Sea surface temperature (SST) is a critical link between the atmosphere and ocean, influencing various climatic phenomena and marine ecosystems. While variations in annual and seasonal mean SST have been studied extensively, the impact of changes in the amplitude of the SST seasonal cycle has received less attention. These changes can significantly influence marine heatwaves, monsoons, precipitation, and the El Niño-Southern Oscillation, along with oceanic oxygen content, a key factor in ocean productivity and biogeochemical cycles. Climate models consistently predict a global intensification of the SST seasonal cycle under future warming scenarios, primarily attributed to thermodynamic effects of atmospheric circulation changes and alterations in mixed layer depth (MLD). However, detecting such changes in historical observational records is challenging due to sparse data and internal variability. This study addresses this gap by analyzing observational data and model simulations to determine whether the intensification of the SST seasonal cycle has already emerged in recent decades and identifying the underlying mechanisms driving this change.

Previous research has established the importance of SST in climate and ecological processes. Studies have focused on variations in annual and seasonal mean SST, but the significance of changes in the amplitude of the SST seasonal cycle has not been fully explored until recently. Climate models consistently project a future intensification of the SST seasonal cycle under various warming scenarios, but the underlying mechanisms remain debated. Some studies emphasize changes in atmospheric circulation influencing surface heat fluxes, while others highlight the role of changes in MLD. As the Earth warms, increased ocean heat uptake leads to enhanced stratification and shallower MLD, reducing thermal inertia and intensifying SST seasonality. While future projections exist, observational evidence of this intensification in historical records has been limited. Existing studies primarily focused on land surface temperature trends, leaving the oceanic component largely unexplored.

The study used three widely-used observational SST datasets: ERSST version 5, HadISST version 1.1, and OISST version 2, covering 1982–2022. Pre-1982 data was excluded due to sparsity, particularly in high latitudes during winter. The amplitude of the SST seasonal cycle was calculated as the difference between annual maximum and minimum values. To separate the effects of internal variability and external forcing, the study employed five CMIP6 simulation ensembles (EXT) encompassing 181 individual realizations. The EXT ensemble included data from a multi-model ensemble (MME) and large ensembles (LENS) from ACCESS-ESM1-5, CanESM5, MIROC6, and MPI-ESM1-2-LR. The study also used DAMIP simulations to attribute the observed trend to four major radiative forcings: greenhouse gases (GHGs), anthropogenic aerosols (AERS), stratospheric ozone (StratO3), and natural forcings (NATs). A mixed layer budget analysis, using CMIP6 simulations and the IAP ocean temperature and salinity analysis, quantitatively assessed the contributions of thermal forcing, horizontal advection, and residual processes to SST seasonal cycle changes. Three monthly objective analyses ocean datasets (IAP, Ishii v7.3, and EN4) were used to examine changes in the seasonal cycle of mixed layer temperature and annual mean MLD. All datasets were interpolated to a 2° × 2° grid for consistent analysis. Statistical significance was determined using Student's t-test and considering an effective sample size to account for autocorrelation. The study also analyzed surface dissolved oxygen and air-sea CO2 flux data to explore implications of the intensified SST seasonal cycle.

The analysis revealed a statistically significant intensification of the global SST seasonal cycle by 0.16 ± 0.07 °C (3.9 ± 1.6% relative increase) from 1983 to 2022. The most substantial intensification occurred in the northern subpolar gyres (up to 11.9%). The EXT ensemble successfully reproduced the observed intensification, indicating that the changes are primarily a forced response to external forcings rather than internal variability. Attribution analysis using DAMIP simulations showed that increased GHGs were the primary driver (∼55.6% of the total enhancement), with a discernible contribution from reduced AERS, particularly in the Northern Hemisphere. Mixed layer budget analysis highlighted the dominant role of shallower MLD in intensifying the SST seasonal cycle, primarily due to enhanced upper-ocean stratification. Decreased MLD amplitude and increased ocean heat uptake also contributed, particularly in the North Pacific and North Atlantic. The intensification extended throughout the mixed layer, leading to an intensified seasonal cycle of dissolved oxygen, particularly in the subpolar North Atlantic, North Pacific, and Southern Ocean, which could exacerbate existing hypoxic conditions. While the intensified seasonal cycle of air–sea CO2 flux was observed, its spatial pattern differed from that of SST, suggesting involvement of other factors like surface winds. Future projections under SSP5-8.5 indicate a substantial intensification (∼10.6%) by 2100.

The findings demonstrate a clear human influence on the intensification of the SST seasonal cycle, primarily driven by GHGs and influenced by reduced AERS. The consistent results across observational datasets and model simulations strongly support the conclusion that external forcings are the primary driver. The dominant role of shallower MLD, due to enhanced upper-ocean stratification, is a key mechanistic explanation. The intensification's impact on the seasonal cycle of dissolved oxygen highlights significant ecological implications, especially concerning potential for increased hypoxia in already vulnerable regions. The spatial discrepancies between SST and air-sea CO2 flux intensification suggest that the mechanisms responsible for changes in air–sea CO2 flux are more complex and require further investigation.

This study provides robust evidence for a significant human-induced intensification of the global SST seasonal cycle, primarily driven by increasing GHGs and influenced by decreasing AERS. The key mechanism is the shoaling of the mixed layer, affecting the thermal inertia of the upper ocean. This intensification has substantial implications for marine ecosystems, particularly regarding the seasonal cycle of dissolved oxygen. Future research should focus on refining the understanding of regional variations, improving model representation of relevant processes, and assessing the cascading impacts on marine biodiversity and biogeochemical cycles. The study underscores the urgent need for mitigation strategies to address climate change and its far-reaching effects on the ocean.

The study acknowledges limitations in observational data, particularly in the Southern Hemisphere, where uncertainties might affect the accuracy of regional estimations. The use of a 2° × 2° grid for analysis could lead to some smoothing of finer-scale features. Additionally, the attribution analysis relies on CMIP6 models, which have inherent limitations and uncertainties. Future research could benefit from improved observational data, higher-resolution model simulations, and exploration of other potential drivers not explicitly considered in this analysis.