Search Results
You are looking at 1 - 10 of 11 items for :
- Author or Editor: D. K. LILLY x
- Article x
- Refine by Access: All Content x
Abstract
A detailed analysis is presented of the large-scale, mesoscale and turbulent-scale features of a major downslope windstorm event in central Colorado on 11 January 1972. The storm is found to be associated with a moderate amplitude baroclinic disturbance moving across the northwestern United States within an intense zonal current. Optimal conditions for strong mountain wave generation are detectable from sounding data 12–24 h in advance and about 1000 km upstream. The mesoscale structure is dominated by a single quasi-hydrostatic wave of extreme amplitude and variable location, with corresponding variations in the windstorm structure.
Severe to extreme aircraft turbulence is observed in a deep boundary layer over the region of strong surface winds and also in a separate mid-tropospheric turbulence zone. Analysis of the latter shows that it originates in a region of intense wave-generated shear and is then carried downstream by the mean flow and upward by the wave motion. Energy generation and dissipation rates of order 1 m2 s−3 are observed. Comparisons of the turbulence features with the theoretical solutions for shearing instability by Tanaka and by Lee and Merkine show fair agreement.
Effects of the wave-windstorm-turbulence event on the larger scales are complex, involving both a substantial removal of westerly momentum and a three-dimensional redistribution of mass.
Hazards to aircraft from this kind of event are illustrated and discussed. Avoidance by vertical path deviation in found to be impractical.
Abstract
A detailed analysis is presented of the large-scale, mesoscale and turbulent-scale features of a major downslope windstorm event in central Colorado on 11 January 1972. The storm is found to be associated with a moderate amplitude baroclinic disturbance moving across the northwestern United States within an intense zonal current. Optimal conditions for strong mountain wave generation are detectable from sounding data 12–24 h in advance and about 1000 km upstream. The mesoscale structure is dominated by a single quasi-hydrostatic wave of extreme amplitude and variable location, with corresponding variations in the windstorm structure.
Severe to extreme aircraft turbulence is observed in a deep boundary layer over the region of strong surface winds and also in a separate mid-tropospheric turbulence zone. Analysis of the latter shows that it originates in a region of intense wave-generated shear and is then carried downstream by the mean flow and upward by the wave motion. Energy generation and dissipation rates of order 1 m2 s−3 are observed. Comparisons of the turbulence features with the theoretical solutions for shearing instability by Tanaka and by Lee and Merkine show fair agreement.
Effects of the wave-windstorm-turbulence event on the larger scales are complex, involving both a substantial removal of westerly momentum and a three-dimensional redistribution of mass.
Hazards to aircraft from this kind of event are illustrated and discussed. Avoidance by vertical path deviation in found to be impractical.
Abstract
An analysis is made of Gage's proposal that the horizontal energy spectrum at mesoscale wavelengths is produced by upscale energy transfer through quasi-two-dimensional turbulence. It is suggested that principal sources of such energy can be found in decaying convective clouds and thunderstorm anvil outflows. These are believed to evolve similarly to the wake of a moving body in a stably stratified flow. Following the scale analysis by Riley, Metcalfe and Weissman it is expected that, in the presence of strong stratification, initially three-dimensionally isotropic turbulence divides roughly equally into gravity waves and stratified (quasi-two- dimensional) turbulence. The former then propagates away from the generation region, while the latter propagates in spectral space to larger scales, forming the −5/3 upscale transfer spectrum predicted by Kraichnan. Part of the energy of the stratified turbulence is recycled into three-dimensional turbulence by shearing instability, but the upscale escape of only a few percent of the total energy released by small-scale turbulence is apparently sufficient to explain the observed mesoscale energy spectrum of the troposphere. A close analogy is found between the turbulence-gravity wave exchanges considered here and the turbulence-β-wave exchanges discussed by Rhines and Williams.
