Search Results
You are looking at 1 - 10 of 26 items for
- Author or Editor: E. Zipser x
- Refine by Access: All Content x
Abstract
This paper describes the two different kinds of downdraft air frequently observed to the rear of some squall lines at low levels. The primary data source is measurements taken during aircraft penetrations of certain low-latitude squall lines; they are supplemented by satellite data, radar data, surface meteorological data, and soundings ahead of and behind the squall lines. A shallow layer of cool, near-saturated air occupies the lowest few hundred meters and is separated by a marked stable layer from a deep layer of highly unsaturated air. The lowest layer is hypothesized to be the product of convective-scale saturated downdrafts, and the drier air is shown to be the result of mesoscale unsaturated downdrafts as described by Zipser (1969).
Over a warm ocean, there is a large latent and sensible heat flux from the surface into the lowest layer, which rapidly becomes a new mixed layer and incorporates the drier air from above by entrainment. Mesoscale sinking in the post-squall region is shown to slow the deepening of the shallow mixed layer. The surface dew point drops during squall passage, but is observed to recover more slowly than the temperature toward ambient values. Frequently, tile dew point reaches its absolute minimum value several hours after squall passage, clearly indicating that enhanced evaporation from the surface can be less than the moisture flux through the top of the mixed layer.
An idealized model describing a class of squall lines is presented and discussed. The thermodynamic transformations that take place in each layer of air are identified hypothetically, and they can account for the observed properties of each airstream both before and after passage through the system. The proposed structure permits the coexistence of convective-scale saturated downdrafts and mesoscale unsaturated downdrafts, the former in the active convective clouds of the squall line, the latter farther to the rear.
Abstract
This paper describes the two different kinds of downdraft air frequently observed to the rear of some squall lines at low levels. The primary data source is measurements taken during aircraft penetrations of certain low-latitude squall lines; they are supplemented by satellite data, radar data, surface meteorological data, and soundings ahead of and behind the squall lines. A shallow layer of cool, near-saturated air occupies the lowest few hundred meters and is separated by a marked stable layer from a deep layer of highly unsaturated air. The lowest layer is hypothesized to be the product of convective-scale saturated downdrafts, and the drier air is shown to be the result of mesoscale unsaturated downdrafts as described by Zipser (1969).
Over a warm ocean, there is a large latent and sensible heat flux from the surface into the lowest layer, which rapidly becomes a new mixed layer and incorporates the drier air from above by entrainment. Mesoscale sinking in the post-squall region is shown to slow the deepening of the shallow mixed layer. The surface dew point drops during squall passage, but is observed to recover more slowly than the temperature toward ambient values. Frequently, tile dew point reaches its absolute minimum value several hours after squall passage, clearly indicating that enhanced evaporation from the surface can be less than the moisture flux through the top of the mixed layer.
An idealized model describing a class of squall lines is presented and discussed. The thermodynamic transformations that take place in each layer of air are identified hypothetically, and they can account for the observed properties of each airstream both before and after passage through the system. The proposed structure permits the coexistence of convective-scale saturated downdrafts and mesoscale unsaturated downdrafts, the former in the active convective clouds of the squall line, the latter farther to the rear.
Abstract
The properties of convective drafts and cores are presented in Part I. By our definition a convective updraft must have a positive vertical velocity for 0.5 km, and exceed 0.5 m s−1 for 1 s; a convective updraft core must exceed 1 m s−1 for 0.5 km. Downdrafts and downdraft cores are defined analogously. Here the properties of the drafts and cores are compared to results of previous work. In addition, the implications of the results in Part I are discussed.
GATE cores and drafts are comparable in size and intensity to those measured in hurricanes but weaker than those measured in continental thunderstorms. The lesser intensity seems related to the nearly moist adiabatic GATE sounding. The mass flux by GATE cores is consistent with large-scale requirements. It is fairly evenly distributed over a range of core size and intensity. Updraft core vertical velocity and diameter are positively correlated, primarily the result of a few large strong events.
The vast majority of GATE convective cores are sufficiently weak, with mean vertical velocities < 3–5 m s−1, that the time scale for air starting at cloud base to reach the upper troposphere can be in excess of 1 h. The microphysical implications of such long time scales are discussed. They include large fractional rainout from the warm part of the cloud, the presence of ice at relatively warm temperatures, and rapid decrease of radar reflectivity with height above the 0°C level.
Usually the clouds in GATE were part of a larger, organized mesoscale system. The typical distribution of cumulonimbus clouds, cores and drafts in such a system is synthesized by combining our results with other GATE results. A schematic updraft core and downdraft core in the middle troposphere are presented, emphasizing that these entities were rather narrow and weak in GATE clouds.
