Ecological Archives E095-025-A6

James E. Palardy, Jon D. Witman. 2014. Flow, recruitment limitation, and the maintenance of diversity in marine benthic communities. Ecology 95:286–297. http://dx.doi.org/10.1890/12-1612.1

Appendix F. Detailed effects of flow on recruitment onto empty plates.

Without exception, the species recruiting onto empty plates represented a subset of the species present in communities at the same site and time. Overall, the number of species recruiting to empty plates in the control flow treatment ranged between 4.1±0.6 per plate in York and 5.1±0.0 per plate in Jakolof, while the number of species recruiting to empty plates in enhanced flows ranged between 5.3±0.8 per plate in Kittery and 6.9±0.8 per plate in Jakolof (Fig. F1). As with the species density of communities, the species density of epifaunal communities on plates scraped clean each sampling period were higher in Alaska (5.9±0.4) than in Maine (4.8±0.4).

When scaled up from recruitment plate, average within-site species richness of recruits on empty plates in the control flow treatment ranged between 10.5 species per observation period in Kittery and 15.1 species in Jakolof, while the number of species recruiting to empty plates in enhanced flows averaged 8.0 more species per observation period, ranging from 18.3 species in Kittery to 22.8 species in Seldovia (Fig. F2).

FigF1

Fig. F1. The number of species recruiting onto empty plates increases with flow. The number of species recruiting onto empty settlement plates (average number of species per 100 cm² ± SE) was higher on plates exposed to enhanced (open circles) flow than on plates exposed to control (filled circles) flow treatments at sites in Maine (York, Kittery) and Alaska (Seldovia, Jakolof) in 2007. Effects of site, flow treatment, time, and time × flow are statistically significant (complete statistical analysis in Table F1). All n = 16.


 

F2

Fig. F2. Species richness of recruitment on empty plates increases with flow. The richness of species recruiting onto empty settlement plates (Chao2 ± SE) was higher on plates exposed to enhanced (open circles) flow than on plates exposed to control (filled circles) flow treatments at sites in Maine (York, Kittery) and Alaska (Seldovia, Jakolof) in 2007. Within all sites, the effect of flow is statistically significant (statistical analysis in Table F1).


 

Table F1. Analysis of recruitment on empty plates. Linear mixed effects model assessing effects of time, flow treatment, enhancer, side of enhancer (nested in enhancer), and left to right position (nested in side) on the number of new species recruiting and the number of recruiting species as a proportion of species present in communities on flow enhancers deployed in 2007.

Term

df

MS

VC

χ²

p

Time

14

290.25

 

819.72

<0.01

Flow

8

488.14

 

299.17

<0.01

Time × flow

7

4.23

 

20.77

<0.01

Site

1

 

0.19

105.05

<0.01

Enhancer

1

 

<0.01

0.92

0.34

Side

1

 

0.01

0.81

0.37

Position

1

 

<0.01

0.13

0.72

Error

 

 

1.19

 

 

The abbreviation "df" is the degrees of freedom used in the χ² test; MS is the mean square of fixed effects; VC is the variance component for random effects. χ² is the result of likelihood ratio tests assessing the importance of model terms using nested models for both fixed (using maximum likelihood) and random (using restricted maximum likelihood) effects.


[Back to E095-025]