync with Florida and Mexico ( e.g., 2011 was a high nesting year in Florida, but for Tortuguero the high nesting year was 2010). The nesting season is similar throughout the range of the DPS, with green turtles nesting from June to November in Costa Rica (Bjorndal et al., 1999), and May through September in the United States, Mexico, and Cuba (Witherington et al., 2006).

B. Summary of Factors Affecting the North Atlantic DPS

1. Factor A: The Present or Threatened Destruction, Modification, or Curtailment of Its Habitat or Range

a. Terrestrial Zone

Within the range of the North Atlantic DPS, nesting beaches continue to be degraded from a variety of activities. Destruction and modification of green turtle nesting habitat results from coastal development, coastal armoring, beachfront lighting, erosion, sand extraction, and vehicle and pedestrian traffic on nesting beaches (Witherington and Bjorndal, 1991; Witherington, 1992; Witherington et al., 1996; Lutcavage et al., 1997; Bouchard et al., 1998; Mosier, 1998; Witherington and Koeppel, 2000; Mosier and Witherington, 2002; Leong et al., 2003; Roberts and Ehrhart, 2007). In addition, sea level rise resulting from climate change poses a threat to all nesting beaches. Portions of the Southern United States and Caribbean are found be to highly vulnerable to sea level rise (Melillo et al., 2014). For instance, along the southern portion of the Florida coastline, one climate change model predicted one meter of sea level rise by 2060, resulting in the inundation of more than 50 percent of coastal wildlife refuges (Flaxman and Vargas-Moreno, 2011). Most green turtle nesting in the United States is concentrated along the southeastern coast of Florida with more than 90 percent of nesting occurring from Brevard to Broward counties ( http://ocean.floridamarine.org/SeaTurtle/nesting/FlexViewer /). Loss of nesting habitat as a result of sea level rise poses a threat to the population. Sea level rise is exacerbated by coastal development and armoring, which prevents the beach from migrating and causes nesting green turtles to abandon their nesting attempts more frequently as a result of their encounter with such structures (Mosier, 1998; Mosier and Witherington, 2000; Rizkalla and Savage, 2011). Females might nest in sub-optimal habitats, where nests are more vulnerable to erosion or inundation (Rizkalla and Savage 2011). As a result, nests would be subject to more frequent inundation, exacerbated erosion, and increased moisture from tidal overwash, which can potentially alter thermal regimes, an important factor in determining the sex ratio of hatchlings.

b. Neritic/Oceanic Zones

Green turtles in the post-hatchling and early-juvenile stages are closely associated with Sargassum algae in the Atlantic and Gulf of Mexico (Witherington et al., 2012), and vulnerable to ingesting contaminants such as tar balls and plastics that aggregate in convergent zones where Sargassum aggregates (Witherington, 2002). Juvenile and adult green turtles and their nearshore foraging habitats are also exposed to high levels of pollutants, such as agricultural and residential runoff, and sewage which result in degraded foraging habitat (Smith et al., 1992). Further, increased nutrient load in these coastal waters causes eutrophication. Eutrophication is linked to harmful algal blooms that result in the loss and degradation of seagrass beds, and possibly fibropapilloma tumors in green turtles (Milton and Lutz, 2003).

In Cuba, Jamaica, Puerto Rico, and Panama, water quality is also affected by sewage and industrial and agricultural runoff. Pollution remains a major threat in the waters of Jamaica. Major sources of pollution are industrial and agricultural effluent, garbage dumps and solid waste, and household sewage (Greenway, 1977; Green and Webber, 2003).

Nearshore foraging habitats such as seagrass beds are affected by propeller scarring, anchor damage, dredging, sand mining, and marina construction throughout the range of the DPS (Smith et al., 1992; Dow et al., 2007; Patrício et al., 2011). Sand placement projects along the Florida coastline affect nearshore reefs as a result of direct burial of portions of the reef habitat and loss of food sources available to green turtles (Lindeman and Snyder, 1999).

The SRT found, and we concur, that the North Atlantic DPS of the green turtle is negatively affected by ongoing changes in both its terrestrial and marine habitats as a result of land and water use practices as considered above in Factor A. The increasing threats to the terrestrial and marine habitats are not reflected in the current trend for the North Atlantic DPS, as it was based on nesting numbers and not on all current life stages. These increasing threats to the population will become apparent when those life stages affected by the threats return to nest, as the trend information is based solely on numbers of nests. This lag time was considered in our analysis. For example, a threat that affects the oceanic juvenile phase would not be detected until those turtles return to nest, approximately 15 to 20 years later. The SRT also found, and we concur, that coastal development, beachfront lighting, erosion, sand extraction, and sea level rise increasingly impact nesting beaches of

2. Factor B: Overutilization for Commercial, Recreational, Scientific, or Educational Purposes

A partial list of the countries within the range of the North Atlantic DPS where ongoing intentional capture of green turtles occurs, includes Costa Rica (Mangel and Troëng, 2001; Gonzalez Prieto and Harrison, 2012), Mexico (Seminoff, 2000; Gardner and Nichols, 2001; Dirado et al., 2002; Guzmán-Hernández and García Alvarado, 2011), Cuba (Fleming, 2001; F. Moncado, Ministerio de la Industria Pesquera, pers. comm., 2013), Nicaragua (Lagueux, 1998; Humber et al., 2014), the Bahamas (Fleming, 2001), Jamaica (Haynes-Sutton et al., 2011), and the Cayman Islands (Fleming, 2001). Harvest remains legal in several of these countries (Humphrey and Salm, 1996; Wamukoya et al. , 1996; Fleming, 2001; Fretey, 2001; Bräutigam and Eckert, 2006).

The commercial artisanal green turtle fishery in Nicaragua continues to be a threat to the Tortuguero nesting population, the largest remaining green turtle population in the Atlantic (Campbell and Lagueux, 2005). Local demand for turtle meat in coastal communities continues (Garland and Carthy, 2010). There is a legal turtle fishery on the Caribbean coast that is located in the most important developmental and foraging habitat for Caribbean green turtles (Fleming, 2001; Campbell and Lagueux, 2005). The hunting of juvenile and adult turtles continues both legally and illegally in many foraging areas where green turtles originating from Florida nesting beaches are known to occur (Chacón, 2002; Fleming, 2001).

