Search Results

You are looking at 1 - 10 of 27 items for

  • Author or Editor: Hann-Ming Henry Juang x
  • Refine by Access: All Content x
Clear All Modify Search
Hann-Ming Henry Juang

Abstract

A new vertical discretization used in the atmospheric dynamics of the National Centers for Environmental Prediction (NCEP) Global Forecast System (GFS) is illustrated, with enthalpy as the thermodynamic prognostic variable to reduce computation in thermodynamic equations while concerning all gas tracers in the model. Mass, energy, entropy, and angular momentum conservations are utilized as constraints to discretize the vertical integration with a finite-difference scheme. A specific definition of a generalized hybrid vertical coordinate, including sigma, isobaric, and isentropic surfaces, is introduced to define pressure at the model levels. Vertical fluxes are obtained by the equation of local changes in variables defined for vertical coordinates at all model layers. The forward-weighting semi-implicit time scheme is utilized to eliminate computational noise for stable integration. Because of time splitting between the dynamic and physics processes, the vertical advection is required both in the model dynamics and model physics, and the semi-implicit time scheme is used both in dynamics and after physics computation.

Three configurations—sigma, sigma pressure, and sigma entropy—from the specific hybrid vertical coordinates with layer definition similar to NCEP operational GFS have been implemented in the NCEP GFS. Results from the sigma-isentropic coordinate show the largest anomaly correlation and the smallest root-mean-square error in tropical wind among all results at all layers, especially the upper layers. The scores from a period of daily forecast up to 5 days with the sigma-isentropic coordinate show the same level of skill as compared to the NCEP operational GFS. The results from the hurricane tracks for the fall of 2005 with sigma-isentropic coordinates show better scores compared with the operational GFS.

Full access
Hann-Ming Henry Juang

Abstract

Short-range (36-h) simulations of a surface cold front that occurred over the Great Plains of the United States during the SESAME-AYE III period are made using a significantly modified version of a hydrostatic, sigma-coordinate, quasi-Lagrangian, primitive-equation gridpoint model. The necessity of using blended data from FGGE Level IIIb and IIb for input is demonstrated. The evolution of the environment associated with the cold front was well predicted by the model when compared with analyses and observations of the cold front location, horizontal temperature gradient, wind field, dry line, temperature inversion, and precipitation. The current model results caught several observed features that were not present in the other mesoscale model results. Also, comparisons between moist and dry versions of the current model showed a stronger frontal upward motion and a faster moving front showed in the moist version of the model experiment than in the dry version, as expected.

The evolution of several local maxima of frontal intensity, defined in terms of temperature gradient, were related to the combination of local maxima of convergence and deformation effects in different locations along the cold front. The genesis of negative relative vorticity along the inverted cold front over the lee of the United States Rockies was found to be related to the effect of solenoidal forcing. The negative solenoidal forcing here was formed by the orographic pressure gradient perpendicular to the strong temperature gradient. The movement of the local maximum of mesoseale lower-level convergence along the cold front moving from the southern portion of the front to north drove the major frontal intensity from south to north in terms of frontogenesis and cyclogenesis.

The model precipitation agreed well with observation in terms of timing and location. The evolution of maximum precipitation followed the evolution of the maximum low-level convergence ahead of the cold front, where ample moisture combined with potential instability. The generation of the temperature inversion in the vicinity of the frontal surface over the southern portion of the cold front prohibited convection there, and localized the frontal precipitation along the northern portion of cold front and east of southern temperature inversion. The temperature inversion was generated by differential temperature advection through the frontal circulation within the PBL, and modified by the vertical stretching effect that stabilized the lower layer behind the cold front and destabilized ahead of the cold front.

