Yongsheng Wu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Yongsheng Wu has a total of 24 co-authors. Here are the top co-authors by number of shared publications:
- Yongxing Ma
Fisheries and Oceans Canada, Bedford Institute of Oceanography, 1 Challenger Drive, Dartmouth, B2Y 4A2, Nova Scotia, Canada
4 shared publications - Kira A. Krumhansl
Department of Fisheries and Oceans Bedford Institute of Oceanography Dartmouth Nova Scotia Canada
2 shared publications - Kumiko Azetsu-Scott
Department of Fisheries and Oceans, Bedford Institute of Oceanography, Dartmouth, B2Y 4A2, Canada
2 shared publications Marc Skinner
2 shared publications- Melisa C. Wong
Bedford Institute of Oceanography, Fisheries and Oceans Canada, Dartmouth, Canada
2 shared publications Michael Z. Li
2 shared publications
Latest Publications
- Open Access
- 22 October 2026
- Published by Elsevier in SSRN Electronic Journal
Parameterizing the Influence of Fixed Monopiles in Phase-Averaged Spectral Wave Models
Monopiles affect waves through scattering and drag-induced dissipation. In this study, these effects are parameterized as two subgrid source terms in ocean wave models. The scattering source is derived from far-field energy fluxes of the MacCamy--Fuchs linear diffraction solution, and the drag sink from a bulk formulation based on the Morison drag law. Both terms are implemented in WAVEWATCH III (WW3), Simulating WAves Nearshore (SWAN), and Wind Wave Model III (WWMIII). In WW3, upper and lower tail means of normalized wave height changes agree with analytical benchmarks to within 3.9% for single monopiles across variations in monopile diameter, water depth, and peak wavelength, and within 2.5% for monopile arrays across variations in monopile spacing. These scattering responses show limited sensitivity to grid spacing. An 11×11 array with 1-km spacing produces a peak downstream wave height deficit of 4.3%, compared with 0.55% for a single monopile, indicating amplification through cumulative scattering. In the diagnostic simulations, scattering-induced wave height changes are approximately an order of magnitude larger than drag-induced changes. The parameterization enables wave models to represent monopile effects without resolving their geometry. Implementation code is provided for all three models.
- 22 October 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessParameterizing the Influence of Fixed Monopiles in Phase-Averaged Spectral Wave Models
Monopiles affect waves through scattering and drag-induced dissipation. In this study, these effects are parameterized as two subgrid source terms in ocean wave models. The scattering source is derived from far-field energy fluxes of the MacCamy--Fuchs linear diffraction solution, and the drag sink from a bulk formulation based on the Morison drag law. Both terms are implemented in WAVEWATCH III (WW3), Simulating WAves Nearshore (SWAN), and Wind Wave Model III (WWMIII). In WW3, upper and lower tail means of normalized wave height changes agree with analytical benchmarks to within 3.9% for single monopiles across variations in monopile diameter, water depth, and peak wavelength, and within 2.5% for monopile arrays across variations in monopile spacing. These scattering responses show limited sensitivity to grid spacing. An 11×11 array with 1-km spacing produces a peak downstream wave height deficit of 4.3%, compared with 0.55% for a single monopile, indicating amplification through cumulative scattering. In the diagnostic simulations, scattering-induced wave height changes are approximately an order of magnitude larger than drag-induced changes. The parameterization enables wave models to represent monopile effects without resolving their geometry. Implementation code is provided for all three models.
