[House Hearing, 118 Congress]
[From the U.S. Government Publishing Office]
WINNING IN WEATHER:
U.S. COMPETITIVENESS
IN FORECASTING AND MODELING
=======================================================================
HEARING
BEFORE THE
SUBCOMMITTEE ON ENVIRONMENT
OF THE
COMMITTEE ON SCIENCE, SPACE,
AND TECHNOLOGY
OF THE
HOUSE OF REPRESENTATIVES
ONE HUNDRED EIGHTEENTH CONGRESS
SECOND SESSION
__________
MARCH 6, 2024
__________
Serial No. 118-36
__________
Printed for the use of the Committee on Science, Space, and Technology
[GRAPHIC NOT AVAILABLE IN TIFF FORMAT]
Available via the World Wide Web: http://science.house.gov
__________
U.S. GOVERNMENT PUBLISHING OFFICE
55-002PDF WASHINGTON : 2026
-----------------------------------------------------------------------------------
COMMITTEE ON SCIENCE, SPACE, AND TECHNOLOGY
HON. FRANK LUCAS, Oklahoma, Chairman
BILL POSEY, Florida ZOE LOFGREN, California, Ranking
RANDY WEBER, Texas Member
BRIAN BABIN, Texas SUZANNE BONAMICI, Oregon
JIM BAIRD, Indiana HALEY STEVENS, Michigan
DANIEL WEBSTER, Florida JAMAAL BOWMAN, New York
MIKE GARCIA, California DEBORAH ROSS, North Carolina
STEPHANIE BICE, Oklahoma ERIC SORENSEN, Illinois
JAY OBERNOLTE, California ANDREA SALINAS, Oregon
CHUCK FLEISCHMANN, Tennessee VALERIE FOUSHEE, North Carolina
DARRELL ISSA, California KEVIN MULLIN, California
RICK CRAWFORD, Arkansas JEFF JACKSON, North Carolina
CLAUDIA TENNEY, New York EMILIA SYKES, Ohio
RYAN ZINKE, Montana MAXWELL FROST, Florida
SCOTT FRANKLIN, Florida YADIRA CARAVEO, Colorado
DALE STRONG, Alabama SUMMER LEE, Pennsylvania
MAX MILLER, Ohio JENNIFER McCLELLAN, Virginia
RICH McCORMICK, Georgia GABE AMO, Rhode Island
MIKE COLLINS, Georgia SEAN CASTEN, Illinois,
BRANDON WILLIAMS, New York Vice Ranking Member
TOM KEAN, New Jersey PAUL TONKO, New York
VACANCY
------
Subcommittee on Environment
HON. MAX MILLER, Ohio, Chairman
BILL POSEY, Florida DEBORAH ROSS, North Carolina,
RICK CRAWFORD, Arkansas Ranking Member
RYAN ZINKE, Montana SUZANNE BONAMICI, Oregon
MIKE COLLINS, Georgia MAXWELL FROST, Florida
C O N T E N T S
March 6, 2024
Page
Hearing Charter.................................................. 2
Opening Statements
Statement by Representative Max Miller, Chairman, Subcommittee on
Environment, Committee on Science, Space, and Technology, U.S.
House of Representatives....................................... 6
Written Statement............................................ 7
Statement by Representative Deborah Ross, Ranking Member,
Subcommittee on Environment, Committee on Science, Space, and
Technology, U.S. House of Representatives...................... 8
Written Statement............................................ 9
Statement by Representative Zoe Lofgren, Ranking Member,
Committee on Science, Space, and Technology, U.S. House of
Representatives................................................ 10
Written Statement............................................ 11
Written statement by Representative Frank Lucas, Chairman,
Committee on Science, Space, and Technology, U.S. House of
Representatives................................................ 11
Witnesses:
The Honorable Dr. Neil Jacobs, Acting NOAA Administrator, 2019-
2021
Oral Statement............................................... 13
Written Statement............................................ 15
Dr. Kevin Petty, Chief Executive Officer, Aeris LLC
Oral Statement............................................... 18
Written Statement............................................ 20
Dr. Scott Weaver, Chief Executive Officer, CLIMET Consulting
Oral Statement............................................... 26
Written Statement............................................ 28
Discussion....................................................... 37
Appendix: Answers to Post-Hearing Questions
The Honorable Dr. Neil Jacobs, Acting NOAA Administrator, 2019-
2021........................................................... 48
Dr. Scott Weaver, Chief Executive Officer, CLIMET Consulting..... 50
WINNING IN WEATHER:
U.S. COMPETITIVENESS
IN FORECASTING AND MODELING
----------
WEDNESDAY, MARCH 6, 2024
House of Representatives,
Subcommittee on Environment,
Committee on Science, Space, and Technology,
Washington, D.C.
The Subcommittee met, pursuant to notice, at 10:01 a.m., in
room 2318 of the Rayburn House Office Building, Hon. Max Miller
[Chairman of the Subcommittee on Environment] presiding.
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
Chairman Miller. The Committee will come to order. Without
objection, the Chair is authorized to declare recesses of the
Committee at any time. Welcome to today's hearing entitled
``Winning in Weather: United States Competitiveness in
Forecasting and Modeling.'' I recognize myself for 5 minutes
for an opening statement.
I'd like to welcome everyone to this morning's hearing,
``Winning in Weather: United States Competitiveness in
Forecasting and Modeling.'' This is the Environment
Subcommittee's third hearing this Congress related to
reauthorizing the Weather Research and Forecasting Innovation
Act of 2017, an important effort for the Committee at large.
Today's hearing continues our examination of United States'
weather policy and how best to update the crucial work started
by the Weather Act 7 years ago.
In November, this Committee unanimously passed the Weather
Act Reauthorization, and now we look toward the future of the
United States' leadership in weather forecasting. Today, we
will hear from witnesses who all have extensive experience in
the United States weather enterprise. Their testimonies will
help us assess the United States' leadership in weather
forecasting and in modeling.
It's easy to see that weather data doesn't just help
determine the day's clothing. It enhances our national economy
and security by assisting long-term decision making. There are
serious economic and humanitarian implications to not being
able to predict weather correctly or precisely, and we cannot
afford to let the United States fall behind on the world stage.
From my conversations with farmers back home in Ohio, I
know that seasonal weather predictions are vital to American
agriculture. Without accurate predictions in this sector,
planting and harvesting is put at major risk. Weather is also
critical to the insurance industry and how to best ensure homes
from Cleveland to Worcester. And last but certainly not least,
modeling and forecasting is crucial to all aspects of
manufacturing, from planning around supply chain constraints
for peak production to determining the safest transportation
routes for the final product.
The purpose of today's hearing is to discuss the impacts of
the Weather Act of 2017 and the Weather Act Reauthorization of
2023 in bolstering U.S. competitiveness in the global weather
community. We'll also discuss the coordination and
collaboration between government agencies, the private sector,
and academia to implement programs that advance new
technologies related to data assimilation and modeling. Whether
the data comes from NOAA (National Oceanic and Atmospheric
Administration), State services, or commercial providers, we
must ensure that all tools at our disposal are used to make the
public aware of extreme weather conditions.
The United States faces the unique challenge of having some
of the widest variety of weather events in the world. Just last
week, the Smokehouse Creek fire in the Texas panhandle and
southwestern Oklahoma grew to over a million acres in a matter
of just days. At the same time, northern California was hit by
a massive blizzard which dumped feet of snow on some areas and
blew gusts of wind over 100 miles per hour. In my home State of
Ohio, there were nine confirmed tornado touchdowns on February
28 just alone.
Through innovation to improve the accuracy and timeliness
of weather models, as well as public awareness, we can all save
lives and property at risk from these events. Today, we will
identify actionable items NOAA can pursue to build trust and
education in weather forecasting products. When an EF-4 tornado
or a wildfire is bearing down on U.S. citizens, there should be
no doubt on the accuracy of the models and what the best course
of action is to be. Additionally, by working together and
increasing partnerships between NOAA and the commercial sector,
the users of weather data will be better equipped to strengthen
both short and long-term weather predictions, benefiting
Americans across all sectors.