Full access
Hann-Ming Henry Juang

Abstract

A reduced spectral transformation is applied to the NCEP atmospheric global spectral model for operational seasonal forecasts. The magnitude of the associated Legendre coefficient provides a basis for this new transformation, which is a simple modification of a traditional reduced grid spectral transform. This transformation can be called a “reduced spectral” method because its Fourier and Legendre transformations need less computation than the traditional uniform full grid or reduced grid methods. In addition, the reduced spectral method saves an extra 50% on Legendre transformations and is easy to load balance for massively parallel computing under certain decompositions.

A comparison, without model physics, among reduced spectral, reduced grid, and full grid transforms indicates that they have negligible differences up to more than a half-month integration and small differences up to a 1- month integration. Extended integrations without physics for up to 4 months show that there is proximity of zonal symmetry between reduced spectral and full grid transforms. When the comparison includes model physics, the results show negligible differences up to 7 days; but the chaotic nature, known as an internal variability, is amplified by physical parameterizations and produces significant differences among these methods after a 1- month integration, which is expected. The seasonally averaged results from 10 years of AMIP-type runs are similar between the reduced spectral method and the full grid method, indicating that they have similar model climatology. These experiments indicate that this reduced spectral transform can be used for short-range as well as seasonal or climate predictions.

Full access
Hann-Ming Henry Juang

Abstract

This paper illustrates a modified nonhydrostatic version of the National Centers for Environmental Prediction regional spectral model (RSM). This nonhydrostatic version of the RSM can simulate atmospheric motions of all scales, especially mesoscale. For simplicity, it is referred to in this paper as the mesoscale spectral model (MSM). The preliminary results of the previous version of the MSM have been published on the year of 1992, with coarse resolution in three dimensions and no model physics. A fine-resolution two-dimensional version has been tested on classical problems and published on the year of 1994.

Instead of an externally determined hydrostatic coordinate as originally designed in 1992 MSM, the internally evolved hydrostatic coordinate as used in the RSM is implemented. This modification makes the MSM closer to the hydrostatic version in model structure and dynamics. Besides the hydrostatic perturbation, related to the external hydrostatic state as perturbation nesting, the nonhydrostatic perturbation related to the internally evolved hydrostatic state is introduced. The same model physics used in the RSM are used in the MSM without the hydrostatic assumption. The major numerical techniques used in the hydrostatic version are used in the MSM as well. They are spectral computation, fourth-order horizontal diffusion, time filter, and semi-implicit adjustment for perturbation. The hydrostatic state, interpolated from the hydrostatic global model, is used as the initial condition without initialization or data assimilation, and it can be integrated up to several days with reasonable predictions.

Extended tests of thermal bubbles and mountain waves in very fine resolutions by this revised MSM showed its behavior to be the same as, but not superior to, those of the previous version. These results are compatible to other model results in the literature. Cases using real data with full model physics as used in the RSM show that the revised MSM has reasonable results and is superior to the previous version as compared with the RSM in a coarse horizontal grid resolution of about 50 km. Furthermore, it shows that it can be successfully nested into the hydrostatic global model at 10- to 20-fold differences in horizontal resolution with a small domain due to the well-behaved perturbation nesting over flat plains, coastal oceans, and steep mountains.

Full access
Yoshi Ogura and Hann-Ming Henry Juang

Abstract

A diagnostic analysis is made for the rapid development of two subsynoptic scale cyclones that coexisted over Canada in the spring season, using the Level IIIb First GARP Global Experiment dataset assimilated by the European Centre for Medium Range Weather Forecasts and National Weather Service operational data. Both cyclones started developing at 0000 UTC 25 April 1979. When development began, these cyclones wore separated from each other by a distance of only about 1300 km. Nonetheless, the physical processes leading to their initial cyclogenesis are found to be different. One of the cyclones remained a weak, shallow surface low for about 48 h after it formed in a zone of developing upper-level westerly waves. It started developing only when it drifted into a region of the horizontal advection of upper-level potential vorticity anomaly associated with a strong tropopause fold. In contrast, the other cyclone formed in a region of strong surface frontogenesis caused primarily by the velocity deformation. Once formed, it developed rapidly and propagated within a narrow zone of small Richardson number, suggesting that the cyclone developed due to localized baroclinic instability. Eventually the latter cyclone also moved into the region of upper-level potential vorticity advection and absorbed the former to become a major synoptic-scale cyclone. Its deepening rate came close to that of explosive cyclogenesis, while heating by latent heat release was found to be of secondary importance in the rapid development from the heat budget analysis in the core of the cyclone.

