63rd%20IHC%20booklet%20final

CITY OF ST. PETERSBURG Rick Baker, Mayor
Office of the Mayor

Welcome to the 63rd Interdepartmental Hurricane Conference!

I would like to welcome each of you attending the 63rd Interdepartmental Hurricane
Conference in our great city of St. Petersburg. We are pleased that you have come to
our community to hold these important meetings!

While you are here, please take time to enjoy our downtown area and The Pier, Baywalk,
the Salvador Dali museum and the wide variety of restaurants and nightlife all within
walking distance of your hotel. We invite you to stroll around downtown and admire
some of Florida’s best Mediterranean-Revival style architecture and two dozen properties
recorded in the National Historic Register. From our white sandy beaches, to our parks,
to the smiling faces greeting you across our city, we hope you relish your days on the
bay this week.

The citizens of St. Petersburg recognize the important work you do for our city and the
country and we appreciate the unique contributions you make to our safety, security and
preparedness. Thank you for coming to St. Petersburg and we send our best wishes for
a successful conference. Please visit us again soon.

Sincerely,

Rick Baker
Mayor

CITY OF ST. PETERSBURG, P.O. BOX 2842, ST. PETERSBURG, FLORIDA 33731, PHONE (727) 893-7201

S
E
S
S
Session 1 I
The 2008 Tropical Cyclone
Season in Review O
N

1

A Year of Destructive Landfalls:
Overview of the 2008 Atlantic Hurricane Season

Jack Beven and Dan Brown
(John.L.Beven@noaa.gov; Daniel.P.Brown@noaa.gov)

NOAA/NWS Tropical Prediction Center/National Hurricane Center

The 2008 Atlantic hurricane season continued the period of above normal activity that started in
1995. Sixteen tropical storms and one tropical depression formed. Eight of the storms became
hurricanes and five became major hurricanes (category 3 or higher on the Saffir-Simpson
hurricane scale). Four hurricanes (Gustav, Ike, Omar, and Paloma) reached category 4 strength.

Six named storms made landfall in the United States, including three hurricanes. These storms
resulted in an estimated $24 billion in property damages. Elsewhere, the Caribbean islands of
Hispaniola and Cuba were affected by a series of cyclones that caused widespread devastation
and loss of life. The overall death toll from the 2008 Atlantic tropical cyclones is over 700.

Session 1 – Page 1

Overview of the 2008 Eastern North Pacific Hurricane Season

Richard J. Pasch and Eric S. Blake
(Richard.J.Pasch@noaa.gov; Eric.S.Blake@noaa.gov)

NOAA/NWS Tropical Prediction Center/National Hurricane Center

Tropical cyclone activity during the 2008 eastern North Pacific season was below average.
There were sixteen tropical storms of which seven became hurricanes. Two of the hurricanes
became major hurricanes. Although the number of tropical storms was near average, the
numbers of hurricanes and major hurricanes were below average. There were several weak and
short-lived storms, and the Accumulated Cyclone Energy index for 2008 was about 75% of the
long-term median value. Among the highlights of the year were Hurricane Norbert, the strongest
storm on record to strike the western Baja California peninsula, and Tropical Storm Alma, which
was the only tropical storm of record to make landfall along the Pacific coast of Nicaragua.


2008 Atlantic and Eastern North Pacific
Forecast Verification

James L. Franklin
NOAA/NWS/National Hurricane Center

Some highlights from a preliminary verification of official track and intensity forecasts from the
National Hurricane Center during the 2008 season are given below. The performance of the
guidance models is also discussed. As of this writing final numbers are not yet available, but
will be presented at the conference.

For the Atlantic basin, official track errors set records for accuracy at all time periods. Forecast
errors continue their downward trends, while skill was also up in 2008. The official track
forecast skill was very close to that of the consensus models, but was beaten by EMXI. EMXI
and GFDI provided best dynamical track guidance. The UKMET, which performed well in
2007, did not do so in 2008. The performance of NOGAPS lagged again. The HWRF has not
quite attained the skill of the GFDL, but is competitive, and a combination of the two is better
than either alone. The best consensus model was TVCN (a variable consensus that includes
EMXI). It was not a good year for the corrected consensus models.

Official intensity errors for the Atlantic basin in 2008 were below the previous 5-yr means, but
the 2008 Decay-SHIFOR errors were also lower than its 5-yr mean, so there was no real change
in skill. Official intensity errors have remained essentially unchanged over the last 20 years,
while skill has been relatively flat over the past 5-6 years. The best model at nearly every time
period was statistical, although GHMI did well at 4 and 5 days. ICON, a simple four-model
consensus (DSHP/LGEM/HWRF/GHMI) beat everything else, including the corrected
consensus model FSSE.

In the eastern North Pacific, the official track forecast errors set records at 24-72 h. The official
forecast beat the individual dynamical models, and also beat the consensus at 4 and 5 days.
Among the guidance models, the GFDL, HWRF, and ECMWF were strong performers, although
the ECMWF had trouble holding on to systems through 5 days. The benefit of a dynamical
model consensus is larger in the eastern North Pacific than it is in the Atlantic; this is suggestive
of different error mechanisms in the two basins.

For intensity, the official forecast mostly beat the individual models and even the consensus at 12
and 36 h. Official wind biases turned sharply negative at 96-120 h, which was also true in 2007.
The best model at most time periods was statistical, and DSHP provided the most skillful
guidance overall. The HWRF continued to have trouble in this basin, while the four-model
intensity consensus ICON performed very well.


Central Pacific 2008 Hurricane Season Summary

James Weyman
Director, Central Pacific Hurricane Center

There was little tropical cyclone activity in the Central Pacific during the 2008 tropical cyclone
season. The first and only tropical cyclone in Central Pacific Basin for 2008 was Kika which
peaked at minimal tropical storm strength. There were no impacts to Hawaiian Islands or other
Pacific Islands from Kika. The presentation will show possible reasons for this low activity and
will also list some of the changes at CPHC which have occurred in 2008 or scheduled for 2009.


A Review of the Joint Typhoon Warning Center
2008 Tropical Cyclone Season

Mr. Robert J. Falvey
(robert.falvey@navy.mil)

Joint Typhoon Warning Center

A review of the 2008 tropical cyclone activity with discussion of significant events for
the JTWC Pacific and Indian Ocean forecast areas of responsibility will be presented.


53d Weather Reconnaissance Review of the 2008 Hurricane Season

LTC Richard M. Harter
(Richard.Harter@keesler.af.mil)

53d Weather Reconnaissance Squadron

The 2008 hurricane season was an above average year for the 53d Weather Reconnaissance
Squadron with 1150 flying hours (compared to 1062 average flying hours per season over the
last 10 years). Overall, the Stepped Frequency Microwave Radiometer performed reasonably
well on ten WC-130J aircraft, providing critical surface wind speed and rainfall rate data to the
customer. The squadron returned to Guam for the first time since 2003 to participate in the
Tropical Cyclone Structure Experiment. The squadron flew 31 missions (263 operational flying
hours) and launched 750 dropsondes, 250 AXBTs, and 24 buoys in support of the project.


NOAA Aircraft Operations Center (AOC)
2008 Seasonal Summary and Future Plans

Jim McFadden, Jack Parrish
(Jim.d.mcfadden@noaa.gov; Jack.R.Parrish@noaa.gov)

NOAA Aircraft Operations Center

After a relatively slow season in 2007 and with the seasonal forecast calling for above average
storm activity in 2008, AOC was ready with its two WP-3Ds and G-IVSP aircraft to perform
often and long during the summer months. In 2007, the NOAA aircraft flew a combined total of
only 45 flights for 272.7 hours – slightly more than were flown in 2006.

The 2008 season turned out to be much busier but still below the 2005 mark. The G-IV flew 24
flights related to its surveillance missions and two research flights in support of the HRD rapid
intensification experiment on Hurricane Paloma. Total time for this aircraft in the 2008 season
was 192.1 hrs.

The two NOAA P-3s had a little better time of it flying 64 flights for 422.1 hours in support of
the 3-Dimensional Tail Doppler Radar effort organized and directed jointly by HRD and
NOAA’s Environmental Modeling Center, rapid intensification research into Hurricane Paloma
for HRD, the NESDIS Ocean Winds experiment and Ocean Heat Content studies in the Gulf of
Mexico flown under the direction of scientists from the University of Miami’s Rosenstiel School
of Marine and Atmospheric Sciences. These missions were flown out of MacDill AFB in
Florida as well as Barbados into Dolly, AL92, Fay, Gustav, Hanna, Ike, Kyle and Paloma

The NOAA Citation jet and Gulfstream jet prop aircraft also participated in the 2008 hurricane
season effort flying 25 damage assessment digital photography flights for 75.2 hours following
the landfall of Hurricane Ike in Texas.

The total for all NOAA aircraft for the 2008 season was 115 flights and 689.4 hrs.

AOC continues to upgrade its aircraft and instrumentation, and in 2009 expects to accomplish
the following:

 Complete Depot Level Maintenance on one of our WP-3Ds, N42RF
 Complete the installation of the tail Doppler radar on the NOAA G-IV
 Begin testing the new G-IV TDR in storm environment
 Accept delivery of and begin instrument installations on N44RF, NOAA’s 3rd P-3
 Complete Hi-Speed Satellite communications aboard the two current P-3s
 Installation of RVP-8 radar digital signal processor on at least one P-3
 Begin operation of 2nd generation AVAPS dropsonde system
 Begin integration of new aircraft data system (AAMPS)


S
E
S
Session 3 S
Observations and I
Observing Strategies for
Tropical Cyclones and O
their Environment, Part 1 N

3

Examination of the Tropical Cyclone Environment through Comparison of COSMIC with
Other Satellite Data

Christopher M. Hill, Patrick J. Fitzpatrick, and Yee Lau
(hillcm@ngi.msstate.edu)

Northern Gulf Institute, Mississippi State University, Stennis Space Center, MS

Recent work has identified casual relationships between tropical cyclone activity and the
influence of Saharan Air Layers (SALs). Dunion and Velden (2004) devised a method for
detecting SALs through geostationary satellite imagery and discussed the negative influence of
SAL interaction on individual and seasonal cyclone activity. Evan et al. (2006) constructed a
long-term climatology of dust prevalence over the tropical Atlantic Ocean from AVHRR
imagery and showed a correspondence of more dust advected from the Saharan Desert with
diminished Atlantic tropical cyclone activity.
This study explores the utility of data from the Constellation Observing System for Meteorology,
Ionosphere, and Climate (COSMIC) in detecting anomalies of moisture near tropical cyclones.
An indirect, yet highly accurate measure of moisture is obtained through a measure of the
refractivity of Earth-occulting GPS signals traversing the atmosphere and intercepted by the
COSMIC satellites (Anthes et al. 2008). This technique is effective for detecting vertical layers
of anomalous moisture, as are found with SALs. COSMIC refractivity data can be analyzed in
conjunction with aerosol subtype data from NASA’s Cloud-Aerosol Lidar and Infrared
Pathfinder Satellite Observation (CALIPSO) satellite to identify the dry air and desert dust
components, respectively, of SAL regions. Atlantic tropical cyclones observed in 2006, 2007,
and 2008 are analyzed with COSMIC and CALIPSO.
Since COSMIC profiles can sample the atmosphere both in the vertical and horizontal directions,
the data can be used to observe the internal thermodynamic structure of a hurricane or typhoon.
A COSMIC profile was found to transect the inner core of Supertyphoon Sepat 2007.
Microwave imagery from AMSU-B, SSM/I, and TMI sensors, transmitted between 2 and 4 hours
after the COSMIC profile, revealed the development of two concentric eyewalls. When
compared against a baseline COSMIC profile located approximately 800 km east of the storm's
center, the transecting COSMIC profile reveals the high moisture regions of the eyewalls with a
significant positive difference of measured refractivity, and similarly reveals the relatively dry
outer eye region with a near-zero difference of refractivity. This analytical method is applied to
other tropical cyclones at various intensity stages. Other examples of additional cyclone-
transecting COSMIC profiles, and identifiable cyclone features, will be presented and analyzed.
This study also adopts the methodologies of Dunion and Velden (2004) and Evan et al. (2006) in
constructing a summertime climatology of SAL prevalence over the tropical Atlantic Ocean
using the SAL-detecting bands of the GOES-8 and the METEOSAT-8 satellites. The influence
of SAL regions, as detected through the geostationary imagery, is investigated for recent
cyclones and on a seasonal basis. COSMIC, CALIPSO, and geostationary satellite data
depictions of SAL are cross-validated where possible.
Ken Knapp of the NOAA National Climatic Data Center is acknowledged for providing the
GOES and METEOSAT imagery. Funding for this study has been provided by NOAA.


The Hurricane Satellite (HURSAT) Dataset

Kenneth R. Knapp
(Ken.Knapp@noaa.gov)

NOAA/NCDC

Historical satellite data of hurricanes are a valuable asset for understanding hurricane structure.
However, few resources provide satellite observations for tropical cyclones. For example, the
Navy Research Laboratory (http://www.nrlmry.navy.mil/tc_pages/tc_home.html) maintains a
thorough catalog of tropical cyclone imagery since 1998. However, this data cannot be used for
quantitative analysis given it is stored simply as image files. The Hurricane Satellite (HURSAT)
dataset developed at NCDC provides satellite observations from a variety of satellites remapped
and reformatted for portability: http://www.ncdc.noaa.gov/oa/rsad/hursat/
Remapping – Brightness temperatures are remapped from the original satellite swath projection
to an equal-angle projection based on the resolution of the satellite data. A grid is selected
centered on the tropical cyclone position at the time of the satellite scan (interpolated from 6-
hourly best track position) generally with a size of 21º latitude and 21º longitude (approximately
1100 km from the center of the storm).
Satellites – HURSAT data presently use data from the AVHRR, SSMI and geostationary data
worldwide (GOES, GMS, Meteosat, etc.). Future NOAA satellites may be included.
Format – All HURSAT data are provided in CF-compliant netCDF format. This maximizes
portability between operating systems and viewing software. Furthermore, other formats are
available if desired. HURSAT-B1 data are used to produce movie loops (MPEG) of the lifetime
of all tropical cyclones in the period of record: 1978-2007. HURSAT-MW was processed to
create imagery of tropical cyclones using SSMI derived parameters: column water vapor, cloud
liquid water, rain rate and surface wind speed. HURSAT AVHRR has been processed to provide
Objective Dvorak imagery. Plans for HURSAT also include processing all data into Google
Earth-compatible imagery using the KML standard.

Figure – Samples of HURSAT imagery: left) AVHRR BD image of an Indian Ocean tropical
cyclone in 1979, middle) SSMI-derived rain rate from 1993 Hurricane Emily, right) Excerpt
image from an MPEG of 2005 Hurricane Katrina derived from HURSAT-B1.


Using Remotely-Sensed Observations to Describe and Predict
Tropical Cyclone Formation

Elizabeth A. Ritchie1, Miguel F. Piñeros2, and J. Scott Tyo3
(ritchie@atmo.arizona.edu)
1
Department of Atmospheric Sciences, University of Arizona
2
Electrical and Computer Engineering, University of Arizona
3
College of Optical Sciences, University of Arizona

The interpretation of a tropical cyclone’s structure from satellite imagery has been a key element
for structure and intensity estimation. One of the most important examples is the Dvorak
technique, which was developed in the 1970s, and is still used in many tropical cyclone
forecasting centers. This technique provides a source of tropical cyclone intensity estimates over
the tropical oceans. The longevity of the Dvorak technique clearly speaks to the utility of
satellite-based observations for tropical cyclone analysis. An aspect of tropical cyclone
development that can clearly benefit from use of satellite-based observations is the genesis
process. Tropical cyclones, for the most part, develop over oceans where traditional observation
platforms such as rawinsonde stations cannot be located. Therefore, the best opportunity for
locating and discriminating those cloud clusters that will develop into tropical cyclones from
those that will not is from analysis of satellite-based observations.

This document introduces an objective technique, based on brightness temperature observations,
which aims to discriminate those tropical cloud systems, which will develop into tropical
cyclones from those that will not. The basis of the technique is that the convective structures
embedded in cloud clusters become more axisymmetric as the wind field of the disturbance
organizes and intensifies. To quantify this evolution, two tasks are required: 1) the center of the
system needs to be located in an objective manner, and 2) the departure from axisymmetry of the
weather system needs to be quantified. Time series of the combined results of these two tasks are
used to examine the evolution of tropical disturbances.

In this presentation results from storms that developed in the Atlantic will be shown. Time series
of the departure from axisymmetry are shown to discriminate between developing and non-
developing cloud clusters with a high rate of success and low false alarm rate. The technique has
utility as a stand-alone system to choose the cloud clusters that will go on to develop, but also
has great potential to be used as the early-detection part of a larger intensity algorithm.


Moving Toward an Operational Satellite Ocean Surface Vector Winds Capability with a
Dual Frequency Scatterometer

Michael Brennan1, Paul Chang2, Zorana Jelenak2, Richard Knabb3, and Joseph Sienkiewicz4
(Michael.J.Brennan@noaa.gov)
1
NOAA/NWS/NCEP/National Hurricane Center, 2NOAA/NESDIS/STAR,
3
NOAA/NWS/Central Pacific Hurricane Center, 4NOAA/NWS/NCEP/Ocean Prediction Center

Remotely sensed ocean surface vector wind (OSVW) measurements continue to be an important
source of data for operational tropical cyclone (TC) analysis and forecasting at the National
Hurricane Center (NHC) and the Central Pacific Hurricane Center (CPHC). These data are also
used heavily for marine analysis and forecasting applications elsewhere within in the National
Weather Service (NWS) at NCEP centers (NHC, OPC) and coastal weather forecast offices
(WFOs).

Currently, OSVW data available in NWS operations are from the NASA QuikSCAT
scatterometer and the Advanced Scatterometer (ASCAT) onboard EUMETSAT’s METOP-A
satellite. Despite the documented shortcomings of the retrievals associated with these
instruments, they have had a positive impact on NWS tropical cyclone and very positive impact
on marine forecast and warning operations. However, QuikSCAT is well beyond its planned
mission life, and ASCAT wind retrievals are inferior to those from QuikSCAT in terms of
coverage and performance at high wind speeds. To move toward an operational satellite OSVW
capability that will meet NOAA’s operational needs, NOAA and NASA JPL are exploring a
partnership with the Japanese Space Agency (JAXA) to fly a Dual Frequency Scatterometer
(DFS) onboard the second mission in the JAXA GCOM-Water cycle (GCOM-W) series,
currently scheduled to launch in 2016. With this launch date, funding must begin in FY2011 to
meet JAXA-imposed deadlines for instrument delivery.

This partnership offers the opportunity for international collaboration to significantly limit
mission while substantially improving the quality of satellite OSVW retrievals available for
NWS tropical cyclone and marine operational warning and forecast applications. While DFS
would not achieve all of the proposed capabilities of the Extended Ocean Vector Winds Mission
(XOWVM), DFS would utilize C-band and Ku-band retrievals to mitigate rain contamination,
which would greatly improve the quality of wind retrievals in the TC environment. DFS would
also offer increased spatial resolution compared to QuikSCAT and provide similar spatial
coverage.

The design of DFS will be discussed, along with the potential benefits of partnership with JAXA.
Simulated retrievals from DFS in the TC environment will be presented. The quality of the DFS
retrievals will be compared and contrasted with simulated retrievals from QuikSCAT and
XOVWM for TC center fixing/identification, intensity (maximum wind) analysis, and the
analysis of the outer TC wind field (wind radii). Preliminary results suggest that DFS retrievals
in the TC environment provide a much more accurate representation of the wind field relative to
QuikSCAT. Finally, a path forward will be outlined to work toward establishing an operational
satellite OSVW capability for NWS operations.


Developing GOES-R Tropical Cyclone Products via Proxies

John A. Knaff1, Donald W. Hillger1, Mark DeMaria1, James J. Gurka2
(John.Knaff@noaa.gov)
1
NOAA/NESDIS/StAR, Fort Collins, Colorado
2
NOAA/NESDIS/OSD, Suitland, Maryland

The next generation GOES satellites (beginning with GOES-R) will include the Advanced
Baseline Imager (ABI) with vastly improved spectral, spatial and temporal resolution relative to
the current GOES I-M series satellites. It will also include the Geostationary Lightning Mapper
(GLM) which, together with the ABI, offers the potential to significantly improve the analysis
and forecasts of tropical weather phenomenon. One of the distinct advantages of GOES-R is the
high temporal resolution that is possible from geostationary orbit. Such capabilities are ideal for
studying/analyzing tropical weather phenomenon, which often evolve quickly and are poorly
sampled by both conventional observations and low-earth orbiting satellites.

This presentation will provide a brief summary of the GOES-R capabilities and describe CIRA
research to develop tropical cyclone analysis and forecast applications using proxy data. Imagery
from the SEVERI instrument on the Meteosat Second Generation (MSG) satellite series is used
as proxy for the ABI and data from the ground-based World Wide Lightning Locator Network
(WWLLN) is a proxy for the GLM. Algorithms have been developed to simulate many of the
ABI channels from combinations of SEVERI imagery. The simulated ABI data will be used to
demonstrate a number of new products for analysis of the storm structure and storm
environment.

Although the WWLLN data only detects the very high current cloud to ground lightning strikes,
it can still capture the basic lightning patterns in tropical cyclones and their environments. The
lightning density (strikes per unit area and time) was calculated for all Atlantic tropical cyclones
from 2005-2008 in a storm-relative cylindrical coordinate system. The WWLLN lightning
density is calibrated by comparison with the annual climatology from the OTD/LIS instruments
on low-earth orbit satellites. This calibration helps to correct for variations of WWLLN network
between 2005 and 2008, and accounts for its low detection rate. Results show that the lightning
density has potential to improve statistical tropical cyclone intensity prediction, and is a good
discriminator of rapid intensity change, provided that the vertical wind shear is accounted for.
The relationship between lightning density and intensity change and rapid intensification are
much stronger in low wind shear regimes.

DISCLAIMER: The views, opinions, and findings in this report are those of the authors and
should not be construed as an official NOAA and/or U.S. Government position, policy, or
decision.


