Search Results
You are looking at 1 - 10 of 45 items for
- Author or Editor: Simon Chang x
- Refine by Access: All Content x
Abstract
Numerical simulations with a primitive equation model which includes parameterized physics are conducted to study the effects of an island mountain range on translating tropical cyclones. The idealized topography with a 200 m peak is introduced over a 12 h growth period. The initial state contains a nonlinearly balanced vortex embedded in a uniform, unsheared, tropical easterly flow.
Many orographic effects are produced similar to those observed for typhoons passing over mountain ranges. The storm tends to translate at about twice the speed of the basic flow near the mountain, while its intensity is reduced. Air flows mostly around the mountain range instead of over it, forming a ridge on the windside and a trough on the leeside slopes. The tropical cyclone's passage induces a mean cyclonic circulation around the mountain with strongest amplitudes at low levels. As a result, the model tropical cyclone makes a cyclonic curvature in its path around the north end of the island mountain.
Further numerical experiments suggest that cumulus heating which maintains the tropical cyclone forces the cyclonic circulation around the mountain. In the experiment with an unforced, quasi-barotropic vortex we found that the lower level circulation is blocked by the mountain range. As the original low-level center fails to pass the mountain range, a secondary low-level circulation center forms in the induced lee trough. The secondary low-level center develops as the upper level center comes into phase.
A vorticity budget is performed for the 700 mb airflow prior to landfall and confirms the importance of diabatic processes in producing the observed orographic effects. Diabatic processes generate convergence to maintain the vorticity of the tropical cyclone. The horizontal advection of positive vorticity in conjunction with the leeside vortex stretching, results in the mean positive vorticity around the mountain.
Abstract
Numerical simulations with a primitive equation model which includes parameterized physics are conducted to study the effects of an island mountain range on translating tropical cyclones. The idealized topography with a 200 m peak is introduced over a 12 h growth period. The initial state contains a nonlinearly balanced vortex embedded in a uniform, unsheared, tropical easterly flow.
Many orographic effects are produced similar to those observed for typhoons passing over mountain ranges. The storm tends to translate at about twice the speed of the basic flow near the mountain, while its intensity is reduced. Air flows mostly around the mountain range instead of over it, forming a ridge on the windside and a trough on the leeside slopes. The tropical cyclone's passage induces a mean cyclonic circulation around the mountain with strongest amplitudes at low levels. As a result, the model tropical cyclone makes a cyclonic curvature in its path around the north end of the island mountain.
Further numerical experiments suggest that cumulus heating which maintains the tropical cyclone forces the cyclonic circulation around the mountain. In the experiment with an unforced, quasi-barotropic vortex we found that the lower level circulation is blocked by the mountain range. As the original low-level center fails to pass the mountain range, a secondary low-level circulation center forms in the induced lee trough. The secondary low-level center develops as the upper level center comes into phase.
A vorticity budget is performed for the 700 mb airflow prior to landfall and confirms the importance of diabatic processes in producing the observed orographic effects. Diabatic processes generate convergence to maintain the vorticity of the tropical cyclone. The horizontal advection of positive vorticity in conjunction with the leeside vortex stretching, results in the mean positive vorticity around the mountain.
Abstract
A planetary boundary-layer (PBL) parameterization based on the generalized similarity theory (GST) was tested in tropical cyclone models. This parameterization, with only one layer, is desired in modeling tropical cyclones for computational speed. The momentum, sensible heat and moisture fluxes are mutually dependent in this parameterization through nondimensional gradient equations. The internal structure of the PBL is determined implicitly through universal functions.
In comparison with a complex, one-dimensional, multilayer PBL model, the GST parameterization yields accurate moisture fluxes, but slightly overestimates the momentum flux and underestimates the sensible heal flux. The GST parameterization produces very realistic dynamics, energetics and thermal structure in an axisymmetric tropical cyclone model. This GST parameterization, although unable to treat the diffusion across the PBL inversion, is judged superior to drag coefficient parameterization and is a good alternative to the more expensive, multilayer parameterization.
Abstract
A planetary boundary-layer (PBL) parameterization based on the generalized similarity theory (GST) was tested in tropical cyclone models. This parameterization, with only one layer, is desired in modeling tropical cyclones for computational speed. The momentum, sensible heat and moisture fluxes are mutually dependent in this parameterization through nondimensional gradient equations. The internal structure of the PBL is determined implicitly through universal functions.
