This survey was developed within the scope of the European Cooperation in Science & Technology (COST) Action CA17122 – “Increasing understanding of alien species through citizen science (Alien-CSI)”, which includes participants from all EU Member States and a few neighboring countries. This COST Action sets out six research coordination objectives, to be first approached through a European wide analysis of existing IAS citizen science initiatives (Roy et al. 2018).

The first version of the survey was tested, revised and validated in a COST Action workshop in Akrotiri, Cyprus, 25 – 28 February 2019. Representatives from 25 countries in the COST Action attended. The survey (Price-Jones et al. 2021) was shared as a Google Form with all COST Action participants, and disseminated online. Responses were collected from June 27, 2019 to April 6, 2020. For each country, existing citizen science projects involving alien species were contacted and requested to complete the survey. All projects are/were active in EU Member States and/or neighbouring countries. A list of projects was compiled, including from a web search and previously available lists of European citizen science projects (e.g., EASIN, Kus Veenvliet et al. 2019), and the missing projects in the survey database were contacted. Finally, to increase reach, the survey was also disseminated through the European Citizen Science Association (ECSA) newsletter and mailing list and respondents were asked to share it with colleagues and local networks via snowball sampling.

Survey questions and attribute values were developed using JRC metadata standards for citizen science projects (European Commission, Directorate-General for Environment 2018) and the project metadata model of PPSR Core, a set of global, transdisciplinary data and metadata standards for Public Participation in Scientific Research (PPSR Core). The survey included 62 questions (Price-Jones et al. 2021), in nine sections: 1) Contact information of the respondent (a project coordinator); 2) General characterization of the project, including a brief summary, geographical scope, time scale, hosting entities, funding, etc.; 3) Information on project scope, including target audience, taxonomic and environmental scope, project aims, type of data collected, etc.; 4) Policy-related information, i.e., policy relevance and inclusion of species listed in the EU IAS Regulation; 5) Information on engagement, such as type of involvement of the general public in the design of the project, engagement methods and social media used, skills needed to participate and frequency of contributions; 6) Information on feedback and support provided to participants by the project, e.g., if the project provides materials for species identification, guidelines, training activities, information on how data from the project are used, feedback mechanisms and support; 7) Data quality and data management, namely validation mechanism for records, registration type, methods of recording, whether data are open and accessible to the general public, data form used to store data, data standards and data licence used, whether a public data management plan was in place; 8) Performance indicators of projects, namely usage of smartphone applications, number of participants and number of records, whether learning is assessed, number and type of publications using data from the project; and 9) Notes and remarks.

Only projects that simultaneously fulfilled the following criteria were included in the analyses: 1) a clearly citizen science-focused project; 2) alien species included in the main scope; and 3) projects developed in Europe (even if not exclusively). As such, national biodiversity networks and portals collecting data on all species were only considered if they had a clear alien species focus. Projects needed to have specific forms of public engagement related to alien species, so projects solely devoted to improving IAS policies but without a typical citizen science component (e.g., data collection using target groups, interaction with volunteers) were not considered. However, projects where data gathering was less relevant, but which had clear educational and outreach goals on IAS, were included.

Due to response rates below 100% for particular questions and the prevalence of responses “Unknown” or “Not applicable”, the number of projects that provided a definite response was determined and used for calculations of percentages for each question.

General characterisation of the project

In total, 129 projects/initiatives completed data for the survey and, of these, 103 fitted the criteria for inclusion and were considered for analysis. Of the 26 that were excluded, 17 were not alien species-focused, seven had no specific forms of public engagement on alien species and two were duplicate entries.

The number of new projects has increased over the past fifteen years with the oldest project recorded beginning in 2005 (Brown et al. 2008) while 21 began in 2019 (Fig. 1). Most projects (76 of 103 projects) are still ongoing. A total of 42 countries were represented in the survey. A majority of projects (66%, 68 of the 103 respondents to this question) were run at the national level, and 85% (87/103) were active in a single country. However, one project, a survey of alien species of Union Concern on iNaturalist, was active in 38 countries. In four countries (Estonia, Malta, Montenegro, and North Macedonia) this represented the sole project.

Figure 1.

Number of new citizen science projects per year on alien species in Europe according to responses to the survey.

The type of organisation responsible for the projects varied between governmental (29%, 30/103) and non-governmental organisations (22%, 23/103), universities (28%, 29/103), public research organisations (22%, 23/103), and private companies, non-profit organisations and individual persons (12%, 12/103). Most projects were fully (54%, 56/103) or partially (19%, 20/103) funded, but 26% (27/103) reported having no funding. Governments were the largest source of funding, although only 36% of projects (28/78) report governments as being their sole source of funding. Otherwise, funding was provided by public entities, the EU LIFE program, NGOs or private sources, or a combination of the above.