- Open Access
Habitat loss does not affect ecological connectivity for a habitat engineer with short dispersal
Habitat loss, directly linked to anthropogenic activities, is a key driver of biodiversity loss globally; in combination with habitat fragmentation, it leads to smaller, more isolated population patches. Such changes in habitat configuration can negatively impact population connectivity (i.e., the supply of individuals locally or from neighboring patches) and compromise metapopulation persistence. For short‐distance dispersers, loss of connectivity between habitat patches may have a smaller impact on persistence than reduction in patch size. We investigated the relationship of habitat loss and associated changes in habitat configuration with demographic connectivity using a temperate marine system of laminarian kelps as a case study. Kelps are an important ecosystem engineer, currently experiencing global declines because of habitat loss. We ran two in silico experiments of habitat loss by removing habitat in 10% increments from 10% to 90%, and 1% increments from 95% to 99%: (1) suitable habitat was removed randomly; and (2) the probability of habitat loss was inversely proportional to patch size, with smaller patches having a greater likelihood of loss. Kelp habitat was clustered into patches based on spatial dependencies derived from Lagrangian particle tracking simulations. We found that area‐weighted habitat loss created a trade‐off between structural and functional habitat fragmentation. While this form of habitat loss resulted in fewer, larger patches (less structural fragmentation) than random loss, patches were more isolated, and consequently, the metapopulation showed lower betweenness centrality and modularity (more functional fragmentation). However, under both random and area‐weighted loss, patches were supported by self‐replenishment for 90% of recruited individuals, receiving minimal inflow from neighboring populations. These results suggest that although rescue potential of patches from neighbors is low, especially in the case of area‐weighted loss, the demise of individual patches may not trigger the loss of neighboring ones. Thus, the probability of metapopulation persistence remains high, given only a few patches need to remain. We showed that some but not all theoretically predicted relationships between habitat configuration and metapopulation dynamics are supported in real‐world ecological systems.
Open AccessHabitat loss does not affect ecological connectivity for a habitat engineer with short dispersal
Habitat loss, directly linked to anthropogenic activities, is a key driver of biodiversity loss globally; in combination with habitat fragmentation, it leads to smaller, more isolated population patches. Such changes in habitat configuration can negatively impact population connectivity (i.e., the supply of individuals locally or from neighboring patches) and compromise metapopulation persistence. For short‐distance dispersers, loss of connectivity between habitat patches may have a smaller impact on persistence than reduction in patch size. We investigated the relationship of habitat loss and associated changes in habitat configuration with demographic connectivity using a temperate marine system of laminarian kelps as a case study. Kelps are an important ecosystem engineer, currently experiencing global declines because of habitat loss. We ran two in silico experiments of habitat loss by removing habitat in 10% increments from 10% to 90%, and 1% increments from 95% to 99%: (1) suitable habitat was removed randomly; and (2) the probability of habitat loss was inversely proportional to patch size, with smaller patches having a greater likelihood of loss. Kelp habitat was clustered into patches based on spatial dependencies derived from Lagrangian particle tracking simulations. We found that area‐weighted habitat loss created a trade‐off between structural and functional habitat fragmentation. While this form of habitat loss resulted in fewer, larger patches (less structural fragmentation) than random loss, patches were more isolated, and consequently, the metapopulation showed lower betweenness centrality and modularity (more functional fragmentation). However, under both random and area‐weighted loss, patches were supported by self‐replenishment for 90% of recruited individuals, receiving minimal inflow from neighboring populations. These results suggest that although rescue potential of patches from neighbors is low, especially in the case of area‐weighted loss, the demise of individual patches may not trigger the loss of neighboring ones. Thus, the probability of metapopulation persistence remains high, given only a few patches need to remain. We showed that some but not all theoretically predicted relationships between habitat configuration and metapopulation dynamics are supported in real‐world ecological systems.
