Midlatitude circulation trends and the efficiency of poleward heat transport in the upper troposphere

急流是地球对流层中强烈的纬向风带，是中纬度地区气候的主要特征。在每个半球，急流的纬度处于热风平衡状态，赤道到极地的经向温度梯度@T=@y。气候模型预测，由于人为强迫，不同高度的纬向平均@T=@y的符号变化不同。随着边界变暖，模型建议在极地表面的快速变暖和热带上对流层(UT)的升温之间展开“拉锯战”[1]。极地表面的快速变暖在低层降低了@T=@y并将急流推向赤道，而热带上对流层(UT)的升温加剧了与急流向极地移动有关的高层的@T=@y。最先进的气候模型对人为强迫下的喷流转移的预测不同，反映了这场拉锯战结果的不确定性。大多数模型预测，到本世纪末，喷气式飞机将小幅向极地移动[2]。然而，过去40年的再分析数据表明，与同期的模型模拟相比，喷流的极地趋势更大[3]。此外，虽然极地表面变暖的放大在再分析数据中被很好地观察到[4]，但热带对流层上层的趋势更不确定[5]，一些观测数据集显示的变暖趋势比模式[6]小。最近的工作引起了人们对对流层上部极地涡热和动量通量的异常趋势的注意[3]，并提出，在纬向平均中，异常的涡流热通量[v*T*]与经向温度梯度通过简单的通量梯度关系[v*T*]=D@T/@yHere，[]联系在一起。表示纬向平均值，D表示某一扩散系数。假设是真实的大气比模型中更有效地将热量和动量输出到热带地区，这与涡流驱动的急流的较强趋势和热带UT较弱的变暖趋势相一致。在这个框架中，对热带加热越敏感的环流对应的扩散系数D值越高。

The jet streams are bands of intense zonal winds in the Earth's troposphere and are a dominant feature of climate in the midlatitudes. In each hemisphere, the jet's latitude is in thermal wind balance withthe meridional equator-to-pole temperature gradient, @T=@y. Climate models predict differently signed changes in the zonal-mean @T=@y at different altitudes in response to anthropogenic forcing. As theclimate warms, models suggest a \tug-of-war" [1] between rapid polar surface warming { which acts to reduce @T=@y at low levels and push the jet equatorward { and enhanced warming of the tropical uppertroposphere (UT), with an increased @T=@y at high levels associated with a poleward shift of the jet. State-of-the-art climate models differ in their projections of jet shifts under anthropogenic forcing,reflecting uncertainty in the outcome of this tug-of-war. Most models predict a small poleward shift of the jets by the end of the century [2]. However, reanalysis data covering the last 40 years suggest largerpoleward trends in the jets than are found in model simulations of the same period [3]. Additionally, while the amplification of polar surface warming is well observed in reanalysis data [4], trends in thetropical upper troposphere are more uncertain [5], with some observational datasets showing smaller warming trends than in models [6].Recent work has drawn attention to anomalous trends in poleward eddy heat and momentum fluxes in the upper troposphere [3] and has proposed that, in the zonal mean, the anomalous eddy heat flux[v*T*] is linked to the meridional temperature gradient by a simple flux-gradient relationship[v*T*] = D @T/@yHere, [.] denotes a zonal mean and D is some diffusivity coefficient. The hypothesis is that the real atmosphere exports heat and momentum out of the tropics more efficiently than in models, consistentwith both the stronger trends in the eddy-driven jets and weaker warming trends in the tropical UT. In this framework, a circulation that is more sensitive to tropical heating corresponds to a higher valueof the diffusivity D.