|Scientific Name:||Hippocampus zosterae Jordan & Gilbert, 1882|
Hippocampus rosamondae Borodin, 1928
Hippocampus regulus Ginsburg, 1933
|Taxonomic Source(s):||Jordan, D. S. and Gilbert, C. H. 1882. Notes on fishes observed about Pensacola, Florida, and Galveston, Texas, with description of new species. Proceedings of the United States National Museum 5(282): 241-307.|
|Red List Category & Criteria:||Least Concern ver 3.1|
|Assessor(s):||Masonjones, H., Hayashida-Boyles, A. & Pollom, R.|
|Reviewer(s):||Rose, E. & Ralph, G.|
Hippocampus zosterae is a coastal seahorse species that inhabits seagrass and other submerged vegetation from Bermuda, the Bahamas, and northeastern Florida and the Gulf of Mexico to the eastern Yucatan. The major potential threats to this species include exploitation for the aquarium trade and seagrass habitat loss. Ample research studies and fisheries-independent monitoring data exist to describe patterns of population change in Hippocampus zosterae extending back 20 years for the northern Gulf of Mexico and around the Atlantic coast of Florida (Florida is 46% of their geographic range). Based on a thorough analysis of available data, overall, populations appear stable, with recently stable and/or increases in seagrass habitat cover throughout areas of their range. Therefore Hippocampus zosterae is listed as Least Concern.
Monitoring is required for the southern Gulf of Mexico portion of their range to get an estimate of population density and population stability over time. In addition, in the area of their densest distribution through their range (Florida Bay/Florida Keys), however, there has been a recent wide-spread die off of seagrass (incident began 2015), and thus continued monitoring throughout their range and timely re-assessment of their status is warranted.
|Previously published Red List assessments:|
|Range Description:||Hippocampus zosterae is distributed in the western Atlantic from Bermuda, the northern Bahamas, and northeastern Florida south along the U.S. coast, into the Gulf of Mexico from the Florida Keys north along the Gulf coast, and south and east to Cancun, Mexico (Lourie et al. 1999, Musick 2000, Baum and Vincent 2005, R. Robertson pers. comm. 2014). Locations for records off Cuba require verification. It is found to depths of 10 m (Lourie 2016). Its estimated global AOO is 23,188 km² (calculated by clipping the distribution polygon to the seagrass layer extracted from WCMC 2005, version 3).|
Native:Bahamas; Mexico (Campeche, Quintana Roo, Tabasco, Tamaulipas, Veracruz, Yucatán); United States (Florida, Louisiana, Texas)
|FAO Marine Fishing Areas:|
Atlantic – western central
|Range Map:||Click here to open the map viewer and explore range.|
|Population:||Summary: Data provided from both research studies and fisheries independent monitoring programs indicate that although they are stochastic and varying in density by location, overall, H. zosterae populations are stable throughout Florida (46% of their AOO - 10,700 km²). In addition, in analysis conducted of fisheries-independent monitoring data for the last 10 years (more than 3 generations for this short-lived species) from 6 subpopulations around the state of Florida, only one has evidence for population decline (Indian River Lagoon, on the Atlantic Coast of Florida), and across all watersheds, populations are stable (10 year analysis of monitoring data, slope = -0.00073; Figure 1).|
Supplementary Information: There have been a variety of research studies and fisheries independent monitoring conducted on H. zosterae throughout their range, using a variety of gear types and inconsistent sampling frequencies and duration's. Throughout this section, reported seahorse densities are low, but this has been observed broadly across seahorse species (Foster and Vincent 2004) and is strongly related to gear specificity in H. zosterae because of its small body size. Between 1950-1951, a total of 1587 specimens were collected off Cedar Key, Florida, but no estimate of sampling effort was possible from the study, and until monitoring began in 1996 (see below), no follow-up work was conducted (Strawn 1958). In a study conducted in 1996 in Laguna Madre, Tamaulipas, Mexico, 6 H. zosterae were observed across 81 sampling stations, but sampling was conducted to survey fish diversity rather than quantify density (Raz-Guzman et al. 2002). No continued monitoring occurred in Laguna Madre to be able to describe population trends.
