Influences of various forcing variables on global energy balance during the period of intensive instrumental observation (1958–1987) and their implications for paleoclimate

Consistent, accurate, and numerous measures of global scale atmospheric variables have been collected since about 1958. A time series of 30 years duration was assembled to investigate contributing factors to the global energy balance. The El Nino-Southern Oscillation (ENSO), CO₂ changes, and variation in solar energy output account for a quarter or more each of the variability in global energy balance. Upper atmospheric aerosols contribute, but less significantly and in a more complex way. The analysis suggests a hypothesis that has bearing on global climatic stability. Global climate fortuitously passed through a shift from a warmer NH to a warmer SH during the study period. The ENSO appears to act as a hemispheric energy balancing mechanism. There were significant changes in global atmospheric function when the hemispheric energy balance shifted in favor of the Southern Hemisphere about 1966. When applied to past climates, hemispheric dominance of global climate and related patterns of periodic stability could explain the rise and fall of some complex hierarchical social systems.     

Transitivity properties of surface temperature and ice cover in the CCM1

A frequently made assumption in simple models of long-term climatic behavior (e.g. ‘energy balance’ models) is that, owing to instability engendered by the ice-albedo feedback, the climatic system (surface temperature, in particular) can exhibit multiple steady states within the paleoclimatologically observed range of temperature. Here we show that, for a more comprehensive model than an energy balance model, the CCM1 general circulation model, such a bimodality does not exist for present values of the solar constant and atmospheric CO₂ even if one excludes the existence of a seasonal cycle of radiative forcing. Thus, no evidence is found to support the whole class of ice-age theories tuned to present CO₂ levels that depend on this bimodality. As a corollary, support is found for the idea that surficial temperature and snow-sea-ice cover are essentially ‘slaved’, fast-response, climatic variables that equilibrate uniquely with the prescribed external forcing and the slow-response climatic variables (e.g. the ice sheets and deep ocean state). It is also implied that, although care should be exercised in the choice of initial conditions to minimize computer time, the selection of initial conditions is not likely to affect the final outcome of general circulation model studies of climate for fixed (near-present) values of the solar constant and CO₂ forcing.     

Mechanisms for projected future changes in south Asian monsoon precipitation

Results are first presented from an analysis of a global coupled climate model regarding changes in future mean and variability of south Asian monsoon precipitation due to increased atmospheric CO₂ for doubled (2 × CO₂) and quadrupled (4 × CO₂) present-day amounts. Results from the coupled model show that, in agreement with previous studies, mean area-averaged south Asian monsoon precipitation increases with greater CO₂ concentrations, as does the interannual variability. Mechanisms producing these changes are then examined in a series of AMIP2-style sensitivity experiments using the atmospheric model (taken from the coupled model) run with specified SSTs. Three sets of ensemble experiments are run with SST anomalies superimposed on the AMIP2 SSTs from 1979–97: (1) anomalously warm Indian Ocean SSTs, (2) anomalously warm Pacific Ocean SSTs, and (3) anomalously warm Indian and Pacific Ocean SSTs. Results from these experiments show that the greater mean monsoon precipitation is due to increased moisture source from the warmer Indian Ocean. Increased south Asian monsoon interannual variability is primarily due to warmer Pacific Ocean SSTs with enhanced evaporation variability, with the warmer Indian Ocean SSTs a contributing but secondary factor. That is, for a given interannual tropical Pacific SST fluctuation with warmer mean SSTs in the future climate, there is enhanced evaporation and precipitation variability that is communicated via the Walker Circulation in the atmosphere to the south Asian monsoon to increase interannual precipitation variability there. This enhanced monsoon variability occurs even with no change in interannual SST variability in the tropical Pacific.     