Measuring the Three-Dimensional Structure of Hurricanes with a Microwave Sounder
Bjorn Lambrigtsen
(lambrigtsen@jpl.nasa.gov)
Jet Propulsion Laboratory, California Institute of Technology
Analysis of data from hurricane aircraft experiments has shown that there is a very high
correlation between microwave sounder observations and radar observations of convective
systems. The correlation is so strong that it is possible to develop a model function that can be
used to transform the radiometric brightness temperatures into vertical profiles of equivalent
radar reflectivity. This new method makes it possible to use current and future satellite sounders
to generate proxy precipitation and cloud radar data. This in turn will enable a wide range of
analyses of convective systems and processes using algorithms and methodologies developed for
the radar systems.
We show some preliminary results from the joint NASA-NOAA Tropical Cloud Systems and
Processes (TCSP) field campaign in 2005. In that experiment NASA deployed a suite of remote
sensing instruments on the high-altitude ER-2 aircraft that included a precipitation radar system
and a microwave sounder. The sounder, the High Altitude MMIC Sounding Radiometer
(HAMSR), was developed at the Jet Propulsion Laboratory under the NASA Instrument
Incubator Program with new technology and is functionally similar to the Advanced Microwave
Sounding Unit (AMSU) that is now operating on several NOAA weather satellites. Comparisons
of HAMSR observations with those from the ER-2 Doppler Radar (EDOP) shows that height
resolved EDOP-equivalent radar reflectivity can be derived from the HAMSR brightness
temperatures with good accuracy, although at lower vertical resolution than is possible with the
radar. In addition, since HAMSR is a cross-track scanning sensor (like AMSU), reflectivity can
be estimated for the entire 3D volume of the atmosphere observed by HAMSR – unlike EDOP,
which is a nadir-only sensor. With this new method HAMSR can be used to map out the
convective structure of tropical convection from the surface to 15 km with a vertical resolution
of roughly 1-2 km.
Efforts to extend this method to the AMSU satellite instruments are under way, and preliminary
results are very promising. This will open up a new avenue for hurricane analysis. In particular,
it will be possible to get a picture of the internal structure of hurricanes as the sensors pass
overhead. Even though the spatial resolution of the satellite sensors is relatively poor, 15-50 km,
the additional information – which is otherwise only available from a very few radar systems – is
expected to be of high value for nowcasting and forecasting as well as retrospective analysis.
This approach will be of particular interest when applied to a geostationary microwave sounder,
with its ability to monitor hurricanes continuously throughout their life cycles. Such a system is
now under development in response to the National Academy of Sciences’ recommendation, in
its recent “decadal survey” of earth satellite missions, to develop the Precipitation and All-
weather Temperature and Humidity (PATH) mission. Key technology required for such a sensor
has been developed at the Jet Propulsion Laboratory, and the “Geostationary Synthetic Thinned
Aperture Radiometer” (GeoSTAR) is ready for implementation. A possible joint NASA-NOAA
mission, perhaps flying GeoSTAR as a demonstration payload on one of the new GOES-R/S/T
satellites, is being explored.

Copyright 2008 California Institute of Technology. Government sponsorship acknowledged.


Arc Clouds in the Tropical Cyclone Environment

Jason P. Dunion1, Christopher S. Velden2, Jeffrey D. Hawkins3
(Jason.Dunion@noaa.gov)
1
NOAA/AOML/Hurricane Research Division
2
University of Wisconsin-CIMSS
3
NRL- Monterey

Arc clouds are common features in mid-latitude thunderstorms and MCSs. They denote the
presence of a density current that forms when dry middle level (~600-800 hPa) air has interacted
with precipitation. The convectively-driven downdrafts that result reach the surface/near-surface
and spread out from the convective core of the thunderstorm. We hypothesize that the mid-level
moisture found in the moist tropical North Atlantic sounding described by Dunion (2009) is
insufficiently dry to generate extensive arc clouds around an African easterly wave (AEW) or
tropical cyclone (TC). However, substantial (100s of km in length and lasting for several hours)
arc clouds consistently form in the tropics in the periphery of these tropical disturbances.
Dunion (2008) did described two addition types of air masses that are frequently found in the
tropical North Atlantic and Caribbean that could effectively initiate the formation of large arc
clouds: the Saharan Air Layer and mid-latitude dry air intrusions. Both of these air masses were
found to contain substantially dry air (~50% less moisture than the moist tropical sounding) in
the middle levels of the atmosphere and can affect the tropical North Atlantic and Caribbean
throughout the summer months. We hypothesize that these arc cloud features can significantly
impact an AEW or TC (particularly smaller, less developed systems) via two mechanisms: 1) as
the arc clouds race away from the convective core region, they create low-level outflow in the
quadrant/semicircle of the AEW or TC in which they form. This outflow pattern counters the
typical low-level inflow that is vital for TC formation and maintenance and 2) arc clouds are
formed when cool, dry air generated by convectively-driven downdrafts rapidly sinks to the
surface/near-surface and races away from the AEW/TC’s convective region. This combination
of downdrafts and cool, dry air help stabilize the middle to lower troposphere and may even act
to stabilize the boundary layer. Examples of arc cloud events around several tropical cyclones
will be discussed and GPS dropsonde observations from a few of these events will be shown.


S
E
S
Session 4
Observations and
S
Observing Strategies for I
their Environment,
Part 2
N

4

A New Approach for Continuous Mapping of Tropical Cyclone ABL Winds to the Surface
from Research and Reconnaissance Aircraft

James Carswell1, Paul Chang2, Stephen Frasier3
(carswell@remotesensingsolutions.com)
1
Remote Sensing Solutions, Barnstable, MA
2
NOAA/NESDIS/STAR, Camp Springs, MD
3
University of Massachusetts, Amherst, MA

Accurate knowledge of the lower atmospheric boundary layer (ABL) winds is a crucial
component in providing accurate intensity forecasts of tropical cyclones and improving our
understanding of these severe weather systems. Airborne Doppler radars have been used to
measure the ABL winds by obtaining multi-look Doppler profile observations of precipitation
within these storms. Using precipitation as a tracer of the winds, and applying velocity-azimuth-
display (VAD) retrieval techniques to the mean Doppler measurements, the vertical profile of the
horizontal ABL winds can be retrieved. However, as the radar views the volume backscatter in
the marine layer and lower atmospheric boundary layer, the Doppler measurements are
contaminated by ocean surface backscatter because of the finite antenna beamwidth. Depending
on the radar’s measurement geometry and aircraft altitude, this contamination can begin to occur
at altitudes from 200 m to 1000 m and continue all the way down to the surface. The net result is
that the mean Doppler estimates have a significant bias which voids the ABL wind retrievals at
these lower altitudes.

Remote Sensing Solutions (RSS), NOAA / NESDIS / STAR and the University of Massachusetts
(UMass) have obtained an extensive set of multi-look, dual wavelength Doppler measurements
of the atmospheric boundary layer (ABL) within the precipitation bands of tropical cyclones.
RSS has developed a new retrieval process that separates the volume and surface backscatter
contributions and allows the ABL winds to be fully profiled to the surface. Further, it overcomes
limitations associated with ocean vector wind scatterometry, permitting the ocean vector winds
to be accurately mapped in the presence of precipitation. This new technique has been applied to
measurements obtained by the Imaging Wind and Rain Airborne Profiler (IWRAP). This
instrument has been renamed to the Advanced Wind and Rain Airborne Profiler (AWRAP) as a
result. Note its sensitivity has also been improved by approximately 15 dB. The measurements
demonstrating contamination free ABL wind profiles collected to the surface will be presented
along with a discussion of the steps we are taking to realize this approach operationally. In the
future, this new technique will enable the NOAA WP-3D aircraft, and potentially the Air Force
aircraft, to provide continuous high resolution (30 m vertical and less than 500 m along / cross
track) mapping of the ABL wind field to the surface as these aircraft transect tropical storms and
cyclones.


Status of the HIWRAP and URAD Conical Scan Radars for Wind Measurements

Gerald Heymsfield1, L. Li2, J. Carswell3, D. Schaubert4, J. Creticos4, M. Vega1
(gerald.heymsfield@nasa.gov)
1
NASA Goddard Space Flight Center
2
University of Maryland, Baltimore
3
Remote Sensing Solutions
4
University of Massachusetts, Amherst

Wind measurements are crucial for understanding and forecasting tropical storms since they are
closely tied to the overall dynamics of the storm. The High-Altitude Imaging Wind and Rain
Airborne Profiler (HIWRAP) is a dual-frequency (Ka- and Ku-band), dual-beam (300 and 400
incidence angle), conical scan, solid-state transmitter-based system, designed for operation on
the high-altitude (20 km) Global Hawk UAV. With the inclusion of Ka-band, HIWRAP will be
able to image the winds through volume backscattering from clouds and precipitation, enabling it
to measure the tropospheric winds above heavy rain at high levels. It will also measure ocean
surface winds through scatterometry, similar to QuikScat. These measurements from higher
altitudes above storms, will be useful for providing higher spatial and temporal resolution than
obtained by current satellites and lower-altitude instrumented aircraft. HIWRAP is currently in
the final stages of development and first flights are planned later in the year on the NASA WB-
57 manned aircraft (60-65kft ceiling). The UAV Radar (URAD) is an X-band conical scan radar
based on a more conventional TWT (tube) design. It is single beam with 2-axis scan so the
incidence angle is adjustable with a maximum value of about 400. It’s current target is a
replacement for the EDOP radar on the NASA ER-2 high-altitude aircraft, but it was designed to
go in the NASA Global Hawk. The presentation will discuss the status of HIWRAP and URAD
and their potential use in hurricanes research, and plans for test flights of HIWRAP on the WB-
57 and Global Hawk.


Results of First Flight Tests of the Wide Swath Radar Altimeter (WSRA) During the 2008
Hurricane Season

Ivan PopStefanija
(popstefanija@prosensing.com)

ProSensing Inc.

Through a NOAA Phase II SBIR Contract, ProSensing Inc. has developed a Wide Swath Radar
Altimeter (SRA) intended for airborne measurement of directional ocean wave spectra. The
WSRA is a novel solid-state digital beamforming radar, designed to replace the aging prototype
Scanning Radar Altimeter (SRA) recently retired by NOAA. The new design includes a flat
panel digital beamforming antenna, which allows the radar to achieve the same beamwidth as the
NASA prototype 36 GHz SRA while operate at less than half the frequency. The ability to
operate at 16 GHz is an important feature of the WSRA, since attenuation due to rain increases
rapidly with increasing frequency. This allows the system to operate through heavy rain at higher
altitudes than the retired SRA.

During August and September of 2008, the WSRA collected data on six hurricane missions aboard
the NOAA AOC P-3. Table 1 shows the flight parameters for each of these missions and the data
volume collected by WSRA.

Table 1 WSRA data collection record during the 2008 hurricane season

Storm Name Takeoff Date/Time Duration Data
UTC (hrs) Collected
Test Flight 05AUG08 / 14:00 3 250GB
(Calm Ocean)
Tropical Storm Fay 18AUG08 / 00:00 8 325GB
Hurricane Gustav 01SEP08 / 08:00 8 350GB
CAT 1
Hurricane Ike 10SEP08 / 08:00 8 425GB
CAT2 and CAT3 11SEP08 / 08:00 8 420GB
12SEP08 / 08:00 8 470GB

WSRA data was collected over a wide range of ocean surface conditions, from calm seas to
category 3 hurricanes. Data gathered on the first mission over a clam ocean were used to
calculated antenna phase calibration coefficients. These coefficients are used to correctly focus
the synthesized antenna beams when performing the digital beamforming algorithm. Preliminary
results show that the WSRA has expanded the operational range of the measurement conditions
over the retired SRA system. WSRA obtained usable signal from a significantly higher altitudes
(12,500 feet vs 5,000 feet). Also, the WSRA signal was not significantly attenuated even under
high rain rate conditions often found in hurricanes. Qualitative assessment of ocean wave
spectra computed from WSRA data showed reasonable values for significant wave height and
wave direction.


Operational Use of Near-Real-Time Sea Surface Directional Wave Spectra
from NOAA Wide Swath Radar Altimeter (WSRA) Range Measurements

Edward J. Walsh

NASA Goddard Space Flight Center

The Wide Swath Radar Altimeter (WSRA) was built by ProSensing under the NOAA SBIR
program to be an operational replacement for the NASA Scanning Radar Altimeter (SRA). It
had its initial test flight on N43RF on August 5, 2008, and subsequently collected data in T.S.
Fay on August 18, Hurricane Gustav on September 1, and Hurricane Ike on September 10, 11
and 12. The WSRA is two orders of magnitude more complex than the SRA. It transmits a long,
chirped pulse whose range extent at 3 km height would be 2900 m whereas the SRA pulse range
extent was only 0.9 m. This allows the WSRA to have better signal level with 10 W peak power
than the SRA had with 1.5 KW peak power. The SRA used a dielectric lens to form a narrow
beam which was scanned mechanically by a rotating mirror. The WSRA sequentially transmits
its chirped pulse on each of 62 slotted wave guide elements which produce narrow beams in the
along-track direction, but fan beams in the cross-track direction. Those 62 return pulses are
dechirped to provide range resolution and then combined coherently with various phase shifts to
produce 80 narrow beams in the cross-track direction. Because of the WSRA complexity,
ProSensing made the prudent decision to record all the raw data during the 2008 season to be
able to identify and correct any unanticipated problems in the hardware. At 2,400 m altitude the
WSRA recorded 1 Giga Byte of raw data every 30 s, limiting data acquisition to about 3.5 hours
per flight. A hardware problem was discovered which caused either the radar range
measurements or the beam angles to deviate from their nominal values and produce a mean
curvature in the cross-track variation of the sea surface. When ad hoc corrections were made to
the WSRA nominal range quantization, excellent sea surface topography was produced whose
wave height and directional wave spectrum were in good agreement with Buoy 42001. The
hardware problem should be resolved soon so that all data from the 2008 season can be
processed in a routine fashion to confirm the system performance and eliminate the need to
record raw data during the 2009 season, enabling wave topography and directional wave spectra
to be produced during the flights.


Transmission of Airborne Doppler Data and Analyses to the Ground from the NOAA WP-
3D Aircraft During the 2008 Hurricane Season:
Quality Control, Analysis and Superobs.

John Gamache1, Peter Dodge1, Frank Marks, Jr.1, Fuqing Zhang2, Yonghui Weng3
(john.gamache@noaa.gov)
1
2
Pennsylvania State University
3
Texas A&M University

For the high-resolution three-dimensional wind observations collected by the NOAA WP-3D
aircraft (and soon the NOAA Gulfstream IV aircraft) to be useful for operations, they must be
quality controlled and transmitted to the ground as quickly as possible during flight. Analyses of
the observations may also be sent to the ground. Since the 2004 Hurricane Season, quality
control and analyses of airborne Doppler have been performed on the aircraft and in 2005 the
first analysis was sent to the National Hurricane Center. Since 2005 these analyses have been
made available (usually in real-time) on an AOML ftp site that is polled by computers at NHC.
Since that season efforts have been made to improve quality control, and to make the software
more accessible to a number of scientists that fly aboard the NOAA P3s, so that the analyses and
QC processing are done in real time. Possibly in 2009, airborne Doppler data will be used to
initialize a real-time parallel run of HWRF, leading, eventually to the assimilation in the real-
time official run of HWRF.

In 2008, a further step was made in the progress toward including data immediately into
numerical simulations, as a research, high-resolution, non-operational run of WRF-ARW was
run using airborne Doppler observations transmitted during the flight to the ground. The
observations were first quality controlled, and then “superobbed” using software developed by
Yonghui Weng and Fuqing Zhang. The observations were then transmitted via ftp AOML,
where they were polled and then assimilated using Ensemble Kalman Filtering techniques, into
the WRF-ARW model. Examples of the influence of the superobs in changing the initial and
early structure of model runs will be shown.

We will also discuss major challenges that face us in 2009. These include the modification of the
software to read new output from the Sigmet RVP-8 processor, that should be running on one P3
during the 2009 hurricane season, and the transmission of data in real time to NCEP Central
Operations.


First Results from a High Altitude Stepped Frequency Microwave Radiometer

Alan S. Goldstein1, Dr. Eric Uhlhorn2
(alan.s.goldstein@noaa.gov, eric.uhlhorn@noaa.gov)
1
NOAA/Aircraft Operations Center, Tampa, FL
2
NOAA/Hurricane Research Division, Miami, FL

The Stepped Frequency Microwave Radiometer (SFMR) is an instrument used to measure wind
speed at the ocean surface. While installed and used operationally for several years on NOAA
and Air Force 'Hurricane Hunter' reconnaissance aircraft, a high-altitude version of the SFMR
was first installed on NOAA's G-IV jet aircraft in the summer of 2008. This presentation
discusses that instrument, including verification of operation, calibration, and data comparison
with in-situ surface measurements in a variety of wind conditions.

The SFMR derives surface wind speed by measuring the emissivity of the ocean over a range of
C-Band frequencies (4.7-7.2 GHz). A special version of this unit was developed with a larger
antenna so that the areal coverage from high altitude (40,000+ ft) is similar to a normal unit at
half of that height. The high altitude unit was installed on the NOAA G-IV high-altitude jet
aircraft in the summer of 2008 and data was collected during surveillance missions, starting with
Hurricane Ike. In addition, a dedicated calibration mission was flown, using a similar
methodology to NOAA P-3 SFMR calibration flights.

The presentation will briefly review basic SFMR theory and describe the G-IV installation. It
will then discuss the calibration accomplished and verification of the unit's performance in
comparison to dropwindsonde data and other surface wind measurements. Finally, it will
highlight any remaining concerns in the instrument's performance and provide a plan for the next
steps in transitioning this instrument to operational status.


Simulation of the Impact of New Aircraft- and Satellite-Based Ocean Surface Wind
Measurements on Estimates of Hurricane Intensity

Eric Uhlhorn1, Robert Atlas2, Peter Black3, Courtney Buckley4, Shuyi Chen5 Salem El-Nimri6,
Robbie Hood7, James Johnson6, Linwood Jones6, Timothy Miller4, Chris Ruf8
(Eric.Uhlhorn@noaa.gov)
1
2
NOAA/Atlantic Oceanographic and Meteorological Laboratory (AOML)
3
Science Applications International Corporation
4
NASA/Marshall Space Flight Center
5
University of Miami/RSMAS
6
University of Central Florida
7
NOAA/Unmanned Aircraft Systems Program
8
University of Michigan

The Hurricane Imaging Radiometer (HIRAD) is a new airborne microwave remote sensor
currently under development to enhance real-time hurricane ocean surface wind observations.
HIRAD builds on the capabilities of the Stepped Frequency Microwave Radiometer (SFMR),
which now operates on NOAA WP-3D, G-IV, and AFRC WC-130J aircraft. Unlike the SFMR,
which measures wind speed and rain rate along the ground track directly beneath the aircraft,
HIRAD will provide images of the surface wind and rain field over a wide swath (~3 times the
aircraft altitude). To demonstrate potential improvement in the measurement of peak hurricane
winds, we present a set of Observing System Simulation Experiments (OSSEs) in which
measurements from the new instrument as well as those from existing platforms (air, surface, and
space-based) are simulated from the output of MM5 high-resolution (~1.7 km) numerical model.
Simulated retrieval errors due to both instrument noise as well as model function accuracy are
considered over the expected range of incidence angles, wind speeds and rain rates. Based on
numerous simulated flight patterns and data source combinations, statistics are developed to
describe relationships between the observed and true (from the model’s perspective) peak wind
speed. These results have implications for improving the estimation of hurricane intensity (as
defined by the peak sustained wind anywhere in the storm), which may sometimes go un-
observed due to sampling limitations from observations along aircraft flight lines.


S
E
S
Session 5
Observations and
S
Observing Strategies for I
their Environment,
Part 3
N

5

Evolving Oceanic and Atmospheric Boundary Layer Measurements During Hurricanes
Gustav and Ike

Lynn K. Shay1, Eric Uhlhorn2, Rick Lumpkin3, Patrick Meyers1, Benjamin Jaimes1, Jodi
Brewster1, George Halliwell1
(nshay@rsmas.miami.edu, eric.uhlhorn@noaa.gov, rick.lumpkin@noaa.gov,pmeyers@rsmas.miami.edu,
bjaimes@rsmas.miami.edu, jbrewster@rsmas.miami.edu, ghalliwell@rsmas.miami.edu)
1
Division of Meteorology and Physical Oceanography, RSMAS, University of Miami
2
Hurricane Research Division, NOAA Atlantic Oceanographic and Meteorological Laboratory
3
Physical Oceanography Division, NOAA Atlantic Oceanographic and Meteorological Laboratory

During the 2008 hurricane season, hurricanes Gustav and Ike moved over the Gulf of Mexico
and interacted with the Loop Current (LC) and the mesoscale eddy field. As part of the National
Center for Environmental Prediction (NCEP)-directed tail Doppler Radar Missions, concurrent
oceanic and atmospheric measurements were acquired from sixteen NOAA WP-3D research
flights for pre, during and post-storm conditions. Such coupled measurements are absolutely
necessary to improve models at the National Centers.

Briefly, over two-hundred global positioning system (GPS) sondes were deployed during the in-
storm flights across the Gulf of Mexico to document the evolving atmospheric structure over
warm and cool ocean features. Each research flight deployed airborne expendable
bathythermographs (AXBT) to document the evolving upper ocean thermal structure across the
entire Gulf of Mexico for the first time. In support of operational modeling efforts at NCEP,
more than four-hundred AXBTs were deployed on these research flights to document the
coupled boundary layer responses. To complement these data, twenty-one drifters (10 with 150-
m thermistor chains) and floats were deployed by the Air Force WC-130J northwest of the LC.
Over this array, forty-five GPS sondes were deployed and surface winds were mapped from
Stepped Frequency Microwave Radiometer (SFMR). These data are being processed to assess
data quality and the levels of the observed ocean thermal response in the LC and Gulf Common
Water, and to relate upper ocean changes to the evolving boundary layer structure.

This prototype experimental approach will be conducted during the 2009 hurricane season from
research aircraft as part of the NOAA Hurricane Forecasting Intensity Program Experiment and
Minerals Management Service (MMS) sponsored Dynamics of the Loop Current Study. Several
MMS-sponsored moorings will be deployed for about 30 months this summer in the LC-eddy
shedding area. This region is where both hurricanes Katrina and Rita rapidly deepened to
category-5 status in 2005. In addition to GPS sondes and SFMR measurements, ocean
measurements this summer will involve the deployment of aircraft expendable current,
temperature and salinity profilers (AXCPs, AXCTDs) prior, during and subsequent to hurricane
passage over the eddy shedding area and possibly drifters and floats in the warm eddy and the
Gulf Common Water. These measurements will provide signals to examine the oceanic response
and assess its’ feedback to the hurricane boundary layer aimed at improving coupled forecast
models.


