The Flexible Global Ocean-Atmosphere-Land System Model, Spectral Version 2: FGOALS-s2

The Flexible Global Ocean-Atmosphere-Land System model, Spectral Version 2 （FGOALS-s2） was used to simulate realistic climates and to study anthropogenic influences on climate change. Specifically, the FGOALS-s2 was integrated with Coupled Model Intercomparison Project Phase 5 （CMIP5） to conduct co- ordinated experiments that will provide valuable scientific information to climate research communities. The performances of FGOALS-s2 were assessed in simulating major climate phenomena, and documented both the strengths and weaknesses of the model. The results indicate that FGOALS-s2 successfully overcomes climate drift, and realistically models global and regional climate characteristics, including SST, precipita- tion, and atmospheric circulation. In particular, the model accurately captures annual and semi-annual SST cycles in the equatorial Pacific Ocean, and the main characteristic features of the Asian summer monsoon, which include a low-level southwestern jet and five monsoon rainfall centers. The simulated climate variabil- ity was further examined in terms of teleconnections, leading modes of global SST （namely, ENSO）, Pacific Decadal Oscillations （PDO）, and changes in 19th-20th century climate. The analysis demonstrates that FGOALS-s2 realistically simulates extra-tropical teleconnection patterns of large-scale climate, and irregu- lar ENSO periods. The model gives fairly reasonable reconstructions of spatial patterns of PDO and global monsoon changes in the 20th century. However, because the indirect effects of aerosols are not included in the model, the simulated global temperature change during the period 1850 2005 is greater than the observed warming, by 0.6℃. Some other shortcomings of the model are also noted.

Changes in Winter Stratospheric Circulation in CMIP5 Scenarios Simulated by the Climate System Model FGOALS-s2

Diagnosis of changes in the winter stratospheric circulation in the Fifth Coupled Model Intercomparison Project （CMIP5） scenarios simulated by the Flexible Global Ocean-Atmosphere-Land System model, second version spectrum （FGOALS-s2）, indicates that the model can generally reproduce the present climatology of the stratosphere and can capture the general features of its long-term changes during 1950-2000, including the global stratospheric cooling and the strengthening of the westerly polar jet, though the simulated polar vortex is much cooler, the jet is much stronger, and the projected changes are generally weaker than those revealed by observation data. With the increase in greenhouse gases （GHGs） effect in the historical simu- lation from 1850 to 2005 （called the HISTORICAL run） and the two future projections for Representative Concentration Pathways （called the RCP4.5 and RCP8.5 scenarios） from 2006 to 2100, the stratospheric response was generally steady, with an increasing stratospheric cooling and a strengthening polar jet ex- tending equatorward. Correspondingly, the leading oscillation mode, defined as the Polar Vortex Oscillation （PVO）, exhibited a clear positive trend in each scenario, confirming the steady strengthening of the polar vortex. However, the positive trend of the PVO and the strengthening of the polar jet were not accompa- nied by decreased planetary-wave dynamical heating, suggesting that the cause of the positive PVO trend and the polar stratospheric cooling trend is probably the radiation cooling effect due to increase in GHGs. Nevertheless, without the long-term linear trend, the temporal variations of the wave dynamic heating, the PVO, and the polar stratospheric temperature are still closely coupled in the interannual and decadal time scales.

Seasonal Evolution of Subtropical Anticyclones in the Climate System Model FGOALS-s2

