bjorn lambrigtsen jet propulsion laboratory - california institute of technology...
TRANSCRIPT
Bjorn LambrigtsenJet Propulsion Laboratory - California Institute of Technology
Jet Propulsion LaboratoryCalifornia Institute of TechnologyPasadena, Californiawww.nasa.gov This work was carried out at the Jet Propulsion Laboratory, California Institute of Technology under a contract with the National Aeronautics and Space Administration. Copyright 2008 California Institute of Technology. Government sponsorship acknowledged
A “Geostationary AMSU”A “Geostationary AMSU”
Hurricanes & severe storms 55-GHz warm core anomaly (continuously, in real time)
Surface pressure anomaly = Intensity Equivalent radar reflectivity (continuous coverage of entire life cycle)
Intensification/weakening, microphysics, convective structure Complete diurnal-cycle observations (< 15-minute time resolution)
Model improvements Real-time atmospheric stability indices (including under full cloud cover)
Severe storm warnings, tornado precursor detection AMV-inferred wind vectors (regardless of cloud cover)
Improved forecasts
Synergies Complement GEO IR sounders (cloud clearing) Complement LEO sounders (swath-gap & temporal-gap filling) Complement GEO imagers (resolution enhancement of MW) Global Precipitation Mission (provide spatio-temporal continuity)
Key component of GPM “super-constellation”
Technology
Proof-of-concept prototype- demonstrated in 2005
The GeoSTAR prototype
National Aeronautics and Space Administration
• Antenna size is determined by distance and “spatial resolution”• AMSU antenna is 6” in dia. and gives 40-km resolution from 705 km• GEO orbit is ~37000 km ≈ 50 x 705 km• AMSU-antenna must then be 50 x 6” to give 40-km res. from GEO• This is 25 feet (8 meters)! Can’t be done!• To get 50-km res. we need 20 feet. Still can’t be done•Solution: Synthesize large antenna GeoSTAR
Low-earth-orbiting MW sounder (AMSU)
Key element:The antenna
Principle of aperture synthesis
Science
Dissecting hurricanes: 3D internal structure
Height resolved “Radar reflectivity” Use radar algorithms to derive• Precipitation rate• Ice water path• Convective intensity• Vertical structure
Vertical slicing through hurricane Emily - July 17, 2005
Scattering index - HAMSR water vapor channels
0.0
0.2
0.4
0.6
0 50 100 150 200 250 300Distance [km]
Fractional Tb depression
167: 3-4km183±3: 8km183±1:10km
EDOP
HAMSR
Nadir along-track view
2 km
Scan swath view
MW sounderIs equivalent
to radar!
Correlation between MW-sounder Tb and radar reflectivity exceeds 90% at all levels except near surface
Inferredfrom
HAMSR
3 km
4 km
5 km
6 km
7 km
8 km
9 km
10 km
11 km
12 km
13 km
14 km
15 km
Mature products: Parameter Horizontal Vertical Temporal Accuracy
Tb (50 GHz) 50 km (6 channels) 3 min per ch. < 1/3 K Tb (183 GHz) 25 km (4 channels) 5 min per ch. < 1/3 K Temperature 50 km 2 km 20 min 1.5-2 K Water vapor 25 km 2 km 20 min 25% Liquid water 25 km 3 km 20 min 40%
Stability index 50 km N/A 20 min N/A TPW 25 km N/A 20 min 10% LWC 25 km N/A 20 min 20% SST 100 km N/A 1 hour < 0.5 K
Evolving experimental products: Parameter Horizontal Vertical Temporal Accuracy Rain rate 25 km N/A 20 min 2 mm/hr
Convect. intens. 25 km N/A 20 min N/A IWC 25 km N/A 20 min 30%
Wind vector 25 km 2 km 30 min TBD
Data products
It works!
Outdoor scene(far field)
Indoor scene(near field)
Calibration
“Near Field range”, JPL
GeoSTAR
”Earth” target
Temperaturecontrolled
pads
Computed from model
Measured with GeoSTAR
|Obs - Calc| < 2 K(Will be improved as system processing is optimized)
Calibrationfacilityat JPL:
Earth-like disc
viewedagainst
cold-skybackground
mimickingEarth-viewfrom GEO
Develop
Mathmodel
AssimilateCompare
Obser-vations
InitializeAdjust
Forecast
Models and observations
Weather & climate models are deficientCloud formation, convection & precipitation is
not completely understood; “microphysics” is deficient
Diurnal-cycle variations are not well modeled; storm life cycles are not well modeled
Observations used to “feed” models are incomplete
Most satellite sensors do not penetrate clouds nor observe the internal “microphysics” of clouds & storms
Most satellites provide only brief snapshots twice a day, when the satellite passes overhead
Therefore, we need…Continuous observations of the entire process:
diurnal cycle, storm cycle, Observations under all weather conditions GeoSTAR!
Improve
Applications Numerical weather prediction (regional/global NWP)
—Assimilation of radiances; 4DVAR applications
Hurricane now-casting (15-minute refresh cycle)
—Intensity assessment, detect rapid intensification/weakening—Observation of internal dynamics, kinematics & microphysics
Severe storm development (cloudy observations)
—Atmospheric stability (CAPE, LI, etc.) in cloudy regions—Detect/assess tornado precursor conditions
Weather hazard assessment (continuous observations)
—Life cycle storm observations; total rainfall—Predict/observe flood conditions
Climate studies (stable sensor with “legacy continuity”)
—Continuous time series; diurnal cycle fully resolved—Basin-scale intra-seasonal to interannual analyses—Cross-calibration of LEO climate sensors & data series
Temperature
soundingchannels
Water vaporsoundingchannels
Physical basis of microwave sounding
Absorption spectrum
Weighting functions
Simulated brightness temperature images(tropospheric temperature sounding channels)
Ready to build!Many accommodation options
Currrent estimate
mass = 230 kgpower = 340 WBW = 1 Mbps
Fly GeoSTAR on GOES-R/S/T?