Elsevier

Journal of Marine Systems

Volume 196, August 2019, Pages 48-64
Journal of Marine Systems

Sensitivity of freshwater dynamics to ocean model resolution and river discharge forcing in the Hudson Bay Complex

https://doi.org/10.1016/j.jmarsys.2019.04.002Get rights and content

Highlights

  • Sensitivity of the HBC to model resolution and river runoff forcing is evaluated.

  • The annually averaged freshwater budget is a balance between river runoff and freshwater advected out of Hudson Bay.

  • Exchange of freshwater between Hudson Bay regions results from Ekman transport and the mean flow.

  • Estimated river runoff residence time for the HBC can be as long as 32 years.

Abstract

Hydroelectric development and regulation have modified the temporal and spatial distribution of runoff entering the Hudson Bay Complex (HBC). To understand the impacts and future of regulation in this region, the numerical ocean model, NEMO, run with the Arctic and Northern Hemispheric Atlantic (ANHA) configuration, is used to model present day freshwater dynamics associated with river runoff and sea ice melt. The present work establishes the freshwater budget in each subregion of the HBC, in addition to evaluating the sensitivity to model resolution and estimates of river discharge forcing. It is shown that the annually averaged HBC freshwater budget is mainly a balance between river discharge and freshwater advected out of the region, while surface fluxes (ice melt and growth, and precipitation and evaporation) are the dominant term on seasonal time scales. Runoff forcing is found to impact the long term mean volume and freshwater fluxes out of the HBC, while increased resolution has minimal effect on these fluxes, with the exception of the Southampton–Baffin Island gate. Quantitative estimates of turbulent, mean, and Ekman components of freshwater exchange between the interior and boundary regions of Hudson Bay are also presented. We use offline Lagrangian passive tracers to estimate the HBC runoff residence time, which is as long as 32 years.

Introduction

Anthropogenic changes such as global warming, which is causing an intensification of the hydrological cycle in the Arctic region (Zhang et al., 2012; Déry et al., 2009), as well as hydroelectric development, are changing the river discharge in northern Canada (Déry et al., 2011, Déry et al., 2016, Déry et al., 2018; MacDonald et al., 2018). One such region undergoing these changes is the Hudson Bay Complex (HBC), which includes Hudson Bay, James Bay, and neighbouring basins, Foxe Basin, and Hudson Strait and Ungava Bay, shown in Fig. 1. The HBC receives about 900 km3/year of river runoff, equivalent to roughly three times the Mackenzie River (Shiklomanov and Shiklomanov, 2003; Holmes et al., 2012), causing this region to be quite fresh compared to the Arctic Ocean. The main pathway of heat, mass, and freshwater exchange between the HBC, the Arctic, and the North Atlantic is via Hudson Strait. This river water flows out of Hudson Strait and along the coast of Labrador in the Labrador Sea, where deep convection occurs (Aagaard and Carmack, 1989; Straneo, 2006; Lazier et al., 2002; Lozier et al., 2019). The role of the fresh Hudson Strait outflow in these processes, however, is still largely unknown. This study aims to provide multi-year estimates of the HBC freshwater budget, so as to understand the role of model resolution and river discharge on freshwater fluxes within the HBC as well as to the North Atlantic.

Isolated from large scale ocean circulation, the main sources of freshwater to the HBC are river discharge and sea ice melt (Prinsenberg, 1988). On time scales less than a year, sea ice melt/growth has a much larger role in the freshwater budget compared to river discharge (Prinsenberg, 1988). Freshwater sourced from river discharge is found mainly along the coast, while freshwater from sea ice melt is distributed more equally around the bay (Granskog et al., 2007, Granskog et al., 2011).

Spatially, the distribution of freshwater within Hudson Bay can be divided into two regions, the outer boundary region and the interior region. The exchange of freshwater between the interior and boundary regions is mainly driven by Ekman transport (St-Laurent et al., 2011). In summer, the freshwater is imported into the interior, and is released during the fall (St-Laurent et al., 2012).

