Search Results

You are looking at 1 - 10 of 60 items for

  • Author or Editor: J. Michael Fritsch x
  • Refine by Access: All Content x
Clear All Modify Search
J. MICHAEL FRITSCH

Abstract

A procedure to use the spline interpolation technique on an arbitrarily prescribed two-dimensional data field is described. For using this technique, one must obtain an initial approximation to the data at the grid points. This is achieved by fitting spherical surfaces to the data. Bidirectional spline interpolation is then applied repeatedly on the grid point estimates of the data to produce convergence to the true surface.

The spline interpolation technique and another objective analysis technique developed by Cressman are tested against an exact solution, and the resulting analyses are compared. Real temperature, geopotential height, and wind data for various pressure surfaces are analyzed by the spline method; and the results are compared to subjective analyses of the same data.

Full access
J. Michael Fritsch

Abstract

Modification of mesoscale convective weather systems through ice-phase seeding is briefly reviewed. a simple mathematical framework for estimating the likely mesoscale response to convective cloud modification is presented, and previous mesoscale modification hypotheses are discussed in the context of this mathematical framework. Some basic differences between cloud-scale and mesoscale modification hypotheses are also discussed. Numerical model experiments to test the mesoscale sensitivity of convective weather systems are reviewed, and several focal points for identifying mesoscale modification potential are presented.

Full access
J. MICHAEL FRITSCH

Abstract

No Abstract Available.

Full access
Peter J. Sousounis and J. Michael Fritsch

A high-resolution numerical model is employed to examine effects of the Great Lakes aggregate, defined to be the five major Great Lakes, on regional and synoptic-scale weather. Simulations wherein the effects of the lakes are included and then excluded are performed on a selected cold air outbreak episode during late autumn when the lakes are still ice-free.

Examination of the differences between the model simulations reveals that several dynamical effects result from heating and moistening by the lake aggregate. These effects are manifested primarily in the form of a 4-km-deep, 2000-km-wide, lake-aggregate mesoscaie disturbance (circulation) that develops slowly over the region. The simulated lake-aggregate circulation splits a synoptic-scale high into two distinct centers and redirects and intensifies a weak synoptic-scale low, as verified by existing observations. These modifications of the synoptic-scale environment result in additional precipitation over, downstream, and upwind from the lakes.

The model simulations also reveal that the developing lake-aggregate circulation influences significantly the lake shore surface winds. In some locations, the surface winds switch from onshore to offshore or vice versa. Because it is well known from observations that the location and orientation of lake-induced snow bands are very sensitive to the low-level wind direction over the lakes, it is concluded that the exact locations of heavy snowfall are the result of a complex multiscale interaction among circulations on three different scales: synoptic, individual lake, and lake aggregate.

In addition to the developing primary lake-aggregate circulation, a secondary dynamic response appears at a distant location, adjacent to the eastern seaboard. The organization of this secondary circulation suggests that the lakes may play a direct role in some cases of East Coast cyclogenesis.

Full access
David J. Stensrud and J. Michael Fritsch

Abstract

A series of five mesoscale convective systems (MCSs) developed within a weakly forced large-scale environment on 11 and 12 May 1982. Two of these systems had a large component of motion against the midtroposphoric flow and propagated in a direction nearly opposite to that of the traveling upper-level disturbances. This description of the evolution of convection is very different from traditional ones in which convection develops and moves more or less in phase with traveling upper-level disturbances. Observations indicate that the initiation and evolution of convection are tied to mesoscale features that are not well observed by the conventional observing network, making the structure of the model initial condition a potentially crucial factor in the success or failure of any subsequent numerical simulation.

It is found that the initial conditions created using the conventional initialization procedure of The Pennsylvania State University-National Center for Atmospheric Research Mesoscale Model do not include several of the mesoscale-sized features observed at 1200 UTC 11 May 1982, 9 h before the development of the first MCS. This is attributed to the lack of observed data with mesoscale resolution, and, therefore, likely is a deficiency in most initialization procedures in use today. Although it is true that new operational observing systems, such as the WSR-88D radar and the 404-MHz radar wind profilers, provide more detailed information, the data density on the mesoscale remains subcritical. A methodology to include mesoscale features, based upon using subjective interpretations of all the available observations, is developed. 11 is found that the mesoscale initial condition created using this subjective approach produces a more reasonable representation of the observed mesoscale features in comparison with the conventionally produced initial condition.