- Open Access
- 20 July 2026
- Published by Elsevier in Ocean Modelling
- Vol. 204
Numerical assessment of wave responses to wind wakes induced by an offshore wind energy area on the Scotian Shelf
- 20 July 2026
- Published by Elsevier in Ocean Modelling
- Vol. 204
Open AccessNumerical assessment of wave responses to wind wakes induced by an offshore wind energy area on the Scotian Shelf
- Open Access
- 9 March 2026
- Published by Elsevier in SSRN Electronic Journal
Numerical Assessment of Wave Changes Induced by Offshore Wind Farm Wake Effects on the Scotian Shelf
Motivated by understanding how the offshore Proposed Wind Energy Area (PWEA) may modify local wave conditions on the Scotian Shelf, this study investigates wave responses to turbine-induced wind speed (WS) deficits. An unstructured-grid spectral wave model based on SCHISM–WWM-III was configured and forced with hourly 10-m winds from ERA5. The WS fields modified by turbine wakes were simulated using PyWake. Model results show good agreement with moored buoy and ADCP observations, with high skill for significant wave height (Hs) and moderate skill for peak wave period (Tp). Simulations with constant wind conditions show that the turbine-induced reduction in Hs depends non-monotonically on background WS, with the strongest response occurring at moderate winds (around 8 m s-1) and weaker responses at both lower and higher wind speeds. The mean wave period (Tm) exhibits spatially varying changes, with slight increases inside the PWEA and decreases downstream, and the magnitude of these changes decreases with increasing WS. Turbine wakes also induce a dipole-like pattern in mean wave direction (θm), with the largest directional changes occurring under weak wind conditions. ERA5-forced simulations with turbines quantified the cumulative impact of turbine wakes under realistic atmospheric forcing. For a turbine spacing of 10 rotor diameters (D) (402 turbines), reductions in time-mean Hs within and downstream of the PWEA reach about 2.4 cm (1.5%) in summer and 1.9 cm (0.8%) in winter. A denser 5 D layout produces larger wave height reductions, with maximum decreases of up to 7 cm (4.3%). In contrast, time-mean Tm generally increases slightly, by up to 0.07 s for the 10 D layout and up to 0.2 s for the 5 D layout, while changes in time-mean θm are typically less than 1°. This study provides a quantitative assessment of how future offshore wind development may influence regional wave characteristics on the Scotian Shelf, with potential implications for ocean mixing and the transport and dispersal of pelagic particles.
- 9 March 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessNumerical Assessment of Wave Changes Induced by Offshore Wind Farm Wake Effects on the Scotian Shelf
Motivated by understanding how the offshore Proposed Wind Energy Area (PWEA) may modify local wave conditions on the Scotian Shelf, this study investigates wave responses to turbine-induced wind speed (WS) deficits. An unstructured-grid spectral wave model based on SCHISM–WWM-III was configured and forced with hourly 10-m winds from ERA5. The WS fields modified by turbine wakes were simulated using PyWake. Model results show good agreement with moored buoy and ADCP observations, with high skill for significant wave height (Hs) and moderate skill for peak wave period (Tp). Simulations with constant wind conditions show that the turbine-induced reduction in Hs depends non-monotonically on background WS, with the strongest response occurring at moderate winds (around 8 m s-1) and weaker responses at both lower and higher wind speeds. The mean wave period (Tm) exhibits spatially varying changes, with slight increases inside the PWEA and decreases downstream, and the magnitude of these changes decreases with increasing WS. Turbine wakes also induce a dipole-like pattern in mean wave direction (θm), with the largest directional changes occurring under weak wind conditions. ERA5-forced simulations with turbines quantified the cumulative impact of turbine wakes under realistic atmospheric forcing. For a turbine spacing of 10 rotor diameters (D) (402 turbines), reductions in time-mean Hs within and downstream of the PWEA reach about 2.4 cm (1.5%) in summer and 1.9 cm (0.8%) in winter. A denser 5 D layout produces larger wave height reductions, with maximum decreases of up to 7 cm (4.3%). In contrast, time-mean Tm generally increases slightly, by up to 0.07 s for the 10 D layout and up to 0.2 s for the 5 D layout, while changes in time-mean θm are typically less than 1°. This study provides a quantitative assessment of how future offshore wind development may influence regional wave characteristics on the Scotian Shelf, with potential implications for ocean mixing and the transport and dispersal of pelagic particles.
