Northeast Fisheries Science Center Reference Document 04-09
Gary R. Shepherd
Estimation of Striped Bass Discards in the Multispecies Groundfish Fishery, during the 2002 Fishing Year (May 2002 – April 2003)
National Marine Fisheries Serv., Woods Hole Lab., 166 Water St., Woods Hole MA 02543
Web version posted June 21, 2004Citation: Shepherd, G.R. 2004. Estimation of striped bass discards in the multispecies groundfish fishery during the 2002 fishing year (May 2002 - April 2003). Northeast Fish. Sci. Cent. Ref. Doc. 04-09; 15 p.
Information Quality Act Compliance: In accordance with section 515 of Public Law 106-554, the Northeast Fisheries Science Center completed both technical and policy reviews for this report. These predissemination reviews are on file at the NEFSC Editorial Office.
Estimation of discards in commercial fisheries is generally dependent on a subset of information collected by at-sea observations, which is then expanded to the total fishery. The method of expansion depends on the type of fishery. General methods for expansion include use of the ratio of discard weight of species A to kept weight of species A (Dspp a / Kspp a); the ratio of discard weight of species A to the aggregate kept weight of other species (Dspp / Kagg); or the ratio of discard weight of species A per unit of fishing effort. For striped bass fisheries which are not legal in federal waters, estimation of discards based on discard to kept ratio is not appropriate. Estimation of discards based on using a fishing effort ratio is difficult because the duration of commercial fishing trips varies among fisheries and ports, and measurement of effective effort is difficult to quantify consistently. Therefore, the most appropriate estimator for striped bass discards in the multispecies groundfish fishery is the ratio of striped bass discards to aggregate kept weight of species targeted by the fishery.
A simple ratio of cumulative discard to aggregate kept weight on observed multispecies groundfish trips is not appropriate due to seasonal and geographic variations in multispecies landings between Maine and North Carolina. To account for this variability, the ratios were stratified by month and statistical area. Statistical area information (Figure 1) for landings was only available from vessel logbook data (VTR data). All trips reporting landing of multispecies in the 2002 VTR logbook database were initially included. Species landed weight was summed within trip, then across trips by month, 3-digit statistical area and gear type. The data were limited to otter trawl and sink gillnets since most of the multispecies groundfish observer trips were made on vessels using these gear types; these gear types are also the most likely to capture striped bass. The VTR results were compared to the reported dealer landed weight data for species comprising the multispecies group. With the exception of white hake (which are landed in a variety of configurations, e.g. headed, or gutted) and halibut (which comprise a very small % of the total), the VTR data averaged 5% less than weighout data (Table 1). Therefore, the VTR landings were adjusted upward by 5% to account for all landings. Monthly landings that did not have area reported were re-distributed based on proportion of landings within each area. The 2002 VTR and observer data sets were subset to May through December to correspond to the fishing year beginning May 1st.
The observer data consist of haul specific information from trips targeting multispecies groundfish. The aggregate weight of multispecies groundfish kept per haul was summed across trip, gear, area and month. The data were also limited to sink gillnet and otter trawl gear. In addition, the weight of striped bass discarded was summarized by month, area and gear type. The data were insufficient to stratify at any finer geographic level than statistical area. A ratio of striped bass discards to aggregate landed weight of the multispecies complex was then calculated by month, area and gear. Observer data for the period January to April 2003 contained no record of striped bass caught or discarded; therefore the expanded estimate of striped bass discarded by the multispecies groundfish fishery in these months was zero (0). VTR reported landings from cells (month, area, gear) with observer coverage accounted for 89% of total landings in the otter trawl fishery and 73% from sink gillnets (Table 2 and Table 3). A monthly summary of landings and observed striped bass discards is provided in Table 4.
In cells (gear, month, area groups) with observer coverage (Table 5 and Table 6), the discard ratio was applied to expanded VTR aggregate landings data. In cells with no observer coverage for the multispecies fleet, a ratio for the general geographic area was applied. Areas were grouped as follows: Gulf of Maine (511-515); Rhode Island to the Great South Channel (521,526-539); western Georges Bank (522,525); eastern Georges Bank (542-543,551-562) ; Long Island south (611 and higher). An overall ratio for an area group was calculated as the sum of observed striped bass discards in the group/sum of observed multispecies landings within the same group. The resulting ratio was applied to VTR landings for each area in the group (Tables 5 and 6). The resulting striped bass discard estimates (Table 7 and Table 8) were summed across cells by gear type, with the assumption of 100% discard mortality in both gear types.
The result was a total of 289,808 lbs of striped bass discards in the multispecies groundfish fishery: 287,019 lbs. from otter trawls in May-December 2002 (Table 7); 2,789 lbs from sink gillnets in May-December 2002 (Table 8); and no discards during January to April for either gear. Beginning in June-July, there appears to be a north to south progression of striped bass discards in the trawler fleet. The majority of discards occurred in statistical area 521, which corresponds to the Great South Channel. Discards of striped bas in September and October in Great South Channel accounted for 84% of the total annual estimated trawl discards of bass in the multispecies groundfish trawl fishery. The NEFSC research vessel bottom trawl survey data (Figure 2 and Figure 3) indicate that the Channel has a seasonal concentration of striped bass (Figure 2), particularly in the autumn when migrating bass spatially overlap with spawning Atlantic herring.
This analysis was intended to examine striped bass discards in the multispecies groundfish fishery. Striped bass encounters within this fishery are a relatively rare event (Figure 4). However, on occasion seasonal/geographic aggregations of striped bass result in high levels of discards. Since statistical area was the lowest resolution for geographic stratification, discards may actually be over-estimated for some areas. Expansion by all landings within broad areas may encompass groundfishing locations which are not in the migratory pathway of striped bass and therefore unlikely to result in striped bass discards. In contrast, discards may also be under-estimated in areas with limited or no observer coverage, such as inshore locations where the majority of the striped bass population occur. Since the analysis did not include all fisheries potentially discarding striped bass, the estimate should not be considered a total estimate of commercial striped bass discards. Nonetheless, striped bass discards in the multispecies groundfish fishery appear to be a localized, seasonal event.