Portal de Periódicos da CAPES

Closing the transport budget of the Florida Straits

2014; American Geophysical Union; Volume: 41; Issue: 7 Linguagem: Inglês

10.1002/2014gl059498

1944-8007

Clément Rousset, Lisa M. Beal,

Ocean Waves and Remote Sensing

Geophysical Research LettersVolume 41, Issue 7 p. 2460-2466 Research LetterFree Access Closing the transport budget of the Florida Straits Clément Rousset, Corresponding Author Clément Rousset MPO/RSMAS, University of Miami, Miami, Florida, USA Correspondence to: C. Rousset, rousset.clem@gmail.comSearch for more papers by this authorLisa M. Beal, Lisa M. Beal MPO/RSMAS, University of Miami, Miami, Florida, USASearch for more papers by this author Clément Rousset, Corresponding Author Clément Rousset MPO/RSMAS, University of Miami, Miami, Florida, USA Correspondence to: C. Rousset, rousset.clem@gmail.comSearch for more papers by this authorLisa M. Beal, Lisa M. Beal MPO/RSMAS, University of Miami, Miami, Florida, USASearch for more papers by this author First published: 20 March 2014 https://doi.org/10.1002/2014GL059498Citations: 6AboutSectionsPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinked InRedditWechat Abstract Closing the mean mass budget of the Florida Straits is fundamental to understanding the source waters of the Gulf Stream and for constraining ocean simulations and forecasts in a region of complex topography. Existing transport time series of the inflow through Yucatan Channel and the outflow at 27°N yield a 9 Sv difference in the mean, which cannot be reconciled by the two minor inflows between. Here we present a synthesis of transport estimates from throughout the Florida Straits, using velocity data collected aboard the cruise ship Explorer of the Seas. Our transports form a consistent mass budget and conform with the northern outflow from the Straits of 32.1 ± 3.3 Sv, as measured continuously by cable. In particular, we find an average of 30.0 ± 5.3 Sv for the inflow to southern Florida Straits and estimate that Old Bahama Channel contributes less than an additional 2 Sv. Key Points The mean inflow in the southern Florida Straits is estimated at 30.0 ± 5.3 Sv The mean transports are consistent from Yucatan Channel to North Florida OBC inflow is less than 2 Sv with eastward flow on top and westward flow below 1 Introduction For more than 40 years, the flow through the Florida Straits (Figure 1), which ultimately feeds the Gulf Stream, has been extensively studied, yet its sources are still uncertain because its mass budget remains unreconciled. The mean inflow to the Straits, as measured by a 2 year mooring array across the Yucatan Channel (23.1 ± 3.1 Sv; Canek program) [Sheinbaum et al., 2002; Candela et al., 2003] (1 Sverdrup = 106 m3/s), is almost 30% less than the outflow transport robustly monitored by 25 years of cable data at 27°N (32.1 ± 3.3 Sv) [Baringer and Larsen, 2001; Meinen et al., 2010]. Existing estimates of the two minor inflows from the east, via Northwest Providence Channel (NPC) and Old Bahama Channel (OBC), can account for only 3 Sv of this large difference [Leaman et al., 1995; Atkinson et al., 1995; Rousset and Beal, 2010; Smith, 2010]. Over the last decade, much community effort has been spent trying to reconcile the Florida cable transport at 27°N with those from moorings in the Yucatan Channel. Figure 1Open in figure viewerPowerPoint Bathymetry (from etopo1: Amante and Eakins [2009]) and currents from Explorer of the Seas ADCP data set. (top) Top bin ADCP currents at all seven sections across Yucatan Channel and Florida Straits. Mean currents at sections Y, Z and at 26°N averaged over (middle) the upper 150 m, and averaged over (bottom) the depths 150–400 m. NWP Ch. and OB Ch. stand for Northwest Providence and Old Bahama Channels, respectively. The mooring location from Atkinson et al. [1995] is drawn as a purple star. Most notably, Hamilton et al. [2005] used 21 moorings throughout the Florida Straits, plus a cable from Key West to Cuba in the southern Straits, to estimate transports across six sections over an 11 month period in 1990–1991. Most sections were sampled rather sparsely by only three moorings. Nevertheless, their results were consistent with the cable