1. Introduction
In the current dramatic scenario of worldwide glacier retreat (Rounce and others, Reference Rounce2023), it remains challenging to assess the effects of this phenomenon on ecosystem functioning and services at a global scale (Velasquez Casallas and others, Reference Velasquez Casallas, Khelidj, Morán-Ordóñez and Losapio2025). As far as we know, the effect of glacier disappearance on the biodiversity linked to them has never been directly documented, although numerous speculations have addressed the impacts of glacier retreat on cold-adapted species (e.g., Kaufmann, Reference Kaufmann2002; Brown and others, Reference Brown, Hannah and Milner2007; Cauvy-Fraunié and Dangles, Reference Cauvy-Fraunié and Dangles2019; Ficetola, Reference Ficetola2021). Microorganisms, plants and animals adapted to cold and humid glacial environments (i.e., cryophilic) are at risk of disappearing along with the habitat to which they are exclusively tied (Cauvy-Fraunié and Dangles, Reference Cauvy-Fraunié and Dangles2019; Gobbi and others, Reference Gobbi2021), and, in general, the knowledge on alpine soil biodiversity is rare on a global scale (Praeg and others, Reference Praeg2025), especially in supraglacial habitat (Valle and others, Reference Valle2025), posing challenges for this kind of monitoring worldwide. From this perspective, peripheral mountain ranges – like the Bergamo Alps – are of particular importance due to their high biodiversity (Medail and Quezel, Reference Medail and Quezel1999), which includes species that are exclusive to this area (“endemic”; Trew and Maclean, Reference Trew and Maclean2021). In fact, during the ice ages, the Bergamo Alps remained partially uncovered by ice caps, allowing the local survival of many plant and animal species and giving rise to a hotspot of mountain biodiversity (Tribsch, Reference Tribsch2004).
The Trobio glacier was the most important and one of the most iconic of the Bergamo Alps (“Alpi Orobie,” Lombardy, Italy), a relatively low (maximum elevation 3052 m a.s.l.) peripheral chain of the southern Alps. The high precipitation rate of this area, compared to the neighboring Alpine sectors (Luino and others, Reference Luino2020), allowed the presence of glaciers even below mountains lower than 2900 m (Scotti and others, Reference Scotti, Brardinoni and Crosta2014).
At the peak of the Little Ice Age (LIA), the Trobio glacier was the largest glacier of Bergamo Alps (1.1 km2) and its front extended down to 2330 m (Caccianiga and others, Reference Caccianiga, Ravazzi and Zubiani1994; Scotti and others, Reference Scotti, Brardinoni and Crosta2014; Fig. 1). Relatively large and easily accessible, it soon became the most famous glacier of the Bergamo Alps (Fig. 1c). Afterwards, the glacier showed a steady decline until its complete disappearance in 2023 (D’Adda and Petrone, Reference D’Adda, Petrone, Chiarle, Bondesan, Carturan and Scotti2024; Figs 1b and 2).
The Trobio glacier (Bergamo Alps, Italy) a) in 1935 and b) in 2022; c) the Trobio glacier (center) at the beginning of the 20th century. Pictures a) and b) were taken from the same viewpoint. Photographs respectively by Giuseppe Nangeroni (Comitato Glaciologico Italiano), Stefano D’Adda (Servizio Glaciologico Lombardo), Alfredo Corti (Archivio Corti, Italian Alpine Club, sez. Valtellinese).
(a) The Trobio glacier (Bergamo Alps, Italy) in 17 September 2022; (b) The Trobio glacier in 10 August 2023 soon before the complete vanishing of the last ice patches (the red circle indicates two sampling persons as a scale). The ice patch subdivided in two parts on the bottom of (a) is the same reported in (b), where the two parts are completely separated. Photograph respectively by Stefano D’Adda and Marco Caccianiga.
