The Lepidoptera as Predictable Communities of Herbivores: a Test of Niche Assembly Using the Moth Communities of Morgan-monroe State Forest

—The response of forest insect communities to disturbances such as timber harvest likely will depend on the underlying ecological assembly rules that affect community structure. Two competing hypotheses are niche assembly, which seeks to demonstrate significant species-environment correlations, and dispersal-assembly, which seeks to demonstrate spatial autocorrelation in the absence of species-environment correlations. Unfortunately, many studies of forest management never explicitly test what factors are responsible for maintaining community structure prior to harvest. The goal of this study is to examine variation in the community structure of forest Lepidoptera using the pre-timber harvest data on Lepidoptera from 18 forest sites within 3 management units of the Hardwood Ecosystem Experiment at Morgan-Monroe State Forest (MMSF) in Indiana. We sampled 14,270 individuals representing 277 species of macrolepidoptera from MMSF in 2007. Canonical correspondence analysis suggested significant correlations between the moth assemblage found within a stand and levels of three environmental variables: importance of oaks, log-tree density, and percent basal area of shrubs and saplings. In contrast, Mantel tests suggested that forest moth communities at MMSF were, at most, weakly autocorrelated. The results here suggest some support of nicheassembly processes within forest macrolepidoptera. Still, a significant portion of the variation in species assemblages among forest stands remained unexplained, suggesting that stochastic factors and sampling bias may be important to consider when discussing patterns of lepidopteran diversity in space and time.