Volume 152, Issue 2 p. 253-269
ARTICLE
Open Access

Timing and route of migration of mature female blue crabs in a large, wind-driven estuary

Geoffrey W. Bell

Geoffrey W. Bell

Department of Marine, Earth and Atmospheric Sciences, North Carolina State University, Raleigh, North Carolina, 27607 USA

Search for more papers by this author
David B. Eggleston

Corresponding Author

David B. Eggleston

Department of Marine, Earth and Atmospheric Sciences, North Carolina State University, Raleigh, North Carolina, 27607 USA

Center for Marine Sciences and Technology, North Carolina State University, Morehead City, North Carolina, 28557 USA

Correspondence David B. EgglestonEmail: [email protected]Search for more papers by this author
First published: 30 March 2023
Citations: 1

Abstract

Objective

Quantify the timing of mature female blue crab Callinectes sapidus migration and the routes they take in the Croatan, Albemarle, and Pamlico Estuarine System (CAPES) in North Carolina, USA.

Methods

Mark–recapture data collected by commercial crabbers in northeastern North Carolina during 2008 and 2009 identified the timing, rate, and route of movement for 1049 postcopulatory female blue crabs out of 8400 tagged (17.5% recapture).

Result

Mature female blue crabs consistently moved from their original, up-estuary release sites towards the Oregon Inlet spawning sanctuary in early summer. Although crabs averaged 2 km/d and covered distances in excess of 50 km during the tagging season, no recaptures were reported from inside or within 10 km of inlet spawning sanctuaries connecting the CAPES to the Atlantic Ocean. The vast majority of recaptures were concentrated within the area where Albemarle, Currituck, and Croatan sounds merge, which is likely due to a combination of high fishing effort and the narrowing of the waterways that concentrate mature females as they migrate south towards spawning sanctuaries.

Conclusion

The high concentration of blue crab recaptures in this region suggests that extending no-take, inlet sanctuary boundaries into these waters may protect mature females as they migrate to the inlet spawning grounds.

Abstract

Impact statement

Marine fish and shellfish that spawn in predictable locations are particularly vulnerable to overfishing when fishers target their spawning grounds or along regular migration routes leading to spawning grounds. Through a collaborative mark–recapture study with commercial crabbers, we identified the timing, rate, and location of migration by mature female blue crabs, whose population numbers are at historic lows. Extending no-take spawning sanctuaries to locations with high concentrations of mature females may reduce fishing mortality and enhance spawning success.

INTRODUCTION

Marine fish and shellfish that spawn in predictable locations are particularly vulnerable to overfishing when fishers target their spawning grounds or along regular migration routes leading to spawning grounds (Coleman et al. 1996; Lipcius et al. 2003; Dew and McConnaughey 2005; Nemeth et al. 2007; Lascelles et al. 2014). Information on migration patterns is critical to temporal and spatial management tools, such as seasonal closures, as well as no-take migration corridors and spawning sanctuaries that aim to protect the spawning stock of mobile, commercially exploited species (Roberts and Polunin 1991; Lipcius et al. 2003; Lambert et al. 2006; Eggleston et al. 2009; Lascelles et al. 2014; Dunn et al. 2019).

Linking no-take migration corridors to spawning sanctuaries has shown promise in rebuilding and conserving spawning stocks of endangered or exploited species (Lipcius et al. 2003; Fernandes et al. 2005; Guzman et al. 2008; Pendoley et al. 2014). For example, migration and spawning sanctuaries for blue crab Callinectes sapidus in the Chesapeake Bay, USA, were originally ineffective at protecting the spawning stock (~16% protected); however, the subsequent expansion of sanctuaries to include migration corridors for postmating females increased the proportion of the spawning stock that was protected by 70%, leading to a resurgence in the spawning stock (Lipcius et al. 2003; Lambert et al. 2006). Fishery-independent trawl survey data on blue crabs informed the designation of the deepwater (>13-m depth) sanctuary and corridor system in the tidally driven Chesapeake Bay (Lipcius et al. 2003; Lambert et al. 2006). Working in a relatively small, tidally driven estuary in North Carolina (White Oak River), Eggleston et al. (2015) used a combination of passive and active acoustic tracking of blue crabs and found that blue crabs traveled approximately 14.1 km, mainly in deeper channels, and over a period of 12–26 days from mating areas to spawning grounds. In contrast, the Croatan, Albemarle, and Pamlico Estuarine System (CAPES), the second-largest estuarine system in the United States (Figure 1), has five narrow inlets, is wind driven and well mixed, and has a salinity gradient that drops from 35 psu at Oregon Inlet to 15–18 psu at the confluence of Croatan and Albemarle sounds (~40 km from Oregon Inlet) and to 5–7 psu near Church's Island, approximately 80 km from Oregon Inlet (Figure 1; Durham et al. 2019). The CAPES is also relatively shallow (mean depth = 4.9 m) and therefore lacks the type of deepwater channels found in the Chesapeake Bay and the White Oak River that appear to serve as migration corridors for blue crabs. In 1965, spawning sanctuaries were established in the five major CAPES inlets (Oregon, Hatteras, Ocracoke, Drum, and Barden); the take of crabs, oysters, or clams with commercial fishing equipment is prohibited within these sanctuaries during March–August. The sanctuary boundaries extend 1–2 km soundward and less than 1 km seaward of the inlets, which range in size from approximately 1.8 to 3.5 km2 (see Figure 2 in Eggleston et al. 2009). The Oregon Inlet sanctuary, which is assumed to be the spawning site for the female blue crabs tagged in this study because of its proximity to tagging locations, is the second largest of the sanctuaries (2352 ha). These sanctuaries, however, do not afford adequate protection to mature female blue crabs since, on average, only 0.7% of the mature female blue crab population is protected (Eggleston et al. 2009). Moreover, the proportions of female blue crabs tagged near two of North Carolina's spawning sanctuaries that were recaptured inside versus outside of the sanctuaries were approximately equal, indicating that these two sanctuaries afforded little protection to spawners (Medici et al. 2006). Results from a long-term, fishery-independent trawl survey program (Program 195) conducted in Pamlico Sound since 1987 by the North Carolina Division of Marine Fisheries suggested that there were no clear migration corridors for mature female blue crabs in Pamlico Sound (Eggleston et al. 2009). The inability to detect a distinct migration corridor may have been due to a water depth-related sampling bias in the Program 195 sampling design. Program 195 is designed to sample the relatively deep waters of Pamlico Sound (>2 m) and does not sample the potential shallow migratory routes of mature female blue crabs. Thus, the goal of this study was to characterize spatiotemporal variation in the migration of mature female blue crabs within the northeastern portion of the CAPES. The northeast region of the CAPES displays a relatively high abundance of mature female blue crabs and appears to connect this crab population with an inlet spawning sanctuary (Oregon Inlet; see below) that exhibits among the highest crab abundances observed in any of the spawning sanctuaries in North Carolina (Eggleston et al. 2004, 2009).

