Fig. 1: Overview of river sampling sites and corresponding methane concentrations.![\"Fig.](\"https:\/\/www.newsbeep.com\/ie\/wp-content\/uploads\/2026\/09\/41467_2026_77190_Fig1_HTML.png\")
<\/p>\n
Geological map of central Svalbard (see inset map for location) with the surveyed glacial catchments outlined in black. Methane concentrations (nM) of river samples are displayed by bubble size over aerial or satellite imagery. The size scale applies equally to all transects. Bubble color denotes the carbon isotopic composition of the methane (\u03b413C-CH4, \u2030), with blue indicating a typical microbial signature (<\u221250\u2030), red indicating a typical thermogenic signature (>\u221250\u2030), and yellow indicating that no isotopic data is available. River flowpaths are drawn in light blue. Aerial imagery \u00a9 Norwegian Polar Institute, satellite imagery from Esri World Imagery (sources: Esri, Vantor) and Planet Labs (source: Copernicus Sentinel). The geology basemap is adapted from data sets available at https:\/\/geodata.npolar.no\/<\/a> with permission under a Creative Commons license CC BY 4.0.<\/p>\n Largely thermogenic methane is supplemented by instream methanogenesis<\/p>\n The carbon isotopic compositions (\u03b413C\u2013CH4) of methane emerging with meltwaters at the glacier termini predominantly fall within the thermogenic range. Microbial methane, produced through the microbially-mediated breakdown of organic matter, typically exhibits \u03b413C values between \u2212110 and \u221250\u203027<\/a>. In contrast, thermogenic methane, generated abiotically through the breakdown of organic matter at high temperatures and pressures, is less depleted in 13C, with \u03b413C values ranging from \u221250 to \u221220\u203027<\/a>. Among the 19 glacial rivers studied, only three exhibit isotopic signatures indicative of microbially produced methane (\u03b413C\u2009<\u2009\u221250\u2030) at the glacier terminus, suggesting that the methane transported by meltwaters is predominantly thermogenic in origin.<\/p>\n It is important to note that the isotopic composition of methane can be altered by microbial oxidation, or methanotrophy, a process which has been found to occur in the subglacial environments on Greenland28<\/a>,29<\/a>. Methanotrophy enriches the remaining methane in 13C, causing its isotopic signature to shift toward the thermogenic range. However, we suspect that the thermogenic signatures observed in our study are not primarily the result of subglacial methane oxidation, but instead reflect the abundant sources of thermogenic methane present in Svalbard\u2019s organic-rich geology.<\/p>\n The geology of Svalbard contains extensive hydrocarbon systems30<\/a>,31<\/a> and numerous thick methane-rich source rocks, including coal seams and shale32<\/a>,33<\/a>. Natural seepages of this methane are found both onshore and offshore Svalbard2<\/a>,34<\/a>,35<\/a>,36<\/a>. Ethane and propane concentrations measured in our river samples yield wetness (C1\/(C2\u2009+\u2009C3)) levels ranging from 3.7 to 91.4 (n\u2009=\u200961). When coupled with the carbon isotopic signatures, these wetness levels indicate that the methane originates from oil-associated, early mature or late mature thermogenic gas37<\/a> (Fig.\u00a02<\/a>). These findings are consistent with previous geochemical observations of gaseous hydrocarbons liberated from rocks in the same geological settings on Svalbard, where gases were found to be largely of thermogenic origin and varying degrees of maturity33<\/a>.<\/p>\n Fig. 2: Wetness levels of river water plotted against stable carbon isotopic signatures of methane. \u03b413C\u2013CH4 (\u2030) versus wetness (C1\/(C2\u2009+\u2009C3)) of river transect samples (circles, n\u2009=\u200961 from 12 of our rivers) plotted on top of a genetic diagram adapted from ref. 37<\/a>. The color of the circle is representative of the dominant geology of the catchment: (1) Paleocene and Eocene, (2) Early Cretaceous, (3) Triassic-Middle Jurassic, (4) Carboniferous and Permian, (5) Pre-Devonian Basement. CR: CO2 reduction, F: methyl-type fermentation, SM: secondary microbial, EMT: early mature thermogenic gas, OA: oil-associated thermogenic gas, LMT: late mature thermogenic gas. Black arrows indicate the direction the wetness and\/or isotopic compositions would shift due to methane oxidation or gas migration. The white box represents the region in which all our measured river samples are plotted.