Flight Test Results of a Differential Absorption Microwave Radar for Remote Sensing of
Atmospheric Pressure

Dr. Roland Lawrence1, Mr. Steve Harrah2, Dr. Bing Lin2, Ms. Patrica Hunt3,
Mr. Carl Lipp4
(rlawrenc@odu.edu)
1
Old Dominion University
2
NASA Langley Research Center
3
Lockheed Martin
4
ATI

The accuracy of numerical weather model predictions of the intensity and track of tropical
storms may be improved by large spatial coverage and frequent sampling of sea surface
barometry. Over oceans, barometric pressure measurements are usually provided by a limited
number of in-situ observations conducted by buoy stations and oil platforms. Surface pressure
data is also available via manned aircraft deploying dropsondes and provides useful spatial
sampling with limited coverage. To improve predictions and forecasts of the intensity and track
of tropical storms, large spatial coverage and frequent sampling of sea surface barometry are
critically needed for use in numerical weather models.

Work underway at NASA Langley Research Center (LaRC) will assess the feasibility of an
active microwave radar working at moderate to strong O2 absorption bands in the frequency
range of 50~56 GHz for surface barometric pressure remote sensing. At these radar wavelengths,
the reflection of radar echoes from water surfaces is strongly attenuated by atmospheric column
O2. Because of the uniform mixture of O2 gases within the atmosphere, the total atmospheric
column O2 is proportional to atmospheric path lengths and the total atmospheric column air, and
thus, to surface barometric pressures. Recent research has developed a technique based on the
use a dual-frequency, O2-band radar to overcome many of the difficulties associated with
techniques requiring larger frequency separation. The technique uses dual wavelength channels
with similar water vapor and liquid water absorption characteristics, as well as similar footprints
and sea surface reflectivities, because of the closely-spaced spectra. The microwave absorption
effects due to liquid water, water vapor in the atmosphere and the influences of sea surface
reflection should be effectively removed by use of the ratio of reflected radar signals of the two
channels.

LaRC has developed a prototype Radar based on this measurement technique to estimate
barometric pressure and has recently completed several flight tests of the radar.. This paper will
present an overview of the differential absorption measurement concept and the radar instrument
concept. Further, results of instrument functional testing and initial flight-testing over the
Chesapeake Bay will be presented. Finally, technology issues with respect to developing higher
altitude or spacecraft instruments based on this technique will be discussed.


WISDOM System Description and Initial Test Results

Justyna Nicinska and Alexander E. MacDonald
NOAA Oceanic and Atmospheric Research

NOAA with support from DHS Science and Technology and a consortium of industrial,
academic and other government partners are developing and testing a new system to improve
upper air data in areas that are poorly observed. The Weather In Situ Deployment Optimization
Method (WISDOM) system is designed around the availability of small (about 1-2 meter) super-
pressure balloons, and the availability of small (100 gram) electronics which include over the
horizon Global Positioning System and satellite radio capabilities. The concept is that large
numbers of WISDOM balloons with the GPS payload could be released to optimize weather
prediction at a future time for phenomena of interest. Advanced techniques of assimilation and
modeling are used to determine the release locations of the balloons to optimize their trajectories
to improve future numerical weather prediction. An initial test of the WISDOM system that was
conducted in fall 2008 aimed at late-season tropical cyclone prediction in the Atlantic basin will
be discussed.


Coordinated use of Targeted Observations to Improve Tropical Cyclone Track Forecasts

Sharanya Majumdar1, Sim Aberson2, Pat Harr3, Sarah Jones4, Rolf Langland5, Tetsuo
Nakazawa6, Melinda Peng5, Carolyn Reynolds5, David Richardson7, Chris Velden8, Martin
Weissmann9, Chun-Chieh Wu10, Munehiko Yamaguchi11,
(smajumdar@rsmas.miami.edu)
1
Rosenstiel School of Marine and Atmospheric Science, University of Miami, FL
2
NOAA/AOML Hurricane Research Division, FL
3
Naval Postgraduate School, Monterey, CA
4
Universität Karlsruhe, Germany
5
Naval Research Laboratory, Monterey, CA
6
Meteorological Research Institute, Japan
7
ECMWF, United Kingdom
8
CIMSS/University of Wisconsin, WI
9
DLR Oberpfaffenhofen, Germany
10
National Taiwan University, Taiwan
11
Japan Meteorological Agency and RSMAS/University of Miami, FL

The Field Phase of the ONR Tropical Cyclone Structure 08 (TCS-08) and THORPEX Pacific
Asian Regional Campaign (T-PARC) took place in the Northwest Pacific basin in August-
September 2008. One of the primary goals of the multi-national, multi-agency campaign was to
collect “targeted” observations within areas deemed important for improving tropical cyclone
track forecasts, such as the periphery of a subtropical ridge or a confluence zone near the
cyclone. The supplementary observations collected during intensive observing periods included
dropwindsondes released from coordinated aircraft missions and driftsonde balloons, airborne
Doppler Wind Lidar, off-time rawinsondes, and high-density atmospheric motion vectors
produced when the rapid-scanning mode was activated on Japan’s MTSAT2 satellite. The
decision of whether to deploy targeted observations was based on the potential threat of the
tropical cyclone to land, and the uncertainty in multi-model ensemble forecasts of the track. The
decision of where to deploy the targeted observations was based on guidance from up to 10
different targeting products derived from adjoint- and ensemble-based methods, several of which
were uploaded onto ECMWF’s PREVIEW Data Targeting System. The coordination of the
deployments was achieved via daily virtual conferences spanning 3 continents. The overall
procedure, lessons learned, and preliminary results of the influence of targeted observations on
tropical cyclone track forecasts will be presented, in addition to recommendations for the design
of an adaptive observational network and coordinated targeting procedures in the Atlantic Basin.


A New Strategy for Tropical Cyclone (TC) Reconnaissance
Based on Western Pacific TCS08 Proof of Concept

Peter G. Black1, Jeffrey Hawkins2 and Russell Elsberry3
(Peter.black@nrlmry.navy.mil)
1
SAIC, Inc, Atmospheric Sciences Division at NRL/ Marine Meteorology Division
2
NRL/ Marine Meteorology Division
3
Naval Postgraduate School/ Department of Meteorology

A new strategy for tropical cyclone reconnaissance was implemented during the Office of Naval
Research-sponsored Tropical Cyclone Structure 2008 (TCS08) experiment in the Western
Pacific this past summer. This new strategy was to fly WC-130J at maximum altitude near 300
mb and deploy on average 26 GPS dropsondes and 13 AXBTs per flight with a grid resolution of
one-half to one degree centered on the cloud cluster or incipient circulation center associated
with early stage TC development. AXBTs were deployed at every other grid point and SFMR
was used to map the surface winds, a feat made possible at relatively low winds by the larger
footprint at the high altitude. The grid varied from 3 to 5 degrees and was either square or
rectangular depending on the cloud morphology. This was the first time that AXBTs have been
successfully deployed, and data transmitted in real time, from 30,000 ft and the first time reliable
SFMR measurements have been made from this high altitude, a feat made possible in relatively
low wind environments by the 3-times larger footprint than from nominal 10,000 ft
reconnaissance altitude. Radar images were also recorded on USB ‘Thumb Drives’ and
compared with a constellation of coincident satellite microwave images, revealing evolution of
TC rainband morphology and providing input for satellite intensity and structure validation.
Flights were designed to be coincident with Quikscat and Windsat satellite scatterometer
overpasses for augmentation of surface wind field observations.

This strategy proved to be very successful in achieving TCS08 science objectives for TC genesis,
structure, intensity change and extra-tropical transition goals. We now suggest that this same
strategy may prove effective in meeting emerging observational requirements for model
initialization and validation in the Atlantic Basin as the use of coupled TC prediction models
grows. We feel that this strategy of flying high in weak systems rather than flying low looking
for closed circulations is an idea whose time has come with the advent of the WC-130J capability
to routinely fly at 300 mb altitude and its consequent ability to simultaneously map vertical
atmospheric structure searching for signatures of mid-level and well as low-level spin-up at the
earliest possible time while also mapping the surface wind field with SFMR. Furthermore,
simultaneous ocean vertical structure profiles will reveal signatures of ocean eddies that may
impact TC development and subsequent intensification. This type of strategy, especially in
concert with G-IV and WC-130J surveillance flights in the storm environment would bring a
powerful new set of observations into play that would be ideal not only for coupled model
initialization and validation but for forecaster diagnostics regarding presence of initial vortex and
potential for subsequent spin-up. The additional cost to the reconnaissance program of such a
strategy would be relatively modest in the form of support for roughly 20 extra expendables per
flight, a cost that is less than half that for flight-hour support for a typical flight.


P
O
S
T
E
R
Poster Session
S
E
S
S
I
O
N

Mapping Surface Winds Using Ocean Acoustic Interferometers

Alexander Voronovich and Cécile Penland
(Alexander.Voronovich@noaa.gov, Cecile.Penland@noaa.gov)

We discuss the feasibility of using an acoustic interferometer consisting of a pair of acoustic
hydrophones separated horizontally by a few tens of kilometers to diagnose the location and
intensity of the noise generated by tropical cyclones. Since the intensity of noise is strongly
dependent on wind, noise mapping is essentially equivalent to wind mapping. The method is
based on coherent processing of a broadband ambient noise within a frequency band of tens of
Hertz. We demonstrate that resolution of the order of a few kilometers, at distances of the order
of a thousand kilometers, is achievable in principle. An estimate of the signal to noise ratio is
also provided.

Poster Session – Page 1

Surface Wind Vector and Rain Rate Observation Capability
of the Future Hurricane Imaging Radiometer (HIRAD)

Timothy Miller1, Robert Atlas2, M. C. Bailey, Peter Black3, Salem El-Nimri6, Robbie Hood4,
Mark James5, James Johnson6, Linwood Jones6, Christopher Ruf7, and Eric Uhlhorn2
(Tim.Miller@nasa.gov)
1
NASA Marshall Space Flight Center
2
NOAA/Atlantic Oceanographic and Meteorological Laboratory (AOML)
3
SAIC, Inc at NRL/ Marine Meteorology Division
4
NOAA, Unmanned Aircraft Systems Program
5
NASA Marshall Space Flight Center
6
7
University of Michigan

The Hurricane Imaging Radiometer (HIRAD) is the next-generation Stepped Frequency
Microwave Radiometer (SFMR), and it will offer the capability of simultaneous wide-swath
observations of both extreme ocean surface wind vector and strong precipitation from either
aircraft (including UAS) or satellite platforms. HIRAD will be a compact, lightweight, low-
power instrument with no moving parts that will produce valid wind observations under
hurricane conditions when existing microwave sensors (radiometers or scatterometers) are
hindered by precipitation. The SFMR is a proven aircraft remote sensing system for
simultaneously observing extreme ocean surface wind speeds and rain rates, including those of
major hurricane intensity. The proposed HIRAD instrument advances beyond the current nadir
viewing SFMR to an equivalent wide-swath SFMR imager using passive microwave synthetic
thinned aperture radiometer technology. The first version of the instrument will be a single
polarization system for wind speed and rain rate, with a dual-polarization system to follow for
wind vector capability. This sensor will operate over 4-7 GHz (C-band frequencies) where the
required tropical cyclone remote sensing physics has been validated by both SFMR and WindSat
radiometers. HIRAD incorporates a unique, technologically advanced array antenna and several
other technologies successfully demonstrated by NASA’s Instrument Incubator Program. A
brassboard (laboratory) version of the instrument has been completed and successfully tested in a
test chamber. Development of the aircraft instrument is underway, with flight testing planned for
the fall of 2009. Preliminary Observing System Simulation Experiments (OSSEs) show that
HIRAD will have a significant positive impact on surface wind analyses as either a new aircraft
or satellite sensor. New off-nadir data collected in 2008 by SFMR that affirms the ability of this
measurement technique to obtain wind speed data at non-zero incidence angle will be presented,
as well as data from the brassboard instrument chamber tests.


Methodology for the Determination of 12-Foot Sea Radii for Tropical Cyclones

Jessica Schauer Clark, Hugh Cobb, and John Cangialosi
(Jessica.Clark@noaa.gov, Hugh.Cobb@noaa.gov, John.P.Cangialosi@noaa.gov)

Tropical Analysis and Forecast Branch, National Hurricane Center, Miami, FL

The advent of satellite altimetry and ocean surface vector wind observations has significantly
increased the amount of observational data available to marine meteorologists. The Tropical
Analysis and Forecast Branch (TAFB) at the National Hurricane Center (NHC) actively
incorporates these cutting edge data sets into their ocean analyses, providing the most accurate
assessment of the current sea state in their tropical North Atlantic and Northeast Pacific areas of
responsibility. This poster will detail the use of the recently-enhanced observational dataset
available to TAFB for the determination of the 12-ft sea radii associated with a tropical cyclone
(TC).

The Jason-1 polar-orbiting satellite observes global significant wave height data at 3-n mi
horizontal resolution with a precision of 0.1 ft. Observations from the Jason altimeter provide
wave height information in data-sparse regions of the tropical Atlantic and East Pacific oceans.
To “fill in” the sea state analysis in other locations, gridded wave heights from a short-term
forecast (most often a six-hour forecast) of a numerical wave model are used as a first-guess field
and adjusted accordingly for any reliable observations. One such model used is the Geophysical
Fluid Dynamics Laboratory (GFDL) version of the Wave Watch model, which incorporates a
more detailed initial wind field for a TC than other numerical wave models. Over data-sparse
regions such as the open waters of the tropical East Pacific or Atlantic oceans, ship data are the
only in-situ observations available to assist in analyzing the 12-ft sea radii for a TC. Once a TC
moves into the Gulf of Mexico, the southwest North Atlantic, or the Caribbean Sea, the
observational networks are denser with NOAA Buoys in the offshore waters and Coastal
Monitoring Automated Network (CMAN) sites in the near shore waters, primarily along the Gulf
Coast. These dense observational networks provide enough data to ensure a much more accurate
analysis of the 12-ft sea radii for the TC as compared to instances when a TC is located in the
open ocean. Additionally, the dense observations in the western portion of the Atlantic basin
allow for a better assessment of the performance of the various numerical models. In the future,
TAFB should be able to perform a real-time evaluation of aircraft-based wave height information
from the Joint Hurricane Test Bed Scanning Radar Altimeter Project, which could further
improve the estimation of the 12-foot sea radii.

The active utilization of remotely-sensed ocean observations allows for a more exact analysis of
the initial sea state in regions of little or no in-situ data and provides an “accuracy check” for
ship and buoy observations that may contain errors. In turn, more reliable sea state analyses
permit a better appraisal of the quality of numerical wave model output. While long-term
verification is difficult due to a lack of historic data, we believe all of these advances have
significantly improved the estimation of the 12-ft sea radii for TCs over previous methods.


Sea Surface Roughness Measurements Using GPS Reflectometry in Hurricane Ike (2008)
Valery U. Zavorotny1, Dennis M. Akos2, Edward J. Walsh3
(Valery.Zavorotny@noaa.gov, dma@colorado.edu, Edward.Walsh@noaa.gov,)
1
NOAA/Earth System Research Laboratory
2
University of Colorado at Boulder/Aerospace Engineering Sciences
3
NASA/Goddard Space Flight Center
A number of airborne experiments have been performed during the last decade that studied a
sensitivity of GPS reflected signals to ocean surface roughness both for steady, uniform wind
conditions, and for hurricane conditions (see, e.g., [1, 2]). A theoretical model [3] is available
that relates the characteristics of reflected signals to the slope statistics, and from there, to a near-
surface wind speed, if the local wind is the only source of sea roughness at the footprint location.
Still a lot of work needs to be done to make this technique more reliable, robust and accurate for
operational needs.
Recently, a compact system, able to record raw GPS intermediate frequency (IF) samples, has
been designed and tested by the GNSS group at the CU/Aerospace Science Engineering. Such an
approach provides the most fundamental measurement, enabling the most advanced and
complete post-processing, with data volumes on the order of 1GB/minute. On September 10 and
11, 2008 this GPS bistatic system was flying on board of the NOAA WP-3D research aircraft
which collected research-mission data on Hurricane Ike in Gulf of Mexico. The aircraft flew
through the hurricane at about 2500 meters performing a characteristic flight track comprising of
several radial lags that transverse the hurricane eye at different azimuthal angles. The GPS
bistatic radar was recording raw data practically over the entire two flights collecting about 800
GB of data. After the flights the hard drives with data have been sent to the lab for post-
processing, and the correlation waveforms for both direct and reflected signals were retrieved
from the raw data for all available satellites.
As it is follows from the theoretical model [3], the shape of the correlation waveform, and the
slope of its trailing edge depends on rms of L-band limited ocean surface slopes. Empirical
spectral models for well-developed seas relate the slope rms to the local wind speed. However,
wave fields generated by hurricanes do not follow such a simple rule. System of swells generated
by various parts of the hurricane complicates the picture and makes the problem of wind retrieval
non-trivial. Presence on board of the aircraft of the Step Frequency Microwave Radiometer
(SFMR) allowed making comparisons between wind retrievals obtained with the SFMR and the
GPS bistatic radar.
References
1. J.L. Garrison, A. Komjathy, V.U. Zavorotny, and S.J. Katzberg, “Wind speed measurement
using forward scattered GPS signals,” IEEE Trans Geosci. Remote Sens., 40, pp.50-65, 2002.
2. S.J. Katzberg, O. Torres, and G. Ganoe, “Calibration of reflected GPS for tropical storm wind
speed retrievals,” Geophys. Res. Lett., 33, L18602, doi:10.1029/2006GL026825, 2006.
3. V.U. Zavorotny and A.G. Voronovich, “Scattering of GPS signals from the ocean with wind
remote sensing application,” IEEE Trans Geosci. Remote Sens., 38, pp. 951−964, 2000.


Development of Miniature SFMR Receiver Modules and the Compact-SFMR

Ivan PopStefanija
(popstefanija@prosensing.com)

ProSensing Inc.

This paper describes the development of a miniaturized Stepped Frequency Microwave
Radiometer (SFMR) receiver module and the next generation compact-SFMR instrument.
Funded through the NOAA SBIR program, these receiver modules are intended for use in multi-
channel imaging radiometers as well as in compact versions of the standard SFMR. ProSensing
is currently testing a prototype compact C-band radiometer receiver (CCR) module, which
employs inexpensive surface mount components to implement the standard SFMR electronics in
a small, lightweight package. The CCR modules will be used in the Hurricane Imaging
Radiometer (HIRad), currently under development by a team of government, university and
industrial partners.

As part of SBIR Phase II effort, ProSensing will develop and deliver:
 Ten single sideband copies of the CCR receiver module suitable for imaging correlation
radiometers such as the HIRad system. The modules include an integral COLFET load,
as well as digitally controlled phase shifters and attenuators to correct for temperature
drift.
 A compact-SFMR system with a total instrument weight of 3 kg, including the CCR
receiver and horn antenna. Two complete compact-SFMR instruments will be delivered
to NOAA, one for deployment on a small UAV, and a second for measuring down
welling atmospheric brightness temperature on the NOAA P-3 for improved SFMR low
wind speed and rain rate measurements.

The first two prototypes of the CCR are currently being tested at ProSensing and are due to be
delivered to NASA MSFC in February, 2009.


The HFIP High Resolution Hurricane Test

Ligia Bernardet1, Louisa Nance2, Barbara Brown2, Shaowu Bao3, Chris Harrop3, Jamie Wolff2,
and Tara Jensen2
(Ligia.Bernardet@noaa.gov)
1
NOAA Earth System research Laboratory – Global Systems Division/Systems Research Group,
Colorado Springs/Boulder, CO
2
National Center for Atmospheric Research, Boulder CO
3
NOAA Earth System research Laboratory – Global Systems Division/ Cooperative Institute for
Research in Environmental Sciences, Boulder, CO

Tropical cyclones are a serious concern for the nation, causing significant risk to life, property
and economic vitality. The NOAA National Weather Service has a mission of issuing tropical
cyclone forecasts and warnings, aimed at protecting life and property and enhancing the national
economy. In the last 10 years, the errors in hurricane track forecasts have been reduced by about
50% through improved model guidance, enhanced observations, and forecaster expertise.
However, little progress has been made during this period toward reducing forecasted intensity
errors.

To address this shortcoming, NOAA established the Hurricane Forecast Improvement Project
(HFIP) in 2007. HFIP is a 10-year plan to improve one to five day tropical cyclone forecasts,
with a focus on rapid intensity change. Recent research suggests that prediction models with
grid spacing less than 1 km in the inner core of the hurricane may provide a substantial
improvement in intensity forecasts. The 2008-09 staging of the high resolution test aims at
quantifying the impact of increased horizontal resolution in numerical models on hurricane
intensity forecasts. The primary goal of this test is an evaluation of the effect of increasing
horizontal resolution within a given model across a variety of storms with different intensity,
location and structure. A secondary goal is to provide a data set that can be used to explore the
potential value of a multi-model ensemble for improving hurricane forecasts.

The Developmental Testbed Center (DTC) and the HFIP Team hosted a workshop at the
National Hurricane Center in Miami, FL, 11-12 March 2008. Experts on hurricanes, numerical
modeling and model evaluation met for two days to discuss the strategy for this test. The test
plan that was put together following this workshop reflects the consensus reached on the
framework for this testing effort. Seven modeling groups expressed interest in participating in
the testing effort at the time of the workshop. This presentation will describe the basic elements
of the test plan and summarize the current status of the test. Only limited examples of the results
will be shown since the bulk of the retrospective hurricane forecasts will not be completed by the
time of the Conference.


Coupling a High Resolution Hurricane Storm Surge Model to Operational Weather and
Ocean Prediction Systems

Yuji Funakoshi1, 2, Jesse Feyen1, Frank Aikman1, Carlos Lozano3, and Hendrik Tolman3
(Yuji.Funakoshi@noaa.gov, Jesse.Feyen@noaa.gov, Frank.Aikman@noaa.gov,
Carlos.Lozano@noaa.gov, and Hendrik.Tolman@noaa.gov)
1
NOAA, National Ocean Service, Coast Survey Development Laboratory
2
University Corporation for Atmospheric Research (UCAR)
3
NOAA, National Weather Service, National Centers for Environmental Prediction

The Coast Survey Development Laboratory of the National Ocean Service (NOS) and the
National Centers for Environmental Prediction (NCEP) of the National Weather Service (NWS)
are cooperating to evaluate strategies for a next-generation storm surge modeling system. The
model used to develop a high resolution Coastal Flooding Model (CFM) is the ADvance
CIRCulation (ADCIRC) finite element model. ADCIRC utilizes an unstructured mesh which
provides high resolution in areas of complex shoreline and geometry to perform storm surge
flood inundation prediction.