In Florida Bay, the densest area of their range (Figure 2), seagrass covers roughly 6,600 km², which is 61% of H. zosterae’s range in Florida and 28% of its overall global distribution (based largely on the Florida Keys and Florida Bay). In the mid 1980’s, a total of 475 specimens were captured, with densities observed of 0.1 (1984), 0.27 (1985) and 0.5 (1985) seahorses/m² (Sogard et al. 1989). Due to concerns in the recovery of fish populations following the seagrass die-off in Florida Bay during the late 1980's, Sheridan and colleagues (1997) used similar sampling techniques, and observed H. zosterae densities of 0.4 (Thalssia seagrass), 0.1 (mudflat), 0.1 (algae), and 0.3 (Halodule seagrass) animals per m² across habitat types, showing level population densities to those seen before the die-off. Matheson and colleagues (1999) compared their results in the same locations with the same gear (throw traps) in the 1990’s to those of Sogard et. al (1989), observing 0.37 animals/m² in 1994/1995 combined, again showing that across habitat types, seahorse densities were stable. Using otter trawls in deeper areas, Thayer et al. (1999) found level population sizes for H. zosterae in sampling in 1984-5 and 1994-6, observing densities of 0.00186 and 0.00177 animals per m². Using the same gear type in 2006-2009, 978 specimens were collected, with an overall density observed of 0.0039 animals per m² sampled on mud banks (Flaherty et al. 2013). These last two studies did not provide data comparable to Sogard et al. (1989), Sheridan et al. (1997), or Matheson et al. (1999), with different and less efficient sampling gear in deeper areas of Florida Bay, which is what most likely led to the two orders of magnitude lower population sizes measured.
In a study conducted in Tampa Bay, Florida between 2005-2007, there were a total of 922 specimens recorded with an overall density of 0.08 (0.009) animals/m², and a range of 0.02 animals/m² (2006 dry season) to 0.18 animals/m² (2007 wet season) (Masonjones et al. 2010). Over this short-term period, densities of H. zosterae significantly decreased. However, with continued population monitoring in Tampa Bay through 2013, we detected a level population size and no evidence of population decline (Figure 3; Masonjones et al. in prep).
Fisheries independent monitoring (FIM) work in progress through the Florida Fish and Wildlife Conservation Commission investigating seven watersheds in Florida provides population data for H. zosterae for the years 1996-2014 (Dunham et al., in prep). These data are critical because they comprise areas of densest H. zosterae distribution and they provide information about changes in population size over time (Figures 2, 4). Data were collected using 21.3m bay seines along shorelines in submerged aquatic vegetation and a 6.1m otter trawl in deeper areas, combined collecting data from 1.5-7.6 m depths (Table 1). Across the time period, no dwarf seahorses were observed in the Jacksonville (JK) area bay system on the northeast Florida coast, providing a firm confirmation of the northernmost limit of their distribution. For the other six watersheds, Indian River Lagoon (IR), Florida Bay (FB), Charlotte Harbor (CH), Tampa Bay (TB), Cedar Key (CK), and Apalachicola Bay (AP), taken together, boosted regression methods indicate stable populations across the 19 year time span with 3078 animals observed (Figure 5). Further, analysing the same data over 10 years indicates that populations are stable, with number of animals caught per sample (area ) over time constant (Figure 6; 10 year analysis of monitoring data, slope = -0.00073; Masonjones et al. unpublished, independent analysis of FIM data for environmental correlates with their distribution; Figure 7, same as Fig. 6, but with FB removed because limited data available). Taken individually, both CH and TB populations appear stable (Dunham et al. in prep; Masonjones et al. unpublished), but neither AP or CK had enough data over time to model population stability using boosted regression or indices of abundance. FB populations, although studied for a shorter period of time (2006-2009), show a dramatic increase in population size using boosted regression, with 29.2% of the variation in seahorse presence/absence explained by year (Masonjones et al. unpublished). Caution should be used in terms of being optimistic about the stability of the FB population, considering the ongoing seagrass die-off in Florida Bay that began during 2015 (see Threats). Indian River Lagoon demonstrated a significant population decline, with few animals seen in the last 4 years of the sampling period, and boosted regression indicating that 16.1% of the variation in presence/absence was explained by year (Figures 8, 9; Dunham et al. in prep; Masonjones et al. unpublished). This is correlated