Coastal sedimentation across North America doubled in the 20th century despite river dams

The study addresses whether coastal depocenters in North America have become sediment-starved due to widespread dam construction and associated reductions in upstream suspended-sediment loads. Intertidal habitats (oyster reefs, salt marshes, mangroves) rely on sediment to keep pace with sea-level rise (SLR) and protect coastal communities, yet many have already been lost. While dams and watershed management have reduced riverine sediment delivery and altered water and nutrient fluxes, it is unclear if these upstream reductions translate to lower sedimentation in coastal depocenters. The authors test the hypothesis that reduced suspended-sediment delivery would be recorded as decreasing sedimentation rates over the last century; alternatively, if downstream sources offset trapped sediment, sedimentation would be constant or increasing. Understanding these dynamics is critical for coastal resilience, restoration planning, and risk assessment under accelerating SLR.

Background literature indicates that land-use change initially increased river sediment loads due to erosion (e.g., deforestation to agriculture), followed by declines associated with watershed management and dam proliferation. River management can decrease sediment delivery while increasing nutrient and pollutant delivery to coasts. Studies document coastal retreat, marsh disintegration, and variable relative sea-level trends influenced by climate and subsidence. Prior work on sediment budgets and geomorphic responses highlights contributions from bank erosion, urbanization, agricultural development, and legacy sediments downstream of dams, as well as the role of storms and long-term accommodation changes due to SLR. However, comprehensive sediment budgets for many watersheds and direct quantification of source contributions to coastal depocenters are lacking, motivating the regional synthesis in this paper.

The authors compiled and analyzed century-scale sediment-core records from 25 coastal depocenters across North America, spanning diverse geologic and climatic settings, including 18 estuaries (average core water depth ~10 m) and 7 inner continental shelf sites (average core water depth ~49 m). They targeted subtidal depocenters with relatively stable depositional processes through time to serve as archives of regional sedimentation, minimizing direct shoreline dynamics. Geochronologies were established primarily using excess 210Pb profiles measured via gamma spectrometry, applying established models where profiles satisfy criteria for reliable dating (stable horizons, sufficient data points, minimal disturbance). Many datasets were originally collected to quantify fluxes of contaminants and organic carbon, but here were reinterpreted to calculate mass accumulation rates (MAR; g cm⁻² yr⁻¹) and sediment accumulation rates (SAR; cm yr⁻¹). The time series were binned into pre-1950 and post-1950 intervals to capture the period of peak dam construction and coastal population growth. For a subset of 13 sites, the authors compared decadal-scale SAR derived from 210Pb with century- to millennial-scale SAR from radiocarbon dates to evaluate potential bias from the Sadler effect; they found decadal SAR were not systematically higher than longer-term SAR within uncertainty, supporting the fidelity of the decadal records. They summarized site-level MAR and SAR statistics (min, max, and pre/post-1950 medians), and compared means (with mean measurement errors) for pre- vs post-1950 periods. They also compared post-1950 SAR to site-specific rates of relative sea-level rise (RSLR) to assess whether sedimentation kept pace with water-level changes. Statistical testing included paired comparisons of pre- and post-1950 means (paired t-tests mentioned) to evaluate increases in MAR and SAR. Site selection emphasized peer-reviewed datasets, minimal direct human disturbance at coring locations, and regional diversity. Quality control involved evaluating constancy in mixing depth, density, texture, and bioturbation across cores over the past ~150 years.