The approach is to simulate storm surge for the coastal region using the high resolution
unstructured mesh CFM. We examine the impact of boundary and meteorological forcing from a
suite of NWS operational weather and ocean prediction models. Open ocean boundary
conditions are provided to the CFM by the NCEP Real-Time Ocean Forecast System (RTOFS),
which is based on the HYbrid COordinate Model (HYCOM) basin-scale ocean circulation
model. The meteorological forcing fields (e.g. the hurricane wind and pressure fields) being
tested include the Hurricane Wind analysis system (H*Wind), the Geophysical Fluid Dynamics
Laboratory (GFDL) hurricane model, and the Hurricane Weather Research and Forecasting
(HWRF) model, all of which are currently operational models for forecasting or analyzing
hurricane track and intensity.

The coupled system predicts inundation via the inclusion of the high resolution ADCIRC coastal
model (the CFM). Furthermore, the outcomes of the coupled system could form decision-support
tools that the coastal community can use. Using the expertise in several parts of NOAA as well
as having partners in other federal and state agencies will ensure that outcomes are based on an
integrated approach that makes the best use of data, modeling approaches, displaying results, and
outreach.


An Integrated Tropical Cyclone Information System for Research

S. Hristova-Veleva1, Y. Chao1, Z. Haddad1, B. Knosp1,
B. Lambrigtsen1, P. P. Li1, D. Vane1, Q. A. Vu1
(svetla.veleva@jpl.nasa.gov)
1
Jet Propulsion Laboratory
In spite of recent improvements in hurricane track forecast accuracy, currently there are still many
unanswered questions about the physical processes that determine hurricane genesis, intensity, track and
impact on large-scale environment. Furthermore, a significant amount of work remains to be done in
validating hurricane forecast models, understanding their sensitivities and improving their
parameterizations. None of this can be accomplished without a comprehensive set of multiparameter
observations that are relevant to both the large-scale and the storm-scale processes in the atmosphere and
in the ocean.
To address this need, we and our colleagues from Marshall Space Flight Center have developed the
framework for a comprehensive Integrated Tropical Cyclone Information System (iTCIS) of high-
resolution satellite, airborne and in-situ observations and model outputs pertaining to: i) the
thermodynamic and microphysical structure of the storms; ii) the air-sea interaction processes; iii) the
larger-scale environment as depicted by the SST, ocean heat content and the aerosol loading of the
environment. Our goal was to create a one-stop place to provide the researchers with an extensive set of
observed hurricane data, and their graphical representation, together with convection-resolving model
output, all organized in an easy way to determine when coincident observations from multiple
instruments are available.
Currently the iTCIS is populated with satellite observations of all tropical cyclones observed globally
during 2005. Our future plans are to extend the database both forward in time till present as well as
backward to 1998. Furthermore, NASA field campaign data will be included and displayed using Goggle
Earth field-campaign-specific approach that allows the creation of overlays on-demand. This innovative
approach, known as the Real Time Mission Monitor (RTMM) has been developed at the Marshall Space
Flight Center. It has proven very valuable during past field campaigns, both during the campaign as well
as in the post-campaign analysis. The RTMM is described in a separate presentation.
We began the development of analysis tools that will communicate with the database and will allow for
online investigations without the need to download large data volumes. The analysis tools will be used to
compute single and multiparameter statistics and to determine spatial, temporal and multiparameter
covariances that are needed to evaluate model performance, provide information for data assimilation and
characterize and compare observations from different platforms.
We envision that the developed hurricane information system will help in the validation of the hurricane
models, in the systematic understanding of their sensitivities and in the improvement of the
parameterizations employed by the models. Furthermore, it will help in studying the physical processes
that affect hurricane development and impact on large-scale environment.
This talk will describe the structure of the satellite-based component of the iTCIS, the analysis system and
the future plans. A poster will accompany the talk. It will illustrate how the iTCIS can be used to
discriminate between high resolution WRF model simulations that employ different parameterizations
with the goal to determine the model setup that produces most realistic simulations.
The work described here was performed at the Jet Propulsion Laboratory, California Institute of
Technology, under contract with the National Aeronautics and Space Administration.


Using NASA’s Real Time Mission Monitor to Manage and Conduct Airborne Science
Missions
Michael Goodman1, Richard Blakeslee1, John Hall2, Yubin He2, Kathryn Regner2 and Helen
Conover2
(michael.goodman@nasa.gov, rich.blakeslee@nasa.gov, john.hall@nasa.gov,
mhe@itsc.uah.edu, kregner@itsc.uah.edu, hconover@itsc.uah.edu)
1
National Aeronautics and Space Administration,
2
University of Alabama in Huntsville

The Real Time Mission Monitor (RTMM) is a visualization and information system that fuses
multiple Earth science data sources, to enable real time decision-making for airborne and ground
validation experiments. Developed at the National Aeronautics and Space Administration
(NASA) Marshall Space Flight Center, RTMM is a situational awareness, decision-support
system that integrates satellite imagery, radar, surface and airborne instrument data sets, model
output parameters, lightning location observations, aircraft navigation data, soundings, and other
applicable Earth science data sets. The integration and delivery of this information is made
possible using data acquisition systems, network communication links, network server resources,
and visualizations through the Google Earth virtual globe application.

RTMM has proven extremely valuable for optimizing individual Earth science airborne field
experiments. Flight planners, mission scientists, instrument scientists and program managers
alike appreciate the contributions that RTMM makes to their flight projects. We have received
positive feedback from a broad spectrum of scientists who used RTMM during field campaigns
including the hurricane-focused 2006 NASA African Monsoon Multidisciplinary Analyses
(NAMMA), 2007 NOAA-NASA Aerosonde Hurricane flight, 2007 Tropical Composition,
Cloud, and Climate Coupling (TC4), 2008 Arctic Research of the Composition of the
Troposphere from Aircraft and Satellites (ARCTAS) missions, and the 2008 Soil Moisture
Active-Passive Validation Experiment (SMAP-VEX).

Improving and evolving RTMM is a continuous process. RTMM recently integrated the
Waypoint Planning Tool, a Java-based application that enables aircraft mission scientists to
easily develop a pre-mission flight plan through an interactive point-and-click interface.
Individual flight legs are automatically calculated for altitude, latitude, longitude, flight leg
distance, cumulative distance, flight leg time, cumulative time, and satellite overpass
intersections. The resultant flight plan is then generated in KML and immediately posted to the
Google Earth-based RTMM for interested scientists to view the planned flight track and
subsequently compare it to the actual real time flight progress.

We are planning additional capabilities to RTMM including collaborations with the Jet
Propulsion Laboratory in the joint development of a Tropical Cyclone Information System which
will serve as a web portal for access to tropical cyclone data, visualizations and model output.

A description of the system architecture, components, and applications along with reviews and
animations of RTMM during the field campaigns, plus planned enhancements and future
opportunities will be presented. A poster will complement the talk.


Use of Total Precipitable Water to Aid Hurricane Rainfall Prediction

Haiyan Jiang1, Edward J. Zipser1, Jefferey B. Halverson2, and Robert Rogers3,
1
Department of Meteorology, University of Utah, Salt Lake City, Utah
2
Joint Center for Earth Systems Technology, University of Maryland, Baltimore County,
Baltimore, Maryland
3
NOAA Hurricane Research Division

The Tropical Rain Measuring Mission (TRMM) Multi-satellite Precipitation Analysis (TMPA,
or TRMM 3B42) has provided 3-hourly continuous rainfall estimates for tropical cyclones over
the global oceans since 1998. Previous work by the authors has established that a positive
relationship exists between the hurricane rainfall and environmental moisture parameters such as
total precipitable water and horizontal moisture convergence for the Atlantic basin using a 7-yr
TRMM 3B42 Atlantic landfalling tropical cyclone database. It suggested that total precipitable
water could be used as an additional predictor for tropical cyclone rainfall.

In this study, we extend this work to other basins, including east+central Pacific, northwest
Pacific, north Indian Ocean, south Indian Ocean, and south Pacific. We further examine the
relationship between total precipitable water and tropical cyclone rainfall as a function of storm
intensity, SST, shear magnitude, and geographical locations including over-land vs over-ocean
using 10 years of TRMM 3B42 data. Then we construct a statistical regression model by
combining maximum wind speed with total precipitable water as an independent variable for
forecasting Atlantic hurricane rainfall. Environmental parameters are derived from NOGAPS
analysis. The azimuthal mean rain rate is examined as a function of total precipitable water
averaged within 500-km radius from the storm center for different tropical cyclone intensities,
different basins, high/low SST, high/low shear, and over-land or over-ocean conditions. The
correlations among total precipitable water, maximum wind speed, and 12, 24, 36, 48-h future
storm volumetric rain are also derived.


Strong Surface Wind Gusts Associated With Heavy Squalls within Hurricane Danny’s
(1997) Eyewall.

Keith G. Blackwell
(kblackwell@usouthal.edu)

Coastal Weather Research Center
Department of Earth Sciencs
University of South Alabama

Some landfalling hurricanes contain massive collapsing cores (CCs) of heavy precipitation
within their eyewalls. Several papers indicate the likely linkage between downbursts-like
features and strong wind gusts in hurricane eyewalls; however, very few surface-based
measurements of winds associated with these features actually exist.

Hurricane Danny (1997) produced one of the greatest precipitation events on record in the
United States during its landfall along the Alabama coast. Radar observations indicate that
Danny was likely one of the most prolific CC producers during this time of any landfalling U.S.
hurricane between 1994 and 2007. Slow-moving Danny stayed within 100 km of the KMOB
WSR-88D Doppler radar for many hours while in the vicinity of Dauphin Island, Alabama.

Shortly after Danny drifted into Mobile Bay, the increasingly convective southwest eyewall of
the storm buffeted the Coastal Marine Automated Network (C-MAN) site on the east end of
Dauphin Island (DPIA1). Wind gusts up to 88 knots were recorded there during this time.
While in the eyewall, nearly all the strongest wind gusts each hour occurred during the passage
of very heavy squalls often displaying a low-level collapse of their heaviest precipitation cores.

Base velocities from the nearby KMOB WSR-88D indicate the presence of low-level wind
maxima in the vicinity of DPIA1 near the 500 m altitude as these squalls passed. Momentum
from these slightly-elevated wind maxima was likely diverted toward the surface during the
passage of these very heavy precipitation cores, thus providing a favorable situation for strong
near-surface wind gusts.

As Danny moved further into Mobile Bay, the low-level wind maxima in the hurricane’s
southwest eyewall intensified within a nearly continuous collapse of extremely heavy
precipitation on the downwind side of an intense and persistent eyewall convective complex.
Thus, even stronger near-surface wind gusts likely occurred repeatedly over a multi-hour period
over the open waters of Mobile Bay, but were occurring in an area lacking surface observing
equipment.


Hurricanes at Landfall: Rain-Field Sizes and Their Relationship to Environmental
Conditions

Corene Matyas
(matyas@ufl.edu)

Department of Geography, University of Florida

Predicting the size of a hurricane’s rain field can help to determine the overall area that may
receive rainfall from the storm as well as the time that rainfall may begin and the duration of the
rainfall event. This study examines 28 hurricanes that made landfall along the U.S. coast during
1995-2007. Radar reflectivity data are entered into a GIS and the distance from the center of
circulation to the edge of the rain field is measured in each quadrant. To associate the extent of
the rain field with the physical forcing mechanisms helping to shape the storm’s structure,
variables derived from the Statistical Hurricane Intensity Prediction Scheme database, storm
intensity, position, and motion, and coastline geometry are examined. Multiple linear regression
models utilize these variables to predict the distance from the circulation center to the edge of the
rain field for each quadrant. Results confirm that it is important to consider that the physical
forcing mechanisms affect the storm differently in each quadrant. The most asymmetrically-
shaped rain fields are present within hurricanes that are experiencing trough interaction, as the
northeast/right-front quadrant typically extends the farthest from the circulation center, while the
southwest/ left rear quadrant exhibits the shortest extent. More than 80% of the variability in rain
field extent in the northeast quadrant, which tends to be in the downshear direction, can be
explained by the north component of the deep-layer vertical wind shear. In the northwest
quadrant, storm motion and intensity are most closely associated to the extent of the rain field,
and intensity as well as the shallow-layer vertical wind shear are highly correlated to the rain
field extent in the southwest quadrant. The extent of the rain field in the southeast quadrant is the
most difficult to predict; vorticity, latitude, and divergence at 200mb combine to account for less
than half of the variance in rain field extent in this quadrant of the storm. Overall, investigating
the extent of the rain field in each quadrant improves the ability to quantify the overall storm
shape.


Challenges in the Prediction of the Life Cycle of Typhoon Sinlaku (2008)

William A. Komaromi, Munehiko Yamaguchi, Eric D. Rappin, Sharanya J. Majumdar,
and David S. Nolan
(wkomaromi@rsmas.miami.edu)

Rosenstiel School of Marine and Atmospheric Science, University of Miami

One of the most scientifically interesting and societally important tropical cyclones during the
ONR Tropical Cyclone Structure (TCS-08) and THORPEX Pacific Asian Regional Campaign
(T-PARC) field phase was Typhoon Sinlaku in September 2008. This poster will illustrate
several challenges in the prediction of Sinlaku from genesis through recurvature.

The genesis of Typhoon Sinlaku was poorly predicted by the operational global models. Only
ECMWF predicted Sinlaku’s development 1 day in advance of its formation. The ECMWF,
GFS, UKMET, and NOGAPS all indicated development during the 4 days prior to formation,
but with no run-to-run consistency. New simulations with a 2-km version of the Advanced
Research Weather Research and Forecasting (WRF-ARW) model will be presented.

There was also day-to-day and model-to-model inconsistency in the timing and location of the
track, particularly when Sinlaku approached Taiwan. Beta-drift may have dominated, given the
weak steering flow. Sinlaku’s recurvature was most likely due to the influence of a short-wave
trough to its north, as is evident via removal of the short-wave using vorticity inversion in the
WRF model. The importance of inserting a realistic bogus vortex within WRF to produce
accurate predictions of structure and intensity is also emphasized in this study.

In order to diagnose the error sources in Sinlaku’s genesis and track forecasts, initial
perturbations derived from the THORPEX Interactive Grand Global Ensemble (TIGGE) are
used. The vertical structure of the perturbations was found to differ significantly between the
ECMWF, UKMO, NCEP, CMC and JMA ensembles. For an ensemble based on a single model,
differences between the tracks are evidently owed to differences in the steering flow, and not the
cyclone size. On the other hand, for multi-model ensembles, the differences in tracks are due to
changes in both the cyclone size and steering flow.

Future plans include the use of adjoint- and ensemble-based diagnostics to elucidate the
influence of assimilating high-density satellite wind and dropsonde data in “sensitive” areas.


The Impact of the Saharan Air Layer on the Development of Hurricanes Irene (2005) and
Helene (2006)

Scott A. Braun1, Jason Sippel2, David Nolan3
1
NASA/GSFC, Greenbelt, MD
2
Oak Ridge National Laboratory, Oak Ridge TN and NASA/GSFC, Greenbelt, MD
3
University of Miami, Miami, FL

The existence of the Saharan air layer (SAL), a layer of warm, dry, dusty air frequently present
over the tropical Atlantic Ocean, has long been appreciated. The nature of its impact on
hurricanes remains unclear, with some researchers arguing that the SAL amplifies hurricane
development and with others arguing that it inhibits it. The potential negative impacts of the SAL
include 1) low-level vertical wind shear associated with the African easterly jet; 2) warm air
aloft, which increases thermodynamic stability; and 3) dry air, which produces cold downdrafts.
Some investigators have assumed the validity of these proposed negative influences and have
frequently used them to explain the failure of individual storms to intensify or to explain the
relative inactivity of recent hurricane seasons. Multiple NASA satellite data sets and National
Centers for Environmental Prediction global analyses are used to characterize the SAL’s
properties and evolution in relation to hurricanes Irene and Helene. The results will show that
neither jet-induced vertical wind shear nor warm SAL air (high stability) were likely to have
produced significant negative impacts on these storms. Dry air appears to be a key mechanism
for SAL influence. Idealized simulations will be used to evaluate the role of dry air.


Evaluating Ensemble Track Forecasts of Tropical Cyclones

Peter Finocchio, Sharanya Majumdar
(p.finocchio@umiami.edu, smajumdar@rsmas.miami.edu)

University of Miami, Rosenstiel School of Marine and Atmospheric Science

While tropical cyclone track forecasts in deterministic models continue to undergo evaluation
and thereby improve, the verification of probabilistic track forecast products derived from
ensembles is less routine. This study evaluates the skill of the ECMWF and UKMO ensemble
prediction systems for the 2008 Atlantic and West Pacific seasons. First, a “cone of uncertainty”
derived from the ensemble forecast is verified, based on the principle that the best track should
fall within the cone in 67% of all cases. As is done at NHC, the cone is constructed using a
series of circles centered on some “best forecast” at successive forecast times. In this study, the
circle at each time is centered on the ensemble mean forecast position, with its radius defined as
that enclosing 67% of the ensemble forecast members at that time. In the evaluation, the number
of times for which the best track lies within this radius is then counted over the season, for all 1-
10 day UKMO ensemble forecasts and 1-5 day ECMWF ensemble forecasts. Preliminary results
indicate that for the Atlantic basin, the UKMO ensemble is superior to ECMWF at predicting an
appropriate spread of tracks for forecasts up to 2 days, but UKMO then becomes too under-
dispersive for longer-range forecasts with the best track falling outside the cone too often. In
contrast, the ECMWF shows virtually ideal ensemble spread for all forecast times between 2-5
days. Other products to be evaluated include cones of uncertainty constructed around NHC’s
consensus model and official forecast tracks, cones of uncertainty of different sizes (percentage
spread), and ensemble means as compared against deterministic forecasts.


The Communication of Tropical Cyclone Emergency Information through Use of
Geographic Information System (GIS) Products

Alison Krautkramer1, Christopher Lauer1, Christopher Sisko1, Charles Sampson2, Ann Schrader2
and Edward Fukada3.
(Alison.Krautkramer@noaa.gov)
1
National Hurricane Center (NWS/NOAA), Miami, FL
2
Naval Research Laboratory (USN/SAIC), Monterey, CA
3
Joint Typhoon Warning Center (NMFC/USN), Pearl Harbor, HI

The communication of tropical cyclone analysis and forecast information has always been a key
method in providing emergency management and decision makers with the necessary tools to
make informed decisions. Although there are many methods of delivering such information, the
use of GIS provides powerful ways to manipulate, interrogate, visualize and ultimately
understand the data. GIS enables the users to quickly and efficiently customize the information
to meet their specific mission requirements. The interoperability of GIS data facilitates a wide
variety of functions that include merging with other data types, expanded report capabilities and
detailed map production. Unlike static content, GIS allows the end user to quickly process the
information directly into applications; consequently, they can easily interpret large volumes of
information.

The National Hurricane Center (NHC) working with other agencies has expanded its capabilities
to produce and distribute experimental tropical cyclone GIS content. During the 2008 hurricane
season, NHC delivered breakpoints, watch and warning, storm surge, forecast track and forecast
uncertainty cone GIS information in near real-time. Future expectations are to deliver best track
and probability information in order to expand the availability of data layers to the user. This
experimental line up of products is expected to fill the GIS gap in regards to communication of
the real-time forecast and post storm analysis. Furthermore, detailed Meta information
describing the products and their limitations accompany these GIS data and the format of choice
utilized is the popular and highly utilized ESRI Shapefile format specification. These particular
products are targeted for emergency management and decision makers; however, the
interoperability of these products has also benefited a wide variety of other users because the
data can be used quickly and efficiently with off the shelf GIS applications. The progress made
in the area of delivering useful tropical cyclone GIS content in near real-time is closing an
identified gap in products and services offered by NHC.


The Near Real-Time and Science Study Satellite Product Suite for the Tropical Cyclone
Structure (TCS-08) Field Project

Jeff Hawkins1, Joe Turk1, Kim Richardson1, Rich Bankert1, Christian Mitrescu1, Chris Velden2,
Derrick Herndon2, Tony Wimmers2, Tim Olander2, Steve Miller3, John Kent4
1
Naval Research Laboratory
2
Cooperative Institute for Meteorological Satellite Studies (CIMSS),
University of Wisconsin
3
Cooperative Institute for Research of the Atmosphere (CIRA),
Colorado State University
4
Science Applications International Corporation, Monterey, CA

A wealth of satellite imagery and derived products were provided in near real-time during the
Tropical Cyclone Structure (TCS-08) field program (summer 2008) to address multiple
objectives: 1) provide mesoscale and synoptic scale weather nowcasting information used to
coordinate multiple research aircraft, 2) address topics ranging from tropical cyclone genesis,
storm structure, extratropical transition and satellite TC intensity validation, and 3) supplement
major data voids due to limited in-situ observations. In response, comprehensive suite of
geostationary and polar orbiter digital data sets were acquired and processed, resulting in near
real-time products distributed automatically to field program personnel and archived for
numerous TCS-08 scientific studies.

The Naval Research Laboratory in Monterey, CA and the Cooperative Institute for
Meteorological Satellite Studies (CIMSS) in Madison, WI combined to create a suite of satellite
products; 1) visible, infrared (IR), and water vapor imagery over synoptic, mesoscale and storm-
specific domains, 2) geostationary-derived cloud and water vapor-tracked winds, 3) microwave
enhanced imagery and products highlighting atmospheric moisture over large domains as well as
storm-focused rainband and eyewall feature structure, 4) wind shear, convergence and
divergence fields, 5) ocean surface winds from scatterometer and polarimetric radiometers, and
6) TC intensity estimates via microwave and IR algorithms. Products were made available to
field program scientists via dedicated web pages, ftp and automated email alerts, including for
the first time Google Earth dedicated products permitting enhanced geographical orientation,
display and visualization. Digital data was also routinely available via a central catalog at the
Earth Observation Laboratory (EOL) and the National Center for Atmospheric Research
(NCAR) for both real-time utilization and archive purposes.

