NEARShORE CONCENTRATION OF pINk ShRIMp (FarFantepenaeuS duoraruM) pOSTLARvAE IN NORThERN FLORIdA BAy IN RELATION TO NOCTURNAL FLOOd TIdE

We address the question of whether the low abundance of juvenile pink shrimp Farfantepenaeus duorarum (Burkenroad, 1939) in northern-central Florida Bay results from (i) limiting environmental conditions, (ii) a reduced postlarval transport, or (iii) both. To explore this question, postlarvae were collected during the new moon in both summer and fall of 2004 and 2005 at six stations located on a transect from the bay’s western margin to its interior. The highest concentrations of postlarvae occurred at two mid-transect stations located in shallow channels with moderate tidal amplitudes (15–20 cm) and dense seagrass beds. At the two interiormost stations postlarval concentrations decreased together with a reduction of the tidal amplitude (≈ 1 cm). Estimates of the cumulative flood-tide displacement with the semidiurnal M2 constituent indicated that the tide moves a maximum of 15 km in four nights, a distance that corresponds to the location of the highest concentrations of postlarvae. The size of postlarvae also reached a maximum at the location of the highest concentrations of postlarvae. Results suggest that postlarvae move into the bay’s interior by a cumulative flood tidal process, advancing onshore during successive nights as far as they can go with the tide. Analyses indicate that, in addition to the tidal amplitude, cross-shelf wind stress and salinity also affect the concentrations of postlarvae. peaks of postlarvae occurred at times of low salinity and strong southeasterly winds. While tidal transport appears to be insufficient for postlarvae to reach Florida Bay’s interior, salinity and winds may also contribute to the observed distribution patterns of early pink shrimp recruits. The pink shrimp Farfantepenaeus duorarum (Burkenroad, 1939) is a seasonally prominent species in south Florida including the southwest Florida (SWF) shelf of the Gulf of Mexico, the Florida keys, and the inshore estuaries, Florida Bay and Biscayne Bay. This species supports a multimillion dollar commercial fishery near the dry Tortugas where landings and catch-per-unit-effort have been highly variable since the early 1980s (Sheridan, 1996). Ehrhardt and Legault (1999) found no clear relationship between offshore spawning stock and inshore juvenile abundance, suggesting that other processes are responsible for determining trends and variability in fishery landings. Understanding the causes of recruitment variability of this ecologically and economically important species is of great relevance to the management of the fishery and the ecosystems where the pink shrimp’s different life stages occur. pink shrimp spawn northeast of the dry Tortugas on the broad SWF shelf (Roberts, 1986). Circulation on the shelf is forced mainly by tides, winds, and buoyancy fluxes (yang and Weisberg, 1999; he and Weisberg, 2002). Subtidal flows are weak and mainly alongshore as a direct response to wind events (Lee et al., 2002). Larvae develop rapidly, passing through 5 nauplii, 3 protozoea, and 3 mysis stages in about 15 d (dobkin, 1961) and three to six additional planktonic postlarval stages in anBULLETIN OF MARINE SCIENCE, vOL. 86, NO. 1, 2010 54 other 15 d (Ewald, 1965). Transport of pink shrimp larvae from offshore spawning grounds to onshore nursery grounds appears to occur in at least a two-stage process, similar to the model proposed for other estuarine species with offshore-onshore migrations (e.g., Boehlert and Mundy, 1988; Miller and Shanks, 2004; Queiroga et al., 2006). during the first days of development larvae may drift with the alongshore current. There is evidence that protozoae migrate vertically with a diel vertical migration (dvM) but a cross-shelf transport with this dvM behavior has not yet been demonstrated (Munro et al., 1968; Jones et al., 1970; Criales et al., 2007). Later in development late myses and postlarvae have been observed to perform rhythmic vertical migrations synchronized with the semidiurnal tide, near Marquesas at about 20 m of depth (Criales et al., 2007). With this behavior postlarvae experienced a considerable onshore flux. This tidal behavior was observed under conditions of strong density gradients and shear due to the presence of non-linear tides. Many estuarine fish and invertebrate species undergo vertical migrations that are coordinated with