Eight years of BioInvasions Records : patterns and trends in alien and cryptogenic species records

“ BioInvasions Records ” (BIR) is an international journal founded in 2011, with its primary focus the publication of new records of non-native species. We analyzed all published articles in BIR between 2012 and 2019, aiming to: make all georeferenced records openly available; investigate spatio-temporal patterns in reported records, methodologies for species identification, and pathways of invasion; and identify possible biases in reporting alien species occurrences and distributions. In total, 10457 georeferenced records were retrieved from 467 published articles, reporting 628 different species. Terrestrial species were under-represented in the dataset. Chordata dominated in the list of reported species, followed by Arthropoda, Mollusca, and Tracheophyta. Europe was the continent with most recorded species, followed by North America. In terms of species reported by country, USA ranked on top. This geographic bias is in accordance with global patterns of research output, related to the fact that North America and Western Europe are leaders in funding research and development, and this is where the majority of highly ranked universities are situated. The country diversity of reported species exhibited an increasing trend from 28 countries in 2012 to 49 countries in 2019. Single-author papers represented only ~ 5% of all published papers, and the median number of authors has increased from 3 in 2012–2013, to 4 in 2015–2019, following global trends of increased collaborations. The frequency of conducting molecular analyses for species identification has increased from 4.5% of published articles in 2012 to 25.2% in 2019, and is expected to further increase with the continuing development of molecular tools, in particular rapid advances and cost reduction in eDNA, next-generation sequencing, barcoding and metabarcoding analyses. The most common pathway of introduction (based on the CBD classification) was “transport-stowaway”, followed by “escape from confinement” and “corridor”. However, the importance of pathways significantly differed by environment. “Transport-stowaway” was the most important pathway for marine and transitional species, whereas “escape from confinement” was the most important pathway for terrestrial and freshwater species. The most important CBD pathway subcategory was “ship/boat ballast water”, followed by “interconnected waterways/basins/seas”, “natural dispersal across borders”, “ship/boat hull fouling”, “aquaculture/mariculture”, and “pet/aquarium/terrarium species (including live food for such species)”. BIR has provided the means for publishing valuable information on the distribution of alien species, the dynamics of invasions, and pathways of introduction, therefore substantially supporting invasion science and management.


Introduction
Humans have transported (intentionally or unintentionally) many thousands of species beyond their natural ranges, reshaping global biogeography (Turbelin et al. 2017) and breaking down biogeographic barriers (Capinha et al. 2015). The impacts of biological invasions are well documented in the scientific literature. Invasive species constitute one of the main drivers of global change and biodiversity loss (Simberloff et al. 2013;Pyšek et al. 2020), often compromise the flow of ecosystem services (Katsanevakis et al. 2014;Vilà and Hulme 2017;Castro-Díez et al. 2019), and can have substantial negative impacts on human health (Mazza et al. 2014;Pyšek et al. 2020). On the other hand, some alien species can be beneficial for biodiversity and ecosystem services (Katsanevakis et al. 2014;Vimercati et al. 2020) and can be even considered as important contributors in reaching conservation goals by securing ecosystem functioning (Mačić et al. 2018;Katsanevakis et al. 2020a).
Globalization and the intensification of trade and transport have accelerated biological invasions on all continents with no sign of saturation (Seebens et al. 2017). It is estimated that alien species will keep accumulating in the following decades (Seebens et al. 2021), and their future impacts are expected to increase in most socioecological contexts mostly driven by transport, climate change and socio-economic change . Hence, it is of upmost importance to support biological invasions science with relevant datasets of presence, distribution, pathways of introduction, and impacts of alien species. To gain a better understanding of ecosystem functioning, and for the effective management and conservation of ecosystems in an era of global change and cumulative anthropogenic pressures (Gissi et al. 2021), there is a need for adequate spatiotemporal knowledge of biota at relevant scales. In particular, such datasets are needed to support the achievement of global commitments of the signatories of the Convention on Biological Diversity (CBD 2010) and the United Nation's Sustainable Development Goals (UN 2015) to prevent the introduction and significantly reduce the impacts of invasive alien species. Management of invasive species should be evidence-based (Sutherland et al. 2004), and thus access to high quality datasets is essential.