The satellite suite includes the MTSAT geostationary sensor, the operational and research
constellation of microwave polar orbiters (SSM/I, SSMIS, AMSR-E, TMI, and WindSat),
microwave sounders (AMSU-A/B, SSMIS, and MHS), scatterometers (QuikSCAT and
ASCAT), as well as multiple radars (TRMM precipitation radar and CloudSat). Data latencies
ranging from minutes to 1-4 hours enabled the TCS-08 personnel to gather a full range of
environmental information in order to address the extensive near real-time and longer term
research initiatives as funded by the Office of Naval Research (ONR) and the National Science
Foundation (NSF).


Improved GOES Utilization for Tropical Cyclone Forecasting
Mark DeMaria1, John A. Knaff1, Andrea B. Schumacher2, John Kaplan3, Daniel P. Brown4,
Gregory M. Gallina5, James P. Kossin6
(Mark.DeMaria@noaa.gov)
1
NOAA/NESDIS/StAR/RAMMB, Fort Collins, Colorado
2
CIRA/Colorado State University, Fort Collins, Colorado
3
OAR/Hurricane Research Division, Miami, FL
4
NOAA/NWS/TPC, Miami, Florida
5
NOAA/NESDIS/SAB, Camp Springs, Maryland
6
NOAA/NESDIS/NCDC, Madison, Wisconsin
Data from the Geostationary Operational Environmental Satellite (GOES) suite has been used
monitoring, diagnosing and forecasting tropical cyclones for over thirty years, but their
information content is far from exhausted. Several efforts are underway to make even better use
of the GOES products for tropical cyclone intensity diagnosis, intensity and formation
forecasting. These are described below.
Two of these efforts involve the expansion of the tropical cyclone IR imagery database and its
improved used for intensity forecasting. NHC’s operational statistical intensity model (SHIPS)
is run in a two-step process. The first step creates a forecast based on TC current conditions and
the Global Forecast System (GFS) analyses and forecasts. The second step, or correction step,
then adjusts the existing forecast to account for satellite information from (IR imagery) and
Oceanic Heat Content. The satellite-based predictors have recently been extended back in time
so that this two step procedure may no be longer needed. In a related project, direction relative
IR imagery is being decomposed into complex empirical orthogonal functions, which can
account for propagating features as possible new predictors in SHIPS and the related Rapid
Intensification Index (RII). Both of these efforts will be described.
The Dvorak technique is the one of the cornerstones of operational intensity diagnosis, but the
method has some known and suspected biases. Using the historical records of both Dvorak and
aircraft fixes along with the best track intensity estimate, a calibration that accounts for tropical
cyclone intensity, latitude, size (via radius of outer closed isobar), translation speed and intensity
trend is created. Preliminary results of this ongoing study are presented.
In addition to the SHIPS, RII, and Dvorak work, the water vapor (WV) imagery from several
geostationary satellites is being inter-calibrated to form a global tropical water vapor dataset.
Past studies have shown that propagating features, e.g., equatorial waves, can be detected in
satellite-based Outgoing Longwave Radiation (OLR) and used for prediction. Unlike OLR, WV
imagery provides information about both convective persistence and atmospheric moisture
trends. The current study exploits the predictive nature of tropical wave forms as well as the
information content of the water vapor imagery to examine the possible predictive information
available in the tropical water vapor strip for better diagnosing TC formation. Progress thus far
on this project will be described.
decision.


The International Best Track Archive for Climate Stewardship (IBTrACS)

Kenneth R. Knapp1, Michael C. Kruk2, David H. Levinson1
(Ken.Knapp@noaa.gov)
1
NOAA/NCDC,
2
STG, Inc., Asheville, NC

There are currently six Regional Specialized Meteorological Centers (RSMC) and five Tropical
Cyclone Warning Centers (TCWC) around the world that forecast and monitor storms in each of
the tropical-cyclone-prone basins and annually archive best track data: information on a storm’s
position, intensity, as well as other related parameters. Despite the widespread interest in global
best track data, no central repository of such data was maintained. The International Best Track
Archive for Climate Stewardship (IBTrACS) is a project at NOAA’s National Climatic Data
Center under the auspices of the World Data Center for Meteorology – Asheville to collect and
disseminate the historical tropical cyclone best track data from all available centers, merging the
disparate data sets into one comprehensive product for the user community. One of the goals of
the project is to maintain open data processing methods, so that desired user feedback on data
quality is more easily collected. In addition, data provenance is completely recorded so all
observations and corrections, either through rigorous quality control or user feedback, may be
attributed to their source. Data are available in various formats to accommodate the diverse
needs of the tropical cyclone data user community. The presentation will summarize the purpose
and vision of the project, the methods used to merge the data, and a discussion of results of
global and basin-wide tropical cyclone statistics. More information on IBTrACS is available at
http://www.ncdc.noaa.gov/oa/ibtracs/

Figure – Tracks of tropical cyclones from IBTrACS during 1947-2007.


Mapping Hurricane Inland Storm Tides

Mike Turco1, Jeffery W. East1, Michael E. Dorsey2, Ben D. McGee3, Brian E. McCallum4, James
L. Pearman5, Asbury H. Sallenger6, Robert R. Holmes, Jr.7, D. Phil Turnipseed8, Charles
Berenbrock8, Robert R. Mason, Jr. 8
(jpearman@usgs.gov)
1
United States Geological Survey, The Woodlands, Texas; 2 United States Geological Survey,
Austin, TX; 3United States Geological Survey, Ruston, LA; 4United States Geological Survey,
Atlanta, GA; 5 United States Geological Survey, Orlando, FL; 6United States Geological Survey,
St. Petersburg, FL; 7United States Geological Survey, Rolla, MO; 8 United States Geological
Survey, Reston, VA

Historically, hurricane-induced storm-tides were documented through analysis of structural or
vegetative damage and high-water marks. However, these sources rarely provided quantitative
information about the timing of the flooding, the sequencing of multiple paths by which the
storm-surge waters arrived, or the magnitude of waves and wave run-up comprising floodwaters.
Nor could highwater marks be used to evaluate storm-surge model performance along the
dynamic track of a hurricane with its accompanying changes in wind strength and direction.
In response to these deficiencies, the U.S. Geological Survey (USGS) developed and deployed
an experimental mobile storm-tide network to provide detailed time-series data for selected
hurricane landfalls (URL: http://water.usgs.gov/waterwatch/hsss/). As part of this program,
water-level and barometric pressure monitors are deployed to areas of hurricane landfall
resulting in a concentrated network of as many as 80 temporary, tide gages placed along water
channels and nearby overland features such as beaches, wetlands, and constructed environments.
USGS storm-surge networks have been successful deployed for Hurricanes Rita (2005, 32 water-
level monitoring sites); Wilma (2005, 30), Gustav (2008, 80); and Ike (2008, 65); and Tropical
Storm Ernesto (2006, 40 sites for which the data were not published.) Data were collected as
frequently as every 30 seconds for 1-2 days prior to landfall and for as much as 2 weeks
afterwards. Data at some sites in southwest Louisiana were collected for multiple storms. The
data are available at (URL: http://water.usgs.gov/osw/programs/storm_surge.html).

The USGS first deployed the mobile storm-tide network for Hurricane Rita. As Hurricane Rita
approached the Texas and Louisiana coasts in September 2005, the USGS deployed 32 water-
level and 14 barometric pressure sensors at selected locations from Sabine Pass to Abbeville,
Louisiana, at distances ranging from a few feet of beaches to 30 miles inland. Sensors were
programmed to record data every 30 seconds, placed in 1.25 inch diameter pipes and strapped to
permanent objects such as power poles and piers. Temporary benchmarks were driven into or
chiseled on to the same or near by objects to serve as elevation controls for the sensors and
highwater marks that were expected to result from the flooding.

Sensors were recovered following passage of the storm. Tape-downs from the temporary
benchmarks to the surface of the water and to available high-water marks were made to confirm
sensor accuracy. Traditional line of sight leveling was employed to tie the sensors, the water
marks, and the benchmarks to a common datum and that datum was tied to the NGVD 1988
through use of GPS surveying.


S
E
S
Session 6 S
Tropical Cyclone Model I
Development and
Technology Transfer, O
Part 1 N

6

Advancement of the HWRF for Next Generation Hurricane Prediction at NCEP’s
Environmental Model Center

Naomi Surgi1, Robert E. Tuleya2, Qingfu Liu2, Vijay Tallapragada2, Young Kwon2
(naomi.surgi@noaa.gov)
1
NOAA/NCEP/EMC
2
NOAA/NCEP/EMC/SAIC

Following the initial operational implementation at NCEP of the Hurricane Weather and
Research Forecast system (HWRF) in 2007, upgrades were made to the system for the 2008
hurricane season. These upgrades addressed a known weak bias in HWRF intensity forecasts.
Forecasts from the 2008 season will be highlighted showing the impact of these upgrades.
Further improvements for the 2009 implementation will be highlighted to include advancements
to the GSI and HWRF physics to include a gravity wave drag parameterization. The latter has
helped improve the synoptic flow impacting both track and in some cases intensity with
particularly positive results in EPAC.

Future advancements of the HWRF will also be discussed in terms of advanced coupling to
ocean-wave-land models, data assimilation strategies for both atmosphere and ocean,
formulation of physics for advanced high resolution and prototype high resolution hurricane
ensembles to enhance NHC’s hurricane intensity prediction capabilities.

Plans for bringing the HWRF into the DTC repository for community use will also be discussed
along with a joint hurricane tutorial with NCAR scheduled for January 2010.


Hurricane Model Transitions to Operations at NCEP/EMC
A Joint Hurricane Testbed (JHT) Program

Robert E. Tuleya, Young Kwon, Vijay Tallapragada,
Zhan Zhang, Yihua Wu, Qingfu Liu,
Janna O'Connor and Naomi Surgi
(Robert.tuleya@noaa.gov)

NOAA/NCEP/EMC/SAIC

The emphasis for this 2005-2007 JHT project has concentrated on HWRF development, its
operational implementation, and current and future upgrades. HWRF has been accepted into
operations and has been run successfully for the 2007 and 2008 seasons. This development was
designed to take into account the strengths of the WRF software system, the use of the well
tested NMM dynamic core, and the physics packages of the GFDL highly successful forecast
system. The HWRF system continued to make advancements as shown by the results of the busy
2008 tropical Atlantic season. HWRF was quite competitive with its 2008 track forecasts being
among the best dynamical models for this active Atlantic season. HWRF was comparable to both
the GFDL and GFS forecasts systems with 5 day forecast errors of ~250nm. Furthermore HWRF
performed better in intensity forecasts than the GFDL dynamical model up to 84h and had nearly
the same skill as the baseline Decay SHIPS statistical model throughout the 120h forecast time
period. The SAIC JHT project was an integral contributor to this progress.

Once the model was made operational, this project moved in the direction of solving production
model problems and model improvement. One nagging problem noticed in the 2008 season was
the unrealistic land surface temperatures predicted during the 2008 tropical season in HWRF. It
was discovered that this was a sporadic problem involving spurious radiative surface fluxes. This
was resolved by calling radiation for the nest domain at frequent constant intervals. It is
anticipated that this upgrade will lead to better forecasts of intensity and track during landfall.
Some examples will be shown.

As mentioned, the physics packages have been brought in line with the GFDL model. The
HWRF group is now in position to make improvements and upgrades over and above the
capabilities of the GFDL model. One such example is the implementation of the NOAH LSM
(land surface model) into the HWRF forecast system in test mode. As an additional benefit, the
runoff from the LSM can be used to drive a basic stream flow model. Preliminary results indicate
better rainfall forecasts with less bias. Another example is the inclusion of gravity wave drag into
the operational HWRF for the 2009 season. Preliminary results of these upgrades will be shown.

Besides improving physics, the HWRF group has made advances in developing a high resolution
version of the production version of HWRF. Other work continues on the development of a more
complete diagnostic package so more objective evaluations can be made over and above the
standard evaluations of absolute errors of track and intensity.


Developing Coupled Tropical Cyclone-Wave-Ocean Models for Transition to Operations

Isaac Ginis, Yalin Fan, Richard Yablonsky and Biju Thomas
(iginis@gso.uri.edu)

Graduate School of Oceanography, University of Rhode Island

We will discuss the progress in developing coupled tropical cyclone-wave-ocean models for
operational implementations at NOAA and Navy. The key element of our coupled modeling
approach is the air-sea interface model (ASIM) developed by our research group at URI that
consists of the wave boundary layer model and the air-sea heat and momentum flux budget
model. The ASIM is imbedded into the TC-wave-ocean coupled model and calculates all the flux
boundary conditions for the atmospheric, wave and ocean models.

We will evaluate the performance of the NOAA’s operational WAVEWATCH III under a very
strong tropical cyclone wind forcing. We will compare the model results with field observations
of the surface wave spectra from a scanning radar altimeter, NDBC time series and satellite
altimeter measurements in Hurricane Ivan (2004). The results suggest that the operational model
tends to overestimate the significant wave height and the dominant wave length, and produces a
wave spectrum that is higher in wave energy and narrower in directional spreading. When an
improved drag parameterization is introduced and the wave-current interaction is included, the
model yields improved forecast of significant wave height and wave spectral energy, but
underestimates the dominant wave length.

We will also present the results of numerical simulations of the hurricane induced sea surface
temperature cooling in the vicinity of a warm core eddy. When considering hurricane intensity,
one often neglected aspect of warm core ocean eddies is the preexisting anticyclonic circulation
around the eddy. Depending on the translation speed of the hurricane and the location of the
eddy relative to the storm track, this anticyclonic circulation may impact the location and
magnitude of the hurricane-induced sea surface cooling. One significant result from these model
simulations indicates that when a warm ocean eddy is located to the right of the storm track, the
interaction of the preexisting eddy circulation with the hurricane-induced cold wake can cause, in
some cases, increased sea surface cooling under the core of the storm relative to the case where
no ocean eddy is present at all. Therefore, the presence of a warm core eddy in advance of a
hurricane may in some cases create a less favorable condition for hurricane intensification;
similarly, the presence of a cold core eddy may in some cases create a more favorable condition
for hurricane intensification.


Coupled Hurricane Atmosphere – Ocean Modeling System (HWRF-HYCOM)
and Hurricane Forecast Performance

Hyun-Sook Kim,
and B. Hundermark, D. Iredell, Y. Kwon, L. Liu, Q. Liu, C. Lozano, J. O’Connor, N. Surgi, V.
Tallapragada, B. Tuleya, Z. Zhan

Environmental Modeling Center (EMC)/ NCEP/NOAA

Real-time hurricane forecast, using a coupled Hurricane Weather Research and Forecast model
and HYbrid Coordinate Ocean Model (HWRF-HYCOM) system, has been successfully
conducted in parallel to the NCEP Central Operational (NCO) model. The primary difference
between the HWRF-HYCOM and NCO system is the ocean dynamic model (POM for the NCO
system), and the method for ocean initialization. The ocean horizontal resolution is O(10km)
and O(20km), and the vertical resolution is hybrid 26 layers vs. 23 levels, for HYCOM and POM
respectively. Significant differences between these two systems include 1) the step for the
correction of forecast storm location and intensity in the HWRF-HYCOM; 2) the approach in the
preparation of the ocean initialization: We derive ocean initial and boundary conditions from the
Atlantic-Basin scale, RTOFS (Real-Time Ocean Forecast System) that is currently running once
a day in NCEP with data assimilation. Whereas, the ocean model in the operational utilizes
remotely sensed sea surface temperature to provide meso-scale features imbedded in a GDEM
climatology with their current locations and intensities.

The HWRF-HYCOM hurricane forecast is comparable to that of the NCO system. Forecast
errors computed against a best track are generally less than 200-km for track and  20% for
intensity after five days. Improvements for the HWRF-HYCOM systems are expected,
particularly on the hurricane intensity estimates, by employing better initial conditions and
tuning mixing and bulk parameterizations of air-sea fluxes.


Evaluation and Improvement of Ocean Model
Parameterizations for NCEP Operations: Ivan (2004)

Lynn K. Shay1, G. Halliwell1, J. Brewster1, W. J. Teague2
(nshay@rsmas.miami.edu, ghalliwell@rsmas.miami.edu, jbrewster@rsmas.miami.edu,
teague@nrlssc.navy.mil)
1
Division of Meteorology and Physical Oceanography, RSMAS, University of Miami
2
Department of Oceanography, U.S. Naval Research Laboratory, Stennis Space Center

The ocean response to hurricane Ivan (Sept 2004) was simulated within the Gulf of Mexico
using the Hybrid Coordinate Ocean Model (HYCOM). In this study, the ocean model response to
hurricane Ivan is evaluated against satellite SST measurements and moored ocean current
observations. Hurricane Ivan passed directly over 14 Acoustic Doppler Current Profiler (ADCP)
moorings that were deployed as part of the Navy Research Laboratory Slope to Shelf Energetics
and Exchange Dynamics (SEED) project (Teague et al., JPO, 2007). The model was initialized
by oceanic fields provided by the latest generation of the U. S. Navy ocean nowcast-forecast
system developed at the Naval Research Laboratory. The modeling effort builds upon a previous
NOAA JHT grant that eliminated two mixing schemes from contention leaving Mellor Yamada
(MY), K- Profile Parameterization (KPP) and Goddard Institute of Space Sciences (GISS)
schemes. A baseline experiment is performed that is forced by atmospheric fields from the 27 km
resolution COAMPS model, but with high-resolution wind speed and stress fields obtained from
the HRD HWIND analysis patched in for the storm region. HWIND vector wind fields are first
patched into COAMPS fields, and then wind stress is calculated using bulk formula with the
Donelan et al. (GRL, 2003) drag coefficient prior to model runs. The model is nested within a
GOM data-assimilative hindcast that uses the U. S. Navy NCODA system. It is run with 26
vertical layers and KPP vertical mixing is used. Surface turbulent fluxes are calculated during the
model run using the COARE 2.6 algorithm bulk formula.

Thirteen alternate experiments are run, each differing from this control experiments in a single
aspect to isolate sensitivity to spatial resolutions, vertical mixing schemes, surface drag and
enthalpy coefficients, surface forcing fields, model initializations, and three-dimensional ocean
dynamics, the latter achieved by altering the model code to solve independent one-dimensional
momentum and thermodynamical balances at each grid point. Model sensitivities are evaluated
by calculating the changes in SST forced by Ivan by calculating turbulent heat flux from ocean to
atmosphere following the storm track, and by comparing upper-ocean velocity profiles to SEED
mooring measurements. RMS differences between those produced by the baseline experiment
and each of the other experiments quantify the sensitivity to the individual model property and
parameterization from the baseline experiment. Accurate initialization of ocean features is the
most important single factor for improving the accuracy of the ocean response. The impact of
three-dimensional ocean dynamics on the SST response pattern and on heat flux from ocean to
atmosphere is nearly as large as the initialization of ocean features. Four factors are of
intermediate importance: (1) parameterization of surface momentum flux through the drag
coefficient; (2) choice of vertical mixing scheme; (3) horizontal resolution; and (4) accurate
representation of storm structure in the surface forcing. The least important factor is vertical
resolution. The parameterization of surface heat flux through the sensible and latent heat drag
coefficients has little influence on SST and current response but has a large impact on heat flux
from ocean to atmosphere.


Upgrades to the GFDN Model for 2009 and Beyond

Morris A. Bender1, Isaac Ginis2, Biju Thomas2, Richard Yablonsky2, Carey Dickerman3, and Roger
Stocker3
(Morris.Bender@noaa.gov)
1
Geophysical Fluid Dynamics Laboratory, NOAA
2
Graduate School of Oceanography, University of Rhode Island
3
Fleet Numerical Meteorology and Oceanography Center

Major physics and resolution upgrades to the Navy’s version of the GFDL Hurricane Prediction
System (GFDN) were made operational at the Fleet Numerical Meteorology and Oceanography
Center in October, 2008. These changes included doubling of the finest resolution from 1/6 to 1/12
degree, coupling with the Princeton Ocean Model (POM) in all ocean basins (three-dimensional
coupling in the Atlantic and one-dimensional coupling elsewhere), and incorporation of dissipative
heating and NCEP’s Ferrier microphysics package. In all ocean basins except the Atlantic, the
ocean model in this new GFDN coupled system is now initialized from the real-time NCODA (Navy
Coupled Ocean Data Assimilation) global analysis. NCODA provides more detailed and accurate
representation of the 3-dimensional temperature and salinity fields. These improvements were made
in close collaboration between scientists at GFDL and URI (University of Rhode Island). Since the
new model became operational toward the end of the hurricane and typhoon seasons, their were not
enough cases to evaluate the impact of these changes. However, preliminary testing prior to
operational implementation suggested that these changes should lead to significant reduction in
model errors for both track and intensity.
While this new version of the GFDN forecast system was being tested for operational
implementation, work began on extending the coupling in the western Pacific from one-dimensional
to a full three-dimensional coupled system. With the NCODA global analysis already operational in
GFDN, this important upgrade should enable the model to properly resolve the effect of oceanic
mesoscale features such as warm and cold core rings and the Kuroshio current. Recent studies have
suggested this can lead to rapid intensity changes in typhoons that pass over these features. This new
3-D coupled system will be delivered to FLEET by March, with implementation planned for the
summer. Several examples will be shown of the impact of this upgrade as well as a summary of the
overall improvements in the model forecasts. Developmental work is also continuing to extend the
three-dimensional coupling to the eastern Pacific as well, with possible implementation by later in
the summer.
Future planned improvements of the GFDN model will also be discussed. These include replacing
the GDEM climatology currently used in the Atlantic, with the NCODA 3-D temperature and salinity
analysis already operational in the other ocean basis in GFDN. As increased computer power
becomes available, the finest horizontal grid spacing will be increased from 1/12 to 1/18 of a degree.
Major upgrades to the physics of the air-sea momentum and heat fluxes are also planned for the
GFDN model in 2011, with implementation of the URI air-sea interface model (AISM) that has been
under development at URI for the past several years. Since the air-sea fluxes depend on surface
wave related processes near the air-sea interface and are highly variable in space and time, a fully-
coupled ocean-wave-atmospheric model is being developed, which uses the current coupled system
and NCEP’s WWIII model, together with new strategies for wind-wave-current interactions. These
will be discussed in more detail. Finally, in collaboration with GFDL scientists, the radiation package
in the GFDN model will be upgraded to include interaction with micro-physics species.