<\/p>\n Our findings are in contrast to those from glacial rivers in Greenland, where crystalline bedrock dominates, and the methane in meltwaters is of microbial origin, produced subglacially through the breakdown of overridden vegetation16<\/a>,19<\/a>,21<\/a>,29<\/a>. While subglacial environments on Svalbard also presumably offer the conditions and substrate necessary for microbial methane production38<\/a>, the abundance of thermogenic methane from organic-rich bedrock overwhelmingly dominates in most catchments. During glacial erosion, the grinding of organic-rich bedrock liberates thermogenic methane, which accumulates in the subglacial environment. Glacial grinding has been found to liberate methane at a greater rate than microbial methane is generated in the same subglacial sediments39<\/a>.<\/p>\n For example, a study of the Vall\u00e5krabreen glacier on Svalbard found that its meltwaters transport up to six times more methane per glacier area during the melt season than the meltwaters of Leverett Glacier in Greenland18<\/a>. This difference arises because Leverett lies on crystalline bedrock, which is unsuitable for hydrocarbon storage, and its meltwaters contain primarily subglacially-produced microbial methane16<\/a>. In contrast, Vall\u00e5krabreen\u2019s subglacial waters flush thermogenic methane from fractures in the underlying organic-rich shale, resulting in greater methane mobilization18<\/a>. Additionally, hydraulic pressure during periods of high subglacial water flow may contribute to rock fracturing40<\/a>,41<\/a>, further enhancing methane release. These natural processes of erosion and fracturing are highly effective at mobilizing methane in subglacial environments and are likely common in glaciers overriding organic-rich rocks such as shale\u2014environments that are widespread across many parts of the Arctic.<\/p>\n Our transects generally show a downstream decrease in methane concentrations along the rivers, likely due primarily to outgassing to the atmosphere. However, we also observe frequent additional recharge of methane along the river flowpaths. Abrupt increases in methane concentration downstream from the glacier terminus, such as those observed in the Gl\u00f8ttfjellebreen (2) and Vestre Gr\u00f8nfjordbreen (5) rivers (Fig.\u00a03<\/a>), are likely caused by discrete inputs of methane-rich groundwaters, which are commonly found in glacier forefields and can be from glacial or non-glacial sources2<\/a>,18<\/a>,42<\/a>,43<\/a>,44<\/a>. Furthermore, isotopic changes along some rivers suggest that, while methane is being lost to the atmosphere through outgassing, it is also being replenished by the microbial production of methane (methanogenesis) in riverbed sediments. During the microbial breakdown of organic matter into methane, kinetic isotopic fractionation enriches the pool of methane in lighter carbon isotopes (12C)27<\/a>. In contrast, isotopic fractionation during gas transfer from water to air is negligible, especially in turbulent conditions45<\/a>. The isotopic enrichment in 12C is evident in some rivers, where downstream isotopic compositions become progressively more negative (e.g., Vall\u00e5krabreen (16) and Mettebreen (17) rivers, Fig.\u00a03<\/a>), suggesting a small, continuous source of microbial methane. Additional evidence for instream methanogenesis comes from an incubation experiment with meltwater from Svalbreen (12), which yielded a\u2009>\u200940% increase in methane concentration after 30\u2009h (Supplementary Fig.\u00a01<\/a>).<\/p>\n Fig. 3: Transects of methane concentration and carbon isotopic composition. Methane concentrations ([CH4], nM) and the carbon isotopic compositions of methane (\u03b413C\u2013CH4, \u2030) are plotted in transects along the length of each river. Glacier catchments are numbered according to the labels in Fig.\u00a01<\/a>. The \u03b413C\u2013CH4 is colored on a scale according to its value, with blue indicating the typical microbial range (<\u221250\u2030) and red indicating the thermogenic range (>\u221250\u2030). Each plot uses the same color scale denoted in the legend.<\/p>\n In contrast, we observe the opposite isotopic trend in some rivers, where the methane becomes progressively enriched in 13C, and thus the \u03b413C\u2013CH4 values become less negative along the flowpath (e.g. in the second half of the Gl\u00f8ttfjellbreen (2) and Penckbreen (11) rivers, Fig.\u00a03<\/a>). This is suggestive of microbial methanotrophy, a process in which methane is consumed and oxidized into carbon dioxide under aerobic conditions. Like methanogenesis, methanotrophy is associated with kinetic isotopic fractionation, which enriches the residual methane in heavier carbon isotopes (13C)27<\/a>. Methanotrophy acts as a methane sink and has the potential to moderate methane emissions in glacial environments20<\/a>,28<\/a>,46<\/a>,47<\/a>. We have evidence of active methanotrophy in groundwater streams feeding into the glacial rivers, supported by an incubation study (Supplementary Fig.