In the context of climate change, the length of the ice free season in the HBC is increasing, with both earlier break up in spring (Gough et al., 2004a; Gagnon and Gough, 2005; Castro de la Guardia et al., 2017; Hochheim and Barber, 2014; Kowal et al., 2017) and later freeze up in the fall (Gagnon and Gough, 2005; Castro de la Guardia et al., 2017; Hochheim and Barber, 2014; Kowal et al., 2017). These changes have been found to be related to the region's air temperature (Hochheim and Barber, 2014; McGovern and Gough, 2015). Hochheim and Barber (2010) found, for every 1C increase in the region's mean air temperature, it can result in a decrease of 105,000–117,000 km2 in late November sea ice extent with concentrations >80%. Sea ice thickness, on the other hand, is weakly related to air temperatures (Gough et al., 2004b). Ice thickness derived from satellite altimetry (Landy et al., 2017) shows a significantly asymmetrical spatial pattern across Hudson Bay in spring due to the strong cyclonic ice drift in winter. Their study also estimated 742 ±10 km3 of freshwater is stored in sea ice within the bay in April.

Anthropocentric influences have also impacted the HBC river discharge. Discharge entering the HBC has increased (Déry et al., 2016), which is associated with the intensified hydrological cycle in the context of Arctic warming (Déry et al., 2009, Déry et al., 2011, Déry et al., 2016; Zhang et al., 2012; Rawlins et al., 2010). Seasonally, hydroelectric development has increased winter HBC streamflow (Déry et al., 2011). Increasing air temperature also leads to earlier spring peak runoff (Déry et al., 2005; Gagnon and Gough, 2002), however, this varies regionally (Gagnon and Gough, 2002). Under the 1.5 and 2C future warming scenarios, MacDonald et al. (2018) found discharge increased in all seasons, except summer, with the largest increases occurring in spring.

In light of these current trends, it is still unclear as to the impact that these changes will have on the freshwater dynamics in this region. The annual net freshwater flux of river discharge is large, thus, changes in river runoff, by seasonal and spatial redistribution, or long term trends, lead to changes in seawater density and stability. A high runoff year and regulated discharge have been shown to lead to a decrease in salinity, along with a general increase in sea ice thickness (Saucier and Dionne, 1998).

To date, there have been no multi-year evaluations of the freshwater budget in this region. Other questions remain regarding the freshwater budget, for instance, how important are small scale processes in HBC dynamics? Does the freshwater budget change with changes in river runoff? To determine the sensitivity of the HBC to runoff forcing as well as model resolution, we use a general circulation ocean model coupled with a sea ice model to evaluate the freshwater budgets, pathways, and boundary-interior exchange processes of each simulation. The following section contains a description of the model, as well as the various datasets used in the numerical experiments. In Section 3, an evaluation of the model and the freshwater budgets for each subregion in the HBC, as well as boundary-interior freshwater exchanges and riverine water residence time, are shown. Our analysis of the residence time is in Section 3.4, preceding the summary and conclusions. This work is part of the BaySys project, a bay-wide initiative to investigate effects of hydroelectric regulation and climate change on various aspects of the Hudson Bay environment, such as the biogeochemical, biological, and physical components of the system.

Section snippets

Numerical model

We use a general circulation ocean model, based on the Nucleus for European Modelling of the Ocean version 3.4 (NEMO; Madec and the NEMO team, 2008), which is coupled to the sea ice model, Louvain-la-neuve Ice Model version 2 (LIM2) with elastic-viscous-plastic (EVP) rheology (Hunke and Dukowicz, 1997), and includes both thermodynamic and dynamic processes (Fichefet and Maqueda, 1997), for our simulations. We use the Arctic and Northern Hemisphere Atlantic (ANHA) configuration, which has two

Model evaluation

To evaluate the model, we first show spatial SST for the model and observations in Fig. 4a–j for both summer and fall. During the winter (January, February, March) and spring (April, May, June), the simulated SSTs are close to the freezing point (not shown) since the bay is ice covered. In fall, simulated SSTs (Fig. 4a–d) agree very well with observations (Fig. 4e). The temperature gradient from north to south in Hudson Bay is captured well by all simulations. Most simulations are too cold

Discussion

In this study we investigated the sensitivity of freshwater in the HBC to model resolution and runoff forcing. The results obtained here also have implications for pathways and residence times of various nutrients or pollutants commonly found in river discharge.

