Future mesoscale horizontal stirring in polar oceans intensified by sea ice decline

nature climate change Article https://doi.org/10.1038/s41558-025-02471-2 Future mesoscale horizontal stirring in polar oceans intensified by sea ice decline Received: 17 May 2024 Accepted: 25 September 2025 Gyuseok Yi 1,2 , June-Yi Lee 1,2,3 , Eun Young Kwon 1,2, Sun-Seon Lee Myeong-Hyeon Kim 1,2, Wonsun Park 1,2, Karl Stein 1,4 & Axel Timmermann 1,4 , 1,4 Published online: xx xx xxxx Check for updates Mesoscale horizontal stirring (MHS) is ubiquitous in the oceans, influencing heat and carbon transport, phytoplankton blooms and fish larvae dispersal. The current generation of Earth system models lacks sufficient resolution to properly resolve MHS-relevant small-scale phenomena, such as oceanic mesoscale eddies, leaving it largely unknown how MHS will change in response to greenhouse warming. Here we determine how CO2 doubling and quadrupling will change the surface MHS statistics in Community Earth System Model simulations with 1/10-degree ocean resolution. MHS is analysed using the finite-size Lyapunov exponent, a Lagrangian diagnostic that measures the separation of close trajectories. Projected increases in MHS are expected in the Arctic Ocean and coastal Antarctic regions, driven by enhanced time-mean ocean flow and turbulence which predominantly result from sea ice reduction. The enhanced horizontal stirring in polar oceans implies substantial yet uncertain consequences for tracer transport, nutrient supply and ecosystems under higher CO2 conditions. Stirring is a turbulent process that deforms and stretches fluid elements into elongated filaments, thereby dispersing fluid properties and generating sharp gradients1,2. These gradients are smoothed by diffusion and, together with stirring, this leads to irreversible homogenization (mixing). In the ocean, where horizontal velocities dominate over vertical ones across most scales, mesoscale horizontal stirring (MHS) is the primary dynamical process. MHS is closely linked to meso scale features such as eddies, meanders, fronts and filaments, which span tens to hundreds of kilometres and persist for days to months. It plays a pivotal role in regulating the transport of heat, carbon and other tracers3–6, phytoplankton blooms7–10, the dispersal of larvae and fish eggs11,12 and broader ecosystem interactions13–15. Given the wide-ranging impacts of MHS, understanding how MHS will respond to future climate change is particularly important in high latitudes, where warming is most strongly amplified. Recently, rapid sea ice decline due to greenhouse warming16–18 has driven major environmental changes in polar oceans, altering ocean temperature, salinity and surface momentum flux19–22, with potential consequences for diverse physical and biological processes23–25. Recent studies using state-of-the-art climate models have reported marked changes in upper-ocean circulation22 and a substantial increase in eddy activity26 in the Arctic under warming scenarios. Notably, a kilometre-scale high-resolution simulation showed a threefold increase in eddy kinetic energy (EKE) in the upper Arctic Ocean under a 4 °C-warmer climate, associated with enhanced baroclinic instability driven by sea ice loss26. Such dynamical changes are expected to markedly influence MHS. While a Lagrangian-based network theory study linked increased kinetic energy to stronger horizontal stirring in the Mediterranean under a warmer climate27, comparable analyses in polar regions are still lacking. Addressing this gap is crucial to improve understanding of oceanic responses to greenhouse warming in the most rapidly changing regions of the world. Here focusing on the polar oceans, we explore spatiotemporal changes in surface MHS under varying CO2 conditions. To characterize and assess changes in MHS, we use the finite-size Lyapunov exponent (FSLE)28,29, a Lagrangian diagnostic which measures the continuous Center for Climate Physics, Institute for Basic Science (IBS), Busan, Republic of Korea. 2Department of Climate System, Pusan National University, Busan, Republic of Korea. 3Research Center for Climate Sciences, Pusan National University, Busan, Republic of Korea. 4Pusan National University, Busan, Republic of Korea. e-mail: ; 1 Nature Climate Change Article Southern Ocean changes Arctic Ocean changes https://doi.org/10.1038/s41558-025-02471-2 a PD e PD i PD b 4 × CO2 f 4 × CO2 j 4 × CO2 c PD g PD k d 4 × CO2 h 4 × CO2 l 0 0.05 0.10 0.15 0.20 0.25 0.30 FSLE (d−1) 10 −2 −1 10 5 × 10 Current speed (m s–1) −1 0 PD 4 × CO2 20 40 60 80 100 Sea ice concentration (%) Fig. 1 | FSLE, current speed and sea ice concentration in the polar oceans under PD and 4 × CO2 conditions. a,b,e,f,i,j, Arctic Ocean snapshots of FSLE at 15-m depth (a,b), current speed (log scale) at 15-m depth (e,f) and sea ice concentration (i,j) from the PD (a,e,i) and 4 × CO2 (b,f,j) simulations at the vernal equinox (20 March) of the first analysed year (model year 130 for PD and model year 160 for 4 × CO2). c,d,g,h,k,l, Southern Ocean snapshots of FSLE at 15-m depth (c,d), current speed (log scale) at 15-m depth (g,h) and sea ice concentration (k,l) from the PD (c,g,k) and 4 × CO2 (d,h,l) simulations at the autumnal equinox (22 September). Credit: Basemaps from NASA Visible Earth (https://visibleearth. nasa.gov). exponential rate of separation between nearby particle trajectories, indicating how quickly a patch of passive tracers is stretched (Methods). The FSLE provides spatially and temporally resolved estimates of transport and mixing, revealing fine-scale features such as filaments and spirals that are often overlooked by Eulerian methods29,30. In this study, FSLE is calculated as the time-based growth rate of separation from an initial distance (δ0) to a final distance (δf ), defined as a tenfold increase. To target mesoscale structure, δ0 and δf are set to 0.1° and 1.0°, respectively. This choice of δ0 aligns with the 0.1° horizontal resolution of the ocean model, while δf is set to a scale comparable to that used in previous studies28,31 for consistency (see Supplementary Information for δ0–δf sensitivity tests). The FSLE technique has been successfully applied to identify complex spatial and seasonal patterns of surface MHS in both the Mediterranean and global ocean under present-day conditions28,31. To investigate how future greenhouse warming affects MHS, we analyse idealized century-long time-slice simulations conducted using the fully coupled ultra-high-resolution Community Earth System Model v.1.2.2 (CESM-UHR)32–34, with horizontal resolution of 0.25° for the atmosphere and 0.1° for the ocean. Notably, in the Arctic the ocean model uses a tripolar grid with 2.5-km eddy-permitting resolution. The three experiments34–37 use constant atmospheric CO2 Nature Climate Change Article https://doi.org/10.1038/s41558-025-02471-2 a 4 × CO2 2 × CO2 PD 1 b 0 10 −1 10 4 × CO2 2 × CO2 PD 1 10 Density (d) Density (d) 10 0 (...truncated)