SALTONSTALL-KENNEDY GRANT NA06FD0170

JANET SMOKER
FISHERIES INFORMATION SERVICES
JUNEAU, ALASKA SEPTEMBER 2000

INTRODUCTION

Purpose of the Report

Skates and sharks are the major components of the "other species" quotas under the Fisheries Management Plans for Groundfish in the Bering Sea and Aleutian Islands (BSAI) and the Gulf of Alaska. In 1998 the BSAI quota of "other species" was exceeded by 15%. In 1999 bycatch of skates by some vessels participating in Community Development Quota (CDQ) fisheries was higher than anticipated, threatening boat quotas. In 1998 the Alaska Board of Fisheries took action to close the state commercial fishery for sharks and establish permit requirements for fishing for skates and rays in state waters. Given these concerns and the wider international concern about shark conservation, the North Pacific Fisheries Management Council has recently initiated action to develop amendments to the two Plans, affecting sharks and skates.

This project is intended to expand information about the distribution of sharks and skates in a way useful to fisheries planners, and to hook-and-line fishermen, the group of Alaskan fishermen who most frequently encounter these species. It also provides recent information about utilization; given that some level of bycatch is inevitable in hook-and-line fisheries, there may be potential to develop markets for fish that are currently being discarded.

Description of Fisheries

Groundfish fisheries off Alaska pursued by hook-and-line (H&L) vessels include Pacific cod, Greenland turbot and sablefish. The Bering Sea/Aleutian Islands (BSAI) H&L Pacific cod season opens January 1 and continues until its first trimester quota is reached, usually in late March or April. The Gulf of Alaska Pacific cod fisheries, also starting January 1, vary by area and may close due to cod quota or halibut bycatch cap, but are generally all closed to H&L gear by late winter. Fishing for Greenland turbot in the BSAI opens May 1 and usually closes the same month. The sablefish fishery is managed under an individual fishery quota system and is open March 15 to October 15, with most of the fishing taking place during the more clement months May to September. The second season BSAI cod season opening has been either September 15 or September 1 (there was no second season in 1993). It lasts at least through October, and sometimes into December, depending on quota. In 1999 and 2000, about a dozen H&L catcher/processors with Community Development Quota (CDQ) cod quotas fished outside of the open-access season, including summer months when cod fisheries have historically been closed.

Data Source

Data provided in this report are from records taken by at-sea observers, which have been audited, put into final format and stored in the NORPAC database. The bulk of the Pacific cod and Greenland turbot catch is taken by large catcher/processors, freezer longliners which have fisheries observers on board at least 30%, and in most cases 100% of their fishing time. In contrast, only about 50 of the more than a thousand sablefish boats carry observers, and rarely for more than a week or two during the season; there are few observer records compared to those of other H&L fisheries.

PROCEDURE

FIS requested from the NORPAC database all H&L records with skate or shark catches indicated. Data were delivered as a single Excel database, including fields for trip, date of set, location of set, number of hooks, total catch, species, species number and weight, and estimated retention.

I imported this file into Microsoft Access, and for ease of analysis sorted it by year and produced Excel files with annual data. Next I sorted each annual file by species. For sharks, I calculated Catch per unit effort (CPUE) in number of fish per 1000 hooks deployed, as this is the effort standard in use for analyzing bycatch of a similarly infrequent species group, birds. For skates, I calculated kilograms per 1000 hooks.

For each set I also calculated the weight of species retained, by multiplying the observer's estimate of retained percent by the set weight.

Next I converted the latitude/longitude information to the decimal degree format that could be used by the Geographic Information Systems (GIS) program ARCVIEW.

Most of the time I spent on this project was in determining the appropriate time/area strata, and testing options of how to display the data in order to make it most understandable.

Time/Area Strata

I originally anticipated using monthly time strata. After producing drafts of monthly maps for two sample years, it became clear that this level of detail for was not particularly helpful. Changes in skate bycatch distribution in particular mainly reflected the movements of the fleet as determined that approximate the seasons of the different longline fisheries described above. I did the same for sharks, except that for 1998 I developed monthly maps.

Data points

Variable point data can be portrayed with varying symbol size (as I did in my 1998 S-K project where I mapped halibut bycatch in H&L Greenland turbot fishery), or color variations (as in NOAA technical memorandum NMFS-AFSC-88). With a very large number of points, color variations are easier to interpret. I portrayed three classes (data ranges) to portray, similar to the NOAA technical memorandum which used high, medium and low CPUEs.

I classified the CPUE data by the "natural breaks" method, which is the default classification method in ArcView. Natural Breaks finds groupings and patterns inherent in data by finding breakpoints between classes using Jenk's optimization method. The Jenk's method minimizes the sum of the variance within each of the classes. Of course each subset of data produced somewhat different natural break points, and it was necessary to pick common cut-off points for each map in order to achieve consistency. For skates, I choose breaks at 75 and 250 Kg/1000 hooks. For sharks, I choose 1 and 10 individuals/1000 hooks.