An examination of climate sensitivity for idealised climate change experiments in an intermediate general circulation model

Radiative forcing and climate sensitivity have been widely used as concepts to understand climate change. This work performs climate change experiments with an intermediate general circulation model (IGCM) to examine the robustness of the radiative forcing concept for carbon dioxide and solar constant changes. This IGCM has been specifically developed as a computationally fast model, but one that allows an interaction between physical processes and large-scale dynamics; the model allows many long integrations to be performed relatively quickly. It employs a fast and accurate radiative transfer scheme, as well as simple convection and surface schemes, and a slab ocean, to model the effects of climate change mechanisms on the atmospheric temperatures and dynamics with a reasonable degree of complexity. The climatology of the IGCM run at T-21 resolution with 22 levels is compared to European Centre for Medium Range Weather Forecasting Reanalysis data. The response of the model to changes in carbon dioxide and solar output are examined when these changes are applied globally and when constrained geographically (e.g. over land only). The CO₂ experiments have a roughly 17% higher climate sensitivity than the solar experiments. It is also found that a forcing at high latitudes causes a 40% higher climate sensitivity than a forcing only applied at low latitudes. It is found that, despite differences in the model feedbacks, climate sensitivity is roughly constant over a range of distributions of CO₂ and solar forcings. Hence, in the IGCM at least, the radiative forcing concept is capable of predicting global surface temperature changes to within 30%, for the perturbations described here. It is concluded that radiative forcing remains a useful tool for assessing the natural and anthropogenic impact of climate change mechanisms on surface temperature.     

Climatic variability within an equilibrium greenhouse simulation

A simulation of the possible consequences of a doubling of the CO₂ content of the atmosphere has been performed with a low resolution global climatic model. The model included the diurnal and seasonal cycles, computed sea ice amount and cloud cover, and used implied oceanic heat fluxes to represent transport processes in the oceans. A highly responsive 2-layer soil moisture formulation was also incorporated. Twenty year equilibrated simulations for control (1 × CO₂) and greenhouse (2 × CO₂) conditions were generated. The major emphasis of the analysis presented here is on the intra-annual and interannual variability of the greenhouse run with respect to the control run. This revealed considerable differences from the time-averaged results with occasions of marked positive and negative temperature deviations. Of particular interest were the periods of negative temperature departures compared to the control run which were identified, especially over the Northern Hemisphere continents. Temporal and spatial precipitation and soil moisture anomalies also occurred, some of which were related to the surface temperature changes. Substantial sea surface temperature anomalies were apparent in the greenhouse run, indicating that a source of climatic forcing existed in addition to that due to doubling of the CO₂. Comparison of the intra-annual and interannual variability of the control run with that of the greenhouse run suggests that, in many situations, it will be difficult to identify a greenhouse signal against the intrinsic natural variability of the climatic system.     

二维能量平衡模式对若干气候问题的研究

Based on a two-dimensional energy balance model, the studies on some climatic issues such as the re- lationship between ice cap latitude and solar constant, desertifieation, and the warming effect of carbon dioxide, have been reviewed and discussed. The phenomenon that a fixed solar constant might correspond to different equilibrium ice cap latitudes is determined by the continuity of albedo distribution. The disconti- nuity in albedo distribution increases the number of equilibrium ice cap latitudes. Desert would expand both northward and southward when desert surface albedo is increasing. This would deteriorate the ecological environment in border regions, and then threaten the existence of local inhabitants. Melting of the polar ice would not be accelerated, with increasing carbon dioxide concentration. The ice cap latitude would move northward slowly, with some “hiatus” periods, under the slowly increasing global average surface tempera- ture. According to the current research, future development of the two-dimensional energy balance model and possible progress are also forecasted.     

