According to the exchange indices, water exchange at Little Pecan Lake (without accounting for 

 flow reductions through stop-log operation and flap-gate closures) was reduced much more (at least 

 15.8 times more) than at the unmanaged Grand Bayou site. If ingress were directly proportional 

 to the volume of inflow, then the reduction in water exchange should cause a proportional 

 reduction in white shrimp production. However, estimated white shrimp production from Little 

 Pecan Lake is only half that measured at the unmanaged Grand Bayou site. It appears that 

 relatively severe reductions can be affected in water exchange without a proportional reduction in 

 white shrimp production. 



This hypothesis implies that recruitment of estuarine organisms is not a totally passive 

 phenomenon. Field and laboratory research has shown that penaied shrimp are capable of 

 detecting difference in salinity (Hughes 1969; Penn 1975) and using vertical migration (Hughes 

 1969; Staples 1980; Hartman et al. 1987) and lateral migration (Copeland and Truitt 1987; Hartman 

 et al. 1987) to immigrate into nursery areas. Arnold and Cook (1980) called this behavior selective 

 tidal stream transport. Similar phenomena have also been documented for numerous immature 

 estuarine-dependent fishes (King 1971; Fore and Baxter 1972; Sabins and Truesdale 1974; Hartman 

 et al. 1987). 



Having detected the plume of fresher water exiting Little Pecan Lake through the open water 

 control structure, post-larval white shrimp may have been able to position themselves for 

 recruitment on subsequent flood tides. This active means of recruitment may explain why 

 recruitment and production of white shrimp is not directly proportional to water exchange. 



An operable water control structure, such as a flap-gated structure, will not provide any ingress 

 opportunities unless it is opened during the recruitment "season." Because high densities of post- 

 larval white shrimp sometimes occur sporadically (Smith 1979), the water control structure should 

 be kept open for as long as possible during the recruitment season in order to maximize ingress. 



In studying the effect of a low-level fixed-crest weir Bradshaw (1985) and Herke et al. (1987b) 

 concluded that the weir caused a substantial reduction in recruitment of white shrimp and other 

 estuarine organisms. The comparison of white shrimp production within Little Pecan Lake versus 

 that from the low-level fixed-crest weir site suggests that a properly operated variable-crest flap- 

 gated culvert may allow greater recruitment of white shrimp than the low-level fixed-crest weir. 

 If this is so, one would expect the flap-gated variable-crest culvert would provide even greater 

 recruitment opportunities than a standard fixed-crest (crest elevation 15.28 cm below marsh level) 

 weir. Furthermore, the weir crest length at the Grand Bayou site was much longer (over 500%) 

 than that recommended by current Soil Conservation Service guidelines. The culvert capacity at 

 Little Pecan Lake was 31% greater than that recommended by Soil Conservation Service guidelines. 

 If water control structures were installed at both sites according to current Soil Conservation 

 Service guidelines, white shrimp production from the Grand Bayou site would likely be even lower 

 relative to that from Little Pecan Lake. 



If recruitment of white shrimp and other estuarine organisms occurs according to the hypothesis 

 of selective tidal stream transport, one might expect that white shrimp production from the low- 

 level fixed-crest weir site would be higher than it is. However, post-larval white shrimp may not 

 have actively selected this site for recruitment if fresher water was not flowing out over the weir 

 on falling tides. Data indicate that salinities within semi-impoundments regulated by the low-level 

 fixed-crest weir and a standard fixed-crest weir (crest elevation 15.2 cm below marsh level) were 

 higher than adjacent outside waters during the white shrimp recruitment season (Rogers et al. 1987; 

 Herke et al. 1987b). 



178 