with a significant loss in seagrass cover and water quality in Indian River Lagoon (see Threats). In addition, the state of Texas also has a fisheries independent monitoring program focused on 8 coastal regions organized north to south: 1. Sabine-Neches Estuary (Sabine Lake), 2. Trinity-San Jacinto Estuary (Galveston Bay), 3. Lavaca-Colorado Estuary (Matagorda Bay), 4. Guadalupe Estuary (San Antonio Bay), 5. Mission-Aransas Estuary (Aransas Bay), 6. Nueces Estuary (Corpus Christi Bay), 7. Upper Laguna Madre Estuary (Upper Laguna Madre), and 8. Lower Laguna Madre Estuary (Lower Laguna Madre) (Texas Parks and Wildlife Department FIM data). Between 1977 and 2010, 56 dwarf seahorses were collected, with zero observed in sites 1 and 2, 4 at site 3 (1987-2003), 5 at site 4 (1985-2010), 8 at site 5 (1981-2010), 5 at site 6 (1977-1998), 27 at site 7 (1979-2010), and 7 at site 8 (1986-2010). Because of the sparse dataset, statistics for population trends were not possible, so additional monitoring in Texas is warranted.
One of the biggest problems in assessing the population status of H. zosterae was gear selectivity; because of their small size, cryptic habit, and ability to hold on to the substrate, their catch rates can vary independently of their actual population size. To help address this, Masonjones and colleagues (in prep) conducted a long term mark-recapture study (2005-2009) and then a short-term, more focused study in 2010 (Masonjones and Masonjones, in review). In the initial study, using pushnets across large (100 m diameter) study plots, recapture rates were so low (0.33%) that population estimates were not possible. In the follow-up study with smaller plot sizes (45m diameter) and sampling plots to depletion, recapture rates increased to 33%. Survival across ten weeks was high (φ = 97.8± 1.0% ), with relatively low recapture probability (p = 17.9 ± 4.7%). POPAN was used to estimate population size across the entire 2006 m² area surveyed at 190.5±70.0 individuals, which provides an estimated population density across the study plots of roughly 0.095±0.035 seahorses m². This estimated value based on the POPAN results was three times higher than that calculated from the actual collection process in this study (0.032 seahorses m-²). This study highlighted the detectability issue with dwarf seahorses, because traditional collection methods rarely catch them, and only very small mesh nets with focused collection locations work to recapture animals reliably for population estimates.
|Current Population Trend:||Stable|
|Habitat and Ecology:||Hippocampus zosterae is found in floating vegetation and shallow seagrass flats (Lourie et al. 1999). Across seagrass species in both Tampa Bay and the Florida Keys, seahorses do not appear to demonstrate a preference, found evenly across turtlegrass (Thalassia testudinum), manatee grass (Syringodium filiforme), and mixed species seagrass beds (Thalassia, Syringodium, and Halodule wrightii (shoal grass); Masonjones et al. 2010, Masonjones et al. in prep). In addition, males, females, and juveniles use seagrass habitats consistently, with no stratification across space or time. Copepods make up the majority of their diet, and they use a “sit-and-wait” strategy of capturing prey (Tipton and Bell 1988). |
Overall, both populations and body sizes were higher during wet seasons (May-October) than dry seasons (November-April; Masonjones et al. 2010). Across the three sites surveyed in Tampa Bay over time, sex ratios were female-biased, ranging from 0.35 to 0.40, based on a total of 841 animals across 2 years. In a continuation of this same study using similar collection methods, sex ratio remained female-biased at 0.40 across 1070 animals observed through 2010 (Masonjones et al. in prep). A mark-recapture study following movement patterns across spring 2010 in 64 animals, females displayed higher recapture rates (35%) and a higher probability of multiple recaptures, but genders moved similar distances from collection to collection, indicating small movements and supporting patterns observed in other seahorse species of small home range sizes and high site fidelity (3.2 (0.6) m, (Masonjones and Masonjones in review, Vincent and Sadler 1995, Perante et al. 2002, Vincent et al. 2005). They are found at low population densities (Masonjones et al. 2010), are genetically monogamous (Rose et al. 2014), and exhibit complex courtship displays (Masonjones and Lewis 1996). Like other seahorses, this species is ovoviviparous and the males brood the embryos in a pouch prior to giving live birth (Foster and Vincent 2004). Compared to other seahorse species, fecundity is low, with the largest clutch size observed being 55 offspring (Foster and Vincent 2004, Strawn 1958). They are estimated to live for 1 year (Foster and Vincent 2004).