Full access
David J. Stensrud and J. Michael Fritsch

Abstract

During a 24-h period, beginning 1200 UTC 11 May 1982, a series of mesoscale convective systems developed within a weakly forced large-scale environment. Two of these systems had a large component of motion against the midtropospheric flow and propagated in a direction nearly opposite to that of the travelling upper-level disturbances. This evolution of convection is very different from traditional ones in which convection develops and moves more or less in phase with traveling upper-level disturbances. It presents a tremendous challenge for three-dimensional numerical models, since the initiation and evolution of convection are tied to mesoscale features that are not well observed by the conventional upper-air network and may not be well approximated in the model parameterization schemes.

Mesoscale model simulations are conducted to evaluate the ability of the model to reproduce this complex event and to examine the model sensitivities to differences in the convective trigger function and model initial condition. Results suggest deal mesoscale models may be capable of producing useful simulations of convective events associated with weak, large-scale forcing, including quantitative precipitation forecasts with the correct magnitude and approximate location of heavy rainfall. If the important mesoscale circulations are incorporated into the model initial condition and a sufficiently realistic trigger function is used. However, model sensitivities to both the initial condition and the convective trigger function are large. Results indicate that the effects of boundary layer forcing must be included in the trigger function in order to initiate convection at the proper time and location. Timing errors in the initial development of convection of greater than 4 h occur if an unpresentative trigger function is used. Mesoscale features in the model initial condition also play an important role in the development and evolution of convection. The locations of heavy rainfall are shifted by greater than 100 km, or disappear altogether, if particular mesoscale features are not included subjectively in the initial condition. These sensitivities suggest that using an ensemble forecast approach to mesoscale model output needs to be considered seriously as mesoscale models move into the operational forecasting environment.

Full access
Joseph J. Charney and J. Michael Fritsch

Abstract

Surface discrete frontal propagation in a wintertime, nonconvective environment is documented using conventional surface and upper-air data and simulated using the PSU–NCAR mesoscale model.

Synoptic and mesoscale surface analyses show a cold front associated with a synoptic-scale low-pressure system propagating from northwest to southeast across the central United States. Apparently discrete frontal propagation occurs when the surface front dissipates and a new front forms approximately 500 km ahead of the original front, with no compelling evidence of frontal passage in the intervening space. Upper-air analyses indicate the infusion of three different airstreams into the frontal region, resulting in the formation of a ribbon of low static stability air parallel to and several hundred kilometers in advance of the original front. This static stability structure appears to be involved in the observed evolution of the front. The development of precipitation over the intervening zone between the old and new frontal positions suggests that precipitation-induced diabatic processes also played a role in the discrete frontal propagation.

The numerical simulation captures the essential surface, upper-air, and precipitation features associated with the discrete propagation. Cross-section analyses of the simulated atmospheric fields indicate that the front propagated discretely only at the surface and in the lowest 200 hPa of the atmosphere, while the midtropospheric trough associated with the surface front propagated continuously though the region. The cross sections also indicate that the vertical winds associated with the frontal system adjust very quickly to the new frontal location while the horizontal winds and mass fields adjust more slowly. Analysis of frontogenetical forcing verifies that the new surface front forms at the expense of the original front. A careful examination of the temperature budgets within the simulation shows that the mass field redistribution associated with the discrete frontal propagation occurred as a result of the lifting of a strong temperature inversion in the prefrontal environment combined with precipitation induced diabatic cooling.

Based on the results of the model simulation, a conceptual model of discrete frontal propagation is presented that incorporates the observed and simulated sequence of events.

