Discovery of the Imperiled Miami Blue Butterfly (Cyclargus thomasi bethunebakeri) on Islands in the Florida Keys National Wildlife Refuges, Monroe County Paula Cannon1, Tom Wilmers2,* and Katie Lyons3 Abstract - A once common butterfly found only in Florida, Cyclargus thomasi bethunebakeri (Miami Blue), declined to near extinction during the 1990s. On 26 November 2006, we discovered this species on Boca Grande Key, 76 km west of Bahia Honda Key. Previously, the butterfly was only known to survive on Bahia Honda Key. Over the following 8-month period, we searched periodically for this butterfly on islands in the Great White Heron National Wildlife Refuge (GWHWNR) and the Key West National Wildlife Refuge (KWNWR). The Miami Blue was not found on 6 islands in GWHNWR, but was present at 8 of 10 areas in KWNWR. Seven of the occupied sites were on islands in the Marquesas Keys. The number of Miami Blue adults varied greatly by island and season, but were highest at the two largest colonies when Pithecellobium keyense (Blackbead; both a larval host and adult nectar plant) was flowering and producing new leaves. Eight other plant species were visited by adult Miami Blues for nectar. Threats to the newly discovered colonies are discussed. Cyclargus thomasi bethunebakeri Comstock and Huntington (Miami Blue) was once widely distributed in coastal areas of southern and central Florida (Calhoun et al. 2002, Minno and Emmel 1994). The butterfly declined and disappeared from much of its former range by the 1980s, but still survived in the Florida Keys (Leston et al. 1982). In the years following Hurricane Andrew in 1992, the Miami Blue seemed to disappear from the Keys as well (Calhoun et al. 2002). Then in 1999, the Miami Blue was discovered at Bahia Honda State Park in the lower Keys (Ruffin and Glassberg 2000). Here we report on the discovery of additional Miami Blue colonies on islands in the Key West National Wildlife Refuge, west of Key West. We also present notes on the abundance, biology, and threats to this imperiled butterfly. Study Area and Methods. Our work was performed on uninhabited islands in the Key West National Wildlife Refuge (KWNWR) and the Great White Heron National Wildlife Refuge (GWHNWR). These refuges are situated in the lower Florida Keys where the climate is subtropical. Islands in the study area are nearly flat. Most are less than 1 m above sea level, with none >2 m in elevation. We searched for adult Miami Blues in 16 areas (1 to 9 visits per area), 6 in the GHWNR and 10 in the KWNWR (Table 1). Of the latter, 7 were in the Marquesas Keys, a circular array of islands 29 to 37 km west of Key West, fl. The largest island in the Marquesas Keys contains 3 upland tracts, separated from each other by 500–3500 m of mangrove forest. Because unsuitable habitat only 30 m wide may curtail Miami Blue dispersal (Saarinen and Daniels 2005), we considered each of these 3 uplands separate study sites called Long Beach, Snook Beach, and East Cove. All other sites in the study area were on individual islands. Twelve of the 16 study sites had narrow coastal strands and dunes fronted by beaches (range = 50–2900 m long) (Table 1). Two sites contained only a hardwood hammock, and two others consisted of upland coastal strands fronted by mangroves (Table 1). The size of the upland areas differed greatly among islands, ranging from less than 0.1 ha to ≈15 ha. Notes of the Southeastern Nat u ral ist, Issue 9/4, 2010 847 12661 Central Avenue, Big Pine Key, fl33043. 2Florida Keys National Wildlife Refuges, Big Pine Key, fl33043. 38204 County Road 3140 Mountain View, MO 65548. *Corresponding author - thomas_wilmers@fws.gov. 848 Southeastern Naturalist Notes Vol. 9, No. 4 We used checklist counting, a method that is more efficient than transects for initially determining species presence (Royer et al. 1998), to document the distribution and abundance of Miami Blues. With checklist counting, an unrestricted search of potentially suitable habitat is made, and target species are counted without regard to other factors (Royer et al. 1998). We recorded the number of adult Miami Blues observed at each site and the time spent searching. We also made systematic counts of adult Miami Blues over the entire upland areas of Boca Grande Key on 3 February, 9 April, and 23 July and at Main Beach in the Marquesas