In terms of the freshwater budget, our estimates of surface fluxes are comparable to Prinsenberg (1988), with our peak freshwater fluxes in the summer of 18 km3/day compared to their 12 km3/day (including areas of both Hudson Bay and

Acknowledgements

We would like to thank Environment and Climate Change Canada for the use of the CGRF forcing fields, as well as the producers of GLORYS for the reanalysis data that we use to initialize our model simulations as well as providing our model with open boundary conditions. Thank you to Dr. Gregory Smith who provided the CGRF atmospheric forcing to force our ocean model. This work is part of the BaySys project, thus we thank the Natural Sciences and Engineering Research Council of Canada and

References (75)

  • F. Straneo et al.

    The outflow from Hudson Strait and its contribution to the Labrador Current

    Deep-Sea Res. I

    (2008)
  • K. Aagaard et al.

    The role of sea ice and other fresh water in the Arctic circulation

    Journal of Geophysical Research: Oceans

    (1989)
  • J.C.M. Andersson et al.

    Key factors for improving large-scale hydrological model performance

  • J. Bamber et al.

    Recent large increases in freshwater fluxes from Greenland into the North Atlantic

    Geophys. Res. Lett.

    (2012)
  • F.G. Barber

    Current observations in Fury and Hecla Strait

    J. Fish. Res. Board Can.

    (1965)
  • B. Blanke et al.

    Variability of the tropical Atlantic Ocean simulated by a general circulation model with two different mixed-layer physics

    J. Phys. Oceanogr.

    (1993)
  • B. Blanke et al.

    Kinematics of the Pacific equatorial undercurrent: an Eulerian and Lagrangian approach from GCM results

    J. Phys. Oceanogr.

    (1997)
  • B. Blanke et al.

    Warm water paths in the equatorial Atlantic as diagnosed with a general circulation model

    J. Phys. Oceanogr.

    (1999)
  • E. de Boisséson et al.

    Origin, formation and variability of the Subpolar Mode Water located over the Reykjanes Ridge

    Journal of Geophysical Research: Oceans

    (2012)
  • P. Bougeault et al.

    Parameterization of orography-induced turbulence in a mesobeta-scale model

    Mon. Weather Rev.

    (1989)
  • L. Castro de la Guardia et al.

    Sea ice cycle in western Hudson Bay, Canada, from a polar bear perspective

    Mar. Ecol. Prog. Ser.

    (2017)
  • A. Dai et al.

    Estimates of freshwater discharge from continents: latitudinal and seasonal variations

    J. Hydrometeorol.

    (2002)
  • A. Dai et al.

    Changes in continental freshwater discharge from 1948 to 2004

    J. Clim.

    (2009)
  • M. Defossez et al.

    Comparing winter and summer simulated estuarine circulations in Foxe Basin, Canada

    Atmosphere-Ocean

    (2012)
  • S.J. Déry et al.

    Flow alteration impacts on Hudson Bay river discharge

    Hydrol. Process.

    (2018)
  • S.J. Déry et al.

    Characteristics and trends of river discharge into Hudson, James, and Ungava Bays, 1964–2000

    J. Clim.

    (2005)
  • S.J. Déry et al.

    Observational evidence of an intensifying hydrological cycle in northern Canada

    Geophys. Res. Lett.

    (2009)
  • S.J. Déry et al.

    Recent trends and variability in river discharge across northern Canada

    Hydrol. Earth Syst. Sci.

    (2016)
  • K.F. Drinkwater

    On the mean and tidal currents in Hudson Strait

    Atmosphere–Ocean

    (1988)
  • D.S. Dukhovskoy et al.

