1996 Vol. 13, No. 1
Display Method:
1996, 13(1): 1-18.
doi: 10.1007/BF02657024
Abstract:
Different versions of a new nine-layer general circulation model which is rhomboidally truncated it zonal wave number 15 (L9R15) are introduced in this paper. On using the observed global monthly sea surface temperature (SST) and sea ice (SI) data from 1979 to 1988 offered by the international Atmospheric Model Inter-comparison Program (AMIP), these different model versions were integrated for the ten-year AMIP period. Results show that the model is capable of simulating the basic states of the atmosphere and its interannual variability, and in performing reasonable sensitivity experiments
Different versions of a new nine-layer general circulation model which is rhomboidally truncated it zonal wave number 15 (L9R15) are introduced in this paper. On using the observed global monthly sea surface temperature (SST) and sea ice (SI) data from 1979 to 1988 offered by the international Atmospheric Model Inter-comparison Program (AMIP), these different model versions were integrated for the ten-year AMIP period. Results show that the model is capable of simulating the basic states of the atmosphere and its interannual variability, and in performing reasonable sensitivity experiments
1996, 13(1): 19-32.
doi: 10.1007/BF02657025
Abstract:
A preferred growing perturbation concept has been introduced into the dynamical study on the generation of atmospheric teleconnection patterns. It is shown that all the important teleconnection patterns observed in the real atmosphere may be established through internal barotropic dynamical processes
A preferred growing perturbation concept has been introduced into the dynamical study on the generation of atmospheric teleconnection patterns. It is shown that all the important teleconnection patterns observed in the real atmosphere may be established through internal barotropic dynamical processes
1996, 13(1): 33-42.
doi: 10.1007/BF02657026
Abstract:
By employing Arnol’d’s method (energy-Casimir), this paper has studied the nonlinear stability of the two-layer generalized Phillips’ model for which the top and bottom surfaces are either rigid or free, and obtained some nonlinear stability criteria. In addition, some linear stability criteria are obtained by normal mode method. The re-sults reveal the influences of the free surface parameter on the stability of atmospheric and oceanic motions
By employing Arnol’d’s method (energy-Casimir), this paper has studied the nonlinear stability of the two-layer generalized Phillips’ model for which the top and bottom surfaces are either rigid or free, and obtained some nonlinear stability criteria. In addition, some linear stability criteria are obtained by normal mode method. The re-sults reveal the influences of the free surface parameter on the stability of atmospheric and oceanic motions
1996, 13(1): 43-58.
doi: 10.1007/BF02657027
Abstract:
Ocean response to atmospheric forcing in the CZ ocean model is analyzed. The results show that Nino 3 index from the CZ ocean model driven by linear composite of biennial, ENSO and even annual time scale wind stress anomalies is consistent well with composite of responding two or three components of observed Nino 3 index during the El Nino period while the La Nina phenomena cannot be reproduced by the linear composite. It implies that linear response process for ocean response to atmospheric forcing is dominated during the El Nino period while nonlinear response might be main process during the La Nina period Simulated results also suggest that optimal response frequency of the CZ ocean model is the frequency lower than annual variability and ocean response to the atmospheric forcing with annual time scale can give rise to incorrect signal-errors in the simulated SSTA field
Ocean response to atmospheric forcing in the CZ ocean model is analyzed. The results show that Nino 3 index from the CZ ocean model driven by linear composite of biennial, ENSO and even annual time scale wind stress anomalies is consistent well with composite of responding two or three components of observed Nino 3 index during the El Nino period while the La Nina phenomena cannot be reproduced by the linear composite. It implies that linear response process for ocean response to atmospheric forcing is dominated during the El Nino period while nonlinear response might be main process during the La Nina period Simulated results also suggest that optimal response frequency of the CZ ocean model is the frequency lower than annual variability and ocean response to the atmospheric forcing with annual time scale can give rise to incorrect signal-errors in the simulated SSTA field
1996, 13(1): 59-66.
