Matlock et al.: Estimating deepwater fish populations 



97 



merit and emigration rates are unknown, but were 

 probably low. If they occurred, recruitment must have 

 been less than emigration because the populations were 

 depleted in the study area. As recommended by Ricker 

 (1975), our fishing effort was concentrated into "a 

 rather short period of time" to minimize the effects of 

 violating this assumption and that of no natural 

 mortality. 



The grouper population based on submersible fish 

 data may have been overestimated because the esti- 

 mated number of fish exceeded the estimated number 

 of burrows and double counting of fish probably oc- 

 curred. On one dive, the same fish (based on a scar on 

 the lower jaw) was seen four different times. 



Additional research is needed to determine the im- 

 pacts of each of the above factors on fish population 

 estimates based on counts made from submersibles. 

 Future burrow counts should include all burrows, not 

 just apparently active ones. Transect width should be 

 accurately measured by counting burrows only within 

 the range of a fixed physical extension from the sub- 

 mersible. Occupancy rates for tilefish and yellowedge 

 grouper should be determined in randomly selected 

 areas at night when they are most likely to be in their 

 burrows. 



Although no significant relationship between CPUE 

 and cumulative catch was found for grouper, a more 

 intensive effort should be made before discounting this 

 technique. Additional longline collections over a longer 

 period for yellowedge grouper are needed to determine 

 if using the Leslie method is feasible. 



Longlines can potentially impact stocks of tilefish. 

 The population estimate of tilefish in the study area 

 (39-128) and the catch made by the intensive fishing 

 effort (79) indicate that from 62 to 100% of the fish 

 were taken out of the area by an effort of approximate- 

 ly 6000 hook-hours, which is a 1.5-2 day effort by a 

 commercial longliner (Prytherch 1983). Catch rates in 

 the northern Gulf of Mexico in 1982 averaged 1-6 fish 

 per 100 hook-hours (Prytherch 1983). Based on the 

 estimated population size within the area, the initial 

 catch rates indicate that the longline effectively catches 

 all fish out of an area that is at least 12 m wide. Some 

 fish are attracted from greater distances (Grimes et 

 al. 1982), and some near the longline are not caught. 

 But the number removed from the population is equal 

 to the length of the longline x a width of 12 m x fish 

 density. 



Estimates of the total portion of the Gulf of Mexico 

 inhabited by tilefish have not been developed, but the 

 optimal areas are limited by depth, temperature, and 

 bottom type (Grimes et al. 1980, 1986; Grossman et al. 

 1985). This, combined with slow growth rate, longev- 

 ity, and low natural mortality (Turner et al. 1983, Har- 

 ris and Grossman 1985), indicate that overfishing could 



easily take place if substantial effort is expended in 

 tilefish habitat. This is especially true in light of the 

 susceptibility to mass mortalities caused by sudden 

 temperature reductions (Hachey 1955). Data from 

 South Carolina tilefish habitat show a substantial 

 decline in catch rate and mean size over a 4-5 year 

 period with low to moderate effort (Low et al. 1983). 

 Further, the number of tilefish burrows per km 2 in the 

 Middle and South Atlantic Bights in the early 1980s 

 was 241 and 125, respectively (Able et al. 1987). These 

 estimates are much lower than the 1600 burrows per 

 km 2 in the Gulf of Mexico estimated in this study. 



More extensive longline studies of yellowedge 

 grouper catches are required to assess the effect of 

 longlines on these populations. The population estimate 

 of yellowedge grouper in the yellowedge study site 

 from fishing activities was not significant, but the best 

 estimate (26 animals) from the non-significant regres- 

 sion may be realistic. The regression indicated that 

 similar fractions of the yellowedge grouper population 

 (40%) would be caught at similar levels of effort as com- 

 pared with tilefish, and similar impacts from the long- 

 line fishery might be expected. However, the results 

 may not be analagous because different gear were used 

 in the two areas. 



While the association with hard substrate and high 

 relief was expected for yellowedge and other groupers, 

 the burrowing habits were not expected. A detailed 

 description of grouper habitat and burrow character- 

 istics have been provided by Jones et al. (1989). The 

 finding that this species also inhabits burrows was 

 especially significant. If this were the only habitat, it 

 would limit their distribution and increase their sus- 

 ceptibility to fishing once they are located. However, 

 this species is also associated with rock and reef habitat 

 typical of other grouper species. This diversity of 

 habitat should enhance the survivability of the species 

 overall, but it makes a part of the population more 

 susceptible to longline fishing. 



The uneven distribution of tilefish and yellowedge 

 grouper between fished and unfished areas was also 

 unexpected. Reasons for the differential distribution 

 are not apparent. But the effects of depth, temper- 

 ature, and bottom type on the fish were probably 

 involved. 



This study demonstrates the need for additional 

 research to estimate population sizes and life-history 

 parameters for deepwater Gulf fishes to quantify the 

 amount of fishing they can support. Routine monitor- 

 ing of these populations could be accomplished with 

 longlines fished during August through October. 

 Limited data on tilefish and yellowedge grouper have 

 been collected with bottom longlines by the National 

 Marine Fisheries Service since 1968 (Table 4). 

 However, the data are insufficient to identify trends. 



