The seasonal cycle in coupled ocean-atmosphere general circulation models

We examine the seasonal cycle of near-surface air temperature simulated by 17 coupled ocean-atmosphere general circulation models participating in the Coupled Model Intercomparison Project (CMIP). Nine of the models use ad hoc “flux adjustment” at the ocean surface to bring model simulations close to observations of the present-day climate. We group flux-adjusted and non-flux-adjusted models separately and examine the behavior of each class. When averaged over all of the flux-adjusted model simulations, near-surface air temperature falls within 2?K of observed values over the oceans. The corresponding average over non-flux-adjusted models shows errors up to ～6?K in extensive ocean areas. Flux adjustments are not directly applied over land, and near-surface land temperature errors are substantial in the average over flux-adjusted models, which systematically underestimates (by ～5?K) temperature in areas of elevated terrain. The corresponding average over non-flux-adjusted models forms a similar error pattern (with somewhat increased amplitude) over land. We use the temperature difference between July and January to measure seasonal cycle amplitude. Zonal means of this quantity from the individual flux-adjusted models form a fairly tight cluster (all within ～30% of the mean) centered on the observed values. The non-flux-adjusted models perform nearly as well at most latitudes. In Southern Ocean mid-latitudes, however, the non-flux-adjusted models overestimate the magnitude of January-minus-July temperature differences by ～5?K due to an overestimate of summer (January) near-surface temperature. This error is common to five of the eight non-flux-adjusted models. Also, over Northern Hemisphere mid-latitude land areas, zonal mean differences between July and January temperatures simulated by the non-flux-adjusted models show a greater spread (positive and negative) about observed values than results from the flux-adjusted models. Elsewhere, differences between the two classes of models are less obvious. At no latitude is the zonal mean difference between averages over the two classes of models greater than the standard deviation over models. The ability of coupled GCMs to simulate a reasonable seasonal cycle is a necessary condition for confidence in their prediction of long-term climatic changes (such as global warming), but it is not a sufficient condition unless the seasonal cycle and long-term changes involve similar climatic processes. To test this possible connection, we compare seasonal cycle amplitude with equilibrium warming under doubled atmospheric carbon dioxide for the models in our data base. A small but positive correlation exists between these two quantities. This result is predicted by a simple conceptual model of the climate system, and it is consistent with other modeling experience, which indicates that the seasonal cycle depends only weakly on climate sensitivity.     

Surface flux response to interannual tropical Pacific sea surface temperature variability in AMIP models

 A systematic comparison of observed and modeled atmospheric surface heat and momentum fluxes related to sea surface temperature (SST) variability on interannual time scales in the tropical Pacific is conducted. This is done to examine the ability of atmospheric general circulation models (AGCMs) in the Atmospheric Model Intercomparison Project (AMIP) to simulate the surface fluxes important for driving the ocean on interannual time scales. In order to estimate the model and observed response to such SST variability, various regression calculations are made between a time series representing observed ENSO SST variability in the tropical Pacific and the resulting surface flux anomalies. The models exhibit a range of differences from the observations. Overall the zonal wind stress anomalies are most accurately simulated while the solar radiation anomalies are the least accurately depicted. The deficiencies in the solar radiation are closely related to errors in cloudiness. The total heat flux shows some cancellation of the errors in its components particularly in the central Pacific. The performance of the GCMs in simulating the surface flux anomalies seems to be resolution dependent and low-resolution models tend to exhibit weaker flux responses. The simulated responses in the western Pacific are more variable than those of the central and eastern Pacific but in the west the observed estimates are less robust as well. Further improvements in atmospheric GCM flux simulation through better physical parametrization is clearly required if such models are to be used to their full potential in coupled modeling and climate forecasting. Received: 24 August 1999 / Accepted: 11 September 2000     

Simple general atmospheric circulation and climate models with memory

This article examines some general atmospheric circulation and climate models in the context of the notion of “memory”. Two kinds of memories are defined： statistical memory and deterministic memory. The former is defined through the autocorrelation characteristic of the process if it is random （chaotic）, while for the latter, a special memory function is introduced. Three of the numerous existing models are selected as examples. For each of the models, asymptotic （at t →∞） expressions are derived. In this way, the transients are filtered out and that which remains concerns the final behaviour of the models.     

