Fildes Peninsula (62°11′ S, 58°58′ W) is located at the southwestern tip of King George Island in the South Shetland Islands of the maritime Antarctic region (Fig. 1). The 38 km² peninsula is surrounded by the waters of the Drake Passage, Fildes Strait and Maxwell Bay. It is free of ice except for its northeast end which is covered by the Collins Glacier (Fig. 1b). Numerous lakes that formed following glacial recession during the Late Holocene are present on the peninsula (Simonov, 1977), some of which are used as water sources for various stations. Our study included three Fildes Peninsula lakes (Fig. 1b): two are situated adjacent to stations, while one is more distant from human infrastructure. The lakes included Hotel Lake, which is adjacent to the Teniente Marsh Aerodrome and was formerly its water supply, and which has ceased to be used for drinking water due to elevated levels of heavy metals (Peter et al., 2013); Las Estrellas Lake, which previously supplied water to Escudero Station (Chile), and is ~ 200 m from the nearest roads and infrastructure; and Mondsee Lake, situated 3.5 km from Frei Station (Chile) and 2.6 km from Artigas Station (Uruguay).

Sediment cores were taken in November-December 2016 (Las Estrellas and Mondsee lakes) and in December 2013 (Hotel Lake), at the deepest known point of each lake. Holes were drilled in the lake ice with a manual ice auger, and cores were recovered using a universal corer (Aquatic Research Instruments) with tenite butyrate core tubes of either 67 or 95 mm internal diameter. Based on sedimentation rates obtained from earlier studies, a general target length of 30 cm was established for the sediment cores, which was long enough to far exceed the period of human presence on Fildes Peninsula. The sediment-water interfaces were stabilized with sodium polyacrylate (Tomkins et al., 2008), and cores were then sealed and transported whole, cold and in the dark to Université Laval (Canada). The cores were then split lengthwise and sectioned at fine intervals for the analysis of diatoms, DNA, metals and radioisotopes. Subsamples followed visible layers in the stratigraphy of each sedimentary record and ranged from 0.1 to 0.5 cm thickness. Samples were studied to sufficient depths to reach at least the mid-20th century, in order to encompass the entire period of human influence on Fildes Peninsula and reach baseline conditions prior to local anthropogenic impacts. More details about the number of samples analyzed for each proxy as well as the sediment range included are available in Table S.1. All subsampling was carried out under sterile protocols to preclude contamination, and separate subsamples were taken for analysis of diatoms, DNA and radioisotopic dating, placed in sterile Whirl-Pak bags, and kept frozen at -80°C until analysis.

Sediment chronologies were generated from ²¹⁰Pb activities measured by alpha spectroscopy at Chronos Scientific Inc. in Ottawa, Canada. Freeze-dried samples sectioned at fine intervals (~ 4 mm) were ground and spiked with ²⁰⁹Po, after which HNO₃ (nitric acid) and HCl (hydrochloric acid) were added, and the samples were heated at 80°C for 16 hours. The solutions were then centrifuged and evaporated to dryness three times, with HCl added after each cycle. Finally, the Po isotopes were electroplated on silver disks and measured using alpha spectroscopy. Age-depth models were generated with the R package serac using the constant rate of supply (CRS) model, with interpolated ages generated to cover the specific depths of the subsamples for all lakes and indicators (Appleby, 2001; Bruel and Sabatier, 2020).

Metal concentrations (i.e., Cd, Ni, Cr, Pb, Co, Zn, Cu, Fe, Mn, As, Se, Ti) were measured on ~ 1 g (dry weight) sediment subsamples using inductively coupled plasma-mass spectrometry (ICP-MS) in the Laboratory for the Analysis of Natural and Synthetic Environmental Toxins (LANSET) at the University of Ottawa, Canada. Duplicates, blanks and standard reference materials were used for quality assurance/quality control. Values are expressed in milligrams per kilogram of dry sediment (mg kg^− 1).

Metal enrichment factors (EF) were calculated in each lake using the following equation:

where x refers to the metal measured by ICP-MS; Ti is titanium, a reliable indicator of changes in allochthonous sedimentation (Davies et al., 2015) used to normalize metal concentrations; and reference represents pre-human background values, calculated using the last sample in each core that age models indicated to be from the 19th century (i.e., 1890, 1895, and 1876 CE in Mondsee, Hotel and Las Estrellas lakes, respectively). EF values around 1 show no sign of enrichment, while values over 3 and 10 show moderate and severe enrichment, respectively (Chen et al., 2007). EFs for Ti were calculated as concentrations relative to those of the reference sample.

