Search Results
You are looking at 11 - 20 of 21 items for
- Author or Editor: B. Stankov x
- Refine by Access: All Content x
Abstract
Data from the Boulder Atmospheric Observatory (BAO) are used to investigate the wave and turbulence structure of the convective atmospheric mixed layer and the overlying inversion. Three cases are discussed, one in considerable detail, in which the depth of the mixed layer is below the top of the 300 m tower at the BAO and is nearly steady state for several hours. Velocity and temperature variances and spectra, coherences between vertical velocity and temperature, and vertical velocities at different levels on the tower are used to show that although the mixed-layer behavior is for the most part similar to that found in previous studies, there are some significant differences due mainly to the relatively large shear term in the turbulence energy equation compared with buoyancy, both within the mixed layer and in the capping inversion. For example, the wavelength of the spectral maximum for vertical velocity in the upper half of the mixed layer is about three times the boundary-layer height, which is about twice that estimated in a previous experiment. The wavelength is up to 5.5 times the mixed-layer height above the top of the mixed layer. Within the mixed layer, terms in the turbulence kinetic energy equation are similar to previous studies. Above the mixed layer, shear production becomes large, and is approximately balanced by the sum of the buoyancy, dissipation and transport terms. The temperature variance and flux budgets also have large terms and significant residuals in the overlying inversion.
Abstract
Data from the Boulder Atmospheric Observatory (BAO) are used to investigate the wave and turbulence structure of the convective atmospheric mixed layer and the overlying inversion. Three cases are discussed, one in considerable detail, in which the depth of the mixed layer is below the top of the 300 m tower at the BAO and is nearly steady state for several hours. Velocity and temperature variances and spectra, coherences between vertical velocity and temperature, and vertical velocities at different levels on the tower are used to show that although the mixed-layer behavior is for the most part similar to that found in previous studies, there are some significant differences due mainly to the relatively large shear term in the turbulence energy equation compared with buoyancy, both within the mixed layer and in the capping inversion. For example, the wavelength of the spectral maximum for vertical velocity in the upper half of the mixed layer is about three times the boundary-layer height, which is about twice that estimated in a previous experiment. The wavelength is up to 5.5 times the mixed-layer height above the top of the mixed layer. Within the mixed layer, terms in the turbulence kinetic energy equation are similar to previous studies. Above the mixed layer, shear production becomes large, and is approximately balanced by the sum of the buoyancy, dissipation and transport terms. The temperature variance and flux budgets also have large terms and significant residuals in the overlying inversion.
Abstract
On 2 July 1987 a nonmesocyclone tornado was observed in northeastern Colorado during the Convection Initiation and Downburst Experiment (CINDE). This tornado, reaching FI–F2 intensity, developed under a rapidly growing convective cell, without a preceding supercell or midlevel mesocyclone being present.
The pretornado environment on 2 July is described, including observations from a triangle of wind profilers, a dense surface mesonet array, and a special balloon sounding network. Important features contributing to tornado generation include the passage of a 700-mb short-wave trough; the formation of an ∼70-km diameter, terrain-induced mesoscale vortex (the Denver Cyclone) and its associated baroclinic zone; the presence of a stationary low-level convergence boundary; and the presence of low-level azimuthal sheer maxima (misovortices) along the boundary.
Vorticity budget terms are calculated in the lowest 2 km AGL using a multiple-Doppler radar analysis. These terms and their spatial distributions are compared with observations of mesocyclone-associated supercell tornadoes. Results show that vorticity associated with the 2 July nonsupercell tornado was generated in a more complicated manner than that proposed by previous nonsupercell tornadogenesis theory. In particular, tilting of baroclinically generated streamwise horizontal vorticity into the vertical was important for the formation of low-level rotation, in a manner similar to that previously proposed for supercell tornadic storms.