ENSO dynamics and seasonal cycle in the tropical Pacific as simulated by the ECHAM4/OPYC3 coupled general circulation model

 The new version of the atmospheric general circulation model (AGCM), ECHAM4, at the Max Planck Institute for Meteorology, Hamburg, has been coupled to the OPYC3 isopycnic global ocean general circulation and sea ice model in a multi-century present-day climate simulation. Non-seasonal constant flux adjustment for heat and freshwater was employed to ensure a long-term annual mean state close to present-day climatology. This study examines the simulated upper ocean seasonal cycle and interannual variability in the tropical Pacific for the first 100 years. The coupled model’s seasonal cycle of tropical Pacific SSTs is satisfactory with respect to both the warm pool variation and the Central and Eastern Pacific, with significant errors only in the cold tongue around April. The cold phase cold tongue extent and strength is as observed, and for this the heat flux adjustment does not play a decisive role. A well-established South Pacific convergence zone is characteristic for the new AGCM version. Apart from extending the southeast trades seasonal maximum to midbasin, wind stress pattern and strength are captured. Overall the subsurface structure is consistent with the observed, with a pronounced thermocline at about 150 m depth in the west and rising to the surface from 160 °W to 100 °W. The current system is better resolved than in some previous global models and, on the whole, has the expected shape. The equatorial undercurrent is correctly positioned but the core is only half as strong as observed. The north equatorial current and counter-current also have reduced maximum speeds but the April minimum is captured. As with the companion publication from Roeckner et al. this study finds pronounced tropical Eastern and Central Pacific interannual variability. Simulated and observed NINO3 sea surface temperature (SST) variability is represented by a single, rather broadband, maximum of power spectral density, centered on about 28 months for the simulation and four years for the observations. For simulation and observations, SST, windstress, and upper ocean heat content each exhibit a single dominant large-scale amplitude and phase pattern, suggesting that the model captures the essential dynamics. The amplitude of the essentially standing oscillation in SST in the NINO3 region attains the observed strength, but is weaker at the eastern boundary. Anomalies of upper ocean heat content show off-equatorial westward and equatorial eastward propagation, the latter’s arrival in the east of the basin coinciding with the SST anomalies. Equatorial wind stress anomalies near the date line provide the appropriate forcing and clearly form a response to the anomalous SST. Received: 14 June 1996 / Accepted: 11 November 1997     

An event-by-event assessment of tropical intraseasonal perturbations for general circulation models

We detect and characterize each large-scale intraseasonal perturbation in observations (1979–2009) and in coupled general circulation models of Institut Pierre Simon Laplace (IPSL) and of Centre National de Recherches Météorologiques (CNRM). These ensembles of intraseasonal perturbations are used to assess the skill of the two models in an event-by-event approach. This assessment addresses: (1) the planetary-scale (i.e. the whole Indo-Pacific area) extent of wind and rainfall perturbations and the reproducibility of the perturbation patterns for a given season; (2) the size and amplitude of rainfall and wind anomalies at basin-scale (i.e. for a particular phase of the perturbation) and; (3) the evolution of the vertical structure of the perturbations (U, T and RH) for selected events. The planetary-scale extent of rainfall perturbations is generally too small for both models. This extent is also small for the wind perturbation in the IPSL model, but is correct, or even too large in boreal winter, for the CNRM model. The reproducibility of the planetary-scale patterns is exaggerated for wind perturbations in the CNRM model and is very poor for all parameters in the IPSL model. Over the Indian Ocean during boreal winter, rainfall and wind anomalies at basin-scale are too large for the CNRM model and too small for the IPSL model. The CNRM model gives a realistic baroclinic perturbations structure for wind, moisture and temperature, but with too large amplitude due in part to a zonally extended rainfall anomaly over the eastern Indian Ocean and the Maritime Continent. The IPSL model gives a realistic response for low-level wind only. Temperature and moisture perturbations are barotropic with a wrong warm anomaly at rainfall maximum and there is no gradual increase in low-level moisture prior to this rainfall maximum. These results suggest that this version of the IPSL model is unable to initiate the coupling between the convection and the dynamic necessary to develop the perturbation. It is difficult to say if this is due to, or is at the origin of the lack of basin-scale organization of the convection. We discuss the likely role of the convective schemes in the differences found between these two versions of the CNRM and IPSL models.     

