1 Introduction

Among snow-active arthropods (Leinaas 1981a; Vanin and Turchetto 2007; Hågvar 2010), springtails (Hexapoda: Collembola) are particularly active on the snow surface, and are well known for showing large mass occurrences (e.g., 4000 animals per m2 snow surface in coniferous forest in Norway; Hågvar 2000; Leinaas 1981a), also visible with naked eyes by unskilled observers.

In particular, the supranival environment hosts specialized winter-active springtails (Leinaas 1981b,c) — typically living in colonies (Leinaas 1983) — that could use this seasonal habitat not only for feeding (Zettel et al. 2002; Hao et al. 2020), but also for effective dispersal and migration (Hågvar 1995, 2000; Zhang et al. 2017). Using sun position, and probably thanks to their high mobility via jumps on the smooth and moist snow surface (Hågvar 2000), migrations of hundreds of meters per day are possible (Hågvar 1995; Zhang et al. 2017). In many podurids (Poduromorpha; e.g., Hypogastrura socialis, Ceratophysella sigillata), this is possible also thanks to the presence of specialized anal vesicles that allow them to adhere to the substratum after landing (Leinaas 1988; Zernecke 1999). Usually, new colonies are established after migration (Hågvar 2010; Leinaas 1983). In addition, this winter behavior may be an efficient way of exchanging genes between populations that live separated during summer (Leinaas 1981a). In fact, these winter-active species usually live in a patchy and temporary micro-habitat during the snow-free period (Leinaas 1983; Hågvar 2000), as it is known for Hypogastrura socialis, typically inhabiting in summer well-drained heaps of needle litter below spruce trees (Leinaas 1981a). Thus, during winter, snow acts as an ecological corridor. In general, intraspecific aggregations are promoted in springtails by pheromones (Tosi et al 1977; Mertens and Bourgoigne 1977; Krool and Bauer 1987) facilitating: (i) the attraction of the opposite sex, and (ii) the aggregation on feeding resources (Hopkin 1997). For example, Zernecke (1999) observed that H. socialis feeds on spruce pollen.

The peak of activity on snow occurs during mild (i.e., sunny days with partial snow melting, maybe also because they need sun for orientation) and humid weather (i.e., in a dry period, springtails swarm at the first rain), and springtails move down in the subnivean environment, more temperature-buffered (average temperature of ca. 0 °C), when the weather gets colder (Leinaas 1981a; Hågvar 2000). During the day, activity peak occurs in hours when mild temperatures are recorded (Zhang et al. 2014). Activity at low temperatures requires physiological adaptations; for instance, the presence of antifreeze agents in certain Alpine springtails was demonstrated by Zettel (1984).

In Europe, specialized species that are known to be regularly active and migrating on the snow surface are Hypogastrura socialis, H. lapponica, Vertagopus westerlundi, Desoria hiemalis, D. tolya (Hågvar 2010; Leinaas 1981c), and Ceratophysella sigillata (Zettel et al. 2002). Hypogastrura peloponnesica (Danyi 2013) is a recently described species, found swarming on the Menalo Mts. (Greece) among melting snow patches. In China, Zhang et al. (2014) and Hao et al. (2020) documented the activity on the snow surface of three unidentified species belonging to the genus Desoria.

If, for Northern Europe, detailed ecological studies have already been published (e.g. Leinaas 1981c; Hågvar 2000, 2010), the knowledge of snow-active species in the European Alps is still incomplete (Thaler 1999): for Northern European Alps, only some ecophysiological studies have been performed on few documented snow-active springtails (Zettel and Zettel 1994; Zettel et al. 2002) and no studies are available for Southern European Alps. Our expectation is that snow habitats in Southern European Alps would host the same species that inhabit the snow habitat of Northern Alps and Northern Europe, since those species are widespread in Europe (e.g., Desoria hiemalis, Hypogastrura socialis; Ceratophysella sigillata; Potapov 2001; Thibaud et al. 2004).