Details are in the caption following the image
Map of the study area in North Carolina, showing the locations of major water bodies, islands, Oregon Inlet, and the four blue crab release sites (denoted by open crosses): Durant Island, North River, Perquimans River, and Church's Island. Note the presence of two release locations for the Church's Island site; in 2009, the release point was moved to the southeast. Small closed circles denote the locations of all recaptures combined for 2008 and 2009.
Details are in the caption following the image
(A) Number of tagged mature female blue crabs that were recaptured by crabbers during May–December (pooled across both years of the study [2008 and 2009]); and (B) percentage of mature female blue crabs tagged in each month that were recaptured (values are average ± SE across study sites; n = 4).

BLUE CRAB LIFE HISTORY AND STOCK STATUS

Like many marine invertebrates, the blue crab has a complex life cycle. Female blue crabs molt to maturity and mate during the spring through fall months in the oligohaline/mesohaline waters of sounds, bays, and rivers throughout the temperate areas of the U.S. East Coast. After mating, females either remain in the oligohaline/mesohaline waters to build up energy stores for migration (Turner et al. 2003) or begin their seaward migration to euhaline waters, where they extrude a fertilized egg mass onto their abdomen (i.e., “sponge”) and release their larvae into the ocean from the mouths or inlets of rivers, bays, and sounds (Judy and Dudley 1970; Aguilar et al. 2005; Medici et al. 2006; Eggleston et al. 2015). In the Chesapeake Bay, mature females are presumed to migrate along relatively deep corridors in the center of the main bay, where they are funneled to the mouth of the bay (Lipcius et al. 2003). The CAPES is relatively shallow (mean depth < 2 m), with a mosaic of shallow shoals separated by deeper water. Moreover, the tidally driven salinity cues that are used by migrating blue crabs (Tankersley et al. 1998; Welch and Forward 2001) are absent from most of the CAPES because it is well mixed by winds and because tides do not exhibit much influence on its hydrodynamics (Reyns et al. 2007). Therefore, it is difficult to predict the potential migratory paths of postcopulatory female blue crabs based on migratory patterns from other systems and mechanistic behavioral studies. The most promising initial approach for answering questions about the timing and route of migration is a large-scale study that tracks the movement patterns of migrating mature female blue crabs.

NORTH CAROLINA BLUE CRAB FISHERY

The blue crab resource supports North Carolina's most valuable commercial fishery. During 1994–2016, commercial landings of blue crabs ranged from a low of 9.7 million kg/year to a high of 30.4 million kg/year (North Carolina Division of Marine Fisheries 2018). During the last decade (2007–2016), an average of 12.2 million kg/year was landed by the commercial fishery, with the primary fishing gear being crab pots (North Carolina Division of Marine Fisheries 2018). The most recent stock assessment for the blue crab in North Carolina (North Carolina Division of Marine Fisheries 2018) indicates that the resource is overfished with a probability of 98%, given that the average spawner abundance in 2016 was estimated at 50 million (below the threshold estimate of 64 million). Recommendations for rebuilding the blue crab spawning stock in North Carolina include (1) seasonal closures and/or (2) expansion of the spawning sanctuaries (North Carolina Division of Marine Fisheries 2018). It is unclear whether mature female blue crabs use specific corridors to reach the inlet spawning sanctuaries. Thus, the overarching goal of this study was to generate information that can be used to guide spatial and temporal management of the blue crab spawning stock in North Carolina by conducting a mark–recapture study to quantify the movement patterns of postcopulatory female blue crabs as they migrated to their inlet spawning grounds.

METHODS

In this study, mark–recapture data on migrating mature female blue crabs were collected by commercial crabbers during 2008 and 2009 in the northeastern region of the CAPES. A combination of GIS mapping of mark–recapture data, traditional statistics, and circular statistics determined the (1) timing, (2) rate, and (3) route of movement by postcopulatory female blue crabs to their spawning grounds.

