Citation: Hayashida, H., Matear, R. J., Strutton, P. G., & Zhang, X. (2020). Insights into projected changes in marine heatwaves from a high-resolution ocean circulation model. Nature Communications, 11(4352). https://doi.org/10.1038/s41467-020-18241-x
As the Earth’s climate warms, the ocean heats up as well (Figure 1). But due to its size and the complexity of its currents, the entire ocean doesn’t warm at exactly the same rate. Instead, some parts of the ocean remain at normal, predictable temperatures, while others experience events called marine heat waves (MHWs). These heat waves can be concerning for many reasons. Marine mammals can be pushed past their heat tolerance levels, local mangrove and kelp forests may be driven to local extinction due to heat stress, and coral bleaching can occur (Figure 2). Invasive species may also expand their range due to temperature fluctuations, altering local biodiversity. MHWs can therefore be devastating.
Furthermore, satellites that measure sea surface temperature have demonstrated that MHWs are becoming more of a problem with increasing climate change – in three specific ways. Heat waves are becoming more frequent (there are more individual events), longer (each event lasts for a longer period of time), and more intense (temperatures during the heat wave are warmer). A group of scientists from the Australian Research Council Centre of Excellence for Climate Extremes teamed up to determine where these events were likely to occur, and just how damaging they would be.
Building a Better Model
To do this, the scientists developed an extremely high-resolution ocean circulation model to simulate sea surface temperature in the past and future. “High-resolution” refers to the number of individual grid squares the ocean was divided into. In this Australian analysis, gridlines were drawn at every 0.1º of latitude and longitude on the ocean’s surface, with an additional 51 vertical water depth layers in a 3-dimensional “z” direction.
Before the Australian analysis, similar studies had relied on very coarse (low-resolution) models to estimate MHW characteristics. While coarse models are generally accurate when averaged over the whole ocean, they’re particularly bad at predicting events in nearshore areas, where coastlines and local currents add a layer of complexity (Figure 3). Given the number of critical marine resources in coastal regions (fisheries, mangrove and coral habitats, and tourism, to name a few), local governments remain very interested in more accurate predictions of MHWs in nearshore areas.
By comparing their new high-resolution model with 23 previously-developed coarser models, the researchers determined that their own model was the best at predicting future MHWs, especially near the coast. But without a time machine, how is this possible? Future heatwaves haven’t happened yet, so how could the researchers know that their own model was best at predicting them? After building their model, the Australian scientists fed it historical data, and compared the model’s projections for 1982-2018 with what actually happened. This is called model validation. Their model more accurately predicted real events than previous models, allowing them to be more confident about its performance.
By projecting into the future, the Australian model predicted that there will be large regional variability in ocean temperature between now and 2050. In particular, heatwaves within western boundary currents (fast coastal currents that run along the western boundary of each ocean; Figure 4) will become far more intense, though the length of individual heatwaves won’t increase much. Western boundary currents – such as the Gulf Stream, the Agulhas, and the Kuroshio Currents – are especially important in transporting warm water from the tropics to the polar regions. They help distribute the sun’s heat across the globe. In addition, these coastal currents flow through critically important ecosystems. A large increase in heatwave intensity may have far-reaching implications for local biodiversity and the coastal economy.
The Australian authors clearly demonstrated that high-resolution models are the gold standard for predicting dangerous marine heatwaves. By utilizing the extra computing power, such models expose the vulnerabilities of coastal ecosystems, and may allow coastal communities to more effectively adapt to a warmer future.