Abstract

The observed wintertime intraseasonal variability of the Northern Hemisphere midtropospheric circulation is analyzed within the framework of an equivalent barotropic model. The analysis centers on the wave domain empirical orthogonal functions (E0Fs) of the 500 mb streamfunction anomalies. The projection of the dynamical model onto the EOFs leads to a system of quadratically nonlinear equations involving the EOF coefficients.

A major result of this study is the identification of the barotropically unstable wintertime mean flow as a potentially important energy source for some of the dominant low frequency EOFS. These EOFS are associated with such hemispheric scale variations as an index cycle, the Pacific/North American pattern, and the North Atlantic Oscillation.

The dominant EOFs are most strongly influenced by nonlinear interactions and this decreases as one goes to the higher order modes. In contrast, the beta and mean flow-EOF interaction terms (which are highly negatively correlated) have a relatively weak influence on the first few EOFs while the strongest influence is on the intermediate EOFs (15–25). The parameterized terms (orography, friction and long-wave correction) have a secondary effect on the EOFs when compared to the advection terms.

Generally, the dominant EOFs represent more unstable flow regimes when compared with the time mean flow. The negative instance of EOF 9, which resembles North Atlantic blocking, is particularly unstable and an inspection of the time mean EOF interactions supports the idea that interactions with the dominant low-frequency modes act to destroy this pattern. In the present model, blocking is most likely to occur as a quasi-linear response to the inhomogeneous forcing which enters into the model as a balance requirement of the time-averaged horizontal flow.

Abstract

The winter-time circulation of the Northern Hemisphere is examined for the following time-scale classes. A) greater than 45 days, B)20–45 days C) 10–20 days D) 6–10 days and E) 2.5&ndash 6 days. The spatial structure of variability within each of these frequency buds is determined by an empirical orthogonal function expansion of the coupled vertical mean and difference streamfunction fields. In general, the dominant low-frequency modes (clan or filter A) exhibit zonally elongted patterns with an equivalent barotropic structure. The intermediate time scale modes (classes B and C) show a tendency for more circular anomaly patterns accompanied by a more baroclinic (westward tilt with height) structure for the class C modes. The dominant frequency modes (classes D and E) are characterized by a strong baroclinic component. The latter exhibit meridionally elongated patterns in the storm track regions together with an upstream tilt to the north and south of the jet exits. In general, lag cross correlations between dominant low-frequency modes tend to be small and/ or symmetric about zero. In contrast, the highest frequency modes exhibit highly asymmetric sinusoidal crow correlation functions reminiscent of travelling waves, while some intermediate and short-time scale modes exhibit correlations suggesting an association with the decaying phases of blocking in the Pacific.

The winter average barotropic and baroclinic energy sources and sinks associated with the time-mean winter flow am examined in the context of a two-layer quasi-geostrophic model. The mean flow kinetic energy (KE) conversion terms reflect the geometry of the anomalies such that the dominant low-frequency modes gain energy, the intermediate frequencies are approximately neutral and the higher frequency modes low energy to the mean flow. The magnitude of the KE conversion is largest in the upper troposphere and is dominated by the lowest-frequency modes. The magnitude conversion from the mean flow to the anomalies is positive for the dominant modes in each band. However, the highest frequency modes associated with traveling waves in the North Atlantic and North Pacific are most efficient in the conversion process. For a typical Rossby radius of deformation, it is found that the net barotropic and baroclinic conversions are of comparable magnitude for periods greater than 10 days while the baroclinic conversion dominates for shorter periods. The majority of the conversions are accomplished by just a few (∼ five) of the dominant EOFs in each frequency band.

Abstract

A method is demonstrated for evaluating global and zonally averaged heat balance statistics based on a four-dimensional assimilation with an atmospheric general circulation model (GCM). The procedure, which provides observationally constrained model diagnostics, uses the GCM of NASA's Goddard Laboratory for Atmospheric Sciences to evaluate the atmospheric heat balance for the February 1976 Data Systems Test period. The global distribution of the adiabatic and diabatic components of the heat balance are obtained by sampling the continuous GCM assimilation shortly after the insertion of conventional synoptic observations. Sampling times of 6 and 9 h after data insertion were chosen to provide adequate damping of high-frequency oscillations in the vertical velocity field caused by the data insertion.

Salient features of the February 1976 analysis include the following: Maximum rising motion in the mean vertical velocity field at 500 mb over South America, south-central Africa, Australia and the Indonesian archipelago. These regions also were characterized by large values of diabatic heating due to convective latent heat release. The cyclogenetically active regions over the North Atlantic and North Pacific oceans were characterized by maxima in latent heat release due to supersaturation cloud formation, and also maxima in the upward and northward transient eddy heat fluxes. In contrast, the continental west coasts showed a tendency for large downward and southward transient eddy beat fluxes.