Keys on 8 January and 10 March 2007. For the systematic counts, we divided the uplands into 3 parallel transects (each >350 m) so that each of us could cover a different area simultaneously to reduce the possibility of double counting. We walked very slowly along the transects and used binoculars that focused to <2 m to view and positively identify the butterflies. Each observer used a digital counter to tally the number of adults seen. The duration of the systematic counts ranged from 3.1 to 4.5 hours. In KWNWR, but not GWHNWR, dense patches of dead trees and shrubs blown down by Hurricane Wilma were intermixed with live vegetation, precluding a search of parts of the uplands. On all islands, we searched for the butterfly’s known host plants (Minno and Emmel 1993, Minno et al. 2005) and incidentally documented their use by the Miami Blue. We also recorded incidental observations of plants visited for nectar by this butterfly. Observations and Discussion. In the KWNWR, Miami Blues were found on Boca Grande Key and five islands in the Marquesas Keys, including three different sites on one island (Tables 1, 2). Upland areas (i.e., dune/coastal strand) occupied by the Miami Blue ranged from ≈0.1–15 ha (Table 1). All areas harboring this butterfly contained Pithecellobium keyense Britton (Blackbead), the observed larval host plant. Table 1. Characteristics of areas surveyed for Miami Blues (MB) in the Florida Keys National Wildlife Refuges; H = hardwood hammock only; BDC = dune and coastal strand fronted by beach, NB = coastal strand landward of a mangrove fringe, N/A = no beach or dune. Area MB present Upland size (ha) Upland type Beach length (m) Great White Heron NWR: Johnson Key No 0.3 H N/A Johnston Key No 0.1 H N/A Porpoise Key No 0.1 NB N/A Sawyer Keys (NE) No 0.5 NB N/A Sawyer Keys (NW) No 2.0 BDC 100 Snipe Point No 2.0 BDC 125 Key West NWR: Boca Grande Key Yes 4.0 BDC 1220 Man Key (east) No 1.0 BDC 50 Marquesas Keys: Island 1 (East Cove) Yes 0.1 BDC 215 Island 1 (Long Beach) Yes 15.0 BDC 2900 Island 1 (Snook Beach) Yes 1.0 BDC 525 Island 2 (Main Beach) Yes 2.0 BDC 1065 Island 3 (New Beach) Yes 0.2 BDC 85 Island 4 (Short Beach) Yes 0.5 BDC 450 Island 5 (Third Beach) Yes 1.0 BDC 500 Woman Key No 3.0 BDC 1300 2010 Southeastern Naturalist Notes 849 We made our first discovery of Miami Blues on 26 November 2006 at Boca Grande Key (9 adults), located 19 km west of Key West and about 76 km west of the known colony on Bahia Honda Key. Several Miami Blues were photographed, including a female laying eggs on Blackbead (Fig. 1) . The occupied habitat consisted Figure 1. A female Miami Blue ovipositing on a Blackbead flower bud on Boca Grande Key, 26 November 2006 (A); female Miami Blue nectaring on a Blackbead flower on Marquesas Keys, 3 December 2006 (B); egg found on Blackbead flower buds on Boca Grande Key, 26 November 2006 (C). Table 2. Number of adult Miami Blues (MB) observed in the Key West National Wildlife Refuge, November 2006–July 2007. Bold text indicates replicate transects. All others are checklist counts in selected uplands. Asterisk denotes MB presence but no count taken. Island Date No. MB Boca Grande 11/26/2006 9 Boca Grande 1/23/2007 100 Boca Grande 2/3/2007 441 Boca Grande 4/9/2007 51 Boca Grande 4/20/2007 30 Boca Grande 5/28/2007 0 Boca Grande 6/18/2007 1 Boca Grande 7/16/2007 * Boca Grande 7/23/2007 89 East Cove 1/23/2007 12 Long Beach 12/3/2006 7 Long Beach 6/4/2007 3 Long Beach 6/18/2007 24 Main Beach 12/3/2006 50 Main Beach 1/8/2007 521 Main Beach 3/10/2007 91 Main Beach 5/2/2007 2 Main Beach 7/23/2007 39 New Beach 1/23/2007 8 Short Beach 12/3/2006 10 Short Beach 6/18/2007 1 Snook Beach 1/8/2007 75 Snook Beach 6/18/2007 119 Snook Beach 7/30/2007 27 Third Beach 12/3/2006 30 Third Beach 7/23/2007 10 850 Southeastern Naturalist Notes Vol. 9, No. 4 of a dense patch of Blackbead flanked by abundant flowering herbaceous species such as Melanthera nivea (L.) Small (Snow Squarestem) and Cakile edentula (Bigelow) Hook (Coastal Searocket). On 23 January 2007, we searched about 30% of the uplands on Boca Grande Key and observed 100 adult Miami Blues over a 1.5-hr period. We performed systematic counts (each >3 h) on 3 February, 9 April, and 23 July 2007 and observed 441, 51, and 89 adult Miami Blues, respectively. On 3 December 2006, Miami Blues were abundant on Main Beach. We