PowerPoint プレゼンテーション

(1)

Kazumasa Aonashi (MRI/JMA)

PEHRPP Workshop

Dec. 3, 2007 Geneva

GSMaP Passive Microwave

Precipitation Retrieval

Algorithm

(2)

Passive Microwave Precipitation Retrieval

GSMaP

Retrieval Algorithm

G

lobal

S

atellite

Ma

pping of

P

recipitation Project

started in 2003.

Leader: Prof. Ken’ich Okamoto (Osaka Pref. Univ.)

Funded by JST/CREST

The goal is to produce accurate precip map using

mainly satellite microwave radiometer.

Passive microwave precip retrieval (Aonashi) Passive microwave precip+ IR wind (Dr. Ushio)

(3)

Outline

Introduction

Algorithm Description

Forward Calculation (precip cloud models)

Retrieval Part

Validation (TRMM PR, Ground-based Obs.)

Future Directions

Improvement of Scattering part

Summary

(4)

Algorithm Description

Forward Calculation

Retrieval Part

(5)

Observed TBs Precip. FLH Precip Profiles DSD Mixed phase inhomogeneity Look-up Table

Forward calculation Retrieval Calculation

Basic Idea of the Retrieval Algorithm

Precip Cloud Models RTM Screening Inhomogeneity estimation Scattering part Radiation part

Find the optimal precipitation that gives RTM

Find the optimal precipitation that gives RTM--calculated calculated TBsTBs fitting best with the observed

fitting best with the observed TBsTBs: : PCT37, PCT85 (land)

PCT37, PCT85 (land)

TB10v,TB19v, PCT37, PCT85 (sea)

(6)

Freezing Level Rain Freezing Level Height Frozen Precip Mixed-phase model

Particle Size Distribution

Precipitation Profile model

Precipitating Cloud Model for forward calculation

Atmosphere & Surfac (GANAL)

Precipitation typ Stratiform/

(7)

Atmospheric variables

(Temp,FLH), surface

variables(Ts, SSW, SST)

are derived from the Global

Analysis data of JMA

Parameters used in the Algorithm:

Atmospheric & surface variables

Freezing

Level Height for Jan.1,

2003

Temperature bias of GANAL against sonde

(8)

Precip type classification

Precip Profile

Data base

Precip profile data base

Precipitation Profile Model

Rainfall rate [mm/h]

He

ight from 1 de

g level [km]

1℃ level Example：

TRMM PR averaged preciptation profiles for each type, surface precip, conv/stra (land) 0: thunderstorm, 1: shower, 2: shallow, 3: frontal rain, 4: organized rain 5: highland

(sea) 6: shallow 7:frontal rain, 8:transit, 9:organized rain 10 types (land 6, sea 4) are

classified from TRMM PR data (2.5 deg, 3 monthly)

(9)

Particle Size Distribution

Data base of conv. Epsillon Averaged for each precip type

DSD for rain: Kozu model (2A25 average distribution

calibrated with averaged epsillon)

epsillon =1 for stratiform rain

PSD for frozen particles: Marshall-Palmer distribution

)

exp(

)

(

D

N

₀

D

(10)

Nishitsuji

(Mixed

_(Mixed

-

_-

Phase) Model

_{Phase) Model}

for Stratiform Rain

On the basis of the filed experiment, the following parameters are modeled Volume liquid water fraction （Pw)

shape parameter of the dielectric constant (U) DSD parameter (B) is a function of Pw

Density ρ=√Pw

Fall velocity Magono-Nakamura(1965) for snow and Foot and Du Toit for rain

0 0.5 1 1 0.5 0 100 102 104 106 PW Di st a n c e b e lo w BB ( k m ) U U PW U P U P U P U _a a a i i i w w w s s + − + + − + + − = + − ε ε ε ε ε ε ε ε 1 1 1 1 cm) in radius : (a ) ( 10 ) (D = N₀ − m−3mm−1 N Ba

Pw and U profile Relationship between B and Pw

10–3 10–2 10–1 100 20

40

PW

(11)

LUT calculation (1)

TBs

for homogeneous

precip

Radiative Transfer Code (Liu,1998)

One-dimensional model (Plane-parallel)

Mie Scattering (Sphere)

4 stream approximation

Calculate TBs for homogeneous, convective &

stratiform precip with each precip types.

function phase is P K K K dz K K where d d TB P T TB d dTB sc ab sc sc ab , /( ), , cos ) , , ( ) , , , , ( 4 ) ( ) 1 ( ) , , ( 0 ' ' ' ' ' ' 0 0 + = + = = − − − =

∫

∫∫

ω τ θ μ ϕ μ ϕ μ τ ϕ μ ϕ μ τ π ω τ ω τ ϕ μ τ μ

(12)

LUT calculation (2):

LUTs

for

inhomo

precip

T

The calculated he calculated TBs TBs are are converted into

converted into TBs TBs for for inhomogeneous

inhomogeneous precip precip with with Aonashi and Liu

Aonashi and Liu’’s method s method (2000

(2000).).