The aim of this work is to provide a first insight about snow-active springtails in a protected area of the Southern European Alps, the Adamello-Brenta Nature Park.

We selected this area because: (i) it represents a peripheral mountain range of relevant biogeographic interest given the a large amount of endemic species of plants and arthropods (Gobbi et al. 2021; Rota et al. 2022); (ii) the Brenta mountain group belongs to the Dolomites chain, and thus, it is an UNESCO World Heritage Site (https://whc.unesco.org/en/list/1237/, accessed on 16 June 2023); (iii) it is deeply affected, as the rest of the Southern European Alps, by the ongoing climate change in terms of glacier disappearance and snow cover reduction (Edwards et al. 2007; Žebre et al. 2021; Marta et al. 2023).

2 Material and methods

2.1 Study area

The study was performed in the Adamello-Brenta Nature Park (“PNAB”; Central-Eastern Italian Alps), the largest (620 km2) protected area in Trentino-Alto Adige region with an altitude that spans between 477 and 3.558 m a.s.l. (Fig. 1). The park exhibits a great environmental heterogeneity — including habitats from the valley floor (e.g., broadleaved and coniferous forests, seminatural and natural grasslands) to glacial environment — also because of the occurrence of two distinct lithological, and thus geomorphological, areas: the granitic massif of Adamello-Presanella and the limestone massif of Brenta Dolomites, separated by the Rendena Valley (46°07′07.1"N 10°45′10.7"E).

Fig. 1
figure 1

Position of THE Adamello-Brenta Nature Park on European Alps (A) and sites where springtail populations have been sampled (B); Legend for population codes: see Table 1

Full size image

2.2 Sampling activity

Sampling was performed in the period December 2019 – April 2021, by collecting springtail bloomings all the year (Table 1). The term “blooming” refers to two different situations: “dense” blooming with at least 20 cm2 of the surface completely covered by springtails, and “sparse” blooming in which springtails extend for hundreds of meters, but they do not cover completely the surface. The sampling activity was performed by an environmental assistant of the Park (M.Z.) during faunistic transects aimed at monitoring the presence and distribution of target species of vertebrates and of red wood ant nests (Formica rufa) between 800 and 2100 m a.s.l.; for each springtail “blooming” found on snow, a variable number of individuals (range: 30–200) were collected and preserved in 96% ethanol. In addition, two other springtail assemblages were found on the bare ground (i.e., not on snow) and on water surface, and they were also sampled as a comparison.

Table 1 Species found and information on micro-habitat (snow, water and ground) and type of blooming (sparse/dense: see “sampling activity” chapter). Station data and sampling data are here reported, and also air temperature at the sampling hour, obtained from weather station data corrected with the thermal gradient (see “sampling activity” chapter)
Full size table

Videos and photos in macro-mode were taken by M.Z. during the sampling procedures to obtain original documentation about morphology and behavior. Multimedia material was taken with two cameras: Nikon Coolpix P900 and Nikon D5500, both potentiated with a Ryanox macroscopic lens (Model M-250) for macro videos and photos (Appendix Fig. 5, Supplementary material 1).

All sampling sites and all populations found are reported in Table 1 and in Fig. 1, while in Appendix Fig. 5, pictures of the sampled species and their habitat are reported.

Air temperature of each sampling site at the sampling hour has been obtained by applying a thermal gradient of 0.6 °C/100 m to meteorological data (meteotrentino.it) of the two main areas in which populations have been found: Molveno weather station (835 m a.s.l.) for population “MOL” and “HMA”, occurring in the Valley of Molveno lake, and Giustino weather station (877 m a.s.l.) for the other ones, all belonging to the Rendena Valley. Since weather data have been collected every 15 min, sampling hours were rounded accordingly.

