Webb and Kneib: Individual growth rates and movement of Litopenaeus setiferus in a tidal marsh nursery 



379 



Shrimp were marked and released only in the upper 

 reaches of each tidal creek. During the marking process, 

 small batches of shrimp (<50 individuals) were collected 

 and held in insulated plastic coolers. Water in the coolers 

 was exchanged each time a new batch of shrimp was col- 

 lected. Only active individuals, 40-80 mm total length 

 (TL, tip of the rostrum to end of the telson) and in appar- 

 ently good condition, were candidates for marking. The 

 marking process required a minimum field crew of two 

 researchers. One measured shrimp and recorded data, 

 while the other injected tags and released marked shrimp 

 at the edge of the tidal creek. 



We attempted to mark 1000 shrimp during a 3-day pe- 

 riod in each tidal creek near the beginning of every month 

 from July to November in both years. It was not possible 

 to tag shrimp in both creeks simultaneously; therefore 

 shrimp were marked and released in week 1 during the 

 first week of the month, then in creek 2 during the second 

 week of the month. The remainder of the month was spent 

 collecting marked shrimp (Fig. 2). Inclement weather 

 and mechanical problems with the tagging equipment 

 sometimes prevented us from achieving the goal of tag- 

 ging 1000 shrimp per creek within the first week of each 

 month. In August 1998, the tag injector malfunctioned on 

 the first day of tagging in SC and was unavailable for sev- 

 eral weeks while it was being refurbished. Consequently, 

 sampling was suspended in SC during that time. Although 

 the same problem occurred while we attempted to mark 

 shrimp from the EF in September 1999, we continued 

 sampling in an attempt to recapture shrimp tagged in 

 previous months. 



A total of 6 sampling events after marking were planned 

 in each creek per month (Fig. 2). A sample consisted of the 

 combined contents of 10 haphazard casts of the net along 

 the edge of the tidal creek within each station per sampling 

 date. Shrimp populations were usually sampled at 3-day 

 intervals for 21 days beginning from the midpoint of the 

 marking period. A consistent exception was the second 

 sample in the series, which occurred at a 6-day interval to 

 accommodate the marking effort in the second creek and 

 to keep the sampling effort consistent in all creeks. Inclem- 

 ent weather interrupted the sampling schedule on occasion 

 and when unfavorable conditions persisted for more than 3 

 days, some of the planned sampling events after marking 

 were cancelled; some months were represented by fewer 

 than 6 sampling events. Sampling was terminated when, 

 as a result of normal seasonal emigration from the nursery 

 areas, shrimp densities declined to the point that they could 

 no longer be consistently collected from the tidal creeks by 

 cast net (19 November 1998 and 21 November 1999). 



Catches of marked shrimp from each station were re- 

 tained in separate plastic bags, placed on ice, and trans- 

 ported to the laboratory. A subsample of shrimp from each 

 station (every tenth individual) was measured (TL, mini 

 to construct size distributions. If a sample included fewer 

 than 100 shrimp, all were measured. Sex was not deter- 

 mined. All individuals were scanned for the presence of 

 tags and when a marked shrimp was detected, it was mea- 

 sured (TL, mm) before the tag was removed and stored in a 

 plastic vial for reading at a later date. For each recapture. 



10 15 20 25 



Days of the month 



30 



Figure 2 



Schedule for monthly marking and postmarking field 

 schedule for the juvenile white shrimp study. 



we recorded date, creek, station of recapture (i.e., distance 

 from original release site) and size offish. 



Daily instantaneous (specific) growth rates (mm/[mm/ 

 d] ) were calculated as 



[(In L 2 - In Lj)/t], 



where L 2 = total length (mm) of an individual on the date 

 of recapture; 

 L 1 = initial total length (mm) on the date of tag- 

 ging; and 

 t = number of days at large. 



Daily absolute growth rates (mm/d) also were calculated 

 (L 2 -L 1 /t) to facilitate comparison with estimates from 

 previous studies. Displacement (distance moved) was 

 determined by comparing the location of recapture with 

 the original location at marking. Residence time within a 

 tidal creek was determined from a plot of the proportion of 

 recaptured individuals against time-at -large for each creek 

 system. First, using the Regression Wizard in the com- 

 puter program SigmaPlot® (version 8.0, SPSS, Inc. Chicago, 

 ID, we fitted the data to an exponential decay function: 



(y=a x e~ bt ) , 



where y = the proportion of total recaptures; 



t = time at large; and 

 a and b = the estimated parameters. 



Constraints imposed on the fit were a = l (because the 

 proportion of total recaptures could not exceed 1) and 

 fe>0 (because this was an exponential decay function). 

 Mean residence time for shrimp in each creek was then 

 estimated from the area below the fitted curves describing 

 the proportion of recaptures with time at large. This was 

 calculated with the macro function "area below curves" 

 included in the "Toolbox" menu selection of SigmaPlot® 

 (vers. 8.0, SPSS, Inc. Chicago, ID which uses the trapezoi- 

 dal rule to estimate the area under curves. 



Statistical analyses 



Most of the data analyses used statistical procedures in 

 version 8.0 of the computer software package Systat® 