Abstract
On 2 July 1987 a nonmesocyclone tornado was observed in northeastern Colorado during the Convection Initiation and Downburst Experiment (CINDE). This tornado, reaching FI–F2 intensity, developed under a rapidly growing convective cell, without a preceding supercell or midlevel mesocyclone being present.
The pretornado environment on 2 July is described, including observations from a triangle of wind profilers, a dense surface mesonet array, and a special balloon sounding network. Important features contributing to tornado generation include the passage of a 700-mb short-wave trough; the formation of an ∼70-km diameter, terrain-induced mesoscale vortex (the Denver Cyclone) and its associated baroclinic zone; the presence of a stationary low-level convergence boundary; and the presence of low-level azimuthal sheer maxima (misovortices) along the boundary.
Vorticity budget terms are calculated in the lowest 2 km AGL using a multiple-Doppler radar analysis. These terms and their spatial distributions are compared with observations of mesocyclone-associated supercell tornadoes. Results show that vorticity associated with the 2 July nonsupercell tornado was generated in a more complicated manner than that proposed by previous nonsupercell tornadogenesis theory. In particular, tilting of baroclinically generated streamwise horizontal vorticity into the vertical was important for the formation of low-level rotation, in a manner similar to that previously proposed for supercell tornadic storms.
Abstract
Remote soundings of precipitable water vapor from three systems are compared with each other and with ground truth from radiosondes. Ancillary data from a mesoscale network of surface observing stations and from wind-profiling radars are also used in the analysis. The three remote-sounding techniques are: (a) a reflectance technique using spectral data collected by the Airborne Visible-Infrared Imaging Spectrometer (AVIRIS); (b) an emission technique using Visible-Infrared Spin Scan Radiometer (VISSR) Atmospheric Sounder (VAS) data acquired from the National Oceanic and Atmospheric Administration's (NOAA) Geostationary Operational Environmental Satellite (GOES); and (c) a microwave technique using data from a limited network of three ground-based dual-channel microwave radiometers. The data were taken over the Front Range of eastern Colorado on 22–23 March 1990. The generally small differences between the three types of rernote-sounding measurements are consistent with the horizontal and temporal resolutions of the instruments. The microwave and optical reflectance measurements agreed to within 0. 1 cm; comparisons of the microwave data with radiosondes were also either that good or explainable. The largest differences between the VAS and the microwave radiometer at Elbert were between 0.4 and 0.5 cm and appear to he due to variable terrain within the satellite footprint.
Abstract
Remote soundings of precipitable water vapor from three systems are compared with each other and with ground truth from radiosondes. Ancillary data from a mesoscale network of surface observing stations and from wind-profiling radars are also used in the analysis. The three remote-sounding techniques are: (a) a reflectance technique using spectral data collected by the Airborne Visible-Infrared Imaging Spectrometer (AVIRIS); (b) an emission technique using Visible-Infrared Spin Scan Radiometer (VISSR) Atmospheric Sounder (VAS) data acquired from the National Oceanic and Atmospheric Administration's (NOAA) Geostationary Operational Environmental Satellite (GOES); and (c) a microwave technique using data from a limited network of three ground-based dual-channel microwave radiometers. The data were taken over the Front Range of eastern Colorado on 22–23 March 1990. The generally small differences between the three types of rernote-sounding measurements are consistent with the horizontal and temporal resolutions of the instruments. The microwave and optical reflectance measurements agreed to within 0. 1 cm; comparisons of the microwave data with radiosondes were also either that good or explainable. The largest differences between the VAS and the microwave radiometer at Elbert were between 0.4 and 0.5 cm and appear to he due to variable terrain within the satellite footprint.