In front of the Trobio glacier tongue, Nebria tresignore (Szallies and Huber, 2014) (Coleoptera: Carabidae), a stenoendemic ground-dwelling, surface-runner (wingless) ground beetle was observed (Tampucci and others, Reference Tampucci2015; Gobbi and others, Reference Gobbi, Tampucci and Caccianiga2018a; Bonelli and others, Reference Bonelli2025). This species is closely linked to glacial environments like glacier surface, proximity of glacier tongue, glacierets, of the Bergamo Alps (Gobbi and others, Reference Gobbi, Tampucci and Caccianiga2018a; Bonelli and others, Reference Bonelli2025). The springtail Desoria orobica Valle, 2025 (Collembola: Isotomidae) was discovered in 2020 and recently described for the Trobio glacier (Valle and others, Reference Valle2025), but its exact distribution is still unknown (it is probably an endemic Alpine species; Valle and others, unpublished). Both organisms are cryophilic species at extremely high risk of extinction as they are strictly linked to the microhabitat conditions ensured by the presence of permanent ice and long-lasting snow cover. No data are available on Nebria tresignore feeding strategy and diet; it is likely an olfactory predator, similarly to the other alpine species belonging to the same genus (Wheater, Reference Wheater1989; Kielty and others, Reference Kielty, Allen-Williams, Underwood and Eastwood1996), feeding on small arthropods like aphids, spiders, flies and springtails (Sint and others, Reference Sint, Kaufmann, Mayer and Traugott2018), thus probably also on Desoria orobica. The specific diet of Desoria orobica is still unknown, but it possibly consists of dead organic matter and biofilms, similarly to other springtails species of glacial and proglacial habitat (Hågvar and Pedersen, Reference Hågvar and Pedersen2015).
The aims of this work are (1) reporting the glaciological trends of Trobio glacier from the LIA till its disappearance and (2) providing data about the monitoring of the abundances and distribution of the two iconic endemic indicator species, Nebria tresignore and Desoria orobica, in relation to the glacier retreat.
2. Materials and methods
2.1. Historical glaciological data
To reconstruct the past extent and dynamics of the Trobio glacier, we consulted a variety of historical sources, including cartographic records, written descriptions and photographic evidence. Early qualitative descriptions, such as those by Maironi Da Ponte (Reference Maironi Da Ponte1808) and Ricci (Reference Ricci1911), provided firsthand accounts of the glacier’s characteristics and movement. Historical maps, including the 1833 Austrian military map (Caccianiga and others, Reference Caccianiga, Ravazzi and Zubiani1994; Bonardi and others, Reference Bonardi, Rovelli, Scotti, Toffaletti, Urso and Villa2012), offered valuable cartographic representations of its extent during the LIA.
Additionally, scientific literature detailing glaciological surveys (e.g., Caccianiga and others, Reference Caccianiga, Ravazzi and Zubiani1994; Scotti and others, Reference Scotti, Brardinoni and Crosta2014) and inventories such as those by Porro (Reference Porro1925) and Nangeroni (Reference Nangeroni1932) and C.N.R. – Comitato Glaciologico Italiano (1961) documented the glacier’s fluctuations over time. Photographic records, including images by Alfredo Corti, further complemented our analysis by providing visual evidence of the glacier’s evolution throughout the 20th century.
This comprehensive review of historical sources associated with a field-based and remote sensing geomorphological interpretation of glacier deposits and trim-lines enabled us to reconstruct the LIA glacier’s maximum surface extent. By doing this, we were able to compare past glacier extents with recent field observations, facilitating a long-term assessment of the Trobio glacier retreat and disappearance.
2.2. Post-LIA glacier area change
In order to quantify post-LIA glacier change we reconstructed the glacier’s extent in five different time steps. Glacier’s extents in 1954, 1989 and 2003 are upgrades from Scotti and others (Reference Scotti, Brardinoni and Crosta2014). The glacier’s limits in 1954 have been formerly stereographically interpreted on paper copies of black and white aerial photographs and slightly corrected by terrestrial historical images and field surveys. Limits from 1989 rely on the 1992 Lombard glacier inventory (Galluccio and Catasta, Reference Galluccio and Catasta1992) checked on 26 September 1988 b/w orthophoto (Volo Italia 1988–89); 2003 (0.5 m grid) orthophoto mosaic and 21 August 2022 Google Earth® images have been used for the more recent time-steps, while in the year 2023 the complete glacier extinction was witnessed directly in the field.