Direct take of eggs is also an ongoing threat in Panama (Evans and Vargas, 1998). Green turtles nesting on Belize's beaches and foraging along its coast are harvested in the Robinson Point area and sold in markets and restaurants (Searle, 2003). Large numbers of green turtles are captured in the area southeast of Belize, an area which may be an important migratory corridor (Searle, 2004). There are important feeding grounds in the Banc d'Arguin, Mauritania. While the frequency of green turtle nesting in Mauritania is not known, green turtle nests are reported as being harvested there (Fretey, 2001; Fretey and Hama, 2012).

Commercial harvest of green turtles was a factor that contributed to the historic decline of this DPS. Current harvest of green turtles and eggs, in a portion of this DPS, continues to be significant threat to the persistence of this DPS.

3. Factor C: Disease or Predation

Fibropapillomatosis (FP) has been found in green turtle populations in the United States (Hirama, 2001; Ene et al., 2005; Foley et al., 2005; Hirama and Ehrhart, 2007), the Bahamas, the Dominican Republic, Puerto Rico (Dow et al. , 2007; Patrício et al., 2011), Cayman Islands (Wood and Wood, 1994; Dow et al., 2007), Costa Rica (Tortuguero; Mangel and Troëng, 2001), Cuba (Moncada and Prieto, 2000), Mexico (Yucatan Peninsula; K. Lopez, pers. comm., as cited in MTSG, 2004), and Nicaragua (Lagueux, 1998).

FP continues to be a major problem in some lagoon systems and along the nearshore reefs of Florida. It is a chronic, often lethal disease occurring predominantly in green turtles (Van Houtan et al., 2014). A correlation appeared to exist between these degraded habitats and the prevalence of FP in the green turtles that forage in these areas but no direct link was established (Aguirre and Lutz, 2004; Foley et al., 2005). Indeed, across green turtle populations, it is widely observed that FP occurs most frequently in eutrophied and otherwise impaired waterways (Herbst, 1994; Van Houtan et al., 2010). A recent study establishes that eutrophication substantially increases the nitrogen content of macroalgae, thereby promoting the latent herpes virus which causes FP tumors in green turtles (Van Houtan et al., 2014) although it is argued that there is no inferential framework to base this conclusion (Work et al., 2014). Despite the high incidence of FP among foraging populations, there is no conclusive evidence on the effect of FP on reproductive success (Chaloupka and Balazs, 2005).

Harmful algal blooms, such as a red tide, also affect green turtles in the North Atlantic DPS. In Florida, the species that causes most red tides is Karenia brevis, a dinoflagellate that produces a toxin (Redlow et al., 2002). Since 2007, there were two red tide events, one in 2007 along the east coast of Florida, and one in 2012 along the west coast of Florida. Sea turtle stranding trends indicated that these events were acting as a mortality factor (A. Foley, Florida Fish and Wildlife Conservation Commission, pers. comm., 2013). These events may impact a population's present and future reproductive status.

Predators such as raccoons ( Procyon lotor ), feral hogs ( Sus scrofa ), foxes ( Urocyon cinereoargenteus and Vulpes vulpes ), and coyotes ( Canis latrans ) may take significant numbers of turtle eggs (Stancyk, 1982; Allen et al., 2001). Nest protection programs are in place at most of the major nesting beaches in the North Atlantic DPS, although they are managed at varying levels and degrees of effectiveness (Engeman et al., 2005). Predator species that are particularly difficult to manage include red fire ants ( Solenopsis invicta ) and jaguars ( Panthera onca ) (Wetterer, 2006; Prieto and Harrison, 2012).

Although FP disease is of major concern, with increasing levels in some green turtle populations in this DPS, it should be noted there is uncertainty of the long-term survivability and effect on the reproductive effort of the population. Predation is known to occur throughout this DPS, and we find it to be a significant threat to this DPS in the absence of well managed nest protection programs.

4. Factor D: Inadequacy of Existing Regulatory Mechanisms

At least 15 regulatory mechanisms that apply to green turtles regionally ( e.g., U.S. Magnuson-Stevens Fishery Conservation and Management Act) or globally ( e.g., Convention on International Trade in Endangered Species of Wild Fauna and Flora) apply to green turtles within the North Atlantic Ocean. The analysis of these existing regulatory mechanisms assumed that all would remain in place at their current levels.

In the United States, regulatory mechanisms that protect green turtles are in place and include State, Federal, and international laws. The green turtle was listed under the ESA in 1978, providing relatively comprehensive protection and recovery activities to minimize the threats to green turtles in the United States. Considering the dependence of the species on conservation efforts, significant concerns remain regarding the inadequacy of regulatory mechanisms. The development and implementation of Turtle Excluder Devices (TEDs) in the shrimp trawl fishery was likely the most significant conservation accomplishment for North Atlantic green turtles in the marine environment since their 1978 ESA listing. In the southeast United States and Gulf of Mexico, TEDs have been mandatory in shrimp and flounder trawls for over a decade. These regulations are implemented and enforced to varying degrees throughout the Gulf and U.S. Southeast Atlantic. For example, the State of Louisiana prohibits enforcement of TED regulations and tow time limits. In other States, enforcement of TED regulations depends on available et al., 2014). There are also regulatory mechanisms in place that address the loss of nesting habitat, such as the Florida Administrative Code Rule 62B-33.0155, which addresses threats from armoring structures. However, these regulatory mechanisms allow for variances and armoring permits continue to be issued along nesting beaches.

Other threats, such as light pollution on nesting beaches, marine debris, vessel strikes, and continued direct harvest of green turtles in places like Nicaragua, are being addressed to some extent by regulatory mechanisms, although they remain a problem. In addition, other regional and national legislation to conserve green turtles (often all sea turtles) exists throughout the range of the DPS. The extent to which threats have been reduced as a result of these efforts is difficult to ascertain. When the SRT assessed conservation efforts, it assumed that all conservation efforts would remain in place at their current levels. The following countries have laws to protect green turtles: The Bahamas, Belize, Bermuda, Canary Islands, Cayman Islands, Costa Rica, Cuba, Dominican Republic, Guatemala, Haiti, Honduras, Jamaica, Mauritania, Mexico, Nicaragua, Panama, and the United States (including the commonwealth of Puerto Rico).

