Is there truly “no saturation in the accumulation of alien species worldwide”?

sea-water-port-industryIn reading group last week (May 11th) Hannah Fraser brought the group a paper reflecting a synergy between QAECO interests and her current role in the Centre of Excellence for Biosecurity Risk Analysis. This paper by Seebens et al. (2017) “No saturation in the accumulation of alien species worldwide” evaluates the trends in the introduction of alien species from 1800 to 2000. They have three hypotheses

  1. The rate at which taxa that are deliberately introduced (mammals, birds and vascular plants) should be declining
  2. The rate at which taxa are accidentally introduced should be increasing due to increases in trade
  3. The rates of alien species introductions should vary between countries depending on country’s history and biosecurity regulations

They show that, overall, the rate of introductions is increasing and that this increase is particularly stark for algae, fungi and invertebrates which are thought to be primarily introduced by accident. They note that the rate of mammal and fish introduction has declined in recent years, possibly consistent with their hypothesis that the rate of deliberate introductions is falling.

Next, Seebens et al. (2017) evaluate the relationship between introduction rate and trade volume across taxa by fitting a Michaelis–Menten curve to the data and evaluating its explanatory power. They found high r2 values for several taxa that are thought to be introduced by accident: algae, fish, and invertebrates. The relationship between trade volume and introduction rate was not as compelling for other taxa including some that are likely to be introduced accidentally (e.g. arachnids) as well as those that are likely to be deliberately introduced (e.g. mammals).

Their analyses neatly provide evidence that the rate of alien species introductions are increasing worldwide but have some substantial limitations that are not drawn out in the text and mean that the findings should be interpreted with caution.

  • As their title suggests, Seebens et al (2017) discuss this information in the context of ‘saturation’. Theoretically, a region can only contain a certain number of species, after this point it is saturated and any further introductions of species will either be unsuccessful or displace an existing species. In order to saturate a country with alien species, these species must be spread across the whole country. Otherwise, there will still be portions of the country that are unsaturated and vulnerable to species introductions. However, the evidence used in their study is based on first records which is also taken as the time when the species first occurs, at which point it definitely has not spread across the whole country. Some discussion of this limitation would put their study in a more realistic context.
  • The study covers 16 taxa and 282 regions over 200 years but does not account for the effect of changing sampling effort or effectiveness. Over the past 200 years, the amount of effort dedicated to searching for alien species and the effectiveness of these searchers has increased dramatically. This means that species are much more likely to be found and identified as alien species in the year 2000 than they were in 1800. The trajectory of this increase will be different between countries and taxa. Countries place differing emphasis on biosecurity and the environment which is reflected in the quality and quantity of data they collect. Further, it is easier to identify species using modern techniques and taxonomies. This is especially pertinent to historically understudied taxa (e.g. invertebrates) and taxa with species that are very difficult to distinguish (e.g. fungi, algae). Accounting for this bias could greatly reduce the reported increase in introduction rate but is very difficult to do, particularly over a global dataset. In reading group discussions we drew out some suggestions that could help address this issue and, we believe, make the findings of the study more accurate and reliable.
    • Evaluating the increase in the rate of first records against a null model that reflects a realistic increase in reporting rate over time would provide a more realistic estimate of introduction rate
    • Accounting for differences in the detection of first records between taxa by assuming that the likelihood of detecting and alien species from a particular taxa is proportional to the identification rate of species in that taxa. For example, the identification rate of fungi and invertebrates is still increasing which suggests that we should expect the rate of first detections in that taxa to still be increasing – as we record more alien species that may have been present for many years but only identified as an alien species recently. This would give a coarse measure of the differences in detection rate between taxa that could potentially be used to correct for inter-taxa differences.
    • The previous two options may not be feasible on a global scale due to differences in the quality and availability of data. However, it would be possible to demonstrate the kind of effect that accounting for reporting rate might have on results by drawing out an example from a region with high quality data. Drawing out a detailed case study where it is possible to account for reporting rate would show the kind of effect that changes in sampling effort might have on these results.
  • When considering the relationship between trade and taxa specific introduction rate the authors only consider one model form (the Michaelis–Menten curve) which forces the curve to pass through the origin and plateau off. This is based on an understanding of how introduction rate is supposed to relate to trade. However, the model form does not appear to fit all taxa well. For example, crustaceans may be better modelled using a linear relationship. Comparing the fit of several model forms would allow for a more nuanced evaluation of differences in response to trade.

These complaints aside, Seebens et al. (2017) have put a lot of effort into compiling and cross-checking an amazingly large dataset of first record rates, providing fantastic opportunities for future research. Further, their study investigated changes in introduction rates at an unprecedented scale, providing useful insights, as long as the results are taken with a grain of salt.

This entry was posted in Uncategorized. Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s