Large-Scale Drivers of Marine Heatwaves Revealed by Archetype Analysis

Abstract

Extreme oceanic events, such as marine heatwaves, can have disastrous impacts on ecosystems and marine industries. Given their potential consequences, it is important to understand how broad-scale climate variability influence the probability of local extreme marine events. Here, for the first time, we employ an advanced data-mining methodology, archetype analysis, to identify large scale climate drivers and teleconnections that lead to marine extremes in certain regions. This methodology is applied to the Australasian region, where it identifies instances of anomalous sea-surface temperatures, frequently associated with marine heatwaves, as well as the broadscale oceanic and atmospheric conditions associated with those extreme events. Additionally, we use archetype analysis to assess the ability of a low-resolution climate model to accurately represent the teleconnection patterns associated with extreme oceanic temperatures, and discuss the implications for the predictability of these impactful events.

Archetypal Analysis++: Rethinking the Initialization Strategy

“…Less common initialization strategies for archetypal analysis include -means, as used by Han et al (2022), and a coreset (Mair and Brefeld, 2019), as used by Black et al (2022) and Chapman et al (2022). Note that the coreset was proposed as a way to condense the data set into a smaller set for a more e cient training of archetypal analysis rather than as a way for initializing it.…”

“…Archetypal analysis has been applied, e.g., for global gene expression (Thøgersen et al, 2013), bioinformatics (Hart et al, 2015), apparel design (Vinué et al, 2015), chemical spaces of small organic molecules (Keller et al, 2021), geophysical data (Black et al, 2022), large-scale climate drivers (Hannachi and Trenda lov, 2017;Chapman et al, 2022), and population genetics (Gimbernat-Mayol et al, 2022).…”

“…Here, boundary points refer to points on the boundary of the convex hull of the data. Since then, FurthestSum has established itself as one of the default initialization strategies for archetypal analysis (Thøgersen et al, 2013;Hinrich et al, 2016;Mair and Brefeld, 2019;Abrol and Sharma, 2020;Beck et al, 2022;Black et al, 2022;Chapman et al, 2022;Gimbernat-Mayol et al, 2022).…”