With regard to the United States, the key law currently protecting green turtles is the ESA. This law has been instrumental in conserving sea turtles, eliminating directed take of turtles in U.S. waters unless authorized by permit and reducing indirect take. In addition, the Magnuson-Stevens Fishery Management and Conservation Act has been effective at mandating responsible fishing practices and bycatch mitigation within fleets that sell fisheries products to the United States, and the Marine Turtle Conservation Act authorizes a dedicated fund to support marine turtle conservation projects in foreign countries, with emphasis on protecting nesting populations and nesting habitat. In addition, at least 12 international treaties and/or regulatory mechanisms apply to the conservation of green turtles in the North Atlantic DPS.

Outside of the United States, there are some national regulations that address the harvest of green turtles as well as the import and export of turtle parts. These regulations allow for the harvest of green turtles of certain sizes, months, or for “traditional” use. Gear restrictions and TED requirements exist in a few countries, although the compliance level is unknown. Our Status Review did not reveal regulatory mechanisms in place to specifically address marine pollution, sea level rise, and other effects of climate change that continue to contribute to the extinction risk of this DPS.

5. Factor E: Other Natural or Manmade Factors Affecting Its Continued Existence

a. Incidental Bycatch in Fishing Gear

Fisheries bycatch in artisanal and industrial fishing gear continues to be a major threat to green turtles in the North Atlantic DPS. The adverse impacts of bycatch on sea turtles has been documented in marine environments throughout the world (National Research Council, 1990b; Epperly, 2003; Lutcavage et al., 1997). The lack of comprehensive and effective monitoring and bycatch reduction efforts in many pelagic and near-shore fisheries operations throughout the range of the North Atlantic DPS still allows substantial direct and indirect mortality (NMFS and USFWS, 2007).

i. Gill Net and Trawl Fisheries

Gill net fisheries may be the most ubiquitous of fisheries operating in the neritic range of the North Atlantic DPS. In the United States, some states ( e.g., South Carolina, Georgia, Florida, Louisiana, and Texas) have prohibited gill nets in their waters, but there remain active gill net fisheries in other U.S. states, in U.S. Federal waters, Mexican waters, Central and South America, and the Northeast Atlantic. Finfish fisheries accounted for the greatest proportion of turtle bycatch (53 percent) in Cuba. In Jamaica, fish traps and gill nets are the gear primarily identified in sea turtle bycatch. Purse seine and gill nets are used commonly in the waters of the Dominican Republic (Dow et al., 2007). In Costa Rica, gill nets, hook and line, and trawls are the main gear types deployed (Food and Agriculture Organization of the United Nations, 2004). Shark-netting operations in Panama are known to capture green turtles (Meylan et al., 2013).

The development and implementation of TEDs in the U.S. shrimp trawl fishery was likely the most significant conservation accomplishment for North Atlantic green turtles in the marine environment since their 1978 ESA listing. In the southeast United States and Gulf of Mexico, TEDs have been mandatory in shrimp and flounder trawls for over a decade. However, compliance varies throughout the States, and green turtle mortality continues in the Gulf of Mexico, where shrimp trawling is the highest (Lewison et al., 2014). With the current regulations in place, an estimated 3,000 green turtles are captured (1,400 killed) by shrimp trawls each year in the Gulf and U.S. Southeast Atlantic ( http://sero.nmfs.noaa.gov/protected_resources/section_7/freq_biop/documents/fisheries_bo/shrimp_biop_2014.pdf ). These regulations are implemented and enforced to varying degrees throughout the Gulf and U.S. Southeast Atlantic (see discussion in Factor D).

ii. Dredge Fishing

Dredge fishing gear is the predominant gear used to harvest sea scallops off the mid- and northeastern U.S. Atlantic coast. Sea scallop dredges are composed of a heavy steel frame and cutting bar located on the bottom part of the frame and a bag made of metal rings and mesh twine attached to the frame. Turtles can be struck and injured or killed by the dredge frame and/or captured in the bag, where they may drown or be further injured or killed when the catch and heavy gear are dumped on the vessel deck.

b. Channel Dredging

In addition to the destruction or degradation of habitat as described in Factor A above, periodic dredging of sediments from navigational channels can also result in incidental mortality of sea turtles. Direct injury or mortality of green turtles by dredges has been well documented in the southeastern and mid-Atlantic U.S. (National Research Council, 1990b). From 1980 to 2013, 105 green turtles were impacted as a result of dredging operations in the U.S Atlantic and Gulf of Mexico. Solutions, including modification of dredges, have been successfully implemented to reduce mortalities and injuries to sea turtles in the United States (73 FR 18984, April 8, 2008; 77 FR 20728, April 6, 2012), and NMFS imposes annual take limits based on the expected number of green turtles impacted that will not, directly or indirectly, appreciably reduce the likelihood of survival and recovery of the green turtle in the wild.

c. Vessel Strikes and Boat Traffic

Boat strikes have been shown to be a major mortality source in Florida (Singel et al., 2003). Vessel strikes are a growing concern and, as human populations increase in coastal areas, e.g., Orós et al., 2005).

d. Effects of Climate Change and Natural Disasters

While sea turtles have survived past eras that have included significant temperature fluctuations, future climate change is expected to happen at unprecedented rates, and if turtles cannot adapt quickly, they may face local to widespread extirpations (Hawkes et al., 2009). Climate change and sea level rise have the potential to affect green turtles significantly in the North Atlantic DPS. North Atlantic turtle populations could be affected by the alteration of thermal sand characteristics of beaches (from warming temperatures), resulting in the reduction or cessation of male hatchling production (Hawkes et al., 2009; Poloczanska et al., 2009). Increased sea surface temperatures may alter the timing of nesting for some stocks (Weishampel et al., 2004), although the implications of changes in nesting timing are unclear. Changes in sea temperatures will also likely alter seagrass, macroalgae, and invertebrate populations in coastal habitats in many regions (Scavia et al., 2002). Further, a significant rise in sea level, as is projected for areas within the range of the North Atlantic DPS (Flaxman and Vargas-Moreno, 2011), could significantly restrict green turtle nesting habitat due to coastal development. Structures on the landward side of the beach can effectively prevent access to nesting habitat and reduce available nesting habitat (Mosier, 1998). The increasing interaction between the structures and the hydrodynamics of tide and current, due to sea level rise, often results in the alteration of the beach profile seaward and in the immediate vicinity of the structure (Pilkey and Wright, 1988; Terchunian, 1988; Tait and Griggs, 1990; Plant and Griggs, 1992), increased longshore currents that move sand away from the area, loss of interaction between the dune and the beach berm, and concentration of wave energy at the ends of the structure (Schroeder and Mosier, 1996). Impacts from global climate change induced by human activities are likely to become more apparent in future years (IPCC, 2007).