Natural climatic variability and the Norse settlements in Greenland

A multi-millennial simulation with the CSIRO Mark 2 coupled global climatic model has been used to determine whether climatic conditions approximate to those experienced by the medieval Norse settlers in Greenland could be identified. The aim of this analysis was to see whether such conditions could be replicated by the natural climatic variability in this unforced simulation, in order to counteract claims that the current observed global warming is merely another example of this type of climatic regime. This view has been expressed in the media in an attempt to refute the existence of a CO₂-induced global warming. A 291-year period of above-average temperature followed by a 41-year cooler period were identified in one millennium of the simulation, and subsequently used as an analogue of conditions representative of the time of the Norse settlements. Considerable interannual variability existed in both these periods, but with noticeable positive and negative surface temperature anomalies in the warm and cold periods respectively. Thus the warm period was not a time of uniform benign conditions. Above-average precipitation was also associated with the warm period, and these climatic conditions would have enhanced pasture growth and hay production (the only crop the Norse produced) thereby sustaining the livelihood of the Norse Greenlanders. The climatic conditions associated with the cold period in the model were probably sufficient to limit the survival prospects of the settlers, especially when other, probably more critical, deleterious factors are taken into account. The temperature anomalies replicated in the simulation are similar to the limited proxy data, but may be smaller in magnitude: nevertheless they appear to be sufficiently large to have affected the viability of the Norse Greenlanders. After considering possible climatic mechanisms that could have contributed to these warm and cold periods it was concluded that they are simply a consequence of stochastic influences generated by nonlinear processes in the simulation. Thus this simulation provides no support for the contention that the current global warming is a manifestation of conditions prevailing during the Norse settlements in Greenland.     

Atmospheric radiative feedbacks associated with transient climate change and climate variability

This study examines in detail the ‘atmospheric’ radiative feedbacks operating in a coupled General Circulation Model (GCM). These feedbacks (defined as the change in top of atmosphere radiation per degree of global surface temperature change) are due to responses in water vapour, lapse rate, clouds and surface albedo. Two types of radiative feedback in particular are considered: those arising from century scale ‘transient’ warming (from a 1% per annum compounded CO₂ increase), and those operating under the model’s own unforced ‘natural’ variability. The time evolution of the transient (or ‘secular’) feedbacks is first examined. It is found that both the global strength and the latitudinal distributions of these feedbacks are established within the first two or three decades of warming, and thereafter change relatively little out to 100 years. They also closely approximate those found under equilibrium warming from a ‘mixed layer’ ocean version of the same model forced by a doubling of CO₂. These secular feedbacks are then compared with those operating under unforced (interannual) variability. For water vapour, the interannual feedback is only around two-thirds the strength of the secular feedback. The pattern reveals widespread regions of negative feedback in the interannual case, in turn resulting from patterns of circulation change and regions of decreasing as well as increasing surface temperature. Considering the vertical structure of the two, it is found that although positive net mid to upper tropospheric contributions dominate both, they are weaker (and occur lower) under interannual variability than under secular change and are more narrowly confined to the tropics. Lapse rate feedback from variability shows weak negative feedback over low latitudes combined with strong positive feedback in mid-to-high latitudes resulting in no net global feedback—in contrast to the dominant negative low to mid-latitude response seen under secular climate change. Surface albedo feedback is, however, slightly stronger under interannual variability—partly due to regions of extremely weak, or even negative, feedback over Antarctic sea ice in the transient experiment. Both long and shortwave global cloud feedbacks are essentially zero on interannual timescales, with the shortwave term also being very weak under climate change, although cloud fraction and optical property components show correlation with global temperature both under interannual variability and transient climate change. The results of this modelling study, although for a single model only, suggest that the analogues provided by interannual variability may provide some useful pointers to some aspects of climate change feedback strength, particularly for water vapour and surface albedo, but that structural differences will need to be heeded in such an analysis.     