at 27°N and with previous estimates of inflows through North West Providence and Old Bahama Channel. The main inflow between Key West and Cuba was estimated to be 25.2 Sv, within 1 standard deviation of the Yucatan mooring transport. However, applying continuity to these flows, Hamilton et al. [2005] found a 3 Sv discrepancy in the mean. They theorized that it was due to an underestimation of the inflow via OBC. In a previous study we showed that velocity data from 90 crossings of the Yucatan Channel, made by acoustic Doppler current profilers (ADCPs) installed in the hull of ship-of-opportunity Explorer of the Seas, suggest that the mean transport is 7 Sv greater than from the mooring array (30.3 ± 5 Sv) [Rousset and Beal, 2010] and therefore consistent with the transport farther north. However, Explorer data suffer gyrocompass biases and miss deep flows below 1200 m, rendering the estimate less than definitive. Furthermore, the mooring-based measurements could be expected to yield a more robust mean transport, owing to better temporal resolution. Hence, we return to the Explorer data for two further measures of transport within the southern Florida Straits to try to settle the issue. We find that transports from all Explorer transections in the southern Straits, representing 120 crossings over 5 years, are statistically consistent and point to a flow from Yucatan Channel of 30.0 ± 5.3 Sv (Figure 1). We also find that the vertical transport structure of the Florida Current, which is well resolved by ADCP, is consistent across these data. The OBC inflow structure and transport is also examined and appears to contribute no more than 2 Sv of mean inflow. All our transports are consistent with the Florida Current cable data at 27°N. 2 Observations and Methods The geography of the Florida Straits and its marginal channels is shown in Figure 1. Between the inflow at 22°N/85–87°W through Yucatan Channel and the outflow at 27°N/79–80°W lie the enclosed Gulf of Mexico and two minor passages, Old Bahama Channel (OBC) just north of Cuba at 23°N/78°W and Northwest Providence Channel (NPC) between the Bahamian Islands at about 26.3°N/78°W. Over a period of 5 years, between 2001 and 2006, the Royal Caribbean cruise ship Explorer of the Seas measured ocean currents along seven different transects throughout the Straits of Florida and Yucatan Channel, representing a total of 276 complete crossings (Figure 1 (upper)). Data were collected over the upper 1200 m and 250 m using two hull-mounted acoustic Doppler current profilers (ADCPs) of 38 and 150 kHz, respectively. Data quality and processing are detailed in Beal et al. [2008] and Rousset and Beal [2010]. South of 26°N, crossing the Straits between Miami and Cuba, we analyze data along two previously unanalyzed Explorer transects, nominally sections Z and Y, as shown in Figure 1 (bottom). These transects represent the inflow transport of the Florida Straits, equivalent to that through the Yucatan Channel, since there are no geographical gaps between (only the enclosed Gulf of Mexico). Section Y also transits through OBC before leaving the Cuban coast and passing east of Cay Sal Bank, allowing a characterization of the minor inflow via OBC from the east. Ship tracks were not always repeated exactly, depending on weather and other factors, so standard transections Z and Y were obtained by dividing tracks into four segments and averaging ADCP profiles, track by track, into bins of length 10 km along those segments. Along section Y, 29 tracks out of 45 contained no gaps (e.g., unsampled 10 km bins), covering the years 2002 and 2003. We used these 29 tracks to estimate the mean and variability of currents and transports. In contrast, along section Z all 93 tracks had measurement gaps of varying extent, owing to large bubble contamination along this long, oblique transection of the current. Therefore, we conservatively combined all the binned profiles to obtain a single composite mean velocity structure and transport estimate for section Z. Therefore, we consider section Z as being fully sampled only once (one "composite" track). Finally, velocity data were detided as explained in Rousset and Beal [2010] and extrapolated to the