Project scope

Plants were the most common target taxonomic group (30%, 31/103; Fig. 2a), the general public was the most common target audience (89%, 92/103; Fig. 2b), and terrestrial habitats the most common environment considered in the projects (57%, 59/103). There was no association between target audience and taxon (p = 0.2779), but an association was observed between target audience and environment (p = 0.0049). Two trends in the data included the prevalence of terrestrial projects aimed at land managers, and freshwater and marine projects aimed at fishers. The marine environment was also the environment type most frequently involving scientists and students.

Figure 2.

Percentage (indicated by numbers on radar plots) of projects that gave selected responses to project scope questions: a target taxon b target audience, and c stated project aim.

84% of projects (87/103) focused solely on alien species and 9% (9/103) focused partially on alien species; 7% (7/103) responded that alien species were not the main focus, yet alien species data were collected and received some emphasis. Most projects had multiple aims, the most common being mapping of alien species distribution (Fig. 2c). Most projects also collected more than one type of data, with species presence and/or abundance being the most common.

Project performance

The usage of applications, number of participants, number of records and number of publications all show a distribution of responses that peaked in lower numbers and fell off quickly at higher numbers (Fig. 3), but these values were often unknown or not applicable. Only 33% of projects (21/63) assessed learning of the participants. A similar number of projects produced scientific peer-reviewed publications (94%, 33/37) and science communication publications aimed at the general public (85%, 30/37). Most of these publications directly presented data from the project (47%, 24/51) or were descriptive in nature (43%, 22/51).

Figure 3.

Survey responses concerning project performance, participation and publication. a number of participants (n = 72) b number of records (n = 73) c number of publications (n = 62) and d percentage of usage of App (n = 17).

According to responses to our survey, the UK had more alien species citizen science projects (21) than any other country, followed by Italy (13), Portugal (9) and France (9) (Fig. 4a). However, when project counts per region are normalised by population, Northern Europe has the highest ratio, followed by the UK and ROI, Southern Europe, Eastern Europe and Western Europe (Fig. 4b). There was no association between geographic region and the target audience (p = 0.51), taxon (p = 0.41) or environment (p = 0.16).

Figure 4.

Geographic distribution of projects: a number of projects by country and b number of projects per million inhabitants by region.

Project duration had a significant, positive impact on the three performance indicators tested, i.e., number of participants (z = 2.78, df = 1, p = 0.0054), publications (z = 3.38, df = 1, p = 0.00073) and records (z = 3.01, df = 1, p = 0.0026). Projects that provided a map also outperformed projects that did not in number of participants (z = 2.13, df = 2, p = 0.033), publications (z = 2.77, df = 2, p = 0.0056) and records (z = 2.84, df = 2, p = 0.0045).

Provision of training was positively related to the number of publications (z = 2.85, df = 1, p = 0.044), as was use of social media (z = 2.35, df = 1, p = 0.019) and provision of guidelines (z = 2.01, df = 1, p = 0.045). Projects that required advanced prior knowledge resulted in more publications than projects that required limited (z = -2.80, df = 2, p = 0.0052) or no (z = -2.74, df = 2, p = 0.0061) prior knowledge. The same result was seen in terms of number of records, with projects that required advanced prior knowledge performing better than projects that required limited (z = -2.47, df = 2, p = 0.014) or no (z = -2.02, df = 2, p = 0.043) prior knowledge.

Provision of feedback positively impacted the number of publications (z = 2.01, df = 1, p = 0.044) but negatively impacted the number of records (z = -2.01, df = 1, p = 0.044). Provision of support negatively impacted the number of publications (z = -2.59, f = 1, p = 0.0096).

The dominance of national projects in our results is consistent with that observed for other research on management of biological invasions (Hulme et al. 2008). Several factors contribute to this tendency, including nationally-derived funding, differing degrees to which countries are exposed to or aware of alien species, species alien in one country being native to another, logistical convenience (Hulme et al. 2008) and uneven distribution of expertise (Hulme et al. 2009). However, international coordination is necessary to better protect native ecosystems from IAS (Perrings et al. 2010; Katsanevakis et al. 2013), as reflected by international agreements, from the Convention on Biological Diversity (United Nations 1992) to EU Regulation 1143/2014. There was a small degree of international coordination evident in our dataset, with most multinational initiatives being active in six or fewer geographically clustered countries.