I want to once again thank all of our witnesses for being
here, and I look forward to each of your testimonies as we are
here this morning.
[The prepared statement of Chairman Miller follows:]
I want to welcome everybody to this morning's hearing,
Winning in Weather: U.S. Competitiveness in Forecasting and
Modeling. This is the Environment Subcommittee's third hearing
this Congress related to reauthorizing the Weather Research and
Forecasting Innovation Act of 2017, an important effort for the
Committee at large.
Today's hearing continues our examination of U.S. weather
policy and how best to update the crucial work started by the
Weather Act seven years ago. In November, this committee
unanimously passed the Weather Act Reauthorization and now we
look towards the future of the United States' leadership in
weather forecasting.
Today, we will hear from witnesses who all have extensive
experience in the U.S. weather enterprise. Their testimonies
will help us assess the United States' leadership in weather
forecasting and modeling.
It's easy to see that weather data doesn't just help
determine the day's clothing. It enhances our national economy
and security by assisting long term decision making. There are
serious economic and humanitarian implications to not being
able to predict weather correctly or precisely, and we cannot
afford to let the United States fall behind on the world stage.
From my conversations with farmers back home in Ohio, I
know that seasonal weather predictions are vital to American
agriculture. Without accurate predictions in this sector,
planting and harvesting is put at major risk.
Weather is also critical to the insurance industry and how
to best insure homes from Cleveland to Wooster. And last but
certainly not least, modeling and forecasting is crucial to all
aspects of manufacturing--from planning around supply chain
constraints for peak production to determining the safest
transportation routes for the final product.
The purpose of today's hearing is to discuss the impacts of
the Weather Act of 2017 and the Weather Act Reauthorization of
2023 in bolstering U.S. competitiveness in the global weather
community. We'll also discuss the coordination and
collaboration between government agencies, the private sector,
and academia to implement programs that advance new
technologies related to data assimilation and modeling.
Whether the data comes from NOAA, state services, or
commercial providers, we must ensure that all tools at our
disposal are used to make the public aware of extreme weather
conditions.
The United States faces the unique challenge of having some
of the widest variety of weather events in the world.
Just last week, the Smokehouse Creek fire in the Texas
panhandle and southwestern Oklahoma grew to over a million
acres in a matter of days. At the same time, Northern
California was hit by a massive blizzard, which dumped feet of
snow on some areas and blew gusts of wind over 100 miles per
hour. In my home state of Ohio, there were 9 confirmed tornado
touchdowns on February 28 alone.
Through innovation to improve the accuracy and timeliness
of weather models, as well as public awareness, we can save
lives and property at risk from these events.
Today we will identify actionable items NOAA can pursue to
build trust and education in weather forecasting products. When
an EF-4 tornado or a wildfire is bearing down on U.S. citizens,
there should be no doubt on the accuracy of the models and what
the best course of action is.
Additionally, by working together and increasing
partnerships between NOAA and the commercial sector, the
``users'' of weather data will be better equipped to strengthen
both short- and long-term weather predictions, benefitting
Americans across all sectors.
I want to thank all our witnesses for being here. I look
forward to each of your testimonies.
Chairman Miller. I now recognize the Ranking Member, the
gentlewoman from North Carolina, for an opening statement.
Ms. Ross. Thank you, Chairman Miller, for holding this
important hearing.
As extreme weather increasingly impacts the lives of the
American people--and also, I want to thank all of our witnesses
for sharing your time and expertise with us today. It's much
appreciated.
Accurate weather forecasts and warnings are essential to
the protection of American lives, property, and commerce.
Americans rely on dependable and accessible weather information
to make important decisions every single day. Timely and
accurate weather prediction is especially important in the face
of extreme weather events. According to NOAA, last year, the
United States experienced a record number of individual weather
and climate disasters that each caused over $1 billion dollars
in damage.
Climate change is causing an increase in the occurrence and
severity of extreme weather events in the United States. That's
a fact. Forty years ago, these billion-dollar disasters
occurred once every 4 months. Today, they occur about once
every 3 weeks. My home State of North Carolina has experienced
many of these disasters. The severity and frequency of extreme
precipitation events has been increasing over the past few
decades and is only expected to worsen.
The dangers and damage associated with severe storms cannot
be stopped entirely, but they can be mitigated, including by
providing the impacted communities with appropriate warnings.
To that end, it is critical that we ensure our weather
forecasting and communication abilities are of the highest
possible quality to protect American lives and property.
The importance of partnerships and collaboration among our
government, academic, and industry partners cannot be
understated. The integration of high-quality data collection
with cutting-edge weather prediction technologies and
fundamental science forms the backbone of reliable weather
forecasting in this country. And this is only achieved through
the collaboration of the U.S. weather enterprise.
Collaboration within government is also important. There is
expertise and weather-relevant research being carried out
across our government. The Interagency Council on Advancing
Meteorological Services, or ICAMS, that was born out of this
Committee's work in the Weather Act of 2017 seeks to advance
U.S. weather forecasting capabilities by synthesizing the
efforts of multiple government agencies, along with private-
public partnerships.
This Committee has heard from several experts, including
our North Carolina State Climatologist, that these types of
collaborative efforts are essential to improving weather
forecasting and warnings in the United States. One major hurdle
to improving our Nation's forecasting system is a lack of
computing resources available for NOAA to support operational
forecasts and data dissemination.
Last year, I co-sponsored the Advanced Weather Model
Computing Development Act, a bipartisan bill introduced by
Chairman Miller that passed the House. This important
legislation enables NOAA to leverage the Department of Energy's
high performance computational resources and collaborate on
resource-research to improve weather and climate modeling.
While the refinement of accurate weather forecasts is
highly important, we must also consider the unique needs of
various end users across the United States. This is as much a
social science challenge as a data or modeling challenge. The
integration of timely, accurate weather information with
effective and accessible communication is essential to our goal
of protecting American lives and property.
I look forward to hearing from our witnesses about how
Congress can continue to support NOAA, other government
agencies, academia, and industry in improving the Nation's
weather forecasting and warnings capabilities.
Thank you, Mr. Chairman, and I yield back.
[The prepared statement of Ms. Ross follows:]
Thank you, Chairman Miller, for holding this important
hearing, as extreme weather increasingly impacts the lives of
the American people; and thank you to our witnesses for sharing
your time and expertise with us today.
Accurate weather forecasts and warnings are essential to
the protection of American lives, property, and commerce.
Americans rely on dependable and accessible weather information
to make important decisions every single day. Timely and
accurate weather prediction is especially important in the face
of extreme weather events. According to NOAA, last year, the
U.S. experienced a record number of individual weather and
climate disasters that each caused over one billion dollars in
damage.
Climate change is causing an increase in the occurrence and
severity of extreme weather events in the United States. Forty
years ago, these billion-dollar disasters occurred once every
four months. Today, they occur about once every three weeks. In
my home state of North Carolina, the severity and frequency of
extreme precipitation events has been increasing over the past
few decades and is only expected to worsen.
The dangers and damage associated with severe storms cannot
be stopped entirely, but they can be mitigated, including by
providing impacted communities with appropriate warnings. To
that end, it is of critical importance to ensure that our
weather forecasting and communication abilities are of the
highest possible quality to protect American lives.
The importance of partnerships and collaboration between
our government, academic, and industry partners cannot be
understated. The integration of high-quality data collection
with cutting-edge weather prediction technologies and
fundamental science forms the backbone of reliable weather
forecasting in this country, and this is only achieved through
the collaboration of the U.S. Weather Enterprise.
Collaboration within the government is also important.
There is expertise and weather-relevant research being carried
out across our government. The Interagency Council on Advancing
Meteorological Services, or ICAMS, that was born out of
this committee's work in the Weather Act of 2017, seeks to
advance U.S. weather forecasting capabilities by synthesizing
the efforts of multiple government agencies along with private-
public partnerships. This Committee has heard from several
experts, including our North Carolina State Climatologist, that
these types of collaborative efforts are essential to improving
weather forecasting and warnings in the United States.