In comparison with a complex, one-dimensional, multilayer PBL model, the GST parameterization yields accurate moisture fluxes, but slightly overestimates the momentum flux and underestimates the sensible heal flux. The GST parameterization produces very realistic dynamics, energetics and thermal structure in an axisymmetric tropical cyclone model. This GST parameterization, although unable to treat the diffusion across the PBL inversion, is judged superior to drag coefficient parameterization and is a good alternative to the more expensive, multilayer parameterization.
Abstract
An efficient, multilayer model for predicting the diurnal variations in the thermal and momentum fields in the planetary boundary layer (PBL) is proposed for incorporating into mesoscale or large-scale dynamical models. The ground temperature is given by a soil slab heated (or cooled) by net radiation and sensible heat from the atmospheric surface layer and a ground thermal reservoir. The surface heat flux can be generated by two mechanisms: 1) the convective mixing depending on the temperature difference between the ground and the screen level and 2) the mechanical mixing depending on the wind stress. Following Blackadar (1976), a prediction equation is employed for the screen-level temperature. In the PBL, the heat and momentum exchanges are computed by a Richardson number adjustment scheme. Heat and momentum exchanges occur mainly due to thermal instability under convectively unstable conditions and due to shear instability under convectively stable conditions. A case study shows good agreement between model results and observation. Additional experiments are performed to test the scheme under calm and stronger wind situations. Since no explicit diffusion coefficient is needed in the adjustment scheme, the model time step is not restricted by computational stability requirements of the diffusion term. This PBL parameterization scheme is therefore very appealing for use in numerical models that use large time steps yet have good vertical resolutions in the PBL.
Abstract
An efficient, multilayer model for predicting the diurnal variations in the thermal and momentum fields in the planetary boundary layer (PBL) is proposed for incorporating into mesoscale or large-scale dynamical models. The ground temperature is given by a soil slab heated (or cooled) by net radiation and sensible heat from the atmospheric surface layer and a ground thermal reservoir. The surface heat flux can be generated by two mechanisms: 1) the convective mixing depending on the temperature difference between the ground and the screen level and 2) the mechanical mixing depending on the wind stress. Following Blackadar (1976), a prediction equation is employed for the screen-level temperature. In the PBL, the heat and momentum exchanges are computed by a Richardson number adjustment scheme. Heat and momentum exchanges occur mainly due to thermal instability under convectively unstable conditions and due to shear instability under convectively stable conditions. A case study shows good agreement between model results and observation. Additional experiments are performed to test the scheme under calm and stronger wind situations. Since no explicit diffusion coefficient is needed in the adjustment scheme, the model time step is not restricted by computational stability requirements of the diffusion term. This PBL parameterization scheme is therefore very appealing for use in numerical models that use large time steps yet have good vertical resolutions in the PBL.
Abstract
An axisymmetric, hydrostatic ocean model containing a rigid bottom and a free surface is constructed to study the barotropic and baroclinic response in the upper and deep ocean to a wind stress corresponding to a stationary tropical cyclone. The numerical model covers a domain of 800 km and 1475 m in r- and z-directions, respectively, with a uniform radial resolution of 20 km and a stretched vertical resolution from 5 to 54 m. The vertical mixing is parameterized based on a local Richardson number and a mixing length.
The model ocean is spun up with the wind stress of Hurricane Eloise. A strong tangential circulation develops that extends to the ocean floor with a maximum speed of 1.2 m s−1 at the surface. The circulation on the r-z plane, which also extends to the ocean floor, oscillates with time with a maximum upwelling of 0.1 cm s−1 at the center. Surface height has a maximum depression of 57 cm. The deep overturning causes density changes deep in the ocean. A maximum temperature decrease of 3°C occurs in the mixed layer at the center; a maximum temperature increase of 0.45°C is found just below the thermocline at a radius of 200 km. The recovery of both the mass and momentum fields is very slow during the spindown. Inertial oscillations dominate in the spindown even in the deep ocean. Adjustments between the momentum and mass fields seem to converge to a state quite different from the prestorm state.