In recent years, a remarkable increase in the availability of biodiversity data has been observed (La Salle et al. 2016). In particular, global data on alien species and their distribution have greatly improved over the last couple of decades through a multitude of targeted projects, comprehensive accounts, citizen science initiatives, and online databases (e.g. Katsanevakis et al. 2015;Fuller and Neilson 2015;Pagad et al. 2018;Giovos et al. 2019;Ahyong et al. 2020). Nevertheless, comparisons of online databases has revealed low level of overlap, indicating that there is an amount of data not fully exploited (Gatto et al. 2013;Seebens et al. 2020). For many taxonomic groups, data on alien species distributions remain limited (Thakur et al. 2019), and regional biases in terms of data availability exist (Pyšek et al. 2008;McGeoch et al. 2010). Often information on alien species distribution remains buried in the personal files of researchers or in scattered repositories (Katsanevakis et al. 2020b). Such information is vital to assess the role of alien species in the ongoing changes of biodiversity patterns, their invasion progress, impacts and temporal dynamics, and to develop species distribution models to forecast their present and future distributions. Hence, publication of alien species records and distributions is vital to support invasion science and management.
Journals publishing first records of alien species, checklists, or extensive datasets of alien species records, often amended with information on native ranges, pathways of introduction, and impacts, have been invaluable in feeding large databases (e.g. Trombetti et al. 2013;Katsanevakis et al. 2020b), and thus substantially contribute in risk analyses and management of biological invasions (Lucy et al. 2016). "BioInvasions Records" (hereafter BIR) is a primary example of this journal type, with its focus on the "publication of accounts of new records of both aquatic and terrestrial non-native species and other applied papers in the area of biological invasions" (Lucy and Panov 2012). BIR has aimed to "provide solutions to the growing issues of (1) environmental and biodiversity-related data sharing (Costello 2009), and (2) the support of online databases of invasive alien species with geo-referenced species records data" (Lucy and Panov 2012).
We herein reviewed all published articles in BIR between 2012 and 2019, aiming to (1) compile all published georeferenced species records in BIR and make them openly available in a single database; (2) investigate spatial patterns and temporal trends in reported records, methodologies for species identification, reported pathways of invasion; and (3) identify possible geographical biases, methodological gaps, or other obstacles in the smooth flow of information of alien species occurrences and distributions.

Materials and methods
We reviewed all published articles in BIR between April 2012 (first issue) and December 2019. For each article, the following information was retrieved: (1) bibliographic information such as DOI, short reference (author, year), year of publication, volume, issue, pages, number of authors, number of countries in affiliations; (2) environment (terrestrial, freshwater, marine, transitional); (3) continent (Europe, Asia, Africa, North America, South America, Australia, Antarctica); (4) country; (5) marine biogeographic realm (only for marine species) according to Spalding et al. (2007) (Arctic, Temperate Northern Atlantic, Temperate Northern Pacific, Tropical Atlantic, Western Indo-Pacific, Central Indo-Pacific, Eastern Indo-Pacific, Tropical Eastern Pacific, Temperate South America, Temperate Southern Africa, Temperate Australasia, Southern Ocean); (6) terrestrial biogeographic realms (for terrestrial, freshwater, and transitional species) in line with Olson et al. (2001) (Neartic, Neotropic, Palearctic, Afrotropic, Indo-Malay, Australasia, Antarctic, Oceania); (7) number of alien species reported; (8) reported species; (9) Phylum; (10) approach for taxonomic identification (only morphological, only molecular analysis, combination of morphological and molecular analyses); (11) if an introduction pathway is suggested (yes/no); (12) main introduction pathway suggested, classified according to the CBD main categories, i.e. "release in nature" ("species intentionally transported and released in the (semi)natural environment with little to no dedicated anthropogenic assistance" post-release), "escape from confinement" ("species that have escaped from the confined or controlled environments where they were kept and cared for a number of purposes, e.g. provide food, resources, services or companionship"), "transport-contaminant" ("species introduced unintentionally or accidentally through the movement of other organisms or organic materials and products"), "transportstowaway" ("species introduced into natural environments as accidental stowaways or hitchhikers on a variety of vectors"), "corridor" ("species spreading to new regions along artificially created infrastructure corridors such as bridges, tunnels, canals"), and "unaided" ("species that spread to new regions by natural dispersal, without action or assistance by humans, from regions in which they are alien and were introduced by one of the other introduction pathways") (Harrower et al. 2017); (13) main introduction pathway suggested, classified according to the CBD subcategories (e.g. "ship/boat hull fouling", "hitchhikers on ship/boat", "contaminant on animals", "parasites on animals", "horticulture", "food contaminant" etc; see Harrower et al. 2017, for a full account and detailed definitions); (14) first record in country (yes/no); (15) first record in continent (yes/no).