2.3 Specimens’ preparation, morphological identification and preservation

For slide preparation, springtails were passed in boiling alcohol for removing fats and cleared by short immersion in 10% KOH solution (1–5 min). Then, after a Chlorophenol bath, they were mounted on slides using Swann medium as a preservative solution (Rusek 1975). Morphological observations and pictures were made with a Leica DM2500 with phase and DIC contrasts and drawing arm and Carl Zeiss Axiolab 5 with phase-contrast light microscopes. Identification has been performed according to Thibaud et al. (2004), Potapov (2001), Skarżyński and Smolis (2003), Danyi (2013). Specimens preserved in alcohol and permanent slides are preserved at the MUSE-Science Museum of Trento (Italy).

For comparison, we extracted DNA and sequenced the barcode fragment (5’-end of the COI mitochondrial gene) from 5 specimens of H. kelmendica Peja, 1985 (population “HKE”, sampled the 13/02/2022 in Poland, near Olsztyn, Kraków-Częstochowa Upland; 50°44′55.5"N 19°16′34.6"E, 300 m asl. Leg. D.S.).

2.4 Molecular data

According to Potapov et al. (2020), both morphological and genetic analyses of integrative taxonomy were performed, to check cryptic diversity. Genomic DNA was extracted from five specimens (whole animal) from each population (with the exception of HBR, for which only one sequence is available), using the Wizard®SV Genomic DNA Purification System (Promega, Madison, WI, USA); DNA was eluted in 50 μl of H2O. A fragment of the mitochondrial cytochrome c oxidase subunit 1 (cox1) was amplified with the primers 5′-GGTCAACAAATCATAAAGATATTGG-3′ (LCO1490) and 5′-TAAACTTCAGGGTGACCAAAAAATCA-3′ (HCO2198) (Folmer et al. 1994). PCR reactions were prepared in a 25 μL volume containing: 2.5 μL of both forward and reverse primers (10 μM), 16 μl of H2O, 4 μl of 1:10 diluted DNA template and lyophilised PCR beads (illustraTM PuReTaq RTG PCR-Cytiva). Amplifications were run with the following conditions: an initial denaturation of 5 min at 95 ℃, followed by 34 cycles of denaturation (30 s at 95 ℃), annealing (30 s at 52 ℃) and extension (40 s at 72 ℃). A final extension of 10 min at 72 ℃ was added. For the HBR population the PCR product had to be reamplified (with the same conditions) to obtain enough material for sequencing. PCR products were sequenced on both strands using 3730xl DNA Analyzer by Macrogen Inc. Sequences were assembled in Geneious 8.1.9 and aligned, then manually corrected, and aligned using BIOEDIT version 7.0.5.3 (Hall 1999).

2.5 Genetic analysis

The sequences produced for this work were analyzed with those from the same species (when available) and other species from the same genera (GeneBank accessions numbers and BoldDatabase sequence ID are reported in Supplementary material 2).

Distance analyses were performed with MEGA7 (Kumar et al. 2016), using a Neighbor-Joining (Saitou and Nei 1987) algorithm with the Kimura-2 parameter model (K2P − Kimura 1980) to estimate genetic distances. The robustness of nodes was evaluated through bootstrap re-analysis of 1000 pseudoreplicates. The trees were replotted using the R package ggtree (Xu et al. 2022).

3 Results

3.1 Biodiversity of snow-dwelling and “blooming” springtails

Overall, nine assemblages of springtails were found, with eight of which monospecific and one comprising two species (Table. 1, Fig. 1, Appendix Fig. 5). In total, five species were found: Hypogastrura cf. kelmendica sp1, Hypogastrura cf. kelmendica sp2, Hypogastrura cf. peloponnesica Danyi 2013, Desoria hiemalis (Schött, 1893) (Table 1, Appendix Fig. 5) and Ceratophysella sigillata (Uzel, 1891). Desoria hiemalis and Hypogastrura cf. kelmendica sp2 have been found only on snow, while the others were also collected in other microhabitats. Specifically, Hypogastrura cf. kelmendica sp1 and Ceratophysella sigillata were found also on the water surface and C. sigillata also on ground.