One major hurdle to improving our nation's forecasting
system is a lack of computing resources available for NOAA to
support operational forecasts and data dissemination. Last
year, I co- sponsored the Advanced Weather Model Computing
Development Act, a bipartisan bill introduced by Chairman
Miller that passed the House. This important legislation
enables NOAA to leverage the Department of Energy's high
performance computational resources and collaborate on research
to improve weather and climate modeling.
While the refinement of accurate weather forecasts is
highly important, we must also consider the unique needs of
various end-users across the United States. This is as much a
social science challenge as a data or modeling challenge. The
integration of timely, accurate weather information with
effective and accessible communication is essential to our goal
of protecting American lives and property.
I look forward to hearing from our witnesses about how
Congress can continue to support NOAA, other government
agencies, academia, and industry in improving the nation's
weather forecasting and warnings capabilities.
Thank you, Mr. Chairman, and I yield back.
Chairman Miller. Thank you, Ms. Ross.
I now would like to recognize the Ranking Member of the
Full Committee, the gentlewoman from California, for an opening
statement.
Ms. Lofgren. Thank you, Mr. Chairman.
This hearing features experts with relevant expertise in a
whole range of issues that we'll explore today surrounding the
U.S. weather enterprise's efforts to improve the Nation's
weather and climate forecasting, as well as modeling
capabilities. A topic that I've been focused on in these
weather hearings is the need for improvement in the subseasonal
to seasonal or S2S--there's got to be an acronym for
everything--forecasts and products, the S2S predictions are
important to community and to industry, including and maybe
even especially agriculture so that adequate decisionmaking and
planning can be made relative to water management,
infrastructure development, electricity market demands, as well
as flood protection. Advancements are going to require
additional research and development in quite a number of areas,
and I look forward to hearing from these experts about that.
While improving modeling is important, as the Ranking
Member just mentioned, there needs to be also a focus on
improving the end products and the delivery of services. Dr.
Petty, in his testimony, mentioned that when weighing the
success in forecasting, there needs to be other meaningful
metrics that are more aligned to NOAA's mission to protect
lives and property. Who is the audience? What information will
the end users find most useful? How should that information be
delivered in the most effective way?
In reaching this goal of improving meteorological services,
there need to be effective collaboration between Federal
agencies that have relevant expertise and among the broader
U.S. weather enterprise. When different agencies compete for
limited funding, they become even more protective of what they
have, and that sometimes constrains collaboration. At a time of
constrained resources, what more could Congress do to encourage
interagency collaboration? And how can the private sector and
academia engage in these efforts and support advancement of
meteorological sciences? These questions need to be answered to
ensure that we're leveraging the Nation's full expertise to
further the development of weather and climate research,
forecasting, and communication.
On the topic of funding today, later today, we'll be voting
on the Fiscal Year 2024 appropriation agreement that includes
funding for NOAA. I'm glad that, as compared to previous
budgets that would have cut NOAA by 22 percent, the budget now
would keep NOAA at level funding. And obviously, level funding
in a time of inflation is not an increase, but it is certainly
an improvement over the 22 percent proposed cut.
I also want to mention the Weather Act Reauthorization bill
that Chairman Lucas and I advanced to improve NOAA's
forecasting and modeling abilities by expanding existing and
creating new programs. I hope this bill makes it to the
President's desk this year. However, I worry that stagnant
funding provided to NOAA would undoubtedly slow advancement,
even with the good provisions of this bill.
I want to thank the Chairman, the Ranking Member, and each
of the expert witnesses who are here today, and I yield back,
Mr. Chairman.
[The prepared statement of Ms. Lofgren follows:]
Good morning. Thank you to Chairman Miller and Ranking
Member Ross for holding today's hearing on weather and climate
modeling and forecasting. Today's hearing features experts with
relevant expertise in a range of issues that we will explore
today surrounding the U.S. Weather Enterprise's efforts to
improve the nation's weather and climate forecasting and
modeling capabilities.
A topic that I continue to be focused on in these weather
hearings is the need for improvement in the subseasonal to
seasonal, or S2S, forecasts and products. S2S predictions are
critical to communities and to industry, including agriculture,
for informing adequate decision making and planning related to
water management, infrastructure development, and electricity
market demands. Advancements in S2S prediction will require
further research and development in a number of areas,
including in coupled numerical weather-climate systems. I look
forward to hearing the expert panel's recommendations for
improving subseasonal to seasonal modeling.
While improving modeling is important, there needs to be
just as much focus on improving the end products and delivery
of services. As Dr. Petty mentions in his testimony, when
weighing our success in forecasting, there need to be other
meaningful metrics that are more aligned to NOAA's mission to
protect lives and property. It is essential to understand who
the audience is, what information the end user finds most
useful, and how that information should be delivered in the
most effective way.
In reaching this goal of improving meteorological services,
there needs to be effective collaboration between federal
agencies that have relevant expertise, and among the broader
U.S. Weather Enterprise. When different agencies are competing
for limited funding, they become even more protective of what
they have. At a time of constrained resources, what more can
Congress do to encourage inter-agency collaboration? How can
the private sector and academia engage in these efforts and
support advancement of meteorological services? These questions
need to be answered to ensure that we are leveraging the
nation's full expertise to further the development of weather
and climate research, forecasting, and communication.
On the topic of funding, later today, we will be voting on
the fiscal year 2024 appropriation agreement that includes
funding for NOAA. I am pleased to see the reversal of the
previous stance by the majority to cut NOAA's budget by as much
as 22%. The budget now would keep NOAA's spending essentially
flat. While the reversal is welcome, I remain concerned about
how NOAA will be able to achieve its goal to improve models and
forecasts rapidly.
The Weather Act Reauthorization bill that I cosponsored
with Chairman Lucas aims to improve NOAA's forecasting and
modeling abilities by expanding existing and creating new
programs. While I do hope this bill makes it to the President's
desk this year, the stagnant funding level provided to NOAA
will undoubtedly slow advancement.
Thank you to the expert panel for being here this morning.
I look forward to your testimony and the discussion. I yield
back.
Chairman Miller. Thank you, Ms. Lofgren.
I ask unanimous consent to enter into the record the
opening statement of Chairman Lucas. Without objection, so
ordered.
[The prepared statement of Chairman Lucas follows:]
I want to thank the gentleman from Ohio, Chairman Miller,
for holding today's hearing.
This hearing couldn't come at a more crucial time. Last
week, the Smokehouse Creek fire burned over 1 million acres in
Western Oklahoma and the Texas Panhandle. This has earned it
the unpleasant distinction of being the second largest wildfire
in U.S. history.
While evaluation is still on going, early reports indicate
that about 500 structures were destroyed, thousands of cattle
were killed, and tragically one life was lost. Last week, I
took some time to drive through the western most parts of my
district to see the damage, and I can tell you that many
families will struggle to fully recover for years to come.
Where I'm from, land is your livelihood. Even if your house
is still standing, you can lose the very backbone of your
career and income if your property is damaged or thrown off
from the delicate balance you have cultivated.
If it wasn't for some strong south to north wind at just
the right time last week, I could have been in that category. I
had to move cattle into fields that were less likely to burn if
the fire kept moving our way. It's not a fun activity, but a
necessary one to protect my family's business.
This personal experience very much relates to today's
hearing. I looked at the Oklahoma Mesonet to see the most up to
date conditions in my area. I closely monitored the National
Weather Service's fire weather outlooks for the coming days, so
I knew what to expect. And I made critical decisions based on
the weather data, models, and forecasts that were available to
me.
It's been seven years since the passage of the Weather Act
of 2017. The tools and data I used last week are more accurate
and readily available than they were before that bill. But I
believe there is more we can do for the U.S. weather enterprise
to be at the forefront of forecasting and modeling.
The Weather Act Reauthorization of 2023 and H.R. 4866, the
Fire Weather Development Act of 2023, are both bills that
passed with bipartisan support out of this Committee. Both
bills will further improve the weather data, models, and
forecasts to meet the unique needs of our country and protect
lives and property.
