An assessment of the Doppler measurements with a Ku-band spaceborne
precipitation radar
Abstract
Currently, a future satellite mission of precipitation observations is
discussed in Japan. From a low-orbit satellite, it is difficult to
directly observe temporal evolution of precipitating clouds. The
dynamical structure of precipitation helps better understandings of the
lifecycle of precipitating clouds. Thus, the Doppler capability of a
spaceborne precipitation radar is expected to provide global information
of the motion for various precipitating clouds. However, the Doppler
measurements of precipitation from space is challenging because of a
fast-moving platform and a radar’s finite field of view (FOV). Since the
radar onboard the spacecraft quickly passes above precipitating clouds,
the decorrelation of precipitation signals due to the beam broadening
effect degrades the Doppler measurement accuracy. Moreover, a spatial
variability of precipitation within the FOV causes mixing of the motion
between precipitating particles and spacecraft, which is called as an
effect of the non-uniform beam filling (NUBF). This study investigates
the Doppler capability of the spaceborne precipitation radar based on
simulation experiments by using the high-spatial resolution ground radar
and numerical model data. Here, we discuss two Ku-band Doppler radar
systems: A) a large one antenna system and B) a two-antenna system.
Since the contamination of the platform motion is proportional to the
platform velocity and the radar’s beamwidth, the large antenna system
mitigates the contamination due to the platform motion. On the other
hand, the two-antenna system adopts the displaced phase center antenna
(DPCA) technique. A signal processing with two antennas cancels out the
platform motion so that mitigation of the beam broadening and NUBF
effects is expected even if the FOV is coarse than the large antenna
system. A quantitative evaluation between the two systems is conducted.
For the large antenna system (FOV of 2.5 km), the mean Doppler velocity
error of precipitation (> 15 dBZ) is evaluated in the range
from 2.3 to 5 .0 m/s. Although the large error is originated from a
residual error of the imperfect NUBF correction, the error is mitigated
from 0.7 to 1.5 m/s when a 5-km average in the along-track direction is
applied. For the two-antenna system (FOV of 5 km), the error is
evaluated in the range from 0.6 to 1.1 m/s.