Tag-recapture data grant researchers the opportunity to synthesize a vast array of information for species they are studying. Sharks, in particular, are well suited for tagging studies due to their migratory behavior, longevity, and relatively large size, which allows tagging at all life stages ( Bonham et al., 1949). Valuable information gained from recapture data includes greater understanding of species-specific migratory patterns, stock structure, spatial and temporal distribution, site fidelity/residence times, and life histories ( Kohler & Turner, 2001).

Among life history parameters needed to best manage populations of fishes, robust longevity estimates are of paramount importance, particularly for iteroparous species, such as sharks. For example, assuming age at maturity, reproductive periodicity, and brood size are constant, the lifetime fecundity of a given species is directly linked to its lifespan. Longevity estimates for sharks are generally derived from von Bertalanffy Growth Function (VBGF) parameter estimates (i.e. growth constant); however, VBGF parameter estimates can be heavily affected by low sample sizes, incomplete sampling among size classes and/or difficulties associated with age estimation of large individuals ( Goldman, 2004; Francis et al., 2007). Obtaining accurate longevity estimates can be hampered by the low probability of catching older individuals as they represent a small portion of the entire population, difficulty associated with capturing large specimens that are more capable of escaping gear than smaller conspecifics, and the reduction of older individuals in the population due to fishing pressure ( Bishop et al., 2006). Additionally, age underestimation has been documented in multiple shark species including long-lived species such as the Porbeagle, Lamna nasus (Bonnaterre, 1788) ( Francis et al., 2007) and the Australian School Shark, Galeorhinus galeus (L. 1758) ( Kalish & Johnston, 2001), and species with intermediate life histories such as the Bonnethead, Sphyrna tiburo (L. 1758) ( Frazier et al., 2014). Due to the above factors, tag-recapture records, when available, are the most reliable source of longevity data as maximum time-at-liberty for any individual within the population would represent the highest directly known longevity. Based on tag-recapture data and direct age estimates, herein we report on the longevity of Atlantic Sharpnose Rhizoprionodon terraenovae (Richardson, 1836) and Blacknose Carcharhinus acronotus (Poey, 1860) Sharks, both of which are common in the coastal waters off the southeastern United States.

Sharks were captured and tagged during survey operations conducted by the South Carolina Department of Natural Resources Adult Red Drum and Coastal Shark Longline Program (SCDNR) (see Ulrich et al., 2007 for longline protocol). Collection, handling and tagging of specimens was authorized and controlled under the SCDNR scientific permit issued to employees of SCDNR.

Upon capture, the fork length (FL) and sex of each shark was recorded. For those sharks deemed to be in good health, a Hallprint ^© nylon dart tag, labeled with unique identifying numbers, was inserted into the dorsal musculature at the base of the first dorsal fin prior to release. As tagged sharks were at liberty in the wild, recaptured specimens were reported by a number of sources, including recreational anglers, commercial fishermen and SCDNR. Estimated measurements at recapture were obtained from recreational and commercial fishermen and direct measurements were recorded from specimens recaptured by SCDNR. When possible, a sample of 8–10 vertebrae were removed from the cervical region of the vertebral column from recaptured sharks.

In the case of recaptured specimens from commercial fishermen, sharks were already deceased and vertebrae were removed. In the case of recaptured specimens from SCDNR, a IACUC protocol approved for graduate students who had previously worked with SCDNR on elasmobranch studies was followed; the vertebral column was served by serrated knife in two cervical locations.

To estimate age at recapture, vertebrae were prepared for analyses following the protocol of Frazier et al., 2014. Initial vertebral growth band counts were conducted by three unbiased readers with no knowledge of specimen length or time-at-liberty. If band counts differed among readers, sections were independently re-read until agreement was reached. Age at recapture was estimated under the assumptions: (1) the birthmark was formed prior or shortly after parturition, (2) the second band was formed 6 months later during the first winter, and (3) the third band was formed 1 year later. Therefore, we subtracted 1.5 from each total band count to calculate age in years (e.g. Loefer & Sedberry, 2003; Driggers et al., 2004).

Age estimates at recapture were also determined by adding time-at-liberty to backtransformed ages at tagging. Species and sex-specific von Bertalanffy growth function parameters from Loefer & Sedberry, 2003 ( R. terraenovae) and Driggers et al., 2004 ( C. acronotus) were used to backtransform age at tagging using the following equation:

Age= ( 1n ( 1- L t L ∞ ) -k ) +t o

    Where:

     L _t = length at age t,

     L _∞ = theoretical maximum length,

     k = coefficient of growth,

     t _o = theoretical age at which length equals zero.

A paired t-test was used to determine if age estimates from vertebral sections were significantly different than backtransformed age estimates. Statistical results were considered significant at α < 0.05.

The backtransformed and direct age estimates documented herein greatly increase the known longevity for R. terraenovae and C. acronotus, which could significantly affect population dynamics models that include longevity as a parameter. Interestingly, the maximum directly estimated ages for both species were associated with male sharks. A review of published shark age and growth studies shows that, in most studies, that females have a greater or equal longevity than males (e.g. Carlson & Baremore, 2003; Carlson & Parsons, 1997; Carlson et al., 1999; Driggers et al., 2004; Drymon et al., 2006; Frazier et al., 2014). Therefore, we believe that we did not sample older females and suggest that the longevity estimates maximum observed longevity we observed for each species be applied to males and females.

Backtransformed age estimates from recaptures were in all cases larger than direct age estimates. This could be evidence that direct ages were underestimated, however direct age estimates only differed by an average of 1.8 years (range 1.0–2.5 years). Observed differences could be due to low sample size or individual variability in growth. The rearranged VBGF gives an estimate of average age-at-length based on parameter estimates. Therefore, specimens could have been younger or older than the average estimated age at time of tagging thus accounting for the discrepancy. Conversely, if the species-specific VBGF do not adequately describe the growth of R. terraenovae and C. acronotus then the observed differences would be expected. However, the growth models utilized were from studies with robust sample sizes that included all size classes ( Loefer & Sedberry, 2003; Driggers et al., 2004).

The age estimates from this study greatly increase longevity for both species; however, actual life spans could be even longer. The specimens recaptured from this study were tagged in the first few years of the SCDNR longline program (1994–1996). Given project species-specific recapture rates of less than 3% and reported tag shedding rates for nylon dart tags as high as 41% to 63% ( Xiao et al., 1999), the chances of recapturing a shark at liberty for 10+ years are small. The fact that seven were encountered and all exceeded published longevity estimates lends support to this assertion. The data gathered from these recaptures also highlights the importance of continuing long term surveys and tagging efforts. While return rates of tagged sharks are notoriously low, our data demonstrate that continued tagging efforts are essential to provide the most up to date and reliable estimates of maximum longevity. Ideally, to obtain the best possible longevity estimates for a given species, future efforts should focus on tagging neonate sharks with the hope of recapturing them as they approach the end of their lifespans.