Contributions from CIRA to the GOESR Satellite Proving Ground

1
Contributions from CIRA to the GOES-R Satellite
Proving Ground
J10.2

• Steve Miller1, Mark DeMaria2, Deb Molenar2,
  Renate Brummer1, Don Hillger2, Ed Szoke1, Arunas
  Kuciauskas3, Joe Turk3, and Hiro Gosden1
• Cooperative Institute for Research in the
  Atmosphere
• Colorado State University, Ft. Collins, CO
• National Oceanic and Atmospheric Administration
• Regional and Mesoscale Meteorology Branch
• Naval Research Laboratory, Monterey, CA
• 16th Conference on Satellite Meteorology and
  Oceanography
• 2009 American Meteorological Society Annual
  Meeting
• Phoenix, AZ

2
Talk Outline

• Proving Ground Overview
• Selected Product Examples
• Technical Interface (AWIPS)
• Initial Users / Needs
• Plans and Summary

Hurricane Katrina as viewed by a GOES-R NPOESS
blended data GeoColor product, to be included
in the Proving Ground.
3
About the GOES-R Proving Ground

• Who Developers at NOAA Cooperative Institutes
  (CIMSS, CIRA) and NASA (SPoRT) working with NOAA
  operational users (NWS, NHC, SPC), NASA Johnson
  Space Center, and other collaborators (e.g., U.
  Alaska-Fairbanks, U. Oklahoma).

• What A project to engage the NWS forecast and
  warning community in pre-operational
  demonstrations of future NOAA GOES-R sensor
  (e.g., ABI, GLM) capabilities.

• Where Application development occurs at
  participant facilities, products are distributed
  nationally to NWS WFOs, interactions occur via
  e-mail/phone/site-visits.

• When The project was initiated in 2008, with
  some participants (CIMSS, SPoRT) having
  already-established activities. The Proving
  Ground is planned to continue through the launch
  of GOES-R in 2015 and potentially beyond.

• Why As part of risk reduction activities for the
  next generation GOES-R series, there is a need to
  ensure Day 1 user readiness for new
  sensor/system capabilities.

• How Use current operational/research-grade
  satellite observing systems and model simulations
  as proxies to GOES-R, demonstrated within the
  operating paradigm of NWS forecasters the
  Advanced Weather Information Processing System
  (AWIPS).

Providing a sustained interaction between the
developers and end users for the purposes of i)
training, ii) product evaluation, and iii)
development based on user feedback.
4
Related Work at CIRA

• Emulation of the NWS Operational Environment
• NOAAPort data ingest
• AWIPS D-2D systems
• Weather Events Simulator

• Experimental GOES-R Risk Reduction Products
• Hazards (fog, smoke, fires, volcanic ash)
• Severe convective weather
• Wintertime weather
• Tropical cyclone track and intensity
• Cloud climatology
• Synthetic imagery

• Ongoing NWS Forecast Office and NCEP
  collaborations
• VISIT, SHyMet and COMET training programs
• GIMPAP program
• Joint Hurricane Testbed
• NCEP Storm Prediction Center Severe Weather
  Testbed
• AWIPS development with OAR/ESRL

5
Online Materials
(!) The Proving Ground is not a web interface.
The information provided online is intended to
serve only as a menu and high-level training
for operational users and the general public.
http//cimss.ssec.wisc.edu/goes_r/proving-ground.h
tml
6
The GOES-R Proving Ground provide the NexSat
project with a conduit to operational forecasters
for demonstration of blended NPOESS/VIIRS
GOES-R / ABI applications.
7
Snow/Cloud Discrimination
Improved detection/discrimination of snow cover
and cloud layers by introducing the GOES-R ABI
1.38 ?m band for cirrus filtering. ? Improved
short term forecasting of (e.g.) nocturnal fog
formation
8
Dust Plume Detection
Improved detection/discrimination of mineral dust
plumes via blue-light absorption techniques
available to GOES-R ABI. ? Improved short term
forecasting of air quality and visibility
conditions
9
Complementary Dust Analysis
Dust Plumes
Current GOES dust products provide a high
time-refresh baseline by which to evaluate many
of the intermittent polar-orbiter products. ?
The aim is to be both instructionally and
operationally relevant.
10
True Color Imagery

• Natural or True color satellite imagery is
  preferred by analysts over panchromatic visible
• Visually intuitive, less ambiguous, higher
  information content (feature recognition).
• Today, broadcast meteorologists attempt poor
  renditions from conventional VIS/IR data.

