First, we conducted an online survey targeting recreational anglers in Iceland to explore their experience with and perceptions of European flounder between October 2019 and June 2020. The anonymous survey, along with information about the aims of the scientific project, was predominantly shared through public Facebook pages of research institutions and a dedicated Icelandic recreational angling Facebook group (15.924 members as of November 2024). We chose Facebook as a tool to reach a wide range of potential participants, as this is a highly popular social media platform in Iceland that around 65% of the population frequently use (Kemp 2024). In November 2019, an article published in a national Icelandic newspaper (Statistics Iceland 2024) that covered the research project (Jónsson 2019) further encouraged participation in the survey. In total, 209 people submitted responses to the survey.

In this survey we asked participants to provide locations where they encountered, either seen and/or caught, European flounder. The occurrence data from the survey was manually reviewed, removing those locations that we could not confidently assign to a specific waterbody. Due to the linguistic characteristics of the place names of Icelandic freshwater systems there are multiple rivers or lakes with the same name that can often only be differentiated when additional geographic indications are provided (e.g. the rivers “Varmá í Mósfellsbæ” and “Varmá í Hveragerði”). For each location provided by survey participants we checked whether at least one of the following conditions applied: 1) it is a recreational angling river/lake with a unique place name not requiring additional geographic information; 2) additional geographic information were provided allowing a clear identification; 3) additional information provided by the participant throughout the survey clearly identify the location such as through information on the region of Iceland they spent most of their time fishing. Locations were removed when none of these conditions were met.

Second, following the concept of iEcology (Jarić et al. 2021), we collected occurrence data from recreational fishermen in Iceland based on conversations within the Facebook group „Veiðidellan er frábær…“ [The fishing passion is great...], a group highly popular among the target group with 15.4 k members as of 02.05.2024. We evaluated different social media sources for their data availability on European flounder in Iceland as well as biodiversity platforms such as iNaturalist. While most platforms at the end of 2022 indicated a low number of data points relevant to this study (for example only three records had been submitted to iNaturalist prior to 2023) Facebook represented a popular tool among the Icelandic public (Kemp 2024) where European flounder has been frequently addressed. We manually identified Facebook posts and threads mentioning European flounder resulting in 50 recovered conversations occurring between 2013 and 2022 (see detailed information on the extraction and reach of the Facebook group in Henke et al. 2024). From those 50 conversations location reports of either catching or observing European flounder were extracted. Within this manuscript, this data is referred to as iEcology.

As neither the online survey nor the locations extracted from Facebook provided specific coordinates, we determined representative locations for each site. For rivers, representative locations were chosen near the lowest part of the river, for lakes near the mouth of the river through which European flounder most likely entered the lake, and for fjords within major estuaries. These locations were chosen under the assumption that they represent the minimum spread of the species. Where specific location names within habitats were provided, such as fishing beats within rivers, we pooled these locations together to create unique location records.

Data on European flounder was extracted from all available data in the MFRI marine database resulting in European flounder occurrences from major annual surveys, such as, the spring groundfish survey (SMB), the autumn groundfish survey (SMH), and the gillnetting survey (SMN). The SMB and SMH are annual trawl surveys that sample widely around Icelandic waters in February-March and September-October respectively (Sólmundsson et al. 2022; Jakobsdóttir et al. 2023). The SMN is a spring gillnetting survey sampling inshore waters around Iceland (Bogason et al. 2024). There were also European flounder occurrences from less regular surveys, such as, a discontinued near-shore beam-trawl survey (2017–2022) (Thorlacius et al. 2024) and a demersal seine survey (1995–2013) (Pálsson and Sólmundsson 2017). In addition to these surveys, European flounder was also occasionally reported in marine research projects (Table 1). Survey- and research-based locations extracted from the MFRI marine database were pooled together under the group “marine surveys”.

Similarly, the freshwater MFRI records comprised European flounder location data from two sources. First, we extracted all locations where European flounder had been caught in freshwater surveys conducted by the institute. The MFRI sampled widely across recreational fishing rivers in Iceland with the main goal of annual salmonid stock assessment, this sampling is conducted at the same time of the year, usually in late summer. Second, we extracted flounder catches documented in logbooks from recreational fishing rivers, a database maintained by the Freshwater Division of the MFRI. Under the Law on salmon and trout fishing (Act 61/2006), fishing associations in Iceland are required to submit catch information on salmonids in their rivers. While catch information was traditionally submitted in the form of physical logbooks, the option of electronic submissions has been available since 2011 and in 2023 the MFRI established an online form allowing anyone to register their catches.

In addition to European flounder occurrence in surveys and logbooks we used data from the “rare fish database” managed within the Demersal Division of the MFRI. The rare fish database is an ongoing project logging reports and catches reported by stakeholders, often fishermen but anyone can report catches. In 2006, European flounder became part of regular surveying in MFRI marine surveys and was no longer reported to the rare fish database (personal communication Klara Jakobsdóttir, MFRI).

