Abstract

This study is concerned with assessing the extent to which extratropical low-frequency variability may be viewed as a response to geographically coherent stochastic forcing. This issue is examined with a barotropic model linearized about the long-term mean wintertime 300-mb flow with zonal and meridional structure. The perturbation eigenfunctions of the model are stable (i.e., decaying) for a realistic 5-day drag, so transient eddy activity can be maintained against the drag only with forcing. In a statistical steady state, a fluctuation–dissipation relation (FDR) links the covariance structure of the eddy vorticity to the covariance structure of the forcing. This relation is used in a forward sense to determine the covariance of eddy vorticity for a specified covariance of forcing. It is also used in a backward sense to infer the covariance of forcing required to maintain the observed covariance of eddy vorticity. The focus is on explaining the observed variability of 10-day running mean anomalies of the 300-mb flow in the northern winters of 1985–93.

When used in the backward sense described above, the FDR yields a forcing covariance matrix that is not quite positive definite. This immediately implies that the low-frequency variability cannot be rigorously viewed as a linear barotropic response to white noise forcing. Nonetheless, retaining only the positive definite part of the forcing matrix and using the forward FDR gives a reasonable approximation to the observed vorticity covariance. The approximation can be improved by specifying a stronger drag in the barotropic model. However, the simulation of the 5-day lag covariance of vorticity, which is poor using the 5-day drag, is made worse with the stronger drag. In other words this model cannot correctly simulate the time development of low-frequency variability. Thus extratropical low-frequency variability cannot be understood as a linear barotropic response to geographically coherent white noise forcing.

A slightly red stochastic forcing, with a decorrelation timescale of 1–2 days, produces only a modest improvement in the 5-day lag results. A very red forcing, with a decorrelation timescale of 20 days, gives better results at 0- and 5-day lags, but not at 10- or 20-day lags. Modeling the forcing separately as a first-order Markov process, with the model parameters estimated from observations, gives almost perfect results at 0- and 5-day lags. However, further analysis shows this to be an artifact of comparing the empirical–dynamical model simulations with dependent data. When the noise model parameters estimated from one-half of the data record are used to explain low-frequency variability in the other, the results are again poor. It is concluded that extratropical low-frequency variability cannot be viewed as randomly forced barotropic Rossby waves evolving on a zonally and meridionally varying climatological 300-mb flow. The spatial and temporal structure of the observed variability cannot be explained without also taking into account the detailed spatial and temporal structure of the forcing, respectively.

Abstract

The dynamical stability of the Northern Hemisphere wintertime mean atmosphere is investigated in a linearized primitive equation model. In the absence of any damping on the perturbation, exponentially growing modes are found for the zonal-mean and zonally varying basic states. Their growth rates are 0.41 and 0.38 days⁻¹, respectively. Both have the form of midlatitude baroclinic wave trains.

Three distinct idealized profiles of linear damping are then imposed on the perturbation vorticity and temperature. The damping is strongest below 800 mb and weak or nonexistent in the rest of the troposphere. It is specified to be proportional at all levels to a single parameter, R _s , the strength of damping at the surface.

For the zonal-mean basic state, as R _s is increased linearly, the growing modes decrease their growth rates almost linearly, and change their structure only slowly. For an average damping timescale in the boundary layer of about one day (R _s = 2 days⁻¹), the growing baroclinic modes are effectively neutralized. The wavy basic state is also rendered neutral when R _s reaches this value. It is argued that this magnitude of damping is within the range of observable parameters in the atmosphere. However, the precise position of the neutral point is sensitive to the relative magnitudes of temperature and vorticity damping. The latter is more efficient in stabilizing the system.

For the wavy basic state, a second mode replaces the undamped mode as the fastest growing just before the neutral point is reached. This mode also resembles a midlatitude baroclinic wave train, but has a longer zonal wavelength. Zonal-mean transient fluxes of eddy temperature and momentum, and eddy kinetic energy calculated for this mode, show an improvement over the undamped and zonal-mean modes when compared with observations. It is argued that this improvement may be meaningful, particularly in an atmosphere that is close to neutral.