- Open Access
- 4 March 2026
- Published by Springer Nature in Estuaries and Coasts
- Vol. 49 (3) , 61
Acclimation To Seasonal and Localized Physical Conditions Influences the Temperature-Productivity Response of Eelgrass, Zostera marina
Increasing ocean warming and marine heatwaves will have important consequences for the world’s seagrasses. Predicting impacts of thermal stress on seagrasses depends on understanding their temperature-productivity responses. Here, we examine these responses for eelgrass ( Zostera marina ) and how they are influenced by plant acclimation to seasonal and localized physical conditions. Eelgrass was collected seasonally from two sites in Nova Scotia (NS) Canada that were characterized by contrasting physical conditions, with temperature regime being a dominant feature. Laboratory incubations measured oxygen production and consumption rates by eelgrass leaves and roots-rhizomes from 5 to 40 °C. Unimodal temperature dependence of net photosynthetic rate and whole plant net productivity was modelled and maximum productivity ( P max ) and optimal temperatures ( T opt ) were estimated. Both net photosynthetic rate and net plant productivity were more peaked and bell-shaped across increasing temperature for plants from a warm highly variable temperature regime compared to plants from a cool less variable regime. P max and T opt were highest in warm relative to cool regime plants across all seasons. Furthermore, P max was lower and T opt was higher when seasonal ambient temperatures increased. T opt for whole plant productivity (22.5 to 28.5 °C) was used to create a thermal risk map for eelgrass, which indicated that 20%, 8.5%, 4.6%, and 1.3% of predicted eelgrass habitat in NS exceeds 22 °C, 24 °C, 26 °C and 28 °C, respectively, for at least 1 h annually. Our results show that even northern temperate eelgrass can be heat stressed, and that plant acclimation to seasonal and localized conditions should be considered when evaluating seagrass thermal risk.
- 4 March 2026
- Published by Springer Nature in Estuaries and Coasts
- Vol. 49 (3) , 61
Open AccessAcclimation To Seasonal and Localized Physical Conditions Influences the Temperature-Productivity Response of Eelgrass, Zostera marina
Increasing ocean warming and marine heatwaves will have important consequences for the world’s seagrasses. Predicting impacts of thermal stress on seagrasses depends on understanding their temperature-productivity responses. Here, we examine these responses for eelgrass ( Zostera marina ) and how they are influenced by plant acclimation to seasonal and localized physical conditions. Eelgrass was collected seasonally from two sites in Nova Scotia (NS) Canada that were characterized by contrasting physical conditions, with temperature regime being a dominant feature. Laboratory incubations measured oxygen production and consumption rates by eelgrass leaves and roots-rhizomes from 5 to 40 °C. Unimodal temperature dependence of net photosynthetic rate and whole plant net productivity was modelled and maximum productivity ( P max ) and optimal temperatures ( T opt ) were estimated. Both net photosynthetic rate and net plant productivity were more peaked and bell-shaped across increasing temperature for plants from a warm highly variable temperature regime compared to plants from a cool less variable regime. P max and T opt were highest in warm relative to cool regime plants across all seasons. Furthermore, P max was lower and T opt was higher when seasonal ambient temperatures increased. T opt for whole plant productivity (22.5 to 28.5 °C) was used to create a thermal risk map for eelgrass, which indicated that 20%, 8.5%, 4.6%, and 1.3% of predicted eelgrass habitat in NS exceeds 22 °C, 24 °C, 26 °C and 28 °C, respectively, for at least 1 h annually. Our results show that even northern temperate eelgrass can be heat stressed, and that plant acclimation to seasonal and localized conditions should be considered when evaluating seagrass thermal risk.
