The Mechanism for the Scale Selection of Tropical Intraseasonal Oscillations. Part I: Selection of Wavenumber 1 and the Three-Scale StructureSource: Journal of the Atmospheric Sciences:;1988:;Volume( 046 ):;issue: 012::page 1779Author:Itoh, Hisanori
DOI: 10.1175/1520-0469(1989)046<1779:TMFTSS>2.0.CO;2Publisher: American Meteorological Society
Abstract: The mechanism for the scale selection of tropical intraseasonal oscillations (ISO) is examined in a stepwise procedure with the use of two models, which differ only in the cumulus parameterization. In the first model, moisture budgets are treated explicitly. Using this model, the ISO with the wavenumber 1 pattern is successfully reproduced and some possible causes responsible for the selection of wavenumber 1 can be specified. The second model does not employ moisture and cumulus convection is related to low-level convergence only. The possible causes obtained by the use of the first model are parameterized and can be implemented or eliminated from the second model to clarify which property is an essential mechanism of the wavenumber selection. In order that the wavenumber 1 pattern is selected, cumulus convection must not be formed over large having low-level convergence, but must be confined to narrow regions. This can be realized by limiting the areas of the wet regions, or preventing weak low-level convergence from generating cumulus convection. Another important result is that the cumulus convection should make clusters with the scale of a few thousand kilometers (the ?Super cluster?) Using these results, the selection of wavenumber 1 can be understood by the following mechanism. Low-level moisture convergence near the equator prevails due to the Hadley circulation. If cumulus convection is generated in places where moisture convergence occurs, cloud clusters will be formed in more than one region, since cumulus is produced over large areas. Therefore, in order for the super cluster to be confined to only one region, it is necessary that cumulus convection occurs under a more severe condition. This condition is that dry region exist over wide areas in the tropics. As a result, the generation of cumulus is suppressed in these regions, even the large areas where weak moisture convergence exists, so that cumulus convection occurs only in one narrow confined area. Thus, the wavenumber 1 pattern is selected. Next, it is shown that the super clusters are not the minimum scale in the ISO, but am composed of smaller-scale clusters, which generally move westward as observed. This small scale corresponds to the most unstable scale predicted by linear wave-CISK theory. Thus, the ISO can be simulated as a three-scale structure: the scale of wavenumber 1, the super-cluster scale and the small scale.
|
Collections
Show full item record
| contributor author | Itoh, Hisanori | |
| date accessioned | 2017-06-09T14:29:03Z | |
| date available | 2017-06-09T14:29:03Z | |
| date copyright | 1989/06/01 | |
| date issued | 1988 | |
| identifier issn | 0022-4928 | |
| identifier other | ams-20102.pdf | |
| identifier uri | http://onlinelibrary.yabesh.ir/handle/yetl/4156294 | |
| description abstract | The mechanism for the scale selection of tropical intraseasonal oscillations (ISO) is examined in a stepwise procedure with the use of two models, which differ only in the cumulus parameterization. In the first model, moisture budgets are treated explicitly. Using this model, the ISO with the wavenumber 1 pattern is successfully reproduced and some possible causes responsible for the selection of wavenumber 1 can be specified. The second model does not employ moisture and cumulus convection is related to low-level convergence only. The possible causes obtained by the use of the first model are parameterized and can be implemented or eliminated from the second model to clarify which property is an essential mechanism of the wavenumber selection. In order that the wavenumber 1 pattern is selected, cumulus convection must not be formed over large having low-level convergence, but must be confined to narrow regions. This can be realized by limiting the areas of the wet regions, or preventing weak low-level convergence from generating cumulus convection. Another important result is that the cumulus convection should make clusters with the scale of a few thousand kilometers (the ?Super cluster?) Using these results, the selection of wavenumber 1 can be understood by the following mechanism. Low-level moisture convergence near the equator prevails due to the Hadley circulation. If cumulus convection is generated in places where moisture convergence occurs, cloud clusters will be formed in more than one region, since cumulus is produced over large areas. Therefore, in order for the super cluster to be confined to only one region, it is necessary that cumulus convection occurs under a more severe condition. This condition is that dry region exist over wide areas in the tropics. As a result, the generation of cumulus is suppressed in these regions, even the large areas where weak moisture convergence exists, so that cumulus convection occurs only in one narrow confined area. Thus, the wavenumber 1 pattern is selected. Next, it is shown that the super clusters are not the minimum scale in the ISO, but am composed of smaller-scale clusters, which generally move westward as observed. This small scale corresponds to the most unstable scale predicted by linear wave-CISK theory. Thus, the ISO can be simulated as a three-scale structure: the scale of wavenumber 1, the super-cluster scale and the small scale. | |
| publisher | American Meteorological Society | |
| title | The Mechanism for the Scale Selection of Tropical Intraseasonal Oscillations. Part I: Selection of Wavenumber 1 and the Three-Scale Structure | |
| type | Journal Paper | |
| journal volume | 46 | |
| journal issue | 12 | |
| journal title | Journal of the Atmospheric Sciences | |
| identifier doi | 10.1175/1520-0469(1989)046<1779:TMFTSS>2.0.CO;2 | |
| journal fristpage | 1779 | |
| journal lastpage | 1798 | |
| tree | Journal of the Atmospheric Sciences:;1988:;Volume( 046 ):;issue: 012 | |
| contenttype | Fulltext |