I thank our witnesses for sharing their expertise with us
and I look forward to a productive discussion. Thank you and I
yield back Mr. Chairman.
Chairman Miller. Let me introduce our witnesses. Our first
witness today is Dr. Neil Jacobs. He is the Chief Science
Advisor for the Community Unified Forecast System (UFS), part
of the University Corporation for Atmospheric Research's
Cooperative Programs for the Advancement of Earth System
Science and hosted at North Carolina State University. Dr.
Jacobs previously served as both Acting Under Secretary of
Commerce for Oceans and Atmosphere and Assistant Secretary of
Commerce for Environmental Observation and Prediction. Dr.
Jacobs holds bachelor's degrees in mathematics and physics from
the University of South Carolina and master's and doctoral
degrees in air and sea interaction and numerical weather
prediction from the North Carolina State University. Man, that
is a lot. I could barely get that out, Doctor.
Our second witness is Dr. Kevin Petty. He is the Chief
Executive Officer of Aeris LLC. He has also worked as a Project
Scientist and Scientific Program Manager with the National
Center for Atmospheric Research and as an Accident Investigator
and Senior Meteorologist for the National Transportation Safety
Board. Dr. Petty graduated with an M.S. and Ph.D. in
atmospheric sciences from the Ohio State University, OH--there
it is--and a B.S. in mathematics/secondary education from
Illinois College.
Our third witness is Dr. Scott Weaver. He is Founder and
Principal Scientist of CLIMET Consulting--or CLIMET Consulting
LLC and brings over 20 years of experience as a scientist and
executive, leading climate and weather scientific research and
advancing interagency science policy across the Federal
Government, most recently, serving as the National Director for
Windstorm Impact Reduction at NIST (National Institute of
Standards and Technology) and Executive Director of the
Interagency Council for Advancing Meteorological Services at
the White House Office of Science and Technology Policy. Dr.
Weaver has also held scientific research and leadership
positions in NASA, NOAA, NIST, and the private sector focused
on weather, climate variability, climate change, and disaster
impact reduction.
Welcome. And a warm welcome to all of our witnesses, and
truly, thank you for being here today. It's much appreciated.
I now recognize Dr. Neil Jacobs for 5 minutes to present
his testimony.
TESTIMONY OF THE HONORABLE DR. NEIL JACOBS,
ACTING NOAA ADMINISTRATOR, 2019-2021
Dr. Jacobs. Chairman Miller, Ranking Member Ross, and
Members of the Subcommittee, thank you for the opportunity to
testify at this hearing on a topic that is extremely important
for the future of our country's scientific leadership. Over the
last few years, there have been several major successes owed to
the Weather Research and Forecasting Innovation Act of 2017.
Considerable progress has been made toward developing a
community-friendly Unified Forecast System, which streamlines
the software architecture, while embracing an open innovation
approach that is mutually beneficial to the public, private,
and academic sectors. This is a major paradigm shift because
previous versions of the code would only work on NOAA's
internal supercomputers. Engaging the broader community will
not only result in greater access to innovative solutions, but
it'll simultaneously train the future workforce.
The Earth Prediction Innovation Center (EPIC) is a program
charged with supporting this effort. The goal of EPIC is to
empower the scientific community to develop, test, and
contribute their own innovations to the UFS through a
transparent and objective evaluation process. This
transformation is underway, but there was a lot more work that
needs to be done. The UFS, which is a community model, still
needs to finalize its governance that is by the community for
the community in which NOAA is a critical stakeholder.
Numerical weather prediction is an initial value problem.
Correctly predicting the future state of any Earth system
requires observations with greater resolution and accuracy.
Many of these new observations are being produced by private
industry. Commercial data buy programs such as GPSRO (Global
Positioning System Radio Occultation), commercial aircraft
data, and the National Mesonet Program are cost-effective ways
to greatly improve model accuracy.
While acquiring data for more advanced observing systems is
a critical step to improving forecasts, we cannot lose sight of
an equally attainable opportunity, which is extracting more
value from the data that we already collect. The process of
data assimilation such as the JEDI (Joint Effort for Data
Assimilation Integration) framework is often overlooked, yet it
is one of the most important aspects of numerical weather
prediction where significant returns on investment are easily
attainable.
Another challenge we are facing is how to manage the new
data when those file sizes are growing exponentially. As files
get larger, transfer speeds get slower. The mission of the
National Weather Service is to protect life and property where
often seconds matter. Increasing the latency of data
availability, even if it is more accurate, is
counterproductive.
Whether the satellite data forecast models or processing
and visualization tools like AWIPS (Advanced Weather
Interactive Processing System), cloud is the answer. We must
evolve beyond a process that relies on moving data. Not only
does cloud computing eliminate the latency problem, the
elasticity of on-demand commercial cloud also eliminates the
bottleneck of limited access to compute research for research
and development.
Efficiencies throughout the value chain of forecasting can
also be gained from artificial intelligence (AI). I am
encouraged by NOAA's rapid adoption of various AI and deep-
learning techniques and would like to thank the Committee for
supporting this emerging capability in the Weather Act
Reauthorization.
While these are a few technical gaps that need to be
addressed, there's broader overarching aspects that are
essential to accelerating progress. Workforce recruitment,
retention, and development is a growing challenge. There are
very few opportunities for the current workforce to upskill or
reskill to adapt to rapidly evolving technological
advancements. While software engineering is a common field at
most universities, finding engineers with a solid understanding
of atmospheric physics and dynamics is not easy.
Additionally, the programming skills needed for model
development are similar to those required in other job markets
where the starting salaries are far more competitive. There's a
growing number of Earth system science degree programs within
the U.S. universities, but many are limited in their ability to
introduce students to model code.
Versions of the UFS through support of NOAA and the
National Science Foundation are being developed for use in the
classroom as a teaching tool. Being able to run a global model
without access to large supercomputers was unimaginable just a
few years ago. However, we now have versions of the UFS that
can run on public cloud, as well as a laptop. While this is a
game-changing capability to train the future workforce, we are
now just approaching the start line for this race.
Chairman Miller, Ranking Member Ross, and Members of the
Subcommittee, thank you again for inviting me to participate
today. I would be pleased to answer any questions you may have.
[The prepared statement of Dr. Jacobs follows:]
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
Chairman Miller. Thank you very much, sir.
I now would like to recognize Dr. Kevin Petty for 5 minutes
to present the testimony.
TESTIMONY OF DR. KEVIN PETTY,
CHIEF EXECUTIVE OFFICER, AERIS LLC
Dr. Petty. Chairman Miller, Ranking Member Ross, and
Members of the Subcommittee, I want to thank you for the
opportunity to share some thoughts and perspectives related to
the weather enterprise, particularly from the standpoint of
U.S. competitiveness and weather forecasting and modeling now
and into the future. I've had a long career working in the
weather enterprise. I've been fortunate enough to work in the
public, private, and academic sectors. Most recently, I joined
the leadership team at Aeris LLC, a small, Colorado-based team
that is focused on the development and application of
environmental modeling at meter-scale resolutions.
These diverse positions and experiences have helped to
shape my perspective on the weather enterprise, particularly
what has worked well and what has not. Here in the United
States, we face distinctive weather-related challenges that are
not seen in other parts of the world. Factors such as a
changing climate, population growth, increased complexities
associated with a built environment are all shaping the need
for accurate, timely weather data and information.
The ability to achieve significant measurable improvements
in weather analysis and forecasting is rooted in four
fundamental areas: in observations, in computing, and modeling
and information delivery. Unfortunately, no single entity or
sector of the enterprise has the capacity to meet the depth and
breadth of the demands associated with each of these domains.
That said, our country has successfully developed something
that no other country has been able to replicate, and that is a
strong, robust weather enterprise where public, private,
academic, nongovernmental organizations have the ability to
support the common goal of protecting life and property and
ensuring a strong, vibrant economy. The U.S. competitiveness in
weather forecasting and modeling is highly dependent upon our
ability to effectively leverage the strengths of each segment
of the weather enterprise.