Most projects target the general public, which is logical given our inclusion criteria. This strategy aligns with the philosophy of informing (Genovesi et al. 2015), engaging with and inspiring a passion for nature in as many participants as possible (Roy et al. 2015). Plants and insects are the most common target-taxa of projects, possibly because both are broad and speciose groups and are among the taxa containing the most invasive species with recorded ecological and/or economic impacts (Vilà et al. 2010; Haubrock et al. 2021). Furthermore, they can be easily accessible, with many urban species. The under-representation of groups such as birds is notable, and is likely a result of the concentration of data specific to these groups on large crowd-sourcing platforms that did not fit our criteria.

The prevalence of projects in the terrestrial environment similarly reflects convenience for the public, as reported for other citizen science projects (Aceves-Bueno et al. 2017). It is also highlighted by the association we found between target audience and target environment.

The most common aim is mapping of alien species, and participants are often asked to submit species presence and/or abundance data. Species presence is easy to observe, report and validate (Hyder et al. 2015), and works well in conjunction with mapping (e.g., Malek et al. 2018; Dissanayake et al. 2019; Kumar et al. 2019). Nonetheless, presence-only data have limitations: for example, the lack of absence reporting and the assumption that species were not present because they were not observed (Johnson et al. 2020).

The region with the most recorded projects is the United Kingdom and the Republic of Ireland, reflecting a long history of citizen science in ecology (Silvertown 2009). After this region, more projects in Western and Southern Europe may reflect a higher level of IAS awareness in these regions due to a relatively higher number of funded IAS projects (e.g., LIFE projects in Italy). On the other hand, a low number of projects, e.g., in the Netherlands and Belgium, may be explained by the presence of single, dominant national biodiversity portals not being included in our survey. For example, in Belgium, biodiversity recording is dominated by the general reporting portal www.waarnemingen.be which has a dedicated app and upon which early warning tools for IAS are built (Vanderhoeven et al. 2015; Swinnen et al. 2018). Additionally, there is a possible language bias, if projects from non-English speaking countries were not reached or not motivated to participate in the survey, which was only available in English.

Studies evaluating data quality and management in citizen science projects sometimes have contradictory conclusions (Crall et al. 2011). Overall, volunteer contributions have been regarded favourably by scientists, e.g., 73% of papers positively described in analysis by Aceves-Bueno et al. (2017). Various tangible benefits have been noted, such as increases in the predicted spatial distribution of IAS by models trained with data from citizen science (César de Sá et al. 2019). However, Aceves-Bueno et al. (2017) also concluded that differences between volunteer data and professional data were significant in 38.4% of projects. In addition, some projects use their data for removal or management of regulated IAS, and as such correct species identification is of utmost importance. The most prominent approaches to validate citizen science data are peer and expert validation, often aided by automatic filtering techniques (Balázs et al. 2021). This is clearly also the case for IAS citizen science in Europe (Adriaens et al. 2021), with our results showing that most projects use expert validation.

Data generated by citizen science are often referred to as dark data: unreproducible, becoming more valuable over time, and at high risk of being lost (Costello and Wieczorek 2014). The implementation of a well-defined data management plan (DMP) can be used to prevent such loss of data. Nonetheless, few of the surveyed projects claimed to have a DMP and our survey did not assess adherence to the DMP for projects that had one. Many citizen science projects are relatively small scale and probably lack experience and/or access to tools for data management planning (Schade et al. 2017). Data management planning could improve the accessibility of data, an important component of FAIR (Findability, Accessibility, Interoperability, and Reusability) data management (Wilkinson et al. 2016; Reyserhove et al. 2020).

Opening alien species data is important to unlock their full potential for science, policy and management (Groom et al. 2015, 2017a,b). However, although some projects deposit their data on national or institutional repositories, less than one third make them freely available on an open data repository, e.g., through GBIF publication. We also found that most alien species citizen science projects produced peer-reviewed papers, but these were not necessarily open access; however, most also produced scientific communications aimed at the public. Reasons for avoiding open data may be multiple, including licensing issues, funding limitations, technical barriers or the privacy of the participants (Ganzevoort et al. 2017). Ganzevoort et al. (2017) found that half of the citizen scientists they surveyed believed that data collected by the citizen science organisation was a public good, but only 12% supported unconditional use. The question of data ownership is complex and can be addressed in legal terms by choice of license. We found that 92% of projects that provided licence information had a licence allowing public use. Overall, these parameters around data accessibility are consistent with the findings of Wiggins and Crowston (2011) and Schade and Tsinaraki (2016), indicating a willingness to provide access to data.