doi: 10.1007/BF02657028
Abstract:
It is impossible, mathematically, to use a time series which is regarded as a set or observational facts of a dynamic system to reconstruct the particular system. Physically, however, with a few assumptions put, a dynamic system can be rebuilt approximately by means of observational facts. This is the goal of the so called invariant quantity method (IQM), whose research and experiment are of potential significance to atmospheric sciences
It is impossible, mathematically, to use a time series which is regarded as a set or observational facts of a dynamic system to reconstruct the particular system. Physically, however, with a few assumptions put, a dynamic system can be rebuilt approximately by means of observational facts. This is the goal of the so called invariant quantity method (IQM), whose research and experiment are of potential significance to atmospheric sciences
1996, 13(1): 67-90.
doi: 10.1007/BF02657029
Abstract:
Without any other approximations apart from the spectral method which is employed, the energy spectra corre-sponding to two kinds of "negative temperatures" are simulated with a symmetric trapezium truncation. The simu-lated results with either of the two negative temperatures are reasonably consistent with those from the statistical the-ory of turbulence. The more usual case for two positive temperatures evolves differently from the theoretical predic-tion.The viscosity influence on the ergodicity is discussed. It is shown that two-dimensional (2D) ideal flows on the sphere have a less pronounced tendency to be ergodic than those on planar geometry due to the curvature of the spherical surface that weakens the interaction between different parts of the flow, enabling these parts to behave in more relative isolation. The expressions for the standard deviations from a canonical ensemble for the two different options of coefficients are shown to be proportional to (N is the total number of independent modes in the sys?tem), independent of the initial conditions of the system
Without any other approximations apart from the spectral method which is employed, the energy spectra corre-sponding to two kinds of "negative temperatures" are simulated with a symmetric trapezium truncation. The simu-lated results with either of the two negative temperatures are reasonably consistent with those from the statistical the-ory of turbulence. The more usual case for two positive temperatures evolves differently from the theoretical predic-tion.The viscosity influence on the ergodicity is discussed. It is shown that two-dimensional (2D) ideal flows on the sphere have a less pronounced tendency to be ergodic than those on planar geometry due to the curvature of the spherical surface that weakens the interaction between different parts of the flow, enabling these parts to behave in more relative isolation. The expressions for the standard deviations from a canonical ensemble for the two different options of coefficients are shown to be proportional to (N is the total number of independent modes in the sys?tem), independent of the initial conditions of the system
1996, 13(1): 91-102.
doi: 10.1007/BF02657030
Abstract:
In order to determine nonlinear energy exchanges into individual triad interactions in the frequency domain, spectral formulas are derived by use of the cross-spectral technique. First time attempt has been made to understand the problem of maintenance of low frequency waves for tropical weather system by using this technique. The TOGA basic level III daily wind analyses on a 2.5 degree square grid around the global zone from 20oS-30oN at 200 hPa for 92-day periods covering June, July and August of 1988 are used. Kinetic energy is gained at high frequencies and lost at low frequencies. In the planetary scale dynamics over tropics, barotropic nonlinear energy transfer plays a negative role. Low frequency wave of period 45 day loses maximum amount of energy when it interacted with frequencies of periods 92 day and 30 day. 45 day cycle is also the main source of energy for other frequencies. Distrubances of period 15 day gain maximum amount of energy. The major contribution comes from the triad interaction of the frequencies . North of 20oN low frequency waves of period 30 to 92-day gain energy through nonlinear triad interaction with the maximum gain at 22.5oN. The study may help to investigate the rapid loss of predictability of low frequency modes over tropics
In order to determine nonlinear energy exchanges into individual triad interactions in the frequency domain, spectral formulas are derived by use of the cross-spectral technique. First time attempt has been made to understand the problem of maintenance of low frequency waves for tropical weather system by using this technique. The TOGA basic level III daily wind analyses on a 2.5 degree square grid around the global zone from 20oS-30oN at 200 hPa for 92-day periods covering June, July and August of 1988 are used. Kinetic energy is gained at high frequencies and lost at low frequencies. In the planetary scale dynamics over tropics, barotropic nonlinear energy transfer plays a negative role. Low frequency wave of period 45 day loses maximum amount of energy when it interacted with frequencies of periods 92 day and 30 day. 45 day cycle is also the main source of energy for other frequencies. Distrubances of period 15 day gain maximum amount of energy. The major contribution comes from the triad interaction of the frequencies . North of 20oN low frequency waves of period 30 to 92-day gain energy through nonlinear triad interaction with the maximum gain at 22.5oN. The study may help to investigate the rapid loss of predictability of low frequency modes over tropics
1996, 13(1): 103-114.