Abstract
The properties of convective drafts and cores are presented in Part I. By our definition a convective updraft must have a positive vertical velocity for 0.5 km, and exceed 0.5 m s−1 for 1 s; a convective updraft core must exceed 1 m s−1 for 0.5 km. Downdrafts and downdraft cores are defined analogously. Here the properties of the drafts and cores are compared to results of previous work. In addition, the implications of the results in Part I are discussed.
GATE cores and drafts are comparable in size and intensity to those measured in hurricanes but weaker than those measured in continental thunderstorms. The lesser intensity seems related to the nearly moist adiabatic GATE sounding. The mass flux by GATE cores is consistent with large-scale requirements. It is fairly evenly distributed over a range of core size and intensity. Updraft core vertical velocity and diameter are positively correlated, primarily the result of a few large strong events.
The vast majority of GATE convective cores are sufficiently weak, with mean vertical velocities < 3–5 m s−1, that the time scale for air starting at cloud base to reach the upper troposphere can be in excess of 1 h. The microphysical implications of such long time scales are discussed. They include large fractional rainout from the warm part of the cloud, the presence of ice at relatively warm temperatures, and rapid decrease of radar reflectivity with height above the 0°C level.
Usually the clouds in GATE were part of a larger, organized mesoscale system. The typical distribution of cumulonimbus clouds, cores and drafts in such a system is synthesized by combining our results with other GATE results. A schematic updraft core and downdraft core in the middle troposphere are presented, emphasizing that these entities were rather narrow and weak in GATE clouds.
Abstract
Analysis of serial rawinsonde observations over the Line Islands during March and April 1967 reveals a multi-layered wind structure, especially in the meridional component, which changes sign as many as eight times below 20 km. The layering is strongest above 14 and below 9 km, and is most marked near the equator. Very large vertical wind shears are observed, occurring most frequently near the tropopause. It is likely that the most extreme shears are accompanied by considerable turbulence and may, therefore, represent significant kinetic energy sinks. The layered winds above 14 and below 9 km may be associated with vertically propagating waves. The wind variations in the 9–14 km region appear to be linked to changes in position and intensity of circulation features in either hemisphere rather than with a regular progression of wave disturbances.
Abstract
Analysis of serial rawinsonde observations over the Line Islands during March and April 1967 reveals a multi-layered wind structure, especially in the meridional component, which changes sign as many as eight times below 20 km. The layering is strongest above 14 and below 9 km, and is most marked near the equator. Very large vertical wind shears are observed, occurring most frequently near the tropopause. It is likely that the most extreme shears are accompanied by considerable turbulence and may, therefore, represent significant kinetic energy sinks. The layered winds above 14 and below 9 km may be associated with vertically propagating waves. The wind variations in the 9–14 km region appear to be linked to changes in position and intensity of circulation features in either hemisphere rather than with a regular progression of wave disturbances.
Abstract
This study presents a systematic comparison of the distributions of mesoscale convective systems (MCSs) and lightning for 19 geographical regions classified as land, ocean, or a mixture of land and ocean between 35°N and 35°S over four 3-month periods beginning in June of 1995. The 85-GHz brightness temperatures and the lightning data are from the Special Sensor Microwave Imager (SSM/I) and the Optical Transient Detector, respectively. The MCSs are defined and classified according to their 85-GHz polarization-corrected brightness temperature (PCT), and the lightning flashes are grouped into lightning clusters. In each of the four periods, the land bias among the lightning clusters is much stronger than among the MCSs. For instance, ocean regions contain only 15%–21% of the total lightning cluster population in a given period (compared with 56%–66% over land), and the majority (>80%) of the oceanic lightning clusters are weak, with few flashes. In contrast, MCSs are more evenly distributed between land and ocean regions with 37%–41% and 40%–45% occurring over the land and oceans, respectively, in a given period. In land regions, MCSs with moderate to strong ice-scattering signatures (minimum 85-GHz PCT ≤ 190 K) and lightning clusters with moderate to high flash rates (four or more flashes) are both relatively numerous, with tropical Africa typically dominating all regions in terms of ice-scattering intensity and lightning flash rates. However, the lightning–ice-scattering relationship is less clear over the oceans. Moderate to strong ice-scattering MCSs occur with far greater frequency over ocean regions than do the moderate-to-high-flash-rate clusters. In addition, the lightning flash densities and flash-to-MCS ratios computed for each region show order-of-magnitude or larger differences between land and ocean. This result suggests that, even when normalized for the intensity of 85-GHz ice scattering, a land MCS is more likely to produce lightning than is an MCS over the ocean. This fact implies differences in the ice microphysics processes between land and ocean convective storms. These differences are under active investigation.