Periodic hurricanes and other weather events are generally localized and rarely result in whole-scale losses over multiple nesting seasons. However, storm intensity and frequency are predicted to increase as a result of climate change (Melillo et al., 2014). The negative effects of hurricanes on low-lying and/or developed shorelines may be longer-lasting and a greater threat to the DPS overall when combined with the effects of climate change, and particularly sea level rise.

e. Effects of Cold Stunning

Cold stunning is the hypothermic reaction that occurs when sea turtles are exposed to prolonged cold water temperatures. Cold stunning of green turtles regularly occurs at several locations in the United States, including Cape Cod Bay, Massachusetts (Still et al., 2002); Long Island Sound, New York (Meylan and Sadove, 1986; Morreale et al., 1992); the Indian River Lagoon system and the panhandle of Florida (Mendonça and Ehrhart, 1982; Witherington and Ehrhart, 1989; Foley et al., 2007); and Texas inshore waters (Hildebrand, 1982; Shaver, 1990). Cold-stunning events at these foraging areas (Witherington and Ehrhart, 1989; McMichael et al., 2006) leads to mortality of juvenile and adult green turtles, which may affect the present and future green turtle population trend.

f. Contaminants and Marine Debris

Several activities associated with offshore oil and gas production, including oil spills, operational discharge, seismic surveys, explosive platform removal, platform lighting, and drilling and production activities, are known to affect sea turtles (National Research Council, 1996; Davis et al., 2000; Viada et al., 2008; Conant et al., 2009; G. Gitschlag, NMFS, pers. comm., 2007, as cited in Conant et al., 2009). Oil spills near nesting beaches just prior to or during the nesting season place nesting females, incubating egg clutches, and hatchlings at significant risk from direct exposure to contaminants (Fritts and McGehee, 1982; Lutcavage et al., 1997; Witherington, 1999), and have negative impacts on nesting habitat. The Deepwater Horizon (Mississippi Canyon 252) oil spill, which started April 20, 2010, discharged oil into the Gulf of Mexico through July 15, 2010. Witherington et al. (2012) note that the Deepwater Horizon oil spill was particularly harmful to pelagic juvenile green turtles. Due to their size, turtles in these stages are more vulnerable as a result of ingesting contaminants (Witherington, 2002).

Green turtles are affected by anthropogenic marine debris (including discarded fishing gear) and plastics throughout the North Atlantic DPS. Juvenile green turtles in pelagic waters are particularly susceptible to these effects as they feed on Sargassum in which there is a high occurrence of debris (Wabnitz and Nichols, 2010; Witherington et al., 2012). In recent decades, there has been an increase in stranded green turtles reported as affected by discarded fishery gear throughout the southeastern United States (Teas and Witzell, 1996; Adimey et al., 2014).

C. Conservation Efforts for the North Atlantic DPS

In the North Atlantic, nest protection efforts have been implemented on two major green turtle nesting beaches, Tortuguero National Park in Costa Rica and Florida, and progress has been made in reducing mortality from human-related impacts on other nesting beaches. Tortuguero National Park was established in 1976 to protect the nesting turtles and habitat at this nesting beach, which is by far the largest in the DPS and the western hemisphere. Since that time, the harvest of nesting turtles on the beach has been reduced by an order of magnitude (Bjorndal et al., 1999). At Tortuguero, Sea Turtle Conservancy researchers and volunteers regularly monitor green turtle nesting trends, growth rates and reproductive success, and also conduct sea turtle lighting surveys, education, and community outreach.

In Florida, a key effort was the acquisition of the Archie Carr National Wildlife Refuge in Florida in 1991 by Federal, State, Brevard and Indian River counties, and a non-governmental organization, where nesting densities range from 36 nests/km (22 nests/mi) to 262 nests/km (419 nests/mi) (D. Bagley, University of Central Florida, pers. comm., 2014; K. Kneifl, USFWS, pers. comm., 2014). Over 60 percent of the available beachfront acquisitions for the Refuge have been completed as the result of a multi-agency land acquisition effort. In addition, Hobe Sound National Wildlife Refuge, as well as coastal national seashores such as the Dry Tortugas National Park and Canaveral National Seashore, military installations such as Patrick Air Force Base and Canaveral Air Force Station, and State parks where green turtles regularly nest, provide protection for nesting turtles. However, despite these efforts, alteration of the coastline continues and, outside of publicly-owned lands,

Considerable effort has been expended since the 1980s to document and reduce commercial fishing bycatch mortality. In the Atlantic and Gulf of Mexico, measures (such as gear modifications, changes to fishing practices, and time/area closures) are required to reduce sea turtle bycatch in pelagic longline, mid-Atlantic gill net, Virginia pound net, scallop dredge, and southeast shrimp and flounder trawl fisheries. However, enforcement of regulations depends on available resources, and bycatch continues to contribute to mortality. Since 1989, the United States has prohibited the importation of shrimp harvested in a manner that adversely affects sea turtles.

As a result of conservation efforts, many of the intentional impacts directed at sea turtles have been lessened. For example, harvest of eggs and adults has been reduced at several nesting areas, including Tortuguero, and an increasing number of community-based initiatives are in place to reduce the take of turtles in foraging areas. However, despite these advances, human impacts continue throughout the North Atlantic. The lack of effective monitoring in pelagic and near-shore fisheries operations still allows substantial direct and indirect mortality, and the uncontrolled development of coastal and marine habitats threatens to destroy the supporting ecosystems of long-lived green turtles.

D. Extinction Risk Assessment and Findings for the North Atlantic DPS

In the North Atlantic DPS, there are several regions that support high density nesting concentrations, including possibly the largest in the world at Tortuguero, Costa Rica. Green turtle nesting population trends have been encouraging, exhibiting long-term increases at all major nesting sites, including Tortuguero (Troëng, 1998; Campbell and Lagueux, 2005; Troëng and Rankin, 2005) and Florida (Chaloupka et al., 2008; B. Witherington, Florida Fish and Wildlife Conservation Commission, pers. comm., 2013). The North Atlantic DPS is characterized by geographically widespread nesting at a diversity of sites, both mainland and insular. The increasing threats are not reflected in the current trend for the North Atlantic DPS as it was based on nesting numbers and not all current life stages. These increasing threats to the population will become apparent when those life stages affected by the threats return to nest as the trend information is based solely on numbers of nests. This lag time was considered in our analysis. However, the 5-factor (section 4(a)(1) of the ESA) analysis revealed continuing threats to green turtles and their habitat that affect all life stages.