Predictability of SST forced climate signals in two atmospheric general circulation models

 The predictability of atmospheric responses to global sea surface temperature (SST) anomalies is evaluated using ensemble simulations of two general circulation models (GCMs): the GENESIS version 1.5 (GEN) and the ECMWF cycle 36 (ECM). The integrations incorporate observed SST variations but start from different initial land and atmospheric states. Five GEN 1980–1992 and six ECM 1980–1988 realizations are compared with observations to distinguish predictable SST forced climate signals from internal variability. To facilitate the study, correlation analysis and significance evaluation techniques are developed on the basis of time series permutations. It is found that the annual mean global area with realistic signals is variable dependent and ranges from 3 to 20% in GEN and 6 to 28% in ECM. More than 95% of these signal areas occur between 35 °S–35 °N. Due to the existence of model biases, robust responses, which are independent of initial condition, are identified over broader areas. Both GCMs demonstrate that the sensitivity to initial conditions decreases and the predictability of SST forced responses increases, in order, from 850 hPa zonal wind, outgoing longwave radiation, 200 hPa zonal wind, sea-level pressure to 500 hPa height. The predictable signals are concentrated in the tropical and subtropical Pacific Ocean and are identified with typical El Ni?o/ Southern Oscillation phenomena that occur in response to SST and diabatic heating anomalies over the equatorial central Pacific. ECM is less sensitive to initial conditions and better predicts SST forced climate changes. This results from (1) a more realistic basic climatology, especially of the upper-level wind circulation, that produces more realistic interactions between the mean flow, stationary waves and tropical forcing; (2) a more vigorous hydrologic cycle that amplifies the tropical forcing signals, which can exceed internal variability and be more efficiently transported from the forcing region. Differences between the models and observations are identified. For GEN during El Ni?o, the convection does not carry energy to a sufficiently high altitude, while the spread of the tropospheric warming along the equator is slower and the anomaly magnitude smaller than observed. This impacts model ability to simulate realistic responses over Eurasia and the Indian Ocean. Similar biases exist in the ECM responses. In addition, the relationships between upper and lower tropospheric wind responses to SST forcing are not well reproduced by either model. The identification of these model biases leads to the conclusion that improvements in convective heat and momentum transport parametrizations and basic climate simulations could substantially increase predictive skill. Received: 25 April 1996 / Accepted: 9 December 1996     

Winter cyclone frequencies in thirteen models participating in the Atmospheric Model Intercomparison Project (AMIP1)

Various aspects of the simulated behaviour of cyclones in thirteen models participating in the AMIP1 exercise are presented. In the simulation of the winter climatological mean sea level pressure field for the Northern Hemisphere, the models produce reasonable simulations of the "semi-permanent" features of the climatology. The greatest departures from the observed climatology occur near the exit regions of the oceanic storm tracks; i.e., over northwestern North America, over and to the west of the British Isles and in the Mediterranean. The departures in the three geographical areas are very systematic in that at least eleven of the models exhibit similar departures from observations. In the Southern Hemisphere the intensity of the circumpolar trough is generally well simulated but positioned slightly too far north. Most models exhibit errors south of Africa, New Zealand, and South America. The simulations of the cyclone events show that the models are reasonably successful in reproducing the large-scale aspects of observed cyclone events but deficiencies in the details of the simulations are apparent. The paucity of simulated events to the south of the Alps and to the east of the Rockies suggests that the models have difficulty simulating lee cyclogenesis. Over much of North America, the models have difficulty simulating the correct level of synoptic activity as demonstrated by the low numbers of both cyclone events and anticyclone events. The models have difficulty simulating the distribution of cyclone events as a function of central pressure. The most common problem is that the models exhibit an ever increasing deficit of events with decreasing central pressure. This problem is more apparent in the Southern Hemisphere than in the Northern Hemisphere and does not appear to be resolution dependent. There is an apparent ENSO signal in the observed Northern Hemisphere interannual variability of intense winter cyclone events. With the exception of ECMWF, the models fail to reproduce this phenomenon. There is some evidence that the models do indeed respond to the interannual variability in the SSTs, but the response tends to be negatively correlated with that of the real atmosphere. In the Southern Hemisphere, there does not appear to be ENSO-induced interannual variability in the observed numbers of cyclone events. Consequently, it could be argued that the models have been reasonably successful in the Southern Hemisphere since they, like the observations, do not exhibit any ENSO-induced interannual variability.     