We anticipated that higher levels of feedback, support and engagement would improve performance, e.g., the number of participants, records and publications, through the generation of commitment and empowerment. As expected, provision of maps, training and guidelines related positively to one or more of the performance indicators. Unexpectedly, provision of feedback related positively to the number of publications, and negatively to the number of records; provision of support also negatively related to the number of publications. While citizen scientists often claim that receiving feedback is important to their continued participation (Geoghegan et al. 2016; Anđelković et al. 2022), a reduced sample size and the fact that many projects are relatively new (17 started in 2017 and 21 in 2019), may have influenced the results. In addition, the number of publications may be influenced by several other factors such as the publishing dynamic of the project team, or the level of knowledge of the survey respondent on the publications stemming from the project. Also, the many missing values in the survey responses might partly explain some of these results.

Only 39% of the projects were designed collaboratively, thus in most cases citizens were contributing in a predetermined way (usually data collection). Even so, a priori fewer projects were expected to be collaborative (e.g., Pocock et al. 2017 analysed more than 500 ecological and environmental citizen science projects and only 4% were collaborative) and so we suspect that this question may have been misunderstood. We define a collaboratively-designed project as a project with citizen scientist participation in the initial conception of the project and all subsequent steps. However, respondents may have considered other roles, such as feedback from participants on project design, as collaboration.

Surveyed projects mostly required low levels of time commitment for learning and participation, possibly recognising that most citizen scientists are amateur observers (Bonney et al. 2016). However, from the authors’ own experience, even though many projects target the general public, in reality many of the participants do have some level of expertise in the taxonomic group they report. Another unexpected result was that projects with limited or no skill requirements were related with a significantly lower number of records and publications. Possibly, participants with advanced skill levels, having already invested the time in learning, have a stronger commitment to contributing. Nevertheless, encouraging anyone to participate is highly relevant to the goal of reconnecting people with nature (Devictor et al. 2010) and increasing the chances of prevention and early detection of IAS. Furthermore, if a contributor can both learn and teach (e.g., through peer validation), knowledge is transferred without the need for training or prior expertise. It should be noted, however, that the lessons that can be drawn from this result are limited by how the levels ‘none,’ ‘low’ and ‘advanced’ skills/knowledge were not defined in our survey, and so may have been interpreted differently by different respondents.

Although it depends on project goals, besides engaging participants, projects often encourage continued participation (Penner 2002). The positive relation between provision of maps, training and guidelines and some of the performance indicators suggest that these investments may encourage participation. It has been shown that publicly displayed maps allow recognition of citizen scientists’ efforts (Williams and DeSteno 2008; Crowston and Prestopnik 2013).

Finally, the majority of projects used an internet-based engagement method, such as a website or social media, reflecting the ubiquity of these technologies in Europe (Kemp 2021).

Several lessons can be drawn from the results of our survey. First, sustainability of projects is key to their performance in terms of the number of records they gather, participants they involve and publications derived from them. Second, many citizen science projects apparently have not yet opened their data. Open data publication maximises the use of the data in policy processes, such as their use by EASIN in the implementation of the EU IAS Regulation (Schade et al. 2019) and provides better return to citizen scientists on their contribution and value of their data.

One partial solution to openness and data management issues might be the drafting of DMPs, which are missing in many projects, despite these facilitating better storage, maintenance, and use of data. Although small projects may struggle to create their own, they may take advantage of existing plans, and strategies can be designed to make data openly accessible, for example on the platform GBIF. Few respondents provided information about their scientific outputs, and there is often no information on project web pages about where they publish their datasets.

To further improve outreach and onboarding of new citizen scientists, and sustained participation, our results suggest that the provision of maps with sightings and the provision of training are important. Future work could also be undertaken to compare the performance of different validation procedures and provide recommendations to new projects to improve data quality (Probert et al. 2022).

Our results show an increasing number of new alien species citizen science projects in the last few years that contribute to IAS mapping and policy implementation, but some regions still hold untapped potential for new citizen science initiatives related to alien species. Existing projects may be made accessible to new audiences through language translation or simplification, and through tailoring of aims and species lists to geographic regions (e.g., Invasive Alien Species in Europe application; Trichkova et al. 2021).

The UN’s SDGs provide an excellent model for how citizen science can be relevant to setting and achieving goals at a global level. Although SDGs were not initially developed with citizen science in mind, data gathered through citizen science can be used directly for feeding SDG indicators (Fritz et al. 2019; Bishop et al. 2020), can increase the temporal and spatial scale of data collection (Schade et al. 2019) and can engage people with science and the environment (Pocock et al. 2014). Nonetheless, Fraisl et al. (2020) noted poor alignment of citizen science initiatives with target 15.8 on IAS.