Abstract
This study presents a systematic comparison of the distributions of mesoscale convective systems (MCSs) and lightning for 19 geographical regions classified as land, ocean, or a mixture of land and ocean between 35°N and 35°S over four 3-month periods beginning in June of 1995. The 85-GHz brightness temperatures and the lightning data are from the Special Sensor Microwave Imager (SSM/I) and the Optical Transient Detector, respectively. The MCSs are defined and classified according to their 85-GHz polarization-corrected brightness temperature (PCT), and the lightning flashes are grouped into lightning clusters. In each of the four periods, the land bias among the lightning clusters is much stronger than among the MCSs. For instance, ocean regions contain only 15%–21% of the total lightning cluster population in a given period (compared with 56%–66% over land), and the majority (>80%) of the oceanic lightning clusters are weak, with few flashes. In contrast, MCSs are more evenly distributed between land and ocean regions with 37%–41% and 40%–45% occurring over the land and oceans, respectively, in a given period. In land regions, MCSs with moderate to strong ice-scattering signatures (minimum 85-GHz PCT ≤ 190 K) and lightning clusters with moderate to high flash rates (four or more flashes) are both relatively numerous, with tropical Africa typically dominating all regions in terms of ice-scattering intensity and lightning flash rates. However, the lightning–ice-scattering relationship is less clear over the oceans. Moderate to strong ice-scattering MCSs occur with far greater frequency over ocean regions than do the moderate-to-high-flash-rate clusters. In addition, the lightning flash densities and flash-to-MCS ratios computed for each region show order-of-magnitude or larger differences between land and ocean. This result suggests that, even when normalized for the intensity of 85-GHz ice scattering, a land MCS is more likely to produce lightning than is an MCS over the ocean. This fact implies differences in the ice microphysics processes between land and ocean convective storms. These differences are under active investigation.
Abstract
Aircraft flight-level data from 787 radial legs in 20 hurricanes are analyzed to identify the composite kinematic structure in the hurricane eyewall, and especially with secondary horizontal wind maxima (SHWM) that occur outside the eyewall. Similar to previous studies, analysis of the flight-level wind data in the eyewall reveals radial convergence near the radius of maximum wind (RMW), and the highest frequency of updraft and the largest upward mass transport radially inward of the RMW.
More than 20% of the flight legs contain substantial secondary horizontal wind maxima of specified strength and length. The kinematic structure associated with SHWM is similar to that of the hurricane eyewall with radial convergence near the radius of maximum wind and a preferred location for maximum upward motions and upward mass transport just inside the RMW. Statistical analysis confirms the similarity in characteristics between radial and vertical velocities of the eyewall and near the SHWM. In addition, for both the eyewalls and SHWM, the radial velocity composite results show that the radial mass transport in the planetary boundary layer must be largely confined to the lowest 1000 m.
Lower fuselage radar reflectivity data from 13 of the hurricanes are used to assess whether the outer wind maxima are associated with rainbands, and vice versa. In the radial legs with SHWM for which radar data were available, the secondary horizontal wind maximum was frequently associated with a mesoscale reflectivity feature (rainband). In contrast, many rainbands, more than 70%, were without wind maxima. The results from this study show that to some extent an outer eyewall or rainband with SHWM can act as a barrier to inflow to the inner eyewall. Additionally, it is possible that thermodynamic modification of inflow air may occur as a result of convective-scale vertical motions associated with a rainband. In those cases when an outer rainband encircles the eyewall, it is possible that these factors act together with subsidence to weaken the inner eyewall.
Abstract
Aircraft flight-level data from 787 radial legs in 20 hurricanes are analyzed to identify the composite kinematic structure in the hurricane eyewall, and especially with secondary horizontal wind maxima (SHWM) that occur outside the eyewall. Similar to previous studies, analysis of the flight-level wind data in the eyewall reveals radial convergence near the radius of maximum wind (RMW), and the highest frequency of updraft and the largest upward mass transport radially inward of the RMW.
More than 20% of the flight legs contain substantial secondary horizontal wind maxima of specified strength and length. The kinematic structure associated with SHWM is similar to that of the hurricane eyewall with radial convergence near the radius of maximum wind and a preferred location for maximum upward motions and upward mass transport just inside the RMW. Statistical analysis confirms the similarity in characteristics between radial and vertical velocities of the eyewall and near the SHWM. In addition, for both the eyewalls and SHWM, the radial velocity composite results show that the radial mass transport in the planetary boundary layer must be largely confined to the lowest 1000 m.