counted 50 adult Miami Blues along ≈30 m of the 1065-m dune before a rapidly falling tide forced our departure. We performed systematic counts on 8 January (4.5 h) and 10 March 2007 (3.4 h) and counted 521 and 91 adult Miami Blues, respectively. Partial searches of Main Beach (2 May 2007) and Boca Grande Key (28 May and 18 June 2007) revealed only 2, 0, and 1 Miami Blues, respectively. We observed little new leaf growth and no flowering of the Blackbead on these 2 islands after February 2007. Although the dune/coastal strand on Snook Beach is considerably smaller than that of Main Beach and Boca Grande Key (Table 1), we counted 75 and 119 Miami Blues, respectively, during >2-h searches on 8 January and 18 June 2007. Thus, it is an important area for the Miami Blue. New Beach had a very short beach (85 m) with a narrow coastal strand. For >2 hours, we searched the entire area for Miami Blues and their known host plants. We found a tiny Caesalpinia bonduc, (L.) Roxb. (Gray Nickerbean), a known Miami Blue host plant (Minno et al. 2005), and only five small Blackbead, four of which were aggregated in one area. Although herbaceous food plants were abundant and flowering, only 8 Miami Blues were observed. We believe a paucity of host plants was a limiting factor at this site. On East Cove, a very narrow (<2 m wide) dune was bordered by such a dense tangle of live and dead Blackbead that we could not even enter the coastal strand. Twelve adult Miami Blues were counted on the strand’s periphery during a 1.5-h search. On Short Beach, the once-abundant stands of Blackbead (T. Wilmers, pers. observ.) were nearly obliterated by Hurricane Wilma, with only scattered stressed individuals remaining. We counted 10 Miami Blues on the extreme south end of the island, where a narrow upland extended some 75 m landward of the dune. At Third Beach, Blackbead was scarce and limited primarily to the southwest side of the beach. Despite this, 30 Miami Blues were counted there during a brief search (<1 h) on 3 December 2006. The dune on Long Beach was the only one in the study area that had been blanketed by a thick mat of calcareous Halimeda algae washed ashore by Hurricane Wilma’s storm surge (T. Wilmers, pers. observ.). This algae deposit was likely a factor for the very slow recovery of the dune vegetation, which even by July 2007 remained sparse. Long Beach was far too large for us to survey, and our work there was limited to a small area. We counted 24 Miami Blues along 500 m of the dune/ coastal forest interface on 18 June 2007. Given the great length of its dune/coastal strand, Long Beach has the potential, once the habitat fully recovers from hurricane damage, to harbor more Miami Blues than any other surveyed site. We found no Miami Blues on east Man Key or Woman Key. These islands had abundant nectar sources, but a paucity of host plants. No Miami Blues were found in the Great White Heron NWR. No host plants were found on northwest Sawyer Key, but herbaceous nectar sources were abundant. Northeast Sawyer Key harbored a few small Blackbead and several Gray Nickerbean but few nectar sources. Johnston Key (west GWHNWR) contained a dearth of nectar 2010 Southeastern Naturalist Notes 851 sources and no host plants. Johnson Key (east GWHNWR) had only a few scattered Blackbead and few herbaceous nectar sources. The tiny upland area (10 x 20 m) on Porpoise Key lacked any herbaceous nectar source and was too small to harbor a Miami Blue population. Snipe Point contained scattered Gray Nickerbean and Blackbead and abundant herbaceous nectar sources. Including Blackbead, adult Miami Blues visited the flowers of nine plant species for nectar (Table 3). Of these, four were woody plant species: Argusia gnaphalodes (L.) Heine (Sea Lavender), Suriana maritima (L.) (Bay Cedar), Erithalis fruticosa (L.) (Black Torch), and Blackbead. All of the Miami Blue colonies in KWNWR were on remote islands (19 to 36 km from the nearest habitation, Key West) having federal wilderness designation. Mosquito spraying, which is potentially detrimental to the Miami Blue (Carroll and Loye 2006) and widespread on the mainline Florida Keys (islands linked by roads), has never occurred on or near any of the islands in the KWNWR. All areas occupied by the Miami Blue had coastal strands and dunes fronted by beaches. Important requirements