Direct comparison with observations is difficult because the model is only two-dimensional. Nevertheless, recent observations seem to suggest the existence of the barotropic response in the deep mean. The model suggests that the observed rapid response in the deep ocean is caused by the barotropic pressure gradient force, which arises from the storm-induced perturbation of the free surface.
Abstract
An axisymmetric, hydrostatic ocean model containing a rigid bottom and a free surface is constructed to study the barotropic and baroclinic response in the upper and deep ocean to a wind stress corresponding to a stationary tropical cyclone. The numerical model covers a domain of 800 km and 1475 m in r- and z-directions, respectively, with a uniform radial resolution of 20 km and a stretched vertical resolution from 5 to 54 m. The vertical mixing is parameterized based on a local Richardson number and a mixing length.
The model ocean is spun up with the wind stress of Hurricane Eloise. A strong tangential circulation develops that extends to the ocean floor with a maximum speed of 1.2 m s−1 at the surface. The circulation on the r-z plane, which also extends to the ocean floor, oscillates with time with a maximum upwelling of 0.1 cm s−1 at the center. Surface height has a maximum depression of 57 cm. The deep overturning causes density changes deep in the ocean. A maximum temperature decrease of 3°C occurs in the mixed layer at the center; a maximum temperature increase of 0.45°C is found just below the thermocline at a radius of 200 km. The recovery of both the mass and momentum fields is very slow during the spindown. Inertial oscillations dominate in the spindown even in the deep ocean. Adjustments between the momentum and mass fields seem to converge to a state quite different from the prestorm state.
Direct comparison with observations is difficult because the model is only two-dimensional. Nevertheless, recent observations seem to suggest the existence of the barotropic response in the deep mean. The model suggests that the observed rapid response in the deep ocean is caused by the barotropic pressure gradient force, which arises from the storm-induced perturbation of the free surface.
Abstract
The impact of satellite-sensed winds on the intensity forecasts of tropical cyclones is evaluated by a simulation study with an axisymmetric numerical model. The parameterized physics in the forecast model are deliberately made different from those in the model that generates the observation. Model-generated “observations” are assimilated into forecasts by 12 h dynamic initialization.
A series of 24 h forecasts with and without assimilation of satellite-sensed winds are conducted and compared with the observations. Results indicate that assimilation with marine surface (or low-level) wind alone does not improve intensity forecasts appreciably, that a strong relaxation coefficient in the initialization scheme causes model rejection of the assimilation, and that an attenuating relaxation coefficient is recommended. However, when wind observations at the outflow level are included in the assimilation, forecasts improve substantially. The best forecasts are achieved when observations over the entire lower troposphere are assimilated.
Additional experiments indicate the errors in the satellite observations contaminate the forecast. But the assimilation of inflow and outflow winds still improve the intensity forecast if the satellite observation errors are less than or about the same magnitude of those in the initial wind field.
Abstract
The impact of satellite-sensed winds on the intensity forecasts of tropical cyclones is evaluated by a simulation study with an axisymmetric numerical model. The parameterized physics in the forecast model are deliberately made different from those in the model that generates the observation. Model-generated “observations” are assimilated into forecasts by 12 h dynamic initialization.
A series of 24 h forecasts with and without assimilation of satellite-sensed winds are conducted and compared with the observations. Results indicate that assimilation with marine surface (or low-level) wind alone does not improve intensity forecasts appreciably, that a strong relaxation coefficient in the initialization scheme causes model rejection of the assimilation, and that an attenuating relaxation coefficient is recommended. However, when wind observations at the outflow level are included in the assimilation, forecasts improve substantially. The best forecasts are achieved when observations over the entire lower troposphere are assimilated.
Additional experiments indicate the errors in the satellite observations contaminate the forecast. But the assimilation of inflow and outflow winds still improve the intensity forecast if the satellite observation errors are less than or about the same magnitude of those in the initial wind field.
Abstract
An axisymmetric, multilayer, numerical tropical cyclone model with a well-resolved planetary boundary layer is used to test the response of local, instantaneous changes of sea surface temperature (SST). One experiment shows that the storm's intensity is steadily decreased as the SST in the inner 300 km is instantaneously cooled by 2°C. However, in the second experiment, in which the SST is cooled by 2°C outside the radius of 300 km, the storm shows no immediate and appreciable weakening. The intensity of the tropical cyclone in this case is maintained by enhanced evaporation in the inner 300 km and increased baroclinicity.