- Open Access
- 2 January 2026
- Published by Springer Nature in Communications Earth & Environment
- Vol. 7 (1)
Blue carbon sequestration dominated by dissolved organic carbon pathways for kelp forests and eelgrass meadows in Nova Scotia, Canada
Existing estimates of carbon sequestration by coastal vegetated ecosystems (e.g., kelp forests, seagrass meadows) remain highly uncertain because they utilize few measurements with high uncertainty or focus on individual stocks or fluxes. Here, we combine empirical data with modeling to generate detailed carbon budgets for kelp forests and eelgrass meadows in Nova Scotia, Canada. Our budgeting showed that the release and export of dissolved organic carbon accounts for substantially more carbon sequestered by these habitats than has been estimated globally (98.5 ± 64.6% for kelp and 84 ± 36.0% for eelgrass), suggesting this carbon pathway has been undervalued. Further, kelp forests are estimated to sequester more carbon than eelgrass by ~1.3 orders of magnitude (27 ± 19% and 12.1 ± 6.3% of annual net primary production for kelp and eelgrass) mainly due to their large habitat area, highlighting their significant but largely overlooked role in ocean carbon sequestration relative to other blue carbon ecosystems.
- 2 January 2026
- Published by Springer Nature in Communications Earth & Environment
- Vol. 7 (1)
Open AccessBlue carbon sequestration dominated by dissolved organic carbon pathways for kelp forests and eelgrass meadows in Nova Scotia, Canada
Existing estimates of carbon sequestration by coastal vegetated ecosystems (e.g., kelp forests, seagrass meadows) remain highly uncertain because they utilize few measurements with high uncertainty or focus on individual stocks or fluxes. Here, we combine empirical data with modeling to generate detailed carbon budgets for kelp forests and eelgrass meadows in Nova Scotia, Canada. Our budgeting showed that the release and export of dissolved organic carbon accounts for substantially more carbon sequestered by these habitats than has been estimated globally (98.5 ± 64.6% for kelp and 84 ± 36.0% for eelgrass), suggesting this carbon pathway has been undervalued. Further, kelp forests are estimated to sequester more carbon than eelgrass by ~1.3 orders of magnitude (27 ± 19% and 12.1 ± 6.3% of annual net primary production for kelp and eelgrass) mainly due to their large habitat area, highlighting their significant but largely overlooked role in ocean carbon sequestration relative to other blue carbon ecosystems.
- Open Access
- 27 June 2025
- Published by American Association for the Advancement of Science (AAAS) in Science Advances
- Vol. 11 (26) , eadw1952
Pathways of blue carbon export from kelp and seagrass beds along the Atlantic coast of Nova Scotia
Coastal vegetated ecosystems are recognized for their role in cycling and storing carbon in the world’s oceans (i.e., blue carbon); however, high uncertainty in carbon sequestration rates is partly due to an absence of studies estimating carbon export to the deep sea. We modeled export from nearshore kelp forests and seagrass beds, showing variability by orders of magnitude across spatial scales (3 to hundreds of kilometers), kelp and seagrass species, seasons, and carbon forms, raising caution in using generalized export rates in blue carbon accounting. Our results also show rapid (20 to 30 days) and extensive export of neutrally buoyant dissolved organic carbon particles to the shelf break (up to 44% within 90 days), contrasting sinking particulate organic carbon particles that largely remained within 100-meter water depth in the nearshore. These results improve estimates of carbon sequestration by blue carbon ecosystems and reveal contrasting patterns of export relative to other regions of the globe.
