Ecological Archives E091-061-A4

Adam M. Siepielski, Keng-Lou Hung, Eben E. B. Bein, and Mark A. McPeek. 2010. Experimental evidence for neutral community dynamics governing an insect assemblage. Ecology 91:847–857.

Appendix D. Detailed results of the observational study.

Individual Enallagma species per capita growth rates (Table D1), mortality rates (Table D2), and relative abundances (Table D3) were not correlated among species. In fact, the only correlation with an uncorrected significance value of P < 0.05 was a positive one (Table D2, which is not what is predicted if niche differences between species promote their coexistence (Chesson 2000). After appropriate Bonferroni correction, no correlations were statistically significant. These results suggest no evidence of interspecific trade-offs among species that would promote their coexistence, and given that the lake environments are heterogeneous (Appendix B, Table 1), this also suggests no evidence of, for example, a competitive hierarchy, which would indicate fitness differences among the different species among lakes.

Consistent with this latter result, individual Enallagma species per capita growth rates (Table D4), mortality rates (Table D5), and relative abundances (Table D6) were not correlated with any of the measured environmental variables. The lack of correlation between relative abundance and environmental factors is also supported by the multivariate cannonical correspondence analysis between individual species relative abundances and environmental variables (E. divagans: Wilk’s λ  = 0.398, F6, 3 = 0.75, P = 0.649; E. ebrium/hageni: Wilk’s λ = 0.608, F6, 10 = 1.07, P = 0.437; E. geminatum: Wilk’s λ = 0.693, F6, 10 = 0.74, P = 0.632; E. pictum: Wilk’s λ = 0.720, F6, 6 = 0.39, P = 0.862; E. vesperum: Wilk’s λ = 0.773, F6, 10 = 0.49, P = 0.803). Therefore, we find no evidence that each species is regulated by different ecological factors, implying that coexistence mechanisms are absent or at best very weak.

In contrast, we found a strong statistically significant correlation between total per capita mortality rate (r2 = 0.40, P = 0.038; Fig. D1a) in relation to total Enallagma abundance, but no correlation between total per capita growth rate (r2 = 0.31, P = 0.08; Fig. D1b) and total Enallagma abundance. However, the mortality and growth rates of individual Enallagma species were not statistically significantly correlated with total Enallagma abundance (Table D8). Overall, these patterns are consistent with our experimental study, and suggest that abundance is regulated at the level of total Enallagma abundance. Although no individual pair-wise correlations between total demographic parameters and environmental factors were significant (Table D7, after correcting for multiple comparisons) the cannonical correspondence analysis using the total Enallagma matrices of demographic parameters and environmental variables revealed that, overall, mortality and growth were significantly associated with environmental variables (Wilk’s λ = 0.0009, F18, 6.14 = 3.75, P = 0.052). Mortality rate increased with increasing fish and macrophyte densities (the latter two are positively correlated [r = 0.667, P = 0.025], suggesting macrophytes limit fish densities), whereas growth rate increases as prey abundance increased and fish abundance decreased. Total abundance was weakly positively associated with increases in prey abundance.

These patterns make intuitive sense and show that environmental variation among lakes does play an important density-dependent regulatory role. Greater predator densities frequently cause increased prey mortality (Sih et al. 1985). In this system, mortality imposed by fish predation appears to be key; a finding that further reinforces the experimental results. Similarly, greater prey availability should positively influence growth rates, if growth rate is food limited. Indeed, fish lake Enallagmagrowth rates are food limited as experimentally adding prey increases growth rates (McPeek 1990, 1998). These observational results are thus consistent with previous studies showing that competition and predation impart strong-negative density dependence on Enallagmagrowth rates and mortality rates in lakes with fish (McPeek 1990, 1998), but they indicate total abundance is the regulated abundance. These results also indicate that we did measure important environmental variables, however, the demographic consequences of environmental variation do not act to regulate individual species abundances only total abundance, as expected for ecologically identical species (e.g., Hubbell 2001, and our model developed above).

TABLE D1. Pearson’s correlations between Enallagma species per capita growth rates across lakes in New Hampshire, USA. No correlations were statistically significant (all P > 0.05).

Species

E. divagans

E. geminatum

E. ebrium/hageni

E. pictum

E. vesperum

E. divagans

1.000

 

 

 

 

E. geminatum

0.403

1.000

 

 

 

E. ebrium/hageni

0.0006

0.842

1.000

 

 

E. pictum

0.017

n.a.

0.995

1.000

 

E. vesperum

0.180

-0.299

0.557

0.508

1.000

 

TABLE D2. Pearson’s correlations between Enallagma species per capita mortality rates across lakes in New Hampshire, USA.

Species

E. divagans

E. geminatum

E. ebrium/hageni

E. pictum

E. vesperum

E. divagans

1.000

 

 

 

 

E. geminatum

0.991

1.000

 

 

 

E. ebrium/hageni

n.a.

*0.994

1.000

 

 

E. pictum

n.a.

n.a.

n.a.

n.a.

 

E. vesperum

0.243

0.590

-0.981

n.a.