in such settings were apparently the host plant, Blackbead, and other abundant nectar sources. Grey Nickerbean, the main host plant used by the Miami Blue at Bahia Honda State Park (Carroll and Loye 2006), was rare in KWNWR, with only a few small plants on Boca Grande Key and the Marquesas Keys. Even before Hurricane Wilma, the plant was not common in the KWNWR (T. Wilmers, pers. observ.). The growth stage of Blackbead, coupled with abundant nectar from herbaceous plants, likely influenced Miami Blue abundance. The highest counts on both Boca Grande Key and Main Beach occurred when Blackbead was flowering profusely and producing new leaves. During late November 2006 through January 2007, an unusually intense profusion of Blackbead flowering occurred. During this period, we frequently observed Miami Blues nectaring on Blackbead flowers (Fig. 1). Female Miami Blues laid eggs on Blackbead flower buds (Fig. 1) and emerging Blackbead leaves. We did not observe any evidence of egg-laying on or larval consumption of mature Blackbead leaves. Thus, even if abundant suitable nectar sources were available year-round, Miami Blue numbers would be regulated by the availability of young Blackbead leaves and buds for egg-laying. Although Hurricane Wilma severely damaged or killed many Blackbead, the storm likely enhanced Miami Blue foraging habitat, if only temporarily, on both Boca Grande Key and Main Beach by removing the extensive dense stands of Uniola paniculata L. (Sea Oats) adjacent to the coastal strand. This plant is neither a nectar source nor host plant for the Miami Blue. In its place grew an abundance of early successional Table 3. A list of plants used as a nectar source by the Miami Blue in the Key West National Wildlife Refuge. Scientific name Common name Argusia gnaphalodes Sea Lavender Alternanthera flavescens Yellow Joyweed Cakile lanceolata Coastal Searocket Erithalis fruticosa Black Torch Heliotrophium curassavicum Seaside Heliotrope Pithecellobium keyense Blackbead Melanthera nivea Snow Squarestem Suriana maritima Bay Cedar Sesuvium portulacastrum Sea Purslane 852 Southeastern Naturalist Notes Vol. 9, No. 4 herbaceous plants that were flowering profusely and providing plentiful nectar for the Miami Blue during the early months (December–February) of our study. Further, in the coastal strand, the hurricane created open patches in what had been a shaded woody understory (T. Wilmers, pers. observ.). Herbaceous plants emerged in these open patches, providing additional nectar for the Miami Blue. (T. Wilmers, pers. observ.). That the plentiful new Blackbead leaves and buds for egg deposition were present concurrently with abundant nectar from the flowers of both Blackbead and herbaceous plants lead us to believe that habitat conditions after Hurricane Wilma were particularly favorable for Miami Blues from December 2006 to February 2007 on Boca Grande Key and Main Beach. By March, Blackbead was no longer budding or flowering, and by May we observed that herbaceous plant flowering had decreased drastically on both islands. During the latter month, we found 0 and 2 Miami Blues on Boca Grande Key and Main Beach, respectively. Ultimate factors regulating butterfly populations remain unresolved (see McLaughlin et al. 2002). Minno and Minno (2009) noted that butterfly metapopulations in the Florida Keys rise and fall dramatically, with core populations expanding greatly during exceptionally favorable conditions, resulting in dispersal and temporary colonization of even marginal habitats. That Miami Blues were found at such a large number of sites in the Marquesas Keys is fortuitous. These islands are remote and seldom visited by humans. How frequently Miami Blues disperse between these islands is unknown. The proximity and circular arrangement of the islands may afford some measure of protection to the butterflies during tropical storms and moderate hurricanes. Threats. Boca Grande Key is >10 km closer to Key West, a tourism hub, than the nearest island in the Marquesas Keys. Human visitation is sometimes high on Boca Grande Key, particularly on weekends (T. Wilmers, unpubl. data), and trampling of dune vegetation has been a long-standing problem. On this island and the Marquesas Keys, the large amount of dead vegetation intermingled with live Blackbead makes the threat of fire, natural