Among the five species found, two were already described in literature: Desoria hiemalis and Ceratophysella sigillata. Two species are likely new for science and morphologically close to H. kelmendica: Hypogastrura cf. kelmendica sp1 and Hypogastrura cf. kelmendica sp2. The last one, H. cf. peloponnesica, morphologically corresponds to the H. peloponnesica recently described from Greece and only known from the type locality. All specimens of Desoria hiemalis exhibit the typical “winter form” (sensu Potapov 2001). The most frequent species is C. sigillata, found in three populations, two of them temporally and spatially very close to each other (CME, CML; Table 1). All the other species, except H. cf. peloponnesica found in a single site, have been found in two populations.

3.2 Genetic analysis

In the three genera analyzed, a clear barcode gap and a clear differentiation was found between intra and interspecific distances (GenBank accession numbers in Appendix Fig. 6).

In Ceratophysella (Table 2, Fig. 2), among the species from the armata group — excluding the C. sigillata from PNAB and Germany — mean intraspecific distance is 0.38% (range: 0.06–1.08%) and mean interspecific distance is 24.47% (range: 21.25–27.87%). The PNAB/German C. sigillata cluster exhibits an intraspecific divergence of 0.33% (with 2.34% divergence between PNAB and the German sequences) and a mean interspecific divergence with all the other species included in the analysis of 22.93% (range: 18.95–28.24%). In Hypogastrura (Table 2, Fig. 3), mean intraspecific distance is 0.44% (0–2.42%) and mean interspecific distance is 21.14%. Without H. cf. kelmendica sp1 and H. cf. kelmendica sp2., mean interspecific distance is 21.34% (range: 17.67–26.14%) and without H. cf. peloponnesica is 21.15%. These two potential new species close to H. kelmendica, exhibit a mean interspecific distance of 15.53% and respectively a mean intraspecific divergence of 0.34% and 0.41%. Their respective mean divergences with the H. kelmendica cluster is 13.08% and 15.20%. In Desoria (Table 2, Fig. 4), without the D. hiemalis cluster, mean intraspecific distance is 0.15% (range: 0–0.46%) and mean interspecific distance is 22.13% (range: 19.26–25.38%). Within D. hiemalis, two clusters are identified (A Norway/German pre-Alps’ cluster and a PNAB one), with a distance of 8.29% (Fig. 4, Table 2). The mean intercluster distance between these two D. hiemalis clusters was 8.29%. Intraspecific variability inside D. hiemalis PNAB cluster is 1.64% (Table 2).

Table 2 Mean intra and interspecific K2P distances in Ceratophysella, Hypogastrura, and Desoria. PNAB = Adamello-Brenta Nature Park
Full size table
Fig. 2
figure 2

Neighbor joining tree (K2P) based on the COI 5’ sequences of (a) specimens of C. sigillata collected in PNAB / Germany along with those from (b) six other Ceratophysella species from the armata group. The upper and lower sides of the triangle represent respectively the maximum and minimum of genetic distances within the species. Nodes supported by ≥ 99% bootstrap values are marked with an asterisk

Full size image
Fig. 3
figure 3

Neighbor joining tree (K2P) based on the COI 5’ sequences of (a) specimens of H. cf. kelmendica (sp1 and sp2) and H cf. peloponnesica collected in PNAB along with those from (b) six other Hypogastrura species (comprising sequences from specimens formally identified as H. kelmendica). The upper and lower sides of the triangle represent respectively the maximum and minimum of genetic distances within the species. Nodes supported by ≥ 99% bootstrap values are marked with an asterisk

Full size image
Fig. 4
figure 4

Neighbor joining tree (K2P) based on the COI sequences of (a) specimens of D. hiemalis collected in PNAB and Norway / German pre-Alps along with those from (b) five other Desoria species. The upper and lower sides of the triangle represent respectively the maximum and minimum genetic distances within the species. Nodes supported by ≥ 99% bootstrap values are marked with an asterisk