Confidentiality

A general rule that has been applied to use of NMFS-collected data is to produce only summaries that include 4 or more reporting entities. While this is easily done for graphs or tables, it is difficult to apply with newer techniques of data visualization such as GIS. It can be accomplished by pooling data, either by time or area strata. FIS's first (1993) S-K project pooled data by month and 1 degree by 1/2 degree rectangles, consistent with similar efforts at the time, and sufficient for its purpose of an overview of halibut bycatch. However, that approach to pooling data defeats the purpose when the intent is to map areas of concentrations or scarcity of a species.

I originally planned to map all data provided, portraying Gulf and BSAI separately. After producing the first draft set of maps I realized that areas to the extreme west and east (West of 175 degrees west in the Aleutian Islands and east of 146 degrees in the Gulf) included observations from very few trips (boats). In some time-strata only one or two trips were included. In consideration of confidentiality of data from these boats, I truncated the map. This lead to a loss of one piece of important information: relatively large numbers of sharks are taken in the summer in the Eastern, particularly Southeastern, Gulf of Alaska. Possibly NMFS researchers currently involved in shark research in this area can use this NORPAC information to aid in their ongoing studies.

OBSERVATIONS (all table, figures, and maps can be linked to through a table at the end of this document)

Data summaries

Table 1. summarizes weights and numbers of all shark species reported, and two skate categories (unidentified and identified), by year. An average size is also calculated. Because the category "unidentified sharks" is so large, I am reluctant to make assumptions about trends in catches of the identified shark species. Less than 0.1% of the skates were identified by species.

Table 2. summarizes NMFS' "best blend" estimated catch of "other species" by hook-and-line gear for years 1993 to 1999 (1992 is unavailable) and compares it to total catch weight of sharks and skate in NORPAC database for each of these years.

Maps

Pages 1-24 are maps of skate locations and CPUE category by trimester for years 1992-1999. Pages 25-42 are similar maps for sharks by trimester for years 1992-1997. I prepared monthly maps for 1998; these are on pages 43-53. Maps 54, 55 and 56 are for trimesters of 1999.

Retention

There are only 11 records in the entire database that indicate any retention of sharks. These include four records from 1997 which indicated 9 spiny dogfish shark and 3 Pacific sleeper shark were retained, four records from 1998 which indicated 3 spiny dogfish shark and 43 Pacific sleeper shark were retained, and 1 record from 1999 showing retention of 5 spiny dogfish shark.

For skates, I looked at the last three years in detail. During these years, observers made estimates of the percent of fish retained in each observed set. In 1997, 22 of 176 observer cruises included some records with some skates retained. Comparable numbers for 1998 are 21 of 189 cruises, and in 1999, 43 of 198 cruises. The increase in 1999 is a result of the requirement of full retention of all groundfish species on CDQ vessels. This also accounts for the increase in actual amounts, shown in figures 1-3.

DISCUSSION

There has been recent interest in pursuing a directed fishery for skates when Pacific cod closes. By choosing rates in excess of 250 kg/1000 hooks, these maps display rates that are in the upper 10% of CPUEs (rather than upper third of CPUEs as shown in the NOAA tech memo). It is more likely, though clearly not guaranteed, that boats would encounter sufficient concentrations of skates in those areas in order to qualify for a skate target. Conversely, fishing in areas where skate CPUEs are less than 75 kg/1000 hooks decreases the probability that skate bycatch will exceed a CDQ boat quota, or that large quantities of skates will be caught, discarded and wasted.

Discard of skates (as calculated from this database) was 79% in 1997, 86% in 1998 and 84% in 1999. Although there are monthly variations in retention, they are not consistent among years. The higher retention rates in the fall of 1999 are at least partly the result of required retention in CDQ fisheries that occurred during that time. Improvement in skate retention during other target fisheries will be driven by market conditions and the degree to which individual boats can be flexible in their processing operations.

Sharks comprise from 2-5% of the "other species" H&L bycatch on an annual basis. As such they do not present much of a concern for fishermen as far as contributing to bycatch quotas, or as an opportunity to develop new products or markets. Thus, the chief value of the shark distribution maps may be as a contribution to the developing information base about sharks in North Pacific waters. Much is still to be learned about the distribution and habits of all the species of sharks, a group unique life history characteristics separates them from most of the other fish taken by H&L fishermen under current management plans.

REFERENCES

Fisherman's Guide to Catch and Bycatch in Alaskan Hook-and-Line Fisheries. 1994. Fisheries Information Services.

A Study of Bycatch Avoidance by Hook-and-Line Fishermen in Greenland turbot fisheries. 1998. Fisheries Information Services.

Fritz, L.W., A. Greig and R.F. Reuter. 1998. CPUE, length and depth distributions of major groundfish and bycatch species in the Bering Sea, Aleutian Islands and Gulf of Alaska regions based on groundfish fishery observer data. U.S. Dep. Commerce, NOAA Tech. Memo. NMFS-AFSC-88.

EA/RIR/IRFA for Amendment 62/63 to the Fishery Management Plans for the Groundfish Fisheries of the Bering Sea/Aleutian Islands and Gulf of Alaska to Revise Management of Sharks and Skates. 1999. Staff, North Pacific Fishery Management Council and Alaska Dept. Fish and Game.

AKNOWLEDGEMENTS

I thank Dr. Dan Ito of the Observer Program and his staff for provision of the NORPAC data. I also thank Thorn Smith, director of North Pacific Longline Association, and owners of freezer-longliners who provided the industry support for this project.