Taxonomic bias and lack of cross‐taxonomic studies in invasion biology

© The Ecological Society of America www.frontiersinecology.org Taxonomic bias and lack of cross-taxonomic studies in invasion biology Peer-reviewed letter Invasion biology is the study of nonnative species and their introduction to – as well as potential establishment, spread, and impact within – exotic ranges. Although the concepts of invasion biology are applicable across taxonomic boundaries, Pyšek et al. (2008) reported a strong taxonomic bias in the discipline. Such a bias is critical; if scientists study a biased subset of species, or disregard some taxa altogether, then important differences among taxa may be overlooked, thereby leading to incomplete or inaccurate generalizations. Here, we extend the work of Pyšek et al. (2008) by investigating whether the taxonomic bias in invasion biology has changed over time. Although past articles on non-native species may have focused on only a few groups of organisms, the field may have since become less taxonomically biased; we test this hypothesis on the basis of a systematic literature search using Thomson Reuters Web of Science. We also examine articles that consider nonnative species of multiple major taxonomic groups, assessing whether such cross-taxonomic articles have become more frequent over time. We present results for (1) articles about non-native species in general and (2) empirical studies focusing on six of the discipline’s major hypotheses – biotic resistance, island susceptibility, invasional meltdown, novel weapons, enemy release, and the tens rule (see WebPanel 1 for details) – which can be applied to organisms of different taxonomic groups. Of the investigated articles, most focused on plants, followed by vertebrate and invertebrate animals; few studies examined other taxonomic groups (Figure 1; cf Pyšek et al. [2008] who analyzed articles published until mid-2006). This taxonomic bias appears to be relatively stable since the late 1980s. Articles on plants were more dominant in the 1970s than today but, at that time, publications on non-native species were rare. The total number of publications per year about non-native species has been rising through time (Richardson and Pyšek 2008; Kühn et al. 2011). Plant-associated taxonomic bias detected in invasion biology in general was found to be more pronounced for certain hypotheses (biotic resistance, novel weapons, enemy release), less pronounced for others (invasional meltdown, tens rule), and even reversed for one hypothesis where most studies have focused on vertebrates (island susceptibility; however, this hypothesis is the one affiliated with the fewest studies; Figure 2). The main reason why non-native plants are studied more frequently than other non-native species is probably because most known nonnative species are plants. Considering the large number of recognized nonnative plants in Europe, for instance, this taxonomic group indeed seems to be understudied (Figure 1; cf Pyšek et al. 2008). Conversely, abundance estimates of other non-native taxa (eg algae, fungi, bacteria) are unavailable because of the paucity of studies about such organisms. One concern is whether numbers of known nonnative species are the cause or consequence of observed differences in research across taxonomic groups. Also, although there are many nonnative plants, the percentage of harmful non-native plants is relatively low compared with other taxonomic groups. According to Vilà et al. (2010), only 5.6% of non-native terrestrial plants in Europe have ecological impacts, as compared with 30.4% of non-native terrestrial vertebrates. With regard to absolute numbers of species with ecological impacts, nonnative terrestrial invertebrates outnumber non-native plants (Figure 1). These are strong arguments for conducting biological invasion research that is less taxonomically biased. If results from one taxonomic group could simply be extrapolated to another group, then the best strategy would be to focus research on a taxonomic group where it is most costeffective. This extrapolation is not possible, however, because of taxonomic differences in (1) the impacts WRITE BACK WRITE BACK WRITE BACK