The study does not show that the structure preserved all marsh functions or that nitrogen removal was unchanged at every site; it documents selective functional convergence within a single artificialized salt-marsh landscape. The investigation used a space-for-time comparison at five locations in the Rivière-du-Loup embayment on the south shore of the Upper St. Lawrence Estuary.

Two unvegetated sites sheltered behind the breakwater were classified as impacted, two sites beyond the structure served as references—one vegetated by cordgrass, Sporobolus alterniflorus, and one unvegetated—and a site immediately seaward of the breakwater was treated as intermediate. The seaward site resembled the references across the measured sediment, oxygen and benthic-flux variables, suggesting relatively little direct breakwater influence there, although the investigators cautioned that it might also have captured offshore effects. By comparing locations rather than sampling the same sites before and after construction, the design associated spatial differences with long-term exposure to the structure without directly reconstructing the construction history at each site.

The breakwater was installed in 2009 as a 300 m-wide barrier of boulders to reduce erosion threatening a major highway. A storm coupled with high tides in 2010 transported substantial sediment that became trapped landward of the structure. The main sampling campaign took place in June 2021, near the beginning of the salt-marsh growing season, when plant roots were active but the aboveground canopy remained short enough to incubate intact vegetated cores.

A supplementary campaign in June 2022 filled specific gaps at the intermediate site and repeated selected sediment measurements; the spatial patterns were consistent between the two years. Seawater temperature and salinity were similar during sampling, at 9 °C and 19.6 in 2021 and 10 °C and 18.7 in 2022. Sediments across the study area were silty clay, but the two landward sites differed clearly from the references.

Their median grain size was smaller, while organic carbon and total nitrogen tended to be higher. This pattern was consistent with the low-energy depositional environment created behind the high-crested structure, where wave energy is reduced and fine, organic-rich particles accumulate. Carbon-isotope signatures remained similar among sites, indicating that the observed difference was primarily in the quantity of deposited material rather than the apparent source of the carbon.

The intermediate seaward location again resembled the reference sediments. Such differences in grain size, organic-carbon content and nutrient content are expected when engineered structures reorganize sediment delivery, but their ecological significance depends on how the altered habitat affects organisms and microbial processes. Those physical changes altered the internal chemistry of the sediment.

At the impacted locations, diffusive oxygen uptake was as much as 2.3 times higher than at reference sites, and oxygen penetrated less deeply, consistent with greater oxygen consumption during the breakdown of accumulated organic matter. Nitrate declined rapidly with depth and fell below the detection limit at roughly 4 cm in impacted sediment, compared with about 6 cm in reference material. Ammonium and phosphate concentrations increased more sharply with depth behind the breakwater, consistent with enhanced organic-matter mineralization and reductive processes.

Reactive iron extracted with ascorbate was concentrated near the surface at the impacted sites, implying that aerobic reactions were confined to a thin surface layer. By comparison, the vegetated reference had a deeper oxidized zone that the researchers associated with marsh-root oxygenation. These profiles show that similar rates measured later at the sediment-water interface emerged from markedly different sediment structures.

The biological community changed as clearly as the sediment. Sampling recovered 2,657 macrofaunal individuals representing 14 taxa, and hierarchical clustering separated the impacted sites, the vegetated reference and the unvegetated reference into three distinct assemblages. Behind the breakwater, high densities of small oligochaetes and the polychaete Hediste diversicolor characterized the fauna.

These organisms are associated with fine, organic-rich sediments, and both can act as substantial bioturbators. The vegetated reference supported small grazers including amphipods and gastropods, while the unvegetated reference had relatively high densities of the bivalve Mya species and greater total biomass. Functional richness was higher at the two references than at the impacted sites, although functional evenness did not differ significantly among them.

Thus, habitat alteration did more than change which species were present: it reduced the breadth of biological functions represented within the communities, even where selected process rates converged. Intact-core incubations showed no simple separation of all affected and reference areas in dark total benthic oxygen uptake, nitrate exchange or ammonium release. These fluxes did differ significantly among individual sites, so the result was not an absence of site-level change.

Instead, their broader pattern was more complex than a uniform breakwater effect: several values were higher at the southern impacted site and the vegetated reference than at the northern impacted, intermediate and unvegetated sites. Total denitrification also differed significantly among sites. It ranged from 6 to 75 µmol m−2 h−1, within the range previously reported for temperate intertidal mudflats and salt marshes, and was highest at the vegetated reference and southern impacted site.

