Plants of Southern California: Distribution of Natives Vs. Non-Natives

Because non-native plants have a comparative advantage over native plants (see "Unnatural" Competition From Non-Native Plants), I expected that I would find a difference in the distribution of non-native taxa among my plant trail guides compared to the distribution of native taxa. In particular, I expected that I would find a much higher percentage of non-native taxa to be found on many trails compared to the percentage of native taxa found on many trails.

Note that this is different from analysis of the physical numbers of a species in a given area; the analysis here is on how many plant trail guides contain at least one specimen of each taxon. The abundance of a given taxon in each area will be a separate analysis.

Example: arundo generally becomes a horribly invasive weed whenever it is introduced to a riparian area. In numbers of plants in such an area, it would quickly dominate. However, most trails among my plant trail guides do not contain significant riparian areas, and hence the number of plant trail guides that contain arundo is relatively low; only 8 plant trail guides of 91 contain arundo. In the following analysis, the species Arundo donax is associated with the number N=8 representing those 8 plant trail guides and N=23, representing 23 floras in my database.

The results astonished me. The overall distribution of non-native plants among my 91 plant trail guides and among the 47 floras I've digitized matches that of native plants quite well. In both cases, common taxa are rare; rare taxa are common; see How Common Are The Plants Of Southern California?.

The data show that there is indeed the effect I expected, but it is much weaker than I expected. The percentage of non-native taxa that are widespread shows, by one measure, an increase of 34% in my plant trail guides, and 19-57% in the floras I've digitized, when compared to the percentage of native taxa that are widespread.

Specifically, 22% of non-native plant taxa are fairly widespread in my 91 plant trail guides, defined as being present in at least 11 trail guides, whereas 17% of native taxa are widespread using the same definition.

Similar results are obtained from the floras I've digitized. Using the same cutoff of 11 floras, the enhancement is 57% (21% vs 13%). To try to make the comparison more similar, I adjusted the cutoff point for being widespread to be the same 11/91 of the 47 floras, which is 6 of the 47 floras. Using that cutoff of at least 6 floras, the enhancement is 19% (34% vs 19%). Neither number is exactly comparable to the number for the plant trail guides, due to my trails not having the same geographic distribution as the floras.

This does not invalidate the fact that non-native taxa have an unfair advantage over native species; it simply means that the fundamental law of biology about plant distributions has not been repealed. I.e., most plant species still find only certain conditions and habitats that they like, and that doesn't change when they are introduced to an area where they are not native. Instead, their unfair advantage shows up in their numbers in a given area. Non-native plants are represented by many more plants in a given area than these same species would have in any area in which they are native.

An example is the prickly pear cactus, native to the western U.S. Prickly pears rarely become a problem and take over an entire area in their native home (except sometimes when non-native cattle overgraze an area). Yet in Australia, in habitats similar to their native habit, prickly pears multiplied uncontrollably until some of their predators from the U.S. were introduced to help control them. Yet even in Australia, you won't find prickly pears growing in riparian or montane areas, since they can't tolerate such conditions.

Similarly, arundo is never going to be found outside riparian areas in Southern California. The number of plant trail guides in which it could theoretically be present is limited to the number of plant trail guides that contain riparian areas to its liking. Its advantage as a non-native is seen when it quickly dominates the entire ecosystem in riparian areas where it is a non-native. However, in its native lands, it never dominates the entire ecosystem because it is controlled by a host of pests that are found in its native range.

Here are the plots showing the results, using my database as of 7 July 2005. The plots show the percentage of species found on N different lists. A complete explanation of what is being shown here is given in How Common Are The Plants Of Southern California?. See that reference if you don't understand these plots.

The first plot for each list type shows the basic distribution. More taxa are found on only a single trail or flora (N=1) than are found for any other value of N, such as N=2 (taxa found on exactly two trails or floras), etc. In both the plant trail guides and the floras, about 30% of all taxa are found on only a single trail or flora. The numbers drop dramatically with higher values of N; only 15-19% of all taxa are found on exactly two trails or floras. Thus nearly half of all taxa are found only on single trail or in a single flora or on exactly two trails or two floras.

The numbers are explicitly given for 1-10 lists (N=1 to 10), and all numbers for more than 11 lists are lumped in the point plotted at a value of 11. This lumping is done simply to quickly show the percentage of species found at all values with N above 10. Thus, for example, 22% of all non-native taxa are found on 11 to 55 trail plant guides, and this number of 22% is plotted at the value of N=11.

The second plot for each list type shows the tail of the histogram, for number of lists larger than 10. This second plot shows how that number of 22% is broken down into for individual values of N. For example, a single non-native taxon, 1 out of 256 non-native taxa (1/256 = 0.4%) was found on 55 plant trail guides.

The numbers of taxa in this analysis were as follows:

Plant Trail Guides: 256 non-native taxa and 1,194 native taxa

Floras: 594 non-native taxa and 2,996 native taxa

Important Caveat: these numbers come from a non-homogeneous database in which plants are not always identified down to the same taxonomic level. That is, some trail guides and floras lists only the species for species that have subspecies or varieties which are given in other trail guides or floras. Hence the precise values for taxa found on N trails will change somewhat when I eventually homogenize my database by obtaining the same level of ultimate taxon identification in all lists. This will not change any conclusions above; only the precise numbers. Any other analysis using the numbers here may be strongly affected by these inhomogeneities; user beware!

Go to:

Copyright © 2005 by Tom Chester
Permission is freely granted to reproduce any or all of this page as long as credit is given to me at this source:
Comments and feedback: Tom Chester
Last update: 7 July 2005