\u00a02<\/a>) and previous isotopic analyses18<\/a>. Thus, it is likely that some methanotrophy occurs within these glacial rivers. However, the highly turbulent flow of the rivers promotes rapid methane outgassing48<\/a>, which likely dominates over rate-limited methanotrophy.<\/p>\n Subglacial geology and thermal conditions determine meltwater methane concentrations<\/p>\n The glaciers selected for this study represent a range of sizes (0.56-110\u2009km2), geological settings (Fig.\u00a01<\/a>), and thermal regimes. Our data reveal that the bedrock geology beneath the glaciers controls the potential for the melt river to contain high concentrations of methane. Statistical analysis shows that geology, when categorized into five units (as shown in Fig.\u00a04<\/a>), significantly predicts the log-transformed methane concentrations of upstream samples (analysis of variance, F4,13\u2009=\u20093.50, P\u2009=\u20090.038, n\u2009=\u200918) and explains more than half of the variance in measured concentrations (R2\u2009=\u20090.52, n\u2009=\u200918). Glaciers overlying shale-bearing geological formations, principally the Middle Triassic, Late Jurassic, and Early Cretaceous units on Svalbard, have the highest potential for methane-rich meltwaters emerging from their termini (Fig.\u00a04<\/a>).<\/p>\n Fig. 4: Basal temperate ice fractions and meltwater methane concentrations. Methane concentrations (nM) of the most upstream sample of each glacial river are plotted according to the main geological unit on which the glacier lies. Bubble size indicates the a, basal temperate ice fraction (n\u2009=\u200918), or the fraction of the bed that has temperate ice present or b, the area of the glacier in km2 (n\u2009=\u200922). The number of glaciers differs between the plots because GPR data was not available to calculate the basal temperate ice fraction for one glacier (Hessbreen (9)). Three glaciers (from the Ny-\u00c5lesund region on Svalbard) were included in addition to the 19 glaciers in b, but these three glaciers were not included in the wider study because transect sampling was not conducted.<\/p>\n However, the extent of methane supersaturation in the glacial outflows appears to be additionally governed by how hydrologically active the glacier bed is. We use the fraction of temperate ice (ice at or above its pressure melting point) at the base of the glacier (Fig.\u00a04<\/a>) as a proxy for how much of the bed is thawed and therefore hydrologically active. The presence of temperate ice at the glacier bed is strongly indicative of a thawed bed that is hydrologically connected to the drainage system. Glaciers with no temperate ice can be assumed with high certainty to be frozen to their bed, and thus there is little hydrological contact with geology11<\/a>,12<\/a>,13<\/a>,14<\/a>. When the basal temperature ice fraction is added to a statistical model and is considered in addition to the geology, they together explain three-quarters of the variance of log-transformed methane concentrations (analysis of covariance, F5,12\u2009=\u20097.35, P\u2009=\u20090.0023, R2\u2009=\u20090.75, n\u2009=\u200918). This indicates that the basal temperate ice fraction explains roughly one-quarter of the variance beyond geology.<\/p>\n Our findings indicate that the bed must be thawed and hydrologically active where the glacier meets the shale-rich geology in order for the meltwater to effectively acquire methane. Thus, meltwaters flowing from a glacier above shale-rich rocks will likely contain high methane concentrations if temperate ice is present and the bed is thawed (e.g., Ragna Mariebreen (18) (1700\u2009nM) in Fig.\u00a05<\/a>), but low concentrations if there is little or no temperate ice (ie., Gl\u00f8ttfjellbreen (2) (35\u2009nM) in Fig.\u00a05<\/a>).<\/p>\n Fig. 5: Glacier cross-sections and subglacial geology. Cross-sections of a selection of glaciers from this study with modeled subglacial geology and temperate ice zones measured by ground-penetrating radar (GPR)66<\/a>. The lines in the geological units represent the dip angle. The Middle Triassic, Late Jurassic, and Early Cretaceous units contain abundant organic-rich shale layers. More detailed geological descriptions of each unit can be found in the Supplementary Information. Neighboring glaciers of similar sizes and geological settings were used as proxies in two instances when GPR surveys were not possible on the study glacier: M\u00f8ysalbreen was used to represent Gl\u00f8ttfjellbreen (2), and Filantropbreen was used to represent Nobelbreen (14).