0.1 g of dried sediment was prepared for diatom analysis by oxidizing organic matter with 30% H₂O₂. Following digestion, samples were allowed to settle for 24 h, the supernatants aspirated, and slurries were rinsed several times with distilled water to return them to neutral pH. Slurries were then placed on cover slips to dry on in a dust-free environment and fixed on microscope slides using Naphrax mounting medium. A minimum of 300 valves were counted in each sample using a Zeiss Axioscop 2 microscope equipped with differential interference contrast (DIC) optics at 1000x magnification under immersion oil. Primary taxonomic references for the identification of the diatom flora included Zidarova et al. (2016), Van de Vijver and Kopalová (2014), Sterken et al., (2015) and Wetzel et al., (2015). The identification of some species was verified using a scanning electron microscope (SEM) at Centres Cientifics i Tecnològics of the Universitat de Barcelona (CCiTUB), Spain. In addition to taxonomic identifications, teratologies were enumerated in separate scans of 500 valves to determine the proportion of diatom deformities in each lake. Scans to determine the proportion of diatom deformities in each lake began with the surface sample and continued downcore until 3 consecutive samples showed an absence of teratologies after 200 valves were counted. The teratology percentage was based only on specimens found in valve view, as teratologies are generally not visible in girdle view (Lavoie et al., 2017).

To detect changes in bacterial community composition, bacterial diversity was analyzed by 16S rRNA amplicon sequencing. All material employed for manipulation of DNA sediment subsamples, as well as the vertical laminar flow cabinet (ESCO Class II, Type A2) where the DNA was extracted, was previously sterilized by irradiation with UV light. 0.5 g of wet sediment were aseptically transferred to sterile microtubes containing ceramic beads and an extraction buffer composed of 1% CTAB (Hexadecyltrimethylammonium bromide) and EDTA (Ethylenediaminetetraacetic acid) and homogenized with a FastPrep homogenizer (MP). After centrifugation at 12000 g the supernatants were subjected 3 times to chloroform:isoamyl alcohol (24:1) extractions and the pellets were precipitated with 0.6 volumes of cold isopropanol at room temperature during 24 h. These precipitates were then subjected to centrifugation for 40 min at 12000 g at room temperature, and the DNA obtained in the pellets was washed with cold ethanol, dried and suspended in ultrapure water overnight at 4°C. The concentration and purity of DNA was determined spectrophotometrically at 260 and 280 nm, and its quality was checked by the amplification of a variable region (V4) of the ribosomal 16S gene. PCR (polymerase chain reaction) products were verified by gel electrophoresis on 1% agarose gel in 0.5X TBE buffer. DNA samples were sent to Novogene for Illumina MiSeq paired-end sequencing of the V4 region, using the 515F and 806R primers (Caporaso et al., 2011).

To correct for sequencing errors and create Amplicon Sequence Variants (ASVs), reads were processed in R using the DADA2 pipeline following a modified version of the DADA2 Bioconductor workflow (Callahan et al., 2017). Briefly, reads were filtered and trimmed by the filterAndTrim function based on the quality score which estimates the error probability of the DNA sequence. Reads with a maximum expected error (maxEE) greater than 2 were removed and based on quality profiles, reads were truncated at 200 and 190 bp for forward and reverse reads, respectively. Sequence variations were inferred with the learnerrors and dada functions. Chimeric sequences were eliminated, and taxonomy assignments from kingdom to genus were performed with assignTaxonomy based on the SILVA database (v138) (Quast et al., 2012). ASV sequences assigned as Archaea, Chloroplasts and Mitochondria were removed. Sequences obtained were submitted to the nucleotide archive GenBank with the BioProject ID PRJNA1365190.

Multivariate analyses were used to explore patterns in the variation over time of metals, diatoms and bacteria. Principal component analysis (PCA) was performed separately for each proxy (metals, diatoms and bacteria); the ICP-MS metals data were log transformed prior to PCA while the diatom and bacteria datasets were Hellinger-transformed.