Abstract
A numerical technique is described for synthesizing realistic atmospheric temperature and humidity profiles. The method uses an ensemble of radiosonde measurements collected at a site of interest. Erroneous profiles are removed by comparing their likelihood with prevailing meteorological conditions. The remaining profiles are decomposed using the method of empirical orthogonal functions. The corresponding eigenprofiles and the statistics of the expansion coefficients are used to numerically generate synthetic profiles that obey the same statistics (i.e., have the same mean, variability, and vertical correlation) as the initial dataset. The technique was applied to a set of approximately 1000 temperature and humidity soundings made in Denver, Colorado, during the winter months of 1991–95. This dataset was divided into four cloud classification categories and daytime and nighttime launches to better characterize typical profiles for the eight cases considered. It was found that 97% of the variance in the soundings could be accounted for by using only five eigenprofiles in the reconstructions. Ensembles of numerically generated profiles can be used to test the accuracy of various retrieval algorithms under controlled conditions not usually available in practice.
Abstract
A numerical technique is described for synthesizing realistic atmospheric temperature and humidity profiles. The method uses an ensemble of radiosonde measurements collected at a site of interest. Erroneous profiles are removed by comparing their likelihood with prevailing meteorological conditions. The remaining profiles are decomposed using the method of empirical orthogonal functions. The corresponding eigenprofiles and the statistics of the expansion coefficients are used to numerically generate synthetic profiles that obey the same statistics (i.e., have the same mean, variability, and vertical correlation) as the initial dataset. The technique was applied to a set of approximately 1000 temperature and humidity soundings made in Denver, Colorado, during the winter months of 1991–95. This dataset was divided into four cloud classification categories and daytime and nighttime launches to better characterize typical profiles for the eight cases considered. It was found that 97% of the variance in the soundings could be accounted for by using only five eigenprofiles in the reconstructions. Ensembles of numerically generated profiles can be used to test the accuracy of various retrieval algorithms under controlled conditions not usually available in practice.
Several ground-based remote sensors were operated together in Colorado during February and March 1991 to obtain continuous profiles of the kinematic and thermodynamic structure of the atmosphere. Instrument performance is compared for five different wind profilers. Each was equipped with Radio Acoustic Sounding System (RASS) capability to measure virtual temperature. This was the first side-by-side comparison of all three of the most common wind-profiler frequencies: 50, 404, and 915 MHz. The 404-MHz system was a NOAA Wind Profiler Demonstration Network (WPDN) unit. Dual-frequency microwave radiometers that measured path-integrated water vapor and liquid water content were also evaluated. Frequent rawinsonde launches from the remote-sensor sites provided an extensive set of in situ measurements for comparison. The winter operations provide a severe test of the profiler/RASS capabilities because atmospheric scattering is relatively weak and acoustic attenuation is relatively strong in cold, dry conditions. Nevertheless, the lower-frequency systems exhibited impressive height coverage for wind and virtual temperature profiling, whereas the high-frequency units provided higher-resolution measurements near the surface. Comparisons between remote sensor and rawinsonde data generally showed excellent agreement. The results support more widespread use of these emerging technologies.
Several ground-based remote sensors were operated together in Colorado during February and March 1991 to obtain continuous profiles of the kinematic and thermodynamic structure of the atmosphere. Instrument performance is compared for five different wind profilers. Each was equipped with Radio Acoustic Sounding System (RASS) capability to measure virtual temperature. This was the first side-by-side comparison of all three of the most common wind-profiler frequencies: 50, 404, and 915 MHz. The 404-MHz system was a NOAA Wind Profiler Demonstration Network (WPDN) unit. Dual-frequency microwave radiometers that measured path-integrated water vapor and liquid water content were also evaluated. Frequent rawinsonde launches from the remote-sensor sites provided an extensive set of in situ measurements for comparison. The winter operations provide a severe test of the profiler/RASS capabilities because atmospheric scattering is relatively weak and acoustic attenuation is relatively strong in cold, dry conditions. Nevertheless, the lower-frequency systems exhibited impressive height coverage for wind and virtual temperature profiling, whereas the high-frequency units provided higher-resolution measurements near the surface. Comparisons between remote sensor and rawinsonde data generally showed excellent agreement. The results support more widespread use of these emerging technologies.