    Greenland freshwater pathways in the sub-Arctic seas from model experiments with passive tracers

    Journal of Geophysical Research: Oceans

    (2016)
  • T. Fichefet et al.

    Sensitivity of a global sea ice model to the treatment of ice thermodynamics and dynamics

    Journal of Geophysical Research: Oceans

    (1997)
  • A.S. Gagnon et al.

    Hydro-climatic trends in the Hudson Bay Region, Canada

    Canadian Water Resources Journal/Revue canadienne des ressources hydriques

    (2002)
  • A.S. Gagnon et al.

    Trends in the dates of ice freeze-up and breakup over Hudson Bay, Canada

    Arctic

    (2005)
  • P. Gaspar et al.

    A simple eddy kinetic-energy model for simulations of the oceanic vertical mixing - tests at station Papa and long-term upper ocean study site

    Journal of Geophysical Research: Oceans

    (1990)
  • A. Gelfan et al.

    Climate change impact on the water regime of two great Arctic rivers: modeling and uncertainty issues

    Clim. Chang.

    (2017)
  • L.C. Gillard et al.

    Meltwater pathways from marine terminating glaciers of the Greenland ice sheet

    Geophys. Res. Lett.

    (2016)
  • W. Gough et al.

    Trends in seasonal sea ice duration in southwestern Hudson Bay

    Arctic

    (2004)
  • Cited by (24)

    • Accuracy of high resolution coastal flow speed simulations during and outside of wind, wave and stratification events (Gulf of Lion, NW Mediterranean)

      2023, Journal of Marine Systems
      Citation Excerpt :

      The better model performance at the shallow stations could be due to the refinement of the horizontal spatial resolution, thanks to the adaptive resolution of the curvilinear grid. Increasing the resolution of model configurations have been tested to improve agreement with other types of observations than flow speeds, sometimes showing predictions improvements (Thoppil et al., 2011; Kirtman et al., 2012; Putman and He, 2013; Ringler et al., 2013; Akhtar et al., 2018; Kvile et al., 2018; Ridenour et al., 2019). In addition to relative bias which indicates goodness of transport predictions, the present study evaluated the correlation between simulated and observed flow speed, an indicator generally disregarded.

    • Storm-driven hydrography of western Hudson Bay

      2021, Continental Shelf Research
      Citation Excerpt :

      The Bay occupies approximately 831,000 km2, making it the world's largest inland sea, and is characterized by a high annual volume of river discharge (712 km3; Déry et al., 2005; 2011) and a dynamic seasonal ice cover that exists from November/December to June/July (Hochheim and Barber, 2010, 2014). Prevailing northwesterly winds (i) create a pronounced gradient in ice thickness across the Bay with a mean ice thickness in April ranging from 1.2 m to 1.7 m from western to eastern Hudson Bay (Landy et al., 2017) and (ii) drive cyclonic circulation of water within the Bay (e.g., Ingram and Prinsenberg, 1998; Saucier et al., 2004; St-Laurent et al., 2011; Ridenour et al., 2019; Dmitrenko et al., 2020). In terms of sea ice, northwesterly winds advect the ice cover across the Bay, creating a recurrent polynya along the floe edge in northwestern Hudson Bay (Gunn, 2014; Bruneau et al., submitted; Fig. 1) and thicker deformed ice in eastern Hudson Bay (Landy et al., 2017; Kirillov et al., 2020).

    • Modelling the advection of pollutants in the Hudson Bay complex

      2021, Journal of Marine Systems
      Citation Excerpt :

      Their work found a strong correspondence in the mean geostrophic velocities between the model integration and satellite observation. Furthermore, Ridenour et al. (2019a) also showed that model's sea ice velocity and thickness compared well with satellite observations over the HBC. An example is shown in Fig. 2, where the sea ice concentration in the Hudson Bay Complex from the model is compared to satellite passive microwave data (available via: https://nsidc.org/data/G02202/).

    View all citing articles on Scopus
    1

    Now at Bedford Institute of Oceanography, Dartmouth, Nova Scotia, Canada.

    View full text