The tropical intraseasonal oscillation in SAMIL coupled and uncoupled general circulation models

Simulations of tropical intraseasonal oscillation(TISO) in SAMIL,the Spectral Atmospheric Model from the Institute of Atmospheric Physics(IAP) State Key Laboratory of Numerical Modeling for Atmospheric Sciences and Geophysical Fluid Dynamics(LASG) coupled and uncoupled general circulation models were comprehensively evaluated in this study.Compared to the uncoupled model,the atmosphere-ocean coupled model improved the TISO simulation in the following aspects:(1) the spectral intensity for the 30-80-day peak eastward periods was more realistic;(2) the eastward propagation signals over western Pacific were stronger;and(3) the variance distribution and stronger signals of Kelvin waves and mixed Rossby gravity waves were more realistic.Better performance in the coupled run was assumed to be associated with a better mean state and a more realistic relationship between precipitation and SST.In both the coupled and uncoupled runs,the unrealistic simulation of the eastward propagation over the equatorial Indian Ocean might have been associated with the biases of the precipitation mean state over the Indian Ocean,and the unrealistic split of maximum TISO precipitation variance over the Pacific might have corresponded to the exaggeration of the double Intertropical Convergence Zone(ITCZ) structure in precipitation mean state.However,whether a better mean state leads to better TISO activity remains questionable.Notably,the northward propagation over the Indian Ocean during summer was not improved in the mean lead-lag correlation analysis,but case studies have shown some strong cases to yield remarkably realistic northward propagation in coupled runs.     

An improved scheme for time-dependent boundary conditions in atmospheric general circulation models

 Atmospheric general circulation models (AGCMs) are often “coupled” with time varying observations of boundary conditions or some other aspect of the climate system. A typical example is the Atmospheric Model Intercomparison Project (AMIP) experimental protocol, which required the specification of sea surface temperature and sea-ice extent from observed monthly means. AGCMs ordinarily incorporate the prescribed conditions by evaluating an interpolating function at each time step. Typical schemes, such as that used in the second generation GCM (GCM2) of the Canadian Centre for Climate Modelling and Analysis (CCC), do not preserve monthly means and have a smoothing effect on the interpolated time series which tends to reduce the amplitude of annual cycle and interannual variability of sea surface temperature (SST). By solving a large set of linear equations, a simple linear time-interpolation scheme that preserves the observed monthly mean SST and hence its variability can be obtained. The new scheme improves upon that used previously in CCC GCM2 by eliminating the substantial loss of interannual variability (up to 20%) and the small attenuation of the annual cycle (less than 4% on average) incurred with the old scheme. The improved linear interpolation scheme is easily adapted to other quantities. Received: 4 August 1997 / Accepted: 26 November 1997     

The seasonal cycle of tropical Pacific sea surface temperature in a coupled GCM

 The mechanisms responsible for the seasonal cycle in the tropical central and eastern Pacific sea surface temperature (SST) are investigated using a coupled general circulation model. We find that the annual westward propagation of SST anomalies along the equator is explained by a two-stage process. The first stage sets the phase of the variation at the eastern boundary. The strengthening of the local Hadley Circulation in boreal summer leads to a strengthening of the northward winds that blow across the equator. These stronger winds drive enhanced evaporation and entrainment cooling of the oceanic mixed layer. The resulting change in SST is greatest in the east because the mixed layer is at its shallowest there. As the east Pacific SST cools the zonal SST gradient in the central Pacific becomes more negative. This development signals the onset of the second stage in the seasonal variation of equatorial SST. In response to the anomalous SST gradient the local westward wind stress increases. This increase drives cooling of the oceanic mixed layer in which no single mechanism dominates: enhanced evaporation, wind-driven entrainment, and westward advection all contribute. We discuss the role that equatorial upwelling plays in modulating mixed layer depth and hence the entrainment cooling, and we highlight the importance of seasonal variations in mixed layer depth. In sum these processes act to propagate the SST anomaly westward. Received: 22 February 1999 / Accepted: 20 March 2000     

The seasonal cycle of mixed layer temperatures in a global ocean general circulation model

