In their introduction Stohlgren et al. (2003) state that as recently as 2002 ecologists proposed that diverse communities can effectively exclude non-native species. They say “It would be comforting to believe that areas with many plant species are less prone to invasions than those with fewer species. Botanical hotspots … might repel the frequent arrival of plants from other regions ... [and] control techniques might be limited to heavily invaded, species-poor areas …” However, they point out that exceptions to this assumption are obvious, such as the spread of tamarisk, Russian olive, and purple loostrife to species-rich wetlands.
The paper is based on two large data sets. The first is from 316 large vegetation monitoring plots in eight states under the auspices of the United States Department of Agriculture (USDA). The plots are spaced (one per every 63,942 ha) throughout the United States (US). Each consists of four 168 m2 subplots with three 1 m2 quadrants in each subplot. Plots were sampled once each summer between 1997 and 2001.
The Biota of North America Program directed the second data collection over 20 years from 44 states. This is based on 229,000 records of native and non-native plants by county.
Figures 1 and 2 in Stohlgren et al. are the Biota of North America data. Both show a positive correlation between non-native and native plants throughout the US.
The authors conclude that areas rich in native plants also support more non-natives; that “the rich get richer.” They acknowledge that there may not be any direct cause-effect relationship. Species richness of both natives and invasives may be correlated to habitat heterogeneity; high species turnover; and/or increased pulses of light, nitrogen, or water. They end with the question: “How are so many species-rich areas successfully invaded?”
In a subsequent issue of Frontiers, Rejmanek questions the data and conclusions of Stohlgren et al.. This letter includes a figure (see Figures/Tables) showing a positive relationship between human population in a state and number of non-native plant species. Rejmanek states that “Admittedly, the whole question of major patterns of non-native species richness in the US is more complicated (data are not independent but auto-correlated). One point, however, is clear: the human population size and the length of intensive disturbance/introduction history are the two primary factors in determining species richness of non-native flora.”
Rejmanek points out that the positive correlation Stohlgren et al.’s Figure 2 is due to an outlier, California, and that removal of these data greatly reduces the significance of the relationship. He uses a stepwise regression analysis of plant data from 50 states to show that human population size (H) in a state explains 60% of the variance for the regression. Further, “… when H is included as an independent variable, together with native species richness, contribution of the last variable becomes non-significant and negative (Table 1, regression 2).” Rejmanek concludes that “Based on stepwise regression analysis, the best available predictors of non-native species richness are H1/2 and the time since the state was admitted to the Union ...[which] is used as a surrogate for the length of intensive disturbance/introduction history.”
In this letter, Renne and Tracy argue that small scale studies do allow for control of environmental factors, however this is not possible in Stolgren et al.’s large scale analysis (see Notes to Faculty for more details).