Files in this item
|(no description provided)|
|Title:||Atmospheric Predictability in the Presence of Tropical Forcing|
|Doctoral Committee Chair(s):||Robinson, Walter A.|
|Department / Program:||Atmospheric Sciences|
|Degree Granting Institution:||University of Illinois at Urbana-Champaign|
|Subject(s):||Physics, Atmospheric Science|
|Abstract:||We investigate the impact of tropical forcing on the predictability of the atmosphere in a two-layer spectral model. The focus is on the improvement of extended range forecasts in the extratropics from knowledge of the tropical forcing. The model-twin approach is used to perform forecast experiments including and excluding midlatitude orography.
Perturbing the two-layer model by an isolated heating anomaly centered on the equator, we obtain a significant model response which resembles the "Gill-Hoskins" teleconnection pattern in the Northern Hemisphere. Barotropic diagnoses based on the notion of the Rossby wave source (RWS) provides an insight into the structure of this teleconnection. The tropical forcing not only affects the eddy field but also generates systematic fluctuation of the zonal means in the tropics through wave/mean flow interaction.
While some fundamental properties of the model's predictability are examined, it is found that there are considerable modulations of the predictability by the presence of the tropical forcing. Proportional to the strength of the stationary eddies, the predictability of the model including any steady external forcing is longer than that of the unforced model.
Although knowing the correct tropical forcing improves the midlatitude forecasts at longer timescales only modestly in a zonally homogeneous model, including a mountain 30$\sp\circ$ downstream of the tropical forcing intensifies the tropical impact significantly over the exit region of the midlatitude jet stream. This local enhancement of the tropical influence, however, is weakened when the mountain is moved 60$\sp\circ$ upstream of the tropical forcing.
The orography modifies the impact of the tropical forcing in two ways: First, the orography amplifies the ensemble mean response of the extratropical atmosphere to the tropical forcing. Secondly, the tropical signal can change the characteristics of the local barotropic instability, by modulating the wavy structure of the time mean flow. This, in turn, generates local difference in error variance between forecasts including and excluding the tropical forcing.
In the case either including or excluding the orography, the significant improvements of midlatitude forecasts from knowledge of the tropical forcing are confined to the middle third of the first month of the model prediction. The dynamical impact of the tropical forcing depends not only on the sensitivity of the ensemble mean response but also on the extent to which the tropical forcing modulates the internal variability. The intensity of the tropical influence represents the relative importance of the forcing effects and the internal variability of the model atmosphere.
Thesis (Ph.D.)--University of Illinois at Urbana-Champaign, 1993.
|Date Available in IDEALS:||2014-12-17|