Lower fuselage radar reflectivity data from 13 of the hurricanes are used to assess whether the outer wind maxima are associated with rainbands, and vice versa. In the radial legs with SHWM for which radar data were available, the secondary horizontal wind maximum was frequently associated with a mesoscale reflectivity feature (rainband). In contrast, many rainbands, more than 70%, were without wind maxima. The results from this study show that to some extent an outer eyewall or rainband with SHWM can act as a barrier to inflow to the inner eyewall. Additionally, it is possible that thermodynamic modification of inflow air may occur as a result of convective-scale vertical motions associated with a rainband. In those cases when an outer rainband encircles the eyewall, it is possible that these factors act together with subsidence to weaken the inner eyewall.
Nimbus IV radiance measurements in the 10.5–12.5-μm window channel from the tropical Atlantic during July and August 1970 are studied. The relative frequency of cold, and presumably high clouds, is determined. The average spatial distribution of equivalent black-body temperatures associated with 38 cloud clusters is reported. Ship cloud and weather observations are studied in conjunction with the satellite data. By combining the evidence provided by the satellite and ship observations, a model of typical clouds and weather associated with cloud clusters is proposed for further study.
Nimbus IV radiance measurements in the 10.5–12.5-μm window channel from the tropical Atlantic during July and August 1970 are studied. The relative frequency of cold, and presumably high clouds, is determined. The average spatial distribution of equivalent black-body temperatures associated with 38 cloud clusters is reported. Ship cloud and weather observations are studied in conjunction with the satellite data. By combining the evidence provided by the satellite and ship observations, a model of typical clouds and weather associated with cloud clusters is proposed for further study.
Abstract
The evolution of an oceanic prefrontal subcloud layer of continental origin was examined by analyzing data gathered on the subsynoptic and mesoscale during Phase III of the Australian Cold Fronts Research Program. Characteristics of the atmosphere ahead of the surface cold front included deep ascent through the troposphere and horizontal low-level warm air advection. Temporal and spatial variation in the thermodynamic structure of the 3.5-km thick prefrontal subcloud layer confirmed that significant local cooling was occurring despite the presence of horizontal warm air advection. Over a 24-hour period this cooling created a new low-level baroclinic zone several hundred kilometers ahead of the front.
Observations from surface stations, rawinsodes, radars, satellites, and research aircraft were combined to demonstrate that the observed local cooling in the prefrontal air mass was most plausibly due to evaporation of precipitation. The region in question contained mesocale bands of precipitation, including convective showers that generated strong downdrafts. The observed mesoscale features are consistent with those obtained in a numerical modelling study of the generation of similar prefrontal rainbands, and appear to account for the observed modification of the baroclinic structure of the frontal zone.
Abstract
The evolution of an oceanic prefrontal subcloud layer of continental origin was examined by analyzing data gathered on the subsynoptic and mesoscale during Phase III of the Australian Cold Fronts Research Program. Characteristics of the atmosphere ahead of the surface cold front included deep ascent through the troposphere and horizontal low-level warm air advection. Temporal and spatial variation in the thermodynamic structure of the 3.5-km thick prefrontal subcloud layer confirmed that significant local cooling was occurring despite the presence of horizontal warm air advection. Over a 24-hour period this cooling created a new low-level baroclinic zone several hundred kilometers ahead of the front.
Observations from surface stations, rawinsodes, radars, satellites, and research aircraft were combined to demonstrate that the observed local cooling in the prefrontal air mass was most plausibly due to evaporation of precipitation. The region in question contained mesocale bands of precipitation, including convective showers that generated strong downdrafts. The observed mesoscale features are consistent with those obtained in a numerical modelling study of the generation of similar prefrontal rainbands, and appear to account for the observed modification of the baroclinic structure of the frontal zone.
Abstract
Very high frequency (VHF) Doppler radar measurements of the horizontal and vertical winds are used to examine three procedures to extract mean profiles of horizontal and vertical winds. These are 1) time averaging of first-moment estimates of radial velocity from the high time resolution Doppler spectra; 2) time averaging of radial velocities estimated from a least-squares fitting of either one or two Gaussians to the spectra in order to account for the double peaks corresponding to turbulent and precipitation scattering that appear in the spectra during heavy rain; and 3) consensus averaging of the least-square-fitted radial velocities. Horizontal winds produced by these procedures were compared to each other and to those from two 5-cm radars operating nearby. Least-squares fitting yielded the best wind estimates, although a slight relaxation of the consensus criterion was sometimes found to be necessary in order to avoid the failure to find a consensus. The simple first-moment method produced comparable results, except below the melting level, where it performed more poorly. Vertical winds from the fitted VHF spectra were compared with those derived from the 5-cm-radar data using the extended velocity-azimuth display (VAD) technique. Reasonable agreement was found at heights above the freezing level.