phases of the tide in order to achieve horizontal movement. This general mechanism is known as Selective Tidal Stream Transport (STST), and more than one behavior has been associated with it (for review see Shanks, 1995; Forward and Tankersley, 2001; Queiroga and Blanton, 2005). A flood-tide transport (FTT) occurs when organisms use the flood phase for shoreward transport and immigration to estuaries (e.g., Morgan et al., 1996; Tankersley et al., 2002; Ogburn and Forward, 2009). An ebb-tide transport (ETT) or ebb-phased migration is used for seaward transport and out-migration from estuaries (e.g., Queiroga et al., 2006; Lopez-duarte and Tankersley, 2007). The FTT is used by pink shrimp postlarvae to enter Florida Bay along its western boundary, as evidenced by their activity in the water column almost exclusively during the night flood tide (Roessler and Rehrer, 1971; Criales et al., 2006). Some other shrimp postlarvae and megalopae crabs that use FTT have shown evidence of advancing into the estuary during successive nights in a type of cumulative or “saltatory” transport (Rothlisberg et al., 1995; Christy and Morgan, 1998). Florida Bay constitutes the primary south Florida nursery ground of the pink shrimp (Costello and Allen, 1966; Costello et al., 1986). The bay is bordered to the east and south by the Upper and Middle Florida keys, to the north by the Everglades and to the west by the inner SWF shelf of the Gulf of Mexico and is comprised of a complex network of shallow basins separated by mangrove islands and extensive mudbanks (Lee et al., 2008). It is a subtropical estuary with unique biological and physical features. Seagrasses, the main habitat of pink shrimp, occur throughout the bay (Robblee et al., 1991; hall et al., 1999). Seagrass development (Zieman et al., 1989) and density of pink shrimp juveniles are greatest in western Florida Bay and decrease with distance into the bay’s interior, with lowest densities of pink shrimp occurring in north-central and eastern Florida Bay (Tabb et al., 1962; Costello et al., 1986). In Florida Bay, estuarine conditions prevail in early winter at the end of the wet season while hypersaline conditions are commonly observed in early summer toward the end of the dry season (Lee et al., 2006; kelble et al., 2007). Extreme hypersaline conditions may occur throughout the bay and all year during extended periods of drought in south Florida. But generally hypersaline conditions are most pronounced, common, and protracted in north-central and eastern Florida Bay, where low densities of pink shrimp have also been reported (Costello et al., 1986; Browder and Robblee, 2009). hypersaline conditions and reduced seagrass habitat in north-central and eastern Florida Bay have been considered potential factors affecting survival and CRIALES ET AL.: FLOOd pOSTLARvAL TRANSpORT 55 growth of young pink shrimp, their abundance, and recruitment to offshore adult populations (Browder et al., 2002; Browder and Robblee, 2009). Another reason for the low abundance of pink shrimp in north-central and eastern Florida Bay may be insufficient tidal transport to carry postlarvae into the interior of the bay. Florida Bay tides are extremely variable because of the interaction of Gulf of Mexico mixed semidiurnal tides and the primary semidiurnal tides of the North Atlantic Ocean entering through channels in the Florida keys (Wang et al., 1994; Smith, 2000, 2005). The bay’s complex bank structure serves to attenuate tidal amplitude and reduce net tidal flow and thus transport. The greatest tidal exchange occurs across the bay’s western boundary (Wang, 1998; Smith, 2000; Lee et al., 2002) where at its northwest corner tidal currents are strong and reach amplitudes of almost 40 cm (Smith, 1997). With distance into the bay, tidal amplitude is reduced to a minimum of about 1 cm, and water level is largely influenced by wind (Lee et al., 2008). In this study we hypothesized that pink shrimp postlarvae, using flood-tide transport (FTT) to move into the bay, would be most abundant at the bay’s western margin where tidal amplitude and currents are greatest and habitat most suitable and decrease with distance into the bay. Our alternative hypothesis was that by rising into the water column during the night flood tide postlarvae would progress into the bay during successive nights in a type of cumulative or “saltatory” transport until tidal transport fails, a transport mechanism that has been documented for other crab and shrimp species (Rothlisberg et al., 1995; Christy and Morgan, 1998). postlarval concentrations might also be reduced due to more hypersaline conditions and less seagrass coverage in the interior of the bay. By exploring these hypotheses, we address the question of whether the abundance of juvenile pink shrimp in the bay’s interior is limited by the ability of planktonic postlarvae to reach this part of the bay or by the availability of favorable habitat conditions. Material and Methods Sampling Stations.