Abstract

The extreme precipitation events with peak observed rainfall of greater than 700 mm over the South Island of New Zealand were simulated using the DAR hydrostatic mesoscale model nested within the ECMWF analyses. The ECMWF analyses for two of the events showed a low-level jet with mixing ratios greater than 12 g kg⁻¹ crossing the South Island of New Zealand during the heavy precipitation near a cold front. The third case, which had smaller mixing ratios, occurred as a low-level jet and crossed the South Island while a low redeveloped downstream.

Three different orographies were used with the 30-km horizontal resolution model runs, with progressively increased terrain heights. The highest orography was created by artificially inserting the effective barrier of the Southern Alps to northwesterly flow in the model grid. Orography had a strong influence on the amount of precipitation: the peak precipitation was related to orographic slope while the area-averaged precipitation was related to the maximum orographic elevation. The model successfully simulated nearly half the peak observed precipitation and over half the area-averaged precipitation (determined by hydrological means) in two of the cases and much less in the third case. Refining the horizontal resolution from 30 to 15 km also increased the peak precipitation amounts. However, the area-averaged precipitation in the 15-km runs was not significantly larger than in the 30-km runs, suggesting more concentrated precipitation over a smaller area.

All simulations, except the artificial barrier orography case, produced a mountain wave consistent with linear theory, in spite of the nonsteady flow, irregular orography, and the large amount of diabatic heating present. The amplitude of the mountain wave increased with mountain height and resolution. The absence of a mountain wave in the run with the artificial barrier orography indicates unrealistic flow for that configuration.

Abstract

The extreme precipitation event that occurred on 27 December 1989 over the South Island of New Zealand was simulated using the DAR hydrostatic mesoscale model nested within the ECMWF analyses. The model simulated nearly half of the peak observed rainfall for this storm (greater than 700 mm) and captured the location and timing of the intense precipitation.

The heavy precipitation developed while a deep layer of moist subtropical air along a cold front ascended the high terrain of the South Island. The intense orographic ascent was associated with a low-level jet core with wind speeds of over 20 m s⁻¹ ahead of the cold front. An upper-level trough and jet streak entrance region were also present upstream of the South Island during the event, aiding the ascent over the mountains and deepening the layer of moist air. The air crossing the mountain was nearly saturated throughout the troposphere and had only weak moist vertical stability near the cold front. Almost all of the simulated precipitation formed in the low troposphere through forced ascent, with only minimal convection behind the cold front.

Two sensitivity experiments were conducted to investigate the effects of orography and latent heating on the development of precipitation in the simulations. Weak upstream blocking by the orography was present, enhancing the ascent upstream and causing a slight moistening of the midtroposphere. The latent heat, maximized near the surface on the upwind side of the mountain, caused increased upward motion and precipitation over the orography and decreased ascent upstream, tending to dry and stabilize the air there. The latent heat release weakened the blocking effect of the orography and altered the mountain wave through reduced effective dry static stability.

Abstract

The energetics of a Southern Hemisphere cyclone wave have been analyzed using ECMWF data and the results of a limited-area model simulation. An analysis of the energy budget for a storm that developed in the eastern Pacific on 4–6 September 1987 showed the advection of the geopotential height field by the ageostrophic wind to be both a significant source and the primary sink of eddy kinetic energy. Air flowing through the wave gained kinetic energy via this term as it approached the energy maximum and then lost it upon exiting. Energy removal by diffusion, friction, and Reynolds stresses was found to be small. The most important conclusion was that, while the wave grew initially by poleward advection of heat as expected from baroclinic theory, the system evolved only up to the point where this source of eddy energy and the conversion of eddy potential to eddy kinetic energy (typically denoted “ωα”) was compensated for by energy flux divergence (dispersion of energy), mainly of the ageostrophic geopotential flux, v_aϕ. Energy exported in this fashion was then available for the downstream development of a secondary system. This finding seems to differ from the results of studies of the life cycle of normal-mode-type waves in zonal flows, which have been shown to decay primarily through transfer of energy to the mean flow via Reynolds stresses. However, this apparent inconsistency can be explained by the fact that while ageostrophic geopotential fluxes can also be very large in the case of individual normal modes, the waves export energy downstream at exactly the same rate as they gain from upstream. The group velocity of the 4–6 September storm, calculated from the ageostrophic geopotential height fluxes, showed that the energy packet comprising the system had an eastward group velocity slightly larger than the time-mean flow.

Abstract

The importance of subcloud evaporation to the thermodynamics and movement of cold fronts is investigated through inclusion of an explicit cloud scheme within a 30-km resolution limited-area model. Two cases are examined: 18 November 1984 and 26 February 1995. The effect of the subcloud evaporation is deduced by comparing simulations with and without the evaporation for these two cases. The implications of these results for weather prediction and climate models are discussed.