Location data from MFRI surveys provided specific coordinates of catches. For freshwater surveys locations within the same river were pooled together to represent unique sites. As the marine habitat cannot be divided by similar geographic boundaries, locations were treated as unique by default and only pooled when the differences in neither longitude nor latitude between two locations were greater than 0.01 (approx. 1.11 km). Logbook locations did not include specific coordinates and were therefore treated like the locations obtained from the online survey and Facebook. A full list of all data used is presented in Suppl. material 1.

Three sites from each quadrant of Iceland (SW, NW, NE and SE) that had been reported in the online survey were selected and subsequently sampled using a beach seine (10 m long, 6 mm mesh) to confirm the presence of European flounder. Local landowners and/or river managers were contacted for recommendations on sampling sites to increase the likelihood of accessing likely areas of European flounder occurrence as well as to ensure safety during the sampling process. Where the safety of the scientists could not be guaranteed due to known, strong currents, we selected an alternative site nearby based on local recommendations. The sampling at these 12 sites took place between July and September 2020. Sites where not at least one European flounder was caught were considered as not validated and therefore excluded from further analysis.

All data handling, statistics and figures were done using R (version 4.3.2, R Development Core Team 2023). First, we tested differences in the geographical representation of data sources by comparing both latitude and longitude between all data sources (marine surveys, freshwater surveys, logbooks, online survey and iEcology) using a pairwise Dunn test with Bonferroni correction implemented in the R package dunn.test (Dinno 2024).

To compare the annual detection of unique sites between data sources with time stamps available (rare fish, marine surveys, freshwater surveys and logbooks), we fitted a generalized additive model (GAM) to the number of unique sites per source using the gam () function of the mgcv package in R (Wood 2017). The statistical family was zero inflated Poisson (-0.593, 1.903) and the link function identity. We fitted the model with year as fixed and a smooth term.

Individual locations per year ~ Year + s (Year, by = Source)

We furthermore examined how well these four sources documented the temporal spread of European flounder in Iceland. As European flounder was first detected in the southwest and southeast of Iceland (Jónsson et al. 2001), we used latitude to approximate the species’ northward spread. We fitted a GAM to the latitude of documented European flounder sites using gam () to investigate for differences in the detection of the species’ spread between the four sources. The statistical family was Gaussian with an identity link function. The model was fitted with year as fixed effect and a smooth term.

Latitude ~ Year + s (Year, by = Source)

For both models the smooth term was Year with source as an interaction term to account for the differences between sources. Smoothing parameters were estimated using restricted maximum likelihood. For the purpose of model diagnostics, we inspected fitted residuals and tested for autocorrelation using the functions “simulate.residuals()” and “testTemporalAutocorrelation()” of the “DHARMa” package (Hartig 2022), respectively.

The results of the online survey targeting recreational fishermen generated 96 unique locations and therefore provided a higher number of occurrence locations than the previously available survey data based on the MFRI databases, which delivered 50 marine, and nine freshwater locations. Efforts to sample European flounder at 12 representative sites around Iceland named in the online survey showed a high validation rate of over 90% as the sampling at 11 out of 12 sites resulted in at least one individual. While the remaining site was consequently omitted from further analysis, the absence of European flounder in the river Hofsá in northeast Iceland is likely explained by the overall scarcity of European flounder in the northeast region of the country as its presence was confirmed in 2023 during monitoring of salmonid stocks in the same area. The knowledge of recreational fishing communities has previously proven valuable in addressing conservation issues in aquatic environments (Giovos et al. 2019; Löki et al. 2023).

Recreational fishers spend substantial time in the aquatic environments gathering experiential knowledge, beyond what is often available to scientists, and can therefore be more likely to encounter rare or new species (Silvano and Begossi 2012; Löki et al. 2023). The quantitative advantage of stakeholder observation-based data we documented in the current study on European flounder is in accordance with the results of these earlier case studies. While scientific surveying can maximize data quality by strategic sampling plans, stakeholder observation-based data are less predictable. In the current study, stakeholders reported unique sites at a lower ratio than scientific surveys, but the overall higher number of locations provided as well as the confirmation of persistent flounder occurrence provided by repeated reporting of some sites, is advantageous to monitoring activities. An additional advantage of stakeholder observation-based approaches is the increased awareness among stakeholders, which is crucial in addressing biological invasions especially at early stages (Dehnen‐Schmutz et al. 2018). The advantages of stakeholder observation-based data were also supported by comparison of the different data sources available within the MFRI. Specifically, large parts of the previously available distribution data on European flounder were already based on stakeholder knowledge, in the form of logbook entries and voluntary submissions to the rare fish database. These submissions were predominantly by recreational anglers and commercial fishermen.

