Results

During the field experiment, mussel mortality was very low on reefs where sea stars were removed (Fig. 4.5A).This suggests that sources of mortality other than predation by sea stars (e.g., dislodgement by waves, consumption by crabs or river otters, etc.) had little effect during these experiments. Thus mortality on reefs where sea stars were at natural densities could be attributed almost entirely to predation by Pisaster. Both rates of predation and sea star density were dramatically reduced during a persistent, cold-water upwelling event (Fig. 4.5B, C), during which water temperature dropped about 3°C (Fig. 4.6A).

This drop in predation could not be explained by variation in potential heat stress (Fig. 4.6B) or maximum wave force (Fig. 4.6C), both of which were relatively consistent among periods. Maximum air temperatures and wave forces tended to be low during upwelling—conditions that should tend to increase rates of predation (Menge and Olson 1990). I used multiple regression to test whether variation in sea star predation (mean mortality after 14 days in four transplants/site/period) among time periods and sites was associated with (1) water temperature (mean during 27 high tides/period), (2) potential heat stress (mean of maximum low tide air temperature on 5 warmest days/period), or (3) maximum wave force (mean of maximum force/day on 5-7 days/period). Sea star

A no seastars upwelling event

18June

2July

17July

1Aug

15Aug

29Aug

Figure 4.5. Results of field predation experiments. The five consecutive 14-day periods are noted on the x-axis. Persistent upwelling occurred during the third period (July 17-August 1). All data are means ± SEM for three study sites. (A) Mussel mortality at end of each 14-day period on reefs with sea stars removed. (B) Mussel mortality in the presence of sea stars. Predation was significantly reduced during upwelling (analysis of variance (ANOVA), contrast period 3 vs. periods 1, 2, 4, 5; F153 = 14.10,p < 0.001). (C) Local sea star density around mussel clumps. Sea star density was also significantly reduced during upwelling (ANOVA, contrast period 3 vs. periods 1, 2, 4, 5; 53 = 5.33,p = 0.025).

18June

2July

17July

1Aug

15Aug

29Aug

Figure 4.5. Results of field predation experiments. The five consecutive 14-day periods are noted on the x-axis. Persistent upwelling occurred during the third period (July 17-August 1). All data are means ± SEM for three study sites. (A) Mussel mortality at end of each 14-day period on reefs with sea stars removed. (B) Mussel mortality in the presence of sea stars. Predation was significantly reduced during upwelling (analysis of variance (ANOVA), contrast period 3 vs. periods 1, 2, 4, 5; F153 = 14.10,p < 0.001). (C) Local sea star density around mussel clumps. Sea star density was also significantly reduced during upwelling (ANOVA, contrast period 3 vs. periods 1, 2, 4, 5; 53 = 5.33,p = 0.025).

predation was associated with water temperature (p < 0.001), but was unrelated to potential heat stress (p = 0.93) or wave forces (p = 0.60). Predation was consistently greater at one of the three sites, so site variables were significant in the model. Together water temperature and site explained 86.7% of the variation in mean mussel mortality on reefs with sea stars.

The reduction of local sea star density during upswelling was

3f 16

0 0

Post a comment