<\/p>\n Thawed areas of the glacier bed can support methane supersaturation through several mechanisms. First, larger zones of unfrozen glacier bed enable greater water flow, supporting a more extensive subglacial drainage system that can flush methane stored in rocks, sediments and cavities below18<\/a>. Second, methane-rich groundwaters may emerge at hydrologically active zones at the glacier bed and integrate into the subglacial drainage system2<\/a>. Third, saturated, anoxic basal sediments provide ideal conditions for methanogenesis if any organic substrates are available49<\/a>. On Svalbard, the first two mechanisms are evidently more significant, as we observe that glaciers with hydrologically active beds situated over geology lacking organic-rich strata\u2014such as carbonate or crystalline bedrocks\u2014exhibit low methane concentrations in their outflows, despite offering suitable conditions for methanogenesis. For example, Fig.\u00a05<\/a> shows that Von Postbreen (1) and Renardbreen (8) are largely or entirely hydrologically active at their beds, yet their outflows contain relatively low methane concentrations (72\u2009nM and 17\u2009nM, respectively) because they lie on metamorphosed basement rocks.<\/p>\n Our findings suggest that as glacier melt continues to accelerate worldwide50<\/a>, larger volumes of meltwater will be produced, increasing the amount of water reaching glacier beds and enhancing the flux of methane mobilized from subglacial environments. We also find evidence that glacier surge events, which are becoming more frequent on Svalbard51<\/a>, may increase subglacial methane mobilization. Three of our rivers with the highest methane concentrations were actively surging or had recently finished a surge at the time of sampling (Penckbreen (11): 825\u2009nM, Vall\u00e5krabreen (16): 477\u2009nM, Bakaninbreen (19): 857\u2009nM). Sapper et al. suggested that subglacial methane production could be enhanced by glacier surging in Yukon, Canada, due to the remobilization of sediments supplying fresh organic material17<\/a>. Here, we suggest that increased basal water pressures and expansion of the hydrologically active zone due to glacier surging52<\/a>,53<\/a> enhances the flushing of subglacial methane stores.<\/p>\n On Svalbard, observations indicate that as valley glaciers recede and thin in a warming climate, they undergo a thermal transition from polythermal to cold-based systems54<\/a>. During this transition, the proportion of thawed bed decreases until the glacier becomes entirely frozen to its bed55<\/a>,56<\/a>. As a result, methane fluxes from glacier meltwaters are likely to increase for some time, driven by enhanced subglacial hydrological activity, but may eventually decline as glaciers shift to cold-based systems where their capacity to mobilize and transport subglacial methane is limited.<\/p>\n Several glaciers in this study have likely reached this point, such as Sagabreen (7) and Bessemerbreen (14). Others, like Gl\u00f8ttfjellbreen (2), may be nearing this transition, which is reflected in low upstream methane concentrations in its melt river, despite its location within the shale-rich Early Cretaceous geological units. However, intra-permafrost groundwaters remain an effective pathway for delivering subsurface methane in these glacial catchments. This is evident across nearly all the rivers in this study, where abrupt increases in methane concentrations downstream from the glacier terminus suggest contributions from groundwater inputs (Fig.\u00a04<\/a>). Notably, peak methane concentrations occur far downstream from the glacier terminus in 68% of the rivers, indicating that the methane flux from groundwater is potentially more critical than the methane delivered by the subglacial drainage system. This is particularly relevant for glacial catchments in more advanced stages of deglaciation, like Gl\u00f8ttfjellbreen, where recently exposed forefields likely contain discontinuous permafrost57<\/a>, allowing for intra-permafrost groundwater flow to sustain methane transport2<\/a>. This finding highlights an important consideration for studies of methane release by glacial systems, which typically concentrate only on water emerging at the glacier terminus and may overlook important methane sources downstream.