On nesting beaches, many portions of the DPS continue to be exposed to, and are negatively impacted by, coastal development and associated beachfront lighting, coastal armoring, and erosion as described in Factor A above. Impacts from such development are further exacerbated by existing and planned shoreline development and shoreline engineering. The current and anticipated increase in armored shoreline along high density nesting beaches, particularly in Florida, is a substantial unresolved threat to the recovery and stability of this DPS as it will result in the permanent loss of nesting habitat.

Nests and hatchlings are susceptible to predation which is prevalent throughout the beaches within the range of the North Atlantic DPS. Predation would be an increasing threat without nest protection and predatory control programs in place.

Nesting beaches are also extremely susceptible to sea level rise, which will exacerbate some of the issues described above in addition to leading to the potential loss of nesting beaches. Along the southeastern United States, one climate change model predicted a 1-meter sea level rise by 2060, resulting in the inundation of more than 50 percent of coastal wildlife refuges (Flaxman and Vargas-Moreno, 2011). Green turtle nesting in Florida is concentrated along coastal wildlife refuges in southern Florida such as Hobe Sound National Wildlife Refuge and the Archie Carr National Wildlife Refuge, with more than 90 percent of nesting occurring along southeast Florida. This increase in sea level will result in the permanent loss of current green turtle nesting habitat. Loss of beach is expected to be worse as a result of the increase in hurricane frequency and intensity (Flaxman and Vargas-Moreno, 2011). The increasing threat of coastal erosion due to climate change and sea level rise is expected to be exacerbated by increasing human-induced pressures on coastal areas (IPCC, 2007).

In the water, fisheries bycatch, habitat degradation, direct harvest, and FP are major threats to green turtles in the North Atlantic DPS. Artisanal and industrial fishing gear, including drift nets, set nets, pound nets, and trawls, still cause substantial direct and indirect mortality of green turtles (NMFS and USFWS, 2007). In addition, degradation and loss of foraging habitat due to pollution, including agricultural and residential runoff, anchor damage, dredging, channelization, and marina construction remains a threat to both juvenile and adult green turtles. Many green turtles in this DPS remain susceptible to direct harvesting. Current legal and illegal harvest of green turtles and eggs for human consumption continues in the eastern Atlantic and the Caribbean. A remaining threat is the directed harvest of turtles in Nicaragua that nest at Tortuguero and thus belong to the largest and arguably the most important population within the DPS (although this population continues to increase in spite of the harvest). However, potential degradation or loss of other, smaller populations is also of concern, as these contribute to the diversity and resilience of the DPS. Finally, the prevalence of FP has reached epidemic proportions in some parts of the North Atlantic DPS. The extent to which this will affect the long-term outlook for green turtles in the North Atlantic DPS is unknown. Nesting trends across the DPS continue to increase despite the high incidence of the disease.

While the Status Review indicates that the DPS shows strength in many of the critical population parameters (abundance, population trends, spatial structure, and diversity/resilience), as indicated above, numerous threats continue to act on the DPS, including habitat degradation (coastal development and armoring, loss of foraging habitat, and pollution), bycatch in fishing gear, continued turtle and egg harvesting, FP, and climate change. Importantly, the analysis of threats in the Status Review was conducted assuming current management regimes would continue.

Many of the gains made by the species over the past few decades are a direct result of ESA protections in the United States, as well as protections by U.S. States and local jurisdictions and other countries within the DPS range that are influenced by the species' ESA status.

Because the green turtle is currently listed under the ESA, take can only be authorized in the United States through the processes provided in sections 7 and 10 of the ESA and their implementing regulations. In the southeastern United States, threats to nesting beaches and nearshore waters include: Sand placement on nesting beaches and associated impacts to nearshore hardbottom habitat; groin, jetty and dock construction; and other activities. Any such activities that are currently funded, permitted and/or authorized by Federal agencies are subject to consultation with USFWS and NMFS, i.e., seagrass beds and nesting beaches). Activities that affect green turtles and do not involve Federal agencies, such as beach driving, some beach armoring, and research, must comply with section 10 of the ESA to avoid violating the statute. Section 10 permits require avoiding, minimizing, and mitigating impacts to green turtles to the extent possible. In addition to the above requirements, the requirement for use of TEDs in fisheries within the United States and in fisheries outside of the United States that export wild-caught shrimp to the United States is tied to listing under the ESA.

This DPS has exhibited increases at major nesting sites, and has several stronghold populations. Green turtles in the U.S. Atlantic have increased steadily since being protected by the ESA (Suckling et al., 2006). ESA driven programs such as land acquisition, nest protection, development of the TEDs, and educational programs provide a conservation benefit to green turtles. The species is conservation dependent or conservation-reliant in that even when biological recovery goals are achieved, maintenance of viable populations will require continuing, species-specific intervention (Scott et al., 2010). Without alternate mechanisms in place to continue certain existing conservation efforts and protections, threats would be expected to increase and population trends may be curtailed or reversed. Considering the conservation dependence of the species, significant concerns remain regarding the inadequacy of regulatory mechanisms (one of the five section 4(a)(1) factors (Factor D), especially when we evaluate the status of the DPS absent the protections of the ESA.

For the above reasons, we propose to list the North Atlantic DPS as threatened. We do not find the DPS to be in danger of extinction presently because of the increasing nesting population trends and geographically widespread nesting at a diversity of sites; however, continued threats are likely to endanger the DPS within the foreseeable future.

VIII. Mediterranean DPS

A. Discussion of Population Parameters for the Mediterranean DPS

The Mediterranean Sea is a virtually enclosed basin occupying an area of approximately 2.5 million square kilometers. The Mediterranean DPS is bounded by the entire coastline of the Mediterranean Sea, excluding the Black Sea. The westernmost border of the range of this DPS is marked by the Strait of Gibraltar (Figure 2).

Nesting in the Mediterranean occurs mostly in the eastern Mediterranean, with three nesting concentrations in Turkey, Cyprus, and Syria. Currently, approximately 452 to 2,051 nests are laid in the Mediterranean each year—about 70 percent in Turkey, 15 percent in Cyprus, and 15 percent in Syria, with trace nesting in Israel, Egypt, the Hellenic Republic (Greece), and Lebanon (Kasparek et al., 2001; Rees et al., 2008; Casale and Margaritoulis, 2010). There are no sites with greater than 500 nesting females. These numbers are depleted from historical levels (Kasparek et al., 2001). In terms of distribution of nesting sites in the Mediterranean, there are 32 sites, with Akyatan, Turkey being the largest nesting site, hosting 25 percent of the total annual nesting (35-245 nesting females; Türkozan and Kaska, 2010).