Population genetic data indicates significant population structuring of Florida dwarf seahorse populations (Fedrizzi et al. 2015), with differences detected between animals sampled in Pensacola (near Apalachicola Bay sampling for FIM data above), the west coast of Florida (near Tampa Bay), the Florida Keys (near Florida Bay), and on the east coast in Indian River Lagoon. These populations have low rates of gene flow with distant populations, but for closer populations, rafting on floating vegetation is the most likely explanation for how recruits arrive, because distances are beyond the possible migration distances for this small species.
|Continuing decline in area, extent and/or quality of habitat:||No|
|Generation Length (years):||0.33|
|Movement patterns:||Not a Migrant|
|Use and Trade:||
Hippocampus zosterae is collected and traded for aquarium use. The numbers of live Hippocampus zosterae that were traded internationally and recorded in the CITES trade database prior to 2010 are 1572 from the wild, and 584 from unknown origins (Koldewey and Martin-Smith 2010). Several hundred live animals have been exported from the US since then (CITES 2016).
Hippocampus zosterae is one of the more popular seahorses in the aquarium trade (Vincent 1996, Wood 2001, Vincent et al. 2011). Florida has a small directed trawl fishery in shallow grass beds off the west coast for H. zosterae where they are landed in a live bait trawl fishery. It previously occupied the 2nd rank of the top 10 fishes exported from Florida for the aquarium trade (Wood 2001). For the period 1990-2014, mean annual commercial catch for the aquarium trade was 38,563 (SE 6015) seahorses, with the peak catch observed of 98,779 animals seen in 1994 (FWC 2016). However, collection limits of 400 fish per day were imposed by the state of Florida in late 2009, and although catch data from 1991-2009 averaged 46,158 (SE 6925) live animals annually, this number dropped to 17,250 (SE 2345) animals per year in the period from 2010-2014 (Florida Fish and Wildlife Commission, 2016). To date, almost a million animals have been commercially harvested from the state of Florida. Harvest in the state of Florida is focused on the west coast between southern Florida (Everglades and the Florida Keys) up to Tampa Bay, with 80% of the trade focused on the region between Tampa Bay and Ft. Myers (FWC 2016).
|Major Threat(s):||This species may be particularly susceptible to decline due to seagrass degradation. Seagrass beds are susceptible to degradation from coastal development and polluted runoff (Waycott et al. 2009, Short et al. 2011). However, recent analysis has indicated that the three species of seagrass H. zosterae are found associated with (Thalassia testudinum, Syringodium filiforme, and Halodule wrightii) are all listed as Least Concern, with evidence for stable or increasing seagrass cover in many places in their range (Short et al. 2011). Florida has the largest coastline along which H. zosterae are distributed, and the largest and densest populations observed of this species are in southern Florida, in Florida Bay. Extensive seagrass die-offs occurred across Florida Bay beginning largely in 1987, and although there was a 5% increase from the mid 1990’s to 2011 in seagrass cover in this location, another massive seagrass die-off began in 2015-2016, affecting over 10% of the existing seagrass habitat in Florida Bay (Hall and Carlson 2015, South Florida Natural Resources Center 2016). Considering the largest proportion of this species' population is concentrated in this area and it is dependent on seagrass for habitat, this represents a major threat. For other regions of Florida, which correspond to the subpopulations of H. zosterae discussed under Population, patterns in seagrass abundance are variable. Seagrass declines are continuing for Indian River Lagoon (1940-2013, 30% decline; Morris 2015), Apalachicola Bay (seagrass declining, unknown rate; Mezich and Smith 2015), and Cedar Key (seagrass declining, unknown rate, Jones et al. 2015). Seagrass increases have been observed in the Charlotte Harbor (2006-2014, 10% increase, Brown et al. 2016) and Tampa Bay regions (1982-2014, 86% increase; Sherwood and Kaufman 2016), where seagrass area is similar to that seen in 1950 (163 km²). Across the species range in the Gulf of Mexico, however, there has been an overall decrease in in seagrass cover (Karnauskas et al. 2013).|