The simulation characteristics of the seasonal evolution of subtropical anticyclones in the Northern Hemisphere are documented for the Flexible Global Ocean-Atmosphere-Land System model, Spectral Version 2 （FGOALS-s2）, developed at the State Key Laboratory of Numerical Modeling for Atmospheric Sciences and Geophysical Fluid Dynamics, the Institute of Atmospheric Physics. An understanding of the seasonal evolution of the subtropical anticyclones is also addressed. Compared with the global analysis established by the European Centre for Medium-Range Forecasts, the ERA-40 global reanalysis data, the general features of subtropical anticyclones and their evolution are simulated well in both winter and summer, while in spring a pronounced bias in the generation of the South Asia Anticyclone（SAA） exists. Its main deviation in geopotential height from the reanalysis is consistent with the bias of temperature in the troposphere. It is found that condensation heating （CO） plays a dominant role in the seasonal development of the SAA and the subtropical anticyclone over the western Pacific （SAWP） in the middle troposphere. The CO biases in the model account for the biases in the establishment of the SAA in spring and the weaker strength of the SAA and the SAWP from spring to summer. CO is persistently overestimated in the central-east tropical Pacific from winter to summer, while it is underestimated over the area from the South China Sea to the western Pacific from spring to summer. Such biases generate an illusive anticyclonic gyre in the upper troposphere above the middle Pacific and delay the generation of the SAA over South Asia in April. In mid- summer, the simulated SAA is located farther north than in the ERA-40 data owing to excessively strong surface sensible heating （SE） to the north of the Tibetan Plateau. Whereas, the two surface subtropical anticyclones in the eastern oceans during spring to summer are controlled mainly by the surface SE over the two continents in the Northern Hemisphere, which are simulated reasonably well, albeit with their centers shifted westwards owing to the weaker longwave radiation cooling in the simulation associated with much weaker local stratiform cloud. Further improvements in the related parameterization of physical processes are therefore identified.

FGOALS-s2 Simulation of Upper-level Jet Streams over East Asia: Mean State Bias and Synoptic-scale Transient Eddy Activity

Upper-level jet streams over East Asia simulated by the LASG/IAP coupled climate system model FGOALS-s2 were assessed, and the mean state bias explained in terms of synoptic-scale transient eddy activ- ity （STEA）. The results showed that the spatial distribution of the seasonal mean jet stream was reproduced well by the model, except that following a weaker meridional temperature gradient （MTG）, the intensity of the jet stream was weaker than in National Centers for Environment Prediction （NCEP）/Department of Energy Atmospheric Model Inter-comparison Project II reanalysis data （NCEP2）. Based on daily mean data, the jet core number was counted to identify the geographical border between the East Asian Sub- tropical Jet （EASJ） and the East Asian Polar-front Jet （EAPJ）. The border is located over the Tibetan Plateau according to NCEP2 data, but was not evident in FGOALS-s2 simulations. The seasonal cycles of the jet streams were found to be reasonably reproduced, except that they shifted northward relative to reanalysis data in boreal summer owing to the northward shift of negative MTGs. To identify the reasons for mean state bias, the dynamical and thermal forcings of STEA on mean flow were examined with a focus on boreal winter. The dynamical and thermal forcings were estimated by extended Eliassen-Palm flux （E） and transient heat flux, respectively. The results showed that the failure to reproduce the tripolar-pattern of the divergence of E over the jet regions led to an unsuccessful separation of the EASJ and EAPJ, while dynamical forcing contributed less to the weaker EASJ. In contrast, the weaker transient heat flux partly explained the weaker EASJ over the ocean.

Attributing Analysis on the Model Bias in Surface Temperature in the Climate System Model FGOALS-s2 through a Process-Based Decomposition Method

This study uses the coupled atmosphere–surface climate feedback–response analysis method（CFRAM） to analyze the surface temperature biases in the Flexible Global Ocean–Atmosphere–Land System model, spectral version 2（FGOALS-s2）in January and July. The process-based decomposition of the surface temperature biases, defined as the difference between the model and ERA-Interim during 1979–2005, enables us to attribute the model surface temperature biases to individual radiative processes including ozone, water vapor, cloud, and surface albedo; and non-radiative processes including surface sensible and latent heat fluxes, and dynamic processes at the surface and in the atmosphere. The results show that significant model surface temperature biases are almost globally present, are generally larger over land than over oceans, and are relatively larger in summer than in winter. Relative to the model biases in non-radiative processes, which tend to dominate the surface temperature biases in most parts of the world, biases in radiative processes are much smaller, except in the sub-polar Antarctic region where the cold biases from the much overestimated surface albedo are compensated for by the warm biases from nonradiative processes. The larger biases in non-radiative processes mainly lie in surface heat fluxes and in surface dynamics,which are twice as large in the Southern Hemisphere as in the Northern Hemisphere and always tend to compensate for each other. In particular, the upward/downward heat fluxes are systematically underestimated/overestimated in most parts of the world, and are mainly compensated for by surface dynamic processes including the increased heat storage in deep oceans across the globe.