<\/p>\n Potential emissions of methane from glacial rivers<\/p>\n We use our findings to provide a rough estimate of annual subglacial methane export from beneath each glacier by assessing the flux of methane emerging with the melt river at each glacier terminus. By combining the methane concentrations at the glacier terminus of each river with modeled total annual runoff for each glacier58<\/a>, we estimate that the methane transport from beneath the glaciers ranges from 1\u2009kg CH4 a-1 (0.9\u20131.1\u2009kg CH4 a-1) for Bessemerbreen (13) to 5130\u2009kg CH4 a-1 (4920-5350\u2009kg CH4 a-1) for Penckbreen (11). The total methane transport for each glacier (mean: 416\u2009kg a-1, median: 27\u2009kg a-1) is detailed in Supplementary Table\u00a01<\/a>. Only one other study has quantified methane export from glacial melt rivers on Svalbard, where seasonal monitoring of the river showed that 618\u2009kg CH4 (486\u2013768\u2009kg) were transported from beneath the Vall\u00e5krabreen glacier during the 2021 melt season18<\/a>. Our estimate for methane export from Vall\u00e5krabreen (16), which is up to 607\u2009kg CH4 a-1 (580\u2013632\u2009kg CH4 a-1), aligns closely with these previous findings, providing confidence in the robustness of our approach.<\/p>\n To approximate the scale of methane transport from beneath glaciers across Svalbard, we upscale our calculations using two methods. The first method applies a discharge-weighted-mean methane concentration to the total modeled land-terminating glacial runoff volume58<\/a> for Svalbard, yielding an estimate of 368 t CH4 a-1 (353-383 t CH4 a-1) in 2023 and 364 t CH4 a-1 (349-380 t CH4 a-1) in 2024. The second method uses a glacier-area-weighted-mean mass flux of methane, which is then applied to the total surface area of land-terminating glaciers on Svalbard22<\/a>, resulting in estimates of 182 t CH4 a-1 (168-198 t CH4 a-1) in 2023 and 219 t CH4 a-1 (202-238 t CH4 a-1) in 2024. These projections are likely conservative, as they are based on methane concentrations measured at the most upstream sampling points, which were not always directly at the glacier terminus. Therefore, we do not account for methane lost from the river upstream of the termini in low-pressure channels or between the termini and the sampling locations. Additionally, our sampling was conducted primarily during mid-summer, which does not capture the early-season pulse of methane that has been stored beneath glaciers during winter and is flushed out at the onset of the melt season. This early-season pulse can represent a substantial portion of the total melt season methane flux18<\/a>.<\/p>\n The methane transported by these rivers becomes available for emission to the atmosphere once it reaches the proglacial area. However, our isotopic measurements indicate that some methane may be oxidized in stream, meaning not all the methane is released directly to the atmosphere. Quantifying the extent of this oxidation is challenging with our current data because in-stream microbial methane production and inputs of methane-rich groundwater between sampling points further alter methane concentrations and isotopic compositions. While methane oxidation rates in rivers vary widely across studies20<\/a>,48<\/a>,59<\/a>, it is generally observed that the rate of methane outgassing from rivers far exceeds the rate of microbial oxidation48<\/a>, especially in shallow rivers with high surface-area-to-volume ratios, such as those in this study. This suggests that while some oxidation may occur, the majority of methane transported by these rivers is likely emitted to the atmosphere.<\/p>\n Our upscaled Svalbard-wide methane flux estimate is a similar order of magnitude to the lower-end estimate for methane emissions from glacial rivers across Greenland (918-2110 t CH4 yr-1)60<\/a>. Given that Greenland’s ice cover is more than 55-times larger than that of Svalbard, the flux per unit area is substantially greater in Svalbard. This disparity reflects the contrasting origins of methane between the two regions. Subglacial methane beneath Greenland\u2019s glaciers is of microbial origin, produced in situ from organic material overridden during past glacial advances60<\/a>. In contrast, Svalbard\u2019s shale and carbon-rich geology supersaturates subglacial meltwaters with ancient thermogenic methane, which overwhelms any potential microbial source. This geologic source is evidently more effective at supersaturating meltwaters and thus yields disproportionately higher methane fluxes per unit area. These results implicate other glaciated regions that are situated within significant hydrocarbon-rich geological provinces, such as parts of Arctic Russia, Canada, northeast Greenland, and sections of Antarctica5<\/a>,61<\/a>, where glacial methane emissions have yet to be investigated.