LUT used for retrieval is

weighted average of

convective & stratiform

convective & stratiform TBsTBs..

STD of

STD of Log(PrLog(Pr) is estimated ) is estimated from STD of Log(rain85) from STD of Log(rain85) statistically. statistically. LOOK-UP TABLE (LUT) STD LN(PR) (PCT85ra < 260K) 2.0 1.5 1.0 .5 0.0 SI G M A 85 2.0 1.5 1.0 .5 0.0 0.414+ 0.678*sigma85o SIGMA85O STDLGPRH

(13)

Flow of the Retrieval Part

Screening of Precip Areas

Inhomo. Estimation / LUT selection

First-guess of Precipitation (scattering)

Minimization of Σ(TBc-TBo)**2 Observed TBs rain flag rain37 Å PCT37+LUT rain85 Å PCT85+LUT Over Ocean Retrivals LOOK-UP TABLE (LUT) Inhomogeneity (STD of Log (Pr)) TB10v,TB19v Over Land Retrievals

(14)

Validation

Comparison with

TRMM PR &

Ground

-

based

observations

(15)

Multi-Satellite Precip Composite

（

GSMaP_MWR, daily precip

）

(16)

Zonal Mean TRMM

Precip

over Ocean

GSMaP

, GPROF,

PR

(1998

～

2006)

PR 2A25V6 TMI 2A12V6 GSMaP V4.8.4 1998-2006 PR swath only

(17)

Comparison with ground

-

based radar

(0.25 x 0.25 deg in lat

-

lon

grid)

Corretation：0.82（No:253）

RMSE：1.37 mm/hr Correlation：0.65（No:1139）_{RMSE：1.78 mm/hr}

COBRA(Okinawa)

4 cases in June 2004

Kwajalein Radar

(18)

Zonal Mean TRMM

Precip

over Land

GSMaP

, GPROF,

PR

(1998

～

2006)

PR 2A25V6 TMI 2A12V6 GSMaP V4.8.4 1998-2006 PR swath only

(19)

Comparison with GPCC data

GSMaP_MWR：monthly mean preciptation (1x1 deg)

GPCC Monthly Precipitation (Monitoring) Product

（Rudolf et al. 2006）： Correlation: 0.80 for 45S～45N （sample: 69440） 0.85 for 15S～15N （sample: 2177） Fitting： y=1.14 x + 10.1 [40S～40N] y=1.21 x + 19.2 [15S～15N]

(20)

Ratio of TMI scattering signals to PR

in terms of

precip

top level (July, 1998)

DTOP 10000 8000 6000 4000 2000 0 -2000 ra in 37 / ra in su rf 50 40 30 20 10 5 4 3 2 1 .5 .4 .3 .2 .1 .05 .04 .03 .02 .01 R 2 乗 = 0.0843 DTOP 10000 8000 6000 4000 2000 0 -2000 ra in8 5/r ai ns ur f 50 40 30 20 10 5 4 3 2 1 .5 .4 .3 .2 .1 .05 .04 .03 .02 .01 R 2 乗 = 0.2697

Precip is over-(under-)estimated for PR high (low) top level. (Rain85/rainsurf) is sensitive to PR top level.

PR top level –FLH (m) PR top level –FLH (m)

(Rain85/rainsurf) (Rain37/rainsurf)

(21)

Future Directions

PSD, densities of

frozen particles

Scattering properties of

Non-spherical particles

(22)

Estimation of realistic PSD, density etc.

PSD (field campaign)

RTM simulation

Observed

Radar &

MWR data

(23)

Scattering properties of non-spherical

frozen particles (Liu, 2004)

Æ An actual target is approximated by an array of dipoles.

DDA :Each of the dipoles is subject to an electric field which is the sum of the incident wave and the electric fields due to all of the other dipoles.

From Mishchenko et al (2000)

• Keeping the single scattering properties

Dmax

D

SP=(D-D0)/(Dmax-D0)

D0

:

diameter of solid sphere

Non-Sphere

(24)

Summary

GSMaP passive microwave precipitation

retrieval algorithm:

Atmopheric & surface variables from GANAL

Precip profile, DSD & inhomo. from PR statistics Mixed-phase model

The retrieved precipitation agreed well with

PR, radar data over ocean.

The over-land algorithm underestimated the

GPCC precipitation, and showed bias in terms

of precipitation top level.

Introduction of the scattering of non-spherical

particles, realistic PSD etc.