Abstract

Global upper-tropospheric kinetic energy (KE) spectra in several global atmospheric circulation datasets are examined. The datasets considered include ERA-Interim, JRA-55, and ERA5 and two versions of NOAA GFS analyses at horizontal resolutions ranging from 0.7° to 0.12°. The mesoscale portions of the spectra are found to be highly inconsistent. This is shown to be mainly due to inconsistencies in the scale-dependent numerical damping and in the large contributions to the global mesoscale KE from the KE in convective regions and near orography. The spectra also generally have a steeper mesoscale slope than the −5/3 slope of the observational Nastrom–Gage spectrum pursued at many modeling centers. The sensitivity of the slope in global models to 1) stochastically perturbing diabatic tendencies and 2) decreasing the horizontal hyperviscosity coefficient is explored in large ensembles of 10-day forecasts made with the NCEP GFS (0.7° grid) model. Both changes lead to larger mesoscale KE and a flatter spectral slope. The effect is stronger in the modified hyperviscosity experiment. These results show that (i) despite assimilating vastly more observations than used in the original Nastrom–Gage studies, current high-resolution global analyses still do not converge to a single “true” global mesoscale KE spectrum, and (ii) model KE spectra can be made flatter not just by increasing model resolution but also by perturbing model physics and decreasing horizontal diffusion. Such sensitivities and lack of consensus on the spectral slope also raise the possibility that the true global mesoscale spectral slope may not be a precisely −5/3 slope.

Abstract

In recent years much attention has been given to Rossby wave propagation and dispersion on representative zonally and meridionally varying background flows in the atmosphere. Particular emphasis has been placed on the 300-mb flow as shaping the structure and evolution of extratropical low-frequency eddies. In this paper an attempt is made to check this hypothesis against the observed evolution of low-pass filtered 300-mb streamfunction anomalies during the eight northern winters of 1985–93. The filter passes periods greater than 10 days. The focus is on explaining the observed evolution of these low-pass anomalies over ∼10 days. This evolution is expected to be nonmodal, that is, a mix of several evolving normal modes rather than a single mode. Two questions are asked: 1) given an initial anomaly, to what extent can one explain the observed subsequent evolution with an unforced barotropic vorticity equation linearized about the climatological 300-mb flow, and 2) in instances of anomaly growth, to what extent is the growth optimally nonmodal, that is, associated with the maximum possible constructive interference of the normal modes.

Concerning question 1, it is found that regardless of the linear drag specified in the model, it cannot reproduce the 10-day lag-covariance structure of the observations. If the model is interpreted as an equivalent barotropic model applied at the 300-mb level, a 5-day drag is appropriate; however, the modeled anomalies lose significant amplitude by day 10 in this case. Question 2 is addressed in two ways. First, a theoretical analysis is performed to determine the optimal as well as expected nonmodal growth of global perturbation kinetic energy in the model. The optimal growth can be as large as a factor of 8 over 3.5 days, even in the presence of the 5-day drag, if certain optimal perturbations (singular vectors) occur as the initial condition. The expected growth, given the statistical structure of the observed “initial” conditions, is however actually a decay. Second, 21 cases of global anomaly growth sustained over at least 7 days are isolated in the data record. The model is run for 7 days with the observed initial condition in each case. In each case, it predicts a decay instead of growth, more consistent with the expected growth (i.e., decay) than optimal growth. The same result is obtained with nonlinear integrations. Somewhat better results are obtained by considering “instantaneous” background flows in the 21 cases; however, the model still predicts a decay. An interesting ambiguity in the interpretation of these latter runs is highlighted.

These results imply that an unforced barotropic vorticity equation linearized about a representative 300-mb flow cannot explain the observed evolution of low-frequency anomalies except possibly in isolated cases. In particular they imply that extratropical low-frequency variability cannot be viewed solely as free Rossby wave propagation and dispersion on a zonally and meridionally varying 300-mb flow, with the forcing acting mainly as a provider of initial perturbations in certain sensitive regions of the atmosphere. Rather the forcing, which in a barotropic model represents the combined effects of diabatic heating, interactions with orography, synoptic-eddy feedbacks, and baroclinic dynamics, is important throughout the development of low-frequency anomalies.

Abstract

Multiple tropical climate regimes are found in an atmospheric general circulation model (AGCM) coupled to a global slab ocean when the model is forced by different values of globally uniform insolation. Even in this simple setting, convection organizes into an intertropical convergence zone (ITCZ) solely due to the effect of planetary rotation, as was found in Kirtman and Schneider, for a single value of insolation. Here the response to a range of insolation values is explored, and surprisingly, multiple climate regimes characterized by radically different ITCZ structures are found. In order from the coldest to warmest climates, these are a symmetric double ITCZ, a near-symmetric equatorial ITCZ, a transient asymmetric ITCZ, and a stable, strongly asymmetric ITCZ.

The model exhibits hysteresis in the transition from the near-symmetric to the strongly asymmetric ITCZ regimes when insolation is increased and then decreased. The initial transition away from symmetry can occur in the absence of air–sea coupling; however, the coupling is essential for the establishment and maintenance of the strongly asymmetric ITCZ. Wind–evaporation–SST feedback as well as the longwave radiative effects of clouds and water vapor on SSTs appear to be important in maintaining the asymmetric regime. The existence of multiple regimes in a single AGCM, and the dependence of these regimes on SST feedbacks, may have a bearing on the ITCZ simulation errors of current coupled climate models.