surface considering uniform velocity over the upper 60 m, as in Beal et al. [2008]. Deep and coastal flows missed by the Explorer measurements are estimated using a 1/12° resolution simulation from the HYbrid Coordinate Ocean Model (HYCOM) [Rousset and Beal, 2011]. One third of section Z, south of 23.8°N and west of 82°W as it enters the Gulf of Mexico, is deeper than the maximum depth sampled (1200 m), and neither section is entirely closed at their southern ends near Cuba. According to the simulation, these missing flows are less than 0.5 Sv in the mean, with one half of this owing to missing deep flows. Transports are estimated from our ADCP transections by cumulating cross-sectional velocity components. The main source of error in the transport calculation comes from the determination of the ship heading by gyrocompass, because of the absence of GPS compass aboard Explorer [Beal et al., 2008]. In the Yucatan Channel and the Straits of Florida at 26°N, Beal et al. [2008] and Rousset and Beal [2010] estimated the gyrocompass bias by comparing the mean transport from all the eastward tracks with that from all the westward tracks. In the Yucatan Channel, biases of +2.7 Sv and −1.3 Sv were found for eastward and westward tracks, respectively. The bias was found to be primarily dependent on the east-west direction and length of the tracks, their exact orientation being less important [Beal et al., 2008]. For the sections discussed here, the ship traveled both eastward and westward along the single composite section Z but only westward along section Y and therefore, we are unable to make a new assessment of gyrocompass bias. Instead, we note that the longitudinal displacement of sections Y and Z is comparable to the Yucatan Channel sections and assume the same bias as obtained previously. The best estimate of the bias correction for each track of section Y is then −2.7 Sv, while at section Z, the proportion of data from eastward tracks (60%) is compared to that from westward tracks (40%) to estimate a mean bias correction of −1.1 Sv. These corrections are applied to each track individually in order to obtain an unbiased estimate of the transports. 3 Florida Straits Sections 3.1 Velocity Structure One of the strengths of our ADCP data over moorings is good cross-sectional coverage. Figure 2 shows the structure of the ocean currents at sections Z and Y, with respect to latitude. Because these sections run along and obliquely to the currents and change their orientation, as does the Florida Current itself as it negotiates the Straits, currents are depicted positive if they are flowing into the Straits and negative if they represent an outflow. Hence, only the eastward surface flow through OBC and the westward flow at the eastern mouth of the Yucatan Channel appear as negative. The structures of the flows across Z and Y are different, largely as a result of their differing orientations and bathymetry. Section Z crosses the southern side of the Florida Current as it flows eastward away from the Gulf and crosses its western side obliquely as it flows northward through the Straits. Section Y is oriented along Old Bahama Channel in the south, before entering the Straits to the east of Cay Sal Bank, at about 23.5°N/79.2°W (Figure 1). In both sections, currents are strongest at the western boundary, above the continental slope off Miami, with maxima over 150 cm/s at the surface, consistent with previous findings [e.g., Schmitz and Richardson [1968]; Niiler and Richardson, 1973; Leaman et al., 1995]. Figure 2Open in figure viewerPowerPoint Ocean speeds (in cm/s) averaged over (left) 4 years (2002–2006) at section Z and over (right) 2 years (2002–2004) at section Y. The coast of Florida is on the left of the panels. Eastward currents near OBC (section Y) and westward currents near Yucatan Channel (section Z) are illustrated by negative values. The approximate locations of Atkinson's current meters are drawn as purple circles on section Y. Section Z exhibits a second current maximum of roughly 90 cm/s at 23.5°N, between Key West and Cuba, with a deep extension (isotach 5 cm/s reaches 1000 m depth), as permitted by deep topography. This feature has been noted before [e.g., Hamilton et al., 2005] and represents water transported eastward from the Loop Current into the Straits (Figure 1). About 27 Sv subsequently turns northward, steered by the western boundary to follow the Straits, while a small fraction (~2 Sv) continues eastward into Nicholas Channel and crosses section Y downstream. The very southern end of section Z is almost at the mouth of the Yucatan Channel, and here a recirculation of the Yucatan Current off Cuba (Figure 1) represents an outflow from the Straits, as characterized by Badan et al. [2005] and Rousset and Beal [2011]. 