Study site and tagging procedures

During 2008 and 2009, four licensed commercial crabbers (Fred Bell, H. L. Bond, Kristina Bridges, and Mike Mixon) captured and tagged 8400 mature female blue crabs within Albemarle and Currituck sounds, located in northeastern North Carolina (Figure 1). Each crabber made monthly sampling trips from June through November within their respective tagging site (Appendix Table A.1) to collect 150–200 mature female crabs for tagging and release (Figure 1). Crabbers used standard wire commercial crab traps fitted with 5.87-cm (2.3125-in) cull rings to capture mature females from their sites. A plastic tag (2.5 × 5.0 cm; Floy Tag, Inc.) was affixed to the top of the carapace of each crab by wrapping the stainless-steel wires that were attached to each end of the plastic tag around the lateral spines. Each tag was coded with a unique combination of color and printed identification number so that individual crabs could be identified. Tags also contained the contact information for the crabber that released the tagged crab and the word “REWARD.” Rewards were offered as an incentive for crabbers to report the tagged crabs that they recaptured. Fluorescent tag colors helped to ensure adequate visibility for the crabbers upon recapture. Mature female blue crabs are ideal for tag-return studies because they generally do not molt (Churchill 1919; Van Engel 1958), so tag loss is minimal. The shape of the carapace is such that a lightweight and noninvasive tag can be attached easily around the lateral spines on the dorsal surface. Tag-return studies of blue crabs have been conducted to examine migration (Fiedler 1930; Cronin 1949; Fischler and Walburg 1962; Tagatz 1968; Judy and Dudley 1970; Turner et al. 2003; Aguilar et al. 2005), to provide estimates of population size (Fischler 1965), and to assess the effectiveness of protected areas (Medici et al. 2006). Prior to release, all mature females were inspected for the presence and stage of a sponge egg mass on their abdomen to provide information on reproductive patterns upon return; sponges were not present on any of the crabs that were caught for tagging. All crabs tagged during sampling trips were released within each tagging site at specific coordinates that were typically 1–2 km from where the crab was captured. The exception was the 2009 Church's Island site, which was located 8.4 km southeast of the 2008 tagging site (Figure 1). To notify the public and crabbers about the project as well as to improve reporting rates, signs were posted at all North Carolina seafood dealers and wholesalers in the region. Tag returns were processed by telephone using phone numbers marked on each tag. Although recapture reports were processed year-round in both years of this study, the vast majority of recaptures (93% in 2008; 81.6% in 2009) occurred during the 4 months (June–September) when commercial crabbers are active. October and November recaptures were typically less than 4% of total recaptures for each year (November 2009 was the exception, accounting for 12.8% of recaptures in that year). Only three crabs were recaptured in total during December 2008 and 2009, and no recaptures were reported from January through April 2009. The reward for each tag was the choice of US$5 or a baseball cap. Crabbers claiming a reward were asked to provide the date and location of recapture, the gear used, and the female crab sponge stage if present. If crabbers could not provide latitude and longitude coordinates for a recapture, they were asked to provide specific distances from well-known landmarks so that coordinates could be estimated. A thank-you letter containing the release information for each returned tag was sent with the reward to everyone who claimed a tag. None of the crabbers involved with the tagging program was allowed to claim a reward for capturing tagged crabs.

Statistical analyses

To determine the timing, rate, and direction of movement for mature female blue crabs tagged at the four release sites, several response variables were generated from the mark–recapture data. The straight-line distance traveled was calculated using the great circle distance formula
Distance = 6371 × acos cos 90 Lat 1 × cos 90 Lat 2 + sin 90 Lat 1 × sin 90 Lat 2 × cos Long 1 Long 2 ,
where 6371 is the radius (km) of the Earth, Lat1 and Long1 are the release coordinates (latitude and longitude in decimal degrees), and Lat2 and Long2 are the recapture coordinates. Movement rate was estimated for each recaptured crab by using the distance traveled and the number of days at large. Finally, the direct-line heading that each crab moved between its tagging and recapture locations was estimated using the standard heading formula used in navigation:
Heading = atan2 cos Lat 1 × sin Lat 2 sin Lat 1 × cos Lat 2 × cos Long 2 Long 1 , sin Long 2 Long 1 × cos Lat 2 .
Each recapture was classified by month, with recaptures from October through December being combined because of the small sample sizes within each of those months. A combination of descriptive statistics, GIS maps, ANOVA models, and circular statistics was used to test for differences in the rate and route of mature female blue crab movement between years, among the five recapture months (June, July, August, September, and October–December), and from the four release sites. Each recaptured crab was considered to be a replicate in our analyses. If a crab had multiple recaptures, only data for the most recent recapture date were used. Moreover, only recapture data for crabs that were at large for at least 3 days were used to remove the small-spatial-scale (<1-km), meandering movement patterns that blue crabs typically make over the course of a few days when foraging (Wolcott and Hines 1989; Bell et al. 2003; Turner et al. 2003). Crabs that were at large for less than 3 days accounted for 29% of the total recaptures. Selecting recaptures of crabs at large for more than 3 days facilitated the detection of relatively large-scale, directed movement patterns.

To assess the timing of spawning migrations, we assumed that recapture months in which mature female blue crabs had relatively fast movement rates and headings in the direction of Oregon Inlet (assumed to be the primary inlet used by spawning female blue crabs that migrate from northeastern North Carolina) would indicate that the crabs had initiated their seaward migration. A two-way ANOVA model tested the main and interactive effects of release site and recapture month on blue crab movement rate. Tukey's multiple comparison tests within ANOVA models were used to test for pairwise differences in movement rate among recapture months for each release site. Although the raw movement rate data were not normally distributed, a log transformation was sufficient to normalize the data and make the variances homogeneous. Circular statistics were used to calculate the mean migration vector for tagged crabs from each combination of release site and month. The mean vector has two properties: (1) direction (i.e., the mean heading μ) and (2) length (r). The vector length ranges from 0 to 1, with values closer to 1 indicating that the observations are distributed more closely around the mean (i.e., nonrandom, directional movement). Separate Rayleigh's uniformity tests (one for each release site × month combination) were used to determine whether the observed distribution of mature female blue crab headings was significantly different (p < 0.05) from a uniform distribution (Zar 1984).