- MAR and SAR more than doubled after 1950 across North American coastal depocenters. Post-1950 median MAR exceeded pre-1950 median MAR at all 25 sites analyzed. Mean MAR and SAR increased after 1950 at each site except Sabine Bay. - Example MAR values illustrating the range: Florida Bay average MAR = 1.405 g cm⁻² yr⁻¹; Corpus Christi Bay average MAR = 0.195 g cm⁻² yr⁻¹; Nastapoka (Natsapoka) Sound average MAR = 0.018 g cm⁻² yr⁻¹. - Pre- vs post-1950 site medians (examples from table): New York Bight median MAR increased from 0.130 to 0.399 g cm⁻² yr⁻¹; San Francisco Bay from 0.863 to 1.143 g cm⁻² yr⁻¹; Santa Clara Shelf from 0.251 to 0.557 g cm⁻² yr⁻¹; Terrebonne Bay from 0.107 to 0.476 g cm⁻² yr⁻¹; Galveston Bay from 0.069 to 0.240 g cm⁻² yr⁻¹. - The observed increases are not attributable to reported changes in bioturbation, surface mixing depth, grain size, or bulk density, implying a genuine increase in sediment supply to depocenters. - Downstream sediment sources offset upstream trapping by dams: contributions include riverbank erosion and incision below dams, urbanization and construction, agricultural development, changes in riparian vegetation, mobilization of legacy sediments, and climate-driven sources (e.g., permafrost thaw delivering sediment to high-latitude depocenters like Nastapoka Sound). - Relationship to sea-level change: Post-1950 SAR generally matched or exceeded relative SLR along the East Coast, Gulf (excluding rapidly subsiding areas), Florida, Alabama, Cuba, and Mexico, indicating stable or shallowing water depths in depocenters. In contrast, at 8 sites in Louisiana and Texas, RSLR exceeded SAR, indicating deepening basins and loss of intertidal area despite increased sedimentation. - Illustrative RSLR variability: Near Nastapoka Sound, sea level fell by ~0.992 mm yr⁻¹ (isostatic rebound), corresponding to low SAR (~0.54 mm yr⁻¹), whereas in Barataria Bay, LA, RSLR was ~4.90 mm yr⁻¹ (subsidence), outpacing SAR. - Coastal population in adjacent counties increased sharply around 1950 and doubled from 1950 to 2001 at many sites, consistent with enhanced downstream sediment production from development and land-cover change.

Findings contradict the prevailing assumption that dam-induced reductions in riverine sediment loads necessarily result in diminished coastal sedimentation. Instead, supplemental downstream sources—bank erosion below dams, urban and agricultural runoff, riparian changes, and mobilized legacy sediments—have increased the flux of fine-grained material to coastal depocenters, producing widespread increases in MAR and SAR since 1950. The coherence of these increases across diverse settings (estuaries and inner shelves) suggests a continental-scale signal related to human landscape modification and increased accommodation from SLR. Where RSLR is moderate, SAR has kept pace or surpassed RSLR, indicating that many depocenters are not sediment-starved and may maintain stable water depths. However, in rapidly subsiding regions (notably Louisiana and Texas), even elevated SAR cannot match high RSLR, leading to deepening basins and loss of intertidal habitats. The results imply that coastal risk assessments should integrate both sea-level projections and regional sediment dynamics, recognizing that downstream sediment generation can partially counteract upstream trapping. Nonetheless, accelerating global SLR will likely outstrip current sedimentation rates in the future, threatening the resilience of coastal habitats unless sediment management is incorporated into restoration and conservation strategies.

Across 25 North American coastal depocenters, sediment mass and thickness accumulation rates increased markedly after 1950, despite widespread dam construction that reduced upstream suspended-sediment loads. Downstream sediment sources have compensated for reservoir trapping, enabling SAR at many sites to match or exceed relative SLR; rapidly subsiding regions along the Texas and Louisiana coasts remain exceptions where RSLR outpaces SAR, contributing to habitat loss. These findings challenge assumptions of universal coastal sediment starvation due to damming and highlight the need to include sediment availability and management in coastal restoration. Future work should quantify the relative contributions of specific downstream sources, develop sediment budgets at watershed-to-basin scales, and assess how accelerating SLR will interact with sediment supply. Restoration in high-subsidence areas will require supplemental sediment to maintain intertidal elevations within the tidal frame.

- The study does not quantify the relative contributions of specific downstream sediment sources (e.g., bank erosion, urban construction, agriculture, legacy sediments) to each depocenter; comprehensive sediment budgets were unavailable and could not be constructed from existing data. - Selection of sites was constrained by the quality and criteria of 210Pb profiles, limiting geographic coverage and potentially excluding some systems. - Some methodological descriptions (e.g., 210Pb source variability and storm-related pulses) indicate potential complications in geochronology, though sites were screened to meet established criteria. - Spatial heterogeneity within basins and alongshore sediment transport introduces uncertainty when extrapolating site-specific cores to entire depocenters. - The analysis relies on pre-/post-1950 binning; finer temporal attribution of drivers (e.g., timing of urbanization, dam completion, storm regimes) is limited. - Relative sea-level trends vary locally due to vertical land motion; uncertainties in RSLR estimates and SAR-MAR measurements affect comparisons. - The study focuses on subtidal depocenters and does not directly resolve intertidal platform dynamics or shoreline change processes at fine spatial scales.