Renaud et al.: Shrimp loss by TEDs in U.S. coastal waters 



135 



Areas In most areas of the study, shrimp 

 catch rates for TED-equipped nets were 

 comparable with those for standard nets 

 (Tables 3-5). Statistically-significant reductions 

 ( 1.2 and 0.5 lb/h) in shrimp CPUE occurred off 

 west Florida in nets equipped with Super 

 Shooter TEDs and Georgia TEDs with fun- 

 nels, respectively. Off Louisiana, standard nets 

 caught 0.3 lb/h more shrimp than the paired 

 Georgia TED-equipped nets with funnels. 

 Shrimp CPUE was higher by 0.7 and 1.0 lb/h 

 for standard nets paired with Georgia TED- 

 equipped nets without funnels offshore of Texas 

 and the east coast (Florida and Georgia), 

 respectively. 



CPUE comparisons with commercial shrimp 

 fleet in the Gulf of Mexico Average shrimp 

 CPUE (heads-off lbs/24 h day/4 nets) by Sta- 

 tistical Area groupings and seasonal groupings 

 for standard nets was compared with CPUE 

 ( heads-off lbs/24 h day/4 nets) for standard nets 

 on other commercial vessels fishing in the same area 

 and season in the Gulf of Mexico. Our initial assump- 

 tion that our data were representative of commercial 

 fishing conditions was supported by standard net 

 CPUEs on commercial observer vessels that were not 

 significantly different (paired r-test, P>0.05) from 

 CPUEs on commercial vessels without observers. Mean 

 differences ranged from a 6.2 lb/h gain by standard 

 nets on TED observer vessels to a 4.9 lb/h gain by 

 standard nets on other commercial vessels. In three of 

 seven season/area combinations, shrimp CPUE from 

 TED-observer vessels was higher than CPUEs of other 

 commercial vessels. Since there were no significant 

 differences in net size during our study, we assumed 

 that this was the case for the rest of the commercial 

 fleet. TED-observer vessels were apparently represen- 

 tative of other commercial vessels in the fleet fishing 

 in similar areas during the same season. Similar analy- 

 ses for the Atlantic fishery could not be made since 

 catch information was not available on a trip-by-trip 

 basis. 



Biological yield models 



Ricker-type yield models (Ricker 1975) developed for 

 each of the three major shrimp species show the same 

 basic curve shape (Fig. 3; Nance & Nichols 1988). The 

 curves are asymptotic where yield estimates are plot- 

 ted for current fishing mortality rates (F-multiplier = 

 1.0). Thus, with current fishing patterns and current 

 fishing mortality rates, little increase or decrease in 

 yield is predicted with the minor reductions in F that 



would be expected due to small losses of shrimp by 

 TEDs. 



Yield estimates were calculated in the model by vary- 

 ing the F-multiplier in increments of 0.02. Mean shrimp 

 loss with TED-equipped vs. standard nets varied from 

 1 to 14% by TED type. A decrease of 5% in F would 

 result in an undetectable change in annual yield in 

 the brown or white shrimp fisheries and a 1% reduc- 

 tion in the annual yield of the pink shrimp fishery in 

 the U.S. Gulf of Mexico. 



Discussion 



Our data were collected by NMFS observers during 

 cooperative cruises with shrimp industry participants. 

 Since this was a voluntary program, TED type, area, 

 and season of sampling were controlled by industry 

 participants. Data came from virtually any vessel 

 whose owner or captain would allow NMFS observers 

 aboard. 



Not all federally approved TED types were tested. If 

 a shrimper could not maintain TED efficiency during 

 a trip, the trip was aborted by the shrimper or the 

 TED was not used again. This resulted in nominal 

 imbalances in the data by area, season, and TED type, 

 including some data sets too small for analysis. 



Mean shrimp catch rates in TED-equipped nets were 

 lower than those in standard nets, varying from a loss 

 of 1.4% with Super Shooter TEDs to a loss of 13.6% for 

 Georgia TEDs without funnels. Nets equipped with 

 Georgia TEDs without a funnel were used mainly dur- 