There are seven sites for which 10 years or more of recent data are available for annual nesting female abundance (a criterion for presenting trends in a bar graph). Of these, only one site—West Coast, Cyprus—met our standards for conducting a PVA. Of the seven sites, five appeared to be increasing, although some only slightly, and two had no apparent trend. However, while the Mediterranean DPS appears to be stable or increasing, it is severely depleted relative to historical levels. This dynamic is particularly apparent along the coast of Palestine/Israel, where 300-350 nests were deposited each year in the 1950s (Sella, 1995) compared to a mean of eight nests each year from 1993 to 2008 (Casale and Margaritoulis, 2010).

With regard to spatial structure, genetic sampling in the Mediterranean has been extensive and the coverage in this region is substantial. Within the Mediterranean, rookeries are characterized by one dominant haplotype CM-A13 and a recent study showed no population substructuring between several rookeries in Cyprus and Turkey (Bagda et al., 2012). However, analysis using unpublished data from additional rookery samples in Cyprus shows evidence for two stocks: Cyprus (Karpaz, North Cyprus and Lara Bay; Bagda et al., 2012; Dutton unpublished data, 2013); and Turkey (Akayatan, Alata, Kazanli, Samandag and Yumurtalık; Bagda et al., 2012). The demography of green turtles in the Mediterranean appears to be consistent among the various nesting assemblages (Broderick and Godley, 1996; Broderick et al., 2002a). This consistency in parameters such as mean nesting size, inter-nesting interval, clutch size, hatching success, nesting season, and clutch frequency suggests a low level of population structuring in the Mediterranean. Mediterranean turtles have not been detected foraging outside the Mediterranean ( e.g., Lahanas et al., 1998; Monzón-Argüello et al., 2010). Despite years of flipper tagging (Demetropoulos and Hadjichristophorou, 1995, 2010; Y. Kaska, Pamukkale University, pers. comm., 2013), few tag recoveries have been reported. However, satellite tracking revealed that post-nesting turtles migrate primarily along the coast from their nesting beach to foraging grounds, increasing the likelihood of interacting with fisheries (Broderick et al., 2002a).

With regard to diversity and resilience, the overall spatial range of the DPS is limited. Green turtle nesting is found primarily in the eastern Mediterranean (Turkey, Syria, Cyprus, Lebanon, Israel, and Egypt: Kasparek et al., 2001). The nesting season is consistent throughout the range of this DPS (June to August; Broderick et al., 2002a), thus limiting the temporal buffering against climate change in terms of impacts due to storms and other seasonal events. The fact that turtles nest on both insular and continental sites suggests some degree of nesting diversity, but with the sites so close together, the benefits of this diversity may be minimal.

B. Summary of Factors Affecting the Mediterranean DPS

1. Factor A: The Present or Threatened Destruction, Modification, or Curtailment of Its Habitat or Range

a. Terrestrial Zone

In the Mediterranean, destruction and modification of green turtle nesting habitat result from coastal development and construction, beachfront lighting, sand extraction, beach erosion, vehicular and pedestrian traffic, and beach pollution (Kasparek et al., 2001; Casale and Margaritoulis, 2010). These activities may directly affect the amount and suitability of nesting habitat available to nesting females and thus affect the nesting success of green turtles, as well as the survivability of et al., 2010). In Turkey and Latakia beach in Syria, beach erosion and sand extraction also pose a problem to green turtle nesting habitat (Türkozan and Kaska, 2010; Rees et al., 2010).

Nesting beaches in the eastern Mediterranean are exposed to high levels of pollution and marine debris, in particular the beaches of Cyprus, Turkey, and Egypt (Camiñas, 2004). In Turkey, marine debris washing ashore is a substantial problem and has degraded nesting beaches, especially Akyatan and Samandağ beaches. In Syria, Jony and Rees (2008) reported that beaches contain a large amount of plastic litter that washes ashore or is blown in from dumps located in the beach dunes; this litter has been documented as accumulating in such large amounts that it can hinder nesting females from locating suitable nesting sites and cause emergent hatchlings to have difficulty crawling to the sea (Rees et al., 2010). In Cyprus, marine debris has also been a significant problem on some beaches, although organized beach clean-ups in recent years have greatly reduced the amount of litter on the beach (Demetropoulos and Hadjichristophorou, 2010; Fuller et al., 2010).

b. Neritic/Oceanic Zones

Dynamite fishing and boat anchors affect green turtles and their habitat in the Mediterranean. Khalil et al. (2009) reported that dynamite fishing offshore of nesting beaches is a common problem in Lebanon. Illegal dynamite fishing also occurs year round in Libya (Hamza, 2010), and, although illegal, explosions at sea that are likely due to dynamite fishing have been reported off the coast of Syria (Saad, unpubl. data, as cited in Rees et al., 2010). Further, the Mediterranean is a site of intense tourist activity, and corresponding boat anchoring also may affect green turtle foraging habitat in the neritic environment.

Because the Mediterranean is an enclosed sea, organic and inorganic wastes, toxic effluents, and other pollutants rapidly affect the ecosystem (Camiñas, 2004). The Mediterranean has been declared a “special area” by the MARPOL Convention (International Convention for the Prevention of Pollution from Ships), in which deliberate petroleum discharges from vessels are banned, but numerous repeated offenses are still thought to occur (Pavlakis et al., 1996).

2. Factor B: Overutilization for Commercial, Recreational, Scientific, or Educational Purposes

Overutilization for commercial purposes likely was a factor that contributed to the historical declines of this DPS. Egg collection and turtle harvest for individual consumption still occurs in Egypt (Clarke et al., 2000; Nada and Casale, 2008). A study found that the open selling of sea turtles in Egypt generally has been curtailed due to enforcement efforts, but a high level of intentional killing for the black market or for direct personal consumption still exists (Nada and Casale, 2008). Several hundred turtles are currently estimated to be slaughtered each year in Egypt (Nada and Casale, 2008). In Syria and Egypt, as reported for other countries, green turtles incidentally captured by fishers are sometimes eaten (Nada and Casale, 2008; Rees et al., 2010). Small quantities of stuffed turtles and juvenile turtle carapaces, presumably of Syrian origin, have been observed for sale in Latakia and Damascus (Rees et al., 2010).

