NOTE Meyer et al : Effects of live-bait shrimp trawling on seagrass beds and fisfi bycatch 



195 



Holding pens were cubical, 1.25 m on each side, 

 constructed of 6.4-mm mesh and had removable mesh 

 tops. A 5-cm diameter polyvinylchloride collar was 

 attached around the top sides of the pen to provide 

 floatation. The mesh top and float collar prevented 

 fish from escaping over the sides, and the top pre- 

 vented avian predation. Lead-core line was sewn to 

 the bottom seams of each pen to maintain pen vol- 

 ume. Loops attached to each corner of the pen en- 

 abled us to anchor the pens with conduit poles. 



Owing to low abundance offish bycatch in Novem- 

 ber, collection methods were modified; the bycatch 

 from both of the paired trawl nets was combined (only 

 a single trawl net was processed in August), and 

 placed into a single holding pen. Handling procedures 

 were also changed to estimate fish bycatch mortal- 

 ity more conservatively by eliminating catch culling 

 time. In November the catch was placed directly into 

 an aerated 76-L transport tank and during trans- 

 port to the hold site, algae, detritus, and hardhead 

 catfish were removed. Thus, bycatch mortality re- 

 sults for the two months were not directly compa- 

 rable. In addition, during November the holding pens 

 were deployed prior to bycatch collection, and the 

 trawl catch was subsequently transferred from the 

 76-L transport tank into individual holding pens. 



Pens were checked for dead fish (no opercular 

 movement) following initial placement at the hold 

 site, and then at intervals of 2, 8, 12, 24, and 36 hours. 

 Checks involved inspecting each holding pen by sys- 

 tematically lifting portions of the pen netting so that 

 the entire volume, net sides, and bottom could be 

 visually inspected for dead fish while allowing sur- 

 viving fish to remain immersed. Dead fish were col- 

 lected at each time check and preserved in 10% for- 

 malin. At 36 hours, all dead and live fish were sepa- 

 rated and preserved. Fish were identified, and stan- 

 dard length and body depth were measured (to the 

 nearest 0.5 mm). The number of individuals and 

 weights of each species (to the nearest 0.01 g) were 

 recorded for each trawl and holding pen time check. 

 Fish with a body depth <7.5 mm were excluded from 

 survival measurements because they were able to 

 escape through the mesh and would cause an unrep- 

 resentative mortality estimate. 



Because the number of individuals per species per 

 trawl was generally <5, we pooled all 30 trawls for 

 each month to assess survival. The more numerous 

 fish species (spotted seatrout, Cynoscion nebulosus, 

 pinfish, Lagodon rhomboides, and silver perch, 

 Bairdiella chrysoura) were subdivided into small (25.0- 

 65.0 mm standard length, SL) and large (65.5-125.0 

 mm SL) size classes for mortality comparisons. Mojarra, 

 Eucinostomus spp. (including silver jenny, Eucino- 

 stomus gula), were smaller and were subdivided into 



small (25.0-55.0 mm) and large (55.5-95.0 mm) size 

 classes so that we could examine size-related mortal- 

 ity. Determination of differential size-class mortality 

 was tested with log-linear model analysis (P<0.05). 



Salinity and air and water temperatures were 

 measured hourly during bycatch trawling. 



Results 



Turtlegrass 



Regression analysis to estimate mean rate of decrease 

 showed no significant (P>0.05) reduction in mean 

 shoot density, number of blades per shoot, longest 

 blade length per shoot, total blade length per shoot, 

 or above- and below-ground biomass with increased 

 trawling during either month. Monthly vegetation 

 measurements at each trawl level (all nine sites com- 

 bined) are shown in Table 1. 



Fish bycatch 



We collected a total of 5901 fish representing 42 spe- 

 cies; 3262 fish (29 species) in August, 2639 fish (36 

 species) in November. Mojarra (including silver 

 jenny) were 79.3% (August) and 52.1% (November) 

 of the total catch (Table 2). Most mojarra were 

 smaller than 40 mm SL and were identified as 

 Eucinostomus spp., whereas all larger Gerreidae 

 could be identified as silver jenny. The unreliability 

 of the identification of smaller Gerreidae (those 

 smaller than 40 mm SL) precluded their definitive 

 identification to species (Matheson and McEachran, 

 1984). Mojarra less than 40 mm composed 60.0% 

 (August) and 24.7% (November) of all fish collected, 

 and silver jenny represented 19.3% (August) and 

 27.4% (November) of all fish collected (Table 2). 



Survival was variable (0-100% ) among species and 

 months (Table 2; Fig. 2). Among the abundant spe- 

 cies (those with at least 20 individuals for each 

 month), high survival was observed for gulf toadfish, 

 Opsanus beta, pigfish, pinfish, and gray snapper, 

 (Table 2; Fig. 2). Abundant species with low survival 

 were silver perch, mojarra, silver jenny, and spotted 

 seatrout. Greatest mortality in most species occurred 

 within the first 8 or 12 hours after collection (Fig. 2). 



Small fish were significantly (P<0.05) more sus- 

 ceptible to trawl-induced mortality than larger fish 

 (Table 3). This was particularly evident for silver 

 perch, spotted seatrout, pinfish, and mojarra (includ- 

 ing silver jenny). 



Species-specific survival was greater in November 

 than August (Table 2; Fig. 2). For those species for 

 which at least 20 individuals were collected each 



