the Odin satellite Preparations for Metop SG Ice Cloud Imager retrievals Patrick Eriksson, Robin Ekelund, Jana Mendrok, Chalmers University of Technology — Bengt Rydberg Molflow — Stefan Buehler, Manfred Brath, University of Hamburg — Anke Thoss, SMHI — Stuart Fox, UK Met Office — Christophe Accadia and Vinia Mattioli EUMETSAT
16
Embed
Preparations for Metop SG Ice Cloud Imager retrievalsipwg/meetings/bologna-2016/Bologna2016... · 2016-10-08 · Preparations for Metop SG Ice Cloud Imager retrievals Patrick Eriksson,
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
the Odin satellite
Preparations for Metop SG Ice Cloud Imager retrievals
Patrick Eriksson, Robin Ekelund, Jana Mendrok,Chalmers University of Technology—Bengt RydbergMolflow—Stefan Buehler, Manfred Brath,University of Hamburg—Anke Thoss,SMHI—Stuart Fox,UK Met Office—Christophe Accadia and Vinia MattioliEUMETSAT
Overview
I Single scattering databaseI Chalmers + Hamburg
I Radiative transfer basicsI Chalmers + Hamburg
I ISMAR airborne instrumentI UK Met Office/ESA + Chalmers + Hamburg + Paris
I Retrieval schemesI SMHI + Chalmers + EUMETSATI HamburgI Paris
posterR. Ekelund
posterJ. Mendrok
posterA. Thoss (PE)
presentationD. Wang
Ice Cloud Imager (ICI)Part of Metop Second Generation (SG)
(courtesy of ESA)
I Launch Dec 2022I 9:30 orbitI Conical scanner, 53◦
I Footprints:I < 16 kmI > 40 % overlap
ChannelsI 183.3±7.0 GHz V pol.I 183.3±3.4 GHz V pol.I 183.3±2.0 GHz V pol.I 243.2 GHz V + H pol.I 325.2±9.5 GHz V pol.I 325.2±3.5 GHz V pol.I 325.2±1.5 GHz V pol.I 448.0±7.2 GHz V pol.I 448.0±3.0 GHz V pol.I 448.0±1.4 GHz V pol.I 664.0 GHz V + H pol.
I + MWI (18.7 - 183.3 GHz)
Why sub-mm? Some example simulationsBased on NICAM model
I Compared to CloudSat/EarthCARE:I larger swath widthI poorer vertical resolution and coverage
Why sub-mm? Some example simulationsBased on NICAM model
I Bridges the gap between IR and existing microwaves
Why sub-mm? Some example simulationsBased on NICAM model
I Main productsI IWP (ice water path)I mean mass size and height
Basic retrieval approach
Retrieval database(set of [x,y])
Retrieval database(set of [x,y])
Inversion approacha: Neural net(s)b: MCI
Inversion approacha: Neural net(s)b: MCI
IWP, Dm and ZmIWP, Dm and Zm
ObservationObservation
Pre-calculations
Retrieval
Atmospheric data1: Model2: CloudSat + ECMWF
Atmospheric data1: Model2: CloudSat + ECMWF
Single scattering dataSingle scattering data
Assumptions, data, ...Assumptions, data, ...
Radiative transfer toolRadiative transfer tool
for operational algorithm,see poster by A. Thoss (PE)
Database of single scattering properties
OverviewI 18 - 887 GHzI 190, 230 and 270 KI Dmax 20µm to 2 cm / x<10I Amsterdam DDAI Total random orientation
I about 30 habitsI Specific alignments
I ? habitsI Can be extended laterI Will be publicly available
I by ARTS web site
StatusI Random orientation
I 20 habits completedI ∼ 50 000 core hoursI more aggregates
will be addedI Specific alignments
I code being tested
posterR. Ekelund
Scattering data, example results
Radiative transfer
I Improving and extending ARTSI ARTS = Atmospheric Radiative Transfer SystemI DISORT and RT4 have been incorporated
I Information content of dual polarisation?
I Relevance of “beam filling” and 3D radiative transfer
posterJ. Mendrok
Example on 3D simulations186.3 GHz, 53◦
Lookingdirection
⇑
Statistical comparison to GMI186.3 GHz, 20◦S - 20◦N, 7 days in Aug 2015
100 120 140 160 180 200 220 240 260 280
Tb [K]
10 -6
10 -5
10 -4
10 -3
10 -2
10 -1
PD
F [#
/K]
GMI, only mid-dayGMI, 24 hSimulations
I CloudSat dBZ→ 3D scenesI Habit = “Hong aggregate”I PSD = Field et al. 2007, tropicalI ARTS Monte Carlo (3D)