(25)

END

(26)

GSMaP

MWR algorithm

_{MWR algorithm}

降水強度 • GANAL（大気、地表面物理量） • 降水タイプ分類（陸上6種、海上4種） • 降水（鉛直）プロファイルモデル • 雨滴粒径分布モデル · 融解層モデル • 層状性降雨、対流性降雨分類降水物理モデル（フォワード計算）各種判定補正降水強度推定 • 衛星が観測するのは、放射・散乱強度の積分値を表す輝度温度である。 • 降水物理モデルを仮定して放射伝達方程式を計算し、輝度温度と降水強度の関係をテーブル化し、観測値に近い輝度温度を与える降水強度を解としている。観測データルックアップテーブル (LUT) 放射伝達方程式 • 散乱アルゴリズム（85, 37GHz） • 放射アルゴリズム（10, 19, 37GHz） • 陸上降雨有無判定 • 海上降雨有無判定 • 海岸降雨有無判定 • 降水の非一様性補正アルゴリズム本体（リトリーバル）降水強度 • GANAL（大気、地表面物理量） • 降水タイプ分類（陸上6種、海上4種） • 降水（鉛直）プロファイルモデル • 雨滴粒径分布モデル · 融解層モデル • 層状性降雨、対流性降雨分類降水物理モデル（フォワード計算）各種判定補正降水強度推定 • 衛星が観測するのは、放射・散乱強度の積分値を表す輝度温度である。 • 降水物理モデルを仮定して放射伝達方程式を計算し、輝度温度と降水強度の関係をテーブル化し、観測値に近い輝度温度を与える降水強度を解としている。観測データルックアップテーブル (LUT) 放射伝達方程式 • 散乱アルゴリズム（85, 37GHz） • 放射アルゴリズム（10, 19, 37GHz） • 陸上降雨有無判定 • 海上降雨有無判定 • 海岸降雨有無判定 • 降水の非一様性補正アルゴリズム本体（リトリーバル）

(27)

Particle Size Distribution Model

Kototabang 50 100 150 200 250 300 350 400 450 Ja n Feb Ma r Ap r Ma y Ju n Ju l Au g Sep Oc t No v De c Ja n Month a kt-5deg98-04 kt-1deg98-05 kt-disd01-03 TRMM PRから推定したZ-R関係の係数aとディスドロメータから推定したaの比較．コトタバン（西スマトラ・山岳地帯）地上ディスドロメータデータによる検証全球で取得可能なDSD パラメータ(TRMM PRの ε)の分布と、降水タイプ分類のパターンの類似性から、降水タイプ分類と関係づけたデータベース

(28)

現状

ア

リ

を

適用

リ

リーバル値とPR（降水レーダ）の地上降水

強度の比較（９８年７月）

DTOP 10000 8000 6000 4000 2000 0 -2000 ra in 37 /r a in sur f 50 40 30 20 10 54 3 2 1 .5 .4 .3 .2 .1 .05 .04 .03 .02 .01 R 2 乗 = 0.0843 DTOP 10000 8000 6000 4000 2000 0 -2000 ra in 85 /r a in sur f 50 40 30 20 10 54 3 2 1 .5 .4 .3 .2 .1 .05 .04 .03 .02 .01 R 2 乗 = 0.2697 降水トップの高い（低い）降水を過大（過小）評価する特に８５ＧＨｚの散乱シグナルは降水トップへの感度が大きい RAINSURF 30 25 20 15 10 5 0 R A IN 3785 30 25 20 15 10 5 0 R 2 乗 = 0.4161 30 25 20 15 10 5 0 30 25 20 15 10 5 0 RAIN37 RAINSURF R 2 乗 = 0.3967 RAIN85 RAINSURF R 2 乗 = 0.3074

(29)

Evaluation using PR Match-up data

We were able to generate 141 AMSR-E vs. PR match-up data within

observation time difference 5 minutes for Jan,Feb,Mar,Jun,Jul,Aug 2003. This figure shows distribution of the match-up locations.

(30)

Evaluation

using

_using

PR Match

_{PR Match}

-

_-

up data

_{up data}

△

△ △ △ △ △

-2 -1 -0.5 0 0.5 1 2

Distribution of the bias

(31)

Evaluation

using

_using

PR Match

_{PR Match}

-

_-

up data

_{up data}

This figure shows the histogram of BIAS which are calculated from 141 match-up data.

Liu

Petty

Aonashi

(32)

Evaluation

using

_using

PR Match

_{PR Match}

-

_-

up data

_{up data}

Distribution of the RMSE

△ △

△

(33)

Evaluation using PR Match-up data

Liu

Petty

Aonashi

This figure shows the frequency of RMSE which are calculated from 141 match-up data.