2.3. Biodiversity monitoring
To assess the presence and distribution of the two cryophilic species Nebria tresignore and Desoria orobica, we conducted surveys during summers 2023 and 2024 verifying previous records from the years 2014 (sampling date 13 September) and 2016 (28 August) for Nebria tresignore (Gobbi and others, Reference Gobbi, Tampucci and Caccianiga2018a) and from the year 2020 (sampling date 18 July) for Desoria orobica (Valle and others, Reference Valle2025) to evaluate potential shifts in species distribution and/or abundance over time. This comparison allowed us to assess population trends and habitat stability after the glacier disappearance. We also added new sampling points (A1-A2-B1-B2) where the last ice patches occurred before complete disappearance. We searched for the two species where the most favorable ecological conditions could be found, i.e. on the last ice patches (in 2023), on snowfields, and in their immediate vicinity. Observations were systematically recorded, with standardized qualitative (for Nebria tresignore) and quantitative (for Desoria orobica) sampling methods, ensuring consistency in methodology, sampling time and/or period and site conditions for a temporal comparison. Nebria tresignore sampling was performed by hand catching performed in nine sampling points of about 25 m2, for 20 minutes by one experienced operator (Gobbi and others, Reference Gobbi, Barragán, Brambilla, Moreno, Pruna and Moret2018b) during the summers 2023 (10 August), 2024 (26 July and 28–29 August) (Fig. 4). Data on abundance of Nebria tresignore were not used for the analysis due to the different duration of the sampling performed in 2014 (1 h – Tampucci and others, Reference Tampucci2015) with respect to those of the following sampling sessions (20 minutes), the different sampling months in each year and the low abundance of recorded individuals (average of 2 ± 1 individuals in each sampling point for each temporal session); thus only occurrence data were used. Desoria orobica sampling was performed in summers 2023 (10 August), 2024 (28–29 August) in the stony debris by flotation methods in three sampling points (Fig. 4, Table 3) in an area of 25 m2 with 22 replicates in 2023 and 28 in 2024, while 17 replicates were carried out in 2020 in one sampling point. At least, three replicates for each sampling point were performed on standard volumes of soils (200 ml; Valle and others, Reference Valle, Gobbi, Brambilla, Borgatti and Caccianiga2023). Desoria orobica was also sampled by hand catching in five sampling points (Fig 4, Table 3), with 4 replicates in 2023, 12 in 2024, none in 2020, performed in each sampling point for 20 minutes by one experienced operator (Gobbi and others, Reference Gobbi, Barragán, Brambilla, Moreno, Pruna and Moret2018b; Fig. 4). Since both species, after their first identification/description for the area, could be easily identified directly on the field using a 10x lens, no individuals were collected in order not to harm the remaining population (Lencioni and Gobbi, Reference Lencioni and Gobbi2021).
3. Results
3.1. Post-LIA glacier change
At the beginning of the 1800s, Giovanni Maironi Da Ponte (Reference Maironi Da Ponte1808) described the Trobio glacier as a “great deposit of perpetual ice.” The 1833 Austrian military map provided a plausible and consistent cartographic representation of the glacier’s extent (Bonardi and others, Reference Bonardi, Rovelli, Scotti, Toffaletti, Urso and Villa2012). Around 1860, the glacier likely reached its maximum extent (Fig. 3), forming the imposing moraine still visible today. A striking photograph by Alfredo Corti (Fig. 1c) shows the glacier in good condition in the early 20th century, when Leonardo Ricci described it in detail, providing a sketch and quantifying an “average annual displacement” of 3.75 m in its middle portion (Ricci, Reference Ricci1911). In the 1920s, the Trobio glacier was mentioned in Porro’s first list of Italian glaciers (Porro, Reference Porro1925) and later described in detail by Nangeroni (Reference Nangeroni1930, Reference Nangeroni1932), who documented its subsequent fluctuations until 1952. Between 1943 and 1945 the original glacier split into two units (Trobio Ovest and Trobio Est). The glacier lost 56% of the former area between its maximum and 1954 (Fig. 3 and Table 1). Like many other Alpine glaciers, a positive phase started from the 1960s, allowing the system to regain volume and even mark a small frontal advance. Since 1987, the trend turned negative again with an average annual area decrease (AAD) that almost doubled from 0.006 km2 a-1 (1860–1989) to 0.010 km2 a-1 (1989–2022) until the complete vanishing of the glacier testified during the biological and glaciological surveys performed in summer 2023 (D’Adda and Petrone, 2024) (Fig. 3 and Table 2).