Abstract
An axisymmetric, multilayer, numerical tropical cyclone model with a well-resolved planetary boundary layer is used to test the response of local, instantaneous changes of sea surface temperature (SST). One experiment shows that the storm's intensity is steadily decreased as the SST in the inner 300 km is instantaneously cooled by 2°C. However, in the second experiment, in which the SST is cooled by 2°C outside the radius of 300 km, the storm shows no immediate and appreciable weakening. The intensity of the tropical cyclone in this case is maintained by enhanced evaporation in the inner 300 km and increased baroclinicity.
Abstract
The interactions between atmospheric vortex pairs are simulated and studied with a nondivergent barotropic model and a three-dimensional tropical cyclone model.
Numerical experiments with nondivergent barotropic vortex pairs show that the relative movements of the vortices are sensitive to the separation distance and the characteristics of the swirling wind of the vortex. No mutual attraction is found in any of the nondivergent barotropic vortex pairs tested.
Results from the three-dimensional tropical cyclone model show that on a constant ƒ-plane with no mean wind, the movements of the two interacting tropical cyclones consist of a mutual cyclonic rotation, attraction and eventual merging, in agreement with Fujiwhara's description. The displacement of one interacting storm in the mutual rotation is proportional to the combined strength of the binary system, but inversely proportional to the size of the storm and to the square of the separation distance. The rate of merging is related to the development of a mean secondary circulation on the radial–vertical plane, and is quite independent of the strength of the two tropical cyclones.
The latitudinal variation of the Coriolis parameter adds a northwest beta drift to the trajectories. Depending on their relative strength and location, the beta drift either speeds up the merging process or separates the two interacting tropical cyclones.
Abstract
The interactions between atmospheric vortex pairs are simulated and studied with a nondivergent barotropic model and a three-dimensional tropical cyclone model.
Numerical experiments with nondivergent barotropic vortex pairs show that the relative movements of the vortices are sensitive to the separation distance and the characteristics of the swirling wind of the vortex. No mutual attraction is found in any of the nondivergent barotropic vortex pairs tested.
Results from the three-dimensional tropical cyclone model show that on a constant ƒ-plane with no mean wind, the movements of the two interacting tropical cyclones consist of a mutual cyclonic rotation, attraction and eventual merging, in agreement with Fujiwhara's description. The displacement of one interacting storm in the mutual rotation is proportional to the combined strength of the binary system, but inversely proportional to the size of the storm and to the square of the separation distance. The rate of merging is related to the development of a mean secondary circulation on the radial–vertical plane, and is quite independent of the strength of the two tropical cyclones.
The latitudinal variation of the Coriolis parameter adds a northwest beta drift to the trajectories. Depending on their relative strength and location, the beta drift either speeds up the merging process or separates the two interacting tropical cyclones.
Abstract
Numerical experiments were conducted to assess the impact of Omega dropwindsonde (ODW) data and Special Sensor Microwave/Imager (SSM/I) rain rates in the analysis and prediction of Hurricane Florence (1988). The ODW data were used to enhance the initial analysis that was based on the National Meteorological Center/Regional Analysis and Forecast System (NMC/RAFS) 2.5° analysis at 0000 UTC 9 September 1988. The SSM/I rain rates at 0000 and 1200 UTC 9 September 1988 were assimilated into the Naval Research Laboratory's limited-area model during model integration.
Results show that the numerical prediction with the ODW-enhanced initial analysis was superior to the control without ODW data. The 24-h intensity forecast error is reduced by about 75%, landfall location by about 95% (reduced from 294 to 15 km), and landfall time by about 5 h (from 9 to 4 h) when the ODW data were included. Results also reveal that the assimilation of SSM/I-retrieved rain rates reduce the critical landfall location forecast error by about 43% (from 294 to 169 km) and the landfall time forecast error by about 7 h (from 9 to 2 h) when the NMC/RAFS 2.5° initial analysis was not enhanced by the ODW data. The assimilation of SSM/I rain rates further improved the forecast error of the landfall time by 4 h (from 4 to 0 h) when the ODW data were used. This study concludes that numerical predictions of tropical cyclone can benefit from assimilations of ODW data and SSM/I-retrieved rain rates.
