|
NPOESS:
From Photons to Forecasts
Providing Critical Data for Weather Predictions
Dave Jones, Mike Haas, and Craig Nelson
This is the sixth in a series of
articles on the National Polar-orbiting Operational
Environmental Satellite System (NPOESS). As the backbone
of the Global Earth Observation System of Systems, NPOESS
will become the main data source for Numerical Weather
Prediction (NWP). NPOESS will help to improve forecasts of
the weather and Earth’s environment through higher
quality and more timely land, sea, atmosphere, and space
observations.
Introduction: Numerical Weather
Prediction
“Coming up next, your seven-day weather
outlook.” “I’ll be back with a look well into next
week….” Ever wonder how the weather forecaster on
television can look into the future each and every day?
Have you thought about where forecasters get the data they
use to make predictions across the globe?
Since the 1950s, meteorologists have used improving
computer models in a scheme called Numerical Weather
Prediction (NWP) to produce the guidance for the forecasts
that are produced for a wide range of purposes and by a
wide assortment of people, from an anxious bride and groom
planning their outdoor wedding to a commander planning a
military operation. Advances in computer technology, our
improved understanding of the underlying physics used in
modeling, and better observations are making weather
forecasts timelier and more accurate. Environmental
satellites play a key role in providing critical input
data worldwide for numerical weather prediction models.
The sophisticated computer models used in weather
prediction are driven by huge volumes of data with more
than 125 million observations coming in daily, mostly from
satellites. According to Dr. Louis Uccellini, the Director
of the National Oceanic and Atmospheric Administration’s
(NOAA) National Centers for Environmental Prediction (NCEP),
more than 99 percent of the data used for making weather
forecasts is derived from satellites, primarily from
polar-orbiting environmental satellites (e.g., NOAA’s
Polar-orbiting Operational Environmental Satellites—POES).
Satellite observations are combined with in situ
observations and fed as input into the models in a process
known as data assimilation. Satellites are especially
important in providing observations in parts of the world
where in situ data are few and far between, such as the
oceans and uninhabited regions. “All forecasts are local
and all [forecasts] start with the Global Observation
System,” said Dr. Uccellini. “The global observations
pass through numerical weather prediction models and are
projected out to 1 to 2 to 3 and on out to 14 days.”
Improved Satellites for Better
Forecasts
With the societal importance of accurate and timely
weather forecasts in mind, the National Polar-orbiting
Operational Environmental Satellite System (NPOESS) is
being designed to improve predictions even further. NPOESS
sounders and imagers will bring increased fidelity
(improved resolution, accuracy, and precision) to the
data, reducing the initial errors introduced into the
models from observations. Equally important is an
improvement in data latency, the time from when a photon
is sensed by the satellite to the time a product is
produced by the system. Current systems take around two
and one-half hours to assimilate data into a model and
produce an output. A large part of the delay comes from
ingesting satellite data. NPOESS will use an innovative
system to reduce the time it takes to obtain data by
nearly one hour. A decreased latency should provide a
tremendous benefit to modelers as well as to the users of
model output. “We get as many kudos from having a 99.5%
on time delivery as we do for increasing accuracy,” said
Dr. Uccellini, “we listen to our customers when they say
lost time in distributing model forecasts is a waste of
resources.”
To prepare for the increased data volume that will
be available from NPOESS, agencies from within the United
States government have banded together to fund the Joint
Center for Satellite Data Assimilation. The center focuses
on data assimilation, bridging the gap from making
satellite observations to ingesting these data into
numerical models. “Five years ago,” said Dr. Uccellini,
“it would take three to four years to test and use new
satellite data; on a five-year satellite you lose 20-40%
of the effective life.” The Joint Center’s goal is to
have data ready for use within one year after launch and
available for use in the environmental community as soon
as possible.
Military Weather Forecasting
In addition to the country’s civilian
meteorologists and public, the users of numerical weather
predictions include those who plan and execute military
operations. Both the Air Force Weather Agency (AFWA) at
Offutt Air Force Base and the U.S. Navy’s Fleet
Numerical Meteorology and Oceanography Center (FNMOC) at
Monterey have worldwide and regional NWP models to support
military missions. Much of their input data comes from
satellites as well. According to Dr. Nancy Baker of the
Marine Meteorology Division of the Naval Research
Laboratory (NRL) in Monterey, the U.S. Navy’s global and
mesoscale numerical weather prediction systems currently
assimilate a wealth of observations from both
geostationary and polar-orbiting satellites. These
observations, particularly in data-sparse regions, are
critically important for initializing the NWP models and
stratospheric analyses used to provide tactical weather
support to the U.S. Navy, Marine Corps, Air Force, and
other Department of Defense (DoD) activities, as well as
some civilian activities.