Climate change and the regulation of the surface moisture and energy budgets

The increase in the vigor of the hydrological cycle simulated in a 2 × CO₂ experiment with the Canadian Climate Centre general circulation model is smaller than that obtained by other models which have similar increases in mean surface temperature. The surface energy budget, which encompasses also the moisture budget for the oceans, is analyzed. Changes in the net radiative input to and sensible heat flux from the surface act to warm it. This is balanced, at the new equilibrium, by a change in the latent heat flux which acts to cool it. Although this same general behavior is seen in other models, the increase in radiative input to the surface in the CCC GCM is smaller than in other models while the change in the sensible heat flux is of similar size. As a consequence, the latent heat flux required for balance is smaller. The comparatively small increase in the net radiative input at the surface occurs because of a decrease in the solar component. On average the decrease in solar input in the tropical region outweighs the higher latitude increase associated with the snow/ice albedo feedback. The notable tropical decrease in solar input occurs because the albedo of the clouds increase enough in this region to outweigh a small decrease in cloud amount. The increase in cloud albedo in the warmer and moister tropical atmosphere is a consequence of the parameterized cloud optical properties in the model which play an important role in the regulation of the surface energy and moisture budgets. The results demonstrate some of the consequences of the negative feedback mechanism associated with increasing cloud albedo in the model. They also suggest that the simulated change in the vigor of the hydrological cycle is not a simple function of the average increase in surface temperature but is a consequence of all of the processes in the model which control the available energy at the surface as a function of latitude.     

Future Changes in Biological Activity in the North Pacific Due to Anthropogenic Forcing of the Physical Environment

Many studies have examined the physical changes expected in the environment as a result of anthropogenic forcing. These physical changes will have an effect on ecosystems as well. In this study, a nitrogen-phytoplankton-zooplankton (NPZ) model is used to examine the effects of changes in the physical environment on primary productivity in the North Pacific ocean. The physical variables considered are mixed layer temperature and depth, solar insolation, and large-scale upwelling. The changes in these fields by the 2090s are taken from a coupled ocean-atmosphere general circulation model forced by projected atmospheric CO₂ and sulfates, then applied to the NPZ biological model. Theresult is a change in the seasonal cycle of phytoplankton and herbivore concentrations across the subpolar North Pacific, moving from a regime characterized by strong variability with low wintertime values and a spring bloom, to much more constant yearly values. A reduction of yearly-averaged primary productivity accompanies much of this shift to more constant year-round conditions. In other regions, productivity increases as warmer surface waters enable higher growth rates. Changes in mixed layer temperature and depth account for almost all the changes in productivity; model-predicted changes in surface insolation and large-scale upwelling have little impact.     

Impact of clouds on the surface radiation balance of the Arctic Ocean

Summary The relationship between clouds and the surface radiative fluxes over the Arctic Ocean are explored by conducting a series of modelling experiments using a one-dimensional thermodynamic sea ice model. The sensitivity of radiative flux to perturbations in cloud fraction and cloud optical depth are determined. These experiments illustrate the substantial effect that clouds have on the state of the sea ice and on the surface radiative fluxes. The effect of clouds on the net flux of radiation at the surface is very complex over the Arctic Ocean particularly due to the presence of the underlying sea ice. Owing to changes in surface albedo and temperature associated with changing cloud properties, there is a strong non-linearity between cloud properties and surface radiative fluxes. The model results are evaluated in three different contexts: 1) the sensitivity of the arctic surface radiation balance to uncertainties in cloud properties; 2) the impact of interannual variability in cloud characteristics on surface radiation fluxes and sea ice surface characteristics; and 3) the impact of climate change and the resulting changes in cloud properties on the surface radiation fluxes and sea ice characteristics.With 11 Figures     

Change in climate variability in the 21st century

As climate changes due to the increase of greenhouse gases, there is the potential for climate variability to change as well. The change in variability of temperature and precipitation in a transient climate simulation, where trace gases are allowed to increase gradually, and in the doubled CO₂ climate is investigated using the GISS general circulation model. The current climate control run is compared with observations and with the climate change simulations for variability on three time-scales: interannual variability, daily variability, and the amplitude of the diurnal cycle. The results show that the modeled variability is often larger than observed, especially in late summer, possibly due to the crude ground hydrology. In the warmer climates, temperature variability and the diurnal cycle amplitude usually decrease, in conjunction with a decrease in the latitudinal temperature gradient and the increased greenhouse inhibition of radiative cooling. Precipitation variability generally changes with the same sign as the mean precipitation itself, usually increasing in the warmer climate. Changes at a particular grid box are often not significant, with the prevailing tendency determined from a broader sampling. Little change is seen in daily persistence. The results are relevant to the continuing assessments of climate change impacts on society, though their use should be tempered by appreciation of the model deficiencies for the current climate.     