We examine mixed layer temperatures in a global ocean general circulation model subjected to seasonally varying climatological forcing. Harmonic analysis of monthly mixed layer temperatures and climatological sea surface temperatures (SSTs) shows that, on the average, the annual harmonic accounts for 90% of the total seasonal variance in both fields, while the semiannual harmonic accounts for about 8%. The semiannual signal is mostly confined to equatorial and high-latitude regions. The model mixed layer temperatures underestimate the mean amplitude of the annual harmonic in middle latitudes (65°||10°) by about 26%, while lagging climatological SSTs by 22 days, on average. In several parameter sensitivity experiments, these differences could be reduced to as little as 12% and 12.5 days, respectively, though most of this gain occurred when the mixed layer was unrealistically shallow (mean depth less than 65 m). At least part of the differences in amplitudes and phases of the annual harmonic is linked to the uncoupled formulation of the surface heat flux, which is computed using specified and seasonally varying climatological air temperatures. In ice-free areas, seasonal amplitudes and phases of air temperatures are almost identical to those of climatological SSTs. Thus, differences between model mixed layer temperatures and climatological SSTs give rise to Newtonian relaxation to SSTs, which then leads to amplitude damping and time lags in mixed layer temperatures relative to the SSTs.     

Connection of stratospheric QBO with global atmospheric general circulation and tropical SST. Part I: methodology and composite life cycle

The stratospheric quasi-biennial oscillation (QBO) and its association with the interannual variability in the stratosphere and troposphere, as well as in tropical sea surface temperature anomalies (SSTA), are examined in the context of a QBO life cycle. The analysis is based on the ERA40 and NCEP/NCAR reanalyses, radiosonde observations at Singapore, and other observation-based datasets. Both reanalyses reproduce the QBO life cycle and its associated variability in the stratosphere reasonably well, except that some long-term changes are detected only in the NCEP/NCAR reanalysis. In order to separate QBO from variability on other time scales and to eliminate the long-term changes, a scale separation technique [Ensemble Empirical Mode Decomposition (EEMD)] is applied to the raw data. The QBO component of zonal wind anomalies at 30?hPa, extracted using the EEMD method, is defined as a QBO index. Using this index, the QBO life cycle composites of stratosphere and troposphere variables, as well as SSTA, are constructed and examined. The composite features in the stratosphere are generally consistent with previous investigations. The correlations between the QBO and tropical Pacific SSTA depend on the phase in a QBO life cycle. On average, cold (warm) SSTA peaks about half a year after the maximum westerlies (easterlies) at 30?hPa. The connection of the QBO with the troposphere seems to be associated with the differences of temperature anomalies between the stratosphere and troposphere. While the anomalies in the stratosphere propagate downward systematically, some anomalies in the troposphere develop and expand vertically. Therefore, it is possible that the temperature difference between the troposphere and stratosphere may alter the atmospheric stability and tropical deep convection, which modulates the Walker circulation and SSTA in the equatorial Pacific Ocean.     

Impact of the convection-wind-evaporation feedback on surface climate simulation in general circulation models

The wind-evaporation-convection feedback in the tropics is demonstrated to strongly affect the mean state of surface climate in atmospheric general circulation models. The feedback is shown to be very effective in channeling perturbations from one component of the climate system to other components, e.g., from evaporation to surface wind and from atmospheric convective activity to evaporation. It also provides an effective channel to pass on atmospheric perturbations in the middle and upper troposphere to the surface. As an illustration, it is shown that surface evaporation over tropical oceans is connected with cloud absorption of shortwave radiation through this feedback. Insufficient shortwave cloud absorption, causing excessive shortwave radiation at the surface as is common in most of the climate models, leads to excessive evaporation. Quantitatively, sensitivity of evaporation to short-wave cloud absorption, when averaged over the whole tropics, can be described by an approximate balance of variations in atmospheric radiative cooling and latent heating. This balance is achieved by the impact of radiation on convection, and then on the surface wind and evaporation. This mechanism calls for the need to include atmospheric processes far beyond the surface for improvements of the quality of surface climate simulation.     

国际大气环流模式比较计划（AMIP）进展

介绍了国际大气环流模式比较计划（AMIP）的概况，它的分析子计划及执行情况、科学进展、相关计划及未来发展等情况，以便国内学者了解并积极参与该研究计划。     

The seasonal climatic simulation of 9000 years before present by using the IAP atmospheric general circulation model

The seasonal cycle of the climate of 9000 years before present was simulated with the IAP two-level atmospheric general circulation model. The incoming solar radiation was specified from the orbital parameters for 9000 years Ago. The boundary conditions of that time were prescribed to the present value because of the small differences between the two. The change in radiation makes temperature to be higher in summer and lower in winter over large areas of the land; and the increased temperature contrast between the land and the ocean strengthens the summer monsoon circulation and increases the precipitation over there. The asymmetry of temperature change between the Northern Hemisphere and the Southern Hemisphere and between summer and winter still exists, which agrees with that get from the previous perpetual experiments.     

