开放获取期刊获得更多读者和引用
700 种期刊 和 15,000,000 名读者 每份期刊 获得 25,000 多名读者
James F. Reinhardt, Richard W. Osman and Robert B. Whitlatch
A mid-domain model was used to examine differences in patterns of species recruitment diversity between native and non-native species groups within shallow sub-tidal epifaunal communities of southern New England. A mid-domain model is a stochastic null model that predicts species richness patterns within a bounded domain. Deviations from the mid-domain model were examined to assess: 1) if each species group’s recruitment diversity differed from the mid-domain model; and 2) if there were differences in the patterns of deviation between groups. In general, the shape of mid-domain null was correlated to empirical patterns of recruitment diversity (i.e., mid-summer peaks in recruitment diversity) although among-year variation occurred in both the pattern of recruitment diversity and the level of fit. Akiake’s information criterion was used to determine if the mid-domain effect contributed to linear regression model predictions of phenological species recruitment patterns. We found distinct differences between the level of fit between native and non-native species groups. Native species did not significantly differ from the mid-domain null in five of seven years examined, while non-native species were significantly different in all years but two. Non-native species also showed a much weaker correlation to the mid-domain model and the seasonal water temperature signal. Collectively, this evidence supports the hypothesis that native and non-native species have different factors controlling the timing of their recruitment. Mid-domain nulls proved to be useful in the analysis of species richness patterns along temporal axes and provide a valuable baseline for further analysis.