Full size image

3.3 Temporal distribution of specimens along the sampling period

Specimens were found in January, April, May, June, November and December (Fig. 4). The longest period in which no swarming springtails could be found was from June 2020 to November 2020 (about 5 months, mainly corresponding to the snow-free period). Ceratophysella sigillata and Hypogastrura cf. kelmendica sp2 were found only during spring (respectively, in April and June the first one and in May the other ones), Desoria hiemalis and Hypogastrura cf. peloponnesica only during winter (respectively, in January and December), while Hypogastrura cf. kelmendica sp1 in autumn and winter (November and December) (Fig. 4). In relation to daytime, of the five populations sampled in spring (HRE, HAR, CCE, CME and CML), four populations were found before 10:30 AM and one at 15:30 (HAR). Otherwise, the four populations sampled in colder seasons, autumn and winter (HBR, MOL, HMA and HTV-DTV), were found after 11:30 AM (Table 1). According to the weather stations used in this study, all springtail “bloomings” were found when temperature was higher than 0 °C (Fig. 4).

4 Discussion

4.1 Snow-swarming biodiversity still to be described

Our data suggest that the snow-active springtail fauna found in the Adamello-Brenta Natural Park is partially different from that of Northern Europe (Hågvar 2010; Leinaas 1981c). Specifically, more than a half of the snow-active collembolan species in the study area are likely new for science (e.g., H. cf. kelmendica sp1 and H. cf. kelmendica sp2) or only recently described (H. cf. peloponnesica; Danyi 2013) (see following paragraph). The remaining taxa, Desoria hiemalis and Ceratophysella sigillata, are widespread species. Desoria hiemalis is a nival species from Central and Northern Europe, already found in the European Alps (Block and Zettel 1980), but this is the first report for Italy. In the European part of Russia, D. hiemalis is “a very common xeromesophilic species”, found in coniferous forests (Potapov 2001), which is compatible with our samples collected in a spruce and beech forest. Ceratophysella sigillata is the only springtail from temperate habitat known to have its main reproductive season in winter and to spend the warm season in dormancy, as reported also from Northern European Alps (Zettel and Zettel 1994; Jureková, et al 2021). We documented that this cold-loving species is active in the Southern Alps in mild spring on snow, as observed in Crete, Greece (Schulz 2010). Ceratophysella sigillata was mainly known from the Northern European Alps (Palissa 1964; Block and Zettel 2003). According to Dallai and Malatesta (1982), C. sigillata was recorded in Italy only in caves, with the exception of two records, from the Dolomites (Marcuzzi 1983a) and from the Apennines on Gran Sasso (Marcuzzi 1983b). Thus, the finding of both species in the Adamello-Brenta Nature Park represents an important record for the Alpine collembolan fauna. Despite the expectations, Hypogastrura socialis was not found. However, two of the three Hypogastrura species found in the Adamello-Brenta Nature Park belong to the Hypogastrura socialisgroup (see following paragraph).

4.2 Morphological and genetic diversity of snow-active springtails in PNAB

In terms of biodiversity, the most interesting group of snow-active springtails found was represented by the genus Hypogastrura, for which we found three different species. Hypogastrura cf. kelmendica sp1 and sp2 are morphologically very close to Hypogastrura kelmendica Peja, 1985 described from Central and Southeast Europe and belonging to the Hypogastrura socialisgroup. They differ from H. kelmendica and from each other in their body and antennae chaetotaxy and tegumentary granulation. In particular, Hypogastrura cf. kelmendica sp1 is morphologically similar to Hypogastrura socialis found in large swarms in Mount Amiata (Central Italy; Dallai and Ferrari 1970) and re-attributed by Skarżyński and Smolis (2003) to the H. kelmendica complex.

The genetic divergence between H. kelmendica and H. cf. kelmendica sp1 and sp2 is lower than the interspecific divergence observed between other well-defined Hypogastrura species (albeit higher than the intraspecific divergence in those species). This lower genetic divergence could be due to a recent — or even ongoing (and therefore incomplete) — speciation (Porco et al. 2018). Their specific status could be further investigated with nuclear gene sequencing; meanwhile, it is important to flag them as distinct genetic entities.