When directly targeting stakeholders or local knowledge holders is not feasible, for example, because of cost, need for prior training etc., iEcology can offer additional approaches to utilize already existing stakeholder observations stored in the form of web-based data such as social media conversations (Jarić et al. 2020a). In the current study the locations scraped from Facebook did not provide the same number of occurrences as the online survey, but it still exceeded the previously documented distribution data based on the freshwater records of the MFRI. Many challenges and limitations have been noted for iEcology approaches, such as, potential spatiotemporal biases and validation of data accuracy (Jarić et al. 2020a; Jarić et al. 2021). However, in the current study 80% of the locations extracted from social media were also named in the online survey, and the approaches produced a comparable distribution pattern, validating iEcology as a tool in the current case study. Overall, in combination with other sources of information and under considerations of potential limitations and biases, social media data can be of great value when addressing conservation issues (Toivonen et al. 2019) as well as recording new species (Cresson et al. 2021).

Despite the apparent advantages of stakeholder observation-based approaches for documenting European flounder occurrence it should be noted that they provided only presence data while data collected as part of scientific surveys most often provides both presence and absence as well as potential information on size, age, diet and environmental variables at the catch site, i.e. salinity, water depth, etc., as well as co-occurring species. These data are all needed to accurately estimate distribution and habitat suitability as well as the ecological impacts of alien species (Robinson et al. 2020). Conversely, involving stakeholder observation-based data in monitoring activities can enable scientists to obtain large amounts of data in a short amount of time, requiring less resources (Cardoso et al. 2017) and it may be particularly suitable to document early stages and spread of invasions.

In addition to documenting occurrences and spread, public and stakeholder observations are often the first records of an alien species (Thomas et al. 2017; Epanchin‐Niell et al. 2021; Kousteni et al. 2022; Pocock et al. 2024). The first official documentation of European flounder in Iceland was based on the submission of a specimen to the MFRI by a member of the public after it was caught at the mouth of the river Ölfusá in southwest Iceland (Jónsson et al. 2001). As shown in this study, stakeholder observation-based sources remain better in reporting European flounder at sites where this species is still rare. Where preventing the arrival of alien species, the most desirable scenario (Browne et al. 2009; Pyšek and Richardson 2010; Schwindt et al. 2023), fails, early detection of alien species becomes crucial (Pyšek and Richardson 2010; Schwindt et al. 2023) but often depends on the premise that the public recognizes and reports unusual observations. Dehnen‐Schmutz et al. (2018) identified raising awareness as one of the main topics that should be prioritized in policies addressing biological invasions.

Following the first detection of an alien species, the experiential knowledge of stakeholders can contribute to the reconstruction of the species’ temporal spread (Latombe et al. 2017). While there was overall no significant difference in the documentation of the European flounders’ northward spread, the voluntary records submitted to the ‘rare fish’ project already indicated a distribution expansion ranging around half of the country in the early 2000s. While the records of European flounder catches in MFRI’s marine surveys generally suggest a similarly fast expansion northward, note that the surveys only officially started recording European flounder after 2006. The recent catch of European flounder during routine freshwater surveying in 2023 provided the first confirmation of the species in the northeast corner of Iceland and the first contribution of formal freshwater surveys to the reconstruction of the temporal spread.

Stakeholder observations-based data highly depends on the willingness of the public and other stakeholders to share their knowledge. The comparably high number of documented locations of European flounder shown in this study indicate that there is quite some willingness among stakeholders, specifically, the recreational fishing community in Iceland to contribute their experiential knowledge for monitoring purposes. Globally, recreational fishermen have been increasingly involved in management and conservation (Granek et al. 2008) and the results of Copeland et al. (2017), suggest that among other incentives there are social factors providing motivation to get involved.

Our data indicates fluctuations in the number of observations submitted to the MFRI by fishermen, which has also been reported by Cresson et al. (2021), who suggest that the reporting of rare species is not only strongly linked to personal motivation and interest of the fisherman but is also likely to decline once the familiarity with this species increases. The reporting of European flounder catches to the MFRI were mostly initiated by stakeholders’ personal motivation although in the years following the first documentation of European flounder news items and reports discussing its status as a new species were relatively common (Henke et al. 2024). The willingness to report could potentially be linked to the general awareness of the public and their interest in the topic of alien species, which for the European flounder in Iceland has been fluctuating over the years (Henke et al. 2024), as well as perceptions of European flounder as novel and “rare” and thereby of interest. In the years following the first official documentation of European flounder in Iceland (Jónsson et al. 2001) when it was still a novelty, occurrences were predominantly recorded as part of the ‘rare fish’ project, but these reports stopped after 2006. The public’s attention towards conservation issues such as biological invasions is generally rather transient (Jarić et al. 2023). The current results suggest that public interest in an alien species is unlikely to be maintained at a level necessary to document spread after initial establishment. However, when stakeholders were approached with an opportunity to share their knowledge and opinions, along with information on the objectives and the anticipated value of their participation, they provided many more locations of European flounder occurrences.