Application of COAMPS-TC to TCS-08

Richard M. Hodur1, S. Chen2, J. Cummings3, J. Doyle2, T. Holt2, Y. Jin2, C.-S. Liou2, K.
Sashegyi2, J. Schmidt2
1
Science Applications International Corporation, Monterey, CA (hodur@nrlmry.navy.mil)
2
Naval Research Laboratory, Monterey, CA (chen@nrlmry.navy.mil, doyle@nrlmry.navy.mil,
holt@nrlmry.navy.mil, jin@nrlmry.navy.mil, liou@nrlmry.navy.mil,
sashegyi@nrlmry.navy.mil, schmidt@nrlmry.navy.mil)
3
Naval Research Laboratory, Stennis Space Center, MS (cummings@nrlmry.navy.mil)

A new version of the Coupled Ocean/Atmosphere Mesoscale Prediction System (COAMPS) is
being developed for the prediction of tropical cyclone track, structure, and intensity (COAMPS-
TC). COAMPS-TC will include new analysis methodologies for the initialization of the TC
structure, and improved air-sea flux and microphysics parameterizations to improve structure and
intensity forecasts. The Tropical Cyclone Structure-2008 (TCS-08) field experiment offered an
excellent opportunity to exercise a prototype of COAMPS-TC in a real-time environment for a
number of tropical cyclones. This enabled us to test the robustness, efficiency, and accuracy of
the analysis and prediction capabilities of COAMPS-TC, and to use the results to guide the next
phase of development of COAMPS-TC.

A real-time automated system was set up to enable COAMPS-TC to be run for every tropical
cyclone in the TCS-08 area (e.g., western Pacific) every 12 hours. COAMPS-TC was run using
nested grids. The TCS-08 configuration of COAMPS-TC used a coarse mesh (45 km) that was
fixed in time and large enough in areal coverage so that the same coarse grid was used for all
TCS-08 applications. The two inner grids (15 and 5 km) were centered on each TC at the start of
each forecast, and moved automatically with each storm so that they always remained centered
on the TC. The COAMPS-TC forecast model automatically tracked the position of the TC in
each forecast. All forecasts were run to 72 h.

A total of 96 forecast were generated over 13 different tropical cyclones. Results show that the
prototype of COAMPS-TC produced track forecasts that were competitive with other currently-
used TC numerical forecast aids. In terms of the intensity prediction of TCs, COAMPS-TC was
generally too weak at the analysis time, and tended to intensify the TCs too much during the 72
forecast, with a bias toward too intense at 72 hours.

These findings have dictated what further research and development needs to be done with
COAMPS-TC. We are now testing improvements to our analysis of the TC circulation to better
define the initial strength and structure, and studying further improvements to the air-sea
interaction and moist physics parameterizations to better model TC intensity changes. In
addition, we are working on the integration of an ocean data assimilation system with the
atmospheric components of COAMPS-TC. We were able to run this two-way coupled
COAMPS-TC for a small number TCs during TCS-08 (and for a number of TCs in other basins,
as well) and have integrated this version of COAMPS-TC in our post-TCS-08 research. The set-
up and results of COAMPS-TC for TCS-08, and current research on this system will be
discussed.


S
E
S
Session 7 S
Development and
Part 2 N

7

Ensemble, Corrected Consensus, and Weighted Consensus TC Track Forecasts

James S. Goerss
(jim.goerss@nrlmry.navy.mil)

Marine Meteorology Division, Naval Research Laboratory, Monterey, CA

The interpolated versions of seven high-quality TC track forecast models are routinely available to the
forecasters at NHC. The seven models are: GFS, GFDL, and the Hurricane WRF (AVNI, GFDI, and
HWFI; NCEP); NOGAPS and GFDN (NGPI and GFNI; FNMOC); the UKMO global model (EGRI);
and the ECMWF global model (EMXI). The operational consensus forecast aid, TVCN, is formed by
giving equal weight to the available forecasts from AVNI, GFDI, HWFI, NGPI, GFNI, EGRI, and EMXI.
Based on the exceptional TC track forecast performance in the Atlantic of the ECMWF global model for
all forecast lengths and of the NCEP GFS ensemble mean (AEMI) for 96 h and 120 h, two experimental
consensus forecast aids were constructed and tested for both the Atlantic and eastern North Pacific basins.
A weighted consensus aid (CNW6) was constructed by giving the EMXI forecasts six times the weight
given to the forecasts from the other models used by TVCN. A second consensus aid (CON3) was
constructed by taking the average of the TVCN, EMXI, and AEMI forecasts.

For the Atlantic in 2008, the TC track forecast errors for EMXI were 46 nm, 80 nm, 126 nm, 167 nm, and
209 nm for 24 h, 48 h, 72 h, 96 h, and 120 h, respectively. The respective errors for GFDI, the second
best model, were 56 nm, 96 nm, 139 nm, 200 nm, and 259 nm while those for TVCN were 48 nm, 86 nm,
126 nm, 169 nm, and 210 nm. The EMXI errors were comparable to those for TVCN and about 10-20
percent smaller than those for GFDI. While the TC track forecast errors for AEMI were considerably
larger than those for GFDI and TVCN for forecast lengths less than 96 h, the respective AEMI errors at
96 h and 120 h were 164 nm and 183 nm. The respective forecast errors for corrected consensus, TVCC,
were 47 nm, 86 nm, 127 nm, 172 nm, and 203 nm, comparable to those for TVCN. The respective
forecast errors for CNW6 were 45 nm, 78 nm, 117 nm, 158 nm, and 198 nm, a 6-9 percent improvement
over those for TVCN. For CON3, the respective errors were 49 nm, 84 nm, 122 nm, 146 nm, and 177
nm, comparable to those for TVCN out to 72 h but 14-16 percent smaller at 96 h and 120 h. In summary,
the forecast errors for CNW6 were the smallest for forecast lengths out to 72 h while those for CON3
were the smallest at 96 h and 120 h.

Unlike the Atlantic, the EMXI TC track forecast errors for the eastern North Pacific in 2008 were
considerably larger than those for TVCN, especially at the longer forecast lengths, and the AEMI forecast
errors were considerably larger than those for EMXI. As a result, it was not surprising to see that the
performance of CNW6 and CON3 for this basin was nothing like that for the Atlantic. The TC track
forecast errors for TVCN were 45 nm, 76 nm, 104 nm, 151 nm, and 192 nm for 24 h, 48 h, 72 h, 96 h, and
120 h, respectively. The respective errors for TVCC were 45 nm, 79 nm, 94 nm, 149 nm, and 205 nm,
about 10 percent better than TVCN at 72 h but 7 percent worse at 120 h. For CNW6, the respective errors
were 45 nm, 75 nm, 106 nm, 159 nm, and 216 nm, comparable to those for TVCN out to 72 h but 5-12
percent larger at 96 h and 120 h. The respective errors for CON3 were 48 nm, 87 nm, 131 nm, 176 nm,
and 229 nm, 14-26 percent larger than those for TVCN for forecast lengths greater than 24 h.

The Atlantic results indicate that weighted consensus TC track forecast aids are capable of producing
improved track forecasts if one or more of the models perform considerably better than the others.
However, whether such improved model performance is consistent from season to season is yet to be
determined. The eastern North Pacific results suggest that such performance and the utility of weighted
consensus TC track forecast aids is likely to be basin dependent.


TC Dressing: A Probabilistic Approach to Providing State-Dependent,
Non-Isotropic Forecast Track Error Guidance

Jim Hansen, Jim Goerss
(jim.hansen@nrlmry.navy.mil, jim.goerss@nrlmry.navy.mil)

Naval Research Laboratory

The Goerss Predicted Consensus Error (GPCE) approach has been extended to independently
predicting expected across-track and along-track error. In addition, a new product called the
Probability of Left/Right/Fast/Slow (P-LRFS) has been produced to provide guidance on 3rd
order forecast moments. GPCE-AX and P-LRFS were originally developed using GUNA over
2002-2005 (dependent) and 2006 (independent). GPCE-AX and P-LRFS have been extended to
CONU and to CONW. The CONU results (training over 2002-2007, testing over 2008) indicate
that GPCE-AX generally outperforms GPCE in terms of reliability (the fraction of time
verification is bound by the 70% uncertainty isopleths) and sharpness (the area bound by the
70% isopleths). The CONW results (training over 2004-2006, testing over 2007) are less
conclusive; the sharpness and reliability results for GPCE and GPCE-AX are very similar to one
another.


Using Partnerships to Meet NOAA’s Needs for its Next Generation Storm Surge System

Jesse Feyen1, Frank Aikman1, Mary Erickson1, Hendrik Tolman2, Wilson Shaffer3, Arthur Taylor3,
Marcia Weaks4, John Kuhn5, Keelin Kuipers6, Jaime Rhome7, and Edward Rappaport7
(jesse.feyen@noaa.gov)
1
NOAA/NOS/Office of Coast Survey/Coast Survey Development Laboratory
2
NOAA/NWS/National Centers for Environmental Prediction/Environmental Modeling Center
3
NOAA/NWS/Office of Science and Technology/Meteorological Development Laboratory
4
NOAA/NOS/Integrated Ocean Observing System Program
5
NOAA/NWS/Office of Science and Technology
6
NOAA/NOS/Coastal Services Center
7
NOAA/NWS/Tropical Prediction Center/National Hurricane Center

Storm surge inundation is a main contributor to the damage and loss of life caused by tropical
cyclones, as seen in recent hurricane storm surge events (e.g., 2008’s Ike, 2005’s Katrina, 2004’s
Ivan). Minimizing the impacts of inundation is an important mission that spans many offices. A
NOAA-wide team has been assembled to guide development of its next generation storm surge
system. Planning and research accomplishments in support of this effort will be highlighted.

Customers have demonstrated a need for clear and timely communication of inundation risk;
meeting this mission requires a multi-faceted approach. Fundamental is prediction of the total water
level of inundation at relevant local scales, regardless of source; factors such as surge, waves, and
hydrologic flooding all need to be considered. However, a wide range of users exist in a broad
community that includes forecasters, emergency managers, coastal planners, and first responders.
Requirements for timeliness and accuracy depend on why they are making decisions:
1) coastal land-use & resiliency planning (years to decades before event),
2) emergency management planning (months to years before event),
3) evacuations (days before event),
4) real time response (duration of event), and
5) post-storm recovery (hours to years after event).

Furthermore, clear and effective communication techniques are required to translate technical
products to decision-making activities. Continued development of tools such as maps and
visualizations of depth of water above land are needed to achieve these goals. Additionally,
effective education efforts are needed to ensure proper use and interpretation of products.

To meet this mission the next generation system needs to be flexible and able to consider the need
for growth in areas such as observations, ensemble modeling, probabilistic prediction, and
mapping/visualization. The system will need to link to federal operational observational and
prediction systems. Furthermore, developments need to be systematically vetted within an
operational environment (e.g., using standardized procedures and a modeling test bed).
Advancement will not be possible without partnerships across the community. By leveraging
resources and technologies across regional associations, government agencies, and academic
institutions, advancements can be shared and research can be quickly transferred to operations.
Finally, effective partnering requires a strong foundation in the development and use of industry-
wide standards, models, tools, and benchmarks, a strategy proposed by the NOAA team.


Sensitivity of the HWRF model prediction for Hurricane Ophelia (2005)
to the choice of the cloud and precipitation scheme

Yuqing Wang and Qingqing Li
(yuqing@hawaii.edu)

International Pacific Research center, University of Hawaii at Manoa,

To demonstrate how the model cloud and precipitation physics may affect the hurricane track,
structure, and intensity prediction by the HWRF model, we conducted several sensitivity
numerical experiments designed to show how the simulation would be different if the cumulus
convective parameterization was used or omitted from the inner nested mesh and how such a
difference is sensitive to the choice of convective parameterization. The model resolution is the
same as that used in the operational HWRF model, namely, about 9 km for the nested domain
and 27 km for the outer domain. But we used a relatively larger domain for the inner nest than
that used in the operational forecast model. Two parameterizations for cumulus convection tested
are the Kain-Fritsch (KF) scheme and the Betts-Miller-Janjic (BM) scheme. Other model physics
include the GFDL shortwave and longwave radiation schemes, the Mellor-Yamada-Janjic TKE
scheme for the planetary boundary layer (PBL), and the LOAH land surface model. These are
identical in all experiments. The case of Hurricane Ophelia (2005) was chosen for a
demonstration. The model was initialized at 00 UTC 09 September 2005 and integrated for 96
hours. The initial and the lateral boundary conditions were defined using the NCEP GFS global
final analysis (FNL).

In this talk, we will demonstrate how the model cloud and precipitation physics may affect the
hurricane track, structure, and intensity prediction by the HWRF model. We will show that the
simulated storm track, intensity, and structure were largely affected by the choosing of the
convective parameterization scheme either used for the inner mesh or just the outer mesh or both
meshes. It was not our attempt to determine which combination is the best since no single
simulation has performed the best in all aspects of the observed hurricane. Although we only
show the simulations for Hurricane Ophelia (2005), the results strongly suggest a need for a
detailed evaluation of the cloud and precipitation physics used in the operational HWF model
and to understand how the discrepancies affect the storm motion, intensity, and structure. Since
some systematic biases have been reported in the operational HWRF model, improvements of
the representation of the cloud and precipitation physics would lead to improved prediction skill
of hurricane track, structure, and intensity by the HWRF model at NCEP/EMC.


Experiments of Hurricane Initialization with Airborne Doppler Radar Data for the
Advanced-research Hurricane WRF (AHW) Model

Qingnong Xiao1, Xiaoyan Zhang1, Christopher Davis1, John Tuttle1, Greg Holland1, and Patrick
J. Fitzpatrick2
(hsiao@ucar.edu)
1
ESSL/MMM, National Center for Atmospheric Research, Boulder, Colorado
2
Northern Gulf Institute, The Mississippi State University, Stennis Space Center, Mississippi

Initialization of the hurricane vortex in weather prediction models is vital to intensity forecasts
out to at least 48 h. Airborne Doppler radar (ADR) data has sufficiently high horizontal and
vertical resolution to resolve the hurricane vortex and its imbedded structures, but has not been
extensively used in hurricane initialization. Using the Weather Research and Forecasting (WRF)
three-dimensional variational (3D-Var) data assimilation system, the ADR data are assimilated to
recover the hurricane vortex dynamic and thermodynamic structures at the WRF model initial
time. The impact of the ADR data on three hurricanes, Jeanne (2004), Katrina (2005) and Rita
(2005), are examined during their rapid intensification and subsequent weakening periods before
landfall.

With the ADR wind data assimilated, the three-dimensional winds in the hurricane vortex
become stronger and the maximum 10-m winds agree better with independent estimates from
best track data than without ADR data assimilation. Through the multivariate incremental
structure in WRF 3D-Var analysis, the central sea level pressures (CSLPs) for the three
hurricanes are lower in response to the stronger vortex at initialization. The size and inner core
structure of each vortex are adjusted closer to observations of these attributes. Addition of
reflectivity data in assimilation produces cloud water and rainwater analyses in the initial vortex.
The temperature and moisture are also better represented in the hurricane initialization.

We conduct 48-h forecasts to evaluate the impact of ADR data using the Advanced-research
Hurricane WRF (AHW), a derivative of the Advanced Research WRF (ARW) model.
Assimilation of ADR data improves the hurricane intensity forecasts. Vortex asymmetries, size,
and rain bands are also simulated better. Hurricane initialization with ADR data is quite
promising towards reducing intensity forecast errors at modest computational expense.


Recent Trends in Dynamical Medium-Range Tropical Cyclone Track Prediction and the
Role of Resolution Versus Physics in the ECMWF Model

Mike Fiorino

Assimilation and Modeling Branch
Global Systems Division
NOAA Earth System Research Laboratory
Boulder, Colorado

A commonly held view in the tropical cyclone (TC) modeling community is that dynamical TC
track and intensity prediction requires: 1) high-resolution (order 1 km); 2) the assimilation of
detailed observations in the inner core; and 3) ensembles of independent deterministic and
perturbed models. However, the recent performance of the operational high-resolution run of
the ECMWF global model challenges these notions and suggests that the critical and dominant
factor in numerical TC prediction skill is the modeling of physical processes. In this paper we
analyze how model changes, in both resolution and in the parameterization of key physical
processes, impact the model forecasts of the global tropical wind field and TC tracks. Since a
change in the cumulus convection scheme in November 2007, the ECMWF TC track forecasts
have outperformed multi-model consensus by 20% globally in the medium-range (72 h). We
compare this finding to trends in model performance vis-à-vis consensus in both the Atlantic and
western North Pacific basins for the period 1992-2008 and demonstrate that this
gain/improvement by the ECMWF model is unprecedented. The results will be updated to
include the 2009 southern Hemisphere season to see if the trend persists and the talk will
conclude with a number of opinions regarding the way forward for improved numerical
hurricane forecasts.


S
E
S
Session 8 S
Development and
Part 3 N

8

Model Development for Typhoons at the Japan Meteorological Agency

Munehiko Yamaguchi, Takuya Komori, Masashi Nagata and Tetsuo Nakazawa
(myamaguchi@rsmas.miami.edu)

Japan Meteorological Agency
(now at Rosenstiel School of Marine and Atmospheric Science, University of Miami)

The Japan Meteorological Agency’s numerical model framework for typhoon prediction is
presented. Two systems are used for typhoon forecasting; one is a high-resolution global spectral
model, which became operational in November 2007; and the other is an ensemble prediction
system, which was implemented in February 2008. The global model has a resolution of
TL959L60 (about 20 km in the horizontal and 60 vertical levels), and runs 4 times a day with a
prediction range up to 216 hours. The data assimilation system is the 4 dimensional variational
system (4DVAR), which has been in operation since 2005, and a typhoon bogus technique is
adopted. The Typhoon Ensemble Prediction System has a resolution of TL319L60 (about 60 km
in the horizontal and 60 vertical levels), and runs 4 times a day with a prediction range of 132
hours. The ensemble size is 11 and a singular vector method is employed to create initial
perturbations.

In the presentation, the verification results of the two systems in 2008 will be presented as well
as the detailed system configuration. In addition, recent participation by JMA in activities such
as THORPEX Pacific Asian Regional Campaign will be briefly introduced.


The Hurricane Research System: Development of a High Resolution, Non-Hydrostatic
Model for Addressing the Need for the Next Generation of Improved Tropical Cyclone
Intensity Prediction

S.G.Gopalakrishnan1, Xuejin Zhang2 , Kevin Yeh2, Robert Rogers1, Jian-Wen Bao3, Frank
Marks1, Sim Aberson1 and Robert Atlas4
(gopal@noaa.gov)
1
NOAA/AOML, Hurricane Research Division, Miami, Florida
2
CIMAS, University of Miami, Miami, Florida
3
NOAA/ESRL, Physical Science Division, Boulder, Colarado
4
NOAA/AOML, Miami, Florida

The hurricane forecast improvement project (HFIP) is a unified NOAA approach to guide and
accelerate improvements in hurricane track, intensity and structure forecasts, with an emphasis
on rapid intensity change. An integral component of the HFIP will be the development of
improved coupled atmosphere-ocean-land surface, high-resolution non-hydrostatic regional
models. The HFIP model development effort will pursue a two-stream development path: one
path that builds on current operational capabilities consistent with current and future computer
limitations for operations, and a second development path that will focus on research using high
resolution models (grid lengths of 1-3 km) which may be necessary to properly resolve the inner-
core processes that influence intensity changes. Developed from NCEP's WRF-NMM/HWRF
modeling system (Gopalakrishnan et al., 2006), the Hurricane Research System (HRS) is the
next generation, high-resolution hurricane modeling system designed for improving forecasts
related to the rapid intensity change problem. This modeling system is currently being developed
at NOAA’s OAR labs (specifically at AOML and ESRL). The development of
This system will be outlined here.

A baseline configuration for the HRS was tested at the current operational moving grid
resolution of 9 km for most of the hurricane season of 2008. Some results from these runs will be
presented. In addition, we have produced forecasts at 3 km grid length from 69 previous
hurricane forecasts from 200?-2007, selected for the challenging nature of their intensity
evolution. We see significant improvements in track, intensity and structure between forecasts
obtained at the operational resolution of 9 km and at the higher resolution of 3 km. The results
from some of these cases will be presented. Development of the third nest valid for inner core
structure prediction down to 1 km resolution is ongoing.

Reference: Gopalakrishnan, S.G., N. Surgi, R. Tuleya, and Z. Janjic, 2006 : "NCEP's Two-way-
Interactive-Moving-Nest NMM-WRF modeling system for Hurricane Forecasting", 27th
Conference on Hurricanes and Tropical Meteorology, 24-28 April 2006, Monterey, California


Impact of Horizontal Resolution on Rapid Intensity Change Forecasts using the Hurricane
Research System
Robert Rogers1, S. Gopalakrishnan1, K. Yeh2, and X. Zhang2
(Robert.Rogers@noaa.gov)

1
NOAA/AOML Hurricane Research Division, Miami, FL
2
CIMAS, University of Miami, Miami, FL

Advances in the prediction of tropical cyclone (TC) intensity and structure have lagged behind
advances in track prediction. A primary reason for this lag is the multiscale nature of the
physical processes relevant for intensity and structure changes. Such processes range in spatial
scale from 1000’s of km to mm and in temporal scale from days to seconds. Numerical models
are a key tool used to understand and predict TC structure and evolution. Only within the past
several years, however, have cloud-resolving models (grid length of O(1 km)) become
commonplace. Such resolution is believed to be important because it is not until that resolution
is reached that the deep convective cores, which produce a large amount of the diabatic heating
and mass flux within the TC core, are explicitly resolved.

Tests of the impact of horizontal resolution on the ability of numerical models to reproduce rapid
intensity change are shown here. The Hurricane Research System (HRS) model, developed from
NCEP's WRF-NMM/HWRF modeling system and adapted by NOAA/AOML and
NOAA/ESRL, has been run at 9-km and 3-km grid lengths on 69 hurricane cases selected for the
challenging nature of their intensity evolution. Results from select cases that had airborne
Doppler radar coverage and demonstrated rapid intensity change will be shown. The inner-core
vortex- and convective-scale structures will be compared with the radar data for these cases in an
attempt to identify model skill and biases and how they vary as a function of resolution.


