Intraspecific patterns of conspecific negative density dependence in tree species across the eastern United States

Background/Question/Methods

An emerging theme in the search for mechanisms that structure and maintain forest diversity is conspecific negative density dependence.  Theory predicts that negative density dependence could be an important process in maintaining diversity and recent empirical evidence supports that prediction.  Building on our previous examinations of conspecific density dependence, we examined variation within species in the strength of density dependence across the majority of many species’ ranges in the eastern US using Forest Inventory and Analysis data.  Through spatial pattern analysis and statistical modeling of the strength of negative density dependence by 2 degree latitude by longitude region, we analyzed individual species patterns and suggest possible mechanisms driving these patterns. We predicted that if natural enemies were driving regeneration patterns then signals of density dependence should be most negative in areas of higher precipitation and temperature where most natural enemies are more active. 

Results/Conclusions

Species level analysis revealed that 18 of the 53 species had significant spatial autocorrelation in the strength of density dependence and the remainder lacked a significant pattern at the regional scale. Initially, 131 species were screened, but 78 species lack sufficient sample size for Moran’s I test of spatial autocorrelation.  Further analyses of the species displaying significant spatial autocorrelation found that 12 of 18 species had significant negative correlations with actual evapotranspiration, which is an index that integrates temperature and precipitation with greater values indicating warmer and wetter conditions.  Only two species had significant positive correlations with evapotranspiration.  The common negative correlation with evapotranspiration is the opposite of what we would expect if seedling establishment were dependent on warmer and wetter conditions.  Although we are not able to definitely conclude that natural enemies are the mechanism causing these broad geographic patterns, the results are suggestive of insect or pathogen activity which tends to be greater in warmer and wetter environments.  The dominant pattern follows the proposal that enemies are likely to be stronger a strong control at lower latitudes and climatic extremes control species ranges at higher latitudes.