Globally, both putative species exhibit a clear morphological and genetic differentiation, suggesting that they could be considered as two independent species. These forms, which are thus likely species new to science, will be formally described in separate taxonomic studies. In general, species of the Hypogastrura socialisgroup seems to include a great diversity of species, with a weak and inconstant morphological diversification; thus, as shown in this study, the use of integrative taxonomy for approaching snow-active Hypogastrura species would be highly advisable.

Concerning H. cf. peloponnesica, despite a correspondence with the diagnosis, we suspect a possible differentiation from the nominal species (a snow-active springtail from Peloponnese, Greece) because of clear differences in the shape of setae and in the body and leg chaetotaxy. Unfortunately, it was not possible to perform a genetic comparison on the original material that could have brought further support to this hypothesis. More investigations are needed to reach conclusions on this topic.

Regarding the Ceratophysella cluster, C. sigillata sequences from PNAB do not cluster with C. pseudarmata sequences from Canadian specimens, i.e., they were found as genetically divergent as the other clusters from the other Hypogastrura species analyzed. This could further support the synonymisation of C. sigillata from North American populations with C. pseudarmata (Babenko et al. 2019).

Regarding the genus Desoria, the distances between the sequences obtained from populations of D. hiemalis from PNAB and from Germany and Norway lay within the known intraspecific range found among populations of a same species in Collembola (e.g., Porco et al. 2014; Schneider et al. 2016). The mean genetic distances found between this cluster (Germany/Norway) and the sequences produced from PNAB populations of D. hiemalis ranged between the mean intraspecific and interspecific distances found among other Desoria species. Nevertheless, in other Collembola groups, interspecific distances were previously found to range from 8% (between well-defined Hypogastrura species—Hogg and Hebert 2004) and up to 11.71% (among Micranurida morphospecies — Porco et al. 2014). Moreover, in this study, the genetic distances found for Desoria species other than D. hiemalis comprised only one population each, thus potentially not reporting the full actual range of the intraspecific variation within species of the genus. From the morphological point of view, no differences was found regarding diagnostic characteristic of the species, but further investigation should be done.

This finding calls for further investigations with nuclear genes and/or more abundant samplings for COI (e.g., Porco et al. 2018). This could help elucidating the nature of the high intraspecific divergence found between PNAB and Norway/Germany populations in D. hiemalis, and deciding if this divergence is at species level or if it is the result of an incomplete ongoing speciation process.

4.3 Snow as a fragmentary habitat and European Alps as biogeographic barrier

On a broader perspective, we could hypothesize that snow, especially in peripheral mountain chains like the Adamello-Brenta mountain group, acts as a fragmentary ephemeral habitat for those cold-adapted springtails that are known to be present in Southern Europe, but mainly active in winter or in cold and stable microhabitats like caves, as suggested by Cassagnau (1973) for H. socialis and C. sigillata. Indeed, Arbea and Pérez Fernández (2020) recently reported the first occurrence of H. socialis for the Iberian Peninsula, in a cave habitat. Also at smaller scale, Jureková et al. (2021) and Raschmanová et al. (2018a) found that C. sigillata prefers cold micro-habitats in Slovak karstic areas. Likewise other species that have a high cold tolerance, this species also shows a remarkable heat tolerance (Raschmanová et al. 2018b) that could enhance its migration capability and explain its wide distribution range in Europe. As suggested by Thaler (1999) for Alpine nival fauna in general, springtails too could have sustained a biogeographic differentiation on Alps; for example, the genetic divergence found between PNAB and German/Norway Desoria hiemalis could support the hypothesis of such a differentiation between Northern and Southern Alps. In addition, the co-presence of the two closely related species of the H. kelmendica complex (H. socialisgroup), probably new for sciences, indicates a local differentiation of Northern species. Based on our data and on morphological descriptions of historical records of H. socialis on European Alps (e.g., Handschin 1924, where H. socialis has only 6–7 sens on antennomerous IV rather than 10–12), we hypothesize that this group also includes several other species still unknown closely related to H. socialis on Alpine mountain chains. In general, European Alps could act as a biogeographic barrier for some nival species between Northern and Southern areas and Southern European Alps seem to constitute a peculiar biogeographic region for snow-active springtails.