Full access
Wilbur Y. Chen and Hann-ming Henry Juang

Abstract

Prediction of blocking flows by a comprehensive general circulation model is still not satisfactory. A large portion of the unskillful forecasts can be traced to the model's inability to predict the evolution of blocking beyond a few days into the forecast. Realizing the fact that blocking is often observed to form following a series of intensive cyclogenesis activities and that the model tends to underestimate the intensity of the synoptic-scale transient eddies, a series of 10-day forecasts were conducted to assess the impact of transient eddies on the establishment of blocking flows. When the fast-propagating synoptic-scale disturbances were suppressed in the initial conditions, the subsequent forecasts completely failed to predict a blocking anticyclone. However, when the transient eddies were enhanced in the initial conditions to compensate for the deficiency of the model, blocking flows were predicted and evolved in remarkable agreement with the observations.

The dynamical processes during establishment of blocking flow were then examined by a series of daily isentropic potential vorticity charts. The role played by the transient eddies can be identified through these charts, which help to explain why the transient eddies are crucial in establishing the blocking flows.

Full access
Hann-Ming Henry Juang and Masao Kanamitsu

Abstract

A nested primitive equation regional spectral model is developed. The model consists of two components—a low-resolution global spectral model and a high-resolution regional spectral model. The two components have identical vertical structure and physical processes. The global model component is a low-resolution version of the operational National Meteorological Center (NMC) global spectral model and uses spherical harmonics as horizontal basis functions. The regional spectral model component is a primitive equation model on a stereographic projection and uses sine-cosine series as horizontal basis functions. The feature of the regional component is that it predicts deviations from the forecast of the global model component, first proposed by Hoyer. A semi-implicit time scheme, time filtering, initialization, and horizontal diffusion are applied to these deviations in the regional domain.

Several sensitivity experiments on “nesting periods,” lateral boundary treatments, and different global model base fields were performed. The results indicate that experiments with 3- or 6-h nesting periods had less noise along the lateral boundaries than those with 1-h nesting period. It was also found that blending along the lateral boundaries may not be necessary but that with relaxation the use of T30 or T62 resolution in the global model was sufficient for a regional model with a horizontal resolution of 80 km.

The model was tested on real-data cases and was shown to have skill comparable to or better than the other NMC operational regional models. The cases shown in this paper included a 48-h prediction of an East Coast disturbance and the “U.S. storm of the century” in March 1993, all done with a horizontal resolution of 80 km, and a 5-day forecast of a hurricane track done with a horizontal resolution of 40 km.

Full access
Song-You Hong and Hann-Ming Henry Juang

Abstract

A simple method to employ an orography blending technique near the lateral boundary of the regional model is proposed. The method blends the orography of the regional model on the lateral boundary with that of the global or coarser resolution regional model. This method is tested in the NCEP Regional Spectral Model. A long-range prediction experiment of one month’s duration is conducted in the East Asian region. It is found that the proposed method provides regional model forecasts consistent with the global model forecasts without a discernible systematic bias. It is likely that modifying the regional model orography following the values of global model or analysis data is more efficient than elaborating the numerics of a lateral boundary scheme to eliminate systematic error.