The first case occurred during the Australian Cold Fronts Research Programme with mesoscale data available to verify the model simulation. The results indicate that the movement of the synoptic cold front was retarded when a prefrontal cool change developed in response to the evaporation of the grid-resolved precipitation. However, the cooling ahead of the front effectively accelerated the cool change, more in line with observations.

The second case involved the prognoses of a cool change crossing Victoria in southeast Australia during a potential bushfire day. In this case, most of the precipitation occurred along and behind the cold front. With precipitation evaporation, the cool change accelerated several hundred kilometers farther in 36 h than in the simulation without precipitation evaporation.

Abstract

The importance of subgrid-scale processes for the simulation of midlatitude frontal clouds by global models is investigated. The case chosen is a frontal cloud associated with a cool change crossing the southern Australian coastline between 17 and 19 November 1984. The Commonwealth Scientific and Industrial Research Organisation limited-area model, Division of Atmospheric Research Limited-Area Model, was run at horizontal resolutions of 30 and 300 km, and the results of the 30-km simulation were then averaged to 300-km resolution. Comparisons and evaluations of the simulations showed that the 300-km simulation failed to develop the frontal clouds. Comparison with the 30-km simulation averaged to 300 km showed the importance of the subgrid-scale vertical motions for this cloud development. In particular, it is found that the covariance of the subgrid-scale terms, although of smaller magnitude when compared with the larger-scale terms, needs to be parameterized to capture correctly the frontal cloud development. It is suggested that parameterization of the subgrid-scale dynamical forcing is important for the correct cloud development in general circulation models.

Abstract

A regional climate model (DARLAM) is implemented over the Australian region and a 20-yr seasonally varying simulation is examined for the presence of tropical cyclone–like vortices (TCLVs). The horizontal resolution of the model is 125 km with nine vertical levels and is forced at its boundaries by the output of the Commonwealth Scientific and Industrial Research Organisation GCM using a mixed layer (or “slab”) ocean. Additional simulations are performed with a horizontal resolution of 30 km and with 18 vertical levels to examine the impact of increasing resolution on storm intensity. A sample of TCLVs from the 125-km resolution simulation is simulated at 30-km resolution to determine whether they reach observed tropical storm intensity at the finer resolution. It is found that stronger vortices in the 125-km resolution simulation are more likely to intensify when simulated at finer resolution than weaker vortices. In this way, a detection threshold for vortices generated in the 125-km resolution simulation is established and then used to detect TCLVs in that simulation. The regional climate model DARLAM provides a good simulation of both cyclogenesis and its seasonal variation under the current climate. The response of the model under enhanced greenhouse conditions is studied. Under 2 × CO₂ conditions, there is no statistically significant change in regions of formation of TCLVs, with only a slight southward shift simulated. Nevertheless, there are statistically significant effects on the poleward movement of TCLVs, with storms generally tending to travel farther poleward in a warmer climate once they have formed. An analysis is undertaken to determine the reasons for this behavior. While the dynamical constraints on the maintenance of TCLV intensity under 2 × CO₂ conditions (e.g., vertical wind shear) are similar to those in the current climate, thermodynamic conditions (e.g., sea surface temperatures) are quite different and are likely to be at least partly the cause of this effect. Other causes include the combination of the slight southward shift in formation and a tendency for TCLV tracks to be more southward in enhanced greenhouse conditions, a consequence of more southward steering winds.

Abstract

The ability of the CSIRO-9 General Circulation Model (GCM) to capture surface cutoff lows over eastern Australia is investigated by comparing composites of ten GCM cases with ten observed lows. The lows are also studied individually to compare their development and movement, as well as synoptic features, which may have been smoothed out in the compositing process. Finally, the incidence of all such lows in the 1 × CO₂ and 2 × CO₂ simulations are examined to determine the possible effects a doubling of CO₂ will have on their occurrence.

The GCM surface lows were found to develop from an upper-level cutoff low in a manner similar to the observed lows. In both sets, this development took place between the upper-level subtropical and polar jets in all seasons except summer, where only one jet was evident. Latent heat release appeared to play an important role in the intensification of the surface lows. The main difference between the two sets of cutoff lows was that the GCM surface lows tended to develop farther to the east of the upper-level cutoff, the upper-level features were less intense and occlusion did not take place. As a result, the GCM lows had a greater eastward translation compared to the observed lows, which often meander along the east coast for several days while they intensify. These features appear to be related to the low resolution of the GCM.

The frequency of east Australian cutoff lows was underpredicted in the WM by about 45% in the 1 × CO₂ simulation, with the greatest underprediction occurring in autumn and winter. Analysis of upper-level jet structure indicated that the GCM produced a poor simulation of the dual jet structure aloft, which may account for this problem. The 2 × CO₂ simulation produced even fewer cutoff lows over eastern Australia. This was probably caused by the reduced baroclinicity due to increased warming of polar regions, which resulted in an even weaker dual jet structure. The cast Australian cutoff lows were found to be more intense in the 2 × CO₂ Simulation, suggesting the greater role played by latent heat effects once development has been initiated.

