Essential krill species habitat resolved by seasonal upwelling and ocean circulation models within the large marine ecosystem of the California Current System

Abstract

Due to their global distribution, high biomass and energy content, euphausiids (krill) are important prey for many mid and upper trophic level marine organisms. Understanding drivers of krill habitat is essential for forecasting range shifts, and to better understand the response of krill predators to climate change. We hypothesized that the distribution and abundance of krill species derived from ecosystem surveys in spring/summer relates to geomorphic features, coastal upwelling during the preceding winter, and spring mesoscale oceanographic conditions. To test this hypothesis, we used boosted regression trees with environmental data and ocean model output to quantify the habitat associations of two primary krill species (Euphausia pacifica and Thysanoessa spinifera) in the central California Current Ecosystem from 2002 to 2018. Models confirmed the neritic distribution of T. spinifera and pelagic, outer slope association of E. pacifica (deviance explained ~35%). Distribution of these species were influenced by depth and bottom rugosity; chlorophyll-a concentrations and increased winter upwelling conditions; and spring surface currents and wind stress. Thysanoessa spinifera and E. pacifica abundance responded negatively (positively) to warm (cold) climate events, confirming known relationships. As an independent evaluation of krill models, observations of krill predator (seabirds, marine mammals) distribution indicated they were present within habitats of predicted high krill species abundance. Our framework indicates species-specific habitat relationships for these foundational forage species and their negative response to large-scale climate variations, such as El Niño and marine heatwave conditions. The approach can be easily transferred to other ecosystems or krill species that respond to similar ocean and climate forcing.

Pelagic Biodiversity, Ecosystem Function, and Services: An Integrated Observing and Modeling Approach

“…Overall, the locations of seabird and mammal biodiversity hotspots (six primary regions) identify important colonies and breeding centers as well as known regional and mesoscale variability of ocean conditions within the CCE (Figure 6b). These hotspots are similar to observed and modeled krill hotspots (Santora et al, 2011;Messié and Chavez, 2017;Cimino et al, 2020;Fiechter et al, 2020), and they overlap with large submarine canyon systems, especially those canyons that intersect the continental shelf and terminate close to the coastline (Figures 6 and 7; Santora et al, 2018).…”

“…The diverse spatiotemporal observation record is a resource for parameterizing and evaluating models that simulate krill population dynamics and distribution (Figure 7). For example, observed krill spatiotemporal dynamics have been used to evaluate the performance of mechanistic coupled ocean-ecosystem models (Santora et al, 2013;Fiechter et al, 2020), Lagrangian individualbased models (particle-tracking models that incorporate diel vertical migration behavior to simulate krill swarms; Dorman et al, 2015;Santora et al, 2017c), and Lagrangian growth and advection models (Messié and Chavez, 2017); they have also been employed to train krill species distribution models (SDMs) built on hydrographic model output and satellite metrics (Cimino et al, 2020). All krill ecosystem models capture the seasonal cycle and resolve the location of observed krill hotspots and their dissociation from strong upwelling centers, as well as previously observed temporal variability.…”

Factors affecting the seasonal distribution and biomass of E. pacifica and T. spinifera along the Pacific coast of Canada: A spatiotemporal modelling approach

“…In agreement with studies further south in the CCE, our models indicate that euphausiid hotspots were orientated alongshore in areas associated with the shelf break and specific depth contours [14][15][16]. This study adds to the large body of literature that suggests that a combination of depth, and the edges that occur along topographic features such as the continental slope and seafloor valleys and canyons, are unequivocally important to the development of areas of higher euphausiid biomass [15,25,[67][68][69].…”

“…We found similarities between the overall distribution of adults, and hotspots of both species were highest over the continental slope, particularly off the west coast of Haida Gwaii. Larger differences may have been observed between the distribution of these two species in waters deeper than 1000m [14], however, unfortunately sampling points in deeper waters were too sparse to be included in our models.…”

“…Euphausiid research in the California Current ecosystem (CCE) has focused on the physical mechanisms that lead to regions of elevated biomass, or 'hotspots', of euphausiids e.g. [12][13][14][15][16][17][18]. The distribution, intensity and temporal persistence of these areas has been quantitatively linked with local euphausiid predators [17,[19][20][21][22][23][24].…”