FNMOC assimilates data from a variety of satellites
ranging from the National Aeronautics and Space
Administration’s (NASA) research missions to operational
NOAA and DoD satellites. Data, such as precipitable water,
wind speed and direction, and energy levels (known as
radiances) are received from sensors similar to those that
will be on NPOESS. This fall, NRL/Monterey will test the
assimilation of vector ocean surface wind retrievals from
WindSat, a risk reduction mission for the Conical-scanning
Microwave Imager Sounder (CMIS) instrument that will be on
NPOESS.
Future plans at NRL include developing data
assimilation methods for the Atmospheric Infrared Sounder
(AIRS) on NASA’s Aqua spacecraft (in preparation for the
NPOESS Cross-track Infrared Sounder (CrIS) sensor), and
adding Global Positioning System (GPS) radio occultation
assimilation. Excellent progress has already been made on
the development of a four-dimensional NAvy Variational
Data Assimilation System using the accelerated representer
method (NAVDAS-AR), a system that will be enhanced with
more timely NPOESS data. Also on the books is a plan to
develop the aerosol assimilation components for the Navy
Aerosol Analysis and Prediction System (NAAPS), again
relying heavily on satellite observations. Overall, NRL is
actively developing a suite of data assimilation
techniques to handle the wealth and variety of satellite
observations that will be available through NPOESS.
AFWA leads the world in meeting the operational
challenges of analyzing and forecasting global cloud
cover. Global and regional products generated from
AFWA’s Cloud Depiction and Forecast System II (CDFS II)
are used to directly support the warfighter. CDFSII
processes data from a variety of polar-orbiting and
geostationary satellites to produce quantitative,
worldwide cloud analyses, and forecasts. The higher
spatial and temporal resolution cloud imagery data that
will be available from the Visible/Infrared Imager
Radiometer Suite (VIIRS) instrument on the NPOESS
Preparatory Project (NPP) and NPOESS will lead to
significant improvements in AFWA’s ability to forecast
clouds accurately and far enough in advance of mission
times to really help the warfighter.
It is anticipated that improved information from
NPOESS will help to provide the best possible input to
numerical weather prediction models and improve weather
forecasts for worldwide users, both military and civil.
Specific payoffs are likely to include better analysis and
prediction of marine winds, slant-path visibility, and
clouds, all of which are important for landing and
launching aircraft from carriers and other tactical
applications; improved specification of low-level
temperature and humidity structures; improved hurricane
track and intensity forecasts (vital for safety of people
and assets); and improved upper atmosphere analyses and
predictions. While the costs of developing the data
assimilation tools for NPOESS observations are
significant, recent successes at NRL and other NWP centers
have clearly demonstrated that substantial gains in
analysis and forecast skill can be made when improved
satellite observations are properly assimilated.
From Weather to Climate
Sea surface temperature (SST) is an important input
or boundary parameter for numerical weather prediction and
climate models. Heat transfer from the ocean to the
atmosphere in part drives weather systems, fueling
hurricanes and tropical cyclones. The use of more accurate
SST information in models is yielding demonstrable
improvements in medium- to long-range forecasts.
Space-based sensors like VIIRS on NPOESS can only measure
the temperature of the surface “skin” (top few
millimeters) of the ocean. However, when SST data are
combined with measurements of sea surface height and in
situ profiles of subsurface temperatures, through
numerical ocean models, the full three-dimensional
structure of the ocean’s temperature field can be
resolved. In the NPOESS era, SST data from VIIRS, sea
surface height measurements from the NPOESS Radar
Altimeter, and subsurface temperature measurements from
observing systems such as the autonomous ocean profiling
Argo floats will provide the primary input data for NWP
models. Over longer time scales, NPOESS ocean temperature
data will be used to monitor the onset and decay of El Niño/La
Niña conditions in the Pacific Ocean and to yield clues
about shifts in the “oceanic conveyor belt” that may
drive large-scale changes in climate.
The study of the environment and prediction of
weather are critical for everyday life, but they also
provide important information for other uses, such as
military operations and in the study of longer-term
climate change. Data from NPOESS will play a key role in
all of these applications. In the next issue, we will take
a more detailed look into satellite data and the study of
the Earth’s changing climate.
About the Authors
Dave Jones is Founder, President and CEO of
StormCenter Communications, Inc. (stormcenter.com).
He is also President of the Foundation for Earth Science
and sits on the Executive Committee of the ESIP Federation
(esipfed.org).
Dave can be reached at [email protected].
Mike Haas is the Principal Director for the
Aerospace Corporation at the NPOESS Integrated Program
Office (IPO). Prior to working for Aerospace, Mike was a
meteorologist for the U.S. Air Force for 27 years. He can
be reached at [email protected].
Craig Nelson is the former Executive Director of
the NPOESS Integrated Program Office (IPO). He is employed
by General Dynamics Advanced Information Systems as a
support contractor to the IPO and can be reached at [email protected].
Back
|