Abstract
The paper describes a convective boundary layer experiment conducted in April 1978 at the Boulder Atmospheric Observatory, and examines the spectral behavior of wind velocity and temperature from the Observatory's 300 m tower, from aircraft flights alongside the tower and from a surface network of anemometers, for evidence of terrain influence on turbulence structure. The gently rolling terrain at the site does not seem to affect the turbulence spectra from the tower in any perceptible manner, except for minor shifts in the vertical velocity and temperature spectral peaks. The aircraft vertical velocity spectra showed different shapes for alongwind and crosswind sampling directions, as in earlier measurements over ocean surfaces, and their peaks are displaced to higher wavenumbers compared with the tower spectra. Long-term spectra of horizontal wind components from surface stations around the tower exhibit no particular sensitivity to site selection. Under near-stationary conditions the peak of the spectrum of the streamwise component tends to reflect more closely the predominant boundary layer. convective scales than does the peak of the lateral wind component. The problem of identifying those scales in the presence of large shifts in wind direction is discussed.
Abstract
The paper describes a convective boundary layer experiment conducted in April 1978 at the Boulder Atmospheric Observatory, and examines the spectral behavior of wind velocity and temperature from the Observatory's 300 m tower, from aircraft flights alongside the tower and from a surface network of anemometers, for evidence of terrain influence on turbulence structure. The gently rolling terrain at the site does not seem to affect the turbulence spectra from the tower in any perceptible manner, except for minor shifts in the vertical velocity and temperature spectral peaks. The aircraft vertical velocity spectra showed different shapes for alongwind and crosswind sampling directions, as in earlier measurements over ocean surfaces, and their peaks are displaced to higher wavenumbers compared with the tower spectra. Long-term spectra of horizontal wind components from surface stations around the tower exhibit no particular sensitivity to site selection. Under near-stationary conditions the peak of the spectrum of the streamwise component tends to reflect more closely the predominant boundary layer. convective scales than does the peak of the lateral wind component. The problem of identifying those scales in the presence of large shifts in wind direction is discussed.
Abstract
An algorithm to compute the magnitude of humidity gradient profiles from the measurements of the zeroth, first, and second moments of wind profiling radar (WPR) Doppler spectra was developed and tested. The algorithm extends the National Oceanic and Atmospheric Administration (NOAA)/Environmental Technology Laboratory (ETL) Advanced Signal Processing System (SPS), which provides quality control of radar data in the spectral domain for wind profile retrievals, to include the retrieval of humidity gradient profiles. The algorithm uses a recently developed approximate formula for correcting Doppler spectral widths for the spatial and temporal filtering effects. Data collected by a 3-beam 915-MHz WPR onboard the NOAA research vessel Ronald H. Brown (RHB) and a 5-beam 449-MHz WPR developed at the ETL were used in this study. The two datasets cover vastly different atmospheric conditions, with the 915-MHz shipborne system probing the tropical ocean atmosphere and the 449-MHz WPR probing continental winter upslope icing storm in the Colorado Front Range. Vaisala radiosonde measurements of humidity and temperature profiles on board the RHB and the standard National Weather Service (NWS) radiosonde measurements at Stapleton, Colorado, were used for comparisons. The cases chosen represent typical atmospheric conditions and not special atmospheric situations. Results show that using SPS-obtained measurements of the zeroth, first, and second spectral moments provide radar-obtained humidity gradient profiles up to 3 km AGL.
