An Index of Abundance for Coastal Species of Juvenile Sharks from the Northeast Gulf of Mexico - Statistical Data Included

Marine Fisheries Review, Summer, 1999 by John K. Carlson, John H. Brusher

Introduction

In 1993, a Fishery Management Plan For Sharks (FMP) (NMFS, 1993) reported the abundance of many species of sharks, particularly large coastal species, could have declined by up to 75% from the 1970's to mid 1980's. To improve management and recovery of shark stocks, the FMP stressed the need for better estimates on the assessment and monitoring of shark populations. Prior to 1993, most estimators of shark stocks were derived using fishery-dependent indices which generally lacked a standardized statistical sampling design. The validity of an abundance estimator (i.e. indices of abundance) depends on its accuracy and precision, and its robustness (the strength of its relationship with recruitment or stock size). Indices of shark abundance are currently used in production model analysis integrated using Bayesian statistical techniques (NMFS(1)).

Fishery-independent estimates of relative abundance are presently limited but can be the best estimator of shark stocks (NMFS(1,2,3)). Currently, only three surveys exist for monitoring shark relative abundance:

1) Musick et al. (1993) reported on a 17-year time series of abundance for sandbar, Carcharhinus plumbeus, and dusky, C. obscurus, shark from areas adjacent to the middle U.S. Atlantic coast.

2) Grace and Henwood (1997) performed pilot studies and have been conducting an assessment of the distribution and abundance of coastal sharks in the Gulf of Mexico and western North Atlantic since 1995.

3) National Marine Fisheries Service (NMFS), Narragansett Laboratory has executed shark longline surveys between Miami, Fla. and southern New England for 1986, 1989, 1991, and 1998 (Casey(4,5,6); Natanson(7)). However, most of these surveys are generally conducted in deeper waters ([is greater than] 10 m) where adult sharks mostly congregate. Neonate and juvenile sharks are commonly found in coastal nursery areas ([is less than] 10 m deep) where they feed and avoid predation (Branstetter, 1990) during summer months.

With the exception of Musick et al.(8) and Merson and Pratt's(9) work on juvenile sandbar sharks along the U.S. east coast, little data exists on juvenile shark stock size and recruitment to the adult portion of the population. Unlike most teleost species, the relationship between stock size and recruitment is direct, owing to the reproductive strategy of low fecundity combined with few, fully formed offspring (Holden, 1977). Quantitative estimates of juvenile abundance can provide promising alternatives to traditional hindcasting models and could improve the ability to assess current and future shark stock size and strength. Herein, we report on a 3-year fishery independent assessment of juvenile coastal shark populations in U.S. waters of the northeastern Gulf of Mexico derived using two methods.

Materials and Methods

Study area

Two regions were established as fixed sampling areas in the northeastern Gulf of Mexico (Fig. 1). The criteria for establishing these areas were based on a priori shark abundance survey information (Trent et al.(10)) and depth strata. The depth strata were between 1-5 m and 5-10 m.

[Figure 1 ILLUSTRATION OMITTED]

The first area (shallow stratum) is located in St. Andrew Sound. This area is a small semi-enclosed marine lagoon with expanses of submerged vegetation, Thalassia spp. and Halodule spp. It is about 14.5 km long and 0.2-2.0 km wide and has mean water depths of 3-5 m. Salinity ranges from 25-36% and tidal amplitude averages 0.42 m. The sound exchanges water with the Gulf of Mexico through passes about 0.5-2.0 km wide.

The second area (deep stratum) is located off St. Vincent Island at the southwest end of the Apalachicola Bay system. This area is about 1-3 km south of St. Vincent Island in the Gulf of Mexico where water depths average 5-10 m. The bay system surrounding this area is largely a line of barrier islands fronting the intersection of the Apalachicola delta and is the only bay system in Florida in which a large river system drains. As a result of river discharge, there is little submerged vegetation due to high turbidity. Salinity fluctuates from 15 to 35% and tidal fluctuation averages 0.66 m.

Sampling Gear and Survey Design

Gillnets

A 186 m long gill net consisting of panels of six different mesh sizes was utilized for sampling. Stretched mesh sizes (SM) ranged from 8.9 cm (3.5") to 14.0 cm (5.5") in steps of 1.27 cm (0.5"), with an additional size of 20.3 cm (8.0"). Panel depths when fishing were 3.1 m. Webbing for all panels, except for 20.3 cm, was of clear mono-filament, double-knotted and double-selvaged. The 20.3 cm SM webbing was made of #28 multifilament nylon, single-knotted, and double-selvaged. When set, the nets were anchored at both ends.

Longlines

The longline was constructed of a mainline made of two 152 m lengths of 425.8 kg test monofilament line. A 15.2 m length of 0.79 cm diameter braided polypropylene line connected each 152 m length, and the entire line when fished was 319.2 m long. Polyethylene floats made of 1.5 m lengths of 136 kg test monofilament line with a snap were attached to the mainline every 30.4 m. A standard longline consisted of 10-20 gangions placed at 15.2 m intervals along the mainline. Gangions were 0.9 m long and composed of snaps, aluminum sleeves, hooks (Mustad(11) #12/0, no 2888), and monofilament lines (136-kg test). Bait was either menhaden, Brevoortia spp., or Atlantic mackerel, Scomber scombrus. The mainline, when set, was tethered to an anchor on each end with a 30.4 m, 0.79 cm polypropylene rope between the anchor and the end of the mainline. A buoy (3.6 m aluminum pole with 1.8-kg weight and 50.8 cm poly float), with a strobe light and flag extended 2.4 m above the float, was attached at each end of the mainline.