3.2 Transports Our estimates for the mean transport within the Florida Straits across sections Z and Y are, respectively, 28.4 Sv and 29.3 ± 6 Sv, where the envelope is the standard deviation about the mean. Note that although section Y crosses a large portion of the OBC inflow, which we will describe later, this inflow must flow back out across the section farther north, since ultimately the ship track crosses from Cuba to Miami. As a result, the transports through sections Z and Y are equivalent and we can combine them to give a weighted average (1:29 occupations, respectively) of 29.3 ± 6 Sv. This value lies within one third its standard deviation of previously published results upstream and downstream from the Explorer data set [Rousset and Beal, 2010], including at Yucatan Channel (30.3 Sv). Hence, we find there is mass continuity throughout the Straits without the need to invoke large undetected inflows through Old Bahama Channel, as hypothesized by Hamilton et al. [2005]. In comparison, Hamilton et al. [2005] found a transport of 25.3 ± 4.3 Sv (their sections B + F − G, see Figure 1 and Table 2) and Candela et al. [2003] reported 23.1 ± 3.1 Sv through the Yucatan Channel. Why are our transports inconsistent with these previous mooring estimates? First, we consider the probability that our Explorer results are in fact consistent with these data. Owing to the large variance of the Explorer transports, which is due in part to small gaps in the transections at the coasts and at depth [Rousset and Beal, 2010], it is statistically possible that these estimates are essentially the same. We can quantify this probability by assuming that all transports, T ± σ Sv, are the mean and standard deviation of random variables with normal distributions. We then calculate the probability A of obtaining the difference between two transports, , under the null hypothesis that the transports are the same [Bendat and Piersol, 2000]: (1)where p(z; T, σ2) is the (Gaussian) probability density function. We find that it there is 41% probability that the our YZ transport and that of Hamilton et al. [2005] are different, while there is 78% probability that the two Yucatan Channel estimates [Candela et al., 2003; Rousset and Beal, 2010] are different. There is only 10% probability that the Explorer transports at YZ and at Yucatan are different. In summary, it is statistically very likely that all our Explorer transports are the same and unlikely that the Yucatan mooring array transport is consistent with them. Examining the analysis of Hamilton et al. [2005], there is evidence that they could have underestimated transport across their sections A and B. First, their assumption of constant velocity from the surface down to 145 m depth would severely underestimate the surface flow since ADCP observations show differences in velocity up to +40 cm/s between 140 m and 60 m depth. Recalculating mean transport at sections Y and Z using Hamilton's surface extrapolation method shows a decrease of about 2 Sv. Second, they assume no-slip boundary condition, while near-bottom velocities are 10–30 cm/s on average along our section Y. Thus, Hamilton's bottom extrapolation would also tend to decrease their transport estimate. Calibration of their cable data between Key West and Havana was dependent on these mooring transports. The stronger discrepancy between Explorer and the denser mooring array at Yucatan, where ADCPs covered the upper water column, is harder to understand. Transport per unit depth is a useful way to examine the fundamental vertical structure of the current, which should not change significantly downstream, despite changes in