1. FY09 GOES-R3 Project Proposal Title Page

1
1. FY09 GOES-R3 Project Proposal Title Page

• Title Hurricane Wind Structure and Secondary
  Eyewall Formation
• Project Type 1
• Status Renewal
• Duration 3 years
• Leads
• Chris Rozoff
• NOAA Collaborator
• Jim Kossin
• Other Participants
• Matt Sitkowski (Ph.D. candidate)

2
2. Project Summary

• The overarching goals of this project are
• Uncover relationships between satellite imagery,
  environmental conditions, and the general
  structure and evolution of hurricane surface wind
  fields.
• To better align with the goals of GOES-R Risk
  Reduction, the main directions of this project
  have been shifted shifted into analyzing
  hurricane structure using synthetic data from
  numerical model simulations and Meteosat Second
  Generation (MSG) data.
• Develop objective tools that can diagnose
  hurricane structure from high spectral resolution
  infrared data.

3
3. Motivation/Justification

• Supports NOAA Mission Goal(s) Weather and Water,
  and Homeland Security.
• The occurrence of secondary eyewall formation
  (SEF) and rapid intensification (RI) are often
  associated with highly anomalous intensity
  changes. In these cases, intensity forecasts
  issued by the NOAA National Hurricane Center have
  significant biases.
• We wish to exploit the hyperspectral capabilities
  of GOES-R in improving the resolution of
  hurricane structural evolution such as SEF by
  obtaining better information about the evolving
  convective and mesoscale structures in
  hurricanes. Existing GOES infrared (IR)
  information have already proven highly successful
  in both our physical and operational
  understanding of hurricane structural and
  intensity change processes and we believe the new
  hyperspectral capabilities of GOES-R will improve
  upon these trends.
• Operational knowledge of the size and structure
  of the hurricane surface wind field is essential
  for accurate assessment of the threats posed by
  an oncoming storm. This is particularly true for
  storms approaching land, as storm surge and
  coastal damage due to wind are directly related
  to storm structure. Diagnosing and forecasting
  intensity and track alone does not provide an
  adequate risk assessment. Storms that are far
  from land still pose a threat to marine interests
  and are often not measured by aircraft
  reconnaissance. Even during active
  reconnaissance, there are significant periods
  when the storm is not being directly measured.
  Objective tools that diagnose structure in the
  absence of aircraft reconnaissance can provide
  this missing data in an operational setting.

4
4. Methodology

• Comprehensively explore the utility of the
  spectral channels in Meteosat Second Generation
  (MSG) data to uncover important features related
  to SEF and structure changes that are not
  detected in traditional operational GOES products
• Obtain all possible MSG data for hurricanes and
  process data into storm-centered datasets ready
  for storm-centered analyses.
• Determine the amount of additional information
  from higher resolution IR imagery to better
  detect the structure and microphysics of
  important cloud features.
• Develop new satellite-derived structural
  predictors for operational use
• Test new predictors in existing Bayesian
  statistical forecast schemes developed for SEF
  forecasting with this GOES-R project and one
  developed for RI under a GIMPAP project.
• High-resolution, full-physics WRF data from a
  simulation of Hurricane Katrina (2005) will be
  used to better understand physical connections
  between features in the synthetic hyperspectral
  data and dynamical and microphysical processes
  related to structural evolution.
• WRF-simulated brightness temperatures in the
  GOES-R channels have already been created at
  CIMSS, providing an ideal test bed for exploring
  the relationships between convective and
  mesoscale features in the model and what will be
  measured by GOES-R and existing MSG data.

5
5. Summary of Previous Results

• A North Atlantic and East Pacific climatology of
  SEF has been created. It was found that SEF can
  be related to GOES IR fields as well as
  environmental features. Using these relationships
  we formed a Bayesian Classifier in a highly
  reduced parameter phase-space and the scheme has
  proven to be skillful. In addition, the
  climatology of environmental and structural
  conditions associated with SEF has shed
  considerable insight into the theory and modeling
  of concentric eyewalls.
• A proposal was submitted to and supported the
  NOAA Joint Hurricane Testbed announcement of
  opportunity for the possibility of including the
  SEF forecasting scheme as a NOAA/NHC testbed
  forecast tool.

6
6. Expected Outcomes

• The additional structural and microphysical
  information gained from the higher resolution IR
  channels in the GOES-R proxy (MSG) data should
  improve our climatological understanding of
  important details in hurricanes, improve existing
  intensity change forecasting schemes (i.e., SEF
  forecasting), provide new ways to interpret
  existing theory and modeling data, and offer
  operationally useful algorithms that can be used
  with GOES-R.
• We expect that the synthetic high-resolution IR
  channels in the WRF simulation of Katrina will
  serve as a primary interface for solidifying
  relationships between observed structural
  features and convective/mesoscale processes that
  influence kinematic and thermodynamic structural
  changes.