1.000

P = 0.0063

 

TABLE D3. Pearson's correlations between Enallagma species relative abundances across lakes in New Hampshire, USA. No correlations were statistically significant (all P > 0.05).

Species

E. divagans

E. geminatum

E. ebrium/hageni

E. pictum

E. vesperum

E. divagans

1.000

 

 

 

 

E. geminatum

-0.318

1.000

 

 

 

E. ebrium/hageni

-0.230

-0.340

1.000

 

 

E. pictum

-0.563

-0.235

0.080

1.000

 

E. vesperum

-0.547

-0.268

-0.480

0.010

1.000

 

TABLE D4. Pearson's correlations between Enallagma species per capita growth rates and measured environmental variables across lakes in New Hampshire, USA. No correlations were statistically significant (all P > 0.05).

Variable

Species

E. divagans

E. geminatum

E. ebrium/hageni

E. pictum

E. vesperum

Lake productivity

-0.003

-0.121

-0.005

0.775

0.304

Macrophyte density

-0.045

-0.120

-0.508

0.746

-0.539

Fish density

-0.037

0.650

-0.414

0.974

-0.598

Newt density

-0.081

-0.269

-0.083

0.904

0.236

Dragonfly larvae abundance

-0.519

0.128

-0.221

-0.862

-0.452

Prey abundance

-0.175

0.083

0.030

-0.119

0.270

 

TABLE D5. Pearson’s correlations between Enallagma species per capita mortality rates and measured environmental variables across lakes in New Hampshire, USA. No correlations were statistically significant (all P > 0.05).

Variable

Species

E. divagans

E. geminatum

E. ebrium/hageni

E. pictum

E. vesperum

Lake productivity

0.534

-0.078

-0.459

n.a.

-0.382

Macrophyte density

0.538

0.025

0.143

n.a.

-0.134

Fish density

0.707

0.196

0.214

n.a.

-0.088

Newt density

-0.106

-0.319

-0.189

n.a.

-0.851

Dragonfly larvae abundance

-0.262

-0.204

0.271

n.a.

0.265

Prey abundance

0.416

-0.066

0.099

n.a.

-0.360

 

TABLE D6. Pearson’s correlations between Enallagma species relative abundance and measured environmental variables across lakes in New Hampshire, USA. No correlations were statistically significant (all P > 0.05).

Variable

Species

E. divagans

E. geminatum

E. ebrium/hageni

E. pictum

E. vernale

E. vesperum

Lake productivity

-0.134

-0.182

0.024

0.014

0.997

0.436

Macrophyte density

0.279

-0.32

0.295

-0.198

-0.848

-0.056

Fish density

-0.4

-0.164

0.264

-0.059

-0.335

0.119

Newt density

-0.36

0.045

0.32

-0.309

-0.420

-0.029

Dragonfly larvae abundance

-0.019

-0.116

0.426

0.088

0.818

-0.344

Prey abundance

-0.631

0.177

0.383

0.151

-0.290

-0.197

 

TABLE D7. Pearson’s correlations between Enallagma species per capita growth and mortality rates and total Enallagma abundance across lakes in New Hampshire, USA. No correlations were statistically significant (all P > 0.05).

Species

Per capita growth

Per capita mortality

E. divagans

-0.451

-0.217

E. geminatum

0.345

0.284

E. ebrium/hageni

0.337

0.317

E. pictum

0.219

n.a.

E. vesperum

0.280

0.029

 

TABLE D8. Pearson’s correlations between total Enallagma abundance, mortality and growth rates and environmental variables across lakes in New Hampshire, USA.

Variable

Total abundance

Total mortality

Total growth

Lake productivity

-0.205

0.208

-0.030

Macrophyte density

0.211

0.248

-0.422

Fish density

-0.069

0.322

*-0.696

Newt density

0.038

-0.037

0.337

Dragonfly larvae density

0.275

-0.045

-0.223

Prey abundance

-0.006

-0.110

0.510

*P = 0.02

 

 

   FIG D1. Relationship between (a) total per capita mortality and (b) total per capita growth rates in relation to the number of Enallagma per 1-m sweep of lake bottom (a measure of density) across lakes in New Hampshire, USA. “Total” is the estimated mortality rate or growth by combining all species.


 

LITERATURE CITED

Chesson, P. 2000. Mechanisms of maintenance of species diversity. Annual Review in Ecology and Systematics 31:343–366.

Hubbell, S. P. 2001. The Unified Neutral Theory of Biodiversity and Biogeography. Princeton University Press, Princeton, New Jersey, USA.

McPeek, M. A. 1990. Determination of species composition in the Enallagma damselfly assemblages of permanent lakes. Ecology 71:83–98.

McPeek, M. A. 1998. The consequences of changing the top predator in a food web: a comparative experimental approach. Ecological Monographs 68:1–23.

Sih, A., P. Crowley, M. McPeek, J. Petranka, and K. Strohmeier. 1985. Predation, competition, and prey communities: A review of field experiments. Annual Review of Ecology and Systematics 16:269–311.


[Back to E091-061]