Seasonal climate and variability of the ECMWF ERA-40 model

The climate and variability of seasonal ensemble integrations, made with a recent version of ECMWF model (used for ERA-40 production) at relatively high horizontal resolution (T_L159), have been studied for the 10-year period, 1980–1989. The model systematic error over the Atlantic-European region has been substantially improved when compared with the earlier model versions (e.g. from the PROVOST and AMIP-2 projects). However, it has worsened over the Pacific-North American region. This systematic error reduces the amplitude of planetary waves and has a negative impact on intraseasonal variability and predictability of the PNA mode. The signal-to-noise analysis yields results similar to earlier model versions: only during relatively strong ENSO events do some parts of the extratropics exhibit potential predictability. For precipitation, there is more disagreement between observed and model climatologies over sea than over land, but interannual variations over many parts of the tropical ocean are reasonably well represented. The south Asian monsoon in the model is severely weakened when compared to observations; this is seen in both poor climatology and interannual variability. Overall, comparing the ERA-40 model with earlier versions, there seems to be a balance between model improvements and deteriorations due to systematic errors. For the seasonal time-scale predictability, it is not clear that this model cycle constitutes an advantage over the earlier versions. Therefore, since it is not always possible to achieve distinct improvements in model climate and variability, a careful and detailed strategy ought to be considered when introducing a new model version for operational seasonal forecasting.     

Effects of increased CO2 on land water balance from 1850 to 1989

Numerous studies have shown that increased atmospheric CO₂ concentration is one of the most important factors altering land water balance. In this study, we investigated the effects of increased CO₂ on global land water balance using the dataset released by the Coupled Model Intercomparison Project Phase 5 derived from the Canadian Centre for Climate Modelling and Analysis second-generation Earth System Model. The results suggested that the radiative effect of CO₂ was much greater than the physiological effect on the water balance. At the model experiment only integrating CO₂ radiative effect, the precipitation, evapotranspiration (ET) and runoff had significantly increased by 0.37, 0.12 and 0.31 mm year^?2, respectively. Increases of ET and runoff caused a significant decrease of soil water storage by 0.05 mm year^?2. However, the results showed increases of runoff and decreases of precipitation and ET in response to the CO₂ fertilisation effect, which resulted into a small, non-significant decrease in the land water budget. In the Northern Hemisphere, especially on the coasts of Greenland, Northern Asia and Alaska, there were obvious decreases of soil water responding to the CO₂ radiative effect. This trend could result from increased ice–snow melting as a consequence of warmer surface temperature. Although the evidence suggested that variations in soil moisture and snow cover and vegetation feedback made an important contribution to the variations in the land water budget, the effect of other factors, such as aerosols, should not be ignored, implying that more efforts are needed to investigate the effects of these factors on the hydrological cycle and land water balance.     

夏季我国干旱、半干旱区陆面过程能量平衡及其局地大气环流

利用NCEP资料分析得出,夏季我国干旱,半干旱区在整个欧亚大陆上是陆面感热通量最强的地方,与此对应的陆面潜热通量则最弱.陆面所接收的太阳短波辐射主要以感热和长波辐射的能量形式释放.该区降水量很少,降水量的年际变率也很弱;因此,该区的陆面热量通量都显出很弱的年际变率;然而,这些通量的年代际变率信号则比较显著.我国干旱、半干旱区大气环流的热力过程与其陆面过程特征密切相关.该区对流层大气的辐射冷却很强,达-3 K d-1.由于缺乏水汽和上升运动,大尺度凝结加热率、深对流加热率、浅对流加热率都非常弱.因此,600hPa以上的大气以绝热下沉加热来平衡辐射冷却;600hPa以下,陆面感热引起的垂直扩散加热率非常强,多达8 K d-1,它除了平衡辐射冷却以外还制造对流层低层的对流运动,以绝热上升冷却来平衡多余的垂直扩散加热.总之,我国干旱、半干旱区的陆面过程特征决定了该区大气运动的特殊垂直结构,即对流层低层对流上升运动及其上层的下沉运动.我国干旱、半干旱区陆面能量平衡及其局地大气环流的年代际变率,是全球气候系统年代际变率的必然结果.     