All georeferenced records reported in the articles were compiled in an additional database and were mapped with the GIS software ArcMap v10.7.1. In addition to descriptive statistics to present the review results, cross-tabulation analysis was conducted to quantify the degree of association between selected pairs of variables, using the software Statgraphics Centurion XVI. The hypothesis of independence between such pairs was tested with chi-square tests.

Results
Overall, data were retrieved from 467 published articles, reporting, in total, 10457 georeferenced records (Supplementary material Table S1) of 628 different species (854 species reports in total), of which 275 were freshwater, 268 marine, 66 terrestrial, and 19 were reported from transitional environments (estuaries, lagoons) ( Figure 1A). There were no articles reporting terrestrial species in 2012 and 2013 but, subsequently, terrestrial submissions increased reaching 26 articles in 2019 (24.5% of published articles) ( Figure 2). Chordata (including mainly vertebrates but also ascidians) dominated in the list of reported species, followed by Arthropoda, Mollusca,  and Tracheophyta ( Figure 1B). Europe was the continent with most reported species, followed by North America and Asia, whereas Australia was the least represented continent, excluding Antarctica, from which there were no records ( Figure 1C). Temperate Northern Atlantic was the marine biogeographic realm mostly represented in marine records, and Palearctic was the terrestrial biogeographic realm mostly represented in both freshwater and terrestrial records. The distribution of records was highly patchy, with vast areas lacking any record, in particular in Asia, Africa and Australia ( Figure 3). Among the species reported, 28.8% were first country records, and 3.9% first continent records.
In terms of species reported in each country, USA ranked on top followed by Ukraine, Canada, and China ( Figure 1D). Some countries ranked high due to a small number of papers reporting national or subnational checklists, e.g. Schofield and Akins (2019) for USA, Semenchenko et al. (2016) for Ukraine, and Xiong et al. (2018) for China, or large-scale studies, e.g. Gartner et al. (2016) for Canada. Reports of alien species from 105 different countries have been published in BIR between 2012 and 2019. Temporal differences in reporting were observed in certain countries, e.g. only one alien species was reported from Libya (which is considered a datapoor country) before 2017 by Milazzo et al. (2012) without the participation of any Libyan authors, while between 2017-2019 fifteen species were reported from that country in six different publications all led by Libyan scientists (Shakman et al. 2017;Rizgalla et al. 2018Rizgalla et al. , 2019a Figure 5A). Single-author papers represented only 5.4% of all published papers. In all years, most articles had 3 or 4 authors but gradually the relative number of multi-authored articles with 5 or more authors has substantially increased ( Figure 5A). In 68.1% of the articles, author(s) were affiliated with organizations of a single country, in 22.3% of cases with organizations of two countries, and in much less cases with three (6.9%) or four countries (2.4%) ( Figure 5B). There were only two articles with affiliations from 5 (Zhulidov et al. 2018) and 6 (Naser et al. 2012) countries.
In 81.5% of studies, species identification was based solely on morphological characters ( Figure 6). In the rest of the studies molecular analyses were conducted either in combination with morphological identification (16.3%) or alone (2.1%). The frequency of conducting molecular analyses for species identification has increased from 4.5% in 2012 to 25.2% in 2019 ( Figure 6). The identified species among all 14 published studies based solely on molecular analyses included species difficult to be identified morphologically or to be distinguished from similar congeneric species. Such species were fish parasites (Chiary et al. 2014    In the 467 reviewed articles, 854 species reports were made, and for 694 a specific pathway was linked to their introduction. The most common pathway (based on the CBD classification) was "transport-stowaway", followed by "escape from confinement" and "corridor" ( Figure 7A). However, the importance of pathways significantly differed by environment (chi-square test, p < 0.001; the same if transitional environment is excluded from the analysis due to low number of records) ( Figure 7B). "Transport-stowaway" was the most important pathway for marine and transitional species, whereas "escape from confinement" was the most important pathway for terrestrial and freshwater species. "Release in nature" was much more important for freshwater species than for the other environments, and "transport-contaminant" for terrestrial species than for marine and freshwater. "Corridor" was the second most important pathway for aquatic species but was not reported for any terrestrial species. The importance of pathways also significantly differed by taxonomic group (chi-square test, p < 0.001; the same if only Chordata, Arthropoda and Mollusca are kept in the analysis) ( Figure 7C). "Escape from confinement" was the most important pathway for Tracheophyta and Chordata, whereas "transport-stowaway" was the most important pathway for Arthropoda, Mollusca, Cnidaria, Annelida and Bryozoa.