RESULTS

Spatial and temporal patterns of recaptures

In total, 1469 (17.5%) of the 8400 tagged mature female blue crabs were recaptured by crabbers throughout the tagging season; 1049 of the recaptured crabs were at large for more than 3 days (Table A.1). Egg masses were not present on any of the released or recaptured crabs. The number of recaptures increased by 34% from 2008 (629) to 2009 (840). This increase was driven primarily by a doubling of the number of recaptured crabs tagged at the Church's Island and Perquimans River sites from 2008 to 2009. Recaptures for crabs tagged at the Durant Island and North River sites were similar between years or slightly greater during 2009. The Durant Island site had the greatest percentage of its tagged crabs recaptured (24.3%), followed by the North River (20.2%), Church's Island (19.5%), and Perquimans River (5.9%) sites (Table A.1).

Although crabbers averaged 185 recaptures/month, there was a strong seasonal pattern to their recaptures. Recaptures were very low during May (only 20 recaptures, all occurring during 2009), which was likely due to crabbers having only 2–3 days in May to recapture crabs after they had been tagged (see Table A.1). Consequently, the number of recaptures rose sharply in June and remained high until August (Figure 2A). From September to December, we observed a decline in the total number of recaptures (Figure 2A) as well as the percentage of crabs recaptured that were tagged in these latter months (Figure 2B). This is likely due to a combination of the reduced effort of crabbers who pull their pots during the fall and crabs becoming increasingly less active as water temperatures decline. There was also a seasonal pattern to the percentage of crabs recaptured from each monthly tagging period. The recapture percentage for crabs that were tagged during the early part of the season (June–August) was commonly over 20% and as high as 51% and was generally greater than the recapture percentage for crabs that were tagged from September to November (Table A.1). Crabs that were tagged during the latter part of the tagging season had relatively low recapture rates—generally less than 10% and commonly 0% (Table A.1). Furthermore, although more than 30% of the crabs that were tagged during June were recaptured, the percentage of recaptures steadily declined with each subsequent tagging month until only 5% of the crabs tagged during November were recaptured (Figure 2B).

Most crabs were recaptured within Albemarle, Currituck, Croatan, and Roanoke sounds; however, the greatest concentration of recaptures occurred within the area where Albemarle, Currituck, and Croatan sounds join (Figure 1). No crabs were reported as recaptured inside any of the inlet spawning areas through December 2009, with only about four mature female blue crabs reaching the southern tip of Roanoke Island or points southward on the western edge of the barrier island chain (Figure 1). Of these crabs, the two closest to Oregon Inlet were approximately 10 km away, while the other two were 11 and 14 km away. All four of these crabs were tagged and released at the two southern-most sites (North River and Durant Island), and three of the four crabs were released in 2008. Most recaptures occurred to the east or south of their respective tagging sites, which corresponded to a general down-estuary migration route toward Oregon Inlet (Figure 1).

Movement headings

Movement of mature female blue crabs was predominantly directed toward the nearest inlet spawning area (Oregon Inlet) at all sites in both years and varied little from June through December except at two sites in 2009. Movement headings were significantly different than random for 32 of the 40 release site × recapture month × year combinations (Table A.2). Crab movement headings were specific to each release site; were generally in the down-estuary direction toward Oregon Inlet; and, with only a few exceptions, did not vary much among years or months. For example, the movement headings of crabs that were released from the Durant Island and North River sites averaged 105° and 152°, respectively, and were remarkably consistent among months (Table A.2) and years (Figures 3 and 4, respectively). Crabs that were released from these sites likely moved eastward out of Albemarle Sound and then south into Croatan and Roanoke sounds toward Oregon Inlet. The movement headings of crabs released during 2008 from the Perquimans River and Church's Island sites were also consistently down-estuary (mean headings: μ = 105.8° and 133.0°, respectively) toward Oregon Inlet (Figures 5 and 6). The monthly mean headings from crabs released at these sites during 2008 were generally consistent, deviating from the yearly mean by only about ±10° (Table A.2). During 2009, however, movement headings from these two sites were variable across recapture months. The headings from crabs released at the Perquimans River site appeared to have a bimodal distribution, with most crabs moving WSW (240–270°) or ESE (90–120°), corresponding to an up-estuary and down-estuary direction, respectively (Table A.2; Figure 5). Similarly, the recapture of crabs tagged at the Church's Island site were distributed mainly in the up-estuary (NNW) and down-estuary (SSE) directions (Figure 6).

Details are in the caption following the image
Polar plots of headings and relative distance traveled between release and recapture sites (i.e., distance from each point to the center of the plot is scaled based on the distribution of distances used in each graph) for each mature female blue crab that was released from the Durant Island site during 2008 and 2009 (there was no difference in headings among months). Plots show individual crab headings (open circles), and the arrow denotes the mean heading for all crabs in the plot.
Details are in the caption following the image
Polar plots of headings and relative distance traveled between release and recapture sites (i.e., distance from each point to the center of the plot is scaled based on the distribution of distances used in each graph) for each mature female blue crab that was released from the North River site during 2008 and 2009 (there was no difference in headings among months). Plots show individual crab headings (open circles), and the arrow denotes the mean heading for all crabs in the plot.
Details are in the caption following the image
Polar plots of headings and relative distance traveled between release and recapture sites (i.e., distance from each point to the center of the plot is scaled based on the distribution of distances used in each graph) for each mature female blue crab that was released from the Perquimans River site during 2008 (there was no difference in headings among months in 2008) and for each recapture month during 2009. Plots show individual crab headings (open circles), and the arrow denotes the mean heading for all crabs in the plot.
Details are in the caption following the image
Polar plots of headings and relative distance traveled between release and recapture sites (i.e., distance from each point to the center of the plot is scaled based on the distribution of distances used in each graph) for each mature female blue crab that was released from the Church's Island site during 2008 (there was no difference in headings among months in 2008) and for each recapture month during 2009. Plots show individual crab headings (open circles), and the arrow denotes the mean heading for all crabs in the plot.