3. Factor C: Disease or Predation

Nest and hatchling predation likely was a factor that contributed to the historical decline of the Mediterranean DPS. There have been no records of FP or other diseases in green turtles in this DPS. In this DPS, green turtle eggs and hatchlings are subject to depredation by wild canids ( i.e., foxes ( Vulpes vulpes ), golden jackals ( Canis aureus ), feral and domestic dogs ( Canis lupus familiaris ), and ghost crabs ( Ocypode cursor; van Piggelen and Strijbosch, 1993; Brown and MacDonald, 1995; Aureggi et al., 1999, 2005; Simms et al., 2002; Akcinar et al., 2006; Jony and Rees, 2008; Khalil et al., 2009; Aureggi and Khalil, 2010; Demetropoulos and Hadjichristophorou, 2010; Fuller et al., 2010; Rees et al., 2010).

4. Factor D: Inadequacy of Existing Regulatory Mechanisms

There are at least 13 international treaties and/or regulatory mechanisms that pertain to the Mediterranean, and nearly all countries lining the Mediterranean have some level of national legislation directed at sea turtle protection. The SRT analysis of these existing regulatory mechanisms assumed that all would remain in place at their current levels.

Regulatory mechanisms are in place throughout the range of the DPS that address the direct capture of green turtles for most of the countries within this DPS. Most Mediterranean countries have developed national legislation to protect sea turtles and nesting habitats (Casale and Margaritoulis, 2010). The following countries have laws to protect green turtles: Albania, Croatia, Cyprus, Egypt, Greece, Israel, Italy, Lebanon, Libya, Syria, Tunisia, and Turkey. In addition, at least 13 international treaties and/or regulatory mechanisms apply to the conservation of green turtles in the Mediterranean DPS. National protective legislation generally prohibits intentional killing, harassment, possession, trade, or attempts at these (Margaritoulis et al., 2003). In addition, some countries have site-specific legislation or conservation designation for turtle habitat protection. These are implemented to various degrees throughout the range of the DPS. There are some national regulations, within this DPS, that specially address the harvest of green turtles.

In western Cyprus, Lara-Toxeftra beaches have been afforded protection through the Fisheries Law and Regulations since 1989 (Margaritoulis, 2007). In northern Cyprus, four beaches (Alagadi Beach, Karpaz Peninsular, South Karpaz, and Akdeniz) have been designated as Special Protected Areas (Fuller et al., 2010). These four areas include the third and fifth most important green turtle nesting beaches in the Mediterranean (Kasparek et al., 2001). In Syria, establishment of a protected area at Latakia beach, the most important green turtle nesting beach in the country, is being sought but is facing strong opposition from the tourism sector (Rees et al., 2010). While it is important to recognize the success of these protected areas, we must also note that the protection has been in place for some time and the threats to the species remain (particularly from increasing tourism activities). It is unlikely that the protective measures discussed here are sufficient for the conservation of the species in the Mediterranean.

Regulatory mechanisms are not in place in many countries within this DPS to address the major threat of sea turtle bycatch. Some of the countries in which this DPS is located limit the number and type of fishing licenses issued but sea turtle bycatch is not considered in these authorizations. It is unlikely that bycatch mortality can be sufficiently reduced across the range of the DPS in

5. Factor E: Other Natural or Manmade Factors Affecting Its Continued Existence

a. Incidental Bycatch in Fishing Gear

Incidental capture of sea turtles in artisanal and commercial fisheries is a significant threat to the survival of green turtles in the Mediterranean. Fishing practices alone have been estimated to result in over 150,000 sea turtle captures per year, with approximately 50,000 mortalities (Lucchetti and Sala, 2009; Casale, 2011) and sea turtle bycatch in multiple gears in the Mediterranean is considered among the most urgent conservation priorities globally (Wallace et al., 2010).

i. Longline Fisheries

In the Mediterranean, surface longline fisheries are a source of green turtle bycatch (Camiñas, 2004). Incidental captures have been reported from Cyprus (Godley et al., 1998), Turkey (Godley et al., 1998), Italy (Laurent et al., 2001), and Egypt (Nada, 2001; Camiñas, 2004). In Egypt, based on fleet data and catch rates reported by fishers during the 2000s, the total number of sea turtles ( i.e., all species) bycaught in longlines was estimated to be over 2,200 per year (Nada and Casale, 2008). Fishers also reported that some of the caught turtles are dead, and the incidence of mortality is particularly high in longlines and gill nets.

ii. Set Net (Gill Net) Fishing

Casale (2008) considered mortality by set nets to be 60 percent, with a resulting estimate of 16,000 turtles killed per year. However, a breakdown of these estimates by turtle species is not available. Most of these turtles are likely juveniles, with an average size of 45.4 cm CCL (n=74, Casale, 2008).

iii. Trawl Fisheries

Green turtles have been reported as incidentally captured in bottom trawls in Egypt (Nada and Casale, 2011), Greece (Margaritoulis et al., 2003), Tunisia (Laurent et al., 1990), Turkey (Laurent et al., 1996; Oruç, 2001), Syria, Israel, and Libya (Casale et al., 2010), but are likely also captured by bottom trawlers in other neritic foraging areas in the eastern Mediterranean (Casale et al., 2010). Laurent et al. (1996) estimated that approximately 10,000 to 15,000 sea turtles were being captured annually by bottom trawling in the eastern Mediterranean. Although most of the turtles taken were loggerheads, they estimated that the number of green turtles taken was 1,000 to 3,000 annually in Turkey and Egypt alone. More recently, Casale (2011) compiled available trawl bycatch data throughout the Mediterranean and reported that Italy and Tunisia have the highest level of sea turtle bycatch, potentially over 20,000 captures per year combined, and Croatia, Greece, Turkey, Libya, Greece, and Egypt each have an estimated 1,900 or more sea turtle captures per year. Further, Albania, Algeria, Cyprus, Morocco, Slovenia, Spain, and Syria may each capture a few hundred sea turtles per year (Casale, 2011). Available data suggest the annual number of sea turtle captures by all Mediterranean trawlers may be greater than 39,000 (Casale, 2011). Although most of the turtles reported by Casale (2011) as taken by bottom trawlers were undoubtedly loggerheads, a few thousand were likely green turtles based on earlier reports (Laurent et al., 1990; Laurent et al., 1996; Oruç, 2001; Margaritoulis et al., 2003; Nada and Casale, 2008).

b. Vessel Strikes and Boat Traffic

Propeller and collision injuries from boats and ships are becoming more common for sea turtles in the Mediterranean, although it is unclear as to whether the events, or just the reporting of the injuries, are increasing. Speedboat and jet-ski impacts are of particular concern in areas of intense tourist activity, such as Greece, Turkey, and Syria. Boats operating near sea turtle nesting beaches during the nesting season are likely to either cause females to abandon nesting attempts or cause their injury or death (Camiñas, 2004). Males may also be affected in high-use boating areas where sea turtle mating occurs (Demetropoulos, 2000; Rees et al., 2010).

c. Pollution

Unattended or discarded nets, floating plastics and bags, and tar balls are of particular concern in the Mediterranean (Camiñas, 2004; Margaritoulis, 2007). Monofilament netting appears to be the most dangerous waste produced by the fishing industry (Camiñas, 2004).

