**4. Discussion**

Until fairly recently, Greece was a country that had relatively few alien fish species compared to other European states [23]. However, nowadays, whole riverine fish assemblages are rapidly changing as non-indigenous fish species spread into several lotic and lentic ecosystems [24]. This study aimed to identify the potential overlapping areas of NIFS against fish species that are under the highest extinction threat category (CR) in Greece. The findings of this study demonstrate that many freshwater ecosystems in the country are under significant invasion pressure and concurrently exhibit high biodiversity conservation value; this problem is now pressing throughout the European Mediterranean basins [25,26].

Predictive patterns of NIFS richness varied largely by utilizing different data sources (i.e., bibliographic data for lentic and field surveys for lotic). These immense differences in NIFS predicted richness between lotic and lentic ecosystems can be attributed to several factors. For instance, bibliographic data usually overestimate the total species richness of an ecosystem, since species that were once reported could now have been extirpated. This is actually true for aliens in several Greek lentic systems, which were once stocked with fishes that were unable to develop viable populations (i.e., *Acipenser baeri*, *Acipenser gueldenstaedtii*, *Ctenopharyngodon idella*, *Hypophthalmichthys nobilis*, *Oncorhynchus kisutch, Oncorhynchus mykiss*) [17]. Thus, these species will be included in checklists and eventually in future analyses despite the fact that they could be currently extirpated. In this study, our combined analysis is indicative of the situation at the country-wide level, as far as possible, in all inland freshwater ecosystems. Correspondingly, results of how bibliographical data may depict overestimated NIFS richness in Greece were reported by a recent study [22], indicating that alien fish species were relatively restricted in Greek rivers, based on sampling data in contrast with bibliographic data.

Similar to other recent studies, the most common NIFS within the overlapping areas were Greece's four most widespread and abundant aliens *G. holbrooki, C. gibelio, L. gibbosus*, and *P. parva* as well as the translocated *C. carpio*, *Salmo farioides*, and *Economidichthys pygmaeus* [14,22]. Evidently, the most recorded NIFS display limnophylic (lacustrine) life history strategies. Indeed, apart from the introductions of cold-water species such as salmonids which are primarily conducted in the upper catchments of rivers and streams, most species introductions in Greece concern warm-water lacustrine species in lowland rivers and lentic ecosystems [27]. In most sites, NIFS richness comprised single species; however, in a single case NIFS richness raised up to 12 fish species (Lake Pamvotis), indicating high invasive pressure. Historically, Lake Pamvotis included only four native species; however, during the 1930s until the late 1990s several NIFS were introduced for purposes of eutrophication control or fisheries enhancement [24]. Overall, lowland riverine areas and lentic ecosystems with well-developed commercial and recreational fisheries indicated higher NIFS richness in comparison with small streams in higher altitudes or in arid regions (e.g., southern Greece and the Aegean islands).

According to the latest Red List assessment at the European level, more than 37% of the freshwater fish species are considered as threatened; 15% as Vulnerable, 10% as Endangered, and 12% as CR status [11]. The IUCN analysis shows that Greece hosts the most species under a threatened status and the most critically endangered freshwater fish species in Europe [28]. Despite this fact, few conservation actions have been applied to protect the CR freshwater fish species of Greece, while how NIFS affect their viability is almost totally neglected. According to current distributional records, two CR species (*P. epiroticus* and *C. stephanidisi*) co-occurring in areas with moderate to high NIFS richness are considered on the brink of extinction [17], while an additional species (*Alosa vistonica)* which indicated low to moderate overlapping areas with the distributions of NIFS has recently been assessed as extinct [29].

Despite the fact that our analysis utilizes broad-scale data of NIFS and CR fish species inhabiting both lentic and lotic ecosystems, we acknowledge its limitations by not incorporating the connectivity of aquatic ecosystems into the approach [30,31]. Generally,

geostatistical techniques are suitable for modeling features that are inherently continuous [32], which contrasts with the fragmentation observed in freshwater ecosystems and the geographic isolation of the islands. However, in an attempt to mitigate the latter issues we used an extensive sampling/site network, thus narrowing the number of maximum neighbors and minimizing the distance points used to perform the interpolation. Furthermore, we acknowledge that some formerly widespread species, with marginal distributions in the country, are now probably totally extirpated as wild native populations from Greece. For instance, regions which include wild populations of sturgeons (*Huso huso*, *Acipenser sturio*, *Acipener naccarri*, *Acipenser stellatus*) [17] may have an overestimated distributional coverage bias of CR fish species.