Movement rate

Movement rates of mature female blue crabs varied among months at the different release sites and did not show distinctive seasonal patterns that would indicate a specific timing to their spawning migrations. There was a significant interactive effect of recapture year and month on the movement rates of crabs tagged at the Durant Island site (two-way ANOVA: n = 510, df = 4, F = 7.85, p < 0.01) and the Perquimans River site (two-way ANOVA: n = 123, df = 4, F = 13.57, p < 0.01). The movement rates of crabs tagged at Durant Island and the Perquimans River were greatest during the earlier months (June–August) in 2008 but did not differ much among months during 2009 (Figure 7). In 2008, movement rates were significantly greater during July compared to all other months for crabs tagged at Durant Island (Tukey's multiple comparison test: p < 0.01; Figure 7A) and were significantly greater during June–August compared to September–December for crabs from the Perquimans River (Tukey's multiple comparison test: p < 0.01; Figure 7A). During 2009, however, movement rates were similar across months for crabs from these sites (Figure 7B). For example, the movement rates for crabs from the Perquimans River site did not differ significantly among recapture months during 2009 (one-way ANOVA: n = 80, df = 4, F = 2.10, p = 0.09). Furthermore, although there was a significant overall difference in movement rate among recapture months for crabs tagged at Durant Island (one-way ANOVA: n = 268, df = 4, F = 4.63, p = 0.001), movement rates during October–December were only significantly different from the movement rates in June and September (Tukey's test: p < 0.01; Figure 7B).

Details are in the caption following the image
Mean monthly movement rates (km/d; ±SE) of mature female blue crabs released during (A) 2008 and (B) 2009 from the two study sites with a significant year × month interaction based on ANOVA (Durant Island and Perquimans River). Letters show the results of Tukey's multiple comparison tests on log-transformed data. Months with different letter designations had significantly different movement rates. Horizontal lines overlap months that did not have significantly different movement rates.

There was no significant interactive effect of recapture year and month on the movement rates of crabs that were tagged at the Church's Island site (two-way ANOVA: n = 408, df = 4, F = 1.33, p < 0.26) and the North River site (two-way ANOVA: n = 414, df = 4, F = 1.34, p = 0.26), indicating that monthly movement rates were similar between years. Therefore, we pooled observations across years and developed one-way ANOVAs for each site to test for differences in monthly movement rates. Blue crabs that were released from the Church's Island and North River sites showed little to no seasonal pattern in their movement rates (Figure 8). There was no significant overall difference in movement rates among months for crabs tagged at the North River site (one-way ANOVA: n = 414, df = 4, F = 2.26, p = 0.06). Although there was a significant month effect on movement rate for crabs tagged at Church's Island (one-way ANOVA: n = 408, df = 4, F = 3.74, p = 0.005), movement rates were only significantly different between July and October–December (Tukey's multiple comparison test: p < 0.01).

Details are in the caption following the image
Mean monthly movement rates (km/d; ±SE) of mature female blue crabs released from the two study sites with no significant year × month interaction based on the ANOVA model (North River and Church's Island). Letters show results of Tukey's multiple comparison tests on log-transformed data. Months with different letter designations had significantly different movement rates. The horizontal line overlaps months that did not have significantly different movement rates.

DISCUSSION

Population declines of many migratory marine species have led to calls for greater knowledge of population connectivity, especially insights from animal tracking studies (Dunn et al. 2019 and references therein). Inclusion of migratory connectivity in the design of conservation and management measures is critical to ensure that they are appropriate for the level of risk associated with various degrees of connectivity.

Timing of migration

We cannot conclusively determine from our study whether the start of spawning migrations occurred soon after mating or was delayed by months or a year because we did not know the copulatory history of our tagged crabs. However, blue crab life history traits and the movement patterns observed in this study and other studies suggest that the timing of individual spawning migrations is spatially and temporally variable. Blue crabs spawn primarily during June–August (Dudley and Judy 1971), recruit to North Carolina estuaries in July–October (a minor peak also occurs in May; Eggleston et al. 2010), live for 3–4 years, and mature after 12–18 months. Thus, although female blue crabs living in Albemarle Sound and its subestuaries may be able to mate after they reach 1 year of age, they may be unable to complete their migration to the inlets and spawn until their second year of benthic life. Assuming that the tagged blue crabs in both years were a mixture of 1- and 2-year-old individuals, the strikingly consistent down-bay orientation of the tagged mature females at all sites from June through December in 2008 and from three of the five sites in 2009 suggests that at least some female blue crabs in northeastern North Carolina may begin their spawning migration during the same year in which they molt to maturity. Our 2009 recapture data from crabs tagged at two sites, however, show that many mature female blue crabs may randomly meander throughout much of the summer instead of migrating to spawning sites (i.e., Church's Island), while others may move in the opposite direction of spawning sites (i.e., Perquimans River). The latter results may also be due to relatively low recapture rates for the Perquimans River site during 2009 (Table A.2). Most previous studies suggest that after copulation, mature female blue crabs in North Carolina and the Chesapeake Bay delay their seaward migration until September or early October to build energy stores (Judy and Dudley 1970; Turner et al. 2003; Aguilar et al. 2005; Medici et al. 2006). For example, Medici et al. (2006) tagged and released mature female blue crabs near our Durant Island study site during summer 2002 and found that mature females exhibited no significant directed movement and remained near their release site for up to 30 days after release. Similarly, tagged crabs in a study by Aguilar et al. (2005) showed no directed down-bay movements until the fall months (September–November). Darnell and Kemberling (2018), however, tagged postcopulatory female blue crabs that were within 2 weeks of their terminal molt and released in three rivers of southeastern North Carolina. They found slow and sustained seaward migration prior to oviposition in summer and fall, which is consistent with the more predominant pattern we observed in 2008 at all sites and in 2009 at all but two sites. To more fully understand the variation in timing of spawning migrations and to develop models that simulate spawning migrations, it is critical for future tagging and movement studies to have information on the copulatory history of the crabs and to determine the environmental and physiological cues that trigger the migratory behavior of mature female blue crabs.