Some differences are obtained between the heating rates calculated with the model parameterizations and through a residual method. Other shortcomings of the procedure include data deficiencies in the Southern Hemisphere, which cause the results to be comparatively more model dependent in the high southern latitudes.

The potential applicability of this method of analysis to the recently acquired FGGE data is noted.

Abstract

Average predictability and error growth are studied in a realistic two-level general circulation model of the atmosphere via a series of Monte Carlo experiments for fixed external forcing (perpetual winter in the Northern Hemisphere). For realistic initial errors, the dependence of the limit of dynamic predictability on total wave number is similar to that found for the ECMWF model for 1980/81 winter conditions, with the lowest wavenumbers showing significant skill for forecast ranges of more than 1 month. For very small amplitude error (1.2 m rms height at 500 mb) distributed according to the climate spectrum, the total error growth is superexponential, reaching a maximum growth rate (2-day doubling time) in about 1 week.

A simple empirical model of error variance involving two broad wavenumber bands (large scales: n < 10 and small scales: 10 ≤ n ≤ 15), provides an excellent fit of the GCM's error growth behavior. The interpretation of the empirical model, based on an analogy with the stochastic dynamic equations developed by Epstein, suggests that the initial rapid increase in the growth rate of errors in the large scales is primarily due to interactions with the small-scale error. These interactions have preferred geographical locations associated with the position of the climate mean jet streams. However, the error growth of the small scales is large unaffected by the presence of the large-scale error. The initial strong growth rate (2-day doubling time) of the small scales is attributed to the model's high level of eddy activity.

Abstract

Low-frequency (20–70 day) variability is examined during Northern Hemisphere (NH) winter based on seven Years (1981–87) of European Centre for Medium Range Weather Forecasts initialized analyses. The dominant 200 mb zonal wind fluctuations in the Pacific sector, determined from an empirical orthogonal function (EOF) analysis, provide the baseline modes of atmospheric variability, which are related to fluctuations in other circulation parameters and outgoing longwave radiation (OLR). The composite circulation associated with the extreme phases of the zonal wind modes are examined for differences in forcing, wave propagation characteristics and stability.

The dominant upper level zonal wind fluctuation (EOF Z1) is associated with an expanded (contracted) region of easterlies in the tropical western Pacific and changes in the shape and intensity of the subtropical jets. The anomalous (difference between composites) eddy streamfunction at 200 mb shows an enhanced pair of anticyclones (cyclones) straddling the equator. These fluctuations are strongly coupled with eastward traveling tropical convection in the western Pacific with a time scale of about 40 days. The composite circulations show marked differences in the propagation of wave activity in the NH at 200 mb. The low phase (reduced easterlies) shows strong propagation away from the dominant source region over East Asia into the tropical western Pacific in conjunction with what appears to be significant reflection from the equatorward flank of the subtropical jet. In contrast, the high composite (enhanced easterlies) shows much weaker equatorward propagation together with reduced vertical propagation over East Asia and the western North Pacific.

The second zonal wind EOF (Z2) displays a more asymmetric structure with respect to the equator, describing a simultaneous decrease (increase) in the easterly extent of the East Asian jet and increase (decrease) in the strength of the jet over southern Australia. The anomalous eddy stream function at 200 mb shows wave trains apparently emanating from the tropical central Pacific extending into both hemispheres: in the winter hemisphere this resembles the Pacific-North American (PNA) pattern. These fluctuations show some coupling with preceding tropical convection anomalies in the western and central Pacific. Stability calculations show that the PNA pattern is maintained through barotropic energy exchanges with the mean flow. For the low composite, an enhanced source of stationary wave activity in the Gulf of Alaska is associated with an increase in synoptic-scale eddy activity.

These results suggest that tropical convection in the western Pacific has a strong modifying influence on (extratropically forced) middle latitude low-frequency variability. The influence is primarily indirect via zonal wind changes which influence the propagation of waves originating in middle latitudes. The zonal wind changes include those associated with the strength and extent of the tropical easterlies as well as more subtle (but important) changes which effect the curvature of the East Asian jet leading in some instances to turning points for middle latitude waves. The PNA also appears to have its main energy source in middle latitudes and in this case the link with the tropics appears to be more tied to phase locking with anomalies forced by tropical convection in the western and central Pacific.