The discharge of chemical substances, including highly toxic chromium compounds from a soda-chromium factory close to the Kazanli nesting beach in Turkey, is cause for concern (Kasparek et al., 2001; Venizelos and Kasparek, 2006).

d. Effects of Climate Change

Both the marine and terrestrial realms will be influenced by temperature increases and will likely undergo alterations that will adversely affect green turtles. Mediterranean turtle populations could be affected by the alteration of thermal sand characteristics (from global warming), resulting in the reduction or cessation of male hatchling production (Kasparek et al., 2001; Camiñas, 2004; Hawkes et al., 2009; Poloczanska et al., 2009). In northern Cyprus, green turtle hatchling sex ratios are already thought to be highly female biased (approximately 95 percent female; Wright et al., 2012). This, in tandem with predicted future rises in temperatures, is cause for concern (Fuller et al., 2010). As temperatures increase, there is also concern that incubation temperatures will reach levels that exceed the thermal tolerance for embryonic development, thus increasing embryo and hatchling mortality (Fuller et al., 2010). Further, a significant rise in sea level would restrict green turtle nesting habitat in the eastern Mediterranean. While sea turtles have survived past eras that have included significant temperature fluctuations, future climate change is expected to happen at unprecedented rates, and if turtles cannot adapt quickly they may face local to widespread extirpations (Hawkes et al., 2009). Impacts from global climate change induced by human activities are likely to become more apparent in future years (IPCC, 2007).

In summary, within Factor E, we find that fishery bycatch and marine pollution that occurs throughout the range of the Mediterranean DPS are significant threats to this DPS. In addition, boat strikes and changes likely to result from climate change are an increasing threat to the persistence of this DPS.

C. Conservation Efforts

Regional and national efforts are underway to conserve green turtles (often all sea turtles) throughout the range of the DPS. The extent to which threats have been reduced as a result of these efforts is difficult to ascertain.

Green turtle nesting primarily occurs in Turkey, Cyprus, and Syria, and a et al., 2001; Fuller et al., 2010). These five protected beaches represent approximately 60 percent of nesting in Turkey (see Canbolat et al., 2009 and Fuller et al., 2010).

There has been success within these protected areas, but as the protection has been in place for some time and the threats to the species remain (particularly from increasing tourism activities), it is unlikely that the protective measures discussed here are sufficient for the conservation of the species in the Mediterranean.

Marine debris is also a significant problem on many green turtle nesting beaches in the eastern Mediterranean, in particular the nesting beaches of Cyprus and Turkey (Camiñas, 2004; Demetropoulos and Hadjichristophorou, 2010; Fuller et al., 2010; Türkozan and Kaska, 2010). Although organized beach clean-ups in recent years on some beaches in Cyprus have greatly reduced the amount of litter on the beach (Demetropoulos and Hadjichristophorou, 2010; Fuller et al., 2010), it is still an overall pervasive problem.

Protection of marine habitats is in the early stages in the Mediterranean, as in other areas of the world. Off the Lara-Toxeftra nesting beaches in western Cyprus, a marine protection zone extends to the 20-m isobath ( i.e., 20-m depth line) as delineated by the Fisheries Regulation (Margaritoulis, 2007; Demetropoulos and Hadjichristophorou, 2010). As mentioned above, establishment of a protected area at Latakia beach in Syria is being sought and would include protection of a section of sea offshore; however, it is facing strong opposition from the tourism sector (Serra, 2008; Rees et al., 2010).

D. Extinction Risk Assessment and Findings

The Mediterranean DPS is characterized by low green turtle nesting abundance at 32 different locations, with many of these sites having only one or two known nesting females and none having greater than 245 nesting females. While some of these sites show stable or increasing trends, the extremely low nesting abundance of this DPS compared to historical abundance creates an intrinsically high risk to the long-term stability of the population. The spatial range of the population is limited to the eastern Mediterranean, and the nesting season is consistent throughout this DPS (June to August; Broderick et al., 2002a), thus limiting the temporal buffering against climate change in terms of impacts due to storms and other seasonal events. The fact that turtles nest on both insular and continental sites suggests some degree of nesting diversity but, with the sites so close together, the benefits of this diversity may be minimal. Mitochondrial DNA studies have identified two stocks but, in general there is low population substructuring in the Mediterranean.

The five-factor analysis in the Status Review reveals numerous significant threats to green turtles within the range of the DPS. Coastal development, beachfront lighting, erosion resulting from sand extraction, illegal harvest, detrimental fishing practices, and marine pollution both at nesting beaches and important foraging grounds are continuing concerns across the Mediterranean DPS, and are insufficiently tempered by conservation efforts. Current illegal harvest of green turtles for human consumption continues as a moderate threat to this DPS. Fishery bycatch occurs throughout the Mediterranean Sea, particularly bycatch mortality of green turtles in pelagic longline, set net, and trawl fisheries. Additional threats from boat strikes, which are becoming more common, and changes likely to result from climate change will negatively affect this DPS.

For the above reasons, we propose to list the Mediterranean DPS as endangered. Based on its low nesting abundance, limited spatial distribution, and exposure to increasing threats, we find that this DPS is presently in danger of extinction throughout its range.

IX. South Atlantic DPS

A. Discussion of Population Parameters for the South Atlantic DPS

The South Atlantic DPS's range boundary begins at the border of Panama and Colombia at 7.5° N., 77° W., heads due north to 10.5° N., 77° W., then northeast to

This text is long and has been trimmed here. Open the source document for the complete record.