7
7a. Major Milestones

• FY08
• Apply our newly constructed SEF database to the
  formation of SEF climatology.
• Accomplished.
• Document NOAA/NHC intensity forecast biases for
  SEF cases.
• Accomplished. We completed this task but in this
  case, the 6-hourly smoothed best track data was
  found to be somewhat inadequate for the task. We
  now plan to apply raw aircraft reconnaissance
  data to this important problem.
• Begin exploration of environmental conditions,
  existing GOES imagery, and GOES-R proxy (MSG)
  data as it relates to SEF.
• Accomplished. We completed the initial
  exploration using SHIPS-based environmental
  parameters and GOES-based parameters. The success
  of this exploration indicates that some
  additional exploration into useful parameters
  could lead to more forecast skill of the new
  model.
• Create extended database of wind profiles and
  critical wind radii using our record of GOES
  imagery (1981-2006).
• Accomplished. Construction of the critical wind
  radii has been completed.

8
8. FY09 Accomplishments

• FY09
• Continue data mining toward forming the parameter
  space for the SEF index.
• Begin constructing and testing the Bayes SEF
  classification algorithm.
• Collect all MSG and synthetic WRF data and
  process into storm-centered format.
• Explore how the additional spectral resolution of
  MSG data can be used to better resolve
  convective-scale and mesoscale structures, and
  ultimately latent heating distributions, in
  developing and mature cyclones.
• Use synthetic IR data from Katrina to classify
  spatial and temporal regimes in convective and
  mesoscale convective activity related to well
  resolved physical and intensity change processes
  in model data.

9
8. FY09 Accomplishments

• Bayesian SEF Scheme results completed
• Results are summarized in the publication
• Kossin, J. P., and M. Sitkowski, 2009 An
  objective model for identifying secondary
  eyewall formation in hurricanes. Mon. Wea. Rev.,
  137, 876892.

10
8. FY09 Accomplishments

• Collect all MSG and synthetic WRF data and
  process into storm-centered format
• MSG-SEVIRI data for a case study, Hurricane
  Helene (2006) has been kindly provided by John
  Knaff at CIRA. Included in the provided data set
  are simulated ABI data (Channels 7-16).
• Simulated ABI data and other model output related
  to a WRF simulation of Hurricane Katrina has been
  kindly provided by Jason Otkin at CIMSS.

11
8. FY09 Accomplishments

• Explore how additional spectral resolution of MSG
  data can be used to better resolve
  convective-scale and mesoscale structures, and
  ultimately latent heating distributions, in
  developing and mature cyclones.
• Preliminary analysis has focused on Hurricane
  Helene (2006), a Category 3 storm that produced
  concentric eyewalls following maximum intensity.

SSMI 85 GHz 9/18 2053Z (Image from NRL)
SSMI 85 GHz 9/18 1017Z (Image from NRL)
Channel 8 ABI 9/18 2100Z
Channel 8 ABI 9/18 1015Z
12
8. FY10 Milestones

• Steps to show how additional spectral resolution
  of MSG-SEVIRI data can be used to better resolve
  convective-scale and mesoscale structures in
  Helene, and ultimately latent heating
  distributions, in developing and mature
  cyclones.
• Complete comparison of GOES-12 and MSG-SEVIRI for
  Helene (July-August)
• Compute spatial correlation of microwave
  precipitation and enhanced simulated ABI infrared
  data for all channels (July-August). Involves
  certain adjustments of the data to properly
  compare.
• Retrieval algorithms to classify cloud phase.
  (August-December)
• Develop new satellite-derived predictors.
• Use synthetic ABI data from a Hurricane Katrina
  (2005) WRF simulation to classify spatial and
  temporal regimes in convective and mesoscale
  convective activity related to well resolved
  physical and intensity change processes.
• Results are currently ongoing. Efforts
  specifically aim to compare WRF diabatic heating
  rates with those that may be retrieved from
  simulated ABI data.

13
9. FY10 Milestones

• FY10
• Derive synthetic data from new simulations. We
  currently have idealized WRF simulations with
  concentric eyewalls that could be of great
  benefit to this project.
• Using our physical knowledge gained from the
  proposed milestones mentioned above, we shall
  develop predictors that can be used in
  operational algorithms.
• This involves revisiting the existing GOES
  archives to further investigate the impact of
  shortwave infrared channels, visible channels
  during daylight, and investigating more
  sophisticated pattern recognition techniques not
  used in Kossin and Sitkowski (2009).
• Examine how our existing SEF forecasting scheme
  may be improved and how the RI forecasting scheme
  developed in a GIMPAP project may improve with
  the additional information added from higher
  spectral resolution of cloud structures.

14
8. Funding Profile (K)

• Summary of leveraged funding
• None

15
9. Expected Purchase Items

• FY07
• (50K) NESDIS/STAR CIMSS Grant for 3 people
  (4-time PI Kossin, 1-time PI Velden, 50-time
  graduate student Matt Sitkowski)
• 1.3K Desktop computer/software for Matt Sitkowski
• FY08
• (50K) NESDIS/STAR CIMSS Grant for 2 people
  (15-time PI Kossin, 25-time graduate research
  assistant)
• Personnel support (including benefits, IT
  charges, overhead, etc) 45K
• Contracts N/A
• Software charges N/A
• Equipment N/A
• Travel (two people - 2008 AMS Hurricane
  Conference) - 3K
• Publication Charges - 2K
• FY09
• (50K) NESDIS/STAR CIMSS Grant for 1 person
  (50-time PI Rozoff)
• Personnel support (including benefits, IT
  charges, overhead, etc) 45K
• Contracts N/A
• Software charges N/A
• Equipment N/A
• Travel (two people - CIRA/CSU) - 3K
• Publication Charges - 2K