iNat as a tool for documenting novel species invasions

The attention we pay to an introduced/non-native species tends to be correlated to the magnitude of the impact of that species (e.g. invasive species), or how widespread the species is in its introduced range. We tend to mostly ignore introduced species when they are localized, and not affecting a large ecological impact on the native ecosystem. This makes sense as a matter of prioritizing attention and resources, but it is also at odds with two major objectives of invasion and conservation ecology, and management:

1. To better understand how and why species become invasive, or widespread after introduction.
2. To contain or eradicate introduced species before they become widespread, problem species.

To address these objectives, we have to devote specific study to newly introduced, or newly naturalizing species. Instead of trying to work out the history of invasions after they have unfolded, we can study invasions as they unfold. In a sense this is akin to the axiom that the news is "the first rough draft of history". Following that analogy, I think that biologists need to make more of an effort reporting the news when it comes to introduced species.

I have found myself increasingly pulled into studying recently introduced, or recently naturalizing, species in SoCal, particularly in the San Gabriel Valley and San Gabriel Mountains foothills, which is my botanical home territory. There are several species at the start of their invasion histories in this area, that have received relatively little attention, the ranges of which are poorly documented. I have found iNaturalist to be incredibly useful for this purpose.

iNat makes the practical work of documenting novel invasion histories very much easier, and more realistic. Take for example the introduced plant Crassula colligata subsp. lamprosperma (Aussie pygmyweed), which was first documented around the mouth of the San Gabriel Canyon in 2003. I stumbled on this species in 2022, when looking for a similar, and related, native species, sand pygmyweed (Crassula connata). Once I was keyed into this species, I began to find it freakin' all over the place in the San Gabriel Valley, and in scattered localities in the San Gabriel Mountains.

When I first noticed Aussie pygmyweed, I had identified it by keying it out in the formidable Jepson Manual, which is not an accessible resource for a casual naturalist to use. At that time Aussie pygmyweed had so few observations on the iNat that it hadn't been included in the computer vision ID model at that point. Consequently, when people found and posted observations of Aussie pygmyweed, the top ID suggestion would usually be the California native, sand pygmyweed, which was included in the computer vision model at the time, and these observations would be incorrectly identified as sand pygmyweed. This was true even of many observations of Aussie pygmyweed in Australia!

I went through a fairly extensive push to comb through the non-research grade observations of sand pygmyweed, and one other widespread, non-native Crassula species, and in doing so found a good number of new localities of Aussie pygmyweed in SoCal, that had previously been posted as C. connata. I also started searching out new localities of Aussie pygmyweed in my area, and posting detailed photos. Shortly thereafter, the iNat computer vision model was updated, and Aussie pygmyweed had enough photos to be incorporated. Since then I have seen new observations of Aussie pygmyweed, with correct IDs, sporadically popping up on iNat. This has already led to the discovery of new localities in SoCal.

What is qualitatively different about the documentation of a novel invasion on a platform like iNat, is that casual, non-expert observers, without public announcements or solicitation, can now meaningfully contribute to the documentation of novel invasion histories, simply by stumbling on new localities and posting them. Previously, this kind of documentation process would be the work of a much smaller number of academics and other trained field biologists, and the sharing of their individual discoveries would be inefficient, or might not happen much at all. Because of platforms like iNat, we now have much more power to document these invasions as they unfold, rather than trying to infer what happened in later historical reconstruction.

Glen R. Morrison.
-29 March 2024

Posted on 29 March, 2024 18:30 by grmorrison