Log²¹⁰Pb activities were low and showed roughly linear decreasing trends over time in each lake, with activities in surface sediments of 95.4, 61.2, and 189.7 Bq kg^− 1, and reaching background at 55, 54 and 31 mm depth in Mondsee, Las Estrellas and Hotel lakes, respectively (Fig. S.1). CRS models indicated that the year 1968 CE, when Bellingshausen base was established, corresponded to ~ 21, ~23, and ~ 5.7 mm depth in Mondsee, Las Estrellas and Hotel lakes, respectively (Fig. S.1).

The metals with the greatest variation throughout the three sediment cores were Pb, Zn, As, Cu, and Cr (Fig. S.2). Mondsee Lake showed little change, with concentrations that ranged between 6.7 and 7.5 mg kg^− 1 for Pb, 51.4 and 59.9 mg kg^− 1 for Zn, As between 3.3 and 8.2 mg kg^− 1, Cu between 84.4 and 111.5 mg kg^− 1 and Cr between 7.9 and 10.2 mg kg^− 1 (Fig. S.2). There was greater variation in metals in Las Estrellas Lake, with concentrations ranging between 5.2–10.6 mg kg^− 1 for Pb, 39.8-148.8 mg kg^− 1 for Zn, 2.6–9.7 mg kg^− 1 for As, 44-107.2 mg kg^− 1 for Cu and 10.6–18.3 mg kg^− 1 for Cr (Fig. S.2). Hotel Lake showed by far the highest metal concentrations as well as the greatest variation, varying between 5.4-683.9 mg kg^− 1 for Pb, 47.3-350.7 mg kg^− 1 for Zn, 1.8–5.4 mg kg^− 1 for As, 62-189.4 mg kg^− 1 for Cu and 9.9–30.5 mg kg^− 1 for Cr (Fig. S.2).

The first two axes of the PCA of metal concentrations explained 80.0% of the variance in the dataset, with PC1 explaining 62.1% and PC2 explaining 17.9%. First axis scores for Mondsee Lake did not change through the core, while Las Estrellas Lake PC1 values undulated slightly between 1950 and the present, and Hotel Lake PC1 values changed markedly beginning between 1940 and 1998 (Fig. 2).

Enrichment factors for Ti were close to 1 and showed little variation throughout all three sediment cores, indicating relatively constant supplies of allochthonous sediment over time (Fig. 3). The highest EFs were observed for Pb, Zn, As, Cu, Cr and Cd (Fig. 3, Table S.2), apart from Mondsee Lake where EFs throughout the core were close to 1 (i.e., no enrichment) except for As, which increased moderately between 2012 and 2017 reaching a maximum EF of 3.0 in 2015 (Fig. 3). EFs for Pb, Cu Cr and Cd in Las Estrellas Lake were close to 1 and varied little in the core, except in 2017 for Pb which was moderately enriched (EF: 2.9) and Cr that increased somewhat but at 1.9 remained below threshold for moderate enrichment (i.e., 3; Chen et al., 2007) (Fig. 3). The metal showing the greatest enrichment in Las Estrellas Lake was Zn, which increased upwards from 1974 to the surface, with moderately enriched values in 1982 and a maximum at the surface of the core (2017) of 5.7 (Fig. 3). As was moderately enriched from 1962 to the most recent sample (2017), with the maximum values in 1974 and 1978 (EFs: 4.9 and 5.6 respectively; Fig. 3). EFs of many metals in Hotel Lake showed pronounced enrichment and had similar trends, increasing slightly in the early 20th century, and greatly after ~ 1976 to a maximum in 1998, and then decreasing somewhat in the surface sediment (Fig. 3). Pb, Zn and Cd showed severe enrichment in 1998 (maximum EFs of 146.2, 14.9 and 10.6) while As, Cu and Cr showed moderate enrichment at this point (maximum EFs: 6.1, 5.5 and 6.2) (Fig. 3).

Ninety-one diatom taxa belonging to 30 genera were identified in the three study lakes (Table S.3, Fig. S.3). In general, assemblages were dominated by small benthic diatoms (< 20 µm), mainly from the genera Achnanthidium, Psammothidium, Planothidium, Sellaphora and Staurosirella. A total of 78 species from 25 different genera was identified in Mondsee Lake (Fig. 4). Some species showed pronounced stratigraphic changes. There was a step-change in Achnanthidium indistinctum, which averaged 4.2% relative abundance prior to ~ 1991 and 26.5% thereafter. This was coincident with an increase of Planothidium frequentissimum and decreases in Staurosirella antarctica and other, less abundant species including Chamaepinnularia australomediocris and Psammothidium confusoneglectum.

