IEEE Geoscience and Remote Sensing Magazine - September 2021 - 33

Figure 2's first column stratifies the observing systems
by hosting platform type-in situ and remotely sensed
surface- and space-based platforms. The space-based systems
are further stratified by sensor type and deployment
strategy, i.e., whether the sensors are deployed individually
or in a constellation. The choice of the satellite observing
system configuration (platform, sensor, and deployment)
affects the usefulness of the observations for
different applications.
Space-based observations must go through a scientificprocessing
step (the second column of Figure 2), which includes
calibration, transformation, aggregation, and validation.
This scientific-processing step is required to produce
useful data products to satisfy the information needs of
applications and services that support user decision making.
Calibrated data and data products are utilized in the
aggregation phase to produce blended data using multiple
techniques, depending on the application area targeted. To
ensure quality, an important parallel validation process occurs
in the background in which the geophysical observables
are compared to independent reference data. Note
that most of the processing steps in the second column
occur both in the originating centers and the applications
centers, especially the weather forecast centers. In particular,
the weather forecast centers continually monitor and
quality-control the observations.
This overview of the observing systems' value chain
and the interconnections that exist among the applications
and the observations show that the value chain is actually
a value network. An observing system supports a multitude
of applications (the third column of Figure 2), and,
in turn, each of these applications is often used to provide
multiple services to many different users (the fourth column
of Figure 2). The components of the EOSC are among
the costliest parts of the network, and efforts that provide
the most cost-effective combination of observations suited
to the intended applications are a key focus of meteorological
and space agencies. Identifying the value network connections
of the observing systems, in combination with
estimates of the costs, is therefore critical to 1) assess the
ROI of existing observing systems and 2) develop strategies
and techniques to design a cost-effective combination
of future observing systems, responding to the full range of
applications and end-user services to which they contribute.
The recent NSOSA study [7]-[9] and the European Organisation
for the Exploitation of Meteorological Satellites'
cost-benefit analysis of the polar system [28] are examples
of such studies.
OVERVIEW OF CURRENT CONSTELLATION
CAPABILITIES
The EOSC is composed of hundreds of satellites observing
Earth's environment. The need for these global observations-continuously
updated-is driven by stringent requirements
these
prediction systems depends strongly on the state of
the environment at the initial time of the forecast.
THE SATELLITES THAT MAKE UP
THE CONSTELLATION
Diverse satellites and sensors in different orbits are necessary
to monitor all of the important environmental parameters
with high spatial and temporal resolution globally. No
one sensor type, satellite, or orbit can make all of the required
useful observations with the necessary performance
level of the attributes given in Table 1.
Although the United States has pioneered the use of
Earth observation (EO) from space and continues to play
a major leading role, a significant surge in the number of
satellites from European and Asian countries in the last
two decades has contributed to the continuous global
coverage of Earth's environment. The sharing of these satellite
data enables an increase in the amount and types of
observations for the common goal of better understanding
and supporting forecasting and other applications,
while the investment requirements remain the same or
even decrease over time for individual nations. This major
benefit for all nations relies on the ability to maintain
and strengthen the time-honored custom of freely sharing
essential environmental observational data among
worldwide national hydrometeorological and meteorological
services.
It is worth noting in this context that the sharing of
surface and upper-air observations (e.g., radiosondes and
aircraft data) is equally important to NWP systems, and,
therefore, the principle of sharing applies to all nations,
whether or not they operate satellite-based meteorological
sensors. This is because the value of the data from the observing
systems is highly dependent on the application and
sometimes on the specific location of the observations. The
volume of the data, at least if considered alone, is not a robust
metric for assessing the value of the data. For instance,
a single radiosonde located in the middle of the Pacific
Ocean, launched at a time when no satellite data were available,
could turn out to be critical for an accurate weather
prediction in regions far distant from that radiosonde. This
should be highlighted to encourage all nations to share
their local measurements and to correct the misconception
that individual local measurements are not as important as
other " voluminous " data.
All satellites have a design lifetime, but, in many cases,
from numerical-environmental monitoring
and prediction systems. The accuracy of the products of
SEPTEMBER 2021 IEEE GEOSCIENCE AND REMOTE SENSING MAGAZINE
they outlive this lifetime. TRMM, launched in 1997, survived
proposals to deorbit in 2005 and continued to provide
data until 2015, well beyond its initial design lifetime. It is
cost-effective to continue making relatively small investments
to maintain such legacy systems as these can provide
valuable data. For example, data-denial experiments
recently conducted by the ECMWF [29] show no saturation
of benefit as MW sounders are added to a baseline experiment.
These include those on the older National Oceanic
and Atmospheric Administration (NOAA) satellites that
33
IEEE Geoscience and Remote Sensing Magazine - September 2021

Table of Contents for the Digital Edition of IEEE Geoscience and Remote Sensing Magazine - September 2021