Full access
David J. Stensrud and J. Michael Fritsch

Abstract

During a 24-h period, beginning 1200 UTC 11 May 1982, a series of five mesoscale convective systems (MCSs) developed within a weakly forced large-scale environment. Analyses indicate that the large-scale flow created a broad region of potential buoyant energy, but owing to a restraining inversion and weak large-scale upward motion, convection was initiated only where lifting associated with mesoscale features was able to eliminate the inversion. The series of MCSs developed sequentially and moved eastward across the moist axis. Two of these systems had a large component of motion against the mean tropospheric flow and propagated in a direction nearly opposite to that of the traveling upper-level disturbances. Each system produced an outflow of cold downdraft air that spread progressively farther south than that from the preceding system. This description of the development and evolution of convection is very different from traditional ones wherein convection develops and moves more or less in phase with traveling upper-level disturbances.

Simple analytic models are used to determine the likely mechanisms of upstream propagation. These results suggest that the combined effects of both density currents and internal gravity waves produce the upstream propagation of the region of convection. Density currents dominate the propagation of convection once it forms, while internal gravity waves may help initiate new convection upstream of the region of existing convection, thereby producing a jump in the region of convective activity.

Full access
J. Michael Fritsch and Kenneth F. Heideman

Abstract

The skill of the limited-area fine-mesh (LFM) model in making 24-h areal quantitative precipitation forecasts (QPF) of ⩾ 0.5 in. (12.7 mm) is evaluated and analyzed for two warm seasons; 1982 and 1983. Differences in skill between the eastern and western United States are investigated and the impact of updated initial conditions is explored. The skill in predicting precipitation from cyclonic versus mesoscale systems is also examined.

It is found that the model skill changed drastically from the summers of 1982 to 1983. Skill levels for 1982 were about 25% to 35%. In 1983, they dropped to an average value of less than 10% and often to zero on a daily basis. The precipitous drop in skill appears to be the result of changes made in key model threshold parameters. These parameters include grid-resolvable saturation criterion and convective cloud base boost. The introduction of constraints on convective precipitation and the vertical advection of moisture also appear to have reduced model skill.

A comparison of skill between the eastern and western United States indicates that the model scored substantially better in the east. It is also evident that, in general, the model skill for quantitative precipitation forecasts for the 12-h period 0000 to 1200 UTC was much greater with initial conditions from 0000 UTC, than with initial conditions from 12 h earlier at 1200 UTC; i.e., updated initial conditions had a large positive impact on the model skill. This was particularly true for mesoscale convective systems. Longer term (> 12 h) predictions of the 0000–1200 UTC period scored better for cyclone-related precipitation than for precipitation from mesoscale systems. When the model was reinitialized at 0000 UTC, there was virtually no difference between skill levels for cyclone-related and mesoscale events.

The results indicate that the lack of a realistic diurnal heating cycle in the LFM may be particularly detrimental to the model's ability to forecast convectively dominated warm-season rainfall. The results also indicate that threshold or critical values for initiating resolvable-scale precipitation or making convective adjustments may introduce artificial seasonal, diurnal, and regional variations in quantitative precipitation forecasts. Due to the inherent nonlinearity of numerical models, attempts to improve model skill for a particular region, season, time of day, or precipitation type by adjusting threshold or critical values may introduce bias in other model parameters and negatively impact overall model skill.

Full access
George H. Bryan and J. Michael Fritsch

Abstract

A benchmark solution that facilitates testing the accuracy, efficiency, and efficacy of moist nonhydrostatic numerical model formulations and assumptions is presented. The solution is created from a special configuration of moist model processes and a specific set of initial conditions. The configuration and initial conditions include: reversible phase changes, no hydrometeor fallout, a neutrally stable base-state environment, and an initial buoyancy perturbation that is identical to the one used to test nonlinearly evolving dry thermals. The results of the moist simulation exhibit many of the properties found in its dry counterpart. Given the similar results, and acceptably small total mass and total energy errors, it is argued that this new moist simulation design can be used as a benchmark to evaluate moist numerical model formulations.

The utility of the benchmark simulation is highlighted by running the case with approximate forms of the governing equations found in the literature. Results of these tests have implications for the formulation of numerical models. For example, it is shown that an equation set that conserves both mass and energy is crucial for obtaining the benchmark solution. Results also suggest that the extra effort required to conserve mass in a numerical model may not lead to significant improvements in results unless energy is also conserved.

Full access