Abstract

Intraseasonal (20–70 day) variability is examined in the Atlantic region during Northern Hemisphere winter using ECMWF analyses and NOAA outgoing longwave radiation (OLR). It is found that the dominant 200-mb zonal-wind fluctuation over the tropical Atlantic, A1, is related to global-scale circulation anomalies with their origins in the Pacific. Compositing techniques are used to investigate the nature of the Pacific-Atlantic teleconnections and related changes in the tropical OLR and moisture convergence.

The OLR anomalies associated with A1 are characterized by eastward propagation over the Indian Ocean and the western Pacific and a standing oscillation over the tropical Atlantic; the latter extends from Northeast Brazil to West Africa and is the dominant component of the Atlantic OLR variability on these time scales. An analysis of the velocity potential and moisture convergence fields suggests that the fluctuations in convection are coupled between the western Pacific and Atlantic via large-scale (zonal wavenumber 1), equatorially trapped, eastward-propagating waves associated with the Madden-Julian oscillation.

The zonal-wind fluctuation, A1, is also related to extratropical waves propagating into the tropics from both the Northern and Southern hemispheres. The Southern Hemispheric wave train, which makes up the dominant contribution to the A1 circulation pattern, appears to emanate from the western South Pacific and amplifies near the west coast of South America. The Northern Hemispheric wave train resembles the Pacific/North American pattern and emanates from the central North Pacific near the East Asian jet exit region.

These results suggest that a major component of the 20–70-day variability over the Atlantic region is remotely forced. The forcing occurs via the Madden-Julian oscillation, which is strongly coupled with eastward-migrating heating anomalies in the western Pacific and Rossby wave trains, which appear to have their origins in the middle latitudes of the Pacific. The Northern Hemispheric wave train appears to be maintained by energy exchange with the East Asian jet, while the nature of the Southern Hemispheric branch is unclear.

Abstract

East Asian countries experienced record-breaking heat waves and drought conditions during the summer monsoon season of 1994. This study documents the large-scale circulation associated with the drought and suggests a forcing mechanism responsible for the anomalous evolution of the East Asian monsoon. The results, based on Goddard Earth Observing System (GEOS) global assimilated data for 1985–94, indicate that the absence of monsoon rainfall during July 1994 over central China and the southern parts of Korea and Japan is due to the unusually early development of the climatological upper-level anticyclonic flow east of the Tibetan Plateau. The anomalous July anticyclonic circulation over the East Asian–northwestern Pacific region and the cyclonic circulation over the subtropical western Pacific, which are more typical of August, acted to reduce the moisture supply from the western Pacific and the Indian Ocean leading to suppressed rainfall over East Asia. The similarity of the July 1994 East Asian circulation anomalies to the climatological July to August change in these quantities suggests that the anomalies may be viewed as an acceleration of the seasonal cycle in which the circulation transitions to August conditions earlier than normal.

Neither tropical nor middle latitude SST anomalies provide a viable forcing mechanism for the 1994 East Asian circulation anomalies: the tropical anomalies are weak and the middle latitude anomalies, while stronger, appear to be primarily a response to atmospheric forcing, though they may feed back to reinforce the atmospheric anomalies. It is suggested, instead, that the anomalous circulation is primarily the result of an orographic forcing associated with zonal wind changes over Tibet. The zonal wind change, characterized by an anomalous northward shift of the East Asian jet is, in turn, tied to unusually persistent stationary waves extending from northern Europe, which developed prior to the onset of the East Asian anticyclone. Several other occurrences of atmospheric anomalies similar in structure (though weaker in amplitude) to the July 1994 anomalies are found in the previous nine summers, suggesting the operative mechanism is not unique to 1994. Such a mechanism appears to operate both for the climatological development of the ridge and for the occurrences of similar anomalies in previous summers: in the former the northward shift of the jet over Tibet is a reflection of climatological seasonal change in the zonal wind, while in the latter, the shift is the result of anomalies similar in structure to the 1994 European–Asian wave pattern.

The indirect role of the Eurasian waves in the development of the East Asian circulation anomalies suggests that useful monthly and longer predictions of the monsoon rests, not only on our ability to predict the occurrence of these waves, but also on our ability to properly model their interaction with orography.

Abstract

The predictability of the 1997 and 1998 south Asian summer monsoon winds is examined from an ensemble of 10 atmospheric general circulation model simulations with prescribed sea surface temperatures (SSTs) and soil moisture. The simulations have no memory of atmospheric initial conditions for the periods of interest.