- 27 June 2025
- Published by American Association for the Advancement of Science (AAAS) in Science Advances
- Vol. 11 (26) , eadw1952
Open AccessPathways of blue carbon export from kelp and seagrass beds along the Atlantic coast of Nova Scotia
Coastal vegetated ecosystems are recognized for their role in cycling and storing carbon in the world’s oceans (i.e., blue carbon); however, high uncertainty in carbon sequestration rates is partly due to an absence of studies estimating carbon export to the deep sea. We modeled export from nearshore kelp forests and seagrass beds, showing variability by orders of magnitude across spatial scales (3 to hundreds of kilometers), kelp and seagrass species, seasons, and carbon forms, raising caution in using generalized export rates in blue carbon accounting. Our results also show rapid (20 to 30 days) and extensive export of neutrally buoyant dissolved organic carbon particles to the shelf break (up to 44% within 90 days), contrasting sinking particulate organic carbon particles that largely remained within 100-meter water depth in the nearshore. These results improve estimates of carbon sequestration by blue carbon ecosystems and reveal contrasting patterns of export relative to other regions of the globe.
- Open Access
- 29 April 2021
- Published by American Geophysical Union (AGU) in Journal of Geophysical Research: Oceans
- Vol. 126 (5)
The Effect of Sea Ice on Tidal Propagation in the Kitikmeot Sea, Canadian Arctic Archipelago
The tides in the Kitikmeot Sea, in the southern Canadian Arctic Archipelago, are found to be heavily influenced by seasonal sea‐ice formation and blocking in a strait. Data from a moored current profiler in Dease Strait exhibit strong tidal damping during wintertime seasonal sea‐ice cover, with 50%–60% reduction in tidal elevation and 65% reduction in tidal velocities. We used a 3D unstratified barotropic numerical tidal model of the region to show that the observed wintertime tidal damping likely requires both high sea‐ice friction and a partial sea‐ice blockage in Victoria Strait (VS), the eastern gateway of the Kitikmeot Sea, where tidal‐induced ridging causes thick, rough ice to accumulate over its shallow sill. Our analysis of the model shows different dynamics for the M2 and K1 tides. Both tidal constituents are dominated by the Atlantic tides entering through VS. Overall, ∼90% of the M2 tidal energy entering VS does not reach Dease Strait because, in addition to dissipation in VS (∼25%), it is significantly diverted into adjoining bays and around an amphidrome in eastern Queen Maud Gulf (∼65%). In contrast, the smaller K1 tide is less diverted into the subsidiary channels, after ∼20% dissipation in VS and only ∼25% in adjoining bays, it propagates far into the region and is amplified due to resonance in Dease Strait and Coronation Gulf, resulting in larger tidal height than the M2 tide in Coronation Gulf.
- 29 April 2021
- Published by American Geophysical Union (AGU) in Journal of Geophysical Research: Oceans
- Vol. 126 (5)
Open AccessThe Effect of Sea Ice on Tidal Propagation in the Kitikmeot Sea, Canadian Arctic Archipelago
The tides in the Kitikmeot Sea, in the southern Canadian Arctic Archipelago, are found to be heavily influenced by seasonal sea‐ice formation and blocking in a strait. Data from a moored current profiler in Dease Strait exhibit strong tidal damping during wintertime seasonal sea‐ice cover, with 50%–60% reduction in tidal elevation and 65% reduction in tidal velocities. We used a 3D unstratified barotropic numerical tidal model of the region to show that the observed wintertime tidal damping likely requires both high sea‐ice friction and a partial sea‐ice blockage in Victoria Strait (VS), the eastern gateway of the Kitikmeot Sea, where tidal‐induced ridging causes thick, rough ice to accumulate over its shallow sill. Our analysis of the model shows different dynamics for the M2 and K1 tides. Both tidal constituents are dominated by the Atlantic tides entering through VS. Overall, ∼90% of the M2 tidal energy entering VS does not reach Dease Strait because, in addition to dissipation in VS (∼25%), it is significantly diverted into adjoining bays and around an amphidrome in eastern Queen Maud Gulf (∼65%). In contrast, the smaller K1 tide is less diverted into the subsidiary channels, after ∼20% dissipation in VS and only ∼25% in adjoining bays, it propagates far into the region and is amplified due to resonance in Dease Strait and Coronation Gulf, resulting in larger tidal height than the M2 tide in Coronation Gulf.