doi: 10.1007/BF02657031
Abstract:
A bogus typhoon scheme, designed for the initialization of a typhoon track prediction model, is developed in this paper. This scheme includes both effects of axisymmetric wind and asymmetric wind. Experimental forecasts using a two-way interactive movable nested mesh model show that the forecast skill of typhoon tracks has clearly improved after introducing the bogus typhoon into the initial fields
A bogus typhoon scheme, designed for the initialization of a typhoon track prediction model, is developed in this paper. This scheme includes both effects of axisymmetric wind and asymmetric wind. Experimental forecasts using a two-way interactive movable nested mesh model show that the forecast skill of typhoon tracks has clearly improved after introducing the bogus typhoon into the initial fields
1996, 13(1): 115-123.
doi: 10.1007/BF02657032
Abstract:
In terms of a baroclinic quasi-geostrophic wave-filtering technique in connection with a dimensionless paramet-er, η(z), of condensation-released latent heat that indicates the CISK mechanism, a model is established for des-cribing tropical atmosphere CISK-Rossby waves alongside its analytical solution. Theoretical study shows that there exists pronounced difference between Rossby waves, CISK-involving and classic, and the former can be used to in-terpret some aspects of the low-frequency oscillation in the tropical atmosphere
In terms of a baroclinic quasi-geostrophic wave-filtering technique in connection with a dimensionless paramet-er, η(z), of condensation-released latent heat that indicates the CISK mechanism, a model is established for des-cribing tropical atmosphere CISK-Rossby waves alongside its analytical solution. Theoretical study shows that there exists pronounced difference between Rossby waves, CISK-involving and classic, and the former can be used to in-terpret some aspects of the low-frequency oscillation in the tropical atmosphere
1996, 13(1): 124-132.
doi: 10.1007/BF02657033
Abstract:
The vertical and horizontal propagation of internal gravity waves forced by a local thermal source at the lower boundary is examined using the irrotational two-dimensional model developed by Zhang et al. (1995). The source generates waves with the same absolute frequency 1 / 15 d-1 but different wavenumber. Internal gravity waves with wide spreading frequencies and wavenumbers appear due to wave-wave interactions. The foreced waves 1W (west-ward-propagating with wavenumber 1) and IE (eastward-propagating with wavenumber 1), whose absolute phase speeds are 31 m s-1, are predominant. Other forced waves with smaller phase speeds are relatively weaker. These waves interact with the zonal mean flow, causing the flow to alternate with easterly and westerly with a 32-month pe-riod. The absolute maximum wind speeds of easterly and westerly are 35 m s-1 which is larger than the largest phase speed of forced waves. The forced waves IE and 1W may accelerate easterly and westerly up to 31 m s-1, respectively. The excessive accelerations are due to the self-acceleration of waves IE and 1W or newly generated high frequency waves IE and 1W The horizontal propagation of the waves are mainly affected by the forcing and wave-low interactions which lead to fast-slow variations of wave phase speed and amplitude.
The vertical and horizontal propagation of internal gravity waves forced by a local thermal source at the lower boundary is examined using the irrotational two-dimensional model developed by Zhang et al. (1995). The source generates waves with the same absolute frequency 1 / 15 d-1 but different wavenumber. Internal gravity waves with wide spreading frequencies and wavenumbers appear due to wave-wave interactions. The foreced waves 1W (west-ward-propagating with wavenumber 1) and IE (eastward-propagating with wavenumber 1), whose absolute phase speeds are 31 m s-1, are predominant. Other forced waves with smaller phase speeds are relatively weaker. These waves interact with the zonal mean flow, causing the flow to alternate with easterly and westerly with a 32-month pe-riod. The absolute maximum wind speeds of easterly and westerly are 35 m s-1 which is larger than the largest phase speed of forced waves. The forced waves IE and 1W may accelerate easterly and westerly up to 31 m s-1, respectively. The excessive accelerations are due to the self-acceleration of waves IE and 1W or newly generated high frequency waves IE and 1W The horizontal propagation of the waves are mainly affected by the forcing and wave-low interactions which lead to fast-slow variations of wave phase speed and amplitude.