orientation and width influenced by topography. Figure 3 shows mean transport per unit depth profiles over the upper 800 m (minimum sill depth in the Straits) from all Explorer sections across the Straits, compared with that from the mooring array across Yucatan Channel [Candela et al., 2003]. The Explorer sections include the Yucatan Channel inflow in the south, sections Z and Y in the Southern Straits, as presented here, and a section at 26°N (also in Figure 1) [Beal et al., 2008]. All Explorer transport profiles are similar and fall within 1 standard deviation (shading), while the mooring profile is an outlier over the upper water column, where transport is much weaker. This is hard to explain physically and may point to instrumentation issues, as implied in a more detailed comparison of the two Yucatan data sets by Rousset and Beal [2011]. Figure 3Open in figure viewerPowerPoint Mean transport per unit depth (in Sv/m) from the Explorer data set in the Straits of Florida at sections Z, Y, 26°N, and in the Yucatan Channel [from Rousset and Beal, 2011]. The transport from the mooring array in Yucatan Channel is also shown in yellow (Canek) [Candela et al., 2003]. Shading depicts standard deviations about the means. 4 Old Bahama Channel On section Y within OBC (Figure 2), the current's maximum speed is at 300 m depth, in contrast to the surface-intensified Florida Current within the Straits. In fact, the surface current flows eastward and out of the Straits of Florida 85% of the time, while below 150 m depth the stronger subsurface current always flows into the Straits. Maximum mean speeds are 19 cm/s in the surface layer and 45 cm/s in the lower layer. This dual core structure is similar to that observed by recent research cruises at the eastern entrance of OBC, which suggest that the subsurface inflow is likely associated with upper and central waters found at those depths in the vicinity of Windward Passage upstream [Smith, 2010]. The sole continuous current measurements in OBC were obtained with a yearlong single current meter mooring [Atkinson et al., 1995; Hamilton et al., 2005], positioned only 3 km north of our section Y. The location of the mooring is illustrated with a purple star in Figure 1, and the three current meters, positioned at 50, 250, and 435 m depth, are illustrated by purple circles in Figure 2. These recorded mean along-channel speeds of 2.6, 50, and 26 cm/s, respectively, all directed toward the Straits. Hence, the mooring and Explorer measurements both show a subsurface-intensified current, but the weak near-surface currents differ in their mean directionality. Atkinson et al. [1995] found a mean transport of 1.9 Sv estimated from the mooring by interpolating velocities over the 495 m deep and 370 km wide channel using splines with zero velocity at the sides and bottom. Following the small Yucatan Channel transport estimate some years later [Sheinbaum et al., 2002; Candela et al., 2003], the vertical coverage of the current meters on Atkinson's mooring was questioned, since more inflow through OBC was anticipated in order to balance mass within the Florida Straits. The mean transport in OBC was hypothesized to be underestimated due to large drawdowns of the mooring line, which appeared to leave the upper 100 m largely unsampled [Hamilton et al., 2005]. However, according to our data, the mooring was well set up to capture the two current cores in OBC (Figure 2), and since we find that the weak surface flow may exit the Straits on average, capturing more flow close to the surface could have actually reduced Atkinson's inflow transport. We can repeat the transport calculation of Atkinson et al. [1995] by averaging all the Explorer velocity profiles from section Y closest to the OBC mooring site and implementing the same extrapolations and no-slip assumptions. We also consider an extrapolation where velocity is assumed uniform across the Channel. In both cases, OBC inflow transport is less than 0.5 Sv in the mean. According to the HYCOM simulation, missing flows over the very shallow bank (<10 m depth) north of OBC account for only hundredths of a sverdrup. We can make two more estimates of the mean OBC channel transport by simply differencing the Explorer transport from 26°N with the transports estimated herein between (1) Miami and Cuba and (2) through Yucatan Channel. Averaging these, we obtain 1.0 ± 6.4 Sv which, despite high uncertainty, is in fair agreement with previous studies: 1.9 Sv from a single mooring and Pegasus sections [Atkinson et al., 1995; Leaman et al., 1995] and 1.7 ± 1.1 Sv from ADCP data at the eastern mouth of OBC [Smith, 2010]. Overall, our new data combined with all previous studies suggest that the mean transport through OBC has no more than a 2 Sv influence on the Florida Current transport downstream. 