Now, it goes without saying that the U.S. weather
enterprise would not be what it is today without NOAA. NOAA is
and will remain a central component of the weather data and
information value chain. Many companies like my own depend on
NOAA to produce very high-quality data, including weather
forecasts. Nonetheless, like other segments of the enterprise,
NOAA must continue to use rapidly evolve if it's going to be a
driving force in terms of being a preeminent leader in weather
forecasting globally.
We've seen other components of the enterprise quickly
hasten the pace of innovation, and this is particularly true
from the private sector's perspective. There--the private
sector has made significant investments and advancements in the
development of new observing systems, in the use of artificial
intelligence, in pushing the boundaries of modeling from the
standpoint of spatial and temporal resolution.
The Weather Research and Forecasting Innovation Act of 2017
has helped to guide recent NOAA activities in several positive
ways, including from the standpoint of encouraging commercial
weather data buys, highlighting the importance of expanding and
upgrading computational resources, and providing concrete
direction on weather research and forecasting priorities.
However, there's more work to be done. If we're going to be the
leader in weather forecasting and modeling, it is imperative
that NOAA have the continued support and direction it needs.
With these elements in place, NOAA should formulate a clear,
concise weather forecast improvement strategy with defined
tradeoffs. This should also include well-defined agreed-upon
metrics for improvement. NOAA should also carefully consider
whether its current structure can foster the desired level of
innovation. And finally, NOAA should build deeper, more
collaborative research and development partnerships with
industry.
In conjunction with NOAA's efforts, Congress should support
NOAA by not overtaxing NOAA with new responsibilities that do
not deliver targeted value, particularly if they don't come
with a supporting budget. Congress should also adhere to the
established timeline for the budgeting process. Federal
agencies such as NOAA can't be expected to be innovative forces
in terms of improved forecasting without the capacity to
effectively plan. The lack of budgets also has a cascade effect
on NOAA's partners.
I want to thank you for this opportunity to present this
testimony, and I welcome any questions or comments you may
have.
[The prepared statement of Dr. Petty follows:]
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
Chairman Miller. Thank you, Doctor.
I now recognize Dr. Scott Weaver for 5 minutes to present
his testimony.
TESTIMONY OF DR. SCOTT WEAVER,
CHIEF EXECUTIVE OFFICER, CLIMET CONSULTING
Dr. Weaver. Thank you. Chairman Miller, Ranking Member
Ross, and Members of the Subcommittee, I'm Dr. Scott Weaver,
Founder of CLIMET Consulting LLC, a new consultancy focused on
science and policy issues and weather and climate. I've also
served in several scientific leadership roles within the
Federal Government, including as a Research Meteorologist for
NASA and NOAA, Director of the interagency National Windstorm
Impact Reduction Program led by NIST, and most recently as the
inaugural Executive Director of the Interagency Council for
Advancing Meteorological Services, or ICAMS, at the White House
Office of Science and Technology Policy. Thank you for the
opportunity to appear before you today.
The ICAMS aspirational goal, as stated in its 2020 charter,
is ``The United States will lead the world in meteorological
services via an Earth system approach, providing societal
benefits with information spanning local weather to global
climate.'' So the question before us today is essentially what
are the requirements for the United States to lead the world in
meteorological science and services and ensure that all sectors
of American society are prepared for the pervasive weather
threats that menace our country on an increasingly frequent
basis?
This is a timely question. Since 1980, the United States
has experienced over $2.5 trillion in economic losses and over
8,000 fatalities from extreme weather events, with $1 trillion
of those losses occurring in the last 7 years alone. These
weather disasters highlight a somewhat uniquely American
experience. No other country in the world faces such a
diversity, frequency, and intensity of extreme weather events.
As such, our national weather prediction efforts are
challenging by comparison, and in some instances, require
targeted phenomenologically based modeling approaches that may
not be as necessary for other countries. Indeed, the United
States experiences approximately 1,400 tornadoes per year on
average, dwarfing the annual tornado count of any other nation
in the world.
Despite these challenges, our Nation has made significant
progress in numerical weather prediction. One shining example
is the reduction in the average 3-day hurricane track forecast
error from 450 miles in 1970 to just 50 miles today, an
incredible achievement. Nevertheless, there still exists a
critical need and an incredible opportunity for the United
States to lead the world in meteorological science and
services. To achieve this requires coordinating a thriving
research and applications enterprise to finally begin to bend
the curve in the Nation's annual weather disaster losses. Given
that ICAMS agencies invest almost $6 billion annually in the
meteorological enterprise, it is critical that the Federal
Government strengthen coordination of its own meteorological
assets.
However, significant obstacles exist. Varying missions,
budget development approaches, and legislative requirements
make it challenging to effectively develop shared priorities
and practices across the range of meteorologically sensitive
Federal agencies. In recognition of these challenges, Congress
created ICAMS to elevate Federal coordination of meteorology to
the White House level.
In response, the White House and the ICAMS agencies
chartered four standing committees, one on observations, one on
computing, one on services and one on research and innovation.
If properly resourced and supported, this structure has the
potential to accelerate progress on an array of important
issues in meteorological science and services, including
numerical weather prediction.
Through these committees, ICAMS has achieved some success
already, including advancing interagency coordination of
meteorological science priorities for wildfire and initiating
the development of the first ICAMS decadal strategic plan, an
effort that must continue for ICAMS to achieve its goal.
Despite these bright spots, I am concerned that ICAMS may
never reach its full potential. Strengthened provisions are
needed in the Weather Act Reauthorization to ensure that ICAMS
remains elevated as a top national priority so that it can
effectively respond to urgent and rapidly evolving national
needs.
Indeed, over the last two decades, the meteorological
enterprise has undergone revolutionary changes, the most
notable of which is the rapid evolution of the private sector
from primarily developing end user products and services to
that of a multibillion-dollar commercial enterprise that is in
direct competition with the Federal Government to provide the
very foundational data, modeling, and information that was once
only the purview of Federal meteorological agencies. This
rapidly evolving landscape requires a nimble and effective
national strategy to make what amounts to an order-of-magnitude
change in how the Federal Government leverages the commercial
sector for mutual benefit.
In closing, I provide several recommendations for
consideration by Congress and the President. These include
mandating the completion of the ICAMS decadal strategic plan,
reimagining the Interagency Meteorological Coordination Office,
unleashing the private sector by aligning relevant Federal and
commercial research and development goals, fostering engagement
at the agency-head level to inform Federal agency budget
development activities, and implementing mechanisms to
encourage and enable the participation of non-Federal
stakeholders in ICAMS.
I thank the Committee for its recognition of the need for a
strong meteorological enterprise. I look forward to our
discussion today, and I'm happy to answer any questions you may
have.
[The prepared statement of Dr. Weaver follows:]
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
Chairman Miller. Thank you very much. And I thank all the
witnesses for their testimony.
The Chair recognizes himself for 5 minutes of questions.
Dr. Jacobs, I'd like to talk about NOAA's commercial
programs because the United States is unique in having many
weather startups and innovators. NOAA has been operating the
Commercial Weather Data Pilot Program for a few years now with
success, despite some--what some might call hiccups in
organization and structure. I understand NOAA intends to offer
short-term contracts for 6 months at a time and then recompetes
those all over again. Do you believe that this has a chilling
effect on the commercial sector or limits the full potential of
a thriving commercial data market? Or even is it correct to say
that the commercial weather data industry is one of the most
valuable assets in--the United States has when compared to
global competitors?
Dr. Jacobs. It's a hugely beneficial program. So a lot of
commercial data that's available, it has a huge impact on the
forecast models. It's really, really valuable. There's more out
there that can be acquired. A lot of these companies are
producing data at a rate faster than NOAA can acquire it.
I would say that from an investor standpoint, I hear often
a lot of issues with confidence in long-term investment in
developing and deploying these programs if they're faced with
short-term contracts. So a lot of times there's what we call
nonrecurring upfront engineering to deploy and build out these
networks to sense things like weather. That's expensive. And
these companies would like to have more confidence that they're
going to get a long-term return on their investment before they
deploy a lot of these assets.