The most important CBD pathway subcategory was "ship/boat ballast water", followed by "interconnected waterways/basins/seas", "natural dispersal across borders", "ship/boat hull fouling", "aquaculture/mariculture", and "pet/aquarium/terrarium species (including live food for such species)" (Figure 8). The importance of each pathway subcategory substantially varied by environment. "Shipping/ballast water" was the most frequent pathway for marine species, followed by "shipping/fouling" and "interconnected waterways/basins/seas", whereas for freshwater species "natural dispersal across borders" was the most common pathway, followed by "interconnected waterways/basins/seas", "ship/boat ballast water", and "aquaculture/mariculture" (Figure 8). Terrestrial species were most frequently introduced through "ornamental purpose other than horticulture", followed by "hitchhikers on ship/boat (excluding ballast water and hull fouling)", "food contaminant (including of live food)", "contaminant on plants (except parasites, species transported by host/vector)", and "horticulture".

Discussion
Europe was the continent with most reported species, followed by North America, and the "Temperate Northern Atlantic" biogeographic realm had by far the most reported marine species. This geographic pattern is common in global reviews or systematic mapping of ecological studies, and has been highlighted by many recent articles (e.g. Mačić et al. 2018;Kytinou et al. 2020;Gissi et al. 2021), including reviews on biological invasions (Pyšek et al. 2008). Globally, the majority of research has been conducted in North America and Europe (e.g. Badenhorst et al. 2016), which can be partly explained by the fact that North America and Western Europe are leaders in funding research and development (UNESCO Institute for Statistics 2020), and this is where the majority of highly ranked universities are situated (Jöns and Hoyler 2013). National spending on research and development, the number of universities in a country, GDP, and English proficiency have been found by various studies to positively correlate with research output (Man et al. 2004;Meo et al. 2013;Jamjoom and Jamjoom 2016;Mueller 2016). Hence, the observed geographic pattern in the distribution of published species records in BIR, should not be perceived as representative of the actual distribution patterns of alien species but it is largely an artifact of global trends in research output. It has to be noted that a similar geographic pattern as the one observed herein, was reported by an analysis of the Global Invasive Species Database and the CABI Invasive Species Compendium (Fig. 1A in Turbelin et al. 2017), with the exception of Australia which had high numbers of invasive alien species according to the latter study but is under-represented in BIR.
Records of aquatic species dominated in BIR's articles. This is contrary to global inventories of alien species, in which terrestrial species, in particular plants and arthropods, constitute the richest groups (Turbelin et al. 2017). For example, in the European Alien Species Information Network (EASIN; Katsanevakis et al. 2015;EASIN 2020), among the 13001 alien and cryptogenic species listed, 11048 (85%) are terrestrial, of which 5950 are Tracheophyta and 3361 are Arthropods. The dominance of terrestrial plants and arthropods has been observed since Elton (1958), who commented that invasive alien species are primarily attributed to the transport of plants and related insect hitchhikers. This low representation of terrestrial species in BIR is probably partly related to the history of the journal, which emerged as a continuation of the former "Aquatic Invasions Records", an electronic supplement of the open access journal "Aquatic Invasions", a journal with clearly aquatic scope. Moreover, there are many other international and national outlets traditionally used for publishing records of alien terrestrial plants ( ). Nevertheless, the increasing trend of terrestrial contributions to BIR will probably lead to a more balanced reporting of alien taxa in the future.