The sensitivity of the global mean surface temperature generally decreases with increasing insolation, a consequence primarily of increasingly negative shortwave cloud forcing. Climate sensitivity measured across a regime transition can be much larger than the sensitivity within a single regime.

Abstract

Away from the tropical Pacific Ocean, an ENSO event is associated with relatively minor changes of the probability distributions of atmospheric variables. It is nonetheless important to estimate the changes accurately for each ENSO event, because even small changes of means and variances can imply large changes of the likelihood of extreme values. The mean signals are not strictly symmetric with respect to El Niño and La Niña. They also depend upon the unique aspects of the SST anomaly patterns for each event. As for changes of variance and higher moments, little is known at present. This is a concern especially for precipitation, whose distribution is strongly skewed in areas of mean tropospheric descent.

These issues are examined here in observations and GCM simulations of the northern winter (January–March, JFM). For the observational analysis, the 42-yr (1958–99) reanalysis data generated at NCEP are stratified into neutral, El Niño, and La Niña winters. The GCM analysis is based on NCEP atmospheric GCM runs made with prescribed seasonally evolving SSTs for neutral, warm, and cold ENSO conditions. A large number (180) of seasonal integrations, differing only in initial atmospheric states, are made each for observed climatological mean JFM SSTs, the SSTs for an observed warm event (JFM 1987), and the SSTs for an observed cold event (JFM 1989). With such a large ensemble, the changes of probability even in regions not usually associated with strong ENSO signals are ascertained.

The results suggest a substantial asymmetry in the remote response to El Niño and La Niña, not only in the mean but also the variability. In general the remote seasonal mean geopotential height response in the El Niño experiment is stronger, but also more variable, than in the La Niña experiment. One implication of this result is that seasonal extratropical anomalies may not necessarily be more predictable during El Niño than La Niña. The stronger seasonal extratropical variability during El Niño is suggested to arise partly in response to stronger variability of rainfall over the central equatorial Pacific Ocean. The changes of extratropical variability in these experiments are large enough to affect substantially the risks of extreme seasonal anomalies in many regions. These and other results confirm that the remote impacts of individual tropical ENSO events can deviate substantially from historical composite El Niño and La Niña signals. They also highlight the necessity of generating much larger GCM ensembles than has traditionally been done to estimate reliably the changes to the full probability distribution, and especially the altered risks of extreme anomalies, during those events.

Abstract

This paper is concerned with assessing the impact of the El Niño–Southern Oscillation (ENSO) on atmospheric variability on synoptic, intraseasonal, monthly, and seasonal timescales. Global reanalysis data as well as atmospheric general circulation model (AGCM) simulations are used for this purpose. For the observational analysis, 53 yr of NCEP reanalyses are stratified into El Niño, La Niña, and neutral winters [Jan–Feb–Mar (JFM)]. The AGCM analysis is based on three sets of 180 seasonal integrations made with prescribed global sea surface temperatures corresponding to an observed El Niño event (JFM 1987), an observed La Niña event (JFM 1989), and climatological mean JFM conditions. These ensembles are large enough to estimate the ENSO-induced changes of variability even in regions not usually associated with an ENSO effect. The focus is on the anomalous variability of precipitation and 500-mb heights.

The most important result from this analysis is that the patterns of the anomalous extratropical height variability change sharply from the synoptic to the intraseasonal to monthly timescales, but are similar thereafter. In contrast, the patterns of the anomalous tropical rainfall variability are nearly identical across these timescales. On the synoptic and monthly scales, the anomalous extratropical height variability is generally opposite for El Niño and La Niña, and is also roughly symmetric about the equator. On the intraseasonal scale, however, the anomalous height variability is of the same sign for El Niño and La Niña in the Atlantic sector, and is antisymmetric about the equator in the Pacific sector. In the North Pacific, these intraseasonal variance anomalies (which are consistent with a decrease of blocking activity during El Niño and an increase during La Niña) are of opposite sign to the monthly and seasonal variance anomalies.

The sharp differences across timescales in the ENSO-induced changes of extratropical variability suggest that different dynamical mechanisms dominate on different timescales. They also have implications for the predictability of extreme events on those timescales. Finally, there is evidence here that these impacts on extratropical variability may differ substantially from ENSO event to event, especially in the northern Atlantic and over Europe.