Seventy-four diatom species from 28 genera were identified in the Las Estrellas Lake core (Fig. 4). There were three horizons of pronounced assemblage change in the core: at ~ 1955, ~1968 and ~ 1976. After ~ 1955, assemblages that had shown relative stability were marked by increases in Achnanthidium indistinctum, Staurosirella antarctica and Psammothidium incognitum, and the period had low abundances of Planothidium renei. Between ~ 1968 and ~ 1976, there was an abrupt decrease of Achnanthidium indistinctum from 22% relative abundance to values ~ 1%, an abrupt increase of Staurosirella antarctica, and gradual increases of Planothidium renei, Stauroneis sofia and Psammothidium abundans. Around 1975, there were sharp decreases of Staurosirella antarctica and Planothidium renei coincident with sharp increases in Sellaphora nigri and Cavinula pseudoscutiformis (Fig. 4).

A total of 50 species from 18 genera was identified in the Hotel Lake core, with no taxa having an average abundance exceeding 20%, representing the lowest diversity among the three lakes (Fig. 4). The species that showed the most evident trend was Psammothidium papilio, which was abundant in the lower part of the core (average 10.3%, 1867–1907) and then decreased dramatically above this horizon (average 0.8%), as well as Psammothidium abundans which also peaked ~ 1907 but had a higher abundance in the upper section of the core (between 10 and 18%). Sellaphora nigri was present at generally low relative abundances during most of the 20th century until the most recent sample (2014) when it increased to a relative abundance of 14% (Fig. 4), and Achnanthidium maritimo-antarcticum increased after 1976, but with lower relative abundances (between 0.3 and 5%) (Fig. 4).

The first axis of the diatom PCA explained 33.8% of the variation in the diatom data while the second axis explained 16.6%. Based on first axis scores, the most pronounced change in diatom composition occurred in Las Estrellas Lake from 1967 to 1979, while compositional changes in Mondsee and Hotel lakes were less notable (Fig. 2).

We detected a total of 14,567 ASVs across the three cores. Rarefaction curves based on the observed ASVs reached plateaus, indicating that sequencing depth was adequate to capture the overall diversity of all lakes (Fig. S.4). As was observed in our previous study (Bertoglio et al., 2025), bacterial communities from each lake were composed of a few very abundant ASVs while most taxa had low abundances. In Mondsee Lake the taxa that showed the most evident trends over time were Sulfurifustis and Phycisphaerae, which decreased after ~ 1956 (Fig. 5). Additionally, certain species dominated in the most recent sample in Mondsee Lake (2017) while others decreased dramatically. For example, Janthinobacterium, Rokubacteriales, Delftia, Clostridium sensu stricto 13, and Desulfosporosinus were more abundant in this sample, whereas Desulfatirhabdium and P9X2b3D02 (phylum Nitrospinota) decreased (Fig. 5).

The abundances of several bacteria in Las Estrellas and Hotel lakes changed notably after ∼1970 (Fig. 5). For example, Holophagaceae and Desulfatirhabdium increased while Phycisphaerae and CCM11a decreased in both lakes (Fig. 5). Other groups also showed considerable shifts but with different patterns in each lake, such as Pseudomonas and Zixibacteria that increased after ∼1970 in Las Estrellas Lake while Phycisphaerae decreased, and MBNT15, Latescibacterota, Rokubacteriales and Zixibacteria decreased in Hotel Lake after ~ 1976.

The first axis of the bacteria PCA explained 28.9% of the variation in bacterial composition data while the second axis explained 18.0%. In addition, based on first axis scores, the most pronounced change was observed in the most recent sample of Mondsee Lake (2017), whereas changes in Las Estrellas and Hotel lakes were more muted (Fig. 2).

Diatom teratologies are deformations of cell morphology that have been observed to increase in contaminated water bodies, particularly those with high concentrations of heavy metals and pesticides (Falasco et al., 2021). In Mondsee Lake they had a maximum occurrence of 1.2% (Fig. 2). Teratologies were most notable in Las Estrellas Lake, where they increased between 1975–1982 to ~ 2% of the enumerated valves, and sharply thereafter with frequencies of 6.6 and 10.0% (in 1997 and 2017, respectively), including 30% of the valves of P. abundans (Fig. 2). The surface sample of Hotel Lake also had more prevalent teratologies, increasing from 1998 to a maximum of 2.9% deformed valves in the surface sample (Fig. 2).