Chairman Miller. Thank you very much. And that all--that
tracks and makes sense, and I appreciate your testimony.
My next question is going to be for all of our witnesses to
chime in on so, you know, feel comfortable to go on in. One
reason that I called this hearing today was to help us better
understand why and how the European global model has been
labeled the gold standard. I understand that the Europeans
largely have the same data that we do, but they focus heavily
on improving their internal model code and data assimilation.
So can each of the witnesses if possible give their opinion on
why has NOAA fallen behind on data assimilation, and what can
we do to catch back up?
Dr. Jacobs. There's a couple different reasons. It's true
the European Centre has had a more skillful global model for
quite some time. I would like to point out that NOAA runs
several different models on different forecast horizons, and
they are by far the leader at all with the exception of the
global model. The U.S. model has been anywhere between second
and fourth as far as forecast skill.
And while there are some datasets that the European Centre
may use that the United States doesn't, it's primarily related
to data assimilation. The European Centre, as well as a couple
other skillful models like the U.K. Met Office use four-
dimensional variational assimilation. The United States doesn't
use 4D-Var yet, although there is a system in development for
4D-Var.
The tradeoff is 4D-Var is extremely computationally
expensive to run, and it requires a tremendous additional
amount of compute both to run it and test it. And that was kind
of the decision back in the day was, with limited compute
resources, how do we divvy up the different models based on the
tradeoffs? And it turns out that, you know, 4D-Var, I think, is
going to add the most value.
Chairman Miller. Thank you.
Dr. Petty. I would add from the standpoint of maybe
organization and what the focus is. As noted in several
testimonies you heard today, the U.S. itself has a number of
different issues that we have to target when it comes to
weather, and the timescales are really broad, everything from
nowcasting to medium-range to longer-term type of forecasting.
When we think about the Europeans and the ECMWF (European
Centre for Medium-Range Weather Forecasts), specifically, they
have a very targeted, focused mission that really helps with
their efforts of advancing what they do. They do not have to
spend the time really deploying resources in so many different
areas because of their focused mission.
The other thing that I might add on top of that is really
the aspect of being diluted is quite challenging for us in the
United States. And if we can have a more focused way of
targeting improvements and have the right metrics, that is key.
What does it mean to have forecast improvement? We can talk
about 500 millibar height, or we can talk about how have you
forecasted tornadoes. And so really having focused metrics is
something that we need to consider.
Chairman Miller. Thank you.
Dr. Weaver. Ten seconds, can't really add much except--no,
that's OK--except to say that I typically view the
meteorological enterprise larger than just the modeling
component. Modeling is foundational, but there's a greater
enterprise out there that we can also be focused on in addition
to just the foundational modeling.
Chairman Miller. Thank you all very much.
I now recognize Ranking Member Ross for 5 minutes of
questions.
Ms. Ross. Thank you again to the witnesses for joining us
today.
While the discussion of our weather models' performance is
highly important, at the end of the day, the goal is for there
to be effective and actionable weather information that aids
decisionmakers and protects lives and property. Researchers
have found that in some cases, increased warning time for
extreme weather can lead to a lack of action, as less urgency
is perceived. There also has been research to suggest that
messaging on personalized impacts can lead to significantly
higher evacuation rates. But at the same time, recipients of
that information found it to be more overblown.
Dr. Jacobs, if the United States were able to achieve a
perfect weather model, there's still a possibility that the
public would not react as intended to certain information. We
certainly see that in North Carolina. How can NOAA and the U.S.
weather enterprise improve the human response to the
information that it puts out?
Dr. Jacobs. Thank you very much for that question. So as
someone who focuses on numerical weather prediction, I'd always
thought the more skillful the forecast, the more lives you
would protect. And you're absolutely right, particularly with
tornado warnings. More lead time doesn't always translate to
more lives saved.
This is really a social science issue. It comes down to how
do you translate this information into actionable decisions
that result in behavior that is going to protect life and
property? I think there needs to be a focus on really
translating this through the social scientists that the
National Weather Service is bringing on board. But
additionally, one of the things that they don't have is they
don't have verification data. They don't have circumstances of
how lives are lost. So it's really challenging to sort of
figure out how do you test and evaluate different sort of
criteria of messaging? Because you may have a tornado warning,
you don't know if the person got the warning, didn't get the
warning, got the warning and didn't know what to do, or got the
warning and knew what to do but didn't have the means to do
anything. There's a lot of different variables other than just
doing damage assessments that we need to actually test and
validate these different metrics and prediction tools that are
coming from the social science community.
Ms. Ross. Dr. Petty, I see you nodding your head. What can
the private sector add? And do you have anything to add to what
Dr. Jacobs has said? And of course, we'll give Dr. Weaver an
opportunity as well.
Dr. Petty. Yes, I think Dr. Jacobs hit the nail on the
head, that the social science component is so important. And
from the private sector standpoint, the private sector stands
poised to be a partner to NOAA when it comes to translating
that information into actionable information that people can
use. And so I think that's where the private sector can really
provide a considerable amount of support.
Ms. Ross. OK. Dr. Weaver, do you have anything to add?
Dr. Weaver. Yes, I have something to add. So when I think
about this from the 10,000-foot level of an organization like
ICAMS and all the different Federal agencies, I've had the
pleasure to work at NIST for several years, which is sort of an
odd place for meteorologists to end up, but what was great
about that was I really got to--I got to work with engineers
and social scientists and all--a very large interdisciplinary
team that even included epidemiology experts, OK, so this is
very interdisciplinary. And this team would study the impacts
of weather disasters, so they would go out after, after the
forecast was made, after the forecast was no longer meaningful,
and they do sometimes year--you know, multiple yearlong
longitudinal studies of the entire disaster, the social
components, how communities recovered, did they get the
information, what was the meteorological science like, the
entire scope. And so I bet you if I told 100 meteorologists
that NIST does this, maybe a few of them would know. And so
this is why something like ICAMS is so important because this
is where that information trading and understanding what
different agencies do can be so important.
So I'll also put a plug in and say, you know, it's time to
reauthorize the National Windstorm Impact Reduction Program.
That is the program that does a lot of this. And it's not in
the Weather Act, per se, but it's a very close relative. And I
think it's vitally important as these--you know, like I said in
my testimony, I'd like to bend the curve in those increasing
billion-dollar disasters. And investigating the different
components, whether it's engineering or social science
afterward, is critical to that effort.
Ms. Ross. Well, Dr. Weaver, we're so glad that you are
testifying today.
And, Mr. Chairman, I'll yield back.
Chairman Miller. Thank you, Ms. Ross.
I now recognize Mr. Posey for 5 minutes of questions.
Mr. Posey. Thank you, Mr. Chairman, and I thank the
witnesses.
Dr. Jacobs, does NOAA weather service's Aircraft
Observation Program have the ability to obtain contrail data
from aircraft, commercial aircraft?
Dr. Jacobs. So they don't actually record the contrail data
itself, but all commercial aircraft record wind and temperature
information, and several commercial airlines also record
relative humidity information through different sensors. And
temperature and relative humidity are the variables in which
you would be able to forecast and predict contrail conditions.
Mr. Posey. Yes, if NOAA doesn't have the ability to do
that, are there other commercial companies that can provide
that data to NOAA?
Dr. Jacobs. There are. In fact, there's a lot of different
commercial companies that pull down different data. All--like I
said, most of the commercial airlines collect this data. It's
just to manage of having a commercial data buy contract to get
the data primarily to cover the cost of communications to run
the data off the plane. But the data does exist, and there are
commercial companies that provide it now.
Mr. Posey. OK. So NOAA does or doesn't have the ability to
model contrails and tell constituents and the people, American
people when and where they will form and if they can predict
them into the future?
Dr. Jacobs. They don't, to my knowledge, do a forecast
specific for contrails, but they do forecast temperature and
humidity at the flight levels, which the planes fly, which you
can use those variables to derive conditions when you would
know contrails would exist.
Mr. Posey. OK. The three of you, yes or no, do you support
NOAA improving the aircraft program so it can be able to
provide more weather data, as well as contrail data from
aircraft?