Impact of Sea Spray on the Balance of Turbulent Kinetic Energy
in the Hurricane Surface Boundary Layer

Jian-Wen Bao, Christopher W. Fairall, Sara A. Michelson, Laura Bianco
(Jian-Wen.Bao@noaa.gov)

NOAA/Earth System Research Laboratory

The feedback effects of sea-spray on the mean wind, temperature and moisture profiles in the
surface boundary layer associated with tropical cyclones is investigated using a 1-D coupled sea-
spray and surface boundary layer (SBL) model. This model is capable of simulating the
microphysical aspects of evaporation of saline water droplets of various sizes and their dynamic
and thermal interaction with the turbulence mixing that is simulated by the Mellor-Yamada 1.5-
order closure scheme. The sea-spray droplet generation is described by a state-of-the-art
parameterization which predicts the size spectrum of sea-spray droplets at a given surface stress
(or wind speed). The results from a series of simulations reveal salient characteristics of the way
in which evaporating droplets of various sizes modify the turbulence mixing near the surface,
which in turn affects further droplet evaporation.

Based on these results, a bulk parameterization scheme of sea-spray mediated air-sea heat and
momentum fluxes has been developed at NOAA/ESRL. Results from testing the scheme with
the current operational setup of the HWRF model indicate that the scheme may be used as a way
to reduce the weak intensity bias since the scheme’s impact on hurricane track prediction is small
enough that it can be neglected. It is found that because the very turbulence that transports
momentum and heat across the air-sea interface is also responsible for the generation of sea
spray, the resultant effect of sea spray on both the momentum and heat fluxes is complicated by
the interaction between the sea spray, turbulence and mean flow.


Evaluation of Planetary Boundary Layer Parameterizations in Tropical Cyclones by
Comparison of in-situ Observations and High-Resolution Simulations of Hurricane Isabel
(2003)

David S. Nolan, Jun. A. Zhang, and Daniel P. Stern
(dnolan@rsmas.miami.edu)

Rosenstiel School of Marine and Atmospheric Science, Division of Meteorology and Physical
Oceanography, University of Miami

The planetary boundary layer (PBL) depicted in high resolution simulations of Hurricane Isabel
(2003) is studied and evaluated by direct comparison with in-situ data obtained during CBLAST.
Two schemes for parameterizing turbulence in the PBL are evaluated: the Yonsei University
(YSU) scheme and the Mellor-Yamada-Janjic (MYJ) scheme. Evaluation of these schemes is
useful since they both widely used and are based entirely different methods for representing the
PBL. Overall, the tracks and intensities predicted by the two schemes are very similar. The peak
surface wind speeds are substantially improved for both schemes when they are modified to have
more accurate formulas for the roughness of the ocean surface as a function of wind speed.

The PBL schemes are also evaluated by direct comparison of boundary layer profiles of wind,
temperature, and humidity to those observed from dropsondes. Overall, both schemes reproduce
the boundary layer structures remarkably well, even more so when the improved formula for
surface roughness is used. A consistent difference between the YSU and MYJ schemes is an
excessive dissipation of low-level momentum in the MYJ scheme, leading to an excessively
large radial inflow and overall secondary circulation. The relationship between the simulated
instantaneous surface wind speeds and 1 minute mean wind speeds is also explored, with
surprisingly different results between the two schemes.


An Integrated Tropical Cyclone Information System for Research

S. Hristova-Veleva1, Y. Chao1, Z. Haddad1, B. Knosp1,
B. Lambrigtsen1, P. P. Li1, D. Vane1, Q. A. Vu1
(svetla.veleva@jpl.nasa.gov)
1
Jet Propulsion Laboratory, Pasadena, CA 91109

In spite of recent improvements in hurricane track forecast accuracy, currently there are still many unanswered
questions about the physical processes that determine hurricane genesis, intensity, track and impact on large-
scale environment. Furthermore, a significant amount of work remains to be done in validating hurricane
forecast models, understanding their sensitivities and improving their parameterizations. None of this can be
accomplished without a comprehensive set of multiparameter observations that are relevant to both the large-
scale and the storm-scale processes in the atmosphere and in the ocean.

To address this need, we and our colleagues from Marshall Space Flight Center have developed the framework
for a comprehensive Integrated Tropical Cyclone Information System (iTCIS) of high-resolution satellite,
airborne and in-situ observations and model outputs pertaining to: i) the thermodynamic and microphysical
structure of the storms; ii) the air-sea interaction processes; iii) the larger-scale environment as depicted by the
SST, ocean heat content and the aerosol loading of the environment. Our goal was to create a one-stop place
to provide the researchers with an extensive set of observed hurricane data, and their graphical representation,
together with convection-resolving model output, all organized in an easy way to determine when coincident
observations from multiple instruments are available.

Currently the iTCIS is populated with satellite observations of all tropical cyclones observed globally during
2005. Our future plans are to extend the database both forward in time till present as well as backward to
1998. Furthermore, NASA field campaign data will be included and displayed using Goggle Earth field-
campaign-specific approach that allows the creation of overlays on-demand. This innovative approach, known
as the Real Time Mission Monitor (RTMM) has been developed at the Marshall Space Flight Center. It has
proven very valuable during past field campaigns, both during the campaign as well as in the post-campaign
analysis. The RTMM is described in a separate presentation.

We began the development of analysis tools that will communicate with the database and will allow for online
investigations without the need to download large data volumes. The analysis tools will be used to compute
single and multiparameter statistics and to determine spatial, temporal and multiparameter covariances that are
needed to evaluate model performance, provide information for data assimilation, and characterize and
compare observations from different platforms.

We envision that the developed hurricane information system will help in the validation of the hurricane
models, in the systematic understanding of their sensitivities and in the improvement of the parameterizations
employed by the models. Furthermore, it will help in studying the physical processes that affect hurricane
development and impact on large-scale environment.

This talk will describe the structure of the satellite-based component of the iTCIS, the analysis system and the
future plans. A poster will accompany the talk. It will illustrate how the iTCIS can be used to discriminate
between high resolution WRF model simulations that employ different parameterizations with the goal to
determine the model setup that produces most realistic simulations.

The work described here was performed at the Jet Propulsion Laboratory, California Institute of Technology,
under contract with the National Aeronautics and Space Administration.


S
E
Session 9 S
Other Research to
Improve the Prediction of
S
Tropical Cyclone Intensity I
and Structure, Track, O
Precipitation, Coastal, and
Inland Inundation,
N
Part 1
9

Differential O2 Absorption Barometric Pressure Radar (DIAR_BAR):
Improvements in Tropical Storm Forecasts

Qilong Min1, Bing Lin2, Yongxiang Hu2, Wei Gong1, Steven Harrah2, Roland Wes Lawrence3,
Dion Fralick2
1
State University of New York, Albany, NY 12203
2
NASA Langley Research Center, Hampton, VA 23681
3
Old Dominion University, Norfolk, VA 23529

Currently, sea surface air pressure measurements can only be observed from in situ
measurements using buoy, ship and dropsonde techniques, which are sparse in spatial coverage
and expensive to implement. There are no pressure remote sensing methods available even in
experimental stages. This study considers use active microwave systems to obtain the
differential O2 absorption at 50-56 GHz bands to fill the observational gap. The numerical
simulation results show that the rms errors of the instantaneous sea surface pressure estimates
can be as low as 4 mb. With multiple radar measurements the uncertainty in radar sea surface
pressure estimates would drop to about 1 mb which is similar to conventional in situ buoy
measurements. A prototype system of the DIfferential O2 Absorption Radar for Barometry
(DIAR-BAR) has been developed and integrated it into an aircraft. The flight test results show
the instrument performance meets or exceeds the all system requirements indicting the concept
of the differential O2 absorption for the pressure remote sensing measurements could be realized
with existing radar technologies.

Motivated by the potential application of DIAR-BAR, the importance of using surface pressure
distribution in three dimensional variational data assimilation system for WRF is investigated in
a case study of hurricane Katrina. Hurricane Katrina crossed Florida peninsula on August 26,
2005, which provides us surface pressure data with good spatial distribution that is rarely
available for other hurricanes in their developing stage over the ocean. Various patterns of
surface pressure distribution are assimilated, and the bias in landfall position for 84-hour
forecasting decreases from 360 km in the control run up to 38 km in assimilation runs. It
demonstrates the DIAR-BAR system will have great potential for weather observations and other
meteorological applications, especially for forecasts of hurricanes.


Astronomic Tides, Flows, and Hurricane Storm Surge Modeling of the Pascagoula River,
Mississippi

Naeko Takahashi, Qing Wang, and Scott C. Hagen, PhD, PE (Presenting Author)
(shagen@mail.ucf.edu)


This study presents simulated astronomic tides, flows and storm surges for the Pascagoula River
located in lower Mississippi along the Gulf of Mexico. The work is motivated by the absence of
an existing model that can accurately describe storm tide propagation up the Pascagoula River
and out of its banks. The research described herein directly incorporates a full accounting of the
hydraulic conditions for flood/river forecasting, including representation of inundation areas.
Astronomical and meteorological tides are obtained from computations performed with
ADCIRC-2DDI a two-dimensional depth integrated hydrodynamic circulation code.

The full Western North Atlantic Tidal (WNAT) model domain includes the Atlantic Ocean west
of the 60 degree west Meridian, the Caribbean Sea and Gulf of Mexico and is modified to
incorporate the Pascagoula River and its tributaries. The domain is described with an
unstructured finite element mesh of triangles that range in side length from approximately five
meters within the river banks to approximately 160 kilometers in the deep ocean. Recently
acquired LiDAR data is used to build the overland DEM and National Landcover Data is
interpreted to characterize the bottom friction. In addition, the atmospheric wind models that
provide wind input to the ADCIRC surge model produce wind speeds that assume open-ocean
marine conditions; therefore, for overland and nearshore areas the wind speeds are adjusted via
wind-reduction factors.

In addition, a shelf-based domain for the Pascagoula River system is developed that can produce
results comparable to the large-scale domain. The simulated results show good agreement with
the large scale domain approach, recorded storm data and observed high water elevations. It is
revealed that the shelf-based model is capable of accurately predicting storm tides generated by
hurricanes along the coastal areas provided the shelf-based model receives accurate elevation
forcing at its open boundary. This work will enable the Lower Mississippi River Forecast Center
in their effort to provide forecasts for this complicated riverine system.


Toward Modeling of River-Estuary-Ocean Interactions to Enhance Operational River
Forecasting in the NOAA National Weather Service

Hassan Mashriqui1, Seann Reed1, Cecile Aschwanden1,2
(hassan.mashriqui@noaa.gov)
1
NOAA, National Weather Service, Office of Hydrologic Development
2
Wyle Information Systems

NOAA National Weather Service (NWS) uses hydrologic models and one-dimensional (1D)
hydraulic routing models to forecast river flows and stages at over 4,000 locations in the United
States. In coastal zones where rivers widen and flow into estuaries, 1D modeling does not
sufficiently represent the two or three-dimensional (2D/3D) hydraulic conditions. Water levels
near river mouths and in estuaries vary in two dimensions depending on tides, winds, freshwater
inflows, and other forcing factors. A logical next step in improving the availability and accuracy
of water forecasts in coastal areas is to develop the capability to robustly couple 1D river models
and 2D/3D estuary/ocean operational models. To achieve the coupling, 2D/3D ocean and
estuary models would receive freshwater inflows from 1D models, and 1D river models would
receive stage boundary conditions from the 2D/3D models.

In this presentation, we will explain the suitability of several 2D/3D, Chesapeake Bay water
forecasting models for this two way coupling. We will evaluate models based on several criteria,
including but not limited to mesh size requirements, computational requirements, wetting and
drying functionality, ability to ingest freshwater flows, model stability under a wide range of
conditions, ease of use, and accuracy.


Sensitivity of an Asymmetric Gradient Wind Model to Sources of Tropical Cyclone Track
Information and Implications for Storm Surge Prediction
CRAIG A. MATTOCKS and CRISTINA FORBES
(cmattock@email.unc.edu) (cforbes@email.unc.edu)
Institute of Marine Sciences, University of North Carolina, Morehead City, NC
The sensitivity of an asymmetric parametric statistically larger than the NHC Best Track
gradient surface wind model to three different radii, both significantly underestimate the
sources of tropical cyclone track information extent of the tropical storm force wind area
is examined to ascertain the impact on (Moyer et al., 2007). Since the magnitude and
predictions of storm surge. The central and expanse of the water surface elevations
background pressures, wind speed intensity produced by storm surge prediction models
and significant (34-, 50-, and 64-kt) wind radii strongly depend on the accuracy of the wind
are obtained from: (1) National Hurricane forecast guidance, it is imperative to quantify
Center (NHC) forecast advisories and the response of the coastal ocean model to
Automated Tropical Cyclone Forecast different sources of wind forcing information.
(ATCF) Best Track data, (2) NOAA H*Wind Therefore, verifications of the simulated wind
analyses, and (3) Colorado State University speeds and phases against their real
(CSU) Multiplatform Satellite Surface Wind meteorological counterparts, of ocean model
Analyses. Employed in a real-time, event- elevations against actual sea surface elevations
triggered storm surge forecasting system measured by NOAA tide gauges, and of
(Mattocks and Forbes, 2008), these parametric simulated depth-averaged current velocities
winds are generated from NHC forecast against Acoustic Doppler Current Profiler
advisories the moment they are inserted into (ADCP) data were performed. The results
the real-time weather data stream, maximizing provide insights on additional operational data
the number of hours of forecast utility. The that is needed to improve the accuracy of
winds are directly coupled to the ADCIRC storm surge predictions during tropical storm
ocean model at every time step and calculated landfall events.
at exact finite-element mesh node locations. A
REFERENCES
directional surface roughness parameterization
that modulates the wind speed at a given Mattocks C. and C. Forbes, 2008: A real-time,
location based on the types of land cover event-triggered storm surge forecasting
encountered upwind from 72 different wind system for the state of North Carolina, Ocean
directions (5° increments) is used to smooth Modelling, 25, 95-119.
the transition of the winds over patchy terrain Moyer, A.C., J.L. Evans, and M.D. Powell,
and across land-water boundaries. Although 2007: Comparison of observed gale radius
H*Wind wind radii have been found to be statistics, Meteorol. Atmos. Phys., 97, 41-55.
FIG 1. Comparison of TS
Hanna wind fields produced
by the asymmetric gradient
wind model using tropical
cyclone track information
from (a) NHC, (b) H*Wind
analysis, and (c) CSU
Multiplatform Satellite
Surface Wind Analysis.


Surface Windfields and Roughness Derived from Mobile Radar Winds

Joshua Wurman, Karen Kosiba
(jwurman@cswr.org, kakosiba@cswr.org)

Center for Severe Weather Research

It is advantageous to obtain measurements of potentially damaging hurricane winds over a large
portion of the landfall area. While an array of anemometers can provide several point
measurements within this area, radar observations allow for the quantification of the two
dimensional wind field over a broad area. However, radar measurements are obtained at
nonstandard heights, and at large heights, are not substantially modulated by surface roughness.
The purpose of this research is to demonstrate the utility of mobile radars in the mapping of a
standard wind field in hurricanes, which will benefit both the engineering and meteorological
communities.

Using a standard logarithmic profile, the 10-m wind speeds (U10) were derived from the
observed Doppler velocities. For the initial analysis, a roughness length of 0.03 m was assumed.
The 10-m winds were calculated at each gate yielding a continuous two-dimensional sector. In
addition, winds at 1, 2, 3.5, and 10 m measured by in situ anemometers are compared with radar
derived winds and surface roughness is deduced.

This standardization procedure was applied to Hurricanes Gustav (where a DOW and several
instruments were deployed near Houma, Louisiana, and Ike (2008) where a DOW and several
instruments were deployed in and near Galveston, Texas. In both hurricanes, the DOW sampled
onshore flow within 5 km of the coast. The10-m wind maps derived for these hurricanes are
shown below (Figures). In both hurricanes, a correction factor was applied to the initial
observations to correct the magnitude of the off-axis wind speeds.

Caption: Raw (left) and 10-m Zo=0.03 wind field (right) at one time in Galveston during Ike.


Performance of an Objective Model for Identifying Secondary Eyewall Formation in
Hurricanes

Matthew Sitkowski1, Jim Kossin2
(sitkowski@wisc.edu, James.Kossin@noaa.gov)
1
UW-CIMSS, Madison, WI
2
NOAA/NESDIS/National Climatic Data Center, Madison, WI

The formation of a second eyewall around a preexisting smaller eyewall is generally the
precursor to an eyewall replacement cycle. These cycles often cause dramatic and rapid changes
in intensity and are very important to recognize in a forecasting setting, particularly when a
hurricane is approaching land. Additionally, the formation of a secondary eyewall occurring
away from land can broaden the hurricane wind field and influence the magnitude of storm
surge. An objective forecast algorithm, based on environmental and geostationary satellite
features, was developed using a Bayesian approach that provides significant skill in forecasting
the likelihood of secondary eyewall formation (SEF, Kossin and Sitkowski 2009). A beta
version of the algorithm was run in real-time during the 2008 Hurricane Season and those results
will be presented.

The algorithm requires the knowledge of prior SEF events (climatology) to output a probability.
Over 4,500 microwave images in combination with aircraft, radar, and additional satellite data
(when available) were used to identify 45 SEF cases in the North Atlantic basin from 1997-2006.
Nearly one third of all hurricanes developed a secondary eyewall at least once during their
lifetime. Comparatively, 70% of major hurricanes (Saffir–Simpson category 3–5) were observed
to form secondary eyewalls at least once during their lifetimes. Secondary eyewalls were
observed to form during each month of the hurricane season except June, while more than 60%
of the total number were observed during September.

The mean intensity at the time of SEF was 109 kt. Intensity changes preceding and following
SEF were varied. The limitations of using the best track to discern intensity changes associated
with the SEF events, and a strategy to apply reconnaissance data to the problem, will be
discussed.

Kossin, J. P., and M. Sitkowski, 2009: An objective model for identifying secondary eyewall
formation in hurricanes. Mon. Wea. Rev., to appear.

Early Online Release available here:
http://ams.allenpress.com/perlserv/?request=get-abstract&doi=10.1175%2F2008MWR2701.1


S
E
Session 10 S
Other Research to
Improve the Prediction of
S
Tropical Cyclone Intensity I
and Structure, Track, O
Precipitation, and Inland
Inundation,
N
Part 2
10

Evaluation of Hurricane Mitigation Hypotheses through an Interactive Program of
Observational Analyses and Numerical Simulations

Dr. William L. Woodley1, Dr. William Cotton2, Dr. Joe Golden3, Dr. Alexander Khain4, Dr. Daniel
Rosenfeld4, Dr. Isaac Ginis5
(williamlwoodley@cs.com; cotton@atmos.colostate.edu; Joe.Golden@noaa.gov;
Khain@vms.huji.ac.il; daniel.rosenfeld@huji.ac.il; iginis@gso.uri.edu)
1
Woodley Weather Consultants,
2
Colorado State University,
3
Cooperative Institute for Research in the Environmental Sciences (CIRES),
4
Hebrew University of Jerusalem,
5
University of Rhode Island

The disastrous hurricane season of 2005 and the devastation of Hurricane Katrina have refocused
attention on the hurricane threat in the United States. An Effort under the Department of Homeland
Security (DHS) to Develop Concepts, Methodologies and/or to Anticipate, Prepare for and/or Mitigate the
Impact of Catastrophic Geophysical Phenomena is underway. The thrust of this program is to reduce the
intensity of Hurricanes before landfall.

This hurricane mitigation program will involve highly interactive hurricane observations and simulations
in order to identify and evaluate viable hurricane mitigation hypotheses. Cutting-edge instruments will be
employed in a hurricane environment in order to: 1) measure the variability of supercooled liquid water
contents in the rainbands, 2) identify cold pools that might result from evaporation of small drops and
updrafts and 3) sample the aerosols, including those acting as cloud condensation nuclei (CCN) within the
hurricane. The measurements made within and around hurricanes will provide the input for the
simulations and serve as the basis of assessing the validity of the results of the simulations.

Because there is no single best model, this program will involve interactions among several models. The
NOAA GFDL ocean-tropical cyclone atmosphere model looks at altering sea-surface temperatures
(SSTs); the Hebrew University of Jerusalem (HUJ) model incorporates explicit bin microphysics with an
aerosol budget where evaporated drops return the aerosols to the atmosphere; the RAMS model uses bulk
microphysics which emulates a full bin model like the HUJ. It has the advantage that its computational
costs are about 1/100 that of the full bin approach yet it explicitly predicts sources of aerosol, transport
and dispersion, activation in clouds, and their regeneration after droplet/ice particle evaporation.

These models are proposed to be combined in coupled atmosphere-ocean simulations that will aim at
producing realistic simulations of hurricane clouds, precipitation, cold-pools, air-sea interactions, sea
spray, and factors affecting tropical cyclone intensity, including the eye wall replacement process. This
will be done for the dual purpose of further improving hurricane forecasts with the added and more
accurate simulations of physical processes and for testing the sensitivity of the storm to the introduction
of various aerosols for both microphysical and radiative effects (e.g.., ultra-fine hygroscopic and/or black
carbon aerosols at various places within the storm).

This research will contribute to NOAA’s excellent work on hurricane track and intensity predictions. The
goal is to reduce the intensity of the winds once they do hit landfall and therefore reduce the wind caused
damage as the storm crosses land. Damage will not be prevented, but the financial savings and potential
to save lives will make the research extremely beneficial.


Review of NOAA Intensity Forecasting Experiment (IFEX) 2008 Accomplishments and
Preliminary Plans for 2009

Eric Uhlhorn1, Frank Marks1, John Gamache1, Sim Aberson1, Jason Dunion2
(Eric.Uhlhorn@noaa.gov)
1
2
University of Miami/CIMAS

A review of the 2008 hurricane season’s NOAA Intensity Forecasting Experiment (IFEX)
accomplishments will be presented, and plans for the upcoming 2009 season will be discussed.


The NHC Visiting Scientist Program

Chris Landsea
(Chris.Landsea@noaa.gov)

NOAA/NWS/National Hurricane Center, Miami

During the 2008 hurricane season, NHC hosted eleven visiting scientists (researchers and outside
forecasters) for up to a week's time each. The goals of the program are to: 1) facilitate better
understanding by researchers/outside forecasters of the NHC hurricane forecasting process
including the tools and techniques utilized by the Hurricane Specialists; and 2) open additional
dialog between NHC and the research/outside forecast community that could lead to
improvements in our analysis and prediction methodologies. The presentation will discuss the
format of the program, who participated, what was the feedback of both participants and hosts of
the program, and how the project may evolve during the upcoming 2009 season.