5 Discussion and Conclusions The mean flow through Yucatan Channel is a basic constraint for simulations and forecasts of the circulation in the Gulf of Mexico, where numerous oil wells pose threat of environmental hazard, and where fragile reef systems are impacted by spatial and temporal patterns of larval recruitment [e.g., Boehlert et al., 1992; Cowen and Castro, 1994; Swearer et al., 1999]. We find that two previously unexplored velocity transects, representing 30 crossings of the southern Florida Straits, yield a mean transport of 29.3 ± 6 Sv. This transport is equivalent to that in the Yucatan Channel, since only the enclosed Gulf of Mexico lies between, and hence can be combined with 90 previous Explorer crossings [Rousset and Beal, 2010] to give an overall best estimate for the inflow to the Straits of 30.0 ± 5.3 Sv. The standard error is 0.5 Sv. The Student's t test gives 99% confidence that the mean transport lies between 28.0 Sv and 32.0 Sv. The mean flow through Old Bahama Channel is characterized by a two-layer system, with surface current order 20 cm/s flowing eastward and subsurface current order 45 cm/s flowing westward below 150 m depth. Overall, synthesizing our analysis with previous studies, we conclude that the inflow is no more than 2 Sv on average. Interannual variability in the ocean can be significant and has been mooted in the past to explain the discrepancy between the Yucatan moorings and cable transport at 27°N. Neither the cable time series (gap in 1999–2001) nor the Explorer data (2001–2006) overlap in time with the mooring array in Yucatan Channel (2000–2001). However, frequency analysis of 25 years of cable data shows interannual variability of order 1.1 Sv or just 13% of the total variance in the Florida Current [Meinen et al., 2010]. Reconciling the Yucatan mooring transport would require interannual variability 8 times greater there. Yet conservation of mass would dictate that both flows should have similar interannual variability, unless NPC and OBC can carry large differences over sustained periods. Kanzow et al. [2007] showed that mass is conserved over the entire North Atlantic at 10 days and longer. Hence, if there were to be a 5 Sv imbalance between Yucatan and the outflow at 27°N, this would require sustained subsurface velocities in OBC order 4 times those measured by Explorer, or over 1.5 m/s, assuming the balance of the flow is shared evenly with NPC. Therefore, we find it unlikely that interannual variability can explain the Yucatan mooring transport. A more likely explanation is that an unfortunate source of measurement error caused erroneously weak velocities over the top 300 m of the array, as evidenced by the anomalous vertical structure of transport when compared to the 276 Explorer sections across the current (Figure 3). Acknowledgments We thank all the reviewers who carefully read this paper and helped improving it by their very thoughtful comments. The Explorer of the Seas shipboard ADCP program is a collaboration between Royal Caribbean International, RSMAS at the University of Miami, and NOAA AOML and is supported by RCL Ocean Fund. The present work was supported by the National Science Foundation grant OCE 0728897. The Editor thanks two anonymous reviewers for their assistance in evaluating this paper. References Amante, C., and B. W. Eakins (2009), ETOPO1 1 arc-minute global relief model: Procedures, data sources and analysis. NOAA Technical Memorandum NESDIS NGDC-24, 19 pp. Google Scholar Atkinson, L. P., T. Berger, P. Hamilton, E. Waddell, K. Leaman, and T. N. Lee (1995), Current meter observations in the Old Bahama