| High-resolution satellite observations show that wind stress is positively correlated with sea surface temperature(SST)at mesoscales in the extratropical oceans,which is quite different from large-scale ocean-atmosphere coupling;meanwhile,mesoscale air-sea coupling processes exert important effects on large-scale ocean circulation and climate.However,there is a lack of a theoretical framework that can systematically explain the mechanism that plays the dominant role in the atmospheric responses to mesoscale SST perturbations,which is still controversial.The effects of mesoscale ocean-atmosphere coupling processes on the ocean are also not well known.It is difficult to accurately simulate mesoscale ocean-atmosphere coupling in low-resolution coupled ocean-atmosphere models.Moreover,great uncertainties still exist in depicting mesoscale ocean-atmosphere coupling with differences in simulations among high-resolution models.The ability of the reanalysis product and high-resolution ocean-atmosphere coupled model is evaluated in terms of the characteristics of mesoscale ocean-atmosphere coupling;their products are used to systematically analyze the mechanisms for controlling the atmospheric responses to mesoscale SST perturbations.Furthermore,the feedback of atmospheric responses is investigated by a numerical model.The main work and conclusions are as follows:(1)The ability of the widely used ERA5 is evaluated in terms of the mesoscale surface wind responses to mesoscale SST perturbations,with a focus on three regions at midlatitudes,Kuroshio Extension(KE),Gulf Stream(GS),and Agulhas Return Current(ARC).The simulated mesoscale surface wind stress response strength from ERA5 is closer to that observed under stronger background wind and more unstable atmospheric boundary layer,which are regionally and seasonally dependent.In the ARC region,simulated wind response strength is close to that calculated from observations all year around;the same goes for the KE and GS regions during winter,while it respectively only accounts for 22% and 43% of that observed during summer.Further analyses highlight that mechanisms for controlling the atmospheric responses also have a dependence on regions and seasons.In the ARC region,pronounced dipole patterns of sea surface pressure(SLP)and vertical velocity perturbation responses,associated with strong vertical mixing indicate that the DMT exerts a dominant effect all the year.In the KE and GS regions,monopole response patterns and weak vertical mixing show the main role played by the PA mechanism for atmospheric adjustment in summer,while dipole patterns indicate the main role played by the DMT mechanism in winter.Both DMT and PA mechanisms play important roles in spring and summer.At the boreal midlatitudes,the PA mechanism is the dominant player during summer with a weak simulated coupling strength,being only half of the observed.As the PA mechanism is mainly manifested as vertical heat transport which adjusts sea surface air pressure,the weak coupling strength is suggestive of the inappropriate parameterization of vertical heat mixing in the atmospheric model.This provides clues for improving high-resolution atmospheric modeling.(2)The extent to which the mesoscale ocean-atmosphere coupling is captured in a high-resolution Community Earth System Model(CESM-HR)is assessed,to validate the authenticity and universality of the conclusion drawn from the analysis of ERA5.Focus on KE and ARC regions,the dominant mechanisms for the atmospheric responses are analyzed during summer and winter,respectively.The results show that both the simulated mesoscale surface wind response strength and the dominant mechanisms are regionally and seasonally dependent.Compared with satellite observations,it is found that the CESM-HR can well depict the characteristics of the mesoscale ocean-atmosphere coupling,with its strength being comparable to that observed;but in the KE region,the simulated strength is only half of the observed during summer.Furthermore,the response patterns of mesoscale atmospheric variable fields to SST perturbations on vertical cross-section are verified by the zonal lead and lag correlation and eddy composite analyses.In the ARC region,dipole patterns of SLP and vertical velocity perturbation responses indicate that the DMT exerts a dominant effect during winter and summer.In the KE region,monopole response patterns show the main role played by the PA mechanism during summer,while dipole patterns indicate the main role played by the DMT mechanism during winter.The results supplement and verify the conclusion of the first part.In addition,the weak coupling strength simulated by CESM-HR in the KE region during summer is also indicative of inappropriate parameterization of atmospheric thermal vertical mixing.This inference is applicable to other high-resolution atmospheric models.(3)A simple mesoscale air-sea coupled model is constructed based on the Regional Ocean Model System(ROMS).Numerical experiments are performed to explore the feedback on the ocean of the responses of mesoscale wind and air temperature to mesoscale SST perturbations: in the CTRL experiment,climatic forcing fields are used to drive the ocean;in the WIND experiment,forcing fields include mesoscale wind perturbations;in the ALL experiment,both mesoscale wind and air temperature perturbations are included in the forcing fields.As for the influences on currents,it is demonstrated that the feedback of mesoscale wind responses leads to negative wind energy input to mesoscale eddies,and more eddy kinetic energy is dissipated.In addition,more eddy kinetic energy(EKE)is converted to mean flow kinetic energy(MKE),and the conversion of eddy potential energy(EPE)to EKE is weaker.The transport of MKE by advection is also reduced.As result,the EKE and MKE are reduced,and the kinetic energy and velocity around the flow axis of the Kuroshio Extension are weakened.When both the feedback of mesoscale wind and air temperature responses are considered,positive wind energy is input to mesoscale eddies,which enhances EKE.In addition,the EKE to MKE conversion is enhanced,which provides energy for the maintenance of mean flow,and the EKE to EPE conversion is also weakened.These result in the enhance of both the EKE and the MKE compared with the WIND experiment,and the kinetic energy and velocity around the flow axis of the Kuroshio Extension are relatively enhanced.For the influences on ocean temperature,mesoscale wind responses suppress the amplitude of mesoscale SST perturbations,cooling warm regions and warming cold regions.The addition of mesoscale air temperature response has a stronger inhibition effect on the amplitude of mesoscale SST perturbations.Such a change in mesoscale SST will further affect the local sea surface net heat flux,that is,the upward heat flux increases in the warm SST regions,while it decreases in the cold SST regions.This change in the local net heat flux will in turn inhibit the mesoscale SST perturbations.In general,mesoscale wind response and mesoscale air temperature response have compensating effects on the large-scale ocean current,while they have superposition effects on the mesoscale SST perturbations.In this thesis,we evaluate the ability of high-resolution reanalysis product and coupled model in terms of the mesoscale ocean-atmosphere coupling and find that it is regionally and seasonally dependent.The analysis of atmospheric responses reveals that the dominant mechanism of the mesoscale wind responses to SST perturbations is also regionally and seasonally dependent.It is further pointed out that the simulated weak coupling strength in the northern mid-latitude regions during summer is indicative of inappropriate parameterization of atmospheric thermal vertical mixing.In addition,the important influence of the mesoscale wind and air temperature responses to the mesoscale SST perturbations on the oceanic largescale and mesoscale flow and temperature is described.These results have important implications for understanding the mechanism of mesoscale ocean-atmosphere interaction in midlatitude regions,improving the cognition level of multi-scale ocean-atmosphere interaction,and enhancing the characterization,simulation,and long-time prediction capability of high-resolution models. |