Society interacts with the ocean most frequently along its boundaries, and the eastern boundary upwelling systems (EBUS) of the Pacific and Atlantic Ocean are among the most biologically productive regions of the world ocean. Primary productivity and biodiversity in EBUS are sustained by coastal upwelling, where the equatorward alongshore winds push surface water offshore, allowing its replacement by cool, nutrient-rich water from deeper depths. The upwelled water nourishes phytoplankton photosynthesis, which fuels a rich marine ecosystem and supports major commercial fisheries (Fig. 1).

Figure 1. Schematic of the upwelling process in the California Current EBUS. Reprinted with permission from https://www.oregonconservationstrategy.org/oregon-nearshore-strategy/habitats/.

While these EBUS occupy only about 1% of the ocean’s surface, they produce about 20% of the global fish catch. The California Current EBUS (CCE), along the U.S. west coast, was worth about $39 billion in 2018 and supports ~220,000 jobs in Washington, Oregon and California. Recreational fisheries add about another 35,000 jobs and $4 billion in sales (NOAA 2021). West coast fishing communities are a critical part of the local economy and are inextricably linked to the health of the coastal ecosystem, where changes in the offshore oceanic temperature structure are displacing marine species towards the poles by up to several hundred kilometers. Thus, it is critical to understand how these regions will respond to continued anthropogenic climate change if we are to anticipate and minimize undesirable outcomes. We were funded by the U.S. National Science Foundation through its Convergence Accelerator program to provide support to the fishing industry along the U.S. west coast that will ensure the sustainability of the regional fishery in the face of such potential changes. Most of our work was conducted at the Texas Advanced Computer Center, using Frontera, one of the most powerful supercomputers in the world (Stanzione et al., 2020). Frontera allowed us to perform the unprecedented set of high-resolution climate simulations that show the greatly improved realism in coastal upwelling we discuss here.

While most large-scale observations and models suggest that average global phytoplankton productivity has declined over the past few decades and will continue to do so because of changes in upper-ocean density structure, the situation in EBUS is more complex. An early attempt to address this issue (Bakun, 1990), assumed that warming will affect the land more than the ocean, and that the increased land-ocean temperature difference will result in stronger pressure gradients, alongshore winds, and more upwelling.

Climate models are used routinely to explore the potential impacts and mechanisms of climate change. A study by Wang et al. (2015), using nominal 1° horizontal resolution models from the Coupled Model Intercomparison Project, Phase 5 (CMIP5), showed stronger alongshore winds and intensified upwelling in EBUS that agreed qualitatively with Bakun (1990). One problem, however, is that these model horizontal scales are too coarse to fully represent key components of the upwelling process, such as oceanic fronts or the observed alongshore, narrow, coastal wind jets. The result is an exaggerated heat transport towards the poles, resulting in water near the coast biased warmer by up to 4°C than observations.

In contrast, our paper (Chang et al., 2023) examines the problem using a much finer resolution model, the Community Earth System Model, version 1.3, which has a nominal horizontal resolution of 0.25° for the atmosphere and land components and 0.1° for the ocean and sea-ice components. A comparison of this high-resolution simulation with a standard, nominal 1° simulation and in situ observations shows evident improvement and much reduced bias. Fig. 2 shows sea surface temperature (SST) comparisons for the CCE, with a much narrower, colder upwelling zone at high-resolution in good agreement with the observations. This is virtually absent in the low-resolution output. The improvements are largely due to improved low-level winds near the coast and their associated wind stress and wind stress curl. The other three EBUS show similar improvements in the high-resolution simulations (see our paper at https://doi.org/10.1038/s43247-023-00681-0)

Figure 2. Comparison between the mean 1991-2020 SST (°C) for (a) the observations (b) the high-resolution, and (c) the low-resolution CESM simulations for the CCE.

 Using this new high-fidelity climate model, we examined projected changes in wind stress, wind stress curl, and ocean temperature over the rest of the century.  Contrary to the results of Wang et al. (2015), our results suggest different effects in the different EBUS, with projected alongshore wind stress reduced in the CCE and Canary Current, but strengthened off Chile and southern Africa, although temperature is projected to increase in all regions. Converting these results to an upwelling index, based on vertical velocities at the Ekman depth (the depth at which the cross-shelf flow is 20% of that at the surface), suggest that upwelling will decrease in future in the CCE and off Peru, increase off Chile and Namibia, and be highly variable off northwest Africa. These projected changes are much more complex than those suggested by Bakun (1990), as it seems that in some EBUS upwelling can increase in some regions even as the winds drop.

We contend that the discrepancy results from Bakun ignoring horizontal heat transport from the tropics. Indeed, heat budget calculations suggest warming from 0.16 Wm^-2 in the Canary system to 0.27 Wm^-2 in Peru and the CCE, but with different components of the heat budget contributing in each region. In the southern hemisphere, horizontal advection is the dominant contributor to warming, counteracted by heat losses via surface fluxes. The opposite occurs in the northern hemisphere, with the net surface heat flux into the ocean being more important.

Summing up, our high-resolution, high-fidelity model projections suggest that the impacts of anthropogenic climate change in EBUS are considerably more complex than previously thought, and that each EBUS has unique dynamics that do not fit the simple Bakun (1990) hypothesis. For the CCE, our model projects a decline in upwelling strength, putting at risk the highly productive ecosystem along the U.S. west coast. Our results also highlight that we must ensure that future models correctly represent fine-scale features such as the coastal wind jets, and differentiate between the effects of local upwelling rates and larger-scale advective processes if we are to correctly predict how EBUS, and the CCE in particular, will evolve. Similarly, we need to add biogeochemistry and fisheries components to these high-resolution models to assess changes in the biology and sustainability of these ecosystems.

References:

Bakun, A. (1990). Global climate change and intensification of coastal ocean upwelling. Science 247, 198–201.

Chang, P. et al. (2023). Uncertain future of sustainable fisheries environment in eastern boundary upwelling zones under climate change (https://doi.org/10.1038/s43247-023-00681-0).

NOAA (2021). Fisheries Economics of the United States 2018. U.S. Dept. of Commerce, NOAA Tech. Memo. NMFS-F/SPO-225, 246 p.

Stanzione, D. et al. (2020). Frontera: The Evolution of Leadership Computing at the National Science Foundation. In Practice and Experience in Advanced Research Computing (PEARC ’20), July 26–30, 2020, Portland, OR, USA. ACM, New York, NY, USA, 11 pages.

Wang, D., Gouhier, T. C., Menge, B. A. & Ganguly, A. R. (2015). Intensification and spatial homogenization of coastal upwelling under climate change. Nature 518, 390–394.