Abstract
Very high frequency (VHF) Doppler radar measurements of the horizontal and vertical winds are used to examine three procedures to extract mean profiles of horizontal and vertical winds. These are 1) time averaging of first-moment estimates of radial velocity from the high time resolution Doppler spectra; 2) time averaging of radial velocities estimated from a least-squares fitting of either one or two Gaussians to the spectra in order to account for the double peaks corresponding to turbulent and precipitation scattering that appear in the spectra during heavy rain; and 3) consensus averaging of the least-square-fitted radial velocities. Horizontal winds produced by these procedures were compared to each other and to those from two 5-cm radars operating nearby. Least-squares fitting yielded the best wind estimates, although a slight relaxation of the consensus criterion was sometimes found to be necessary in order to avoid the failure to find a consensus. The simple first-moment method produced comparable results, except below the melting level, where it performed more poorly. Vertical winds from the fitted VHF spectra were compared with those derived from the 5-cm-radar data using the extended velocity-azimuth display (VAD) technique. Reasonable agreement was found at heights above the freezing level.
Abstract
On 18 January 1987 the three aircraft from the Equatorial Mesoscale Experiment (EMEX) completed a mission designed to identify the mesoscale reflectivity, kinematic, and thermodynamic structure of a convectively active rainband on the leading edge of developing Tropical Cyclone Irma. The reflectivity structure, determined from land-based and aircraft radars, reveals that the band consisted of an unusually wide (80 km) region of convective clouds that did not maintain a linear organization. High-θ e air entered from both sides in the subcloud and lower cloud layer. Thermodynamically, the rainband was benign in the sense that it did not produce cold downdrafts or large regions of low-θ e air that would limit the energy of the low-level inflow to the developing eyewall region. Kinematically, the band was a preferred location for the enhancement of the wind component parallel to the band. The low-level flow was along rather than across the major axis of the band. Above 4.5 km flow was from the inner to the outer side of the band, but this air was also converging, which contributed to the widespread heavy rain encountered by the aircraft. The chaotic reflectivity, minor thermodynamic modification of the low-level inflow, and the tendency of the flow to be along the band are characteristics that contrast sharply with those of convectively active bands observed on the leading edge of mature tropical cyclones.
Abstract
On 18 January 1987 the three aircraft from the Equatorial Mesoscale Experiment (EMEX) completed a mission designed to identify the mesoscale reflectivity, kinematic, and thermodynamic structure of a convectively active rainband on the leading edge of developing Tropical Cyclone Irma. The reflectivity structure, determined from land-based and aircraft radars, reveals that the band consisted of an unusually wide (80 km) region of convective clouds that did not maintain a linear organization. High-θ e air entered from both sides in the subcloud and lower cloud layer. Thermodynamically, the rainband was benign in the sense that it did not produce cold downdrafts or large regions of low-θ e air that would limit the energy of the low-level inflow to the developing eyewall region. Kinematically, the band was a preferred location for the enhancement of the wind component parallel to the band. The low-level flow was along rather than across the major axis of the band. Above 4.5 km flow was from the inner to the outer side of the band, but this air was also converging, which contributed to the widespread heavy rain encountered by the aircraft. The chaotic reflectivity, minor thermodynamic modification of the low-level inflow, and the tendency of the flow to be along the band are characteristics that contrast sharply with those of convectively active bands observed on the leading edge of mature tropical cyclones.
Abstract
The tilt with height of the leading edge of seven mesoscale convective lines in GATE is determined by two independent methods. When averaged over time and along the line axis, the tilt is found to he surprisingly shallow, between 20 and 35 degrees from the horizontal. This is distinct from the slopes of the individual towers, which can be much steeper. The line leading-edge slope corresponds to the ratio of the vertical to horizontal velocity, relative to the line motion, of “representative” embedded convective cores.
Abstract
The tilt with height of the leading edge of seven mesoscale convective lines in GATE is determined by two independent methods. When averaged over time and along the line axis, the tilt is found to he surprisingly shallow, between 20 and 35 degrees from the horizontal. This is distinct from the slopes of the individual towers, which can be much steeper. The line leading-edge slope corresponds to the ratio of the vertical to horizontal velocity, relative to the line motion, of “representative” embedded convective cores.