Abstract
An analysis is made of Gage's proposal that the horizontal energy spectrum at mesoscale wavelengths is produced by upscale energy transfer through quasi-two-dimensional turbulence. It is suggested that principal sources of such energy can be found in decaying convective clouds and thunderstorm anvil outflows. These are believed to evolve similarly to the wake of a moving body in a stably stratified flow. Following the scale analysis by Riley, Metcalfe and Weissman it is expected that, in the presence of strong stratification, initially three-dimensionally isotropic turbulence divides roughly equally into gravity waves and stratified (quasi-two- dimensional) turbulence. The former then propagates away from the generation region, while the latter propagates in spectral space to larger scales, forming the −5/3 upscale transfer spectrum predicted by Kraichnan. Part of the energy of the stratified turbulence is recycled into three-dimensional turbulence by shearing instability, but the upscale escape of only a few percent of the total energy released by small-scale turbulence is apparently sufficient to explain the observed mesoscale energy spectrum of the troposphere. A close analogy is found between the turbulence-gravity wave exchanges considered here and the turbulence-β-wave exchanges discussed by Rhines and Williams.
Abstract
A system is proposed for grid allocation and differencing of apparently general applicability to purely marching-type systems of equations of fluid dynamics. The method is based on casting of the equations into the conservation form, which then permits use of a staggered space-time grid system with interpolations required only in certain linear terms. The method is illustrated by application to two systems of equations, on one of which numerical experiments have been successfully performed. Advantages and drawbacks of the method are described in comparison to other currently used grid systems, and the possibility and desirability of parametric simulation of turbulent eddy exchange processes are discussed.
Abstract
A system is proposed for grid allocation and differencing of apparently general applicability to purely marching-type systems of equations of fluid dynamics. The method is based on casting of the equations into the conservation form, which then permits use of a staggered space-time grid system with interpolations required only in certain linear terms. The method is illustrated by application to two systems of equations, on one of which numerical experiments have been successfully performed. Advantages and drawbacks of the method are described in comparison to other currently used grid systems, and the possibility and desirability of parametric simulation of turbulent eddy exchange processes are discussed.
Abstract
Detailed stratospheric wind and temperature data were gathered by aircraft over the mountains of southern Colorado on 1 March 1970. In a unique operation, two instrumented RB-57F aircraft flew a total of twelve upwind and downwind legs at altitudes of 13 to 20km,.with an average separation of 600 m.
The observational data show that the stratosphere was disturbed by sporadically-turbulent gravity waves of length 20–30 km, which were apparently generated directly or indirectly by the mountainous terrain. Wave amplitudes and turbulence frequency showed a general increase with height up to about 17 km, where they reached a maximum. The maximum amplitude layer was also characterized by a minimum in the mean wind speed.
The horizontal and vertical velocity and temperature variances and covariances were evaluated and found to be generally consistent with predictions of linear gravity wave theory. The spectra of horizontal kinetic and potential energy were also calculated and appear to follow a −3 law over about a decade in wavenumber space. The covariance of the horizontal and vertical velocity perturbations is everywhere negative, with a mean downward momentum flux of about ∼2 dyn cm−2. The correlation spectra show that most of this momentum flux is contributed by the long waves.
Abstract
Detailed stratospheric wind and temperature data were gathered by aircraft over the mountains of southern Colorado on 1 March 1970. In a unique operation, two instrumented RB-57F aircraft flew a total of twelve upwind and downwind legs at altitudes of 13 to 20km,.with an average separation of 600 m.
The observational data show that the stratosphere was disturbed by sporadically-turbulent gravity waves of length 20–30 km, which were apparently generated directly or indirectly by the mountainous terrain. Wave amplitudes and turbulence frequency showed a general increase with height up to about 17 km, where they reached a maximum. The maximum amplitude layer was also characterized by a minimum in the mean wind speed.
The horizontal and vertical velocity and temperature variances and covariances were evaluated and found to be generally consistent with predictions of linear gravity wave theory. The spectra of horizontal kinetic and potential energy were also calculated and appear to follow a −3 law over about a decade in wavenumber space. The covariance of the horizontal and vertical velocity perturbations is everywhere negative, with a mean downward momentum flux of about ∼2 dyn cm−2. The correlation spectra show that most of this momentum flux is contributed by the long waves.