Mature females migrated toward their spawning grounds at variable rates that were mainly between 1.93 and 2.23 km/d (the 95% confidence interval for all crabs). Our migration rate calculations are likely underestimates because we used straight-line distances and crabs would likely meander while migrating prior to being recaptured. Moreover, many of our straight-line paths crossed over land that crabs would have moved around to get to the location where they were recaptured. Using a conservative estimate of 2 km/d, female blue crabs would need approximately 20–50 days to complete the roughly 40–100-km route from the tagging sites to Oregon Inlet. Therefore, postcopulatory females that are present in Albemarle and Currituck sounds during June and July should be capable of reaching Oregon Inlet during the August spawning period. However, none of our tagged crabs was reported as recaptured within 7 km of any major inlet spawning location, only two crabs were caught within 10 km of Oregon Inlet, and none of our returned crabs had sponges. Although this suggests that mature female blue crabs in the region may take more than one growing season after attaining maturity to reach Oregon Inlet and spawn, we cannot rule out the possibility that mature female crabs did complete the journey, as commercial crabbers (our greatest source of recaptures) are prohibited from deploying gear in inlet spawning sanctuaries of North Carolina from March to September. Moreover, it is also possible that tagged crabs were recaptured near the inlets and not reported by crabbers. Effectively communicating the details about scientific mark–recapture studies to local crabbers is extremely important for achieving good reporting rates (Aguilar et al. 2005). A different group of crabbers than those that participated in this study fishes the inlet waters, and information about our tagging study may not have reached that group. However, posters advertising our study were posted at fish houses located near Oregon Inlet (Colington Harbor and Manns Harbor, North Carolina), suggesting that local crabbers around Oregon Inlet should have been aware of the study. Moreover, one of our crabber collaborators, Kristina Bridges, lives in Colington Harbor, where her family traditionally sells blue crabs. Alternatively, poaching within the boundaries of the inlet spawning sanctuaries is somewhat common (Medici et al. 2006), and crabbers may not want to risk penalties for returning tags that have been recovered from crabs caught in illegal waters. Crabs tagged in 2008 should have completed their spawning migration during our 2-year study given the movement rates we recorded and the proximity of many crabs to Oregon Inlet (one of the two crabs that were caught within 10 km of Oregon Inlet was captured in June and could have made it to the inlet to spawn). As mentioned above, during a related study in a relatively small, tidally driven estuary in North Carolina (White Oak River), a combination of active and passive acoustic tracking methods quantified the movement of postcopulatory female blue crabs migrating down-estuary to oceanic spawning grounds (Eggleston et al. 2015). Crabs traveled approximately 14.1 km (mainly in deeper channels) and over 12–26 days from mating areas to spawning grounds (Eggleston et al. 2015). No crabs were detected as migrating down-estuary in autumn, and only 30% were detected as migrating down-estuary in spring. None of the crabs detected near spawning grounds was later detected or recaptured up-estuary, suggesting that they either (1) do not return to the estuary after a 1–2-week period in the spawning area or (2) were captured by fishermen (Eggleston et al. 2015). Thus, the timing between mating and the arrival of postcopulatory female blue crabs to inlet spawning grounds appears to be strongly dependent upon the distance between these two locations. For example, the distance between the mating area and the Oregon Inlet spawning grounds in this study was 40–100 km, whereas the distance from the mating area to the spawning grounds in the White Oak River, which postcopulatory female crabs reached in 12–26 days, was approximately 14 km (Eggleston et al. 2015). Future studies attempting to estimate the timing and duration of spawning migrations by mature female blue crabs should focus on using females of known copulation history and should coordinate with fisheries management agencies to establish recapture programs with crabbers inside spawning sanctuaries, where crabbing is not permitted.

Route of migration

It is difficult to identify potential migration corridors of mature female blue crabs from northeastern North Carolina based on the spatial distribution of recaptures from our fishery-dependent mark–recapture study because fishing effort is not distributed uniformly or randomly, which makes it problematic to discriminate the spatial locations where crabs are present or absent. For example, crabbers do not distribute their pots across all depths because of depth regulations and limited boat access to the shallowest areas. Moreover, crabbers move their crab pots progressively seaward during the fall to catch migrating female blue crabs more efficiently (K. Bridges, personal communication). Therefore, the cluster of recaptures around the union of Albemarle, Currituck, and Croatan sounds (see Figure 1), as well as the restrictive waterway access to inlets in northeastern North Carolina (see Figure 1), suggests that the relatively narrow throughways of Croatan and Roanoke sounds create a de facto migration corridor. As mature female crabs migrate from the large expanse of Albemarle and Currituck sounds during the summer and fall, they are likely to become concentrated along the northern edges and within Croatan and Roanoke sounds, which are relatively narrow water bodies.