Abstract

A linear inverse model (LIM) suitable for studies of atmospheric extratropical variability on longer than weekly timescales is constructed using observations of the past 30 years. Notably, it includes tropical diabatic heating as an evolving model variable rather than as a forcing, and also includes, in effect, the feedback of the extratropical weather systems on the more slowly varying circulation. Both of these features are shown to be important contributors to the model's realism.

Forecast skill is an important test of any model's usefulness as a diagnostic tool. The LIM is better at forecasting week 2 anomalies than a dynamical model based on the linearized baroclinic equations of motion (with many more than the LIM's 37 degrees of freedom) that is forced with observed (as opposed to the LIM's predicted) tropical heating throughout the forecast. Indeed, at week 2 the LIM's skill is competitive with that of the global nonlinear medium-range forecast (MRF) model with nominally O(10⁶) degrees of freedom in use at the National Centers for Environmental Prediction (NCEP). Importantly, this encouraging model performance is not limited to years of El Niño or La Niña episodes. This suggests that accurate prediction of tropical diabatic heating, rather than of tropical sea surface temperatures per se, is key to enhancing extratropical predictability.

The LIM assumes that the dynamics of extratropical low-frequency variability are linear, stable, and stochastically forced. The approximate validity of these assumptions is demonstrated through several tests. A potentially limiting aspect of such a stable linear model with decaying eigenmodes concerns its ability to predict anomaly growth. It is nevertheless found, through a singular vector analysis of the model's propagator, that predictable anomaly growth can and does occur in this dynamical system through constructive modal interference. Examination of the dominant growing singular vectors further confirms the importance of tropical heating anomalies associated with El Niño/La Niña as well as Madden–Julian oscillation episodes in the predictable dynamics of the extratropical circulation. The relative contribution of initial streamfunction and heating perturbations to the development of amplifying anomalies is similarly examined. This analysis suggests that without inclusion of the effects of tropical heating, extratropical weekly averages may be predictable about two weeks ahead, but with tropical heating included, they may be predictable as far as seven weeks ahead.

Abstract

The global and zonal atmospheric angular momentum (AAM) budget is computed from seven years of National Centers for Environmental Prediction data and a composite budget of intraseasonal (30–70 day) variations during northern winter is constructed. Regressions on the global AAM tendency are used to produce maps of outgoing longwave radiation, 200-hPa wind, surface stress, and sea level pressure during the composite AAM cycle. The primary synoptic features and surface torques that contribute to the AAM changes are described.

In the global budget, the friction and mountain torques contribute about equally to the AAM tendency. The friction torque peaks in phase with subtropical surface easterly wind anomalies in both hemispheres. The mountain torque peaks when anomalies in the midlatitude Northern Hemisphere and subtropical Southern Hemisphere are weak but of the same sign.

The picture is different for the zonal mean budget, in which the meridional convergence of the northward relative angular momentum transport and the friction torque are the dominant terms. During the global AAM cycle, zonal AAM anomalies move poleward from the equator to the subtropics primarily in response to momentum transports. These transports are associated with the spatial covariance of the filtered (30–70 day) perturbations with the climatological upper-tropospheric flow. The zonally asymmetric portion of these perturbations develop when convection begins over the Indian Ocean and maximize when convection weakens over the western Pacific Ocean. The 30–70-day zonal mean friction torque results from 1) the surface winds induced by the upper-tropospheric momentum sources and sinks and 2) the direct surface wind response to warm pool convection anomalies.

The signal in relative AAM is complemented by one in “Earth” AAM associated with meridional redistributions of atmospheric mass. This meridional redistribution occurs preferentially over the Asian land mass and is linked with the 30–70-day eastward moving convective signal. It is preceded by a surface Kelvin-like wave in the equatorial Pacific atmosphere that propagates eastward from the western Pacific region to the South American topography and then moves poleward as an edge wave along the Andes. This produces a mountain torque on the Andes, which also causes the regional and global AAM to change.

Abstract

The optimal anomalous sea surface temperature (SST) pattern for forcing North American drought is identified through atmospheric general circulation model integrations in which the response of the Palmer drought severity index (PDSI) is determined for each of 43 prescribed localized SST anomaly “patches” in a regular array over the tropical oceans. The robustness and relevance of the optimal pattern are established through the consistency of results obtained using two different models, and also by the good correspondence of the projection time series of historical tropical SST anomaly fields on the optimal pattern with the time series of the simulated PDSI in separate model integrations with prescribed time-varying observed global SST fields for 1920–2005. It is noteworthy that this optimal drought forcing pattern differs markedly in the Pacific Ocean from the dominant SST pattern associated with El Niño–Southern Oscillation (ENSO), and also shows a large sensitivity of North American drought to Indian and Atlantic Ocean SSTs.