The entire genus Hippocampus is listed in Appendix II of CITES, but there are no species-specific conservation measures in place for Hippocampus zosterae. Full monitoring of the trade is underway in the United States, however this is dependent on traders’ declarations. Seahorses are listed under Title 68 (Rules of the Fish and Wildlife Conservation Commission) of the Florida Administrative Codes. The targeted fishery for the aquarium trade in Florida is monitored and regulations are in place, such as a limitation on the number of commercial harvesters, however the non-selective exploitation is not monitored in any state. Stock assessments are needed in order to evaluate the sustainability of the fishery and establish appropriate management guidelines. Further research on this species biology, ecology, habitat, abundance and distribution is needed.
A petition to list H. zosterae as threatened or endangered under the US Endangered Species Act (ESA) has been presented based on the over-utilisation of the species, the loss and degradation of its habitat, the inadequacy of existing regulatory mechanisms, and other factors. The petition was initiated after the Deepwater Horizon oil spill in April 2010 in the northern Gulf of Mexico. The ESA document was initiated in 2012, completed July, 2016, and is currently (as of Sept. 28th, 2016) in review with NOAA (C. Horn, NOAA, personal communication 2016).
Non-species specific conservation measures in place that protect large areas of their habitat in the state of Florida include harvest closures for all Marine Life Species (Florida designation) in the Florida Keys National Marine Sanctuary (from Ecological Reserves, Special- use Research Only Areas, and Sanctuary Preservation Areas), Biscayne National Park, Dry Tortugas National Park, and Everglades National Park, together which comprise most of Florida Bay, the densest area of their distribution. (FWC 2016). Seahorse commercial harvest is also excluded from John Pennekamp Coral Reef State Park in the Upper Florida Keys. In addition, shrimp trawling closures are in effect from July 1 through August 31st each year in the Big Bend region of Florida, comprising roughly 28% of the coastal Florida seagrass cover, and areas shown in fisheries independent monitoring to have dwarf seahorses (see Apalachicola Bay and Cedar Key in Population section, Mezich and Smith 2015).
The state of Florida has proposed further limitations on harvest of dwarf seahorses in addition to the catch limit imposed in 2009 of 400 animals per trip. The additional regulations being considered are 1) annual harvest limits of 25000 animals, 2) prohibition of harvest north of Anclote Key State Park on the Florida west coast (28.216018, -82.849746) and north of Jupiter Inlet on the Florida east coast (26.948255, -80.084687), 3) reduce the recreational catch limit to a total of 5 seahorses (independent of species) and an annual closure for recreational seahorse fishing from July to September each year, 4) annual closure for commercial dwarf seahorse fishing from July-September each year, and 5) decrease bag limits per person or vessel (whichever is smaller) to 200 animals per day (NOAA Fisheries Service 2015). Based on an analysis of trip data, the trip/bag limit reduction to 200 animals with a full closed season and area closures described would reduce their collection by 61%, and just reduction bag/trip limit would reduce collections by 18%. These proposed regulations have been through 2 rounds of public discussion, and final decisions are pending based on the decision of NOAA to list them under the US Endangered Species Act (N. Sheridan, FWC, personal communication).
|Citation:||Masonjones, H., Hayashida-Boyles, A. & Pollom, R. 2017. Hippocampus zosterae. The IUCN Red List of Threatened Species 2017: e.T10089A46910143.Downloaded on 21 March 2018.|
|Feedback:||If you see any errors or have any questions or suggestions on what is shown on this page, please provide us with feedback so that we can correct or extend the information provided|