地球惑星科学専攻 地球科学輻合部特別講演会

日時: 2010年2月3日(水) 16:30 - 17:30
場所: 京都大学理学部1号館 563号室
題目: TRMM/PRおよびGPM/DPR衛星搭載レーダによる降水強度推定アルゴリズム
講演者: 井口俊夫氏 (情報通信研究機構 電磁波計測研究センター)

講演要旨:
 This paper describes the basic structure of the rain profiling algorithm
for the TRMM Precipitation Radar (PR) and a proposed algorithm for the
GPM Dual-frequency Precipitation Radar (DPR).
 In the PR algorithm, the retrieval of rain rate from radar received power
Pr can be divided into three steps. The first step is the conversion from
Pr to the apparent radar reflectivity factor Zm. Major uncertainties in
this step are the fluctuation in Pr and the background noise. The second
step is the conversion from Zm to the effective radar reflectivity factor
Ze. This step corresponds to the attenuation correction and is the most
crucial step. The third step is the conversion from Ze to rainfall rate
R. Since the attenuation correction and the Ze-to-R conversion depend
most sensitively on the DSD parameters, much effort has been spent on
the selection of appropriate initial DSD parameters and the development
of an algorithm that modifies them according to the extra information
provided by the surface reference technique. The inclusion of attenuation
corrections for cloud liquid water and water vapor in V6 has removed a
small bias in the measured radar echo power and increased the overall
rain estimates. The correction for the effect of inhomogeneous distribution
of precipitation within a footprint also improves heavy rain estimates.
 The assumptions used in the selection of an initial DSD for the PR
algorithm can be loosened to some extent in the GPM's DPR algorithm
because combining the reflectivity data at two radar frequencies enables
us to estimate one additional DSD parameter. It is expected that we can
take advantage of this ability to separate liquid precipitation from
solid precipitation, which provides the actual freezing or melting level,
thereby improving both rain estimates by the DPR and the GPM Microwave
Imager (GMI). On the other hand, radar echoes in the DPR's Ka-band
channel suffer from the attenuation by cloud liquid water and water
vapor much more than in the Ku-band channel, and attenuation correction
for them becomes essential. Since the attenuation in the Ka-band channel
is large, the non-uniformity of rain distribution within a footprint
also creates a large bias in the apparent radar reflectivity. How to
circumvent these new difficulties is a challenge in the DPR algorithm
development.

問い合わせ先: 重 尚一