A transient climate change simulation with greenhouse gas and aerosol forcing: projected climate to the twenty-first century

 The potential climatic consequences of increasing atmospheric greenhouse gas (GHG) concentration and sulfate aerosol loading are investigated for the years 1900 to 2100 based on five simulations with the CCCma coupled climate model. The five simulations comprise a control experiment without change in GHG or aerosol amount, three independent simulations with increasing GHG and aerosol forcing, and a simulation with increasing GHG forcing only. Climate warming accelerates from the present with global mean temperatures simulated to increase by 1.7 °C to the year 2050 and by a further 2.7 °C by the year 2100. The warming is non-uniform as to hemisphere, season, and underlying surface. Changes in interannual variability of temperature show considerable structure and seasonal dependence. The effect of the comparatively localized negative radiative forcing associated with the aerosol is to retard and reduce the warming by about 0.9 °C at 2050 and 1.2 °C at 2100. Its primary effect on temperature is to counteract the global pattern of GHG-induced warming and only secondarily to affect local temperatures suggesting that the first order transient climate response of the system is determined by feedback processes and only secondarily by the local pattern of radiative forcing. The warming is accompanied by a more active hydrological cycle with increases in precipitation and evaporation rates that are delayed by comparison with temperature increases. There is an “El Nino-like” shift in precipitation and an overall increase in the interannual variability of precipitation. The effect of the aerosol forcing is again primarily to delay and counteract the GHG-induced increase. Decreases in soil moisture are common but regionally dependent and interannual variability changes show considerable structure. Snow cover and sea-ice retreat. A PNA-like anomaly in mean sea-level pressure with an enhanced Aleutian low in northern winter is associated with the tropical shift in precipitation regime. The interannual variability of mean sea-level pressure generally decreases with largest decreases in the tropical Indian ocean region. Changes to the ocean thermal structure are associated with a spin-down of the Atlantic thermohaline circulation together with a decrease in its variability. The effect of aerosol forcing, although modest, differs from that for most other quantities in that it does not act primarily to counteract the GHG forcing effect. The barotropic stream function in the ocean exhibits modest change in the north Pacific but accelerating changes in much of the Southern Ocean and particularly in the north Atlantic where the gyre spins down in conjunction with the decrease in the thermohaline circulation. The results differ in non-trivial ways from earlier equilibrium 2 × CO₂ results with the CCCma model as a consequence of the coupling to a fully three-dimensional ocean model and the evolving nature of the forcing. Received: 24 September 1998 / Accepted: 8 October 1999     

A local climatic model of the temperature-cloudiness relationship

Summary Filtered secular series of temperature at Zagreb (Croatia) are negatively correlated with simultaneous series of solar radiation and positively correlated with cloudiness.The local balance of radiative fluxes, with cloudiness interference included, provides a basis for the evaluation of a local deterministic model, which can describe both the positive and negative impacts of cloud radiative forcing upon surface temperature. A theoretical temperature series, generated by the model simulation, proves a prevailing greenhouse effect of clouds upon recent climatic variations of temperature at Zagreb.With 5 Figures     

Climate variability and projected change in the western United States: regional downscaling and drought statistics