Full access
Hann-Ming Henry Juang and Song-You Hong

Abstract

This paper evaluates the performance of the National Centers for Environmental Prediction (NCEP) Regional Spectral Model (RSM) based on the sensitivities of different model domain sizes and horizontal resolutions. The perturbation method and the spectral computation in the NCEP RSM construct the nesting strategy as a “domain nesting” in physical space as well as a “spectral nesting” in spectral space, instead of the conventional “lateral boundary nesting” as used in most regional models. The NCEP RSM has the same model structure, dynamics, and physics as its outer coarse-resolution global model, and it also has a terrain blending along the lateral boundary at the initial time. Both together result in a smooth lateral boundary behavior through one-way nesting. An optimal lateral boundary relaxation reduces the influence of lateral boundary error and generates more areas with small-scale features. The treatment of numerical stabilities, such as a semi-implicit time scheme, time filter, and horizontal diffusion, are applied in perturbation without recomputing or disturbing the large-scale waves. The combination of the aforementioned methods is the uniqueness of the NCEP RSM, which demonstrates the capabilities to conserve the large-scale waves, resolve the mesoscale features, and minimize the lateral boundary errors.

A case of winter cyclogenesis with propagation of the synoptic-scale disturbances through the lateral boundaries was selected to investigate the sensitivities of the NCEP RSM based on different nesting strategies. The results from the experiment over a quarter-sphere domain with similar resolutions between RSM and T126 global model demonstrated that the domain nesting was a success, because the lateral boundary error and perturbation were negligibly small. The experiments in a 48-km resolution with different sizes of the model domain had mixed results. The continental domain had the best performance but inclined to generate erroneous large-scale waves that degraded its performance after 60 h. The results of the regional and subregional domains were proximity to their base field, T126, in terms of root-mean-square differences. They had similar mesoscale features in a 48-km horizontal resolution regardless of the different model domain sizes. The results from the experiments with nesting in different coarse grids over the radar-range domain imply that it can use either a T126 or subregional domain as its base field for similar performances. Nevertheless, more mesoscale features were obtained by the experiment with the base field at higher resolution. The results from the experiments, with 30-day integration, reveal that the performance of the experiment in the subregional domain was much closer to its base field than that in the continental domain. It indicates that the predictability of the global model is the predictability of the NCEP RSM in the regional domain; however, the regional domain could generate higher-resolution features than its base field. This successful long-range integration with the regional domain is because the lateral boundary errors are relatively small and the large-scale waves are preserved through the domain and spectral nesting.

Full access
Hann-Ming Henry Juang and Yoshi Ogura

Abstract

Short-range (36 and 48 h) simulations of a rapid cyclogenesis that occurred over Canada in the spring season are made using a significantly modified version of the hydrostatic, sigma-coordinate, quasi-Lagrangian grid model developed by Mathur. The modifications include the adaptations of a recent variant Kuo cumulus parameterization scheme, a revised computation of the horizontal pressure gradient force, and a high-resolution planetary boundary layer parameterization. Analyzed data are used as an initial field, and to specify the values at the lateral boundaries of the model during integration as well as to verify the results of simulations. The model has several versions in the combinations of dry or moist physics, two different vertical resolutions (18 and 9 layers), and three different horizontal grid spacings (180, 90 and 45 km). The event was diagnostically analyzed in Part I of Ogura and Juang and represents a unique case of two coexisting mesoα-scale surface lows that developed through different physical processes and subsequently merged to form a synoptic-scale cyclone.

The model simulates the rapid development of the two surface lows and their subsequent merging reasonably well. The simulated precipitation pattern also agrees with observations. A simulation using a dry version of the model is essentially similar to that from a moist version of the model. This suggests that diabatic heating by condensation is not essential in the rapid cyclogenesis in this case, unlike many explosive cyclogenesis events for winter storms over the ocean. Instead, horizontal advection of potential vorticity and localized baroclinic instability are the essential ingredients in this case. Impacts of reduced vertical and horizontal resolution of the model on the accuracy of the simulations are examined. Also discussed is an effect of boundary layer friction on the prediction of the sea-level central pressure of the surface lows.

Full access