PCI and PC2 indicating their importance to characterizing samples collected 20 June and 02 July. PC A of each 

 date group separated in the first PCA also failed to segregate flow treatments or disturbance frequency 

 treatments. 



DISCUSSION 



Flow Effects 



The structures used to manipulate flow significandy altered flow velocity in the expected manner (Fig. 5), but 

 they did not significandy affect tvirbidity or macrobenthos overall (Table 2, Fig. 6). Flow treatments accounted 

 for only a small fraction of the total variance in abundance and biomass (<1.9 %). Significant interactions with 

 sampling date and disturbance frequency treatments indicate that altered flow affected macrobenthos only in 

 certain instances (Table 2). The strongest flow effects were found in undisturbed samples only (Table 3). In 

 undisturbed sediments, average abundance and biomass were greater at increased and decreased flow levels 

 than in control plots for all sampling dates except the first (Fig. 7). Increased abundance and biomass in 

 increased and decreased flow plots relative to control plots indicate the physical presence of experimental 

 structures may have affected macrobenthos. It is possible structure proximity afforded the community a refuge 

 from predation or increased food supply by increasing sedimentation rate. The difference may also explain the 

 flow*frequency interaction, because fi-equentiy disturbed (biweekly and monthly) samples had low abundance 

 and biomass due to recent defaunation. 



At the species level, Polydora comuta was found only at increased and decreased flow plots with one exception, 

 indicating a possible attraction to structures. Between 1994 and 1997, P. comuta was not observed at Rincon 

 Bayou (Montagna and Ritter unpublished data). The present study is the only report of its presence at 

 station C. 



Failure to detect significant abundance and biomass differences among flow treatments may arise from 

 inadequacy of the manipulations. Flow treatments did not alter flow velocity enough to significandy affect 

 resuspended sediment. Average flow velocities of weir treatment plots ranged between 10.1 and 16.4 cm s'. 

 These velocities may not have reached critical erosion velocities of the natural cohesive sediment of Rincon 

 Bayou. For example, critical erosion velocities of the Skeffling intertidal mud flat on the Hiunber Estuary, U.K. 

 ranged between 21.8 cm s ' and 30.8 cm s ' (Widdows et al. 1998). Community change driven by altered 

 resuspension rates would not be detected by the manipulations in the present study. 



Changes in flow velocity affects growth of some species. For example, growth of some macrobenthic 

 suspension feeders, such as Membranipora membranacea . Balanus crenatus . and Pseudochitinopoma 

 occidentalis . are inhibited by flow velocities 12-30 cm s ' (Wildish and Kristmanson 1979; Eckman and 

 Duggtns 1993). In contrast, growth rates of spionid polychaetes common to Rincon Bayou increase with 

 increasing flow. Polydora comuta and Streblospio benedicti increase growth rates with flow velocities between 

 9 and 18 cm s ' (Hentschel 1999a; 1999b). For example, P. comuta grew at a rate of 1.4 volumetric doublings 

 per day, reaching sexual maturity within 1 week at 18 cm s ' velocity (Hentschel 1999a; 1999b). Because the 

 response by the two spionid polychaetes common to the study area increased over the range of measured flow 

 velocities, inhibition probably did not occur near structures. It is more likely slight enhancement occurred, 

 explaining the small increase of abundance and biomass in increased flow plots (Fig. 7). 



The increase of abundance and biomass in increased flow (weir structure) plots is due to changes in population 

 size of Streblospio benedicti . Population growth of S. benedicti appears to have been enhanced at increased 

 flow plots prior to the flood of 22 June 1997. On 20 June, S. benedicti populations at increased flow plots were 

 twice that of decreased flow and control plots. Streblospio benedicti grows faster under higher flow conditions 

 (Hentschel 1999a; 1999b). The unique life history of S. benedicti could also be partly responsible for the rapid 

 population growth. Colonists could have been adults carrying a full brood (Levin 1984). 



Disturbance Frequency Effects 



Disturbance frequency treatments appear to demonstrate succession. Longer periods between disturbances 

 (i.e., biweekly to monthly to bimonthly) were associated with increasing abundance and biomass and decreasing 

 evenness (Table 1, Fig. 8). The trend demonstrates early colonization and successively greater abundance, 

 biomass, and less dominance with longer periods between disturbance. Lower evenness and abundance in 



Appendix F ♦ F-7 



