3-D ocean particle tracking modeling reveals extensive vertical movement and downstream interdependence of closed areas in the northwest Atlantic

Abstract

Novel 3-D passive particle tracking experiments were performed in the northwest Atlantic to elucidate connectivity among areas closed to protect vulnerable marine ecosystems. We examined (1) the degree of vertical movement of particles released at different depths and locations; (2) the location of potential source populations for the deep-sea taxa protected by the closures; and (3) the degree of functional connectivity. A long-term oceanographic dataset (EN4) was queried to characterize the temperature and salinity regimes in each of the closed areas as a basis for interpreting recently published climate change projections. Using the Parcels Lagrangian particle tracking framework and the BNAM hydrodynamic model, we found enhanced connectivity over previously developed 2-D models and unexpected, current-driven, strong (to a maximum of about 1340 m) downward displacement at depth (450, 1000 and 2250 m), with weaker upward displacement except for the release depth of 2250 m which showed upward movement of 955 m with a drift duration of 3 months. The current velocities create down-stream interdependence among closed areas and allow redundancy to develop in some of the areas of the network, with some of the larger areas also showing retention. Source populations for sponges in the upstream closure are likely in adjacent waters of the Canadian continental shelf. Collectively this information can be used to inform management decisions related to the size and placement of these closed areas, and vertical velocity surfaces have potential for use in species distribution modeling of benthic species and habitats.

Climate-Change Refugia for the Bubblegum Coral Paragorgia arborea in the Northwest Atlantic

“…(Kenchington et al, 2019b). Experiments were therefore run for 2 weeks, 1 month and 3 months (Wang et al, 2020) to encompass all likely durations. Hilario et al ( 2015) report a range of maximum PLD for five alcyonacean corals, including one gorgonian species, of 7 to 90 days, consistent with this selection of time frames.…”

“…Predictive models of species distributions, when presented with associated levels of uncertainty, enable managers to make informed decisions on the trade-offs implicit in marine spatial planning (Lester et al, 2013;Grorud-Colvert et al, 2014;Carr, 2019;Metaxas et al, 2019), while insight into the impacts of climate change on the size and configuration of suitable habitat, can improve network design (Santos et al, 2020). Meanwhile, Lagrangian particle tracking (LPT) models are valuable for assessing connectivity (e.g., Bracco et al, 2019;Kenchington et al, 2019b;Metaxas et al, 2019;Zeng et al, 2019;Wang et al, 2020;Wang et al, 2021), and can provide independent support for the evaluation of species distribution models (Kenchington et al, 2019b;Wang et al, 2021). In LPT models, virtual "particles" are advected by flow fields derived from numerical ocean models.…”

Life in the Fast Lane: Modeling the Fate of Glass Sponge Larvae in the Gulf Stream

“…However, those results have been challenged (Lanna and Riesgo, 2020) and even if valid, there remains a lack of information on the directionality of sponge larval swimming behavior needed to make use of the information in particle tracking models. Vertical distribution has important consequences for larval transport and dispersal trajectories (Edwards et al, 2007;Gary et al, 2020;Wang et al, 2020). However, the vertical movements of V. pourtalesii larvae, like those of many sponge species, remain unknown (Kenchington et al, 2019), although settlement of V. pourtalesii has been observed on artificial substrates about 5 m off the bottom (Busch et al, 2020).…”

“…However, the two ocean models differ in a number of key aspects: ( 1 (2) BNAM is a prognostic model, with outputs obtained from governing equations directly, which accords with the physical processes. In contrast, GLORYS12V1 is a data-assimilative model, created by assimilating model output from the NEMO 3.1 ocean model and the LIM2 EVP sea ice model with observational data (Reijnders, 2020); (3) The surface forcing of BNAM is taken from a combination of CORE (Coordinated Ocean-ice Reference Experiments) and NCEP/NCAR reanalysis forcing. Model forcing variables include air temperature, wind velocities and humidity; daily short-and long-wave radiation, and total precipitation (rain plus snow) (Wang et al, 2019).…”

“…However, that approach can only provide some of the required information and often does not consider the ability of a species to disperse into isolated areas of predicted suitable habitat. More generally, an understanding of connectivity within and between spatially separated populations is important for habitat management in order to design effective protection networks and enable restoration of damaged habitats (Kool et al, 2013), and thus attracts broad interest from both scientists and managers (Simons et al, 2013;Kenchington et al, 2019;Ross et al, 2020;Wang et al, 2020). Knowledge of connectivity can enhance our comprehension of species distributions and is an important supplement to species distribution modeling, particularly if species responses under future climate projections are to be interpreted reliably (Holloway et al, 2016).…”