<\/p>\n Our calculated methane flux ranges are approximately an order of magnitude lower than the projected annual methane emissions from proglacial groundwater across Svalbard, which are estimated to be up to 2310 t CH4 per year2<\/a>. Therefore, emissions from terrestrial seepage points of methane-rich groundwaters, such as those found downstream in glacial forefields2<\/a> and pingos36<\/a>, appear to play a more critical role in contributing to the Arctic methane budget. This is an important consideration for future studies of methane release from glacial environments, as larger methane sources may exist downstream of the glacier terminus.<\/p>\n Our study demonstrates that the transport of subglacial methane by glacial rivers on Svalbard is governed by the underlying geology and the thermal regime of the glacier bed. Glaciers situated on shale-rich geological formations with thawed, hydrologically active beds mobilize significantly larger fluxes of methane. Subglacial methane on Svalbard is predominantly thermogenic in origin, as corroborated by isotopic signatures and wetness ratios of dissolved gases. While microbial processes such as methanogenesis and methanotrophy are evident in some systems, their contributions are secondary to the dominant geologic influence. Figure\u00a06<\/a> summarizes the processes that mobilize, produce, and transform methane throughout the glacial environment.<\/p>\n Fig. 6: Conceptual model of subglacial methane mobilization. Methane can be introduced to the subglacial drainage system by thermogenic and microbial sources, especially beneath temperate ice zones where the ice-bed interface is thawed and hydraulically conductive. Permafrost beneath cold ice zones, where the ice-bed interface is frozen, substantially reduces hydraulic permeability, and therefore methane sources are limited. Once the glacial river is exposed to the atmosphere, methane begins to outgas, although some methane may be consumed in the river by in-stream methanotrophy. Potential additional sources of methane to the proglacial river include in-stream methanogenesis and inflows of methane-rich groundwater.<\/p>\n As glaciers worldwide continue to melt in response to rising atmospheric temperatures50<\/a>, the increased hydrological activity at glacier beds is likely to enhance methane mobilization. However, as glaciers thin and transition to cold-based systems, methane fluxes from subglacial drainage systems are expected to decline over time, leaving methane-rich groundwaters as the prevailing methane source. This study provides a framework for assessing potential methane fluxes from valley glaciers globally, particularly in regions where glaciers override organic-rich geology. These findings support a growing understanding of the importance of rivers, worldwide, in routing ancient carbon from land to the atmosphere62<\/a> and are critical for improving projections of natural methane emissions in a warming climate.<\/p>\n","protected":false},"excerpt":{"rendered":"Pervasive methane supersaturation in glacial meltwaters Glacial river samples were collected in transects along the length of each…\n","protected":false},"author":2,"featured_media":621264,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[22],"tags":[9850,49279,246,86387,2026,13989,61,60,2027,82],"class_list":["post-621263","post","type-post","status-publish","format-standard","has-post-thumbnail","category-environment","tag-carbon-cycle","tag-cryospheric-science","tag-environment","tag-environmental-chemistry","tag-humanities-and-social-sciences","tag-hydrology","tag-ie","tag-ireland","tag-multidisciplinary","tag-science"],"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/posts\/621263","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/comments?post=621263"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/posts\/621263\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/media\/621264"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/media?parent=621263"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/categories?post=621263"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/tags?post=621263"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}![\"Fig.](\"https:\/\/www.newsbeep.com\/ie\/wp-content\/uploads\/2026\/09\/41467_2026_77190_Fig2_HTML.png\")
<\/p>\n![\"Fig.](\"https:\/\/www.newsbeep.com\/ie\/wp-content\/uploads\/2026\/09\/41467_2026_77190_Fig3_HTML.png\")
<\/p>\n![\"Fig.](\"https:\/\/www.newsbeep.com\/ie\/wp-content\/uploads\/2026\/09\/41467_2026_77190_Fig4_HTML.png\")
<\/p>\n![\"Fig.](\"https:\/\/www.newsbeep.com\/ie\/wp-content\/uploads\/2026\/09\/41467_2026_77190_Fig5_HTML.png\")
<\/p>\n![\"Fig.](\"https:\/\/www.newsbeep.com\/ie\/wp-content\/uploads\/2026\/09\/41467_2026_77190_Fig6_HTML.png\")
<\/p>\n