IGARSS July 2011 1 Thresholds of Detection for Falling Snow from Satellite-Borne Active and Passive Sensors IGARSS 2011 Vancouver, Canada Gail Skofronick Jackson Benjamin Johnson Joe Munchak NASA Goddard Space Flight Center, Greenbelt, Maryland [email protected]
18
Embed
IGARSS July 2011 1 Thresholds of Detection for Falling Snow from Satellite-Borne Active and Passive Sensors IGARSS 2011 Vancouver, Canada Gail Skofronick.
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
IGARSS July 2011 1
Thresholds of Detection for Falling Snow from Satellite-Borne Active and Passive Sensors
(2) Falling Snow Detection Thresholds• Analysis framework• Active thresholds based on instrument sensitivity• Passive thresholds• Comparison between active and passive• Future improvements
(3) Snow Field Campaign (Jan – Feb 2012)
(4) Summary
IGARSS July 2011 3
Percentages from Surface, Snow, & Water VaporL
ake
Eff
ect
2-3k
m t
op
s (0
.5 t
o 1
.0 I
WP
)S
yno
pti
c 5-
7km
to
ps
(0.5
to
1.0
IW
P)
Bli
zzar
d ~
10km
to
ps
(0.5
to
1.0
IW
P)
Bli
zzar
d ~
10km
to
ps
(9 t
o 1
0 IW
P)
“Surface and Atmospheric Contributions to Microwave Brightness Temperatures for Falling Snow Events,” by Gail Skofronick-Jackson and Benjamin Johnson, JGR-Atmos, published Jan 2011.
(a)
(b)
(a)
(b)
Macro and microphysical cloud characteristics affect TB signal
These usedendrite snowflakes
IGARSS July 2011 4
Falling Snow Detection Thresholds
What are the thresholds of detection in terms of IWP or IWC of falling snow?
Analysis Approach:• Use WRF models of Lake Effect and Synoptic snow
•Vertical profiles: IWC, temperature, water vapor profiles•Surface: temperature, land classification, snow depth
• Joint active and passive computations of Z and TB•Use Liu’s 2004 DDA tables for abs, scat, asymmetry, & backscatter
• 11 non-spherical snowflake shapes•Adjust N0 to integrate Liu’s min-max DDA sizes to ensure WRF IWC is preserved
IGARSS July 2011 5
(1) Surface Emissivity Part 1
Urban cropland deciduous evergreen/mixed water
Surface Temperature Vegetation Type Snow Depth
WRF Simulations
Courtesy of W.-K. Tao & team
IWP
(Ja
n 2
0 04
00U
TC
)
IWP
(Ja
n 2
2 06
00U
TC
)
Lake Effect Case Synoptic Snow Case
IGARSS July 2011 6
Radar Calculations
W-Band (-26dBZ) Ka-Band (12dBZ) Ku-Band (18dBZ)
Thresholds of Detection for Falling Snow from Satellite-borne Active and Passive Sensors by G. Skofronick-Jackson, et al., IEEE TGRS, submit 9/11
These use 3-bullet rosette snowflakes
IGARSS July 2011 7
Reflectivities Depend on Particle Shape
W-Band
Ka-Band Ku-Band
IGARSS July 2011 8
Reflectivities Depend on Particle Shape
W-Band
Ka-Band Ku-Band
Ka
IGARSS July 2011 9
Z-Thresholds Depend on Particle Shape Average IWC Detected at Surface
1. Radiometer/DPR Snowfall measurement sensitivities to snow type, rate, surface and tropospheric characteristics
2. Physics of snowfall in the column and relation to extinction characteristics3. Model databases for forward modeling and retrieval development.
Approach: •DFIR instrument clusters (account for measurement uncertainty, mitigate wind, complimentary physics) centered around X/W/Ka-KU/MRR radars and a ground-staring radiometer at CARE site.•Clusters located under C-band/D3R multi-freq/dual-pol radar umbrella; D3R V-point with W and X-bands or cover clusters in scanning/RHI/spectral sampling modes. •Overfly in-situ aircraft in coordination with DC-8 (APR-2 and CoSMIR radiometer);•Pre and post land surface radiative measurements by Ka-Ku and radiometers.
O (60 km)O (10 km)
7-8 km
0.4-0.8 km
Ht.
King City C-band Dual-pol
DFIR Clusters
xGeorgian Bay
CARE D3R
PSD: 2DVD, Parsivel, POSS,SVI
Radar: Ka/Ku,X,W(2),MRR
SWER: Pluvio, Hot Plate
SWE/Depth L-Band + -sensor
(Land/Snow) 10-89 GHz Radiometer
Aircraft: DC-8, Citation
x
IGARSS July 2011 17
Today’s Messages
(1) Falling snow retrievals are complex
Challenges being addressed:• non-spherical particles• surface emissivity
(2) Thresholds of Detection• Theoretical thresholds of detection are promising• Differences between active and passive detection thresholds• Thresholds for passive could be improved with additional
constraints
(3) What matters? IWP, cloud thickness, surface underneath, snow particle shapes and PSD limits, and more
(4) The GCPEx Field Campaign in 2012 will provide data to help address challenges and finalize algorithms.
IGARSS July 2011 18
Questions?
IEEE Geoscience and Remote Sensing Society Administrative Committee (AdCom) Member Voting is open
All GRSS members can vote for new AdCom members
Please vote this week at the GRSS booth or online by Sept. 16, 2011