We tested correlations between different metal EFs and the abundances of certain bacteria that were identified as indicators of contamination (i.e., Anaerovorax, Hungateiclostridiaceae, OPB41, Pseudorhodoplanes, and Leptolinea; Bertoglio et al., 2025). Of the five taxa evaluated, all except Pseudorhodoplanes were positively correlated with enrichments of multiple metals (Table 1). Pb, Zn, Cu, Cr and Cd had positive, significant and high correlation coefficients (i.e., R² = 0.40–0.70) with between two and four taxa each (Table 1).

Fildes Peninsula lakes are subject to multiple stressors, including rapidly rising temperatures and strong UV exposure due to ozone depletion, in addition to any direct anthropogenic impacts. We hypothesized that, due to its remoteness from infrastructure, Mondsee Lake could be used as a control site to determine changes in diatom and bacterial communities due primarily to natural environmental change, including ongoing warming. This hypothesis was supported by the differing trends in metal enrichment between Mondsee Lake and the other two sites (Fig. 3).

The limnological consequences of recent warming in maritime Antarctica include changing lake ice cover duration and thickness and thermal stratification regimes (Izaguirre et al., 2021). Decreases in ice cover duration influence diatom communities, as has been observed in many lakes in the Northern Hemisphere (Lotter and Bigler, 2000; Sorvari et al., 2002; Smol et al., 2005; Rühland et al., 2015). Although somewhat muted, recent diatom shifts in Mondsee Lake likely reflect changes to ice cover duration and extent, including declines of Staurosirella antarctica, which is associated with colder conditions and prolonged ice cover (Lotter and Bigler 2000; Keatley et al., 2008), and increases after ~ 1991 of Achnanthidium indistinctum, a taxon that thrives in nearshore lake environments where external disturbances such as wind are common (McCabe and Cyr 2006; Cyr, 2016). A coincident decline in Staurosirella antarctica in Las Estrellas Lake (after the mid-1970s) also likely reflects the broader effects of climate warming on ice cover duration and thickness, although the species was absent in the 20th century in Hotel Lake. In fact, diatom assemblages in Hotel Lake showed no systematic changes, either in overall assemblage composition or in the abundances of most taxa. Hotel Lake had much thicker ice cover than the other two lakes during spring 2017 and autumn 2018 (1.60 and 0.55 m vs. 0.98 and 0.13 m on average for both seasons, respectively; Bertoglio et al., 2023). Although the precise reasons are uncertain, Hotel Lake is located in the upper section of the Grande Valley, which appears to channel cold winds between the Drake Passage to the west and Maxwell Bay to the east (pers. obs.). Its thick ice may therefore reflect a colder microclimate and imply the delayed onset of the effects of warming in this lake.

Certain bacteria found with higher abundances in the most recent sample of Mondsee Lake may potentially be associated with the consequences of warming. For example, Janthinobacterium produce the pigment violacein which confers resistance to UV radiation (Alem et al., 2020), which is expected to increase as ice cover thins. Moreover, Rokubacteriales and Desulfosporosinus are common soil taxa that were also found in peatlands (Pester et al., 2010; Ivanova et al., 2021), and their increases could be related to increasing runoff and aeolian deposition in longer melting seasons. The increase of Janthinobacterium after ~ 1995 in Las Estrellas Lake may also be a consequence of climate warming; it also increased in Hotel Lake but much more subtly. This reaffirms the muted nature of climate change effects in Hotel Lake as suggested by diatom trends. The lack of increases in other bacteria taxa reflecting climate change in Hotel and Las Estrellas lakes may be because the influence of climate change in shaping communities in these lakes is overshadowed by the pronounced impact of anthropogenic activities. Indeed, in the presence of multiple stressors related to climate change and human activities, lakes may differ in responses compared with those that only experience single stressors (Jackson et al., 2016).

The increase in metal EFs in Las Estrellas and Hotel lakes since the establishment of bases on the peninsula is consistent with our hypothesis that the sites nearest human activities would show increases in contaminants over time, and the contrast with the lack of changes in Mondsee Lake suggests that these increases cannot be ascribed to natural environmental changes. The changes we observed in Las Estrellas and Hotel Lakes therefore represent the cumulative effects of climate change and human impacts. There is growing evidence for human impacts in terrestrial and marine ecosystems in maritime Antarctica. For example, elevated Cu, Zn, Cd and Pb have been found in marine sediments, soil, lichens and mosses from Fildes Peninsula close to stations and contaminated sites (Aronson et al., 2011; Lu et al., 2012; Padeiro et al., 2016; Pereira et al., 2017; Fabri-Jr et al., 2018). These results are comparable to our findings, since heavy metal contamination may be related to intense human activity such as transportation, fossil fuel combustion, accidental oil spills, waste incineration and sewage disposal (Chu et al., 2019). Our results give temporal context to these findings, showing that the affected sites not only had high metal concentrations but that they increased over time, while those in our remote site showed little to no change over the same period.

Biological proxies also showed greater changes in the proximal vs. remote lake and provided evidence of the ecological effects of anthropogenic impact. While shifts in overall community composition of diatoms and bacteria did not differ markedly between the three lakes, we found trends in taxa that were indicators of pollution and may thus have adaptive advantages in metal-impacted environments. The diatom Sellaphora nigri is an indicator species that is found in greater abundances in environments where eutrophication, organic contaminants or pollution by pesticides and heavy metals are observed (Morin et al., 2012, 2014; Wetzel et al., 2015). Given its tolerance to pollution, the increase in S. nigri relative abundances in Las Estrellas Lake around 1975, coincident with increasing human activities as well as with the construction of numerous stations in the subsequent years (Braun et al., 2012; Peter et al. 2013) may therefore reflect contamination due to human activity. The increase of this species in the surface sediments of Hotel Lake, while less than that observed in Las Estrellas Lake, may also reflect such impacts. By comparison, S. nigri always had abundances < 1% in Mondsee Lake, which reinforces the hypothesis of the pristine nature of the lake relative to our other sites.

Diatom teratologies represent an individual-level response to environmental stress (Falasco et al., 2021). Deformed frustules associated with heavy metal stress have been reported in various studies (Falasco et al., 2009; Cantonati et al., 2014; Pandey and Bergey, 2018). While the exact mechanism of the deformations has not been demonstrated, it has been suggested that contaminants alter cell membrane polarity and cause cytoplasmic acidification, leading to disruption of cytoplasmic homeostasis (Pinto et al., 2003). While some authors have attributed diatom teratologies to high UV exposure (summarized in Falasco et al., 2021), such as that due to thinning of the stratospheric ozone layer in the Antarctic, this mechanism cannot explain the differing trends between Mondsee Lake and the other two sites. The increase in teratologies in Las Estrellas and Hotel lakes, and their absence in Mondsee Lake, however, is consistent with the observed changes in metal enrichment and thus demonstrates the ecological effects of pollution.

Within bacterial communities, particular taxa with higher tolerances to contaminants became more abundant over time in Hotel and Las Estrellas lakes, such as Sulfurifustis and Desulfatirhabdium. These sulfur bacteria may be related to the presence of pollutants as they can employ a variety of electron donors or inorganic sulfur compounds as electron acceptors (Balk et al., 2008, Kojima and Fukui 2015), such as those found in contaminants originating from the burning of fossil fuels. However, we do not exclude that the presence of these bacteria may be related to changes in sediment habitats, for example due to anoxia, influencing active bacteria rather than reflecting historical changes. Desulfatirhabdium increased in both Las Estrellas and Hotel lakes after ~ 1982, while Sulfurifustis also increased in Las Estrellas Lake over the same period. Neither increased in abundance in Mondsee Lake where their abundances in fact decreased there after ∼1956 and 1991, respectively. Several groups of sulfur bacteria (e.g. Geobacteraceae, Desulfurivibrionaceae and Rhodobacteraceae) were also previously observed in the bacterial community in water samples from Hotel Lake (Bertoglio et al., 2023). Moreover, the significant relationships between bacterial indicators of pollution and metal EFs (i.e., Hungateiclostridiaceae, OPB41, Anaerovorax and Leptolinea; Bertoglio et al., 2025), provide further evidence that human impacts are modifying aquatic communities in Fildes Peninsula’s lakes.