Using NASA’s Real Time Mission Monitor to Manage and Conduct Airborne Science
Missions

Michael Goodman1, Richard Blakeslee1, John Hall2, Yubin He2, Kathryn Regner2 and Helen
Conover2
(michael.goodman@nasa.gov, rich.blakeslee@nasa.gov, john.hall@nasa.gov, mhe@itsc.uah.edu,
kregner@itsc.uah.edu, hconover@itsc.uah.edu)
1
National Aeronautics and Space Administration
2
University of Alabama in Huntsville

The Real Time Mission Monitor (RTMM) is a visualization and information system that fuses
multiple Earth science data sources, to enable real time decision-making for airborne and ground
validation experiments. Developed at the National Aeronautics and Space Administration
(NASA) Marshall Space Flight Center, RTMM is a situational awareness, decision-support
system that integrates satellite imagery, radar, surface and airborne instrument data sets, model
output parameters, lightning location observations, aircraft navigation data, soundings, and other
applicable Earth science data sets. The integration and delivery of this information is made
possible using data acquisition systems, network communication links, network server resources,
and visualizations through the Google Earth virtual globe application.

RTMM has proven extremely valuable for optimizing individual Earth science airborne field
experiments. Flight planners, mission scientists, instrument scientists and program managers
alike appreciate the contributions that RTMM makes to their flight projects. We have received
positive feedback from a broad spectrum of scientists who used RTMM during field campaigns
including the hurricane-focused 2006 NASA African Monsoon Multidisciplinary Analyses
(NAMMA), 2007 NOAA-NASA Aerosonde Hurricane flight, 2007 Tropical Composition,
Cloud, and Climate Coupling (TC4), 2008 Arctic Research of the Composition of the
Troposphere from Aircraft and Satellites (ARCTAS) missions, and the 2008 Soil Moisture
Active-Passive Validation Experiment (SMAP-VEX).

Improving and evolving RTMM is a continuous process. RTMM recently integrated the
Waypoint Planning Tool, a Java-based application that enables aircraft mission scientists to
easily develop a pre-mission flight plan through an interactive point-and-click interface.
Individual flight legs are automatically calculated for altitude, latitude, longitude, flight leg
distance, cumulative distance, flight leg time, cumulative time, and satellite overpass
intersections. The resultant flight plan is then generated in KML and immediately posted to the
Google Earth-based RTMM for interested scientists to view the planned flight track and
subsequently compare it to the actual real time flight progress.

We are planning additional capabilities to RTMM including collaborations with the Jet
Propulsion Laboratory in the joint development of a Tropical Cyclone Information System which
will serve as a web portal for access to tropical cyclone data, visualizations and model output.

A description of the system architecture, components, and applications along with reviews and
animations of RTMM during the field campaigns, plus planned enhancements and future
opportunities will be presented. A poster will complement the talk.


Bi-Static GPS Result for Hurricane Seasons 2007 and 2008 with Implications to High
Altitude UAV Operations

Stephen J. Katzberg
(Stephen.J.Katzberg@nasa.gov)

NASA Langley Research Center

The GPS ocean reflection experiment (GORE) has acquired results from the 2007 and 2008
hurricane seasons as part of an ongoing effort to assess the use of bi-static techniques in
hurricane research. Previous accomplishments have included the demonstration the L-Band
surface roughness does not saturate at tropical cyclone wind speeds. A preliminary wind speed
calibration using COAMPS high resolution model surface wind fields has allowed the
comparison with other methods used by NHC and HRD. This paper will report comparisons of
the GPS surface reflection results with wind speed results from flight level, H*Wind, SFMR,
dropsondes, etc., for several hurricanes such as Fay, Gustav, Ike, and Paloma in 2008 as well as
Felix and Erin from 2007. A new area of research has been added in which the GPS bi-static
RADAR operates in radiometer mode and some results from this work will be presented.
Finally, some considerations relating to using the bi-static approach at Global Hawk UAV
altitudes will be given.


S
E
S
Session 11
S
Joint Hurricane Testbed: I
Project Updates and Plans O
for the Future
N

11

The Joint Hurricane Testbed (JHT): 2008 Update

Jiann-Gwo Jiing, Christopher Landsea, Shirley Murillo
(Jiann-Gwo.Jiing@noaa.gov, Chris.Landsea@noaa.gov Shirley.Murillo@noaa.gov)

Joint Hurricane Testbed

New analysis and forecasting tools and techniques, developed by the research community, were
tested and evaluated during 2008 at the National Hurricane Center (NHC), in real time for the
eighth consecutive hurricane season, as facilitated by the Joint Hurricane Testbed (JHT). Ten
4th round (FY07-09) projects were tested and evaluated during the 2008 hurricane season,
following any necessary technique modifications or other preparations. These projects include
upgrades to dynamical models and model components, enhancements to observed data and
assimilation techniques, track forecasting algorithms, intensity estimation and forecasting
algorithms. An announcement of Federal Funding Opportunity for new FY09 funding was
released in July. Letters of Intent followed by full Proposals were submitted by Principal
Investigators and evaluated by the JHT Steering Committee for a 5th round of JHT projects to
begin in August 2009.


Validation and Processing Tools for the Air Force Reserve Command 53rd Weather
Reconnaissance Squadron WC-130J Multi-Aircraft SFMR Systems

James Carswell1, Dan Robinson1, Paul Chang2, Eric Uhlhorn3
(carswell@remotesensingsolutions.com)
1
Remote Sensing Solutions, Barnstable, MA
2
NOAA/NESDIS/STAR, Camp Springs, MD
3
NOAA Hurricane Research Division, Miami, FL

The Stepped Frequency Microwave Radiometer (SFMR) has become an integral part of the
operational sensor suite onboard the Air Force Reserve Command 53rd Weather Reconnaissance
Squadron WC-130J and NOAA WP-3D aircraft. Our JHT project has sought to improve the
operational utilization of wind and rain products from these instruments. To this end, Remote
Sensing Solutions (RSS) and NOAA/NEDIS/STAR have developed and deployed a real-time
and post analysis data collection, processing and visualization tool. This novel application
provides hurricane specialists with the ability to visualize and interact with the SFMR and other
aircraft reconnaissance observations together in multiple formats (time series, radial, GIS, storm
relative GIS). This application is allowing hurricane specialists the capability to efficiently and
accurately estimate wind products such as the hurricane wind radii in each quadrant of a tropical
cyclone and storm intensity. In addition to this application, RSS has developed efficient
algorithms that automatically assess the validity of the SFMR retrievals. RSS has also worked to
quantify the effects of shallow water bathymetry and determine potential biases in the SFMR
observations by working with our JHT team member, NOAA Hurricane Research Division. This
presentation will report on the accomplishments in the areas discussed above, and offer an
overview of the features and capabilities of the innovative tools that have been developed during
this effort.


An Improved Wind Probability Program:
A Year 2 Joint Hurricane Testbed Project Update

Mark DeMaria1, Stan Kidder2, Buck Sampson3, John Knaff1, Chris Lauer4 and Chris Sisko4
(Mark.DeMaria@noaa.gov)
1
NOAA/NESDIS, Fort Collins, CO
2
CIRA, Colorado State University, Fort Collins, CO
3
4
NCEP Tropical Prediction Center, Miami, FL

Under Previous JHT support a new program for estimating the probability of occurrence of 34,
50 and 64 kt winds was developed by NESDIS and CIRA. A Monte Carlo (MC) method was
utilized to combine the uncertainty in the track, intensity and wind structure forecasts. The MC
probability program was transitioned to NHC operations for the 2006 hurricane season. A
verification program was also developed and provided to NHC. Verification results showed that
the probabilities were skillful using standard metrics for probabilistic forecasts, and were
relatively unbiased.

In the current project, three improvements to the MC program were proposed. First, the
timeliness is being improved through code optimization. Second, the MC code is being modified
to produce the NHC “wind speed probability table” product. Third, the probability estimates are
being refined by making the underlying track error distributions a function of the forecast
uncertainty. The current MC model uses basin-wide error statistics but recent research has shown
that the spread of track forecasts from various models provides information about the expected
track error. J. Goerss from NRL developed a real-time tool to quantitatively estimate the track
forecast uncertainty (the Goerss Predicted Consensus Error, GPCE).GPCE input is being used by
the MC model to provide additional information about the track error distributions.

The first two tasks were completed in Year 1 or early in Year 2 of the project. The optimized
code resulted in a factor of six increase in speed, and the MC model-based wind speed
probability table was implemented for the 2008 hurricane season. The primary emphasis of the
presentation will be on the third task to make the track error distributions a function of the GPCE
input. A parallel version of the code was developed utilizing the GPCE input, and was tested on
169 forecast cases from the Atlantic 2008 season where the initial storm position was within
1000 km of the U.S. coastline. Examples of the impact of the GPCE input on the MC model will
be presented, along with a quantitative comparison with the operational version using the Brier
Score. Results show that the GPCE input modifies the probabilities by as much as 15% and the
resulting fields are qualitatively reasonable. The quantitative evaluation shows that the Brier
Score is improved with the GPCE input relative to the operational version of the MC model.

decision.


Web-ATCF, User Requirements, and Intensity Consensus

Charles R. Sampson and Ann Schrader
(Buck.Sampson@nrlmry.navy.mil)


The Automated Tropical Cyclone Forecasting System (ATCF) has been in use at the NHC since
1990. Since this time, a concerted effort has been made to develop common data formats and
functionality so that tropical cyclone information can be easily shared among the operational
forecast centers in real-time. The ATCF has become a target platform for many NHC and JHT
products. Development is coordinated between the tropical cyclone forecast agencies and
developers at NRL and NHC. This particular JHT project had three main thrusts:

1) Web-ATCF: The web-ATCF adds powerful capability as a thin client software
bundle that allows a variety of web proxy, web server and ATCF server
configurations for use in demanding, secure IT environments. A prototype will be
deployed at NHC for this season. The intended use is as a display tool for coastal
forecast offices, NASA, and DoD customers.
2) User Requirements: Approximately 50 NHC requirements were addressed in the two
year project. Some of the more interesting updates will be reviewed.
3) Intensity Consensus: Intensity consensus aids developed in 2007 were upgraded for
the 2008 season. ICON is a four-model average (SHIPS, GFDL, LGEM and HWRF)
and IVCN is a five-model average that can be computed if two or more members
exist. Results from the 2008 season indicate that these intensity consensus aids
performed well. Suggestions for 2009 will be discussed.


S
E
S
Session 12
Products, Services, and
S
Lessons Learned during I
the 2008 Tropical Cyclone O
Season,
Part 1
N

12

SAB 2008 Year in Review

Michael Turk
(Michael.Turk@noaa.gov)

NOAA/NESDIS/Satellite Analysis Branch (SAB)

The Satellite Analysis Branch (SAB) derives position and intensity estimates for tropical
disturbances and cyclones around the world using the internationally recognized Dvorak
technique and cutting edge technologies provided by various research groups. A review of the
2008 hurricane/typhoon season from the operational perspective will be provided as well as an
update of NESDIS-funded tropical cyclone initiatives through the Product System Development
and Implementation (PSDI) program.


The Determination of Optimal Thresholds of Tropical Cyclone Incremental Wind Speed
Probabilities to Support Expressions of Uncertainty in Text Forecasts

Pablo Santos, Mark DeMaria, Dave Sharp

As tropical cyclone events unfold, decision-makers require a meteorologist’s most likely wind
speed forecast, along with an accompanying expression of uncertainty. Both are necessary to
effectively manage preparations for life-threatening weather events. The inherent uncertainty in
tropical cyclone forecasts results in the shortcoming of deterministic-only wind speed forecasts
such as those found within the current Zone Forecast Product and Coastal Waters Forecast
issued by National Weather Service (NWS) Weather Forecast Offices. To address this
shortcoming, the NWS offices in Miami and Melbourne have developed an experimental means
to consistently and coherently incorporate uncertainty information in these products through the
creative use of the National Hurricane Center’s incremental wind speed probabilities. These
probabilities are created upon the issuance of each official advisory for 34-, 50-, and 64-knot
winds through 120 hours. The probabilities are derived using uncertainties in the official track
and intensity forecast, and also from a climatology and persistence wind radii forecast model
(with error statistics determined from the previous five years of forecasts). Then, the
probabilities are configured locally, in gridded form, to match the traditional time increments of
the textual Zone Forecast Product and Coastal Waters Forecast. Together with gridded hazard
information (e.g., tropical storm/hurricane watches/warnings) and gridded deterministic wind
speed information, the tropical cyclone incremental wind speed probabilities trigger enhanced
wording which conveys the situational uncertainty for successive forecast periods. The logic has
been encoded within tropical cyclone versions of the respective text formatters which invoke the
prescribed expressions. This includes the zone-based versions which are used to generate the
legacy products, and the dynamic point-n-click (point-based) versions found on Weather
Forecast Office web sites. The logic depends, in part, on the exceedance of preliminary
incremental wind speed probability thresholds as a function of time (e.g., forecast period) for
uncertainty involving tropical storm force winds and hurricane force winds. Experimental and
operational testing during the 2006 to 2008 seasons has yielded encouraging results. Yet, since
the incremental wind speed probability thresholds are a critical component to the formatter logic,
it is prudent to formally identify those values which tend to maximize the responsible detection
of a potential event while minimizing false alarms which can lead to the overuse of enhanced
expressions. Optimal thresholds were determined using Relative Operating Characteristics
Diagrams and Threat Scores. Initial results were recently presented at the 2009 Annual Meeting
of the American Meteorological Society for the 34 and 64 knot incremental wind speed
probability thresholds for coastal points extending across the U.S. Gulf of Mexico and Atlantic
coasts. This paper will recount the original validation but also present a cross validation using
Peirce Scores. More so, validation results for the 50 knot thresholds will be introduced, along
with the inclusion of additional validation points located away from the coast at inland locations.


Envirocast® Vision™ TouchTable:
Linking Federal Partners to Collaborate, Communicate and Visualize the
Hurricane Threat to Decision Makers, Emergency Managers and the Public

Dave Jones
(dave@stormcenter.com)

StormCenter Communications, Inc.

StormCenter Communications, inc. is leading an effort to connect NWS and FEMA through an
advanced technology designed at the Northrop Grumman Futures Lab that promotes real-time
collaboration and visualization of NOAA and NASA data for enhanced decision making.
Currently StormCenter is working closely with NWS SR HQ, FEMA Region VI, NHC, CPHC
and HRD to develop enhanced briefing capability for a more coordinated approach to Federal
preparedness, response and communications. Working closely with NOAA Research, NOAA
Operations, FEMA and NASA, the Envirocast® Vision™ TouchTable is providing an
opportunity to address the visualization and collaboration requirements of AWIPSII and serve as
a model for transition from research to operations for many applications.


S
E
S
Session 13
Products, Services, and
S
Lessons Learned during I
the 2008 Tropical Cyclone O
Season,
Part 2
N

13

The Digital Hurricane Consortium

John Schroeder1, Mike Biggerstaff2, Daniel Cecil3, Kurt Gurley4,
Andrew Kennedy5, Marc Levitan6, Forrest Masters4, Mark Powell7, Josh Wurman8
(john.schroeder@ttu.edu, mikeb@rossby.metr.ou.edu, cecild@uah.edu, kgurl@ce.ufl.edu,
andrew.kennedy@nd.edu, levitan@hurricane.lsu.edu, masters@ce.ufl.edu,
mark.powell@noaa.gov, jwurman@cswr.org)
1
Texas Tech University
2
Oklahoma University
3
University of Alabama Huntsville
4
University of Florida
5
University of Notre Dame
6
Louisiana State University
7
NOAA/AOML Hurricane Research Division
8
Center for Severe Weather Research

Our nation’s ability to determine the severity of and the appropriate response to a hurricane
impact is based on the accuracy and sufficiency of the information recorded and how quickly
actionable information can be collected, interpreted, analyzed, and relayed from the affected
areas to public and private decision makers at the local, state and national levels. Ideally, a dense
network of robust direct and remote observing platforms would measure the physical quantities
needed to define the event. Standardization and synthesis of real-time data would produce a short
term view of the event based on the latest available information. After the event, prompt and
detailed post-storm evaluation would allow a final analysis of record which can be used to define
the storm impact on coastal communities and ecosystems.

Since the late 1990s, university field research programs have deployed monitoring equipment to
characterize and document hurricanes at landfall. With few exceptions, these programs have
operated independently or at times with ad hoc collaborations. Recently the university research
programs responsible for collecting wind, storm surge, wave, precipitation and damage data
convened to discuss in‐field coordination and the production of real‐time data streams which
would serve a variety of stakeholders. The major outcome of this gathering was the formation of
a new umbrella organization called the Digital Hurricane Consortium (DHC) that will provide
“global” coordination at landfall while allowing individual programs to meet their research
deliverables. The formation of the DHC is an important step toward producing integrated, multi-
platform data sets that are standardized and quality controlled.

This presentation will provide background information on the suite of observational assets that
have been deployed into hurricanes (using Hurricane Ike as an example) and define the concept
of developing an adaptive, event-driven observing system which can be applied to any tropical
cyclone. This adaptable network of portable weather stations, mobile Doppler radars, and wave /
surge sensors would be jointly deployed to the impacted areas to bridge the information void
near the coast. Dissemination of the real-time information would result in improved situational
awareness and more efficient response, while detailed post-storm analysis would benefit the
scientific and engineering communities by providing a more detailed evaluation of storm
structure at landfall.


Application of the National Hurricane Center Tropical Cyclone Wind Speed Probability
Product to Quantifying Potential Impacts of Hurricane Forecast Improvements
Andrea Schumacher1, Mark DeMaria2, Dan Brown3, and Ed Rappaport3
(Schumacher@cira.colostate.edu)
1
CIRA, Colorado State University, Fort Collins, CO
2
NOAA/NESDIS, Fort Collins, CO
3
National Hurricane Center, Miami, FL

The Tropical Cyclone Wind Speed Probability product became operational at the National
Hurricane Center (NHC) in 2006, replacing its Strike Probability Program. Since then, several
new applications of this probabilistic product have been explored. One such application involves
the possible use of the product-generated probabilities to provide “first guess” objective guidance
for hurricane watch and warning issuance. A preliminary scheme has been developed that uses a
set of hurricane warning issuance rules based on product probability values.

The current study sought to refine the preliminary scheme by using a larger sample set, which
includes 19 hurricanes making landfall along the U.S. coast from 2004-2008. The original
scheme was run over this larger dataset and the probability threshold values were varied until the
scheme-derived hurricane warnings correlated well with both the storm total coastal lengths (R2
= 0.85) and storm total average durations (R2 = 0.76) of the official hurricane warnings issued
for these 19 hurricanes. The wind speed probability thresholds chosen for this new scheme are
pup ≥ 8.0%, which prompts the issuance of a hurricane warning, and pdown = 0.0%, which
prompts the removal of an existing hurricane warning.

Using this new scheme, a series of tests were conducted to investigate the potential impacts of
forecasted improvements on hurricane warning length and duration. This was accomplished by
artificially reducing the underlying error distributions in the wind speed probability model and
then calculating the reduction in warning size and duration. Both track and intensity error
reductions were considered. Preliminary results suggest that the ratio of reductions in hurricane
warning duration and coastal length and prescribed improvements in track and intensity forecasts
is less than one, and that the impacts of improvements in track and intensity forecasts on
warnings are not additive. The relationships derived from this study, combined with
socioeconomic analyses of the costs and benefits of hurricane warnings, provide a simple
framework for quantifying the value that future improvements to hurricane track and intensity
forecast may bring to society.

decision.


A Proposed New Storm Surge Scale

Pat Fitzpatrick, Nam Tran, Yee Lau, Yongzuo Li, and Chris Hill
(fitz@gri.msstate.edu)

Mississippi State University
Geosystems Research Institute
Stennis Space Center, MS

It is now well-recognized that the original Saffir-Simpson (SS) scale’s relationship of storm
surge based solely with intensity is not valid. The SS scale’s intention was to give an estimate of
the potential flooding and damage to property based on a hurricane's sustained wind speed
(Simpson 1974). However, the surge depends on intensity, storm size, bathymetry, storm speed,
barometric pressure, and local topographical features. A combination of these factors results in
large surge differences between hurricanes of comparable intensity, such as Category 4
Hurricane Charley’s relatively minor surge inundation versus Category 3 Katrina’s record surge.
It has also been proposed by Powell and Reinhold (2007) that Integrated Kinetic Energy (IKE)
may provide a better correlation to storm surge than intensity.

A new scale will be presented which accounts for all surge factors, as well as IKE. These results
are based on many hypothetical surge simulations using the Advanced CIRCulation (ADCIRC)
model. The variations include 5 different bathymetries for continental shelf slope and distance;
small-, average-, and large-sized hurricanes defined by radius of tropical storm-force winds;
slow-, average-, and fast-moving hurricanes; and different intensities. It is found that, if one
partitions by bathymetric zones, 3 tables result for storm size categories. Storm speed turns out to
be a minor component that’s only relevant for intense hurricanes in shallow bathymetry, and can
be used as a correction factor in these tables.

However, the tables can be further simplified if IKE is included. IKE cannot be used as a
replacement for intensity because a large Category 3 hurricane and small Category 5 hurricane
have the same IKE magnitude, but produce different surges since wind stress plays an important
role. However, a nonlinear combination of IKE0.5 and intensity produces a linearly increasing
trend for all bathymetries. This scale varies from 0 to 7X106 kg0.5 J. It is proposed that a 0-7
metric be used to represent storm surge potential based on bathymetric zone. The general public
need not know how the metric is computed, and there is precedence for using metrics in other
scales. For example, users of the Richter scale do not know or care that it is a base-10
logarithmic scale obtained by calculating the logarithm of the combined horizontal amplitude of
the largest displacement on a Wood–Anderson torsion seismometer. Another advantage of the
proposed scale is that it is not quantized like the SS scale, where a 5-knot intensity difference can
result in a unit Category change.

We will also present tables for computing the eastward displacement of 5-ft and 10-ft surge
inundation. This scale could be useful for anticipating the extent of surge inundation east of a
hurricane landfall. These proposed scales are a natural evolution of the SS scale, and intended to
educate the general public on their regional storm surge vulnerability, not to replace storm surge
model forecasts. This work was funded by the Northern Gulf Institute and by NOAA.


63rd%20IHC%20booklet%20final

More Related Content

Viewers also liked (6)

Similar to 63rd%20IHC%20booklet%20final (20)

More from tutorialsruby (20)

Recently uploaded (20)

63rd%20IHC%20booklet%20final