Channel, J. Geophys. Res., 100(C5), 8555– 8560. Wiley Online LibraryADSWeb of Science®Google Scholar Badan, A., J. Candela, J. Sheinbaum, and J. Ochoa (2005), Upper-layer circulation in the approaches to Yucatan Channel, in Circulation in the Gulf of Mexico: Observations and Models, Geophys. Monogr. Ser., vol. 161, pp. 69– 57, AGU, Washington, D. C. Google Scholar Baringer, M. O., and J. C. Larsen (2001), Sixteen years of Florida Current transport at 27°N, Geophys. Res. Lett., 28(16), 3179– 3182. Wiley Online LibraryADSWeb of Science®Google Scholar Beal, L. M., J. M. Hummon, E. Williams, O. B. Brown, W. Baringer, and E. J. Kearns (2008), Five years of Florida Current structure and transport from the Royal Caribbean Cruise Ship Explorer of the Seas, J. Geophys. Res., 113, C06001, doi:10.1029/2007JC004154. Wiley Online LibraryADSWeb of Science®Google Scholar Bendat, J. S., and A. G. Piersol (2000), Random data analysis and measurement procedures, Meas. Sci. Technol., 11(12), 1825, 99– 103. Google Scholar Boehlert, G. W., W. Watson, and L. C. Sun (1992), Horizontal and vertical distributions of larval fishes around an isolated oceanic island in the tropical Pacific, Deep Sea Res., 39, 439– 466. CrossrefADSWeb of Science®Google Scholar Candela, J., S. Tanahara, M. Crepon, B. Barnier, and J. Sheinbaum (2003), Yucatan Channel flow: Observations versus CLIPPER ATL6 and MERCATOR PAM models, J. Geophys. Res., 108(C12), 3385, doi:10.1029/2003JC001961. Wiley Online LibraryADSPubMedWeb of Science®Google Scholar Cowen, R. K., and L. R. Castro (1994), Relation of coral reef fish larval distributions to island scale circulation around Barbados, West Indies, Bull. Mar. Sci., 54, 228– 244. Web of Science®Google Scholar Hamilton, P., J. C. Larsen, K. D. Leaman, T. N. Lee, and E. Waddell (2005), Transports through the Straits of Florida, J. Phys. Oceanogr., 35, 308– 322. CrossrefADSWeb of Science®Google Scholar Kanzow, T., S. A. Cunningham, D. Rayner, J. J.-M. Hirschi, W. E. Johns, M. O. Baringer, H. L. Bryden, L. M. Beal, C. S. Meinen, and J. Marotzke (2007), Observed flow compensation associated with the MOC at 26.5°N in the Atlantic, Science, 317, 938– 941. CrossrefCASADSPubMedWeb of Science®Google Scholar Leaman, K. D., P. S. Vertes, L. P. Atkinson, and T. N. Lee (1995), Transport, potential vorticity, and current/temperature structure across Northwest Providence and Santaren Channels and the Florida Current off Cay Sal Bank, J. Geophys. Res., 100, 8561– 8569. Wiley Online LibraryADSWeb of Science®Google Scholar Meinen, C. S., M. O. Baringer, and R. F. Garcia (2010), Florida Current transport variability: An analysis of annual and longer-period signals, Deep Sea Res., Part I, 57(7), 835– 846, doi:10.1016/j.dsr.2010.04.001. CrossrefADSWeb of Science®Google Scholar Niiler, P. P., and W. S. Richardson (1973), Seasonal variability of the Florida Current, J. Mar. Res., 31, 144– 167. Web of Science®Google Scholar Rousset, C., and L. M. Beal (2010), Observations of the Florida and Yucatan Currents from a Caribbean Cruise Ship, J. Phys. Oceanogr., 40, 1575– 1581, doi:10.1175/2010JPO4447.1. CrossrefADSWeb of Science®Google Scholar Rousset, C., and L. M. Beal (2011), On the seasonal variability of the currents in the Straits of Florida and Yucatan Channel, J. Geophys. Res., 116, C08004, doi:10.1029/2010JC006679. Wiley Online LibraryADSPubMedWeb of Science®Google Scholar Schmitz, W. J., Jr., and W. S. Richardson (1968), On the transport of the Florida Current, Deep Sea Res., 15, 679– 693. CrossrefADSWeb of Science®Google Scholar Sheinbaum, J., J. Candela, A. Badan, and J. Ochoa (2002), Flow structure and transport in the Yucatan Channel, Geophys. Res. Lett., 29(3), 1040, doi:10.1029/2001GL013990. Wiley Online LibraryADSWeb of Science®Google Scholar Smith, R. H. (2010), Atlantic-Caribbean exchange through Windward Passage, M.S. thesis, Univ. of Miami, Coral Gables, Fla. Google Scholar Swearer, S. E., J. E. Caselle, D. W. Lea, and R. R. Warner (1999), Larval retention and recruitment in an island population of a coral reef fish, Nature, 402, 799– 802. CrossrefCASADSWeb of Science®Google Scholar Citing Literature Volume41, Issue716 April 2014Pages 2460-2466 FiguresReferencesRelatedInformation