A combined telemetry – tag return approach to estimate fishing and natural mortality rates of an estuarine fish

Abstract

A joint analysis of tag return and telemetry data should improve estimates of mortality rates for exploited fishes; however, the combined approach has thus far only been tested in terrestrial systems. We tagged subadult red drum (Sciaenops ocellatus) with conventional tags and ultrasonic transmitters over 3 years in coastal North Carolina, USA, to test the efficacy of the combined telemetry -tag return approach. There was a strong seasonal pattern to monthly fishing mortality rate (F) estimates from both conventional and telemetry tags; highest F values occurred in fall months and lowest levels occurred during winter. Although monthly F values were similar in pattern and magnitude between conventional tagging and telemetry, information on F in the combined model came primarily from conventional tags. The estimated natural mortality rate (M) in the combined model was low (estimated annual rate ± standard error: 0.04 ± 0.04) and was based primarily upon the telemetry approach. Using high-reward tagging, we estimated different tag reporting rates for state agency and university tagging programs. The combined telemetry -tag return approach can be an effective approach for estimating F and M as long as several key assumptions of the model are met.Résumé : Une analyse combinée du retour des étiquettes et des données de télémétrie devrait améliorer les estimations des taux de mortalité des poissons exploités; cependant, une telle approche conjointe n'a à ce jour été testée que dans les systèmes terrestres. Nous avons marqué des tambours rouges, Sciaenops ocellatus, avec des étiquettes ordinaires et des ém-etteurs à ultrasons sur une période de trois années sur la côte de la Caroline du Nord, É .-U., pour évaluer l'efficacité de l'utilisation combinée du retour des étiquettes et de la télémétrie. Il existe un fort patron saisonnier dans les estimations du taux mensuel de mortalité due à la pêche (F) faites à partir des étiquettes ordinaires et celles de télémétrie; les valeurs maximales de F s'observent à l'automne et les plus faibles en hiver. Bien que les patrons et l'importance des valeurs mensuelles de F soient semblables à partir du marquage classique et de la télémétrie, l'information sur F dans le modèle combiné provient surtout des étiquettes ordinaires. Le taux estimé de mortalité naturelle (M) dans le modèle combiné est faible (taux annuel estimé ± erreur type : 0,04 ± 0,04) et il se base principalement sur les données de télémétrie. En utilisant un programme de marquage avec de fortes récompenses, nous avons estimé des taux de retour des étiquettes différents pour les programmes menés par l'organisme de l'état et par l'université. L'approche qui combine le retour des étiquettes et la télémétrie peut être efficace pour estimer F et M à la condition que plusieurs des présuppositions principales du modèle soient respectées.[Traduit par la Rédaction]

Estimating population size in the presence of temporary migration using a joint analysis of telemetry and capture–recapture data

“…Using joint telemetry/CMR data has been shown elsewhere to increase precision of mortality rates (Bacheler et al 2009;Nasution et al 2001;Pollock, Jiang & Hightower 2004) and population size (Ivan, White & Shenk 2013). In our case, the size of credible intervals in model T was not necessarily lower than those in models P or C. This is likely because model T encapsulates a greater degree of uncertainty due to migration.…”

“…For example, animals entering different behavioural states (Schick et al 2008), plants going dormant (Speybroeck et al 2006) or cases where dead individuals are not accessible to conventional sampling (Bacheler et al 2009) could all be reliably assayed through non-CMR methods. Such unobservable states are frequently of interest to ecologists, so the ability to better model population-level probabilities of moving in and out of these states will be useful.…”

Evaluating the efficacy of the Florida Keys’ angler-assisted permit tagging program

“…Angler nonreporting is usually measured using high reward tags following Pollack et al (1991;e.g., Jenkins et al 2000;Pollock et al 2001;Denson et al 2002;Taylor et al 2006). Monetary or souvenir rewards have anecdotally encouraged anglers in other tagging programs to report their tags and participate in tagging programs (Kohler and Turner 2001;Ortiz et al 2003;Jech et al 2009). Communication between scientists and managers is a necessary component for a successful cooperative study (NRC 2003;Helvey 2004;Conway and Pomeroy 2006;Humston et al 2007).…”

Using generalized linear models to estimate selectivity from short-term recoveries of tagged red drum Sciaenops ocellatus: Effects of gear, fate, and regulation period

“…In recent years, approximately two-thirds of harvested red drum in North Carolina has come from the recreational hook-and-line fishery, with commercial gill and pound nets accounting for most of the remaining harvest (Takade and Paramore, 2007;Bacheler et al, 2009a); other gears were excluded from all analyses due to low sample sizes of recovered fish. All hook-and-line gears such as hand lines, bottom rigs, and other hook-and-line rigs were considered together here.…”

“…Bacheler et al (2008) used this approach to estimate selectivity patterns for red drum (Sciaenops ocellatus), an estuarine species found in the southeast United States and Gulf of Mexico. Traditional stock assessment models could not be used to estimate red drum selectivity because harvest occurred within a minimum and maximum size limit (i.e., window limit) centered on age-2 fish (with a smaller proportion of age-1 and age-3 fish; Bacheler et al, 2009a); it was impossible for the assessment model to determine if the lack of age-3 harvest was due to reduced selectivity or high mortality rates on previous age groups (Latour et al, 2001). A benefit of the tag-return approach (Bacheler et al, 2008) to estimate selectivity is being able to use all tag recoveries occurring over time.…”