There are several troublesome consequences for the blue crab fishery if crabbers target mature female blue crabs as they move through the narrow passageways in Croatan and Roanoke sounds. First, the increased concentration of crabs in this narrow waterway could lead to increased catch efficiency and contribute to a decline in blue crab spawning stock biomass. Crabbers in North Carolina exploit episodic environmental conditions that concentrate adult blue crabs in high densities. For example, in 1999 after Hurricanes Dennis and Floyd dumped large amounts of rain on coastal North Carolina, blue crabs were pushed out of the rivers and estuaries as they filled with freshwater and were concentrated in a narrow band of higher-salinity water along the eastern edge of Pamlico Sound (Burkholder et al. 2004). Many crabbers moved their pots to this narrow waterway, where their catch efficiency increased dramatically (Eggleston et al. 2004). An increased catch efficiency within the migration corridor may also lead to errors in estimating the catchability parameter used in stock assessments that employ fisheries-dependent catch data (Trenkel and Skaug 2005). Stock assessment models that calculate a catchability estimate require this parameter to be constant in space and time; therefore, increased catch efficiencies in Croatan and Roanoke sounds would violate this assumption and may lead to errors in blue crab stock assessments and management decisions.

This study highlights the large degree of inherent variability among studies that attempt to determine the timing of spawning migrations by mature female blue crabs. Our study also identifies a potential de facto migration corridor created by the narrowing of waterways at Croatan and Roanoke sounds, which female crabs from Albemarle Sound use to arrive at the inlet spawning grounds. We still lack very basic information about the spawning migration behavior of mature female blue crabs; thus, it is difficult to determine and evaluate potential management strategies for conserving and rebuilding the blue crab spawning stock biomass. The high concentration of blue crab recaptures in Roanoke and Croatan sounds in this study, however, suggests that extending the inlet sanctuary boundaries into these waters to protect mature females as they migrate to the inlets may help to reduce the 80% decline in spawning stock biomass that occurs during the migratory period leading up to spawning (Eggleston et al. 2009). However, this stretch of water is small compared to the entire coastal waters that female blue crabs use when migrating to the inlets, and it is impossible to predict the efficacy of such a protected corridor without a more thorough study of blue crab migratory behavior. The lack of answers to key questions limits our ability to develop a large-scale model that simulates the migration of mature female blue crabs. These questions include the following: (1) “What environmental cues do female blue crabs use to determine when to migrate in primarily wind-driven systems?”; (2) “How soon after copulation do female crabs begin migrating?”; and (3) “Where do female crabs move during their postcopulatory migration?”

ACKNOWLEDGMENTS

We thank commercial crabbers K. Bridges, F. Bell, H. L. Bond, and M. Mixon for their participation in the field mark–recapture studies. Funding was provided by North Carolina Sea Grant through the Blue Crab Research Program (09-POP-05R; Marc Turano, program manager).

    CONFLICT OF INTEREST

    There is no conflict of interest declared in this article.

    ETHICS STATEMENT

    There were no ethical guidelines that were applicable to this study.