Abstract
A numerical model is developed for simulating the flow of stably stratified nonrotating air over finite-amplitude, two-dimensional mountain ranges. Special attention is paid to accurate treatment of internal dissipation and to formulation of an upper boundary region and lateral boundary conditions which allow upward and lateral propagation of wave energy out of the model. The model is hydrostatic and uses potential temperature for the vertical coordinate. A local adjustment procedure is derived to parameterize low Richardson number instability. The model behavior is tested against analytic theory and then applied to a variety of idealized and real flow situations, leading to some new insights and new questions on the nature of large-amplitude mountain waves. The model proves to be effective in simulating the structure of two observed cases of strong mountain waves with very different characteristics.
Abstract
A numerical model is developed for simulating the flow of stably stratified nonrotating air over finite-amplitude, two-dimensional mountain ranges. Special attention is paid to accurate treatment of internal dissipation and to formulation of an upper boundary region and lateral boundary conditions which allow upward and lateral propagation of wave energy out of the model. The model is hydrostatic and uses potential temperature for the vertical coordinate. A local adjustment procedure is derived to parameterize low Richardson number instability. The model behavior is tested against analytic theory and then applied to a variety of idealized and real flow situations, leading to some new insights and new questions on the nature of large-amplitude mountain waves. The model proves to be effective in simulating the structure of two observed cases of strong mountain waves with very different characteristics.
Abstract
The lee flow disturbances produced by the Front Range of the Colorado Rockies have been quantitatively observed in a continuing program at the National Center for Atmospheric Research. The results of midtroposphere constant-volume ballon and aircraft flights in the winter of 1966/67 are here presented. The relative merits and limitations of the two methods are compared with respect to various operational and inherent phenomenological difficulties of the subject. The nonstationarity of many flow features is inescapable and poses serious problems for data evaluation and theory. Schematically, we distinguish between smooth, wavy, and hydraulic jump-type flow patterns, but also observe some cases that do not fit well into any of these categories. The stronger stationary wave features can be compared with the “stable” resonance modes computed from stationary linear theory, that is, those modes which are insensitive to small changes in the upstream flow. The frequent occurrence of erratic and nonstationary flows may relate to the frequent existence of “unstable” or sensitive modes in the linear theory predictions. Examples of smooth and hydraulic jumplike flows are also shown and qualitatively compared to current theoretical predictions. Some suggestions are made for improvement of observational techniques in the downslope boundary layer.
Abstract
The lee flow disturbances produced by the Front Range of the Colorado Rockies have been quantitatively observed in a continuing program at the National Center for Atmospheric Research. The results of midtroposphere constant-volume ballon and aircraft flights in the winter of 1966/67 are here presented. The relative merits and limitations of the two methods are compared with respect to various operational and inherent phenomenological difficulties of the subject. The nonstationarity of many flow features is inescapable and poses serious problems for data evaluation and theory. Schematically, we distinguish between smooth, wavy, and hydraulic jump-type flow patterns, but also observe some cases that do not fit well into any of these categories. The stronger stationary wave features can be compared with the “stable” resonance modes computed from stationary linear theory, that is, those modes which are insensitive to small changes in the upstream flow. The frequent occurrence of erratic and nonstationary flows may relate to the frequent existence of “unstable” or sensitive modes in the linear theory predictions. Examples of smooth and hydraulic jumplike flows are also shown and qualitatively compared to current theoretical predictions. Some suggestions are made for improvement of observational techniques in the downslope boundary layer.
Abstract
Analysis is presented of data obtained from instrumented aircraft flying in a mountain wave of moderate amplitude west of Denver, Colo., on 17 February 1970. Emphasis is placed on determination of the downward flux of westerly momentum generated by the wave, for which accurate measurements of vertical velocities on scales of order 50 km are essential. Three different methods are applied and compared: direct aircraft measurement, using vanes and an inertial platform; evaluation from the steady-state equation for conservation of potential temperature; and integration of the steady-state continuity equation. Each method produces errors, but by combining the results of the three methods a profile is obtained which agrees. fairly well with a steady-state theoretical prediction. An important side result is the discovery that gust-probe equipment is apparently not necessary for the direct aircraft measurement of wave momentum flux, but an inertial platform or similarly stable attitude reference level is essential.
