ZIMMERMAN ET AL.: SELECTION OF HABITAT BY PENAEUS AZTECl 'S 



the presence of estuarine macrophytes can be as- 

 sociated with an increase in epifaunal abundance 

 (Heck and Wetstone 1977; Heck and Orth 1980) as 

 well as providing protective cover (Vince et al. 

 1976; Nelson 1979; Coen et al. 1981; Heck and 

 Thoman 1981). For shrimp selecting vegetated 

 marsh, this may translate into a greater variety 

 and abundance of food and some degree of protec- 

 tion from predation. 



Zonal and Areal Relationships 



Penaeus aztecus demonstrated a greater degree 

 of attraction to vegetated habitat in the inner than 

 the outer marsh. Accordingly, shrimp densities 

 were higher among vegetation and lower on non- 

 vegetated bottom in the innermost zone compared 

 with the outer zone. This relationship is 

 adequately reflected by comparing ratios of vege- 

 tated with nonvegetated shrimp density. Using the 

 ratios, the change in selection from the outer, mid- 

 dle, to inner zone was 3.3:1, 11.0:1, and 27.7:1, re- 

 spectively. The percent area covered by S. alter- 

 ni flora (Fig. 2) also increased (by a factor of three) 

 from outer to inner marsh, but as vegetational 

 coverage increased arithmetically selection by P. 

 aztecus increased geometrically (Fig. 6). This im- 

 plies that salt marshes with more vegetational 

 coverage have disproportionately greater attrac- 

 tive value to P. aztecus than do those with less 

 coverage. On a larger scale, Turner ( 1977 ) revealed 

 a positive correlation between extensiveness of es- 

 tuarine vegetation and offshore shrimp yield. 

 However, the relationship may not be simple; it is 

 likely to depend upon characteristics such as the 

 configuration, accessibility, and quality of vegeta- 

 tional patches within a marsh. For instance, an 

 edge effect has been identified which associates 

 large numbers of shrimp with the nonvegetated 

 zone adjacent to vegetation (Mock 1966; Christmas 

 et al. 1976). Since our Spartina habitat was inter- 

 tidal, and often not inundated during low tides, 



the nonvegetated subtidal habitat provided a ref- 

 uge against stranding. We have assumed that it 

 did and that shrimp redistributed accordingly 

 each tidal cycle. It is evident that an increase in 

 the amount of ecotone edge (between habitats) 

 would facilitate movement for the shrimp popula- 

 tion. It is also evident that the amount of edge 

 is proportionally related to the degree of retic- 

 ulation in the marsh (Fig. 2). Thus, reticulation 

 may be an important mechanism for increasing 

 the accessibility of intertidal vegetation to P. 

 aztecus. 



Shrimp Densities 



Density estimates for penaeid shrimp in S. al- 

 terniflora vegetation have not been reported pre- 

 viously. We found a density range for P. aztecus in 

 Spartina habitat of 0.7 to 43.2/m 2 with an overall 

 mean, from March through July, of 11.7/m 2 (1 SD = 

 9.4, n = 81). Comparable densities from adjacent 

 nonvegetated habitat ranged between and 

 18.2/m 2 . All densities were taken when P. aztecus 

 numerically dominated the shrimp population. By 

 August, when P. setiferus first began to dominate, 

 the combined mean for both species in vegetation 

 increased to 50.8/m 2 (1 SD = 31.6, n = 12) and a 

 single sample attained a density of 118.6 shrimp/ 

 m 2 . These data may indicate a potential for higher 

 P. aztecus densities earlier in the season and 

 suggest that P. aztecus were not restricted by lack 

 of space. 



To our knowledge, we have provided the first 

 accurate estimates of shrimp density in marsh 

 vegetation, and our densities are among the few 

 available for any estuarine system. Due to method 

 limitations, most researchers have only reported 

 relative abundances of restricted sizes, usually 

 over nonvegetated bottom. The single exception 

 was data by Allen and Hudson (1970), using a 

 suction sampler in seagrasses in Florida Bay. 

 From 43 trials, they reported a mean of 6.2/m 2 ± 

 3.4 SD for P. duorarum. 



Estimates off! aztecus densities from nonvege- 

 tated bottom in three other Galveston Bay salt 

 marshes were available from the Texas Parks and 

 Wildlife Department (TPWD) from 1976 through 

 1981 (Benefield 1982, footnote 5.). The data were 

 taken using a marsh net (Renfro 1963) which was 

 relatively effective for capturing shrimp on non- 

 vegetated bottom (Table 6 compares a beam trawl, 

 similar to the marsh net, with other sampling 

 devices). Mean TPWD densities for P. aztecus dur- 

 ing the latter half of March were 10.4/m 2 for 1976, 

 5.2/m 2 for 1977, 0.3/m 2 for 1978, 1.3/m 2 for 1979, 

 8.7/m 2 for 1980, and 5.1/m 2 for 1981 with an overall 

 mean of 5.2/m 2 . In our study, on nonvegetated bot- 

 tom, the March mean for P. aztecus was 0.9/m 2 and 

 overall (March through July) the mean was 1.4/m 2 . 

 It is evident that our nonvegetated densities for P. 

 aztecus were within the range, but low compared 

 with the mean calculated from TPWD data. 



These densities of P. aztecus may not be strictly 



5 R. L. Benefield, Bay Shrimp Project Leader, Texas Parks and 

 Wildlife Department, Coastal Fisheries Branch, RO. Box 8, 

 Seabrook, TX 77586, pers. commun. September 1982. 



333 



