Age and Growth of Atlantic Sturgeon in the James River, Virginia, 1997–2011

Abstract

Historically the Chesapeake Bay supported a large population of Atlantic sturgeon Acipenser oxyrinchus, but loss of suitable spawning habitat and overfishing coincided with dramatic in‐system declines throughout the 20th century. Atlantic sturgeon harvest moratoriums were implemented in 1974 for Virginia waters and were expanded coastwide in 1998. In 1997, researchers became aware that commercial fishers in the James River, a tributary of the Chesapeake Bay, were catching juvenile and subadult Atlantic sturgeon as bycatch in various fisheries. Genetic studies showed that the Chesapeake Bay population has maintained genetic integrity and qualifies as a distinct population segment. Between 2007 and 2011, almost 150 adults have been caught in the tidal–freshwater portion of the James River during putative spawning runs. Pectoral fin spines from juveniles and subadults collected in the Burwell Bay (rkm 40) and Cobham Bay (rkm 60) areas and mature adult samples from vessel strikes in freshwater around or above rkm 120 were analyzed to create a length‐at‐age curve for Atlantic sturgeon in the James River. Five models were used to analyze the data, and the double von Bertalanffy (k1 = 0.054, k2 = 0.097, t1 = −2.85, t2 = 1.09, tp = 6.03 years, L∞ = 2241 mm) provided the best fit to the observed data. We estimated an increase in growth coefficient at tp, which could be an artifact of low sample size or due to ontogenetic changes in habitat use as older fish spend more time in oceanic waters than younger fish. Atlantic sturgeon in the 6–9 year age range are rarely encountered in the James River compared with younger and older age‐classes, so a more in‐depth analysis of the increased growth coefficient would require ocean sampling.

Age structure of juvenile Amur sturgeon Acipenser schrenckii and kaluga Huso dauricus in the Fuyuan reach of the Amur River, Northeast China

“…Age estimation is challenging in older adult sturgeon, because of the close spacing of annuli near the edge of the pectoral fin spine (Nathan, et al, 2015), so the higher agreement may be attributable to our specimens being juveniles. In the present study, both sturgeon species presented higher CV between readers than the 1.8% in juvenile lake sturgeon Acipenser fulvescens (Balazik, McIninch, Garman, & Latour, 2012). The CV of Amur sturgeon was lower than the reported 13.6% in juvenile pallid sturgeon Scaphirhynchus albus, while the CV of kaluga was lower than that reported for juvenile pallid sturgeon S. albus (Koch et al, 2011).…”

Growth Estimation of Western Population Segment Gulf Sturgeon Using Length‐at‐Age and Mark–Recapture Data

“…Balazik et al. () found that out of five candidate models, the Gompertz model was the second‐best model for growth estimation of Atlantic Sturgeon A. oxyrinchus oxyrinchus in the James River, Virginia. Additionally, Woodland () found that the Gompertz model seemed to fit length‐at‐age data for Hudson River Shortnose Sturgeon A. brevirostrum , but an information‐theoretic approach was not applied in that study.…”

“…Balazik et al. () selected the double VBGF as the best model among five candidate models for Atlantic Sturgeon growth in the James River based on length‐at‐age data. The inclusion of that model as well as data for older individuals of Gulf Sturgeon and data from different drainages will improve the ability to identify the most appropriate growth model for Gulf Sturgeon across the GOM.…”

“…; Balazik et al. ). Multimodel frameworks may partially overcome some sampling limitations (such as oversampling and undersampling a particular age‐class) because a range of curve options are compared (Smart et al.…”

“…Despite the common use of the VBGF for sturgeon and other species, the VBGF may not be the best growth model to select a priori (Katsanevakis 2006;Katsanevakis and Maravelias 2008). The use of information theory and multimodel inference (Burnham and Anderson 2002) has become increasingly commonplace to discern whether the VBGF or other growth models provide a better fit for age data in teleosts (e.g., Katsanevakis 2006;Dippold et al 2016), elasmobranchs (Smart et al 2016), and acipenserids (e.g., Shuman et al 2011;Balazik et al 2012). Multimodel frameworks may partially overcome some sampling limitations (such as oversampling and undersampling a particular age-class) because a range of curve options are compared (Smart et al 2016).…”