A region of severe turbulence at 100 mb is found to he associated with the source of most of the downward wave momentum flux. Measurements of the loss of total energy along isentropes are found to he consistent with kinetic energy losses estimated from momentum flux divergence and with energy dissipation estimated from inertial-range aircraft measurements of the turbulent energy spectrum.
Abstract
Analysis is presented of data obtained from instrumented aircraft flying in a mountain wave of moderate amplitude west of Denver, Colo., on 17 February 1970. Emphasis is placed on determination of the downward flux of westerly momentum generated by the wave, for which accurate measurements of vertical velocities on scales of order 50 km are essential. Three different methods are applied and compared: direct aircraft measurement, using vanes and an inertial platform; evaluation from the steady-state equation for conservation of potential temperature; and integration of the steady-state continuity equation. Each method produces errors, but by combining the results of the three methods a profile is obtained which agrees. fairly well with a steady-state theoretical prediction. An important side result is the discovery that gust-probe equipment is apparently not necessary for the direct aircraft measurement of wave momentum flux, but an inertial platform or similarly stable attitude reference level is essential.
A region of severe turbulence at 100 mb is found to he associated with the source of most of the downward wave momentum flux. Measurements of the loss of total energy along isentropes are found to he consistent with kinetic energy losses estimated from momentum flux divergence and with energy dissipation estimated from inertial-range aircraft measurements of the turbulent energy spectrum.
Abstract
Vertical diffusion coefficients in the stratosphere are estimated from data obtained in the High Altitude Clear Air Turbulence (HICAT) investigation. The HICAT data sample was obtained from 285 flights of over 800,000 km distance, containing 24,000 flight kilometers of turbulence between 14 and 21 km MSL, and is the only such collection of fine-scale, true gust velocities in the stratosphere.
Our calculations indicate that small-scale diffusion coefficients vary from order 4×103 cm2 sec−1 over ocean regions to order 105 cm2 sec−1 over high mountains when averaged over periods of perceptible turbulence, which include from 2.0 to 5.2%, respectively, of the total flight distance. The values decrease rapidly above 17 km except over mountainous terrain, where mixing appears to be pronounced up to and above 19 km in winter. The overall mean value over North America and Greenland is 190 cm2 sec−1, and the global mean is 120 cm2 sec−1, with an uncertainty of about half an order of magnitude.
Abstract
Vertical diffusion coefficients in the stratosphere are estimated from data obtained in the High Altitude Clear Air Turbulence (HICAT) investigation. The HICAT data sample was obtained from 285 flights of over 800,000 km distance, containing 24,000 flight kilometers of turbulence between 14 and 21 km MSL, and is the only such collection of fine-scale, true gust velocities in the stratosphere.
Our calculations indicate that small-scale diffusion coefficients vary from order 4×103 cm2 sec−1 over ocean regions to order 105 cm2 sec−1 over high mountains when averaged over periods of perceptible turbulence, which include from 2.0 to 5.2%, respectively, of the total flight distance. The values decrease rapidly above 17 km except over mountainous terrain, where mixing appears to be pronounced up to and above 19 km in winter. The overall mean value over North America and Greenland is 190 cm2 sec−1, and the global mean is 120 cm2 sec−1, with an uncertainty of about half an order of magnitude.
Abstract
A new large eddy simulation (LES) stratocumulus cloud model with an explicit formulation of micro-physical processes has been developed, and the results from three large eddy simulations are presented to illustrate the effects of the stratocumulus-topped boundary layer (STBL) dynamics on cloud microphysical parameters. The simulations represent cases of a well-mixed and a radiatively driven STBL. Two of the simulations differ only in the ambient aerosol concentration and show its effect on cloud microphysics. The third simulation is based on the data obtained by Nicholls, and the simulation results from this case are contrasted with his measurements. Cloud-layer dynamical parameters and cloud droplet spectra are in reasonably good agreement with observations.