Dr. Jacobs. Yes.
Dr. Petty. Yes.
Dr. Weaver. Yes.
Mr. Posey. Thank you. Dr. Weaver and Dr. Petty, are there
current commercial data providers whose data would drastically
improve the U.S. forecasting models?
Dr. Petty. Yes, I think there are several companies out
there. I slightly misspoke earlier, but what I wanted to hit on
was that there is an acceleration in innovation. And you can
see that innovation across these new observations, whether they
be space-based or airborne observations or terrestrial. So
there are opportunities out there for NOAA to acquire
additional data sources.
Mr. Posey. What do you think the role of infrared sounding
plays in creating weather forecasts? You know, how can we
incorporate more of those data into U.S. models, and what would
the impact be?
Dr. Weaver. So, I mean, there--as Dr. Petty was discussing,
there's plenty of data sources out there, right? So it's not
just about infrared. And I'm not an expert specifically on
infrared, but I am aware, as I mentioned in my testimony, that
there are several private sector companies that are rivaling
the capabilities of the Federal Government. And this is
something that we have to deal with and figure out a way that
we can use all of the data, whether it's being collected by the
government or the commercial sector. So I know of several
companies that are doing quite revolutionary things in terms of
radars on satellites or radars on the ground, et cetera. And,
you know, these are very valuable data sources that would
certainly--you have to work with your government partners,
though. You can't just throw the data out there and expect them
to be able to know how to assimilate it. It's got to be a co-
production as this goes along through time, right? So--because
there--we have standards that we have to meet in terms of
weather forecasting and the Federal Government. We can't just
throw anything at it. And so it has to be quality controlled
and, you know, assured through some research that these are
adequate datasets for their purpose.
Mr. Posey. What's the role of calibration and data
continuity for both short-term weather and longer-term climate
predictions?
Dr. Weaver. OK. Yes. So, I mean, it's very important for
longer-term climate predictions. I mean, one of the things that
we must do, even in the short-term climate predictions, things
like seasonal, right, which is--you know, varies year to year,
et cetera, those kinds of phenomena, you need to have a very
consistent data base through time to understand them first and
then be able to predict them. So we do have something in the
climate and weather community called reanalysis, which tries to
go back and collect all the observations through time and then
streamline them or quality control, if you will, to make them--
you know, an observation taken in 1980 would be similar to the
one today. But they're not, so you need to figure out ways to
put them on an even keel.
Mr. Posey. Got you. I see my time's up. Mr. Chairman, I
yield back. Thank you.
Chairman Miller. Thank you, Mr. Posey.
I now recognize Ms. Lofgren for 5 minutes of questions.
Ms. Lofgren. Well, thank you, Mr. Chairman. And thanks to
each of these witnesses, very interesting, and I've learned
some things this morning, which is always a treat in the
Science Committee.
Obviously, we have challenges not only in data collection,
but data assimilation and distribution. I have a question
about, you know, once those challenges are addressed, which may
take a while, how are we thinking about using AI if at all? I
mean, NOAA had a pilot project in partnership with a company
that specializes in AI language software to translate weather
information more accurately and efficiently into different
languages, and so that would be more accessible to non-English
language speakers. But it occurs to me that there may be a
greater capability with AI. I'm wondering what each of you
thinks about that. And then also, how do we deal with the
persistent problem of hallucinations? So if each of you could
address that, I would appreciate it.
Dr. Jacobs. I think there's a tremendous opportunity for AI
all across numerical weather prediction, even down to
potentially replicating the entire modeling system. So right
now, the current state-of-the-art is physics-based models. AI
can replicate a lot of what those algorithms do. And the
advantage, I think, with AI is that you can run this on a tiny
fraction of the amount of compute, but you still need the
physics-based modeling systems and the observations and the
data assimilation to build it--at a minimum, a training dataset
to train the AI systems on. So I don't think we'll ever see a
situation where we can predict the weather going forward in the
future without correctly observing the atmosphere because it's
an initial value problem. But I think there's huge potential
benefits to gain from AI in optimizing the speed and efficiency
of which the physics-based models predict the weather.
Ms. Lofgren. Thank you. I mean, I didn't mean to suggest
that we wouldn't need the actual data. Obviously, we do. But
how we assimilate and crunch it could be of value.
Dr. Jacobs. It is. Well, I often hear people talk about how
AI is going to replace physics-based modeling.
Ms. Lofgren. Nope.
Dr. Jacobs. And I don't understand how that's theoretically
possible because you'll always need this type of system to do
the training and the air analysis on the observing systems.
Ms. Lofgren. Correct. Dr. Petty, do you have any additional
comments?
Dr. Petty. I would just echo Dr. Jacobs, but I would add,
even beyond our ability to leverage AI from the context of
forecasting improvement, I think we're going to see it in other
areas of the entire value chain all the way down to how do we
translate data into actionable information? I would admit I was
one of those individuals who is a little bit kind of
standoffish, if you will, on AI, but seeing what has been done
in the last couple of years, it's going to be a powerful tool.
Dr. Weaver. Yes. So obviously, we're at a technological
revolution here. I certainly agree with what my two colleagues
here have just said regarding the training and, you know, where
we're going with this. You know, when I was a research
scientist, you know, and all day every day, you're looking at
that, you're just looking at data and trying to understand
things, you know, we live in a culture of doubt. I know that's
surprising for some people, that we are not as always as sure
as we would like to be. And that's part of the scientific
process. You're always doubting your results.
So when I see these papers coming out that are like, we're
beating, you know, numerical weather prediction, et cetera,
it's very exciting to see that. I'd like to see more. Yes, I've
only seen a few. So maybe there's more, I don't know. But that
could just be the scientists and the culture of doubt. If
this--if what some of these claims are making are really true,
then that's fantastic because, you know, when Neil was talking
about the training, once you train the system, it's over. Then
you just need to run it, right? And that could be done on a
laptop, whatever, right? You just--you don't need the massive
amount of compute. Once you have this----
Ms. Lofgren. Well, you need to train the system----
Dr. Weaver. You do, yes.
Ms. Lofgren [continuing]. But you still need data for it to
crunch.
Dr. Weaver. Hundred percent. I mean, that's--that--yes. So
we need both, but the idea here is the forecast can be done
much quicker because you've trained it already essentially. You
don't have to wait for it to crunch out through. So very
exciting time. I'd like to see a little more personally, but
that's just my own opinion.
Ms. Lofgren. Of course. I know I'm putting you on the spot,
but if you have after this hearing some thoughts on how to deal
with guardrails, as well as the hallucination problem that is
persistent, I'd be eager to hear it after the hearing.
With that, Mr. Chairman, I yield back.
Chairman Miller. Thank you, Ms. Lofgren.
I now recognize Mr. Collins for 5 minutes of questions.
Mr. Collins. Thank you, Mr. Chairman. Dr. Weaver, I kind of
want to pivot off of that just a little bit, and maybe not
pivot, just ask some more questions about it because there's
several things here at play. Like we all have talked about for
the past year about China and the threat has been one of the
greatest threats to our country that there is. And I know that
they do a good job of AI. And I guess first question is where
do we stand in relationship to China on the AI race and maybe
some quantum computing in there as well?
Dr. Weaver. Yes, so I--I'll be honest with you, I'm not an
expert on where we stand with China in terms of AI, so I
wouldn't really have much to add to the hearing on that. I
don't know if any----
Mr. Collins. Do you think that--do you think their AI
modeling is pretty good? Have you seen any of their forecasting
modeling?
Dr. Weaver. I have not. Therefore, I can't really comment
on it.
Mr. Collins. Have any of--anybody?
Dr. Petty. I have not seen their output. I think one thing
that I would maybe caution--and I think this is what Dr. Weaver
was referring to--is, when we hear claims about these models,
it's really important to understand, well, how are these
models--AI models verified, and what were they verified
against? And until we can see, you know, those data and how
that was done, it's very difficult to ascertain how well maybe
the Chinese or others are doing. So there really needs to be a
rigorous----
Mr. Collins. Well, I mean, isn't it--it's easy to verify it
after the fact whether or not it was true or not, right,
whether or not it happened.
