Arctic coastal reactors: Lateral and vertical organic matter dynamics in permafrost lagoons of the western Canadian Arctic

Tanski, George; Fritz, Michael; Vonk, Jorien E.; Strauss, Jens; Chassiot, Léo; Carnero-Bravo, Vladislav; Falardeau, Jade; Klein, Kostantin; Lajeunesse, Parick; Lantuit, Hugues; Speetjens, Niek J.; de Vernal, Anne; Whalen, Dustin; Irrgang, Anna M.

doi:10.5194/egusphere-2026-2314

Preprints

https://doi.org/10.5194/egusphere-2026-2314

Preprints

22 Jun 2026

| 22 Jun 2026

Arctic coastal reactors: Lateral and vertical organic matter dynamics in permafrost lagoons of the western Canadian Arctic

George Tanski, Michael Fritz, Jorien E. Vonk, Jens Strauss, Léo Chassiot, Vladislav Carnero-Bravo, Jade Falardeau, Kostantin Klein, Parick Lajeunesse, Hugues Lantuit, Niek J. Speetjens, Anne de Vernal, Dustin Whalen, and Anna M. Irrgang

Abstract. Arctic warming leads to longer open-water periods, intensified storms, and rising relative sea level, which accelerate permafrost coastline erosion and enhance the lateral transport of sediment and organic matter (OM) from land to sea. Lagoons and embayments are widespread along Arctic coasts, and are located at the heart of the land-sea transition zone, yet their role in OM cycling remains poorly understood. Here, we assess the function of the Kanivaliuraq lagoon (Ptarmigan Bay, western Canadian Arctic) as a biogeochemical reactor for terrestrial OM. We combine shoreline position changes (1950–2018) with sediment core and surface sediment transect sampling to quantify sources, pathways, and trajectories of OM and sediments. Samples were analyzed for total organic carbon (TOC), total nitrogen, stable carbon isotopes (δ¹³C), mineral surface area, and grain size, complemented by sedimentation rates from 210Pb/137Cs dating and water turbidity dynamics from Landsat imagery. We find that OM content declines by more than 50 % along the land–lagoon–ocean gradient, both in TOC (% dry weight) and OC loading (mg OC m^-2), indicating efficient degradation, burial and offshore transport of terrestrial OM within the lagoon. While currents and wind-driven resuspension are expected to further enhance OM redistribution and export, a substantial amount of OM is sequestered and mineralized in lagoon sediments reflecting both effective burial and degradation of erosion-derived OM. Yet, a major portion of OM is also quickly removed offshore. Shoreline erosion rates increased from 0.6 m yr^-1in the 1950s to 0.9 m yr^-1in the 1970s to 3.3 m yr^-1 in 2011 to 2018, paralleled by increasing mass accumulation rates from 0.32 to 0.57 g cm^-2yr^-1. Due to intensified erosion and warming-induced permafrost degradation, lagoons and embayments will be supplied with increasing amounts of terrestrial OM being subject to either mineralization, sequestration or offshore transport, also fueling primary production in these Arctic coastal “reactors” at the same time. Our results highlight that Arctic lagoons, often overlooked components of the land–ocean transition zone, act as dynamic reactors and partial sinks for terrestrial OM. Intensifying coastal erosion as a result of Arctic climate warming is likely to increase their importance in Arctic carbon cycling.

Received: 22 Apr 2026 – Discussion started: 22 Jun 2026

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.

Download & links

Status: final response (author comments only)

RC1:
'Comment on egusphere-2026-2314', Anonymous Referee #1, 30 Jul 2026
This manuscript by Tanski et al. addresses the role of Arctic permafrost lagoons as dynamic biogeochemical reactors in the land‑ocean transition zone. The study combines shoreline change analysis, sediment core chronology, and bulk geochemistry to trace organic matter (OM) dynamics along a terrestrial–lagoon–marine gradient. Findings from this type of studies have implications for carbon budgeting and climate feedbacks. Overall, I suggest moderate/major revisions.
Below are my comments for consideration:
The study provides useful empirical data from a well‑characterized system, but the conceptual novelty is limited. The finding that OM content declines along the land‑lagoon‑ocean gradient is not surprising and has been documented in other coastal systems (e.g., Schreiner et al., 2013; Hanna et al., 2014). In addition, costal erosion and lateral transport have been reported in Beaufort Sea (Ping et al., 2011; Tsai, 2024). The authors should clarify the novelty relative to Jenrich et al. (2021, 2025) and Irrgang et al. (2018).