More than half of denitrification at both impacted sites and the vegetated reference was supported by nitrate produced inside the sediment through nitrification. At the unvegetated reference, less than 10% was attributed to that coupled pathway, indicating greater reliance on nitrate supplied from the water column. Stable-isotope tracing helps explain how the apparently similar outcome was achieved.

The southern impacted site had the densest subsurface bioturbation among the affected locations. Burrowing and irrigation by H. diversicolor and oligochaetes increase oxygen transfer, organic-matter mineralization and nitrification in the upper sediment, thereby supplying nitrate that can be reduced immediately nearby. The vegetated reference reached a broadly similar result through a different mechanism.

Cordgrass roots release oxygen into the rhizosphere, creating a subsurface oxidized layer where nitrification can occur; the deeper reactive-iron maximum was consistent with this process. The resulting nitrate was then consumed by denitrification in adjacent anoxic sediment. Root oxygenation at the marsh and animal-mediated sediment irrigation behind the breakwater thus supported comparable nitrogen-removal functions through distinct biogeochemical routes, with locally produced nitrate serving an important role at both.

The two affected sites were not functionally identical. The northern impacted site contained less macrofauna and maintained more reducing conditions than the southern site, apparently limiting internal nitrate production and therefore denitrification. The southern site’s greater bioturbator density partly compensated for its lack of marsh roots and preserved more of the processes measured in vegetated sediment.

By contrast, the unvegetated reference had low organic-carbon availability, restricting ammonium regeneration and nitrification, while its denitrification depended mainly on external nitrate. Its larger burrowers may also have exerted effects beyond the scale of the intact cores used for flux measurements. Comparisons with an earlier bulkhead study further indicate that impacts from shoreline hardening are structure-specific: the present breakwater accumulated organic-rich sediment and supported bioturbators, whereas vertical structures can cause coastal squeeze and reduce total marsh area even when nitrogen removal per unit area remains similar.

Several limits qualify the findings. The incubations were conducted in darkness to standardize comparisons, excluding photosynthetic oxygen production by microphytobenthos and marsh plants, although root oxygen release can persist without light. Oxygen microprofiles and porewater nutrients were measured on one core per location, creating potential pseudo-replication and requiring cautious interpretation of statistical results.

The terminal inventory of isotope-labelled nitrogen gas in sediment porewater was not quantified, so some nitrogen produced during incubations may not have reached the overlying water. Measurements covered selected benthic processes—respiration, nutrient exchange and denitrification—rather than the full suite of services supplied by a natural marsh. Altered communities may also be more vulnerable to additional pressures even where individual process rates remain similar.

At the seascape scale, compensation appears incomplete. The southern impacted area covers about 18,800 m², roughly three times the approximately 6,300 m² of marsh directly lost through breakwater construction, but less than one-third of the approximately 62,700 m² lost through the combined effects of erosion and construction. A first-order estimate therefore suggests that the measured fluxes in the southern area might offset some of the functional capacity lost from the marsh directly removed by the structure, provided the rates adequately represent spatial heterogeneity.

The results demonstrate selective functional convergence, not equivalence between engineered and natural coastlines. Assessing coastal development consequently requires more than a comparison of average process rates: habitat area, sediment conditions, community composition, the origin of nitrate and vulnerability to additional stressors must all be considered when evaluating the consequences of shoreline hardening.

Subject of Research: Breakwater effects on salt-marsh sediment ecology, macrobenthic communities and benthic nitrogen cycling

Article Title: Different paths, same destination: similar benthic nitrogen cycling sustained by distinct biogeochemical mechanisms in artificialized coastal systems

Article References: Pascal, L., Dansereau-Macias, É., Guay, C.-A., Didier, D., Nozais, C., Sirois, L., Zouaghi, F., & Chaillou, G. (2026). Different paths, same destination: similar benthic nitrogen cycling sustained by distinct biogeochemical mechanisms in artificialized coastal systems. Biogeosciences, 23(20), 7327-7343. https://doi.org/10.5194/bg-23-7327-2026

Image Credits: AI Generated

DOI: 10.5194/bg-23-7327-2026

Keywords: coastal breakwaters, salt marshes, benthic nitrogen cycling, denitrification, sediment biogeochemistry, macrobenthic communities, bioturbation, rhizosphere oxygenation, nitrification, St. Lawrence Estuary, Different, paths