Influence of tropical cyclones on sea surface temperature seasonal cycle and ocean heat transport

Recent studies suggested that tropical cyclones (TCs) contribute significantly to the meridional oceanic heat transport by injecting heat into the subsurface through mixing. Here, we estimate the long-term oceanic impact of TCs by inserting realistic wind vortices along observed TCs tracks in a 1/2° resolution ocean general circulation model over the 1978–2007 period. Warming of TCs’ cold wakes results in a positive heat flux into the ocean (oceanic heat uptake; OHU) of ~480 TW, consistent with most recent estimates. However, ~2/5 of this OHU only compensates the heat extraction by the TCs winds during their passage. Another ~2/5 of this OHU is injected in the seasonal thermocline and hence released back to the atmosphere during the following winter. Because of zonal compensations and equatorward transport, only one-tenth of the OHU is actually exported poleward (46 TW), resulting in a marginal maximum contribution of TCs to the poleward ocean heat transport. Other usually neglected TC-related processes however impact the ocean mean state. The residual Ekman pumping associated with TCs results in a sea-level drop (rise) in the core (northern and southern flanks) of TC-basins that expand westward into the whole basin as a result of planetary wave propagation. More importantly, TC-induced mixing and air-sea fluxes cool the surface in TC-basins during summer, while the re-emergence of subsurface warm anomalies warms it during winter. This leads to a ~10 % reduction of the sea surface temperature seasonal cycle within TCs basins, which may impact the climate system.     

The energy cycle in atmospheric models

The energy cycle characterizes basic aspects of the physical behaviour of the climate system. Terms in the energy cycle involve first and second order climate statistics (means, variances, covariances) and the intercomparison of energetic quantities offers physically motivated “second order” insight into model and system behaviour. The energy cycle components of 12 models participating in AMIP2 are calculated, intercompared and assessed against results based on NCEP and ERA reanalyses. In general, models simulate a modestly too vigorous energy cycle and the contributions to and reasons for this are investigated. The results suggest that excessive generation of zonal available potential energy is an important driver of the overactive energy cycle through “generation push” while excessive dissipation of eddy kinetic energy in models is implicated through “dissipation pull‘’. The study shows that “ensemble model” results are best or among the best in the comparison of energy cycle quantities with reanalysis-based values. Thus ensemble approaches are apparently “best” not only for the simulation of 1st order climate statistics as in Lambert and Boer (Clim Dyn 17:83–106, 2001) but also for the higher order climate quantities entering the energy cycle.     

Stratospheric circulation in seasonal forecasting models: implications for seasonal prediction

Accurate seasonal forecasts rely on the presence of low frequency, predictable signals in the climate system which have a sufficiently well understood and significant impact on the atmospheric circulation. In the Northern European region, signals associated with seasonal scale variability such as ENSO, North Atlantic SST anomalies and the North Atlantic Oscillation have not yet proven sufficient to enable satisfactorily skilful dynamical seasonal forecasts. The winter-time circulations of the stratosphere and troposphere are highly coupled. It is therefore possible that additional seasonal forecasting skill may be gained by including a realistic stratosphere in models. In this study we assess the ability of five seasonal forecasting models to simulate the Northern Hemisphere extra-tropical winter-time stratospheric circulation. Our results show that all of the models have a polar night jet which is too weak and displaced southward compared to re-analysis data. It is shown that the models underestimate the number, magnitude and duration of periods of anomalous stratospheric circulation. Despite the poor representation of the general circulation of the stratosphere, the results indicate that there may be a detectable tropospheric response following anomalous circulation events in the stratosphere. However, the models fail to exhibit any predictability in their forecasts. These results highlight some of the deficiencies of current seasonal forecasting models with a poorly resolved stratosphere. The combination of these results with other recent studies which show a tropospheric response to stratospheric variability, demonstrates a real prospect for improving the skill of seasonal forecasts.