Abstract
An algorithm to compute the magnitude of humidity gradient profiles from the measurements of the zeroth, first, and second moments of wind profiling radar (WPR) Doppler spectra was developed and tested. The algorithm extends the National Oceanic and Atmospheric Administration (NOAA)/Environmental Technology Laboratory (ETL) Advanced Signal Processing System (SPS), which provides quality control of radar data in the spectral domain for wind profile retrievals, to include the retrieval of humidity gradient profiles. The algorithm uses a recently developed approximate formula for correcting Doppler spectral widths for the spatial and temporal filtering effects. Data collected by a 3-beam 915-MHz WPR onboard the NOAA research vessel Ronald H. Brown (RHB) and a 5-beam 449-MHz WPR developed at the ETL were used in this study. The two datasets cover vastly different atmospheric conditions, with the 915-MHz shipborne system probing the tropical ocean atmosphere and the 449-MHz WPR probing continental winter upslope icing storm in the Colorado Front Range. Vaisala radiosonde measurements of humidity and temperature profiles on board the RHB and the standard National Weather Service (NWS) radiosonde measurements at Stapleton, Colorado, were used for comparisons. The cases chosen represent typical atmospheric conditions and not special atmospheric situations. Results show that using SPS-obtained measurements of the zeroth, first, and second spectral moments provide radar-obtained humidity gradient profiles up to 3 km AGL.
Abstract
The NOAA/WPL pulsed coherent Doppler lidar was used during the Texas Frontal Experiment in 1985 to study mesoscale preconvective atmospheric conditions. On 22 April 1985, the Doppler lidar, in conjunction with serial rawinsonde ascents and National Weather Service rawinsonde ascents, observed atmospheric features such as middle-tropospheric frontal and vertical wind shear layers and the planetary boundary layer. The lidar showed unique evidence of the downward transport of strong winds from an elevated vertical speed shear (frontal) layer into the planetary boundary layer. The lidar provided further evidence of atmospheric processes such as clear-air turbulence within frontal layers, and dry convection turbulence within the superadiabatic planetary boundary layer. As a result, high-technology remote sensing instruments such as the Doppler lidar show considerable promise for future studies of small-scale weather systems in a nonprecipitating atmosphere.
Abstract
The NOAA/WPL pulsed coherent Doppler lidar was used during the Texas Frontal Experiment in 1985 to study mesoscale preconvective atmospheric conditions. On 22 April 1985, the Doppler lidar, in conjunction with serial rawinsonde ascents and National Weather Service rawinsonde ascents, observed atmospheric features such as middle-tropospheric frontal and vertical wind shear layers and the planetary boundary layer. The lidar showed unique evidence of the downward transport of strong winds from an elevated vertical speed shear (frontal) layer into the planetary boundary layer. The lidar provided further evidence of atmospheric processes such as clear-air turbulence within frontal layers, and dry convection turbulence within the superadiabatic planetary boundary layer. As a result, high-technology remote sensing instruments such as the Doppler lidar show considerable promise for future studies of small-scale weather systems in a nonprecipitating atmosphere.
Abstract
During June–July 1999, the NOAA R/V Ron H. Brown (RHB) sailed from Australia to the Republic of Nauru where the Department of Energy's Atmospheric Radiation Measurement (ARM) Program operates a long-term climate observing station. During July, when the RHB was in close proximity to the island of Nauru, detailed comparisons of ship- and island-based instruments were possible. Essentially identical instruments were operated from the ship and the island's Atmospheric Radiation and Cloud Station (ARCS)-2. These instruments included simultaneously launched Vaisala RS80-H radiosondes, the Environmental Technology Laboratory's (ETL) Fourier transform infrared radiometer (FTIR), and ARM's atmospheric emitted radiance interferometer (AERI), as well as cloud radars/ceilometers to identify clear conditions.
The ARM microwave radiometer (MWR) operating on Nauru provided another excellent dataset for the entire Nauru99 experiment. The calibration accuracy was verified by a liquid nitrogen blackbody target experiment and by consistent high quality tipping calibrations throughout the experiment. Comparisons were made for calculated clear-sky brightness temperature (T b ) and for precipitable water vapor (PWV). These results indicate that substantial errors, sometimes of the order of 20% in PWV, occurred with the original radiosondes. When a Vaisala correction algorithm was applied, calculated T b s were in better agreement with the MWR than were the calculations based on the original data. However, the improvement in T b comparisons was noticeably different for different radiosonde lots and was not a monotonic function of radiosonde age. Three different absorption algorithms were compared: Liebe and Layton, Liebe et al., and Rosenkranz. Using AERI spectral radiance observations as a comparison standard, scaling of radiosondes by MWR data was compared with both original and corrected soundings.