Abstract

A nested global, limited-area model was used to predict the President's Day cyclone of 18-19 February 1979. Both a low (∼150 km) and a high (∼50 km) horizontal resolution version were used. The model has full physics with a planetary boundary layer, moisture, moist convective adjustment, and radiation.

The low-resolution model using a global analysis for initial and boundary conditions (termed a simulation), was able to capture the general development and movement of the cyclone. Some discrepancies were noted for the intensity of upper-air features between the analyses and the model solution during the first 24 hours. The primary focus of this paper is to determine the effect of initial and boundary conditions, as well as model parameterizations on the accuracy of the predictions. The evolution of the storm is discussed with an emphasis on the quality of the numerical simulation.

The impact of the initial conditions on the model solution was tested by using four different global analyses. It was found that the variability between the solutions was less than the variability between the analyses. Varying the horizontal diffusion in the model produced stronger development with weaker diffusion, but the character of the development did not change significantly. The sensitivity of the simulation to latent heat was tested by running the model without latent heating. A low did develop in this model solution, although it was much weaker and it did not develop vertically as in the cases with latent heating.

The most significant improvement in accuracy in this sensitivity study occurred when the horizontal resolution was increased from 1.25° × 1.0° (∼150 km) to 0.4° × 0.32° (∼50 km). The position and intensity of the surface low were much close to reality, as indicated by comparison with a mesoanalysis and to satellite pictures.

The nested model was also run in forecast mode with boundary conditions for the limited-area model supplied by the (Geophysical Fluid Dynamics Laboratory) GFDL global model forecast. In general, the quality of the limited-area forecast compared very well with the simulations. The overall character and intensity of the development were similar.

The role of lateral boundary conditions was demonstrated by comparing forecasts and simulations with identical initial conditions. The results suggest the increasing importance of the boundary data with time in the limited-area forecast and show high correlation between the errors in the limited-area forecast and the global forecast within the limited-area domain.

Abstract

A multimode Chameleon Surface Model (CHASM) with different levels of complexity in parameterizing surface energy balance is coupled to a limited-area model (DARLAM) to investigate the impacts of complexity in land surface representations on the model simulation of a tropical synoptic event. A low pressure system is examined in two sets of numerical experiments to discuss the following. (i) Does land surface parameterization influence regional numerical weather simulations? (ii) Can the complexity of land surface schemes in numerical models be represented by parameter tuning? The model-simulated tracks of the low pressure center do not, overall, show large sensitivity to the different CHASM modes coupled to the limited-area model. However, the landing position of the system, as one measurement of the track difference, can be influenced by several degrees in latitude and about one degree in longitude. Some of the track differences are larger than the intrinsic numerical noise in the model estimated from two sets of random perturbation runs. In addition, the landing time of the low pressure system can differ by about 14 h. The differences in the model-simulated central pressure exceed the model intrinsic numerical noise and such variations consistent with the differences seen in simulated surface fluxes. Furthermore, different complexity in the land surface scheme can significantly affect the model rainfall and temperature simulations associated with the low center, with differences in rainfall up to 20 mm day⁻¹ and in surface temperature up to 2°C. Explicitly representing surface resistance and bare ground evaporation components in CHASM produces the most significant impacts on the surface processes. Results from the second set of experiments, in which the CHASM modes are calibrated by parameter tuning, demonstrate that the effects of the physical processes represented by extra complexity in some CHASM modes cannot be substituted for by parameter tuning in simplified land surface schemes.

Abstract

A rapidly deepening cyclone that occurred over the South Pacific on 5 September 1987 was investigated in order to assess the possible factors contributing to its development. Cyclogenesis took place when a disturbance in the subtropics merged with a wave in the polar westerlies. Analysis revealed that the evolution of the cyclone system was associated with the interaction of a potential vorticity anomaly from the subpolar region with a subtropical surface disturbance in a manner typical of “Class B” cyclogenesis. As the storm intensified, the subtropical jet merged with the polar jet, producing a strong poleward heat transport characteristic of baroclinic systems. However, the absence of tilt to the frontal zone, together with weak vertical wind shear, was suggestive of a significant barotropic component to the storm. The zonal average of potential vorticity over the storm displayed large regions where the meridional gradients have different signs, indicating that the system could have developed initially by internal instabilities (barotropic and/or baroclinic) without significant external forcings.

Sensitivity experiments were conducted to determine the role of surface processes in the development of the storm. It was found that development was insensitive to both surface heat fluxes and the presence of South American topography, with little change in either the circulation or kinetic energy of the storm. Intensification of the storm was substantially affected by surface frictional effects, as indicated by significant increases in the vertically averaged kinetic energy when the surface roughness was reduced. The results suggest a need to reduce the roughness heights not only over sea ice, but over the ocean in areas of strong winds as well.