Reconstruction of Trobio glacier (Bergamo Alps, Italy) extents from Little Ice Age (LIA) maximum (∼1860) to 2022. Front positions in 1911 and 1929 are extrapolated from Caccianiga and others (Reference Caccianiga, Ravazzi and Zubiani1994). Shaded relief derived from the Lombardy region 5-m Digital Terrain Model.
Trobio glacier (Bergamo Alps, Italy) extents from Little Ice Age (LIA) maximum (∼1860) to 2022. The three parts of the glacier (Trobio Est, Trobio Centrale, Trobio Ovest) are considered individually and collectively (total).
Trobio glacier (Bergamo Alps, Italy) absolute and relative Annual Average Decrease (AAD) in multiple time-steps since LIA maximum (∼1860). The three parts of the glacier (Trobio Est, Trobio Centrale, Trobio Ovest) are considered individually and collectively (total).
3.2. Nebria tresignore and Desoria orobica occurrences
In 2024, the ground beetle Nebria tresignore was found only in close proximity to snow patches and could not be observed in the sites of the previous observations (Fig. 4); it shifted uphill by approximately 300 m in a beeline over the last 10 years (2014–24), corresponding to a shift of approximately 30 m a-1. Monitoring activity performed in 2024 revealed that Nebria tresignore now shares its microhabitat with the niche competitor, cold-adapted, Central Alpine endemic Nebria lombarda K. & J. Daniel, 1890, unlike previous observations where N. tresignore was the sole ground beetle occurring near the glacier front. For the springtail Desoria orobica, only one record was not confirmed in 2024; however, comparing abundances, very low densities were observed in 2024 with respect to the previous records. In particular, comparing average abundances recorded with flotation method, a mean value of 80.8 individuals (st. dev. 164.0) per sampling was found during 2020. The maximum values (660, 359 and 140 individuals) were all recorded within 5 cm from the snow, while the minimum values (0) were recorded at the greatest distances from the snow, from 90 to 150 cm. Conversely, during and after the disappearance of the glacier – 2023 and 2024, respectively – the maximum value recorded dropped to 3.75 (st. dev. 3.3) individuals per sampling (Fig. 4 and Table 3).
Temporal distribution of observation records of the cryophilic ground beetle Nebria tresignore and springtail Desoria orobica at the Trobio glacier (Bergamo Alps, Italy). Nebria tresignore was sampled with hand catching during the years 2014 (13 September; Gobbi and others, Reference Gobbi, Tampucci and Caccianiga2018a), 2016 (28 August; Gobbi and others, Reference Gobbi, Tampucci and Caccianiga2018a), 2023 (10 August) and 2024 (26 July and 28–29 August). Desoria orobica was sampled in 2020 (18 July; Valle and others, Reference Valle2025), 2023 (10 August), 2024 (28–29 August) with flotation methods in the stony debris and also with hand catching (more details in the text). From left to right, Trobio Ovest, Trobio Centrale, Trobio Est glaciers. The glacier extents of 2003 (white line) and 2022 (yellow line) are shown. Desoria orobica abundance values are reported in Table 3 (Orthophoto from Google Earth®).
Average abundance values and standard deviation of the cryophilic springtail Desoria orobica for year and sampling methods. HC = hand catching (expressed as individuals per sampling), F = flotation method (expressed as individuals 200 ml–1). Desoria orobica was sampled in 2020 (18 July; Valle and others, Reference Valle2025), 2023 (10 August), 2024 (28–29 August) with flotation methods in the stony debris and with hand catching on the ice (more details in the text). The exact year of Ice disappearance in ORIEN (Trobio Est) is unknown.