    APPENDIX A

    TABLE A.1. Number of mature female blue crabs tagged and recaptured as well as recaptures expressed as a percent of total tagged for each release site and trip date.
    Release information Recaptures
    Release site Latitude Longitude Date Tagged Number Percent
    Durant Island 35°59.441 N 75°52.992 W May 30, 2008 150 46 30.7
    Jul 3, 2008 151 45 31.8
    Jul 29, 2008 150 48 32.7
    Aug 18, 2008 150 68 46.0
    Sep 13, 2008 150 19 12.7
    Oct 6, 2008 150 8 6.0
    Oct 27, 2008 150 10 6.7
    2008 total 1051 244 23.2
    Durant Island 35°59.403 N 75°53.060 W May 27, 2009 150 53 35.3
    Jun 24, 2009 150 39 26.0
    Jul 25, 2009 150 59 39.3
    Aug 31, 2009 150 40 26.7
    Sep 21, 2009 150 34 22.7
    Oct 27, 2009 150 20 13.3
    Nov 16, 2009 150 22 14.7
    2009 total 1050 267 25.4
    2008 and 2009 2101 511 24.3
    North River 36°09.237 N 75°53.492 W Jun 30, 2008 200 86 43.0
    Jul 2, 2008 100 40 40.0
    Jul 29, 2008 100 25 25.0
    Jul 30, 2008 51 15 29.4
    Aug 16, 2008 100 16 16.0
    Aug 18, 2008 50 2 4.0
    Sep 13, 2008 150 17 11.3
    Oct 3, 2008 151 8 5.3
    Nov 3, 2008 150 4 2.7
    2008 total 1052 213 20.3
    North River 36°09.165 N 75°53.592 W May 27, 2009 150 77 51.3
    Jun 24, 2009 150 52 34.7
    Jul 25, 2009 150 42 28.0
    Aug 28, 2009 150 20 13.3
    Sep 21, 2009 150 13 8.7
    Oct 27, 2009 150 8 5.3
    Nov 16, 2009 150 0 0.0
    2009 total 1050 212 20.2
    2008 and 2009 2102 425 20.2
    Perquimans River 36°05.326 N 76°16.522 W Jun 7, 2008 152 26 17.1
    Jul 14, 2008 60 8 13.3
    Jul 15, 2008 90 4 4.4
    Aug 20, 2008 150 1 0.7
    Sep 5, 2008 60 0 0.0
    Sep 12, 2008 91 0 0.0
    Sep 22, 2008 106 0 0.0
    Sep 23, 2008 44 1 2.3
    Oct 14, 2008 150 3 2.7
    Nov 7, 2008 60 0 0.0
    Nov 10, 2008 90 0 0.0
    2008 total 1053 43 4.1
    Perquimans River 36°05.326 N 76°16.522 W May 28, 2009 150 22 14.7
    Jun 22, 2009 117 0 0.0
    Jun 23, 2009 33 13 39.4
    Jul 31, 2009 144 8 5.6
    Aug 28, 2009 150 8 5.3
    Sep 21, 2009 150 4 2.7
    Oct 29, 2009 150 20 13.3
    Nov 14, 2009 150 6 4.0
    2009 total 1044 81 7.8
    2008 and 2009 2097 124 5.9
    Church's Island 36°26.877 N 75°59.034 W May 30, 2008 153 34 22.2
    Jun 28, 2008 75 22 29.3
    Jun 30, 2008 56 11 19.6
    Jul 9, 2008 18 2 11.1
    Jul 30, 2008 150 18 12.0
    Aug 18, 2008 150 31 20.7
    Sep 11, 2008 150 11 7.3
    Oct 4, 2008 150 0 0.0
    Oct 28, 2008 149 0 0.0
    2008 total 1051 129 12.3
    Church's Island 36°24.021 N 75°54.633 W May 28, 2009 150 69 46.0
    Jun 24, 2009 150 77 51.3
    Jul 29, 2009 99 44 44.4
    Jul 31, 2009 51 15 29.4
    Aug 28, 2009 150 37 24.7
    Sep 21, 2009 150 17 11.3
    Oct 26, 2009 150 12 8.0
    Nov 22, 2009 149 9 6.0
    2009 total 1049 280 26.7
    2008 and 2009 2100 409 19.5
    All sites 2008 4207 629 15.0
    2009 4193 840 20.0
    2008 and 2009 8400 1469 17.5
    TABLE A.2. Circular statistics of blue crab movement headings (μ = mean heading; r = length) for each release site, year, and recapture month (crabs that were at large < 3 days are excluded; see Methods). Significant Rayleigh's tests indicate that headings were not random (**headings were uniformly distributed and confidence intervals [CIs] could not be calculated).
    Release site Year Month n Vector Rayleigh's test (p)
    μ r 95% CI
    Durant Island 2008 Jun 41 114.7° 0.88 105.7–123.8° <0.01
    Jul 21 111.8° 0.97 105.4–118.3° <0.01
    Aug 58 101.2° 0.94 96.1–106.2° <0.01
    Sep 13 101.3° 0.97 93.0–109.7° <0.01
    Oct–Dec 12 104.5° 0.91 88.7–120.3° <0.01
    All 145 106.8° 0.92 102.9–110.4° <0.01
    Durant Island 2009 Jun 54 98.9° 0.70 86.1–111.7° <0.01
    Jul 17 99.9° 0.96 91.9–107.8° <0.01
    Aug 7 109.7° 0.99 100.4–119.0° <0.01
    Sep 29 87.8° 0.84 75.6–100.0° <0.01
    Oct–Dec 45 117.5° 0.93 111.2–123.7° <0.01
    All 152 103.6° 0.82 98.0–109.2° <0.01
    North River 2008 Jun 0 NA NA NA NA
    Jul 54 158.0° 0.98 154.9–161.1° <0.01
    Aug 38 157.5° 0.98 154.0–161.0° <0.01
    Sep 5 143.7° 1.00 140.7–146.7° <0.01
    Oct–Dec 14 148.3° 1.00 145.6–151.0° <0.01
    All 111 155.9° 0.98 153.8–158.1° <0.01
    North River 2009 Jun 59 146.2° 0.98 142.7–149.7° <0.01
    Jul 33 144.8° 0.97 139.6–149.9° <0.01
    Aug 33 146.2° 0.84 134.9–157.6° <0.01
    Sep 21 154.3° 0.99 150.5–158.0° <0.01
    Oct–Dec 27 156.9° 0.98 153.0–160.8° <0.01
    All 173 148.7° 0.95 145.9–151.5° <0.01
    Perquimans River 2008 Jun 16 108.9° 0.99 105.8–112.0° <0.01
    Jul 11 105.0° 0.99 99.7–110.4° <0.01
    Aug 6 113.3° 0.98 102.3–124.3° <0.01
    Sep 4 53.1° 0.68 0.0–106.5° 0.16
    Oct–Dec 4 161.5° 0.20 ** 0.86
    All 41 105.8° 0.85 95.7–115.8° <0.01
    Perquimans River 2009 Jun 21 230.1° 0.51 199.2–261.1° <0.01
    Jul 12 132.8° 0.46 82.1–183.5° 0.07
    Aug 6 238.1° 0.99 231.2–245.1° <0.01
    Sep 9 149.5° 0.27 320.1–338.9° 0.54
    Oct–Dec 31 150.1° 0.56 126.8–173.4° <0.01
    All 79 179.1° 0.39 157.2–201.0° <0.01
    Church's Island 2008 Jun 27 128.5° 0.75 111.9–145.0° <0.01
    Jul 42 117.5° 0.67 101.8–133.2° <0.01
    Aug 45 142.9° 0.94 137.0–148.9° <0.01
    Sep 13 147.2° 0.98 139.9–154.9° <0.01
    Oct–Dec 1 114.4° 1.00 NA NA
    All 127 133.0° 0.79 127.0–140.2° <0.01
    Church's Island 2009 Jun 60 156.2° 0.43 133.7–178.7° <0.01
    Jul 50 166.4° 0.17 99.7–233.1° 0.25
    Aug 54 50.6° 0.26 9.3–92.0° 0.03
    Sep 38 110.5° 0.49 85.6–135.3° <0.01
    Oct–Dec 19 103.3° 0.72 82.6–123.9° <0.01
    All 221 121.3° 0.29 102.9–139.7° <0.01

    DATA AVAILABILITY STATEMENT

    The data that support the findings of this study are available from the first author upon request ([email protected]).