The model simulations show that the 1998 monsoon is considerably more predictable than the 1997 monsoon. During May and June of 1998 the predictability of the low-level wind anomalies is largely associated with a local response to anomalously warm Indian Ocean SSTs. Predictability increases late in the season (July and August) as a result of the strengthening of the anomalous Walker circulation and the associated development of easterly low-level wind anomalies that extend westward across India and the Arabian Sea. During these months the model is also the most skillful, with the analyses showing a similar late-season westward extension of the easterly wind anomalies.

The model shows little predictability or skill in the monthly mean low-level winds over Southeast Asia during 1997. Predictable wind anomalies do occur over the western Indian Ocean and Indonesia; however, over the Indian Ocean the predictability is artificial, because the model is responding to SST anomalies that were wind driven. The reduced predictability in the low-level winds during 1997 appears to be the result of a weaker (as compared with 1998) simulated anomalous Walker circulation, and the reduced skill is associated with pronounced intraseasonal activity that is not captured well by the model. It is remarkable that the model does produce an ensemble mean Madden–Julian oscillation (MJO) response, though it is approximately in quadrature with, and much weaker than, the observed MJO anomalies during 1997.

Abstract

The Great Plains region of the United States is characterized by some of the most frequent and regular occurrences of a nocturnal low-level jet (LLJ). While the LLJ is generally confined to the lowest Kilometer of the atmosphere, it may cover a substantial region of the Great Plains, and typically reaches maximum amplitudes of more than 20 m s⁻¹.

A two-month, springtime simulation with the Goddard Earth Observing System (GEOS-1) atmospheric general circulation model (AGCM) has produced a Great Plains LLJ with a vertical and temporal structure, directionality, and climatological distribution that compare favorably with observations. The diurnal cycle of the low-level flow is dramatic and coherent over a subcontinental area that includes much of the western United States and northern Mexico. This cycle can be interpreted as the nightly intrusion of the anticyclonic, subtropical gyre (associated with the Bermuda high) into the North American continent as surface friction decreases. The AGCM also simulates a pair of northerly LLJ maxima off the California coast, which seem to correspond to observations of a so-called “Baja Jet.” Other apparently related diurnal variations extending well into the upper troposphere are documented and compared with observations.

The time-averaged climatological picture of the low-level flow is dominated over land by the nocturnal phase of the diurnal cycle, in which surface friction is minimal and wind speeds are strongest. This pattern, with its zones of strong convergence, is characteristic of an unsteady, strongly forced flow. Over the open ocean, the mean low-level flow is more reminiscent of a smooth, climatological pattern.

Analysis of the simulated moisture budget for the continental United States reveals a horizontally confined region of strong southerly moisture transport with a strong diurnal cycle in the region of the Great Plains LLJ, as has been found in observations of water vapor transport. The LLJ plays a key role in that budget by transporting almost one-third of all the moisture that enters the continental United States with most of the influx from the LLJ (slightly less than two-thirds of it) entering during the 12 nighttime hours. However, it is the mean flow pattern and not covariances associated with the diurnal cycle that contribute most significantly to the total time-mean moisture transport. Covariances on the synoptic and longer timescales contribute only about one-fifth of the total time-mean transport of moisture in the jet region, and covariances on the diurnal timescale are negative and negligible despite the strong diurnal signal in the wind.

Abstract

The column budget technique described by Oort and Vonder Haar (1976) is used to assess the physical consistency and accuracy of estimates of the earth-atmosphere energy balance. Regional estimates of the atmospheric budget terms, the net radiation at the top of the atmosphere, and the time tendency and flux divergence of energy are calculated for the Special Observing Periods of the FGGE year. The data are assimilated by the Goddard Laboratory for the Atmospheres (GLA) four-dimensional analysis system. Ocean heat storage is obtained from marine temperature records while the energy flux through the surface and ocean heat flux divergence are computed as residuals.

During winter the midlatitude oceans supply large quantities of energy to the overlying atmosphere which then transports the energy to the continental heat sinks, the energy flows in the opposite direction during summer. The energy exchange between continental and oceanic regions is much stronger in the Northern Hemisphere where land coverage and land-sea differences are greater.

The uncertainties in the energy balance calculations are assessed by examining the errors in the observations, the data assimilation system including the GLA general circulation model, and the energy budget procedures. Sensitivity tests, error analyses and comparison with other studies indicate that the uncertainties in the continental-scale atmospheric energy flux divergence and the surface energy flux are approximately 20 W m⁻² and 30 W m⁻², respectively. We conclude that at present it is not possible to estimate accurately the ocean heat divergence and transport using the column budget technique.