4. Discussion
Climate change is already making us witness glacier disappearance in real time. It is therefore crucial to begin monitoring this critical phase not only from a glaciological perspective, but also from a biological and ecological one. The Trobio case exemplifies a process that is occurring worldwide, where the extinction of small glaciers is advancing rapidly, with consequences that extend far beyond local landscapes.
The history of the Trobio glacier is emblematic of the broader transformations that have reshaped Alpine glacial landscapes since the end of the LIA. Despite some brief and difficult-to-quantify phases of advance, the Trobio glacier experienced an almost continuous retreat since the end of the LIA. The physical withdrawal, traced through the analysis of the Average Annual Decrement (AAD), reveals a clear climatic fingerprint: the doubling of the absolute AAD after 1989, mirroring trends observed in other mountain glaciers and sectors of the Alps (Scotti and others, Reference Scotti, Brardinoni and Crosta2014), clearly reflects the sharp global temperature increase occurring since the 1980s (IPCC, Reference Lee and Romero2023). Notably, the European Alps have emerged as one of the most affected regions globally, with glacial recession proceeding at a markedly fast pace (Copernicus Climate Change Service (C3S) and World Meteorological Organization (WMO), 2025). A closer examination of the temporal intervals reveals that the peak of Trobio AAD occurred between 1989 and 2003 (0.012 km2 a-1). Surprisingly, despite increasingly adverse climatic conditions in the European Alps, a partial deceleration was observed in the subsequent period (0.08 km2 a-1). This partial decorrelation, seemingly at odds with the overarching trend, can be explained by glacier’s progressive retreat into a more sheltered morphological niche in the last 20 years. As the ice front migrated beneath steep rock walls, it began to benefit from enhanced shading, increased wind-driven snow deposition and avalanche accumulation, factors that can temporarily buffer the glacier from climatic forcing. These peculiar morphological conditions typically lower the Equilibrium Line Altitude (ELA) of glaciers setting a local topography-driven ELA that is significantly lower than the ELA granted by climatic conditions only, i.e., Regional Climatic Equilibrium Line Altitude (ELArc). In Bergamo Alps this ELA decoupling is typically pronounced and it plays a critical role in sustaining the existence of the remaining glacial bodies (Scotti and others, Reference Scotti, Brardinoni and Crosta2014). However, despite the local factors offering a brief reprieve and a weak increase in glacier resilience, topographical conditions were not sufficient to halt the intensity of recent warming and the decline of the Trobio glacier. The rock walls surrounding the glacier are relatively gentle and of limited vertical extent, unable to provide the strong orographic shielding observed in other, more protected cirques in the surroundings. Conversely, the same gentle morphology, coupled with a sufficiently high elevation, once facilitated the great expansion of the glacier at the LIA apex. During that period, a lowered ELArc favored direct snow accumulations over a vast area. In this regard, therefore, the Trobio glacier presented characteristics atypical for its size, behaving more like a large alpine glacier almost entirely fed by direct snowfall. Yet, this apparent robustness masked an underlying vulnerability. The limited elevation of the hosting cirque, combined with the absence of strongly protective topography, ultimately sealed the glacier’s fate: as climate warming accelerated, these marginal conditions proved insufficient to ensure survival.
The profound physical changes observed in the Trobio glacier do not simply represent a shift in the landscape morphology. Rather, they provide a critical ecological backdrop for understanding the severe and still unfolding consequences for cryophilic fauna tightly bound to these disappearing icy habitats (Fig. 5). Although numerous speculations have addressed the impacts of glacier retreat on cold-adapted species worldwide (e.g., Kaufmann, Reference Kaufmann2002; Brown and others Reference Brown, Hannah and Milner2007; Cauvy-Fraunié and Dangles, Reference Cauvy-Fraunié and Dangles2019; Ficetola and others, Reference Ficetola2021), the direct biological impacts of complete glacier disappearance have never been investigated in the field. Thus, the Trobio glacier offers a unique natural experiment with implications that go well beyond the Alps, shedding light on the fate of cryophilic biodiversity under accelerating global warming. Our findings show that both mobile species (i.e., ground beetles) and microhabitat-dependent taxa (i.e., springtails) exhibit rapid responses to glacial loss at fine temporal scale. In particular, the current occurrence of Nebria tresignore and Desoria orobica (Fig. 5) in the area once occupied by the glacier supports the hypothesis of an “extinction debt” (Kuussaari and others, Reference Kuussaari2009). This phenomenon describes a delayed biodiversity collapse, where the full impact of habitat loss becomes apparent only after a time lag.