Abstract
Numerical experiments were conducted to assess the impact of Omega dropwindsonde (ODW) data and Special Sensor Microwave/Imager (SSM/I) rain rates in the analysis and prediction of Hurricane Florence (1988). The ODW data were used to enhance the initial analysis that was based on the National Meteorological Center/Regional Analysis and Forecast System (NMC/RAFS) 2.5° analysis at 0000 UTC 9 September 1988. The SSM/I rain rates at 0000 and 1200 UTC 9 September 1988 were assimilated into the Naval Research Laboratory's limited-area model during model integration.
Results show that the numerical prediction with the ODW-enhanced initial analysis was superior to the control without ODW data. The 24-h intensity forecast error is reduced by about 75%, landfall location by about 95% (reduced from 294 to 15 km), and landfall time by about 5 h (from 9 to 4 h) when the ODW data were included. Results also reveal that the assimilation of SSM/I-retrieved rain rates reduce the critical landfall location forecast error by about 43% (from 294 to 169 km) and the landfall time forecast error by about 7 h (from 9 to 2 h) when the NMC/RAFS 2.5° initial analysis was not enhanced by the ODW data. The assimilation of SSM/I rain rates further improved the forecast error of the landfall time by 4 h (from 4 to 0 h) when the ODW data were used. This study concludes that numerical predictions of tropical cyclone can benefit from assimilations of ODW data and SSM/I-retrieved rain rates.
Abstract
An axisymmetric, multilayer hurricane model is used to investigate the hurricane's response to sudden changes of sea surface temperature (SST). The model contains a parameterization of the planetary boundary layer (PBL) which includes matched formulations for the surface layer and the mixed layer. The heat, moisture and momentum fluxes are mutually dependent through Monin-Obukhov similarity theory.
The height of the model hurricane PEL is 400–500 m, below which the potential temperature and specific humidity are nearly invariant with height. The flow in the hurricane PBL is characterized by subgradient tangential velocities and nearly uniform cross-isobaric flow angles. The sensible heating from the ocean is insignificant, but the evaporation is large. The magnitudes of the equivalent drag coefficients are approximately one-third those of the exchange coefficients for heat and moisture.
As the SST is suddenly decreased (increased), the steady-state model hurricane experiences two stages of modification. The first stage consists of adjustments of the hurricane PBL featuring a weakened (enhanced) dynamic and thermodynamic coupling of the storm with the ocean. No important changes of intensity occur during this stage, which lasts several hours. The decrease (increase) of kinetic energy dissipation offsets part of the decrease (increase) of kinetic energy generation. The second stage is characterized by a steady modification of storm intensity. The fluctuations of intensity in these experiments are less pronounced than those shown by a similar model with a conventional bulk parameterization of the hurricane PBL.
Abstract
An axisymmetric, multilayer hurricane model is used to investigate the hurricane's response to sudden changes of sea surface temperature (SST). The model contains a parameterization of the planetary boundary layer (PBL) which includes matched formulations for the surface layer and the mixed layer. The heat, moisture and momentum fluxes are mutually dependent through Monin-Obukhov similarity theory.
The height of the model hurricane PEL is 400–500 m, below which the potential temperature and specific humidity are nearly invariant with height. The flow in the hurricane PBL is characterized by subgradient tangential velocities and nearly uniform cross-isobaric flow angles. The sensible heating from the ocean is insignificant, but the evaporation is large. The magnitudes of the equivalent drag coefficients are approximately one-third those of the exchange coefficients for heat and moisture.
As the SST is suddenly decreased (increased), the steady-state model hurricane experiences two stages of modification. The first stage consists of adjustments of the hurricane PBL featuring a weakened (enhanced) dynamic and thermodynamic coupling of the storm with the ocean. No important changes of intensity occur during this stage, which lasts several hours. The decrease (increase) of kinetic energy dissipation offsets part of the decrease (increase) of kinetic energy generation. The second stage is characterized by a steady modification of storm intensity. The fluctuations of intensity in these experiments are less pronounced than those shown by a similar model with a conventional bulk parameterization of the hurricane PBL.