Climate change in the twenty-first century, projected by a large ensemble average of global coupled models forced by a mid-range (A1B) radiative forcing scenario, is downscaled to Climate Divisions across the western United States. A simple empirical downscaling technique is employed, involving model-projected linear trends in temperature or precipitation superimposed onto a repetition of observed twentieth century interannual variability. This procedure allows the projected trends to be assessed in terms of historical climate variability. The linear trend assumption provides a very close approximation to the time evolution of the ensemble-average climate change, while the imposition of repeated interannual variability is probably conservative. These assumptions are very transparent, so the scenario is simple to understand and can provide a useful baseline assumption for other scenarios that may incorporate more sophisticated empirical or dynamical downscaling techniques. Projected temperature trends in some areas of the western US extend beyond the twentieth century historical range of variability (HRV) of seasonal averages, especially in summer, whereas precipitation trends are relatively much smaller, remaining within the HRV. Temperature and precipitation scenarios are used to generate Division-scale projections of the monthly palmer drought severity index (PDSI) across the western US through the twenty-first century, using the twentieth century as a baseline. The PDSI is a commonly used metric designed to describe drought in terms of the local surface water balance. Consistent with previous studies, the PDSI trends imply that the higher evaporation rates associated with positive temperature trends exacerbate the severity and extent of drought in the semi-arid West. Comparison of twentieth century historical droughts with projected twenty-first century droughts (based on the prescribed repetition of twentieth century interannual variability) shows that the projected trend toward warmer temperatures inhibits recovery from droughts caused by decade-scale precipitation deficits.     

Changes of interannual NAO variability in response to greenhouse gases forcing

Observations show that there was change in interannual North Atlantic Oscillation (NAO) variability in the mid-1970s. This change was characterized by an eastward shift of the NAO action centres, a poleward shift of zonal wind anomalies and a downstream extension of climate anomalies associated with the NAO. The NAO interannual variability for the period after the mid-1970s has an annular mode structure that penetrates deeply into the stratosphere, indicating a strengthened relationship between the NAO and the Arctic Oscillation (AO) and strengthened stratosphere-troposphere coupling. In this study we have investigated possible causes of these changes in the NAO by carrying out experiments with an atmospheric GCM. The model is forced either by doubling CO₂, or increasing sea surface temperatures (SST), or both. In the case of SST forcing the SST anomaly is derived from a coupled model simulation forced by increasing CO₂. Results indicate that SST and CO₂ change both force a poleward and eastward shift in the pattern of interannual NAO variability and the associated poleward shift of zonal wind anomalies, similar to the observations. The effect of SST change can be understood in terms of mean changes in the troposphere. The direct effect of CO₂ change, in contrast, can not be understood in terms of mean changes in the troposphere. However, there is a significant response in the stratosphere, characterized by a strengthened climatological polar vortex with strongly enhanced interannual variability. In this case, the NAO interannual variability has a strong link with the variability over the North Pacific, as in the annular AO pattern, and is also strongly related to the stratospheric vortex, indicating strengthened stratosphere-troposphere coupling. The similarity of changes in many characteristics of NAO interannual variability between the model response to doubling CO₂ and those in observations in the mid-1970s implies that the increase of greenhouse gas concentration in the atmosphere, and the resulting changes in the stratosphere, might have played an important role in the multidecadal change of interannual NAO variability and its associated climate anomalies during the late twentieth century. The weak change in mean westerlies in the troposphere in response to CO₂ change implies that enhanced and eastward extended mid-latitude westerlies in the troposphere might not be a necessary condition for the poleward and eastward shift of the NAO action centres in the mid-1970s.     

A Coupling Experiment of an Atmosphere and an Ocean Model with a Monthly Anomaly Exchange Scheme

A nine-layer spectral atmospheric general circulation model is coupled to a twenty-layer global oceanic general circulation model with the “prediction-correction” monthly anomaly exchange scheme which has been proposed at the Institute of Atmospheric Physics (IAP). A forty-year integration of the coupled model shows that the CGCM is fairly successful in keeping a reasonable pattern of the modelled SST although most of the Pacific become warmer than those given by the uncoupled ocean model. The model tends to reach a more realistic state than the uncoupled one in terms of downward surface heat flux into ocean particularly in the equatorial Pacific region. Also, the model is capable to simulate interannual variability of sea surface temperature in tropical region.