Precipitation Changes in Wet and Dry Seasons over the 20th Century Simulated by Two Versions of the FGOALS Model

Seasonal precipitation changes over the globe during the 20th century simulated by two versions of the Flexible Global Ocean-Atmosphere-Land System （FGOALS） model are assessed. The two model versions differ in terms of their AGCM component, but the remaining parts of the system are almost identical. Both models reasonably reproduce the mean-state features of the timings of the wet and dry seasons and related precipitation amounts, with pattern correlation coefficients of 0.65-0.84 with observations. Globally averaged seasonal precipitation changes are analyzed. The results show that wet sea- sons get wetter and the annual range （precipitation difference between wet and dry seasons） increases during the 20th century in the two models, with positive trends covering most parts of the globe, which is consistent with observations. However, both models show a moistening dry season, which is opposite to observations. Analysis of the globally averaged moisture budget in the historical climate simulations of the two models shows little change in the horizontal moisture advection in both the wet and dry seasons. The globally averaged seasonal precipitation changes are mainly dominated by the changes in evaporation and vertical moisture advection. Evaporation and vertical moisture advection combine to make wet seasons wetter and enhance the annual range. In the dry season, the opposite change of evaporation and vertical moisture advection leads to an insignificant change in precipitation. Vertical moisture advection is the most important term that determines the changes in precipitation, wherein the thermodynamic component is dominant and the dynamic component tends to offset the effect of the thermodynamic component.

基于耦合气候系统模式的中全新世黄、东海海表通量分析

对中全新世（6,ka时期）海洋和气候的研究可加深人们对现阶段气候变化和海洋环境的认识，为预测未来海洋与气候环境变化提供一个重要参照．文章分析一个耦合气候系统模式 FGOALS-s2.0的模式结果，首先对其工业革命前（0,ka 时期）东亚地区夏季降水及冬、夏季10,m 风场的模拟结果进行评估，然后进一步对中全新世和工业革命前黄、东海海表大气强迫的季节变化进行了对比．结果显示：模式模拟出0,ka 时期东亚夏季降水从东南洋面至西北内陆减少的空间分布特点，冬、夏季10,m风场亦与观测大体一致；6,ka时期夏季，黄、东海风速较0,ka时期增大约0.8,m/s，16%左右；黄海风应力旋度值为正，东海为负，与0,ka 时期相比旋度绝对值均增大；同时，两海区接收的太阳短波辐射较0,ka 时期均增加，短波辐射的差异是中全新世夏季黄、东海海表的净热吸收增加的主要因子．6,ka 时期冬季，黄、东海北风加强，东海增加量在0.5～1.0,m/s，幅度约为10%，较黄海更为明显；两海区在冬季的净热释放也较0,ka 时期增大，东海释放更甚；冬季黄、东海风应力旋度较0,ka时期则无太大差别．研究表明，由于6,ka时期太阳辐射季节循环的改变，造成了黄、东海夏季风增强，海表净热通量也发生相应变化，该时期大气强迫场的变化可能会使黄、东海表层水温分布趋势发生较大改变，进而影响陆架环流格局．

Evaluation of Snow Depth and Snow Cover Fraction Simulated by Two Versions of the Flexible Global Ocean–Atmosphere–Land System Model

Based on historical runs,one of the core experiments of the fifth phase of the Coupled Model Intercomparison Project （CMIP5）,the snow depth （SD） and snow cover fraction （SCF） simulated by two versions of the Flexible Global OceanAtmosphere-Land System （FGOALS） model,Grid-point Version 2 （g2） and Spectral Version 2 （s2）,were validated against observational data.The results revealed that the spatial pattern of SD and SCF over the Northern Hemisphere （NH） are simulated well by both models,except over the Tibetan Plateau,with the average spatial correlation coefficient over all months being around 0.7 and 0.8 for SD and SCF,respectively.Although the onset of snow accumulation is captured wellby the two models in terms of the annual cycle of SD and SCF,g2 overestimates SD/SCF over most mid-and high-latitude areas of the NH.Analysis showed that g2 produces lower temperatures than s2 because it considers the indirect effects of aerosols in its atmospheric component,which is the primary driver for the SD/SCF difference between the two models.In addition,both models simulate the significant decreasing trend of SCF well over （30°-70°N） in winter during the period 1971-94.However,as g2 has a weak response to an increase in the concentration of CO2 and lower climate sensitivity,it presents weaker interannual variation compared to s2.