—pink shrimp postlarvae were sampled at six stations in northern Florida Bay arranged along a west-east transect of about 31 km extending from Middle Ground on the bay’s western margin to the interior of Florida Bay at dump keys (Fig. 1). Stations 1 and 2 were located in the wide and relatively unconfined western portion of the main channel that connects the bay with the SWF shelf at Middle Ground and at navigational Marker 6, respectively. Station 3 was located in southeast Chonchie Channel, a branch of the main channel, and station 4 was located east of Joe kemp key where the main channel penetrates the western margin of Snake Bight Bank; both stations were located in relatively narrow channels confined by shallow seagrass-covered banks. Station 5 was located on the eastern margin of Snake Bight Bank at Buoy key. Station 6 was located at dump keys in the short pass between eastern Rankin Lake and Whipray Basin about 26 km from Middle Ground and 31 km from the western margin of Florida Bay in the vicinity of Cape Sable. The depth among sampling stations decreases from 3.0 m to 0.9 m from west to east along the transect (Table 1). Benthic habitat at stations 1–3 was characterized by shells, sand, and scattered seagrass beds, and at stations 4–6 by mud, sand, and dense seagrass. The seagrasses Thalassia testudinium Banks & Soland. ex koenig (turtle grass), Syringodium filiforme kütz., (manatee grass), and Halodule wrightii Aschers. (shoal grass), were present along this transect. Monthly Sampling.—Results from our previous studies indicated that the immigration of pink shrimp postlarvae in northwestern Florida Bay has a strong seasonal pattern with a major peak in summer (Criales et al., 2006). There is also evidence that postlarvae are most abundant entering Florida Bay during the new moon dark-flood period (Roessler and Rehrer, BULLETIN OF MARINE SCIENCE, vOL. 86, NO. 1, 2010 56 1971; Criales et al., 2006). Based on these earlier results, sampling of postlarvae was conducted monthly during two or three nights around the new moon from July through November in 2004 and from June through October in 2005. Sampling ended after one sampling day in October 2005 because of bad weather. plankton samples were collected with moored channel nets that swing in response to the ebb and flood of tidal currents. Moored channel nets of 0.75-m2 opening, 1-mm mesh size, and 500-mm mesh in the cod-end were suspended at each station for about 12 hrs, from dusk until shortly after sunrise each night. Sampling depth was regulated by the length of the float ropes that suspended the nets in the water. Nets sampled below the surface (subsurface) at a mean depth of 0.65 m at the three deeper western stations (stations 1–3) and at a mean depth of 0.30 m at the three shallower interior stations (stations 4–6). A General Oceanic flowmeter (low speed rotor) was suspended in the mouth of each net to measure the volume of water filtered through the net and current velocity. pink shrimp postlarvae were sorted from the drift seagrass and algae collected in the net, concentrated in a 0.5 mm sieve, and preserved in 90% ethanol. Surface sea water temperature and salinity were recorded every day and night of sampling at each station during 2005 using a ySI model 85 meter. Carapace length (CL), total length (TL), and the number of dorsal rostral spines (dRS) were measured and counted on 208 postlarvae selected randomly from the six sampling stations to estimate the relative size of postlarvae along the transect and to test our hypotheses. TL was measured (to the nearest 0.1 mm) from the tip of the rostrum to the posterior border of the Figure 1. Map of the study area depicting sampling stations 1–6 (circles, large numbers) in northern Florida Bay extending from Middle Ground (1) on the bay’s western margin to the interior at Dump Keys (6); stations from Smith (1997) (rectangles, small