Standard VIS

• However, the ABI will not provide a standard true
  color capability owing to its lack of a green
  (0.555 ?m) channel.
• There is a workaround we can synthesize the
  green channel to produce an approximation to true
  color imagery.

11
Synthesizingthe Green Band
(MODIS)

• Using training data from MODIS, we can construct
  a look-up-table (LUT) of green expressed as a
  function blue, red and NIR.
• For test data, interpolate the LUT to determine
  green value for current pixel.
• Combine interpolated green with blue, red
  channels to form true color image.

100
B,G,R,NIR
NIR
0
100
R
0
B
0
100
Truth
Gunnison Bay
Gilbert Bay
Great Salt Lake
12
GeoColor
13
Hurricane Katrina in GeoColor
Provides day/night context and proximity of
weather to population centers.
14
Introducing More Layers to GeoColor
GeoColor can be extended to N-layers (e.g., add
dust/aerosol, snow cover, SST/LST fields, ocean
color, etc.) to provide a multi-feature
enhancement utility. ? The Proving Ground allows
us to demonstrate new imagery techniques.
15
Principal Component Image (PCI) Analysis
Volcanic Ash Enhancement
PCI analysis of initial 12-13 July 2008 Okmok
eruption, extracting the dominant image
combinations from the available channel data.
16
Okmok Eruption Via PCI Analysis
Visible Imagery
PCI Enhancement
Several PCIs are combined in this R/G/B image to
enhance the volcanic ash cloud. Clear areas in
the image are deep purple, high water clouds are
mainly green, lower-level water clouds are
yellow, and ash-dominated clouds are orange. ?
Relevant to aviation safety. High priority for
Alaska WFOs.
17
Okmok Eruption Via Blue-Absorption
July 12, 2008 2145 Z
? The Proving Ground allows users to assess the
merits of various algorithms in an operational
setting, while previous future capabilities.
18
The Technical Interface Integrating NexSat
Imagery Products within AWIPS

• Color-mapping of imagery to accommodate display
  limitations within AWIPS.
• Imagery registration (projection, resolution,
  rotation, etc.) must match identically with
  pre-defined AWIPS regions.
• Packaged in HDF with appropriate header
  information.
• Automated processing, transfer, conversion, and
  staging.

19
24-bit to 8-bit Display Reduction

• Many NexSat products are 24-bit depth, but AWIPS
  can display only 8-bit 0,255 depth imagery.
• Requires a reduction of NexSat imagery depth,
  optimal specification of a 256-element color
  palette.
• Photoshop used to define optimal representative
  palette
• Translate each R/G/B pixel to closest palette
  index

20
Mapping Accuracy
Original
21
The Initial Operational User Contacts

• Boulder, CO Weather Forecast Office
• POCs Larry Mooney (MIC) and Eric Thaler (SOO)
• Applications related to aviation (convection,
  fog, etc.)
• Cheyenne, WY Weather Forecast Office
• POCs Melissa Goering (SOO), John Eise (MIC), Ray
  Gomez (ITO)
• Applications related to severe weather, fog, high
  wind events, fires

• Monterey, CA Weather Forecast Office
• POC Dave Reynolds (MIC)
• Applications related to fog, fires, and coastal
  rain/flooding
• Eureka, CA Weather Forecast Office
• POC Mel Nordquist (MIC)
• Applications related to fog, fires, and coastal
  rain/flooding

• Miami, FL Weather Forecast Office
• POC Pablo Santos (SOO)
• Applications related to convection, lightning,
  and tropical storms
• National Hurricane Center
• POC Bill Read (Director)
• Applications related to tropical storm track and
  intensity

22
Plans and Conclusion

• Demonstrate initial set of ABI prototype products
• Selecting those most applicable to WFO-stated
  needs
• Include GOES-R NPOESS blended applications
  (e.g., NexSat)
• Training materials for all demonstration products

• Foster interactions with WFOs
• Establish protocols for product ingest usage
• Include additional offices after proof-of-concept
• Conduct site visits, participate in NWS workshops

• The Research?Operations?Research (R2O2R) cycle
• Modify initial applications based on user
  feedback
• Introduce new applications based on specific user
  needs

The multi-institute/agency GOES-R Proving Ground
team provides users a physical link to algorithm
developers, and in turn, provides developers an
opportunity to transition more relevant and
robust satellite applications to the user
community.