It has been recommended that at a minimum, countries should obtain updated occurrence records of alien species every five years and make these publicly available (Latombe et al. 2017). In this study, we have shown that stakeholder observation-based tools, when designed accordingly, can offer a great contributory source of data where resources for alien species monitoring are limited. Latombe et al. (2017) further suggests that national monitoring of alien species could be improved by building upon already existing structures. In line with previous studies (Jarić et al. 2020a; Löki et al. 2023), we show that there are many ways to collect stakeholder observation-based data, including logbooks, online questionnaires and iEcology. However, there are certain considerations on how to foster a continuous two-way knowledge exchange between stakeholders and scientists to be considered when designing such approaches (Courchamp et al. 2017). Here it can be beneficial to create platforms that are readily accessible and easy to use for the targeted stakeholder group. Recreational anglers in Iceland are required by law to submit catch reports of salmonids (Act 61/2006) and the electronic logbook of the MFRI offers an existing structure to submit catches of Atlantic salmon, brown trout and Arctic charr. A fourth option (“other”) allows the records of other caught species, most applicable to catches of European flounder or pink salmon (Oncorhynchus gorbuscha), an invasive salmonid species that has shown strongly increasing abundances not only in Iceland but many other northern European countries (Lennox et al. 2023). Simply providing options specific to European flounder and pink salmon could raise awareness of this information being both officially recorded and important for management, thereby encouraging anglers to report catches. In turn, the resulting data could greatly improve the monitoring of the overall spread of alien fish species in Iceland and provide indications of abundance at very low cost. However, these considerations regarding accessible platforms for data collections should not end at the public and stakeholder level. It should be noted that while we are confident that the underlying data for this study represents the bulk of the available data on European flounder catches in Iceland, we are aware that other institutions and/or scientists may have collected additional catch data. A holistic database that keeps updated records across institutions would greatly benefit not only the monitoring of alien species in Iceland but also other conservation efforts.

We have shown that stakeholder observations can represent a valuable, complementary source in the monitoring of an alien species. However, considering the context dependency of biological invasions (Robinson et al. 2017; Catford et al. 2022) approaches involving stakeholders need to be carefully designed on a case-by-case basis to appropriately address each invasion individually. In the case of European flounder in Iceland, we selected Facebook over other social media or biodiversity platforms that are more commonly used in similar studies (Pace et al. 2019; Jarić et al. 2021) as Facebook was much more frequently used in this specific case. In addition to choosing the most appropriate data source(s), the reliability of observations regarding the correct species identification needs to be evaluated as this is a crucial aspect of gathering distribution data. In our case, we rated the reliability of stakeholder observations as high based on the fact that European flounder is easily distinguishable from other species in rivers and lakes as it represents the only flatfish species entering these habitats. This was further reinforced by successfully conducted site validation. We recommend that the approach implemented in this study, can be highly valuable and applicable in small nations where resources for formal, scientific monitoring of alien species is limited but collaborations between research institutions, management parties and stakeholder groups are easier to establish.

We acknowledge that, while stakeholder observations are part of Local ecological knowledge (LEK), to holistically incorporate the knowledge of stakeholder groups in the monitoring of alien species, their involvement must be established throughout the entire research process. As the phenomenon of biological invasions is inherently of interdisciplinary and complex nature, embedding LEK in the necessary research and management approaches is becoming increasingly recognized (Caceres‐Escobar et al. 2019; McElwee et al. 2020). While the current study shows the benefits of including stakeholder observations in monitoring it does not fully integrate the available LEK. We recommend broadening the approaches to the monitoring of European flounder in Iceland as well as of other alien species, to include LEK where the context allows, for a more holistic understanding of the alien species and their impacts.

The authors have declared that no competing interests exist.

No ethical statement was reported.

Theresa Henke was funded by Rannsóknasjóður (the Icelandic Research Fund, Grant number 239953-051).

Conceptualization: TH, GÁÓ. Data curation: MT, TH, HB. Formal analysis: GÁÓ, TH. Funding acquisition: TH, GÁÓ. Methodology: GÁÓ, TH. Project administration: TH. Supervision: GÁÓ. Visualization: TH, GÁÓ. Writing – original draft: TH. Writing – review and editing: HB, MT, GÁÓ.

Theresa Henke https://orcid.org/0000-0002-0729-7818

Hlynur Bárðarson https://orcid.org/0000-0001-7336-3919

Magnús Thorlacius https://orcid.org/0000-0002-5050-2880

Guðbjörg Ásta Ólafsdóttir https://orcid.org/0000-0002-2814-9160

All of the data that support the findings of this study are available in the main text or Supplementary Information.