We combine our previous ice-sheet and climate models to address abrupt climate changes pertaining to Heinrich (H) and Dansgaard-Oeschger (DO) cycles as well as last deglaciation punctuated by Younger Dryas (YD). We posit their common origin in the calving of the ice sheet but differentiate thermal triggers by geothermal-heat/surface-melt in calving inland/marginal ice, the respective sources of H/DO-cycles. The thermal switches would produce step-like freshwater fluxes to endow abruptness to the resulting climate signals characterized by millennial timescale due to the internal ice dynamics. For an eddying ocean, its response to the freshwater perturbation entails millennial adjustment to maximum entropy production, which would cause sudden post-H warming followed by gradual cooling to form the H-cycles, and the above-freezing warmth (hence surface-melt) would calve the marginal ice to generate DO-cycles anchored on the cooling trend to form the Bond cycle. Since there is already ablation of the Holocene icecap, there would be self-sustained DO-cycles, which thus retain the same pacing as their glacial counterparts to resolve this seeming puzzle. This millennial pacing also transcends the deglaciation to account for its observed sequence although the occurrence of YD requires a boost of the freshwater flux by the rerouted continental meltwater. It is seen that by differentiating thermal triggers of the ice calving and incorporating MEP adjustment of the ocean, the theory has provided an integrated account of the genesis of the abrupt climate changes and their deduced anatomies bear strong resemblance to the observed ones, in support of the theory.

Most of the classic wind-driven circulation theories based on the Sverdrup balance have neglected the profound influence of eddy mixing on the large-scale distribution of the potential vorticity (PV), thus failing to explain some prominent features of the observed circulation. In this study, using a series of numerical experiments based on the MITgcm, we diagnose the PV balance to quantify the effect of eddy mixing on the subtropical gyre. Four grid-spacings of 1, 1/3.2, 1/10, and 1/32 degrees are selected to compare the structure of the upper-ocean circulation. In the 1° grid case, the structure of the thermocline is as predicted by the Sverdrup dynamics, with its maximum depth located in the subtropical interior where the wind stress curl is strongest. With increasing resolution, however, this maximum depth is displaced toward the subtropical front, which more closely resembles the observed thermocline. From 1° to 1/32°, the enhanced eddy mixing tends to homogenize the macroscopic PV in the subtropical gyre and reduces the latitudinal PV range to about 25% of the non-eddy solution; and the region where the Sverdrup balance holds is relegated to isolated patches, with its area reduced by about 60%. Furthermore, sensitivity experiments show that the observed thermocline structure is well reproduced in eddy-resolving runs, indicating that the PV mixing provides a better explanation of the subtropical circulation than the Sverdrup dynamics. Our results suggest that the Sverdrup relationship should be treated carefully in the eddy-rich region, even in the subtropical interior.

The summer air temperature that regulates the ice margin covaries with the sea surface temperature but precedes the ice volume during glacial cycles, suggesting that the ocean is the intermediary of the orbital forcing of the ice sheet. We formulate a minimal model to elucidate the ocean role in the genesis of glacial cycles. We show that, because of the atmospheric coupling, an eddying ocean exhibits bistates of maximum entropy production, which would translate to ice bistates of polar ice cap and Laurentide ice sheet, enabling large ice-volume signal when the bistable interval is crossed by the forcing. Since this bistable interval is set by the global convective flux, it is lowered during Pleistocene cooling, whose interplay with the ice-albedo feedback may account for the mid-Pleistocene transition from 41-ky obliquity cyles to 100-ky ice-age cycles paced by eccentricity. Quantitative tests of the theory and its parsimony in resolving myriad glacial puzzles suggest that the theory has captured the governing physics of the Pleistocene glacial cycles.

This two-part paper considers the general circulation of the atmosphere (Part 1) and ocean (Part 2) within the deductive framework of our climate theory, which aims to derive the earth’s generic climate state from first principles. Because the planetary fluids are inherently turbulent, such state is a macroscopic manifestation of a nonequilibrium thermodynamic system, whose closure involves the maximum entropy production, a veritable generalization of the second fundamental law. The logical progression detailed in the preceding papers of the theory has reduced the planetary fluids to warm and cold thermal masses and determined their bulk properties, which provide the prior constraints for the present dynamical derivation. Consistent with the asymptotic thermal state, we assume the potential vorticity (PV) to be homogenized in thermal masses to derive the upper-bound general circulations. In Part 1, this upper bound is seen to resemble the prevailing wind, forsaking therefore discordant explanations of the easterly trades and the polar jet stream. In this Part 2, we show again that this upper-bound may reproduce the observed general ocean circulation, suggesting that the latter may be explained by PV mixing — in place of the laminar Sverdrup dynamics. Together with Part 1, we posit that the general planetary circulations are the maximum flow extractable by random eddy mixing when subjected to differential solar heating.