Search

Upper ocean temperature variability in the tropical Atlantic is examined from the Comprehensive Ocean Atmosphere Data Set (COADS) as well as from an ocean model simulation forced by COADS anomalies appended to a monthly climatology. Our findings are as follows: Only the sea surface temperatures (SST) in the northern tropics are driven by heat fluxes, while the southern tropical variability arises from wind driven ocean circulation changes. The subsurface temperatures in the northern and southern tropics are found to have a strong linkage to buoyancy forcing changes in the northern North Atlantic. Evidence for Kelvin-like boundary wave propagation from the high latitudes is presented from the model simulation. This extratropical influence is associated with wintertime North Atlantic Oscillation (NAO) forcing and manifests itself in the northern and southern tropical temperature anomalies of the same sign at depth of 100-200 meters as result of a Rossby wave propagation away from the eastern boundary in the wake of the boundary wave passage. The most apparent association of the southern tropical sea surface temperature anomalies (STA) arises with the anomalous cross-equatorial winds which can be related to both NAO and the remote influence from the Pacific equatorial region. These teleconnections are seasonal so that the NAO impact on the tropical SST is the largest it mid-winter but in spring and early summer the Pacific remote influence competes with NAO. However, NAO appears to have a more substantial role than the Pacific influence at low frequencies during the last 50 years. The dynamic origin of STA is indirectly confirmed from the SST-heat flux relationship using ocean model experiments which remove either anomalous wind stress forcing or atmospheric forcing anomalies contributing to heat exchange.

An earlier study established the existence of globally coherent interannual fluctuations in atmospheric angular momentum (AAM), associated with the El Nino-Southern Oscillation (ENSO) cycle. In this paper, we pursue the origin and the structure of these fluctuations using an ensemble of experiments generated by the National Centers for Environmental Prediction, medium range forecast model version 9. In the control experiments, where the observed sea surface temperatures (SSTs) were used as the lower boundary conditions, the model captures the characteristic V-like structure in time-latitude plots of zonally averaged AAM, while experiments with climatological SSTs and those with either perpetual warm or cold ENSO conditions superimposed on the climatological SSTs failed to reproduce this structure. The numerical results indicate that these AAM structures are related to SST variations associated with transitions between different phases of the ENSO cycle and have both propagating and standing components. The largest zonal wind contribution from the levels studied (850, 500, and 200 hPa) is at 200 hPa, where the tropical convective outflow is the strongest. Composites of zonal wind and geopotential height show a clear relationship between the stages of the global AAM oscillation and the ENSO cycle. The strong similarity between the simulated and observed AAM series attests to the model's ability to realistically simulate the interannual response of the atmosphere to ENSO SST anomalies.

The stratospheric quasi-biennial oscillation (QBO) provides an important source of interannual variations in the Northern Hemisphere. O'sullivan and Salby (1990) related extra-tropical eddy transport with the phase of the tropical QBO. When the tropical wind is easterly, the zero wind line is shifted into the winter hemisphere. Enhanced wave activity in middle latitudes acts to weaken the polar vortex. When the tropical wind is in the westerly phase the situation reverses. Heights at 30 mb and ozone configurations are contrasted in this paper for these two QBO phases. When the winter vortex deforms due to the amplification of planetary waves 1 and 2, extends westward and equatorward, the complementary band of low vorticity air spirals in toward the pole from lower latitudes. Sometimes, these planetary waves break (Juckes and McIntyre, 1987) and an irreversible mixing of air takes place between high and mid-latitudes. Global ozone patterns, as obtained form satellite observations, appear to be affected by planetary wave breaking (Leovy et al. 1985). This mixing results on regions with uniform ozone and potential vorticity. In the Southern Hemisphere (SH), Newman and Randel (1988) using Total Ozone Mapping Spectrometer (TOMS) data and the NMC analyses have found strong spatial correlation between the October mean temperature in the lower stratosphere and total ozone for the 1979 through 1986 years. Recently Nogues-Paegle et al.(1992) analyzed SH ozone and height data from 1986 to 1989. They found that leading empirical orthogonal functions (EOFs) for both ozone and 50 mb heights exhibit zonal wave 1 and 2 and that the correlations between ozone and 50 mb principal components (PCs) are high. The results were found to be consistent with a linear planetary wave advecting a passive tracer. In this paper, the dominant patterns of variability for 30 mb NMC heights and TOMS total ozone are obtained for the winter to summer transition (January to May) in the Northern Hemisphere (NH) for the years 1987-1990.

The objective of the study was to analyze data from the Dynamical Extended-Range Forecast Experiment conducted from January 1986 to March 1987, and to evaluate differences between analysis and forecasts with emphasis on the tropical 30-50-day oscillation. The diagnostic toll used is the projection of analysis and forecast data onto the normal modes of a primitive equation model. Examination of zonal-wind anomalies in the tropics shows that the forecast model predicts propagation of intraseasonal variations more accurately for slow propagation rates. The forecast amplitude is generally weaker than the analyzed amplitude. Analyzed kinetic energy and error fields exhibit similar horizontal scales for internal and external modes. External Rossby-mode components maximize in the extratropics while Rossby internal modes exhibit patterns that extend over the entire globe.

Global NMC height analyses and Nimbus 7 TOMS Grid-T Version 6 data are used to investigate the horizontal structure of ozone and 50-mb height waves for the 1986-1989 springs of the Southern Hemisphere. Empirical orthogonal functions (EOFs) are obtained for the ozone and height data independently to identify the meridional structure of observed long waves as well as their propagation characteristics. The leading EOFs for both the ozone and 50-mb height waves exhibit wavenumber 1 and 2 patterns. The correlation coefficients between ozone and 50-mb height principal components are high. They indicate high/low ozone values associated with high/low values of 50-mb heights. This distribution is consistent with the advection of ozone by planetary waves south of the ozone maximum, as shown by a simple linear wave model. The observed total ozone amplitudes for waves 1 and 2 for most years are consistent with the total ozone predictions by the wave model.

Low-frequency variability of large-scale atmospheric dynamics can be represented schematically by a Markov chain of multiple flow regimes. This Markov chain contains useful information for the long-range forecaster, provided that the statistical significance of the associated transition matrix can be reliably tested. Monte Carlo simulation yields a very reliable significance test for the elements of this matrix. The results of this test agree with previously used empirical formulae when each cluster of maps identified as a distinct flow regime is sufficiently large and when they all contain a comparable number of maps. Monte Carlo simulation provides a more reliable way to test the statistical significance of transitions to and from small clusters. It can determine the most likely transitions, as well as the most unlikely ones, with a prescribed level of statistical significance.