We observed an increasing trend in the number of authors per article. Single-author articles represented only 5% of the published articles, which is quite similar to the results of MacNeil (2019) who assessed the trends in the number of authors of published articles in the journals BioInvasions Records, Aquatic Invasions and Management of Biological Invasions over a range of years until September 2018, and led him to the conclusion that we are probably witnessing "the imminent extinction of that increasingly rare species, the single-author". Similar were the conclusions by Barlow et al. (2018) who examined the trends in the number of authors in the Journal of Applied Ecology, and noted that single-author papers accounted for > 60% of the journal's output in the 1960s but < 4% in the last decade. Such trends had been identified since 1963 by Price (1963) who stated that "the proportion of multi-author papers has accelerated steadily and powerfully, and it is now so large that if it continues at the present rate, by 1980 the single-author paper will be extinct". The rise of research networks and collaborations is a potent aspect of globalization of science (Leydesdorff and Wagner 2008;Gui et al. 2019), and has been reflected in the increasing number of authors in scientific articles in most scientific disciplines (Adams 2012). Such trends observed in BIR, as in other journals and disciplines, are probably related to increasing considerations of larger spatial scales and larger datasets, the growing need of multiple skills (e.g. combining field work, ecological knowledge, taxonomic expertise, and the use of molecular tools), increased internationalization through multinational networks and projects, research funding considerations often encouraging collaborations among different institutes and countries (e.g. the funding schemes of the European Commission), broader contributor recognition and more inclusive author lists often including data collectors and citizen scientists (Barlow et al. 2018;MacNeil 2019).
We found an increasing use of molecular tools for species identification, in particular in cases of high morphological similarities and lack of unique diagnostic morphological characters among related species. It is expected that with the further development of molecular tools, their use in biodiversity studies will increase in the near future (Porter and Hajibabaei 2018). In particular, rapid advances and cost reduction in eDNA, nextgeneration sequencing, barcoding and metabarcoding analyses are promising for early detection of new invasions, overcoming difficulties in traditional morphological approaches for species identification, allowing identification when damaged specimens may not contain the characters needed for identification or when the life stages collected lack such characters, and allowing fast and large-scale biomonitoring of multiple alien species as part of whole communities (metabarcoding), based on water, biofilm or soil samples (Jerde et al. 2011;Muha et al. 2017). In addition to species identification, molecular approaches allow invasion biologists to better understand the mechanisms and dynamics of an invasions by identifying the origin and invasion route, assessing effective population sizes, estimating levels of genetic differentiation, and assessing hybridization with native species (Pauls et al. 2014 and references therein).
Our analysis highlighted the differing importance of the various pathways of introduction amongst environments and taxonomic groups. We are aware that due to the geographic and environmental nonrepresentativeness of the present dataset (as previously discussed) biases may be introduced in the assessment of the importance of pathways. Nevertheless, this is a large dataset (N = 694), which we believe is of high relevance. All global databases suffer to some extent from geographic biases due to country development status, varying funding and sampling effort (McGeoch et al. 2010), and global research trends (as previously discussed). The ranking of the main pathways herein reported agrees with many other studies based on different datasets. For example, for marine species "transport-stowaway" (i.e. shipping) has been highlighted as the most important pathway in both global (Molnar et al. 2008) and regional analyses Pergl et al. 2020). "Corridor" was the second most important pathway, as also found for Europe  due to the impact of the Suez Canal, although in the global review by Molnar et al. (2008) it ranks third after aquaculture. For freshwater species, the European review by Nunes et al. (2015), based on the assessed pathways of 550 species, also concluded that "escape from confinement" was the most important pathway. For terrestrial species, "escape from confinement" and "transport-contaminant" were also the two most important pathways highlighted by Pergl et al. (2020) (for plants and terrestrial invertebrates respectively) based on a subset of the EASIN database (298 plants and 1345 terrestrial invertebrates). Nevertheless, it has to be highlighted that in most cases there was no direct evidence for a documented pathway (e.g. a species actually found in ballast waters) but the pathway was rather based upon speculation and correlation (e.g. a marine species initially found near a major port with no other potentially accountable human activities in the wider area). Apart from intentional introductions, in most other cases more than one possible pathway can be inferred based on the human activities in the locality of the first records (Katsanevakis and Moustakas 2018).
In conclusion, BIR substantially contributed in making open access important information on the distribution of alien species, the dynamics of invasions, and pathways of introduction. Such information can feed global databases and support invasions science and management. Nevertheless, we identified geographic and taxonomic biases, similar to existing global databases, which affect global assessments of biological invasions. We also highlighted important trends in increased research collaborations and the use of molecular tools, which can further promote biomonitoring and allow large-scale and low-cost assessments and reporting of the distribution and dynamics of biological invasions in the future. The present global analysis of pathways, despite the low representation of terrestrial species, highlighted the main pathways and thus contributes to the global efforts to prioritize management efforts for reducing the rates of new species introductions.