The conclusion that lagoons act as both sinks and reactors for terrestrial OM echoes previous work on thermokarst lagoons (Jenrich et al., 2021, 2025a,b,c). The increase in erosion and sedimentation rates over time is consistent with regional trends already reported by the same research group (Irrgang et al., 2018; Couture et al., 2018). The authors should more clearly distinguish their contributions from those of Jenrich et al. (2021, 2025a,b,c), who have already published extensively on thermokarst lagoon carbon dynamics.

The study would benefit from a clearer articulation of what is new beyond confirming existing frameworks. For example, do the authors offer novel insights into the relative importance of degradation versus burial, or the role of wind‑driven turbidity in organic matter processing?

The literature review is comprehensive and appropriately cites key studies on Arctic erosion, permafrost carbon, and lagoon systems. However, there is too much emphasis on local literature (and heavily from their own research network) and too little discussion of broader coastal biogeochemistry. In addition, the authors could cite recent work on DOM dynamics in Arctic nearshore zones to strengthen the discussion of lateral OM export.

The claim that lagoons will become increasingly important carbon reactors under future warming is plausible but not directly tested. Similarly, the authors suggest that lagoons may be significant greenhouse gas (GHG) sources, but no GHG flux measurements are presented, weakening the direct relevance to the permafrost carbon feedback.

The study relies on one‑time sediment sampling and does not capture seasonal dynamics, which are critical for understanding OM processing in dynamic lagoons with seasonal and interannual lateral transport. The authors should justify their choice of constant flux‑constant sedimentation model and discuss potential uncertainties given the dynamic nature of lagoons.

Is there quantitative evidence or reference to support the occurrence of the depocenter and undisturbed sedimentation at the sampling location and thus the interpretation of MAR and OCMAR. In addition, the linkage between physical forcing, sediment dynamics, and OM transformation seems underdeveloped.

The statement that “most of the OM is derived from terrestrial sources” (Section 4.2) is not supported by the data, as marine OM contribution (based on δ¹³C and C/N) is not quantitatively estimated. A two‑end‑member mixing model would strengthen this claim.

In Table 1, OCMAR for the lagoon (PB) is higher than expected from MAR (e.g., 0.48 vs. 0.57). This is mathematically possible only if TOC content is > 0.84 or >80%. However, TOC data shown in SI are normally less than 30% in PB lagoon. Please verify and comment.

To place this study into a broader context, the manuscript would benefit from comparison with recent syntheses on permafrost carbon export and coastal carbon burial.