Dr. Petty. But a lot of times in science when you're doing
that verification, you need to know the methodology that was
followed so that methodology can be replicated and make sure
that that was done correctly. And I--from kind of your question
about the Chinese, I don't know if they're verifying against
500 millibar height or temperature at the surface or, you know,
wind speeds in New York. And really having that verification,
thorough verification is helpful.
Mr. Collins. Right. Well, I've been in small business for
over 30 years, and just seeing the accelerated advancement of
AI even in the transportation industry, which I'm in, it's just
amazing. And I would like to know where--you know, where NOAA
stands and where we stand from a private industry standpoint in
the country because there is a race. And China's spends more
money than the entire world on quantum technology and quantum
computing. And that's in everything, not just defense, not just
hacking, and--because they want to rule the world one day.
Dr. Jacobs, I was walking in, so I'm sorry. We've been kind
of running around like crazy here this morning. You were
talking about cloud computing. Is NOAA ready for artificial
intelligence?
Dr. Jacobs. So right now, they're testing out a couple
different AI systems to replicate a lot of what the physics-
based models are using, and they're showing a lot of promise.
As I was saying earlier, I think the benefit there is you save
on computational resources, at least for the operational
prediction system, but you still need to generate the training
datasets.
When it comes to cloud, NOAA is definitely headed in that
direction, as well as a lot of the private sector, most of the
universities. I think a lot of--at least in the private
sector--I don't really know any companies that own their own
high-performance computing anymore.
Mr. Collins. Right.
Dr. Jacobs. Pretty much all use commercial cloud. The
beauty of commercial cloud is probably more seen in the
research and development than it is in the operational because
you can--you buy on demand. You only buy what you need, and
when your research needs fluctuate, that's an extremely cost-
effective way to do it because if you have a series of
experiments that you want to test with traditional on-premises
high-performance computing, you only have a limited amount of
resources so you run the experiments one at a time. With the
commercial cloud option, you can pay for all of these
experiments to be run in parallel, simultaneously. It costs the
same whether you run them all at the same time or one after the
other. But the beauty of running them all at the same time is
you get the research done faster.
Mr. Collins. Good. I've got one other question. I know
we're down to less than 25 seconds. Dr. Jacobs, I introduced
the Weather Information for Agricultural Act, and I didn't know
if you could take just a minute to explain the advances in
drought weather modeling and how it can best be used by
farmers.
Dr. Jacobs. There's a lot of advancements, and actually,
there's a lot of work being done on seasonal and subseasonal
prediction with AI, too, because a lot of it--there are dynamic
models that run out to 9 months. But a lot of this is pattern
recognition, seasonal patterns like ENSO (El Nino Southern
Oscillation) and PDO (Pacific Decadal Oscillation) and things
like that. That's actually a really unique spot for AI, so I
think there's a lot of advancements being done on those
different forecast horizons. And most of the agriculture
community, they know exactly how to use the data. They just
want to know both what is the prediction----
Mr. Collins. Right.
Dr. Jacobs [continuing]. And what is the confidence in the
prediction.
Mr. Collins. All right. Thank you. Sorry about that, Mr.
Chairman. I yield back.
Chairman Miller. No worries. I'd like Mr. Frost for 5
minutes of questions.
Mr. Frost. Thank you, Mr. Chairman.
I represent central Florida, and in Florida, we get a lot
of hurricanes. Every hurricane season, friends, family,
siblings, myself, we see an increasing number of dangerous
storm systems develop and make their way toward our State. Mr.
Jacobs--or, Dr. Jacobs, I first off want to thank, you know,
the work that you're doing and that NOAA and all the staff are
doing. You know, this past hurricane season, I spent a lot of
time at our county and city EOC (Emergency Operations Center)
of the city of Orlando and county. When I asked them how their
relationship is with NOAA, they have nothing but good things to
say about your staff and the work that y'all are doing and
keeping us safe in the entire State of Florida, so I appreciate
that.
Dr. Jacobs, when it comes to hurricane preparedness, why is
it important to understand the relationship between climate and
weather?
Dr. Jacobs. Well, climate modulates the weather, and so
while we have the ability to predict the weather, particularly
when it comes to hurricanes, the fuel for those is the sea
surface temperatures. And we also have seasonal influences. For
example, this last season, the prior one, we had El Nino, which
in the Atlantic, there tends to be a lot of shear, which
prevents the development of storms. That's nature's way of
removing heat and transporting it toward the poles. And so as
we transition into a La Nina year, I think the seasons will be
a good bit different.
But understanding the relationships between the changing
climate, the influences that feed these storms, that I think
is--there's a lot of research that needs to be done there. The
exciting thing to me is we're--the capabilities and forecasting
these storms, both track and intensity, are really, really
improving.
But one of the areas in hurricane forecasting that I think
needs additional work and focus on is rapid intensification. We
always see storms that when they come into the Gulf,
particularly over the loop current, there's a tremendous amount
of rapid development that the models struggled to handle. And
that, I think, is an area where there needs to be additional
work and research above what's being done right now.
Mr. Frost. Yes, I mean, in the event of a hurricane, good
forecasting and modeling can really save lives. What--where
have you seen gaps in successfully communicating the severity
of these storms to people?
Dr. Jacobs. I think that there's work to be done on the
track forecasting. I've got to give a tremendous amount of
credit to the Hurricane Center. They just rolled out a new way
to sort of visualize the cone based on impacts. But there was--
you know, a translating model, both the projection and the
uncertainty, is tricky because a lot of people tend to focus on
just the center track of the storm when the effects of the
storm could be hundreds of miles away. Having people being able
to visualize the impact of a storm surge, it's one thing to
just say this storm surge may be 5, 10, 15 feet. It's another
thing to actually be able to visualize how high 15 feet of
water is. Getting this message out there, I think, it's largely
a social science problem, where even though we're improving the
skill and the models, we still need to figure out better ways
to message, to visualize the output in a way that solicits a
proper response.
Mr. Frost. Yes, I have a bipartisan bill with Congressman
Webster on this Committee, the Fixing Gaps in Hurricane
Preparedness Act. It's actually included in this Committee, the
Weather Act, and it would direct NOAA to study the ways that
most at-risk populations, seniors, folks with disabilities,
people who speak different languages, how they react to
hurricane forecasts and communications that come from the
government.
Dr. Jacobs, do you think focusing on these types of
populations align with what you think some of the gaps are?
Dr. Jacobs. Absolutely. I think that one of the biggest
challenges--and we've seen this with tornado watches and
warnings, but it's the same thing with the hurricanes is we
need verification data. Do the people understand the warnings?
If they do understand the warnings, do they know what to do? If
they do understand the warnings and they know what to do, can
they take action? There's a lot of different situations where
people, even though they know exactly what to do, for one
reason or another don't take action. Maybe they can't take
action. And this, I think--this needs both verification
information from the circumstances of all these different
populations, as well as tailored messaging.
Mr. Frost. Yes. Yes, the bill directs NOAA to do the study,
but really will have impacts for municipal government, State
government, Federal Government, and all types of nonprofits and
players in this space who helped reduce the amount of lives
that we've seen lost during these intensified storms that are
getting--lasting longer and creating more devastation due to
the climate crisis. So I appreciate your work.
You know, last hurricane season, I had a great offer from
NOAA to hop on the Hurricane Hunters plane. I wasn't able to do
it just because it was my first hurricane season, and I wanted
to make sure I was home. But next time, I think I'll hop on.
Thank you. I yield back.
Chairman Miller. Thank you, Mr. Frost.
I just want to say I want to thank all the witnesses for
their valuable testimony and the Members for their questions.
The record will remain open for 10 days for additional comments
and written questions for Members. Once again, thank you to our
witnesses.
This hearing is adjourned.
[Whereupon, at 11:01 a.m., the Subcommittee was adjourned.]
Appendix
----------
Answers to Post-Hearing Questions
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
[all]