However, population genetic studies would be necessary for assessing gene flow and migration dynamics.

4.4 Conservation issues of snow habitat

It is already known from the literature that ground-dwelling springtails have a general decrease in activity during the warmest and driest seasons (Badejo and van Straalen 1993). Populations collected in the Adamello-Brenta Nature Park confirm this observation and suggest that summer is not a good “blooming” period for springtails; however, we cannot exclude that the lack of “blooming” occurrence in the summer is due to an increased difficulty in detecting springtails when snow is not present. Concerning sampling time, it was observed that, while in spring springtails swarm earlier in the morning, in colder season, springtails swarm usually in later (and milder) hours, confirming the expectations (Zhang et al. 2014). According to this, springtail “bloomings” were always found in Adamello-Brenta Nature Park when the air temperature was above 0 °C.

From the ecological point of view, we consider interesting the hypothesis that snow cover could act as an ecological corridor able to connect, in temperate and cold-temperate regions, populations of cold-adapted springtails that are spatially separated during the snow-free period. It, therefore, favors gregarism and easier dispersal ability for the individuals in a bidimensional space (rather than tridimensional as in the soil). For this reason, it is important to know how many species use snow as habitat for foraging, reproduction, and for connecting populations.

In addition, the use of the snow surface as habitat and/or ecological corridor could be relevant in conservation biology in the context of the ongoing climate changes. Several arthropods, specifically those active during the winter, seem to be negatively affected by the reduction of the duration of the snow cover (see Slatyer et al. 2017; Templer et al. 2012). As reported in the literature, springtail abundance decreases during the snow-free period and, more generally, it is negatively affected by soil drought (Badejo and van Straalen 1993). Changes in the frequency and magnitude of snow cover are significantly impacting the Alpine ecosystems (Marta et al. 2023) that rely on snowmelt to satisfy their water demands. A recent research by Colombo et al. (2023) clearly demonstrated unprecedented snow-drought conditions in the Italian Alps during the early 2020s which is part of a recent pattern of increased intensity and frequency of snow-drought events since the 1990s. Soil arthropods are negatively impacted by extreme warming events (Harvey et al. 2023), also during the cold season (Bokhorst et al. 2012). Thus, we can expect, for the future, a significant reduction of the role of snow as an ecological corridor, potentially bringing the populations to a higher risk of spatial separation, that could lead to a higher extinction risk.

Interestingly, no cryophilic species (i.e., linked to ice) was found during our samplings. Cryophilic Alpine springtails are among the most spatially fragmented and threatened fauna (Valle et al. 2021). Thus, whether the snow could act as an ecological corridor for glacial organisms remains to be tested.

5 Conclusions

Snow is an important ephemeral habitat for springtails, not only for the biodiversity hosted and its ecological importance for cold-adapted springtails. The Southern European Alps could be an important area of diversification for snow-active species, probably for the effect of Alps as a biogeographic barrier.

In practical terms, snow also facilitates the detection and sampling of springtails by untrained persons, whose activity might be useful for finding new species, as it happened in our study where samplings were performed during monitoring activities. This makes the collection of snow-active springtails an ideal topic for citizen science projects. On the other hand, despite the relative simplicity in sampling springtail “blooming” on snow, this biodiversity is still understudied and poorly described. The description of this biodiversity is fundamental for future ecological and zoogeographical research, as they strictly depend on the progress of the taxonomic knowledge (Thaler 1999): without precise taxonomic information, it is not possible even a proper evaluation of fieldwork.