Lastly, the manuscript should explicitly acknowledge limitations on single lagoon, limited number of sediment cores, lack of water-column carbon measurements (no DOC budget), and incomplete carbon mass balance.
Citation: https://doi.org/10.5194/egusphere-2026-2314-RC1
RC2:
'Comment on egusphere-2026-2314', Anonymous Referee #2, 02 Aug 2026
The manuscript presents a multi-proxy dataset from a lagoon system in the western Canadian Arctic, combining 65 years of shoreline-change analysis, a matched land–lagoon–marine sediment geochemistry transect, mass accumulation rates from lagoon and marine short cores, and Landsat-derived turbidity and temperature patterns under contrasting wind regimes. I particularly like the gradient approach and think this represents a valuable dataset and a generally well-argued contribution that addresses an important gap in the Arctic land-to-ocean carbon cycle literature. The combination of multiple proxies and methods across the land–ocean continuum is a clear strength of the study.
My main concerns are less related to the measurements themselves, but rather to how strongly some of the framing and conclusions are stated compared to what the current dataset can actually support. In addition, some aspects of the terminology and methodological description would benefit from further clarification to improve transparency and reproducibility. I also think that there is a missed opportunity to strengthen the main conclusions through a more quantitative assessment ( mass balance , mixing model ). Overall, I think these issues are addressable through moderate revision and would substantially improve the robustness and impact of an already valuable study.
Main comments
Land–lagoon–ocean gradient and transport pathway. The land–lagoon–ocean gradient framing used throughout the manuscript appears to conflate a simple spatial gradient with an actual transport pathway. The manuscript treats the terrestrial sites, lagoon short cores and marine HB cores as representing a continuous land-to-ocean transfer pathway, and geochemical proxies are interpreted along this implied gradient (Figs. 4–6). However, the authors' own description of sediment transport pathways and potential OM sources in the study area and discussion suggests that the HB site may receive sediment and organic matter from several sources that are not directly connected to lagoon export. It therefore seems that the marine HB cores are not necessarily hydrodynamically downstream of the lagoon, but rather represent a nearby marine environment. This is an important distinction and not uncommon. Land–ocean transition zones are generally complex and dynamic systems. Spatial proximity between a lagoon/estuary/river and a coastal ocean site does not necessarily demonstrate connectivity. This point requires further discussion because the reported ~50% OM decline along the gradient is one of the main results, and because the stepwise comparison of terrestrial, lagoon and marine OC pools in Section 4.3 and Table 2 relies on this framing. But the observed land-to-ocean "gradient" could simply reflect differences in OC inputs and transport/degradation histories between sites rather than progressive OC transformation during export. I would suggest either providing stronger evidence for hydrodynamic and sedimentological connectivity between the lagoon and HB site, or alternatively reframing the interpretation throughout as a comparison between distinct terrestrial, lagoon and regional marine end-members. The latter appears to be the more defensible interpretation based on the information currently provided.
This issue also affects Fig. 5. The Jong et al. data points from Herschel are plotted together with the in-lagoon transect, which implies a common spatial axis or transport pathway. However, these represent an independent dataset, and the physical connection between this dataset and the sampled lagoon transect is currently not clear.
Lagoon carbon budget The manuscript provides estimates of OC erosion from shorelines into the lagoon (Table 3) and OC burial rates from lagoon cores (Table 1), but then qualitatively states that "a major portion of OM is also quickly removed offshore". I think this represents a missed opportunity to provide a first-order lagoon carbon budget. Given the available erosion-derived OC fluxes (t yr-1) and lagoon OC mass accumulation rates (kg m-2 yr-1), a simple estimate of the retained fraction could be calculated by scaling OCMAR rates by lagoon area. Such an approach would obviously involve uncertainties and assumptions (which should be discussed), but even a first-order estimate would allow the authors to quantify the relative importance of burial versus export rather than only discussing this qualitatively. This would substantially strengthen the "bioreactor" concept by providing clearer constraints on inputs and outputs. Importantly, while being independent of the connectivity issue raised above regarding the marine cores. Also, a simple summary figure or box showing the conceptual OM budget (erosional input, burial, degradation and export, including uncertainties) would also help readers understand the overall mass balance that underpins the manuscript.
Degradation versus mixing . The authors appropriately acknowledge that d13C enrichment may reflect either degradation or increased mixing with marine OM, and that OC loading is affected by mineral dilution effects. This discussion is useful. However, would a simple end-member mixing calculation help quantify how much of the observed lateral OC decrease could be explained by conservative mixing alone, compared to actual degradation? Such an analysis would provide additional support for the degradation interpretation and would make the "reactor" framing more robust.
OC pool comparison (Tab2) should more explicitly state considered depths/timescales and consider the limitations of comparing different depth/time scales. OC contents in sediments don’t scale linearly with depths and similar depths scales can represent largely different timescales. The authors acknowledge that other studies used different depth intervals (3 m vs 30 cm here), yet still directly compare absolute values in the text (larger than OC pools observed in other regions). Normalize pools to a common depth or show OC density profiles instead of integrated pools. In addition, the conclusion that the lagoon holds two times more OC than the marine OC pool is based on a fixed 0–30 cm depth interval. Given that MAR can differ substantially between the two core locations the top 30 cm likely represents very different time intervals. A depth-fixed comparison would then also reflect differences in burial history rather than in sequestration capacity. Consider either normalizing to a common time window using the age models already available, or explicitly discussing this limitations. Finally, OC fluxes through a 30 cm depths horzion are not burial fluxes in the traditional sense (ie transition of C from the surface to the geological C cycle, see Bradley et al. https://www.nature.com/articles/s41467-022-35112-9 ). Explicitely state what you mean by “burial” here (ie sequestration beyond the 30cm depths horizon) and also by “longterm” (ie on which timescale is C sequestered between the 0-30cm depths horizon, assuming that DIC released within the first 10s of centimeters of sediment can still interact with the surface C cycle on policy relevant timescales).
Critically discuss limitations/representativeness. I really like the multi-proxy, multi-method gradient approach and support the general discussion, but think some conclusions and the manuscript's language sometimes read more confidently than the data justifies. For instance, the central 50% decline in OC along the land–lagoon–ocean gradient rests on two shallow short cores plus 11 surface-transect samples along a single 2 km transect, plus one lagoon depocenter core (PB) and one marine core (HB), which does not necessarily lie along a source to sink transport pathway. This is a very nice but still has its limitations in spatial/temporal resolution, connectivity and is not necessarily representative for a wider area. Please explicitly discuss potential limitations critically and tone down generalizing claims accordingly, or, alternatively, clearly justify why the data supports the strong claim. Also, section 4.5 carefully notes that different lagoon types need to be defined before generalizing but the abstract's framing ("Arctic lagoons... act as dynamic reactors,") reads more generally than a single semi-open, spit-protected, glaciated-terrain site can strictly support. Consider flagging the single-site caveat slightly earlier and more prominently.
Minor comments
Spelling "Qikiqtaruk" or "Qikiqatruk" please correct throughout.