(a) Desoria orobica (average length: about 1.5 mm), cryophilic springtail recently described for the Trobio glacier (28 July 2024); (b) Nebria tresignore (length: 8.5–10 mm) dwelling on the surface of the Trobio glacier (30 July 2018). Photograph respectively by Barbara Valle and Danilo Donadoni.
Our data suggest that the extinction risk is particularly high for Nebria tresignore, whose ecological niche overlaps significantly with N. lombarda, a congeneric species now advancing into the glacier foreland from the surrounding slopes. Similarly to Nebria tresignore and other congeneric species (Raso and others, Reference Raso2014), Nebria lombarda preys on springtails and other arthropods. In addition, intraguild predation among Nebria species has been documented (Raso and others, 2014; Sint and others, Reference Sint, Kaufmann, Mayer and Traugott2018), raising the possibility of direct competition for diminishing food resources.
Even more alarming is the striking population decline of Desoria orobica observed between 2020 and 2024. The combination between the final disappearance of ice and two years with below-average snow cover (namely 2022 and 2023) heavily affected the consistency of the once abundant springtail species. Its survival appears tightly linked to a microclimatic niche – cold, humid and snow-insulated – that is rapidly vanishing under the current warming trends. Yet, since the exact distribution range of this newly discovered species is still not known, the broader implications of this loss remain difficult to quantify.
Ultimately, the fate of Nebria tresignore and Desoria orobica likely depends on their ability to locate and persist in small-scale refugia: cold and humid microhabitats, such as coarse debris and/or sub-superficial ice patches. Current research efforts should aim to identify potential refugial areas where conservation and monitoring efforts can be effectively directed (Morelli and others, Reference Morelli2016). However, field-based evidence on fragile cryophilic biodiversity remains limited and many species still likely remain undescribed (Fjellberg, Reference Fjellberg2010; Valle and others, Reference Valle2025), especially at low latitudes. Furthermore the lack of fine-scale data severely limits our ability to predict species-specific responses to the subtle but decisive role of local topography in shaping survival opportunities under climate stress. Documenting local glacier extinction is therefore not only crucial for local understanding, but also provides key evidence to understand and forecast the global consequences of glacier extinction for biodiversity and ecosystem functioning (and correlated ecosystems services, such as water regulation, nutrient cycling and habitat provision).
Data availability statement
Data are available from the authors upon request.
Acknowledgements
The work was partially funded by Orobie Bergamasche Regional Park, by the National Recovery and Resilience Plan (NRRP), Mission 4 Component 2 Investment 1.4 – Call for tender No. 3138 of 16 December 2021, rectified by Decree n.3175 of 18 December 2021 of the Italian Ministry of University and Research funded by the European Union – NextGenerationEU (Project code CN_00000033, Concession Decree No. 1034 of 17 June 2022 adopted by the Italian Ministry of University and Research, CUP B63C22000650007, Project title “National Biodiversity Future Center – NBFC”) and by EU Biodiversa+, the European Biodiversa Partnership under the 2021–22 BiodivProtect joint call for research proposals, co-funded by the European Commission (Grant Agreement No. 101052342 “PrioritIce-Vanishing habitats: conservation priorities for glacier-related biodiversity threatened by climate change”). The authors thank Erika Cabrini, Agostino della Janna, Elena Eustacchio, Francesco Simone Mensa, Simone Ornaghi, Duccio Tampucci, and Virginia Toscano Rivalta, for the support given on the field. The authors also thank all the field operators that surveyed this glacial site.
Competing interests
The authors declare that they have no competing interests.
Open access