or man-caused, a significant threat to the Miami Blue and its habitat. Illicit campfire pits have been found many times over the past 2 decades in both areas (T. Wilmers, pers. observ.). Rising sea level is an acute threat at all the sites that harbored Miami Blues. Six of 8 occupied sites were <1 m above sea level—including Main Beach, where the largest number of Miami Blues was observed—and none was >2 m. Pronounced beach erosion, undercutting, and narrowing the dunes and/or coastal strand, was particularly evident at Short Beach and Boca Grande Key. Although lesser hurricanes like Hurricane Wilma may temporarily enhance Miami Blue habitat by promoting early successional growth, a stronger hurricane could extirpate this species in KWNWR. The last Category 4 or higher hurricane in the Florida Keys was Hurricane Donna in 1960, but its strongest winds occurred more than 130 km from Boca Grande Key. Although Hurricane Wilma (2005) was at most a Category 3 hurricane in the Florida Keys (National Oceanic and Atmospheric Administration 2005), its winds and storm surge washed away a part of the southern dune on Boca Grande Key, creating a breach that has since progressively widened. The number and severity of hurricanes are projected to increase because of global warming (Webster et al. 2006), ominously further threatening the Miami Blue population in KWNWR. Acknowledgments. We are grateful to the Florida Keys Audubon and the Friends and Volunteers of Refuges of the Florida Keys National Wildlife Refuges for financial support. P. Hughes, K. Meyer, and G. Zimmerman provided helpful comments on an 2010 Southeastern Naturalist Notes 853 early draft of the manuscript. M. Minno provided extensive, insightful comments that greatly improved the manuscript. We thank E. Saarinen for editorial assistance. A. Morkill, Project Leader of the Florida Keys National Wildlife Refuges, unwaveringly supported our efforts. Literature Cited Calhoun, J.V., J.R. Slotten, and M.H. Salvato. 2002. The rise and fall of tropical blues in Florida: Cyclargus ammon and Cyclargus thomasi behtunebakeri (Lepidoptera: Lycaenidae). Holarctic Lepidoptera 7:13–20. Carrol, S.P., and J. Loye. 2006. Invasion, colonization, and disturbance: Historical ecology of the endangered Miami Blue Butterfly. Journal of Insect Conservation 10:13–27. Leston, D., D.S. Smith, and B. Lenczewski. 1982. Habitat diversity and immigration in a tropical island fauna: The butterflies of Lignum Vitae Key, Florida. Journal of the Lepidopterists’ Society 36:241–255. McLaughlin, J.F., J.J. Hellmann, C.L. Boggs and P.R. Ehrlich. 2002. The route to extinction: Population dynamics of a threatened butterfly. Oecologia 132:538–548. Minno, M.C., and T.C. Emmel. 1993. Butterflies of the Florida Keys. Scientific Publishers, Gainesville, fl. 168 pp. Minno, M.C., and T.C. Emmel. 1994. Miami Blue, Hemiargus thomasi bethunebakeri Comstock and Huntington. Pp. 646–648. In M. Deyrup and R. Franz (Eds.). Rare and Endangered Biota of Florida. Volume IV. Invertebrates. University Press of Florida, Gainesville, fl. 798 pp. Minno, M.C., and M. Minno. 2009. A plan to conserve rare butterflies in the Florida Keys. Report to the Florida Keys Mosquito Control District, Key West, fl. 193 pp. Minno, M.C., J.F. Butler, and D.W. Hall. 2005. Florida Butterfly Caterpillars and Their Host Plants. University Press of Florida, Gainesville, fl. 341 pp. National Oceanic and Atmospheric Administration. 2005. Hurricane Wilma. Available online at http://www.srh.noaa.gov/key/HTML/wilma/wilma.html. Accessed 15 December 2008. Ruffin, J., and J. Glassberg. 2000. Miami Blues still fly. American Butterflies 8:28–29. Royer, R.A., J.E. Austin, and W.E. Newton. 1998. Checklist and “Pollard Walk” butterflysurvey methods on public lands. The American Midland Naturalist. 140(2):358–371. Saarinen, E.V., and J.C. Daniels. 2005. The ecology and metapopulation dynamics of the endangered Miami Blue Butterfly. The 2005 Entomological Society of America Annual Meeting and Exhibition. December 13–15, 2005, Fort Lauderdale, Florida. Poster presentation. Available online at http://esa.confex.com/esa/viewHandout.cgi?uploadid=439. Accessed 25 September 2007. Webster, P.J., G.J. Holland, and H.R. Chang. 2006. Changes in tropical cyclone number, duration, and intensity in a warming environment. Science 16:1844–1846.