"Environment Canada Canada" duplicated word.

"CF model" and "CFCS model" are used interchangeably. Pick one term and use it consistently (see comment above).

“long-term” be more precise. Over what timescale?

The abstract states erosion rates increased "from 0.6 m yr-1...to 0.9...to 3.3 m yr-1," but Table A2 gives slightly different "mean erosion rate" values (-0.7, -0.9, -3.7) compare to the the mean shoreline change rate values (-0.6, -0.9, -3.3) used elsewhere. Please clarify in the text which metric is being reported where.

Table 3: the annual area loss for 2011–2018 (26,976 m2 yr-1) appears inconsistent by an order of magnitude with the other two periods (3,926 and 4,088 m² yr⁻¹) relative to the "eroded area" column. A sentence explaining this would help.

The Discussion (4.4–4.5) about the regional extent of lagoons along the Beaufort coast is useful context but reads somewhat like a literature review. Consider tightening or moving some of this to the Introduction, keeping the Discussion focused on interpreting your own data.
Citation: https://doi.org/10.5194/egusphere-2026-2314-RC2

Viewed

Total article views: 130 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	BibTeX	EndNote
70	44	16	130	13	13

HTML: 70
PDF: 44
XML: 16
Total: 130
BibTeX: 13
EndNote: 13

Views and downloads (calculated since 22 Jun 2026)

Month	HTML	PDF	XML	Total
Jul 2026	34	30	9	73
Aug 2026	36	14	7	57

Cumulative views and downloads (calculated since 22 Jun 2026)

Month	HTML	PDF	XML	Total
Jul 2026	34	30	9	73
Aug 2026	36	14	7	57

Viewed (geographical distribution)

Total article views: 98 (including HTML, PDF, and XML) Thereof 98 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 21 Aug 2026

Short summary

As the Arctic warms, permafrost coasts erode rapidly, flushing carbon from frozen ground into the sea. We studied a lagoon in the western Canadian Arctic, collecting sediments and analyzing organic carbon content, stable isotopes, and grain size, alongside satellite imagery. Carbon content fell by over 50 % from land to open ocean, while erosion rates quintupled since the 1950s. Arctic lagoons bury, degrade, and export terrestrial carbon, acting as coastal reactors whose role will grow.


Total:	0
HTML:	0
PDF:	0
XML:	0