292
P. Joe
unambiguous velocities and combined to yield a new estimate of the radial velocity with an
extended unambiguous velocity (Crozier et al., 1991; Sauvageot, 1982). For example, a CBand radar using PRFs of 1200 and 900 Hz have nominal unambiguous velocities of 16 and 12
mis, respectively. The amount of aliasing can be deduced from the difference between the two
velocity estimates to de-alias the velocity to an extended velocity range of ±48 m S-I.
Continuity techniques rely on having sufficient echo to discern that there are aliased velocities
and correcting them by assuming velocity continuity (i.e., assuming no discontinuous jumps of
2 x Vmax ). See Doviak and Zrnic (1993) for a detailed discussion.
There is also a range limitation imposed by the use of high PRFs (~1000 Hz). Echoes beyond
the maximum range will be aliased back into the primary range. For radars with coherent
transmitters (for example, Klystron systems), the echoes will appear within the primary range.
For coherent-on-receive systems, the second trip echoes will have very noisy velocity characteristics due to the randomly transmitted phases and can be eliminated to some extent (Joe et
al., 1995).
12.4.1 Measurement of velocity fields
A great deal of information can be determined in real-time from a single Doppler radar. Color
displays of single-Doppler radial velocity patterns aid in the real-time interpretation of the
associated reflectance fields, and can reveal important features not evident in the reflectance
structures alone. Such a capability is of particular importance in the identification and tracking
of severe storms. On typical color displays, velocities between ± Vmax are assigned one of 8-15
colors or more. Velocities extending beyond the Nyquist interval enter the scale of colors at the
opposite end. This process may be repeated ifthe velocities are aliased more than one Nyquist
interval.
Doppler radar can also be used to derive vertical profiles of horizontal winds. When the radar's
antenna is tilted above horizontal, increasing range implies increasing height, and a profile
of wind with height can be obtained by sinusoidal curve-fitting to the observed data (termed
velocity-azimuth display or VAD technique after Lhermitte and Atlas, 1961). The winds along
the zero radial velocity contour are perpendicular to the radar beam axis. Typical elevated
conical scan patterns in widespread warm frontal precipitation reveal an S-shaped zero radial
velocity contour as the mean wind veers with height (Wood and Brown, 1986). On other
occasions, closed contours representing jets are evident.
Since the measurement accuracy is good, divergence estimates can also be obtained employing
the VAD technique. This technique cannot be accurately applied during periods of convective precipitation around the radar. The VAD technique seems well suited for winds from
precipitation systems associated with extratropical and tropical cyclones.
Moderately powerful, sensitive Doppler radars have successfully obtained VAD wind profiles
and divergence estimates in the optically clear boundary layer during all but the coldest months,
up to heights of 3-5 km AGL. In the radar's clear-air mode, a time-series of measurements of
divergence and derived vertical velocity is particularly useful in nowcasting the probability of
deep convection.
Since the mid 1970's, experiments have been made for measuring three dimensional wind fields
using multiple Doppler arrays. Measurements made at a given location inside a precipitation
area may be combined, by using a proper geometrical transformation, in order to obtain the
three wind components. Many significant advances in the knowledge of cloud dynamics have
been obtained from multiple Doppler radar systems (Ray et al., 1980).
P. Joe
unambiguous velocities and combined to yield a new estimate of the radial velocity with an
extended unambiguous velocity (Crozier et al., 1991; Sauvageot, 1982). For example, a CBand radar using PRFs of 1200 and 900 Hz have nominal unambiguous velocities of 16 and 12
mis, respectively. The amount of aliasing can be deduced from the difference between the two
velocity estimates to de-alias the velocity to an extended velocity range of ±48 m S-I.
Continuity techniques rely on having sufficient echo to discern that there are aliased velocities
and correcting them by assuming velocity continuity (i.e., assuming no discontinuous jumps of
2 x Vmax ). See Doviak and Zrnic (1993) for a detailed discussion.
There is also a range limitation imposed by the use of high PRFs (~1000 Hz). Echoes beyond
the maximum range will be aliased back into the primary range. For radars with coherent
transmitters (for example, Klystron systems), the echoes will appear within the primary range.
For coherent-on-receive systems, the second trip echoes will have very noisy velocity characteristics due to the randomly transmitted phases and can be eliminated to some extent (Joe et
al., 1995).
12.4.1 Measurement of velocity fields
A great deal of information can be determined in real-time from a single Doppler radar. Color
displays of single-Doppler radial velocity patterns aid in the real-time interpretation of the
associated reflectance fields, and can reveal important features not evident in the reflectance
structures alone. Such a capability is of particular importance in the identification and tracking
of severe storms. On typical color displays, velocities between ± Vmax are assigned one of 8-15
colors or more. Velocities extending beyond the Nyquist interval enter the scale of colors at the
opposite end. This process may be repeated ifthe velocities are aliased more than one Nyquist
interval.
Doppler radar can also be used to derive vertical profiles of horizontal winds. When the radar's
antenna is tilted above horizontal, increasing range implies increasing height, and a profile
of wind with height can be obtained by sinusoidal curve-fitting to the observed data (termed
velocity-azimuth display or VAD technique after Lhermitte and Atlas, 1961). The winds along
the zero radial velocity contour are perpendicular to the radar beam axis. Typical elevated
conical scan patterns in widespread warm frontal precipitation reveal an S-shaped zero radial
velocity contour as the mean wind veers with height (Wood and Brown, 1986). On other
occasions, closed contours representing jets are evident.
Since the measurement accuracy is good, divergence estimates can also be obtained employing
the VAD technique. This technique cannot be accurately applied during periods of convective precipitation around the radar. The VAD technique seems well suited for winds from
precipitation systems associated with extratropical and tropical cyclones.
Moderately powerful, sensitive Doppler radars have successfully obtained VAD wind profiles
and divergence estimates in the optically clear boundary layer during all but the coldest months,
up to heights of 3-5 km AGL. In the radar's clear-air mode, a time-series of measurements of
divergence and derived vertical velocity is particularly useful in nowcasting the probability of
deep convection.
Since the mid 1970's, experiments have been made for measuring three dimensional wind fields
using multiple Doppler arrays. Measurements made at a given location inside a precipitation
area may be combined, by using a proper geometrical transformation, in order to obtain the
three wind components. Many significant advances in the knowledge of cloud dynamics have
been obtained from multiple Doppler radar systems (Ray et al., 1980).