Abstract
During June–July 1999, the NOAA R/V Ron H. Brown (RHB) sailed from Australia to the Republic of Nauru where the Department of Energy's Atmospheric Radiation Measurement (ARM) Program operates a long-term climate observing station. During July, when the RHB was in close proximity to the island of Nauru, detailed comparisons of ship- and island-based instruments were possible. Essentially identical instruments were operated from the ship and the island's Atmospheric Radiation and Cloud Station (ARCS)-2. These instruments included simultaneously launched Vaisala RS80-H radiosondes, the Environmental Technology Laboratory's (ETL) Fourier transform infrared radiometer (FTIR), and ARM's atmospheric emitted radiance interferometer (AERI), as well as cloud radars/ceilometers to identify clear conditions.
The ARM microwave radiometer (MWR) operating on Nauru provided another excellent dataset for the entire Nauru99 experiment. The calibration accuracy was verified by a liquid nitrogen blackbody target experiment and by consistent high quality tipping calibrations throughout the experiment. Comparisons were made for calculated clear-sky brightness temperature (T b ) and for precipitable water vapor (PWV). These results indicate that substantial errors, sometimes of the order of 20% in PWV, occurred with the original radiosondes. When a Vaisala correction algorithm was applied, calculated T b s were in better agreement with the MWR than were the calculations based on the original data. However, the improvement in T b comparisons was noticeably different for different radiosonde lots and was not a monotonic function of radiosonde age. Three different absorption algorithms were compared: Liebe and Layton, Liebe et al., and Rosenkranz. Using AERI spectral radiance observations as a comparison standard, scaling of radiosondes by MWR data was compared with both original and corrected soundings.
Field studies in support of the Winter Icing and Storms Project (WISP) were conducted in the Colorado Front Range area from 1 February to 31 March 1990 (WISP90) and from 15 January to 5 April 1991 (WISP91). The main goals of the project are to study the processes leading to the formation and depletion of supercooled liquid water in winter storms and to improve forecasts of aircraft icing. During the two field seasons, 2 research aircraft, 4 Doppler radars, 49 Mesonet stations, 7 CLASS sounding systems, 3 microwave radiometers, and a number of other facilities were deployed in the Front Range area. A comprehensive dataset was obtained on 8 anticyclonic storms, 16 cyclonic storms, and 9 frontal passages.
This paper describes the objectives of the experiment, the facilities employed, the goals and results of a forecasting exercise, and applied research aspects of WISP. Research highlights are presented for several studies under way to illustrate the types of analysis being pursued. The examples chosen include topics on anticyclonic upslope storms, heavy snowfall, large droplets, shallow cold fronts, ice crystal formation and evolution, and numerical model performance.
Field studies in support of the Winter Icing and Storms Project (WISP) were conducted in the Colorado Front Range area from 1 February to 31 March 1990 (WISP90) and from 15 January to 5 April 1991 (WISP91). The main goals of the project are to study the processes leading to the formation and depletion of supercooled liquid water in winter storms and to improve forecasts of aircraft icing. During the two field seasons, 2 research aircraft, 4 Doppler radars, 49 Mesonet stations, 7 CLASS sounding systems, 3 microwave radiometers, and a number of other facilities were deployed in the Front Range area. A comprehensive dataset was obtained on 8 anticyclonic storms, 16 cyclonic storms, and 9 frontal passages.
This paper describes the objectives of the experiment, the facilities employed, the goals and results of a forecasting exercise, and applied research aspects of WISP. Research highlights are presented for several studies under way to illustrate the types of analysis being pursued. The examples chosen include topics on anticyclonic upslope storms, heavy snowfall, large droplets, shallow cold fronts, ice crystal formation and evolution, and numerical model performance.