numbers) used to calculate tidal excursion; Murray Key and Buoy Key stations (triangles) from South Florida Natural Resources Center (SFNRC) DataForEVER Dataset, Everglades National Park; and Long Key C-MANstation (in small map). CRIALES ET AL.: FLOOd pOSTLARvAL TRANSpORT 57 telson, and CL was measured (also to the nearest 0.1 mm) from the postorbital margin of the carapace to the mid-dorsal posterior margin of the carapace, with an ocular micrometer. The number of dRS on pink shrimp postlarvae has been considered a measure of length of planktonic life because postlarvae gain a rostrum dorsal spine with each molt until they become benthic (dobkin, 1961; Allen et al., 1980). Analysis of data.—Wind data from the Coastal Marine Automated Network C-MAN at Long key in the Middle Florida keys were analyzed for the period June 1 to November 30, 2004 and 2005. hourly wind data were slightly smoothed with a 3-h low-pass filter then filtered with a 40-h low-pass filter to remove tidal and sea breeze influences and to reveal low-frequency (subtidal variations). Wind vectors were rotated 180° (direction toward) and resolved into cross-shelf, u = to east (+) and to west (−), and alongshore, v = to north (+) and to south (−) components. Wind vectors were converted to wind stress (the force applied to the sea surface by the wind), calculated as: , C u v V , x y a d x t = ^ ^ h h where τ(x,y) corresponds to the cross-shelf and alongshore wind stress (dynes cm –2) respectively, ρa is air density (1.3 × 10 –3 g cm–3), Cd is a drag coefficient (1.5 × 10 –3), V is the wind velocity vector, and u and v are cross-shore and alongshore wind components respectively. Because the effect of wind on postlarval supply results from the cumulative effect of wind events over several days, wind stresses were constructed over a period of 7 d. previous research indicated that wind events occur on time scales of about 1 wk (Lee and Williams, 1999). Time series of 15-min water temperature, salinity, and water level data for the period of June 1 to November 30, 2004 and 2005 were obtained for two monitoring stations along our transect, Murray key and Buoy key (South Florida Natural Resources Center, SFNRC, dataForEvER dataset, Everglades National park, homestead, FL). Environmental data from Murray key were used in stations 1–3 analyses as Murray key was only 1–8 km distant from these western stations, and Buoy key environmental data were used in analyses of stations 4–6 as the Buoy key monitoring location was only 0–7 km from these eastern stations (Fig. 1). data from the two monitoring stations in 2005 have several gaps because three major hurricanes passed through the region, destroying data recording instruments. Available data were smoothed slightly with a 3-h low-pass filter and subsampled at hourly intervals. precipitation data for Flamingo, Everglades National park, and information on tropical storms and hurricanes during 2004 and 2005 were obtained from the NOAA National Climatic data Center (http://www.ncdc.noaa). We calculated the tidal excursion (ti ) or tide-induced movement of water along our transect using tidal amplitude data from Smith (1997). Based on water level data, Smith (1997) quantified the principal constituents at 37 locations in Florida Bay. Our sampling stations were located close to Smith’s (1997) stations: 1 = 8105N, 4 = Marker 5, 14 = Murray key, 18 = Buoy key, and 19 = Whipray Basin (Fig. 1). Amplitudes (in cm) calculated from water levels at the two most western stations (1 and 4) compared favorably with amplitudes calculated from Table 1. Stations sampled in northern Florida Bay during new moon periods from July to November 2004 and June to October 2005 with their respective depth and volume of water filtered calculated from flowmeter readings. Station Location Depth (m) Volume of water (m3) mean ± SD 1 Middle Ground 3.0 2,315 ± 625 2 Mark 6 2.5 2,946 ± 1,428 3 Conchie Channel 2.0 1,850 ± 967 4 Joe Kemp 1.5 539 ± 373 5 Buoy Key 1.0 207 ± 90 6 Dump Keys 0.9 510 ± 208 BULLETIN OF MARINE SCIENCE, vOL. 86, NO. 1, 2010 58 available current meter data (in cm s–1, Smith, 2000; N. Smith, harbor Branch Oceanographic Institute, unpubl. data). however, current meter data from the interior stations were not available for comparison. Given the tidal amplitudes and period of the dominant semidiurnal M2, we calculated the tidal excursion (ti ) at each of the six stations. The tidal excursion (ti ) was given by: