Holocene shifts in marine mammal distributions around Northern Greenland revealed by sedimentary ancient DNA

doi:10.21203/rs.3.rs-5291490/v1

Download PDF

Article

Holocene shifts in marine mammal distributions around Northern Greenland revealed by sedimentary ancient DNA

https://doi.org/10.21203/rs.3.rs-5291490/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Arctic marine ecosystems have undergone notable reconfigurations in response to Holocene environmental shifts. Yet our understanding of how marine mammal occurrence was impacted remains limited, due to their relative scarcity in the fossil record. We reconstructed the occurrence of marine mammals across the past 12,000 years through genetic detections based on sedimentary ancient DNA from four marine sediment cores collected around Northern Greenland, and integrated the findings with local and regional environmental proxy records. Our findings indicate a close association between the establishment of marine mammals at densities detectable in marine sediments and the deglaciation of marine environments at the onset of the Holocene. Further, we identified air temperature as a significant driver of community change across time. Several marine mammals were detected in the sediments earlier than in the fossil record, for some species by several thousands of years. During the Early-to-Mid Holocene, a period of past warmer climate, we recorded northward distribution shifts of temperate and low-arctic marine mammal species. Our findings provide unique, long-term baseline data on the occurrence of marine mammals around Northern Greenland, providing novel insights into past community dynamics and the effects of Holocene climatic shifts on the region’s marine ecosystems.

Biological sciences/Ecology/Palaeoecology

Earth and environmental sciences/Ocean sciences/Marine biology

Biological sciences/Ecology/Ecological genetics

Ongoing anthropogenic climate change is transforming the Arctic. The region is warming four times faster than average global temperatures¹ and the extent of summer sea ice has already declined by as much as 50% over the past 40 years². Temporarily ice-free conditions in the Arctic Ocean, defined as the first occurrence of a total sea ice area < 1 million km², may already emerge during summer by the end of this decade³. In response to these shifting environmental conditions, the spatial distribution of marine species across the Arctic is changing^4–7. Species endemic to the Arctic region are expected to face a drastic reduction in suitable habitats^8,9, while temperate species may benefit from the reconfiguration of oceanic currents and increased availability of ice-free habitats^10,11.

Arctic marine mammals are sensitive to climate change, especially due to their dependence on sea ice^12,13. Seven species are found around Northern Greenland year-round, including the three Arctic cetacean species [beluga (Delphinapterus leucas), narwhal (Monodon monoceros), and bowhead whale (Balaena mysticetus)], three pinniped species [ringed seal (Pusa hispida), bearded seal (Erignathus barbatus), walrus (Odobenus rosmarus)], and polar bear (Ursus maritimus). In addition, harp seal (Phoca groenlandica) and hooded seal (Cystophora cristata) seasonally migrate northwards along the east and west coasts of Greenland, and depend on the presence of sea ice for reproduction^14,15.

At least 12 temperate cetacean species with distribution ranges not confined to the Arctic have been recorded in Greenland, with species such as fin whales (Balaenoptera physalus) and minke whales (Balaenoptera acutorostrata) seasonally migrating in large numbers to ice-free feeding grounds along the central and north coasts of Greenland^16,17. The observed impacts of ongoing climate change on marine mammals found around Greenland are not uniform¹⁸. For example, the extended open water season may boost the availability of prey (e.g. krill, copepods) for species such as bowhead whales^19,20, but the associated increase in the presence of orcas (Orcinus orca) also poses higher risks of predation²¹.

Monitoring efforts provide crucial data on population dynamics, distribution shifts, and behavioral changes in response to changing environmental conditions^13,22,23. However, we lack an understanding of the longer-term dynamics of Arctic marine ecosystems. Integrating knowledge on the shorter-term response of species to anthropogenic climate change²⁴ with longer-term perspectives on how species distributions have shifted in response to past environmental changes may enhance our understanding of their resilience to global warming²⁵.

Population genetic studies of marine mammals in the Arctic document the impacts of the environmental changes of the Holocene (past 11.7 thousand years [ka]) on present-day population structure^26–28. Holocene paleoclimate records reveal periods with regional air temperatures up to 6°C warmer than the pre-industrial average²⁹, which may be used as near-future climate analogues. Specifically, a period in the Early-to-Mid Holocene was characterized by elevated air and ocean temperatures, as a consequence of higher boreal summer insolation, known as the Holocene Thermal Maximum (here broadly used to refer to the period of peak Holocene warmth ~ 10 − 5.5 ka BP)³⁰. As a result, marine outlet glaciers along the coasts of Greenland retreated dramatically^31–33, sea ice extent and thickness decreased as indicated by driftwood deposition along the northern coast of Ellesmere Island and Greenland^34,35, and oceanic currents such as the West Greenland Current (Fig. 1a) intensified³⁶.

Information on the past occurrence of species can be retrieved from the fossil record. However, discoveries of marine mammal fossils are rare, as most species live and die at sea. Archaeological sites are rich sources of faunal remains, but are limited to the species harvested by humans³⁷ and for Greenland extend back only 4.5 ka³⁸. Although individual, older remains of e.g. bowhead whale, narwhal, ringed seal, walrus, and polar bear, have been recovered from palaeontological contexts^39–42, a detailed, long-term reconstruction of these species´ histories is limited by the scarcity of fossils.

Sediment archives provide an alternative substrate to document the occurrence of species through time⁴³. In the terrestrial realm, sediment archives have been used to broaden our understanding of spatiotemporal patterns of species occurrence. For example, DNA retrieved from permafrost sediments has indicated the presence of species in the landscape for millennia after they were presumed extinct based on their last appearance dates in the fossil record^44,45.

Marine sediments have long been used for paleoenvironmental reconstruction⁴⁶, but their full potential for detecting species occurrence using DNA remains underexplored^47,48. So far, most research on marine sediments has been based on traditional micropalaeontological methods and has focused on obtaining time-series data from organisms at the base of the food web^49–53. However, the retrieval of sedimentary ancient DNA from higher trophic level organisms provides the opportunity to improve our understanding of the dynamics of marine ecosystems through time^54,55.

In this study, we characterize the occurrence of marine mammal species across the Holocene, using sedimentary ancient DNA retrieved from four marine sediment cores collected around the northern coast of Greenland that span the past ca. 12 ka (Fig. 1). We applied two complementary molecular approaches (metagenomic shotgun sequencing and hybridization capture enrichment of mitochondrial genomes) for marine mammal detection, and integrated our findings with novel paleoceanographic proxy data derived from the same or nearby sediment cores that provide information on primary productivity, sea ice conditions, and the influence of Atlantic-derived warmer water masses. Specifically, we investigated (i) the occurrence through time of the seven marine mammals that at present inhabit the Atlantic sector of the Arctic year-round; (ii) the occurrence through time of marine mammal species that at present only inhabit the Arctic seasonally; and (iii) spatiotemporal associations between marine mammal occurrence and Holocene paleoenvironmental proxy data.

Overview of sequencing data and DNA damage

Using a combination of shotgun sequencing and mitochondrial genome hybridization capture, we retrieved DNA time-series data spanning the past ~ 12 ka from four marine sediment cores collected off the coasts of Northern Greenland, albeit with varying temporal resolution (Fig. 1); LK21-IC-st26-GC1 (75.32° N 61.91° W, 912 m water depth, 320 cm sediment recovery; hereafter Melville Bay 26G), Ryder19-24-PC1 (81.62° N 62.30° W, 520 m water depth, 525 cm sediment recovery; hereafter Hall Basin 24PC), Ryder19-12-GC1 (82.58° N 52.53° W, 867 m water depth, 273.5 cm sediment recovery; hereafter Lincoln Sea 12-GC) and DA17-NG-ST07-073G (79.07° N 11.90° W, 385 m water depth, 410 cm sediment recovery; hereafter North-East Greenland 73G). Melville Bay 26G and Lincoln Sea 12-GC comprise a complete record of the Holocene, while Hall Basin 24PC and North-East Greenland 73G covered the past 9.5 cal ka BP (Fig. 1c).

The subsampling of 0.35–1 g of sediment from the dated layers of each core for DNA analysis resulted in 109 samples (11–42 samples per core), of which all were processed using hybridization capture, and a subset was subjected to additional shallow shotgun sequencing (Fig. 1c, Data table 1). When comparing DNA concentrations of total DNA extracts across sediment cores and sample ages, samples < 2 ka BP generally had the highest DNA concentrations (7.4 ± 5.1 ng µL⁻¹), while older samples yielded lower concentrations (1.4 ± 1.5 ng µL⁻¹) or remained below the detection limit of 0.05 ng µL⁻¹ (Fig. 2d, 3d, 3f, 4d).

We used shotgun sequencing to characterize the most abundant eukaryotic taxa and hybridization capture to specifically enrich the genetic traces of marine mammals endemic to the Arctic and of relevant marine mammals, whose present day distribution is not confined only to the Arctic. We investigated the association of the combined dataset with paleoceanographic proxy data, to identify potential drivers of change through time. All genetic detections are reported per taxon and sample, in unique sequence abundances (defined as the number of unique DNA sequences) and in relative sequence abundances (defined as the number of unique sequences divided by the number of quality-filtered sequences). Both metrics are interpreted as confidence of the genetic signal underlying the detection of a taxon; high unique sequence abundance and high relative sequence abundance translate into a detection with higher confidence, and low unique sequence abundance and low relative sequence abundance into a detection with lower confidence.

The analysis of nucleotide deamination rates (DNA damage) in the shotgun sequencing and hybridization capture data enabled us to identify the accumulation of nucleotide deaminations for older samples typical for ancient DNA, and thus provided an authentication of our species detections. We visualized DNA damage by sample age for each core based on sequences assigned to eukaryotes for the shotgun sequencing dataset, and sequences assigned to marine mammals in the hybridization capture dataset (Fig. S1). Overall, the percentage of DNA sequences containing DNA damage was relatively low (< 15%), even for the oldest samples (Fig. S1). Nevertheless, DNA damage was detected for most samples > 3 cal ka BP, especially for taxa with > 100 sequences.

Age-depth models of the studied marine sediment cores

We established age-depth models for the Melville Bay 26G and Hall Basin 24PC cores. Age-depth models are available for Lincoln Sea 12-GC⁵⁶ and North-East Greenland 73G⁵⁷.

For Melville Bay 26G, we constructed an age-depth model using ¹⁴C radiocarbon dates of three mixed planktonic and three mixed benthic foraminifera samples, ²¹⁰Pb excess activity from the uppermost part of the sediment core, and one age constraint (Fig. S4). The lower part of Melville Bay 26G (213–320 cm) was laminated with high magnetic susceptibility and we did not find sufficient microfossils for ¹⁴C dating in this part of the core (Fig. S4). At 213 cm core depth, we noted a shift in the sedimentology from laminated to homogenous and a pronounced increase in the magnetic susceptibility followed by a gradual decrease towards the surface of the core. Due to the observed shift in sedimentology and the absence of datable material in the lower part of the core, we extrapolated the age-depth model only up to the depth where the sedimentology changes (213 cm depth). The upper section of Melville Bay 26G (0–213 cm) was homogeneous with larger drop stones deposited throughout. Due to lower sedimentation rates in the uppermost part of the sediment core (0–14 cm), the temporal resolution of the past two thousand years is limited to two samples (Fig. 1). The age-depth model was constrained in the top of the sediment core to the year of retrieval (2021). A modern age for the surface of the core is supported by the presence of detectable ²¹⁰Pb excess activity in the uppermost centimeter of the core (Table S1). Raw and calibrated ¹⁴C dates are listed in Table S2. and Fig. S4 shows an overview of the final chronology.

For Hall Basin 24PC, no new radiocarbon dates were generated. Rather, existing radiocarbon dates of another sediment core retrieved at the same site (OD1507-18GC)⁵⁸ were migrated onto Hall Basin 24PC, after a common depth scale was constructed through the correlation of bulk density, magnetic susceptibility, and XRF-scanning data (Fig. S5, Data table 4). The age-depth model (Fig. S6) was generated using clam⁵⁹.

Shotgun sequencing data

After quality-filtering the shotgun sequencing data, a total of 3,824 ± 3,752 sequences per sample were taxonomically assigned (Data table 1). Of these, 56–100% (mean: 90%) were assigned to prokaryotes, which were excluded from further analysis. All ten blanks as well as five samples did not contain any taxonomic assignments after quality-filtering. The resulting taxonomic profile contained only six eukaryotic families (or ten, if eukaryotic families detected in less than three samples are considered), as expected from the relatively shallow sequencing depth of 1.5 ± 0.7 M raw sequences per sample. We refrained from deeper shotgun sequencing due to the high cost of data needed to exhaustively describe samples.

Our shotgun sequencing data revealed DNA of codfishes (Gadidae) and pinnipeds (Phocidae) in all four sediment cores, and DNA of diatoms (Chaetoceraceae) in all cores except for North-East Greenland 73G (Data table 1, Fig. S7 - S9). Across sediment cores, we detected a higher number of eukaryotic families in samples > 2 cal ka BP. Additionally, the number of sequences assigned to eukaryotes decreased from past to present, and across sediment cores, all five samples that did not contain any sequences assigned to eukaryotes fall in the most recent time period (< 2 cal ka BP). We recorded low numbers of sequences assigned to codfishes in the five Melville Bay 26G samples dated ≥ 11.7 cal ka BP.

Melville Bay 26G

Using hybridization capture, we detected a total of ten different marine mammal species in Melville Bay 26G (Fig. 2e). No DNA was assigned to marine mammals in the six samples from the lower, laminated part of the sediment core (> 213 cm). The earliest detected marine mammal DNA was assigned to ringed seals (11.0–13.8 cal ka BP). Between 11.4–7.3 cal ka BP, we detected at least three different species in samples with marine mammal DNA. In a sample dated 11.4 ± 1.2 cal ka BP, we recorded the highest number of sequences assigned to one species (670 sequences assigned to harp seal). In a sample dated 7.3 ± 0.6 cal ka BP old, we recorded DNA from eight different marine mammals (Fig. 2e). After 7 cal ka BP, we detected fewer species, especially 3 − 0 cal ka BP. Between 11 − 3 cal ka BP, we recorded fin whale DNA (north of the species’ current distribution range) and DNA of harp seal, hooded seal, beluga, minke whale, and orca; all of which only occur seasonally in the region today (Fig. 2e; Table S4).

The fossil remains of foraminifera, single-celled protists that live in either the water column or at the seafloor, can be used to reconstruct paleoceanographic conditions^49–53. For our analysis of Melville Bay 26G, we grouped the following foraminifera species to indicate influx of Atlantic-derived water masses indicative of relatively warmer and saltier sub-surface waters associated with the West Greenland Current^36,60: the calcareous species Cassidulina neoteretis, Cassidulina reniforme, Islandiella norcrossi and Pullenia osloensis, as well as the agglutinated species Adercotryma glomerata, Lagenammina difflugiformis, Reophax catella, Reophax pilulifer, and Reophax fusiformis. We recorded their absence or near-absence in the oldest samples (> 11 cal ka BP), high counts from 10 − 5 cal ka BP and low counts in the Late Holocene (< 5 cal ka BP) (Fig. 2c, Data sheet 3).

Hall Basin 24PC and Lincoln Sea 12-GC

The earliest detected marine mammal DNA in Hall Basin 24PC was assigned to ringed seals (7.3 ± 0.1 cal ka BP). We detected narwhal and ringed seal using hybridization capture in all Hall Basin 24PC samples < 5 cal ka BP, albeit with decreasing sequence abundance over time (Fig. 3e). Between 5 cal ka BP and 3 cal ka BP, we also recorded harp and hooded seal DNA, two species which have only recently been recorded this far north⁶¹. In addition, we detected beluga whale DNA at 2.5 ± 0.1 cal ka BP, which has never been recorded this far north in present-day and historic surveys (Table S4). The first detection of marine mammals (ringed seal) in Lincoln Sea 12-GC using hybridization capture is 8.4 ± 0.1 cal ka BP, and ringed seal DNA is recorded in all but one younger sample (Fig. 3g). Narwhal DNA was first identified at 7.9 ± 0.1 cal ka BP, and was detected in all but one younger sample, again with decreasing sequence abundance over time. We also detected DNA assigned to hooded seals in three samples (7.8 ± 0.1, 7.1 ± 0.1 & 1.8 ± 0.1 cal ka BP), albeit only with a low number of sequences and low relative sequence abundance.

North-East Greenland 73G

The hybridization capture data from North-East Greenland 73G revealed DNA of four endemic Arctic marine mammals. We also retrieved DNA of harp seal (Fig. 4), which seasonally migrates northwards along the East Greenland coast (Table S4). Most samples of this core are characterized by low numbers of DNA sequences (3–10 sequences per species per sample). For example, ringed seal DNA was detected throughout the core (13 out of 38 samples) but samples always contained < 50 sequences. In a sample dated 7.9 ± 0.2 cal ka BP, narwhal DNA was first retrieved, but only samples < 4.5 cal ka BP showed a continuous genetic signal of narwhals. DNA from harp seals was detected in samples > 3 cal ka BP. The highest number of sequences (145 sequences) was assigned to bearded seals in the youngest sample (~ 0.3 ± 0.1 cal ka BP).

DNA detections in the context of changing paleoenvironmental conditions

The statistical analysis of our DNA detections in conjunction with reconstructed air temperatures, sea ice algal productivity, primary phytoplankton productivity, and foraminifera assemblages, revealed correlations between DNA detections and interpolated paleoenvironmental measurements, and identified air temperature as a significant driver of community change (Fig. S10 & S11). Based on the pairwise Spearman’s rank correlation analysis, we found strong positive correlations between e.g. codfishes and pinnipeds (r_s=0.8; p < 0.01), codfishes and ringed seals (r_s=0.7; p < 0.01), codfishes and interpolated air temperatures (r_s=0.6; p < 0.01), and priapulids (Priapulidae) and bowhead whales (r_s=0.7; p < 0.01) (Fig. S10).

The redundancy analysis (RDA) of DNA data from the two sediment cores Melville Bay 26G and North-East Greenland 73G, revealed air temperature to be a significant explanatory variable (F = 3.01; Pr(> F) = 0.079) albeit explaining only ~ 3% of the observed variance (Fig. S11).

Using a combination of shallow shotgun sequencing and mitochondrial hybridization capture, we recovered unique time series of marine mammal species occurrence covering the past ~ 12,000 years from four coastal marine sediment cores around Northern Greenland. By integrating the faunal data retrieved from the cores with environmental proxy records, we identified periods of environmental changes and associated shifts in species distributions at levels detectable via sedimentary ancient DNA.

Ecosystem-wide changes in Melville Bay during the Holocene

The analysis of the Melville Bay 26G core, retrieved from the continental shelf of north-west Greenland (Fig. 1), revealed the paleoceanographic conditions and marine mammal communities throughout the entire Holocene, following the last deglaciation.

The six oldest samples in the laminated section of this core (> 213 cm depth in core) did not contain marine mammal DNA, nor did we detect foraminifera associated with Atlantic-derived water masses, but we detected low numbers of DNA assigned to codfishes and priapulids in the shotgun sequencing data (Fig. S7d). During the Last Glacial Maximum (~ 26.5 to ~ 19 cal ka BP), the Greenland Ice Sheet extended to the edge of the continental shelf in Melville Bay, and only retreated from the outer coast by 11.6 ± 0.3 cal ka BP, when the marine-based ice sheet collapsed in Melville Bay⁶². In marine sediment cores, retreating ice-sheet deposits can be identified by a transition of laminated sediments containing none or low amounts of foraminifera to homogeneous sediments with dropstones and foraminifera⁵⁸. The six samples retrieved from the laminated section of the core > 213 cm thus originate from retreating ice-sheet deposits characterized by high sedimentation rates and therefore likely represent a short time period at or just prior to 11.6 ka BP.

The first marine mammal DNA we detected was from ringed seal at 11.0–13.8 cal ka BP. This finding pre-dates the earliest Holocene fossil evidence (dated 9.1–9.7 cal ka BP⁶³) of this species in Greenland – and indeed of any marine mammal species – by ~ 2 ka. The same sample revealed DNA of codfishes, diatoms, and pinnipeds in the shotgun sequencing data (Fig. S7d). For our DNA findings of ringed seals and codfishes, we identified a positive correlation, not just for Melville Bay 26G, but across samples from all four cores (Fig. S10). The analysis of Melville Bay 26G for foraminifera revealed near-absence before 10.5 cal ka BP (Data sheet 3, Fig. 2c), albeit an increase in the presence of foraminifera had been recorded in marine sediments of northernmost Baffin Bay at ~ 12.3 cal ka BP^50,64. Ringed seals are physiologically adapted to life in the pack ice and are at present found in the Arctic year-round⁶⁵. Our finding supports the hypothesis that ringed seals were the first marine mammals to move north after the end of the Last Glacial Maximum³⁸.

Between 11.4–7.3 cal ka BP, we detected DNA of marine mammals that are at present found in the region year-round (e.g. narwhal), only seasonally (e.g. harp seal), or not at all (fin whale) (Fig. 2, Table S4), potentially reflecting a non-analogous community structure in the Early Holocene as compared to recent and historic surveys^14,15,66,67. This period also had the highest diversity of detected eukaryotic families and the highest abundance of foraminifera associated with relatively warmer waters of Atlantic origin, which indicates increased intensity of the West Greenland Current (Fig. 2).

The detection of narwhal DNA at 11.4 ± 1.2 cal ka BP predates the earliest Holocene fossil record for narwhals in Greenland (dated 6.1 ± 0.2 cal ka BP, from the northern tip of Greenland 83.65° N)³⁵ by almost five millennia, and an older fossil (dated 6.8 ± 0.1 cal ka BP) from northern Ellesmere Island, Canada (82° N)⁶⁸. Similarly, the detection of bowhead whale DNA at 10.7 ± 1.0 cal ka BP predates the earliest fossil record of this species in Melville Bay (at 9.3–9.0 cal ka BP⁶⁹) by at least 500 years. In contrast to the relative scarcity of bowhead whales in the Greenland fossil record, their fossil occurrence in the Canadian Arctic Archipelago documents their presence at similar latitudes (~ 73° N) as Melville Bay (~ 75° N) as early as 10.5 ka BP^70,71.

Temperate cetaceans, including minke whales and orcas, are known to migrate northward along the west coast of Greenland during the summer months¹⁷, albeit Melville Bay is at the northern limit of their present-day distribution range (Table S4). Fin whales, which we detected at 7.8 ± 0.7 cal ka BP, have not previously been recorded along the west coast of Greenland further north than 71° N in present-day and historic surveys^16,73,74. In addition to temperate marine mammals, we also detected DNA from the low-arctic hooded seal (at 11.4 ± 1.2 and 4 ± 0.4 cal ka BP), harp seal (11.4–0.9 cal ka BP), and beluga (7.9–3 cal ka BP). Both hooded and harp seals migrate northward along the west coast of Greenland during summer, but in contrast to harp seals, which forage in shallower waters^15,75, hooded seals generally stay further offshore¹⁴. For belugas, Melville Bay is known as a migration corridor between their summer distribution hotspot in the Canadian Arctic and their wintering grounds further south along the west coast of Greenland⁶⁷.

The detection of temperate and low-arctic marine mammal species suggests that during the Early-to-Mid Holocene, they were present in the area at densities that are detectable in marine sediments. We hypothesize this expansion or northward shift in their distribution ranges was associated with environmental changes. Specifically, the redundancy analysis (RDA) of DNA detections and reconstructed paleoenvironmental proxy measurements revealed air temperature to be a significant driver of community change (Fig. S11). Furthermore, our analysis of foraminiferal assemblages from Melville Bay 26G revealed high foraminifera counts, indicating increased intensity of the West Greenland Current 10 − 7 cal ka BP (Fig. 2c), which is in line with previous paleoceanographic reconstructions from northernmost Baffin Bay^50,64,76. The biomarkers brassicasterol, dinosterol, and IP₂₅ were also relatively high in Melville Bay between 10.5–6 cal ka BP, indicating a phase of high primary productivity and stable ice-edge conditions^77,78. The West Greenland Current transports relatively warmer waters of Atlantic origin northwards along the west coast of Greenland⁷⁹ and its prevalence has been shown to affect the size and abundance of marine organisms at the base of the food web⁸⁰.

Our detections in the Early-to-Mid Holocene of temperate species such as orcas, fin whales, and minke whales at 75° N may reflect a northward expansion relative to their present-day distribution, similar to the occurrence of boreal mollusc species along the west coast of Greenland (9.2–5.6 cal ka BP), which has been interpreted as a consequence of higher sea-surface temperatures associated with an increased intensity of the West Greenland Current⁸¹. To date, the mechanisms controlling increased northward heat transport are not yet fully understood, but certain atmospheric conditions can facilitate increased intensity of the West Greenland Current⁸².

Overall, our DNA assemblage from Melville Bay 26G contained both Arctic and temperate species, which may indicate the coexistence of species in non-analogous communities. However, the age uncertainty of our individual sediment samples from this core span at least several hundred years (ranging from 0.4–1.6 ka) and therefore we cannot rule out that replacement of species may have occurred at a finer temporal scale (Fig. 2e).

In Disko Bay (69° N), an increased prevalence of temperate cetaceans such as humpback whales has been observed during the summer months since the 18th century⁸³. Humpback whales, which are found at lower latitudes during most of the year, have been observed in the shallower waters of Disko Bay later in the summer, whereas bowhead whales migrate to Disko Bay earlier and spend more time in deeper waters. The co-occurrence rather than replacement of species may be facilitated by bowhead whales feeding on copepods and amphipods, while humpback whales target krill and fish^83,84. However, further south, off the coast of south-east Greenland, the increased presence of temperate marine mammal species correlates with a decrease in Arctic species⁶.

Following this period of higher species diversity, marine mammal detections were more sporadic in the Mid-to-Late Holocene (7.2–3 cal ka BP), although we still detected species occurring seasonally in the region today (Fig. 2e). In the same period, we detected less eukaryotic families (Fig. S7d), and the abundance of foraminifera indicative of increased intensity of the West Greenland Current decreased (Fig. 2c). The timing matches the disappearance of boreal mollusc species in western Baffin Bay after 5.6 cal ka BP⁸¹ and eastern Baffin Island after 3 cal ka BP⁷¹, both attributed to a decline in sea-surface temperature. Similarly, sea-surface temperature reconstructions based on diatom and dinocyst assemblages of a sediment core from northernmost Baffin Bay showed a transition from fluctuating but higher temperatures > 4 cal ka BP, to lower temperatures < 4 cal ka BP⁷⁶.

After 3.2 cal ka BP, marine mammal detections were rare and we only detected DNA of species that occur seasonally in the region today in one sample (harp seal at 0.9 cal ka BP). In this period, our Melville Bay 26G foraminifera data suggests low influx of Atlantic-derived water masses, potentially reflecting a phase of a weaker West Greenland Current, as has been suggested based on a paleo-oceanographic reconstruction using another marine sediment core from southern Melville Bay⁴⁹. Furthermore, data from several other marine sediment cores indicate a phase of increasing sea ice concentrations and decreasing phytoplankton production during the Late Holocene in northern Baffin Bay^52,64,85 and Disko Bay³⁶. A shift from low-arctic species such as harp seal and harbor porpoise > 3 cal ka BP to Arctic endemic marine mammals such as ringed seals and bowhead whales < 3 cal ka BP has also been described based on the analysis of archaeological middens^86,87. The change in the consumption of marine mammals by paleo-inuit communities along the west coast of Greenland may reflect the species that were most prevalent in the environment at a certain time period.

Only two samples were dated to < 2.2 cal ka BP, and thus we have limited temporal resolution for the most recent past for the region. However, one sample (2.2 ± 0.7 cal ka BP) with DNA of narwhal, bearded seal, and ringed seal falls within the Roman Warm Period, which is a period of relatively higher temperatures in the context of the Late Holocene ~ 2.2–1.3 ka BP⁸⁸, during which marine sediment cores in Baffin Bay have recorded increased intensity of the West Greenland Current^36,50. In contrast, the youngest sample (0.2 ± 0.3 cal ka BP) contained only ringed seal DNA and fell within the Little Ice Age – the most recent cold period ca. 0.6 − 0.1 ka BP⁸⁸ – where low intensity of the West Greenland Current has been recorded^36,50. The absence of marine mammals in periods of colder climates and lower intensity of the West Greenland Current could point to a southward shift in distribution, as has been identified for bowhead whales during the Little Ice Age based on fossil remains⁸⁹.

The opening of the Nares Strait

Our two northernmost cores, Lincoln Sea 12-GC (82.58° N) and Hall Basin 24PC (81.62° N) (Fig. 1a), provide unique insights into the Holocene history of marine mammal populations along the northernmost coast of Greenland. Today, the region is regarded as part of the ‘Last Ice Area’, where the persistence of multiyear sea ice may provide a refuge for Arctic species under global warming⁹⁰. During the Late Pleistocene (19 − 13 ka BP), the narrow connection between northernmost Baffin Bay and the Lincoln Sea known as the Nares Strait was covered by shelf-based ice extending between the Ellesmere Island and Greenland ice sheets³⁴. The retreat of shelf-based ice marks the opening of the Nares Strait ~ 9 cal ka BP, although sea ice is still pronounced throughout most of the year today^51,91.

The Lincoln Sea 12-GC core, retrieved north-east of Hall Basin 24PC, extends back ~ 12 ka BP, whereas Hall Basin 24PC only covers the past 9.5 ka BP (Fig. 1c). We did not detect any marine mammal DNA in Lincoln Sea 12-GC > 8.4 cal ka BP (Fig. 3); the Early Holocene has been characterized as a period of seasonal sea ice and relatively higher primary productivity in the southern Lincoln Sea⁵⁶. Even if the previous detection of the sea ice algae biomarker IP₂₅ and primary productivity biomarker brassicasterol in the same core corresponding to the Early Holocene indicate suitable sea-surface conditions⁵⁶, marine mammals may not have been able to access the habitat as long as the Nares Strait was covered by glacial ice. The first detection of marine mammal DNA (ringed seal; 8.4 ± 0.1 cal ka BP) supports this, occurring shortly after the opening of the Nares Strait ~ 9 cal ka BP^51,91. At the same time, the detection of ringed seal DNA is in line with our findings for Melville Bay 26G and provides further evidence that ringed seals are the first marine mammals to move to newly accessible habitats³⁸.

We provide evidence for the occurrence of narwhals in the Lincoln Sea by 7.9 ± 0.1 cal ka BP; we detected narwhal DNA in all samples < 7.9 cal ka BP from Lincoln Sea 12-GC (except one sample dated 5.3 ± 0.2 cal ka BP), and in five out of ten Hall Basin 24PC samples, indicating the presence of the species in the area at a density that is detectable in the cores (Fig. 3). Our findings suggest colonization of the newly accessible habitat by populations migrating north from Baffin Bay through Nares Strait occurred within a few hundred years of the strait opening. Although single narwhal observations have documented the species´ presence in the Nares Strait^92–94, their northernmost recognised management unit (Smith Sound) lies further south (~ 78° N) and only partly includes Nares Strait. As our understanding of past and present distributions of narwhals improves, the consequences for current management units and marine protected areas will need to be considered, especially in the light of expected future northward distribution shifts^8,95.

In the Hall Basin 24PC core (81.62° N), our DNA detections indicate a later establishment of marine mammals than for Lincoln Sea 12-GC, albeit the period 8.5–5 cal ka BP was only covered by two samples (7.3 & 8.1 cal ka BP, Fig. 3). A possible explanation may be the earlier retreat of the Ryder Glacier from the mouth of Sherard Osborn Fjord (estimated at > 10.7 ± 0.4 ka cal BP)³² relative to the retreat of the Petermann Glacier from the Petermann fjord mouth (estimated at 7.5 cal ka BP)⁹⁶. Although the Lincoln Sea 12-GC core (in proximity to the Ryder Glacier) and Hall Basin 24PC core (in proximity to the Petermann Glacier) were collected outside the actual fjords, the dynamics of the two glaciers during the Holocene have had a strong impact on the marine ecosystems, as seen e.g. by the dramatic increase in the abundance of foraminifera < 8.5 cal ka BP in another Hall Basin marine sediment core (OD1507-18GC)⁵⁸.

In the Mid-to-Late Holocene (5 − 2 cal ka BP), Hall Basin 24PC recorded the presence of harp seal, hooded seal, and beluga: three migratory species to the northernmost part of Baffin Bay, of which only harp and hooded seal have been observed north of 80° N in present-day and historic surveys (Fig. 4)^61,67. Albeit based on a single sample, the detection of beluga DNA in Hall Basin 24PC at 2.5 ± 0.0 cal ka BP coincides with the timing of higher air temperatures (+ 0.9° C as compared to the pre-industrial average²⁹), suggesting a northward distribution shift of the species during past warmer climates. This matches habitat suitability models, which predict a northward shift of suitable habitat from areas where belugas occur in summer today (~ 75° N) into parts of Nares Strait (~ 79° N), assuming warming < 2° C as compared to the pre-industrial average⁸.

In the Lincoln Sea 12-GC core, which is further north (82.58° N) than Hall Basin 24PC, all samples with hooded seal DNA were > 0.9 ± 0.3 cal ka BP, prior to the most recent re-advance of the Ryder glacier and the re-growth of its ice tongue to the outer sill of Sherard Osborn fjord at < 0.9 ka BP³², which may have negatively impacted habitat suitability for this species.

Episodic marine mammal detections in north-east Greenland over the past 10 ka

The marine ecosystem of the north-east Greenland shelf is shaped by the East Greenland Current (Fig. 1a), which transports cold water masses and > 90% of the sea ice exported from the Central Arctic Ocean into the North Atlantic⁹⁷, and a deflected branch of the West Spitsbergen Current, which transports relatively warm and saline water masses from the North Atlantic⁹⁸.

In sediment core North-East Greenland 73G (79.07° N), we identified several discrete periods over the past 10 ka where samples contained marine mammal DNA from multiple species, interspersed with episodes where samples were characterized by the absence or near-absence of marine mammal DNA (Fig. 4). Between 10 − 9 cal ka BP, we detected harp seal DNA. In the same period, sea ice biomarkers indicate conditions between open sea and dense drift ice (i.e. marginal ice zone conditions), high primary productivity, and elevated Atlantic water influence^57,99. The environmental conditions may thus have facilitated the seasonal migration of harp seals further north than in periods with increased sea ice cover or low primary productivity. Between 9 − 8 cal ka BP, we recorded the near-absence of marine mammal DNA (Fig. 4e); a time interval characterized by relatively low sea ice algae productivity, low primary productivity, and low Atlantic water influence (Fig. 4c). In contrast, harp seal DNA occurs again (along with narwhal and ringed seal DNA) between 8–6.9 cal ka BP, a period with similar environmental conditions as 10 − 9 cal ka BP.

Between 6.9 cal ka BP and 4 cal ka BP, we retrieve DNA from bowhead whale, bearded seal, and ringed seal, but not harp seal. In this period, sea ice algae and primary productivity remained stable, while Atlantic water influence decreased notably after 6.5 cal ka BP (Fig. 4c). Similar to 9 − 8 cal ka BP, environmental conditions in the region may not have been favorable for harp seals, and their distribution may have shifted further south.

We only detected bowhead whale DNA in one sample 5.7 ± 0.2 cal ka BP, although the current presence of the species in this region has repeatedly been reported^100,101. It has been hypothesized that bowhead whales were present in north-east Greenland in the Early Holocene³⁸, based on their Early Holocene presence in Svalbard¹⁰², where the timing of the last deglaciation (and thus the availability of suitable habitat) was similar to north-east Greenland^103,104. However, few bowhead whale fossils have to date been found in east Greenland; one Early Holocene record further south (72.08° N; 8.6 ± 0.1 cal ka BP)¹⁰⁵, one Mid Holocene at the same latitude as our core (79.34° N; 5.0–5.7 cal ka BP)¹⁰⁶ and one Mid Holocene record further north (81° N; 5.9 ± 0.1 cal ka BP)³⁹.

A key difference between marine ecosystems around Svalbard and north-east Greenland is the interplay of currents transporting cold and warm water masses, leading to sea surface temperatures ~ 5°C lower on the north-east Greenland shelf, as compared to the Svalbard shelf¹⁰⁷. Our low genetic detection of bowhead whales in marine sediments and rare fossil discoveries in the region suggest a limited past presence of the species along the east coast of Greenland.

After 3 cal ka BP, we did not identify DNA assigned to codfishes in any of the North-East Greenland 73G samples. We also did not detect DNA assigned to harp seal, which correlates with the detection of codfishes (Fig. S10). Overall, marine mammal detections are rare, with narwhal DNA recorded in five of 13 samples, and ringed seal DNA in three of 13 samples. The youngest sample (0.3 ± 0.1 cal ka BP) contained the highest number of sequences of any marine mammal (bearded seal) in this core. Both sea ice algal productivity and phytoplankton productivity increase slightly after 1 ka, pointing towards a shift in the environmental conditions that has been interpreted as reflecting the formation of the North-East Greenland polynya⁹⁹. Polynyas are annually recurring ice-free areas in high latitudes characterized by high primary productivity that can support high abundances of Arctic species communities, including marine mammals^85,108. In contrast to the North Water polynya (located in northernmost Baffin Bay), the history of the North-East Greenland polynya has so far received less attention⁹⁹. Our findings of episodic rather than continuous detection of marine mammal DNA in the region encompassing the North-East Greenland polynya potentially reflects the instability of this marine ecosystem and vulnerability to climatic changes as has been inferred for the North Water polynya⁸⁵.

Our findings provide novel insights into 12,000 years of marine mammal occurrence around Northern Greenland. We report the increased prevalence of low-arctic and temperate species during the Early-to-Mid Holocene at Melville Bay 26G and North-East Greenland 73G, a period with regional high air temperatures, and the presence of foraminifera associated with increased influence of Atlantic-derived warmer water masses, and higher primary productivity. Our analysis of Lincoln Sea 12-GC represents the northernmost retrieval of ancient environmental DNA from any sediment to date. Based on data from Lincoln Sea 12-GC and Hall Basin 24PC, we reconstruct the timing of the first establishment of marine mammal populations after the opening of the Nares Strait ~ 9 kya. We detect the earlier occurrence – in some cases by several thousands of years – of several marine mammal species in Northern Greenland relative to their fossil chronology. During the Late Holocene, we detected fewer marine mammal species at all four sites, likely reflecting the decreased influence of Atlantic-derived warmer water masses and lower primary productivity during the neoglaciation. Our study demonstrates the potential of sedimentary ancient DNA for providing long-term baseline data of marine mammal occurrences, and for improving our understanding of the effects of past environmental changes on species distributions and community composition.

Sediment cores

We analyzed four marine sedimentary cores in the present study, which were collected off the coasts of Northern Greenland (Fig. 1a). All cores were cut into 1 m sections onboard the research vessels upon retrieval, and stored at 4°C until subsampling for DNA was carried out. The four cores were collected during three expeditions:

Gravity core LK21-IC-st26-GC1 (75.319° N 61.910° W, 912 m water depth, 320 cm sediment recovery; hereafter Melville Bay 26G) was retrieved from the north-west Greenland shelf during the ICAROS Expedition onboard HDMS Lauge Koch in 2021. A chronology was constructed using a combination of ²¹⁰Pb and ¹⁴C dating. The upper 10 cm were tested for ²¹⁰Pb activity using a Canberra ultralow-background Ge-detector using the Constant Rate of Supply (CRS) model¹⁰⁹. The radiocarbon analysis was carried out using an accelerator mass spectrometer (AMS) mini carbon dating system (MICADAS) with gas targets¹¹⁰ on three mixed planktonic and three mixed benthic foraminifera samples. The age-depth model was modeled in R¹¹¹ using a Bayesian accumulation model code (BACON)¹¹² calibrated with Marine20¹¹³ and a local reservoir correction (ΔR) of -49 ± 59 years¹¹⁴.

Piston core Ryder19-24-PC1 (81.622° N 62.296° W, 520 m water depth, 525 cm sediment recovery; hereafter Hall Basin 24PC) was retrieved from the north Greenland shelf during the Ryder 2019 Expedition onboard R/V Oden. It was taken at the same station as OD1507-18GC, which had been collected by R/V Oden during the Petermann 2015 expedition. The objective was to retrieve a longer core that penetrated the basal diamict. Radiocarbon dates of OD1507-18GC⁵⁸ were migrated onto Hall Basin 24PC after a common depth scale was constructed through the correlation of bulk density, magnetic susceptibility and XRF-scanning data (Fig. S5, Data table 4). The top of the basal diamict was assigned an age of 9600 ± 200 years BP⁵⁸. The age-depth model (Fig. S6) was generated using clam⁵⁹.

Gravity core Ryder19-12-GC1 (82.578° N 52.528° W, 867 m water depth, 273.5 cm sediment recovery; hereafter Lincoln Sea 12-GC) was also retrieved during the Ryder 2019 Expedition and an age-depth model (Fig. S2) and a paleoceanographic reconstruction based on biosterols has been published⁵⁶.

Gravity core DA17-NG-ST07-073G (79.068° N 11.903° W, 385 m water depth, 410 cm sediment recovery; hereafter North-East Greenland 73G) was retrieved from the north-east Greenland shelf during the NorthGreen Expedition onboard R/V Dana in 2017. An age-depth model (Fig. S3) and a paleoceanographic reconstruction based on foraminifera shells has been published⁵⁷.

Subsampling of cores for sedaDNA analysis

The subsampling of Melville Bay 26G and North-East Greenland 73G was carried out at Globe Institute, University of Copenhagen, whereas subsampling of Hall Basin 24PC and Lincoln Sea 12-GC was carried out at the Center for Paleogenetics, University of Stockholm. In both events, a designated clean sub-sampling laboratory was used that is physically isolated from the molecular biology laboratories. All working surfaces and equipment used during the subsampling were soaked in bleach and subsequently cleaned with Ethanol. All people involved in the subsampling process wore appropriate protective clothing, including lab coveralls, two layers of gloves, surgical face masks, and sleeves. We regularly changed gloves and used 5% bleach followed by ethanol to avoid contamination. Initially, each core section was split into one working half and one archive half. We carefully removed a thin layer of sediment (~ 0.2 cm) using sterile plastic cards first and another thin layer using sterile single-use scalpels just before the actual subsampling of the working half was performed. We subsampled using sterile 3 mL plastic syringes, and samples were immediately frozen to avoid further DNA degradation.

Extraction and library preparation

All pre-PCR laboratory work was performed at the ancient DNA clean lab facilities at Globe Institute, University of Copenhagen, where strict precautions are taken to avoid contamination^115,116. A subset of samples was processed using the semi-automated ancient environmental DNA pipeline operated by the GeoGenetics Sequencing Core facility at Globe Institute. For these samples, 0.35 g of sediment were subsampled and extracted using the Qiagen® MagAttract® Power Soil Pro kit, with modifications. After extractions, DNA concentrations were quantified by qPCR and a fixed number of cycles for the library build was determined (17 cycles). The libraries were prepared following the double-stranded protocol¹¹⁸ and unique 10-base pair motifs were used for double-indexing. The amplified libraries were purified and size selected with MagBio beads, targeting 60–600 base pairs (bp; with a 1.6x ratio) and fragment lengths and concentrations were determined using a Fragment Analyzer (Agilent Technologies).

The remaining samples were processed manually using the Qiagen DNeasy® PowerSoil® Pro Kit following the product protocol with minor modifications: an input weight of 0.5–1.0 g sediment was used. We added 25 µL of DTT (1M) and 25 µL of protK (2 mg mL⁻¹) to the bead tube containing the beads and sediment sample. The vortex step was performed using a FastPrep-24™ 5G with 2 x 20s (4 m s⁻¹). Samples were incubated for approximately 24 h at 56℃. In the last step of the protocol, 35 µL of elution buffer was added followed by a 5-min incubation. This step was performed twice to yield a total extract volume of 70 µL.

Single stranded libraries were prepared from 1–14 ng of DNA (as determined by Qubit dsDNA HS assay) following the Santa Cruz Reaction protocol¹¹⁹. For the indexing, two unique 6-bp motifs were used for each library, to minimize the risk of cross-contamination during pooling for sequencing. Blanks were included in both workflows to investigate potential contamination. For shotgun sequencing, indexed libraries were pooled equimolarly (except for blanks which were included with 10% molarity) and sequenced by NovoGene UK on an Illumina NovaSeq 6000 platform.

Hybridization capture

Shotgun sequencing revealed a low number of mammalian sequences, thus we applied hybridization capture to enrich libraries specifically for mitochondrial DNA derived from marine mammals occurring in Greenland seasonally, as well as year-round (Table S3 & S4). In preparation of the panel design, which was conducted in collaboration with Daicel ArborBiosciences myBaits®, the first 80 nucleotides of each mitogenome were added to the end of each mitogenome sequence of interest. Next, runs of ambiguous nucleotides were changed to thymine nucleotides. Finally, to account for the circularity of mitochondrial genomes, a 4x tiling with 80 nt baits was achieved by starting a new bait every 20 nt, resulting in 10,254 unique baits.

The hybridization capture panel used in this study was originally designed for the retrieval of mitochondrial DNA from macrofossils, and therefore includes several species for which we did not expect to find DNA in marine sediments (e.g. polar bears, Ursus maritimus, which mostly live on sea ice and occur in low densities). At the same time, the panel did not include species occurring along the coasts of Greenland in large numbers (e.g. harp seals). However, by lowering the hybridization temperature to 55°C, we could facilitate the retrieval of DNA from closely related species with up to 25% genetic divergence (myBaits® User Manual (v 5.02)), while also accommodating for the expected DNA damage of marine sedimentary ancient DNA⁴⁷.

We followed the “High Sensitivity” protocol of the myBaits® User Manual (v 5.02) with the following additional modifications: the input mass of each indexed library into each capture reaction was 154 ng. Generally, higher inputs are possible (up to 12 µg per enrichment reaction) but we standardized the input mass to account for libraries with lower DNA concentrations across our sample set. The input volume of baits and water in the hybridization mix was adjusted to 1.1 µL and 4.4 µL, respectively. One round of capture was performed.

Bioinformatic Analysis

a. Shotgun sequencing

Raw sequencing data were processed using leeHom¹²⁰ to trim adapters and merge sequencing reads. Merged sequences were complexity-filtered, sequences shorter than 30 bp were discarded and exact duplicates and homopolymers were removed using sga¹²¹. For taxonomic assignment of the shotgun sequencing data, a reference index was constructed from the non-redundant NCBI nucleotide database (downloaded on 1st December 2023), the full NCBI RefSeq database (release 213), and a compilation of Arctic plant and animal genomes⁴⁵. Filtered sequences were taxonomically assigned using bowtie2¹²² and alignment files were compressed using compressbam of the metaDMG¹²³ suite.

The compressed files were parsed to metaDMG¹²³, which was used to filter alignments using a lowest-common-ancestor approach with a 95% sequence similarity threshold and to assess DNA damage. All downstream visualizations were performed in R¹¹¹. The taxonomic assignments of the shotgun sequencing data were additionally filtered by a minimum of 10 sequences per taxon and for Fig S7-S9, only taxa present in at least 2 samples across each core were visualized (for the full list of DNA detections, please refer to Data table 1). Furthermore, we excluded all taxonomic assignments to prokaryotes, thus only retaining detections of eukaryotes. We calculated relative sequence abundances for each taxon per sample by dividing the number of unique sequences by the number of quality-filtered read pairs, thus providing a representation of the sequencing effort and library complexity.

We visualized the mean proportion of nucleotide deaminations for all eukaryotic sequences over the last three nucleotide bases to provide a representation of DNA damage across sample ages and sediment cores (Fig. S1a-d).

b. Mitochondrial capture

The raw mitochondrial capture sequencing data were processed in the same manner as the shotgun sequencing data, with the following modifications: for taxonomic assignments, a reference database was constructed using all complete vertebrate mitochondrial genomes and the complete refseq database of mitochondrial sequences (downloaded from the NCBI databases on 17th Oct. 2023). To avoid inflating the database, duplicate reference sequences were removed, and only one Homo sapiens mitochondrial genome was retained. After the taxonomic assignment, an additional duplicate removal step was included, where sequences with the same start and end position were removed using samtools¹²⁴.

We also performed an analysis of sequence similarity on mitochondrial genomes to establish a sequence similarity threshold appropriate for the observed genetic divergence within the target group of Arctic marine mammals (Fig. S12). Using a low sequence similarity threshold (e.g. 95%) as part of the lowest-common ancestor inference for a target group with moderate genetic divergence between species (interspecific variation) and low genetic divergence within the same species (intraspecific variation) may underestimate the number of detected species, as sequences are classified on a higher taxonomic level (e.g. genus or family). In the same scenario, a high sequence similarity threshold (e.g. 100%) may overestimate the number of detected species based on single nucleotide mismatches deriving from sequencing errors or DNA damage¹²⁵. Specifically, we downloaded all publicly available mitochondrial genomes belonging to the subfamilies Phocinae, Balaenidae and Monodontidae (downloaded from the NCBI on 25th Mar. 2024).

After aligning and trimming the genomes, we generated thousands of sequence fragments of 50bp length from each genome. We calculated pairwise sequence similarities between the simulated fragments of mitochondrial genomes within the same species and between species. Within species, we observed that for any pair of mitochondrial genomes, 95% of the 50'mers showed sequence similarities between 98–100% whereas only 5% showed sequence similarities below 98% (Fig. S12). In contrast, between species of the same subfamily, we observed that for any pair of mitochondrial genomes, only 10% of the 50'mers showed sequence similarities above 98%, whereas 26% showed sequence similarities between 95–98%, and 64% showed sequence similarities below 95% (Fig. S12). We thus argue that a 2% threshold is appropriate for our taxonomic target group, to accommodate for the observed high sequence similarity within species and moderate sequence similarity between species.

The resulting taxonomic assignments were filtered using a minimum of 3 unique sequences per taxa and for Fig. 2–4, only species-level marine mammal detections were visualized (for the full list of DNA detections, please refer to Data table 2).

Paleoenvironmental and paleoclimatic data for comparison

a. Published records

For comparison of our DNA data on marine mammal distributions, we compiled existing records on surface and bottom/subsurface water conditions at each study site. Paleoceanographic proxy data was retrieved from the literature for the same cores when available, or else from published records in proximity to the cores.

For Melville Bay 26G, measurements of IP₂₅ (indicative of sea ice algal productivity), Brassica- and Dinosterol (indicative of primary phytoplankton productivity) were compiled from a nearby marine sediment core (GeoB19927-3)⁷⁸.

For Hall Basin 24PC and Lincoln Sea 12-GC, a Holocene air temperature reconstruction from the Agassiz ice cap²⁹ and IP₂₅ as well as Brassicasterol measurements from Lincoln Sea 12-GC ⁵⁶ were compiled. In addition, trends in δ¹³C values of the benthic foraminifera Cassidulina neoteretis and abundance of the benthic foraminifera Nonionellina iridea–indicative of the opening of Nares Strait–were compiled from HLY03-01-05GC⁵¹.

For North-East Greenland 73G, measurements of IP₂₅, Brassica- and Dinosterol were retrieved from PS93/025–2⁹⁹. Furthermore, we included the abundance of the two benthic foraminifera Cassidulina neoteretis and Pullenia bulloides–indicative of relatively warm, saline Atlantic water–previously published for North-East Greenland 73G⁵⁷.

b. Analysis of foraminifera assemblage (Melville Bay 26G)

For Melville Bay 26G, no foraminiferal data existed and we carried out the analyses for this study: approximately four grams of sediment per sample depth (each sample representing 1 cm of the core) were washed and wet-sieved using a 63 µm sieve. Next, all constituents > 63 µm were left in foraminifera storage solution (prepared using 300 ml of ethanol (96%), 700 ml of distilled water, and 1.5 grams of sodium carbonate) for ca. 30–40 minutes to dissolve any cohesive clay clusters that remained after sieving. Afterwards, all samples were washed once more with the foraminiferal storage solution before the foraminiferal assemblage was analyzed using an Olympus stereomicroscope. The following foraminiferal species were grouped to indicate influx of chilled Atlantic Water associated with the West Greenland Current^36,60: the calcareous species Cassidulina neoteretis, Cassidulina reniforme, Islandiella norcrossi and Pullenia osloensis, as well as the agglutinated species Adercotryma glomerata, Lagenammina difflugiformis, Reophax catella, Reophax pilulifer, Reophax fusiformis.

c. Statistical analysis

We performed a statistical analysis on the association between DNA detections and available paleoclimatic data (Fig. 2a-c, 3a & 4a-c) using correlation analyses and redundancy analysis (RDA) in R¹¹¹. First, we approximated air temperature anomaly, IP₂₅, foraminifera abundance, brassica- & dinosterol measurements for each sample age using linear interpolation, and standardized all interpolated values to have a mean of zero and a standard deviation of 1. We then calculated pairwise Spearman’s rank correlation coefficients for all species-level detections (derived from hybridization capture), family-level detections (derived from shotgun sequencing) and interpolated paleoenvironmental and paleoclimatic measurements. The resulting correlation matrix was filtered for significant (p < 0.1) correlations (Fig. S10).

In addition, we performed a Redundancy Analysis (RDA) on our data for the two sediment cores, for which the paleoenvironmental and paleoclimatic record was most complete (Melville Bay 26G and North-East Greenland 73G). Across samples of the two sediment cores, the contribution of each interpolated proxy measurement to explaining the diversity of the species-level detections based on hybridization capture was evaluated using a RDA followed by an ANOVA-like permutation test (Fig. S11).

Data Availability

A detailed description of the processing of raw sequencing files, metadata files, taxonomic count data for DNA detections (shotgun sequencing and hybridization capture) and foraminifera assemblages used in this study are available at https://github.com/slennart/HHA-sedaDNA. Raw sequencing data files are available upon request.

Funding

This study was supported by the Villum Fonden Young Investigator Programme (YIP+) grant no. 37352 to EDL, Independent Research Fund Denmark (IRFD) grant no. 9064-00039B to SR, and grant no. 0135-00165B to MSS. The study also received funding from the Danish Center for Marine Research (DCH) and the European Union’s Horizon 2020 research and innovation programme under grant agreement No. 846142 (POLARC) to RJ, and the Horizon Europe programme under grant agreement No. 101136480 (SEA-Quester) to MSS.

Author contributions

LS: conceptualization, formal analysis, visualization, writing (original draft), writing (review & editing);KN: investigation, formal analysis, writing (review & editing);FS: investigation, formal analysis, writing (review & editing); ABK: investigation, formal analysis, writing (review & editing); RJ: investigation, formal analyses, writing (review and editing); MSS: funding acquisition, resources, writing (review & editing); CP: investigation, resources, writing (review & editing); MO'R: investigation, resources, writing (review & editing); HHZ: investigation, methodology, writing (review & editing); SR: conceptualization, funding acquisition, resources, supervision, writing (review & editing);EDL: conceptualization, funding acquisition, resources, supervision, writing (review & editing).

Acknowledgements

We thank the captains, crew and scientific parties of NorthGreen 2017 Expedition onboard R/V Dana, ICAROS 2021 Expedition onboard HDMS Lauge Koch, and Ryder 2019 Expedition onboard R/V Oden (special thanks to Martin Jakobsson). Further, Christine Rømer is acknowledged for contributing to the micropalaeontological analysis. We also thank Mikkel Winther Pedersen and Benjamin Vernot for helpful discussions regarding the data analysis and the laboratory leaders and technicians of Globe Institute, University of Copenhagen.

Rantanen M et al (2022) The Arctic has warmed nearly four times faster than the globe since 1979. Commun Earth Environ 3:168
Yadav J, Kumar A, Mohan R (2020) Dramatic decline of Arctic sea ice linked to global warming. Nat Hazards 103:2617–2621
Jahn A, Holland MM, Kay JE (2024) Projections of an ice-free Arctic Ocean. Nat Rev Earth Environ 5:164–176
Brandt S, Wassmann P, Piepenburg D (2023) Revisiting the footprints of climate change in Arctic marine food webs: An assessment of knowledge gained since 2010. Front Mar Sci 10:1096222
Deb JC, Bailey SA (2023) Arctic marine ecosystems face increasing climate stress. Environ Rev 31:403–451
Heide-Jørgensen MP et al (2023) A regime shift in the Southeast Greenland marine ecosystem. Glob Change Biol 29:668–685
Wassmann P, Duarte CM, Agustí S, Sejr MK (2011) Footprints of climate change in the Arctic marine ecosystem. Glob Change Biol 17:1235–1249
Chambault P et al (2022) Future seasonal changes in habitat for Arctic whales during predicted ocean warming. Sci Adv 8:eabn2422
Szuwalski CS, Aydin K, Fedewa EJ, Garber-Yonts B, Litzow MA (2023) The collapse of eastern Bering Sea snow crab. Science 382:306–310
Polyakov IV et al (2017) Greater role for Atlantic inflows on sea-ice loss in the Eurasian Basin of the Arctic Ocean. Science 356:285–291
Polyakov IV et al (2023) Fluctuating Atlantic inflows modulate Arctic atlantification. Science 381:972–979
Albouy C et al (2020) Global vulnerability of marine mammals to global warming. Sci Rep 10:548
Laidre K et al (2008) Quantifying the sensitivity of Arctic marine mammals to climate-induced habitat change. Ecol Appl 18:S97–S125
Andersen JM, Wierma YF, Stenson G, Hammill MO, Rosing-Asvid A (2009) Movement Patterns of Hooded Seals (Cystophora cristata) in the Northwest Atlantic Ocean During the Post-Moult and Pre-Breed Seasons. J Northwest Atl Fish Sci 42:1–11
Pierce GJ (1992) Harp seals, man and ice. Rev Fish Biol Fish 2:270–271
Hansen RG et al (2018) Abundance of Whales in West and East Greenland in Summer 2015. NAMMCO Sci Publ 11
Ugarte F, Rosing-Asvid A, Heide-Jørgensen MP, Laidre KL (2020) Marine Mammals of the Greenland Seas. in Encyclopedia of the World’s Biomes 575–586Elsevier, 10.1016/B978-0-12-409548-9.12485-6
Moore SE, Reeves RR (2018) Tracking arctic marine mammal resilience in an era of rapid ecosystem alteration. PLOS Biol 16:e2006708
George JC, Moore SE, Thewissen JGM (2020) North Slope Borough (Alaska). Department of Wildlife Management, & University of Washington. Center for Ecosystem Studies. Arctic Report Card 2020: Bowhead whales: recent insights into their biology, status and resilience. 10.25923/CPPM-N265
Lewis KM, Van Dijken GL, Arrigo K (2020) R. Changes in phytoplankton concentration now drive increased Arctic Ocean primary production. Science 369:198–202
Breed GA (2021) Predators and impacts of predation. in The Bowhead Whale 457–470Elsevier, 10.1016/B978-0-12-818969-6.00029-7
Heide-Jørgensen MP et al (2010) The effect of sea-ice loss on beluga whales (Delphinapterus leucas) in West Greenland. Polar Res 29:198–208
Moore SE, Huntington HP (2008) Arctic marine mammals and climate change: impacts and resilience. Ecol Appl 18:S157–S165
Shuert CR et al (2022) Decadal migration phenology of a long-lived Arctic icon keeps pace with climate change. Proc. Natl. Acad. Sci. 119, e2121092119
Nogués-Bravo D et al (2018) Cracking the Code of Biodiversity Responses to Past Climate Change. Trends Ecol Evol 33:765–776
Foote AD et al (2013) Ancient DNA reveals that bowhead whale lineages survived Late Pleistocene climate change and habitat shifts. Nat Commun 4:1677
Louis M et al (2020) Influence of past climate change on phylogeography and demographic history of narwhals, Monodon monoceros. Proc. R. Soc. B Biol. Sci. 287, 20192964
Westbury MV et al (2023) Impact of Holocene environmental change on the evolutionary ecology of an Arctic top predator. Sci Adv 9:eadf3326
Lecavalier BS et al (2017) High Arctic Holocene temperature record from the Agassiz ice cap and Greenland ice sheet evolution. Proc. Natl. Acad. Sci. 114, 5952–5957
Axford Y, De Vernal A, Osterberg EC (2021) Past Warmth and Its Impacts During the Holocene Thermal Maximum in Greenland. Annu Rev Earth Planet Sci 49:279–307
Lesnek AJ, Briner JP, Young NE, Cuzzone JK (2019) Maximum Southwest Greenland Ice Sheet Recession in the Early Holocene. Geophys. Res. Lett. 47, eGL083164 (2020)
O’Regan M et al (2021) The Holocene dynamics of Ryder Glacier and ice tongue in north Greenland. Cryosphere 15:4073–4097
Reilly BT et al (2019) Holocene break-up and reestablishment of the Petermann Ice Tongue, Northwest Greenland. Quat Sci Rev 218:322–342
England JH et al (2008) A millennial-scale record of Arctic Ocean sea ice variability and the demise of the Ellesmere Island ice shelves. Geophys Res Lett 35:2008GL034470
Funder S et al (2011) A 10,000-Year Record of Arctic Ocean Sea-Ice Variability—View from the Beach. Science 333:747–750
Perner K, Moros M, Jennings A, Lloyd J, Knudsen K (2013) Holocene palaeoceanographic evolution off West Greenland. Holocene 23:374–387
Crees JJ, Collen B, Turvey ST (2019) Bias, incompleteness and the ‘known unknowns’ in the Holocene faunal record. Philos Trans R Soc B Biol Sci 374:20190216
Bennike O, Wagner B (2021) Quaternary Vertebrates from the North Atlantic Islands. in Biogeography in the Sub-Arctic (eds. Panagiotakopulu, E. & Sadler, J. P.) 147–160Wiley, 10.1002/9781118561461.ch7
Bennike O (1997) Quaternary vertebrates from Greenland: A review. Quat Sci Rev 16:899–909
Bennike O, Andreasen C (2007) Radiocarbon dating of walrus (Odobenus rosmarus) remains from Greenland. Polar Rec 43:361–365
Hjort C (1997) Glaciation, climate history, changing marine levels and the evolution of the Northeast Water polynya. J Mar Syst 10:23–33
Seppä H, Seidenkrantz M-S, Caissie B (2023) Macias Fauria, M. Polar bear’s range dynamics and survival in the Holocene. Quat Sci Rev 317:108277
Willerslev E et al (2003) Diverse Plant and Animal Genetic Records from Holocene and Pleistocene Sediments. Science 300:791–795
Haile J et al (2009) Ancient DNA reveals late survival of mammoth and horse in interior Alaska. Proc. Natl. Acad. Sci. 106, 22352–22357
Wang Y et al (2021) Late Quaternary dynamics of Arctic biota from ancient environmental genomics. Nature 600:86–92
CLIMAP Project Members (1981) Relative abundance of planktic foraminifera in the 120 kyr time slice reconstruction of sediment core GIK12392-1. https://doi.org/10.1594/PANGAEA.51983. 726 data points [object Object]
Armbrecht L et al (2019) Ancient DNA from marine sediments: Precautions and considerations for seafloor coring, sample handling and data generation. Earth-Sci Rev 196:102887
Nguyen N-L et al (2023) Sedimentary ancient DNA: a new paleogenomic tool for reconstructing the history of marine ecosystems. Front Mar Sci 10:1185435
Hansen KE, Giraudeau J, Wacker L, Pearce C, Seidenkrantz M-S (2020) Reconstruction of Holocene oceanographic conditions in eastern Baffin Bay. Clim Past 16:1075–1095
Jackson R et al (2021) Holocene polynya dynamics and their interaction with oceanic heat transport in northernmost Baffin Bay. Sci Rep 11:10095
Jennings A et al (2011) The Holocene History of Nares Strait: Transition from Glacial Bay to Arctic-Atlantic Throughflow. Oceanography 24:26–41
Limoges A et al (2020) Learning from the past: Impact of the Arctic Oscillation on sea ice and marine productivity off northwest Greenland over the last 9,000 years. Glob Change Biol 26:6767–6786
Seidenkrantz M-S et al (2007) Hydrography and climate of the last 4400 years in a SW Greenland fjord: implications for Labrador Sea palaeoceanography. Holocene 17:387–401
Armbrecht L et al (2022) Ancient marine sediment DNA reveals diatom transition in Antarctica. Nat Commun 13:5787
Zimmermann HH et al (2023) Marine ecosystem shifts with deglacial sea-ice loss inferred from ancient DNA shotgun sequencing. Nat Commun 14:1650
Detlef H et al (2023) Seasonal sea-ice in the Arctic’s last ice area during the Early Holocene. Commun Earth Environ 4:86
Pados-Dibattista T, Pearce C, Detlef H, Bendtsen J, Seidenkrantz M-S (2022) Holocene palaeoceanography of the Northeast Greenland shelf. Clim Past 18:103–127
Jennings A et al (2022) Modern and early Holocene ice shelf sediment facies from Petermann Fjord and northern Nares Strait, northwest Greenland. Quat Sci Rev 283:107460
Blaauw M (2010) Methods and code for ‘classical’ age-modelling of radiocarbon sequences. Quat Geochronol 5:512–518
Perner K et al (2011) Centennial scale benthic foraminiferal record of late Holocene oceanographic variability in Disko Bugt, West Greenland. Quat Sci Rev 30:2815–2826
Lomac-MacNair K et al (2018) Seal Occurrence and Habitat Use during Summer in Petermann Fjord, Northwestern Greenland. ARCTIC 71
Søndergaard AS et al (2020) Early Holocene collapse of marine-based ice in northwest Greenland triggered by atmospheric warming. Quat Sci Rev 239:106360
Bennike O, Higgins AK, Kelly M (1989) Mammals of central North Greenland. Polar Rec 25:43–49
Knudsen KL, Stabell B, Seidenkrantz M, Eiríksson J, Blake W (2008) Deglacial and Holocene conditions in northernmost Baffin Bay: sediments, foraminifera, diatoms and stable isotopes. Boreas 37:346–376
Smith TG, Hammill MO, Taugbøl GA (1991) Review of the Developmental, Behavioural and Physiological Adaptations of the Ringed Seal, Phoca hispida, to Life in the Arctic Winter. Arctic 44:124–131
Heide-Jørgensen M (1988) Occurrence and hunting of killer whales in Greenland. Rit Fiskid 11:115–135
Hobbs RC et al (2019) Global Review of the Conservation Status of Monodontid Stocks. Mar Fish Rev 81
Evans DJA (1989) An early Holocene narwhal tusk from the Canadian high Arctic. Boreas 18:43–50
Bennike O (2008) An early Holocene Greenland whale from Melville Bugt, Greenland. Quat Res 69:72–76
Bednarski J (1990) An Early Holocene Bowhead Whale (Balaena Mysticetus) in Nansen Sound, Canadian Arctic Archipelago. Arctic 43:50–54
Dyke AS, Hooper J, Savelle JM (1996) A History of Sea Ice in the Canadian Arctic Archipelago Based on Postglacial Remains of the Bowhead Whale (Balaena Mysticetus). Arctic 49:235–255
Nielsen J (2011) Photo-id of killer whales (Orcinus orca) in Greenland waters
Heide-Jørgensen MP et al (2010) Abundance of fin whales in West Greenland in 2007. J Cetacean Res Manage 11:83–88
NAMMCO (2023) Fin Whale
Folkow LP, Nordøy ES, Blix AS (2004) Distribution and diving behaviour of harp seals (Pagophilus groenlandicus) from the Greenland Sea stock. Polar Biol 27:281–298
Levac E, Vernal AD, Blake W Jr (2001) Sea-surface conditions in northernmost Baffin Bay during the Holocene: palynological evidence. J Quat Sci 16:353–363
Belt ST, Müller J (2013) The Arctic sea ice biomarker IP25: a review of current understanding, recommendations for future research and applications in palaeo sea ice reconstructions. Quat Sci Rev 79:9–25
Saini J et al (2020) Holocene variability in sea ice and primary productivity in the northeastern Baffin Bay. arktos 6, 55–73
Fratantoni PS, Pickart RS (2007) The Western North Atlantic Shelfbreak Current System in Summer. J Phys Oceanogr 37:2509–2533
Munk P (2003) Changes in plankton and fish larvae communities across hydrographic fronts off West Greenland. J Plankton Res 25:815–830
Funder S, Weidick A (1991) Holocene boreal molluscs in Greenland — palaeoceanographic implications. Palaeogeogr Palaeoclimatol Palaeoecol 85:123–135
Myers PG et al (2021) Extreme High Greenland Blocking Index Leads to the Reversal of Davis and Nares Strait Net Transport Toward the Arctic Ocean. Geophys. Res. Lett. 48, eGL094178 (2021)
Laidre K, Heide-Jørgensen MP (2012) Spring partitioning of Disko Bay, West Greenland, by Arctic and Subarctic baleen whales. ICES J Mar Sci 69:1226–1233
Laidre K et al (2010) Spatial associations between large baleen whales and their prey in West Greenland. Mar Ecol Prog Ser 402:269–284
Ribeiro S et al (2021) Vulnerability of the North Water ecosystem to climate change. Nat Commun 12:4475
Seersholm FV et al (2016) DNA evidence of bowhead whale exploitation by Greenlandic Paleo-Inuit 4,000 years ago. Nat Commun 7:13389
Seersholm FV et al (2022) Ancient DNA provides insights into 4,000 years of resource economy across Greenland. Nat Hum Behav 6:1723–1730
Kobashi T et al (2011) High variability of Greenland surface temperature over the past 4000 years estimated from trapped air in an ice core. Geophys Res Lett 38
McLeod BA et al (2009) Bowhead whales, and not right whales, were the primary target of 16th- to 17th-century Basque Whalers in the Western North Atlantic. Arctic 61:61
Heikkilä M, Ribeiro S, Weckström K, Pieńkowski AJ (2022) Predicting the future of coastal marine ecosystems in the rapidly changing Arctic: The potential of palaeoenvironmental records. Anthropocene 37:100319
Georgiadis E et al (2020) Local and regional controls on Holocene sea ice dynamics and oceanography in Nares Strait, Northwest Greenland. Mar Geol 422:106115
Carlyle CG et al (2021) Marine mammal biodiversity and rare narwhal (Monodon monoceros) observations near northern Ellesmere Island, Canada. Ecosphere 12, e03534
Heide-Jørgensen MP, Box JE, Hansen RG, Jakobsson M (2024) Evidence of a narwhal (Monodon monoceros) summer ground in Nares Strait. Polar Res 43
Peary RE (1910) The North Pole. Hodder Stoughton
Whitney CK, Cheung WWL, Ban NC (2023) Considering the implications of climate-induced species range shifts in marine protected areas planning. FACETS 8:1–10
Jakobsson M et al (2018) The Holocene retreat dynamics and stability of Petermann Glacier in northwest Greenland. Nat Commun 9:2104
Rudels B, Friedrich J, H., Quadfasel D (1999) The Arctic Circumpolar Boundary Current. Deep Sea Res Part II Top Stud Oceanogr 46:1023–1062
Rudels B et al (2005) The interaction between waters from the Arctic Ocean and the Nordic Seas north of Fram Strait and along the East Greenland Current: results from the Arctic Ocean-02 Oden expedition. J Mar Syst 55:1–30
Syring N et al (2020) Holocene changes in sea-ice cover and polynya formation along the eastern North Greenland shelf: New insights from biomarker records. Quat Sci Rev 231:106173
Boertmann D, Merkel F, Durinck J (2009) Bowhead whales in East Greenland, summers 2006–2008. Polar Biol 32:1805–1809
Kovacs KM et al (2020) The endangered Spitsbergen bowhead whales’ secrets revealed after hundreds of years in hiding. Biol Lett 16:20200148
Borge T, Bachmann L, Bjørnstad G, Wiig Ø (2007) Genetic variation in Holocene bowhead whales from Svalbard. Mol Ecol 16:2223–2235
Farnsworth WR et al (2020) Holocene glacial history of Svalbard: Status, perspectives and challenges. Earth-Sci Rev 208:103249
Larsen NK et al (2022) Late glacial and Holocene glaciation history of North and Northeast Greenland. Arct Antarct Alp Res 54:294–313
Håkansson S (1976) University of Lund radiocarbon dates IX
Bennike O, Weidick A (2001) Late Quaternary history around Nioghalvfjerdsfjorden and Jøkelbugten, North-East Greenland. Boreas 30:205–227
Beszczynska-Möller A, Fahrbach E, Schauer U, Hansen E (2012) Variability in Atlantic water temperature and transport at the entrance to the Arctic Ocean, 1997–2010. ICES J Mar Sci 69:852–863
Heide-Jørgensen MP et al (2013) The Significance of the North Water Polynya to Arctic Top Predators. Ambio 42:596–610
Andersen TJ (2017) Some Practical Considerations Regarding the Application of 210Pb and 137Cs Dating to Estuarine Sediments. In: Weckström K, Saunders KM, Gell PA, Skilbeck CG (eds) Applications of Paleoenvironmental Techniques in Estuarine Studies, vol 20. Springer Netherlands, Dordrecht, pp 121–140
Mollenhauer G, Grotheer H, Gentz T, Bonk E, Hefter J (2021) Standard operation procedures and performance of the MICADAS radiocarbon laboratory at Alfred Wegener Institute (AWI), Germany. Nucl Instrum Methods Phys Res Sect B Beam Interact Mater At 496:45–51
R Core Team. R: A Language and Environment for Statistical Computing (2022)
Blaauw M, Christen JA (2011) Flexible paleoclimate age-depth models using an autoregressive gamma process. Bayesian Anal 6
Heaton TJ et al (2020) Marine20—The Marine Radiocarbon Age Calibration Curve (0–55,000 cal BP). Radiocarbon 62:779–820
Pearce C, Özdemir KS, Mathiasen F, Detlef R, H., Olsen J (2023) The marine reservoir age of Greenland coastal waters. Geochronology 5:451–465
Gilbert MTP, Bandelt H-J, Hofreiter M, Barnes I (2005) Assessing ancient DNA studies. Trends Ecol Evol 20:541–544
Willerslev E, Cooper A, Review Paper. Ancient DNA (2005) Proc. R. Soc. B Biol. Sci. 272, 3–16
Laine J et al (2024) Late Pleistocene stickleback environmental genomes reveal the chronology of freshwater adaptation. Curr Biol 34:1142–1147e6
Kircher M, Sawyer S, Meyer M (2012) Double indexing overcomes inaccuracies in multiplex sequencing on the Illumina platform. Nucleic Acids Res 40:e3–e3
Kapp JD, Green RE, Shapiro BA (2021) Fast and Efficient Single-stranded Genomic Library Preparation Method Optimized for Ancient DNA. J Hered 112:241–249
Renaud G, Stenzel U, Kelso J (2014) leeHom: adaptor trimming and merging for Illumina sequencing reads. Nucleic Acids Res 42:e141–e141
Simpson JT, Durbin R (2012) Efficient de novo assembly of large genomes using compressed data structures. Genome Res 22:549–556
Langmead B, Salzberg SL (2012) Fast gapped-read alignment with Bowtie 2. Nat Methods 9:357–359
Michelsen C et al (2022) metaDMG – A Fast and Accurate Ancient DNA Damage Toolkit for Metagenomic Data. http://biorxiv.org/lookup/doi/10.1101/2022.12.06.519264 doi:10.1101/2022.12.06.519264
Li H et al (2009) The Sequence Alignment/Map format and SAMtools. Bioinformatics 25:2078–2079
Wang Y, Korneliussen TS, Holman LE, Manica A, Pedersen MW (2022) ngsLCA—A toolkit for fast and flexible lowest common ancestor inference and taxonomic profiling of metagenomic data. Methods Ecol Evol 13:2699–2708
Fetterer F, Knowles K, Meier W, Savoie M, Windnagel A (2017) Sea Ice Index, Version 3. NSIDC https://doi.org/10.7265/N5K072F8
Jennings A et al (2011) The Holocene History of Nares Strait: Transition from Glacial Bay to Arctic-Atlantic Throughflow. Oceanography 24:26–41

There is NO Competing Interest.

SupplementNCOMMS2467262.docx
rs.pdf
Reporting Summary
DatasupplementNCOMMS2467262.xlsx
Data tables 1 - 4

Download PDF

Version 1

posted

You are reading this latest preprint version

Holocene shifts in marine mammal distributions around Northern Greenland revealed by sedimentary ancient DNA

Status:

Version 1

Abstract

Figures

Introduction

Results

Overview of sequencing data and DNA damage

Age-depth models of the studied marine sediment cores

Shotgun sequencing data

Melville Bay 26G

Hall Basin 24PC and Lincoln Sea 12-GC

North-East Greenland 73G

DNA detections in the context of changing paleoenvironmental conditions

Discussion

Ecosystem-wide changes in Melville Bay during the Holocene

The opening of the Nares Strait

Episodic marine mammal detections in north-east Greenland over the past 10 ka

Conclusions

Methods

Sediment cores

Subsampling of cores for sedaDNA analysis

Extraction and library preparation

Hybridization capture

a. Shotgun sequencing

b. Mitochondrial capture

a. Published records

b. Analysis of foraminifera assemblage (Melville Bay 26G)

c. Statistical analysis

Declarations

Data Availability

Funding

Author contributions

Acknowledgements

References

Additional Declarations

Supplementary Files

Status:

Version 1