The Cretaceous period, a 'greenhouse' climate era, is characterized by relatively high global temperatures and a relatively weak latitudinal temperature gradient. Multiple paleoproxy records, such as oxygen isotopes in foraminifera (δ¹⁸O) and the TEX86 index, indicate a wide range of sea surface temperatures (SSTs) across latitudes during this period, with warmer temperatures at lower latitudes and cooler temperatures at higher latitudes. While changes in atmospheric CO₂, solar constant, and paleogeography are recognized as major drivers of SST variations, their individual contributions and underlying mechanisms remain unclear. Understanding the Cretaceous climate is crucial for improving our understanding of climate sensitivity and the response of the Earth's system to significant changes in greenhouse gas concentrations. This research addresses the knowledge gap by investigating the relative contributions of atmospheric CO₂ variability and paleogeographic changes to the mid-latitudinal SST gradient during the Cretaceous, using paleoclimate simulations and a fundamental model of wind-driven ocean gyres. The study specifically aims to isolate the impact of paleogeographically-driven changes in extratropical gyral circulation on the pole-to-tropics SST gradient. The importance of this research lies in its ability to refine our understanding of past climate dynamics and improve predictive capabilities for future climate scenarios, particularly in response to anthropogenic changes in greenhouse gas concentrations and potential changes in landmasses.

Previous studies have explored the influence of paleogeography on pole-to-tropics temperature gradients and ocean circulation during the Cretaceous-Paleogene-Eocene (CPE) period. However, the specific effects of paleogeographic changes on surface ocean circulation have not been thoroughly examined. Existing research has established a correlation between the aspect ratio of ocean basins and the intensity of surface gyres, impacting heat transport and meridional SST gradients. While Stommel's model provides a fundamental understanding of wind-driven gyres, applying it to past geological epochs requires considering the dynamic interplay between wind stress, basin geometry, and continental configurations. This study builds upon previous work by developing a simplified model that isolates the impact of paleogeographic changes on gyral circulation and subsequent effects on SST gradients, utilizing data from existing paleoclimate simulations to test this model.

The study utilizes data from two ensembles of paleoclimate simulations performed using the HadCM3L coupled atmosphere-ocean model. The model incorporates various climate feedbacks, including vegetation. In each ensemble, atmospheric CO₂ concentration was held constant (560 ppmv and 1120 ppmv, respectively, representing ×2 and ×4 preindustrial levels), while paleogeography and solar constant were varied across different geological ages. The chosen CO₂ concentrations aim to isolate the effect of CO₂ versus paleogeography and capture the range of variation observed during the CPE. Paleogeographies were derived from geological data. The solar constant increased monotonically through the simulations, but its impact was secondary compared to CO₂ and paleogeographic changes. The model simulations provided SST data for analysis. To quantify the impact of paleogeography on gyre circulation, the study employs a simplified model based on Stommel's work, considering the meridional and zonal extents of the North Pacific gyral basin (Ly and Lx, respectively) and the maximal volumetric transport (ψmax). The authors calculated Ly and Lx based on the zonally averaged wind stress and continental boundaries, applying a cosine(latitude)-weighted average to Lx. The meridional SST gradient (ΔSSTy) between 20°N and 50°N was calculated, and its relationship with ψmax was analyzed. The analysis assumed a first-order linear relationship between ΔSSTy and poleward heat transport (H), with H further assumed to be proportional to ψmax. A regression analysis was conducted to determine the relationship between ΔSSTy and ψmax, estimating parameters such as ΔSSTRad (meridional gradient at radiative equilibrium) and κ (an empirical constant). The study further compares the findings from the model with existing proxy-based reconstructions of meridional SST gradients.

The analysis of HadCM3L simulation data reveals a strong anti-correlation between the meridional gradient of zonally averaged SST (ΔSSTy) and the maximal volumetric transport (ψmax) in the mid-latitudes. The meridional extent of the North Pacific gyral basin (Ly) decreased non-monotonically through the Cretaceous, while the zonal extent (Lx) remained relatively constant. This decrease in Ly led to a reduction in ψmax and a corresponding decrease in poleward heat transport (H), resulting in an increase in ΔSSTy. The study found that the reduction in ψmax can explain a large portion (~75-80%) of the variance in ΔSSTy from the Berriasian to the Priabonian across both simulation ensembles. The increase in ΔSSTy from the Early to Late Cretaceous is primarily attributed to paleogeography-driven changes in ocean circulation, while the effect of pCO₂ changes on ΔSSTy, while present, is less significant, particularly in the Cretaceous. The model developed (ΔSSTy = ΔSSTRad / (1 + (κ × ψmax × ΔSSTRad)) effectively captures the general trend in ΔSSTy evolution across the CPE. The study notes a steep increase in ΔSSTy between ~130 Ma and ~100 Ma in the higher CO₂ simulation, which is further investigated, although a definitive explanation remains outside the scope of the study. The comparison with proxy-based data shows some discrepancies, potentially due to limitations in the spatial distribution of proxies. A stronger correlation is found between Lx and Ly, linked to increased land area in the polar North Pacific since the mid-Cretaceous. This highlights the importance of considering multiple factors in the study of past climates and their connection to the spatial distribution of oceans and land masses. The study concludes that the changes in ocean circulation, driven by paleogeographic changes, were a key driver in shaping the meridional temperature gradient during the Cretaceous. The influence of paleogeography on ocean circulation highlights the significance of tectonic processes in modulating past and future climate.

The findings of this study significantly advance our understanding of Cretaceous climate by quantifying the relative importance of paleogeographic changes and atmospheric CO₂ variability in shaping mid-latitudinal temperature gradients. The strong anti-correlation between oceanic heat transport and ΔSSTy demonstrates a critical link between tectonic processes and climate dynamics. The proposed model, based on fundamental principles of ocean circulation, successfully captures the observed trends, supporting the key role of paleogeographically driven changes in ocean gyre intensity and heat transport in modulating meridional SST gradients. The observed discrepancy between model results and proxy-based data highlights the need for further research considering the limitations of both model resolution and proxy data distribution. The study underscores the necessity of integrating geological and climate modeling approaches for a comprehensive understanding of past climate change. The insights gained are directly relevant to future climate projections, providing valuable context for understanding the complex interplay between greenhouse gas concentrations, continental configurations, and ocean circulation patterns in shaping global temperature distributions.

This study demonstrates that changes in Northern Hemisphere paleogeography, specifically the reduction in the aspect ratio of the North Pacific gyral basin, played a crucial role in increasing the mid-latitudinal SST gradient during the Cretaceous. While variations in atmospheric CO₂ also contributed to SST changes, paleogeographic changes had a more dominant effect. The simple model developed successfully captures the observed trends, highlighting the importance of considering ocean dynamics when investigating past climate variability. Further research integrating advanced climate models with higher-resolution paleogeographic reconstructions and more comprehensive proxy data is needed to refine our understanding of this intricate interplay.

The study acknowledges several limitations. The simplified model makes first-order assumptions about ocean heat transport and gyre dynamics, neglecting potential complexities of the real ocean. The use of a specific climate model (HadCM3L) might introduce model-specific biases. The limited spatial resolution of the model and the uneven spatial distribution of proxy data introduce uncertainty in comparing model results with paleoclimatic reconstructions. The study focuses mainly on the Northern Hemisphere, limiting its generalizability to other regions. Furthermore, the study focuses primarily on the long-term trend of CO2, neglecting shorter-term fluctuations which may have had significant climatic impacts.