As demonstrated by the results of three large eddy simulations presented in the paper, the cloud microphysical parameters are significantly affected by cloud dynamics. This is evidenced by the sensitivity of the cloud drop spectra itself, as well as by that of the integral parameters of the spectra, such as mean radii and droplet concentration. Experiments presented here also show that cloud microstructure is significantly asymmetric between updrafts and downdrafts. Mixing with dry air from the inversion may significantly enhance evaporation and result in cloud-free zones within the cloud. As a result of mixing, the cloud layer is very inhomogeneous, especially near its top and bottom.
The authors analyze in detail the fine structure of the supersaturation field and suggest an explanation for the formation of the model-predicted supersaturation peak near the cloud top. The LES results suggest that super-saturation may have a sharp increase in near-saturated parcels that undergo forced vertical displacement at the cloud-layer top. The main forcing mechanism that may supply the additional energy for the forced convection in the case presented is from propagating gravity waves. Although radiative cooling may also result in increased convective activity at cloud top, the sensitivity tests presented here suggest that, at least in these simulations, this effect is not dominant.
Abstract
A new large eddy simulation (LES) stratocumulus cloud model with an explicit formulation of micro-physical processes has been developed, and the results from three large eddy simulations are presented to illustrate the effects of the stratocumulus-topped boundary layer (STBL) dynamics on cloud microphysical parameters. The simulations represent cases of a well-mixed and a radiatively driven STBL. Two of the simulations differ only in the ambient aerosol concentration and show its effect on cloud microphysics. The third simulation is based on the data obtained by Nicholls, and the simulation results from this case are contrasted with his measurements. Cloud-layer dynamical parameters and cloud droplet spectra are in reasonably good agreement with observations.
As demonstrated by the results of three large eddy simulations presented in the paper, the cloud microphysical parameters are significantly affected by cloud dynamics. This is evidenced by the sensitivity of the cloud drop spectra itself, as well as by that of the integral parameters of the spectra, such as mean radii and droplet concentration. Experiments presented here also show that cloud microstructure is significantly asymmetric between updrafts and downdrafts. Mixing with dry air from the inversion may significantly enhance evaporation and result in cloud-free zones within the cloud. As a result of mixing, the cloud layer is very inhomogeneous, especially near its top and bottom.
The authors analyze in detail the fine structure of the supersaturation field and suggest an explanation for the formation of the model-predicted supersaturation peak near the cloud top. The LES results suggest that super-saturation may have a sharp increase in near-saturated parcels that undergo forced vertical displacement at the cloud-layer top. The main forcing mechanism that may supply the additional energy for the forced convection in the case presented is from propagating gravity waves. Although radiative cooling may also result in increased convective activity at cloud top, the sensitivity tests presented here suggest that, at least in these simulations, this effect is not dominant.
From 30 September to 2 October 1999 a workshop was held in Gif-sur-Yvette, France, with the central objective to develop a research strategy for the next 3–5 years, aiming at a systematic description of root functioning, rooting depth, and root distribution for modeling root water uptake from local and regional to global scales. The goal was to link more closely the weather prediction and climate and hydrological models with ecological and plant physiological information in order to improve the understanding of the impact that root functioning has on the hydrological cycle at various scales. The major outcome of the workshop was a number of recommendations, detailed at the end of this paper, on root water uptake parameterization and modeling and on collection of root and soil hydraulic data.
From 30 September to 2 October 1999 a workshop was held in Gif-sur-Yvette, France, with the central objective to develop a research strategy for the next 3–5 years, aiming at a systematic description of root functioning, rooting depth, and root distribution for modeling root water uptake from local and regional to global scales. The goal was to link more closely the weather prediction and climate and hydrological models with ecological and plant physiological information in order to improve the understanding of the impact that root functioning has on the hydrological cycle at various scales. The major outcome of the workshop was a number of recommendations, detailed at the end of this paper, on root water uptake parameterization and modeling and on collection of root and soil hydraulic data.