Even though there has been a rapid growth in developing protected areas worldwide [33], such efforts are not usually optimally designed for freshwater biodiversity. Many conservationists accept that freshwater conservation must relate to a separate ecological realm, beyond the terrestrial or marine, whose specific recognition may have important consequences for both biodiversity conservation and the wider water management issues [34]. A recent study that aimed to overlap the ranges of threatened species, based on the IUCN Red List and Natura 2000 delineations in Greece [35], reported that species belonging to the class of Actinopterygii (including exclusively freshwater species) were covered fairly well within protected areas. However, the aforementioned study also included freshwater fish species within the threatened categories of Endangered and Vulnerable, thus this increase in spatial overlap with protected areas of the Natura 2000 sites is expected, since more species occupy wider areas. In our study, by focusing on the highest extinction threat category (CR), we revealed that the largest part of the distributional ranges of CR freshwater fish species fall outside protected areas. Finally, it is evident, based on the absence of relevant scientific literature, that conservation actions targeting CR freshwater fish species in Greece have rarely been applied. In fact, out of the 25 CR freshwater fish species, concrete conservation actions have been applied only for four species mainly by conducting conservation translocations (*A. naccarii,* [36]; *Pungitius hellenicus,* [37], *V. letourneuxi*, and *V. robertae*; [38]).

IUCN and national red listing assessment schemes are scheduling to modify a number of listed species categories as new reviews of vulnerability status are revised (i.e., some species that are Critical may be downgraded to Endangered and so forth). Although the new red listing revisions may alter the status of some species, in our opinion, most of the species utilized in our study should remain as important guiding species for conservation planning. The Hellenic Zoological Society's Greek Red List procedure was last reviewed in 2009, lagging far behind many other EU states; efforts to revise it have recently been initiated through funding scheduled from the Hellenic Ministry of Environment and Energy. Recently, a new review of Mediterranean freshwater fish species has also begun by IUCN experts and is expected to be completed in 2021. In the current work, we solely utilized the highest extinction threat level category as a proxy indicator for identifying areas of outstanding conservation interest in a rapid screening process that may be easily repeated and transferable to other states and for future monitoring.

Conservation planning in its entirety is a highly complex process encompassing several steps in order to develop and implement the protection, conservation, and enhancement of natural resources. As freshwater protected area effectiveness is usually challenging [39], especially regarding NIFS threats [40], the need for well-designed protected areas and management plans for freshwater biodiversity is a necessity for critically endangered fishes and the ecosystems that host them in Greece. Unraveling the spatial overlap between NIFS and critically endangered fish species can support the first stages of targeted conservation planning and contribute to preliminary actions to minimize the potential negative effects. Future studies should aim to a) identify and assess broader important fish areas for conservation and b) assess how the combined effects of various stressors (i.e., NIFS, water abstraction, pollution, river fragmentation, etc.) can affect the populations of the threatened freshwater fish species of Greece under a climate change

context. Moreover, there is a need for high-quality comprehensive reviews regarding the ichthyofaunal compositions of Greek lentic ecosystems, which will exclude all species with questionable occurrences and clarify the actual and current status of NIFS introductions. Finally, future work should also address the analysis of species traits of the mixed fish assemblages (native and NIFS) to better understand the functional organization of these novel ecosystems.

**Author Contributions:** Conceptualization, N.K., L.V., and S.Z.; methodology, N.K., L.V., and S.Z.; formal analysis, N.K. and L.V.; investigation and data curation, N.K., S.Z., and L.V.; writing—original draft preparation, N.K., S.Z., and L.V.; writing—review and editing, N.K., S.Z., L.V., and O.-I.K.; visualization, N.K. and L.V.; supervision, O.-I.K. and S.Z.; project administration, O.-I.K., N.K., and S.Z. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research is co-financed by Greece and the European Union (European Social Fund-ESF) through the Operational Programme «Human Resources Development, Education and Lifelong Learning 2014–2020» in the context of the project "Interdisciplinary assessment of non-indigenous fish fauna in the freshwater ecosystems of Greece" (MIS 5047785).

**Institutional Review Board Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

**Conflicts of Interest:** The authors declare no conflict of interest. The founding sponsors had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; and in the decision to publish the results.
