IEEE Geoscience and Remote Sensing Magazine - September 2021 - 30

sensor attributes include characteristics such as the noise
level, swath, and calibration stability. These attributes are
not included in Table 1. In other words, during the design
process, the sensor and platform attributes are optimized
to generate the desired geophysical performance in terms
of their targeted attributes.
The types of derived geophysical variables determine
the applications for which the observing system is most
useful. Some applications require global coverage (e.g.,
global NWP), and others require only regional coverage
(e.g., sea-ice monitoring). Resolving power is the ability of
a sensing system to separate or distinguish small, closely
adjacent features, either horizontal or vertically. Horizontal
resolving power depends on both footprint size and
TABLE 1. THE GEOPHYSICAL OBSERVABLE ATTRIBUTES.
ATTRIBUTE DEFINITION
Number of
variables
observed
Types of
variables
observed
Geographic
coverage
Horizontal
resolution
Footprint
size
Resolving
power
Temporal
refresh
Vertical
extent
Vertical
resolution
Accuracy
and
precision
Validity
range
Continuity
Robustness
Resiliency
Data
timeliness
The number of geophysical variables that can be derived
from the observing system measurements.
The types of geophysical information provided by an
observing system and its geophysical capability.
The horizontal domain of the observations.
The average horizontal distance or spacing between individual
observations. For imagers, this is often equivalent
to the footprint size.
Horizontal scale represented by a single observation
The ability of a sensing system to separate or distinguish
small, closely adjacent features.
The number of times a given spot on the globe is
observed in a given time period. Alternatively, the time
required to observe the horizontal domain.
The bottom and top levels of the derived observations.
The number of independent pieces of information
[degrees of freedom (DoF)] derived at an observation
location (GIFOV, in most cases). This may be expressed
in kilometers as the vertical extent divided by the DoF.
Often lumped together as the size of errors, which consists
of bias and random errors. Accuracy and precision refer
to bias and random errors, respectively. There are various
metrics of these errors, such as root-mean-square error.
Only those values between the low and high limits of the
validity range are reliably determined.
The time period of overlapping intercalibrated sensors
that can be used to monitor the geophysical variable.
The robustness of the availability of this observation. We
define this as the number of sensors that can be used to
retrieve a given geophysical variable.
How quickly a sensor can be replaced.
The time between the observation time and when the
observation is available for use in applications.
horizontal resolution. A system with high resolving power
has small footprints and high horizontal resolution. For
most observing systems, there are differences between the
ground-instantaneous field of view's (GIFOV's) size and
resolution; for example, the ATMS oversamples, and its
GIFOVs overlap. There is also a distinction between the
horizontal resolution of the physical measurements (e.g.,
brightness temperatures) and that of the derived observables
(e.g., the temperatures and water vapor) because the
physical measurements may be remapped prior to the
retrieval step.
An important aspect of observational quality is the
prior dependence on other data used by the processing
algorithm. Some satellite products are generated by an
EXPLANATION OF IMPORTANCE AND/OR EXAMPLES
The capability of an observing system to measure or retrieve a wide
range of variables makes it relevant to a wider number of applications.
The geophysical capability of a system defines its ability to satisfy the
applications' needs. A scatterometer provides both ocean surface wind
and land surface information.
Polar satellites provide global coverage of Earth's surface. A geostationary
satellite, on the other hand, covers a specific area of the globe.
It is the combination of horizontal resolution and footprint size that
determines resolving power.
Small footprints represent fine scales (e.g., nearly point measurements)
and permit high resolving power if the resolution is high.
High resolving power permits the detection of fine-scale spatial features,
e.g., fronts, thunderstorms, and individual convective cells. It is
also necessary to support precision agriculture, which requires highly
resolved soil and vegetative-state variables.
For some slow-moving environmental features (such as vegetative
phenology or sea ice motion), this might not be a critical attribute; for
hurricanes, storms, or other phenomena with a rapid evolution or diurnal
variation, short temporal refresh is critical.
The full vertical extent of phenomena of interest should be observed. The
vertical extent for global NWP and space weather are quite different.
The vertical resolution is critical for capturing fine-scale vertical features
of the profile, whether it determines the height of the tropopause, detects
temperature inversions, or resolves the cloud layers' heights. The
RO sensors provide excellent vertical resolution for temperature.
This is a core characteristic of an observing system. An observation with
large errors might be of limited value even if it has great vertical and
horizontal resolutions and is provided with a global, dense coverage.
It is possible that the most important values are not measured or
retrieved. For example, current scatterometer designs are not sensitive
to wind speeds greater than 30 m/s, but, for marine warnings, the
maximum wind speed in a storm is critical.
Continuity is particularly important for climate-trend detection.
The robustness of the observing system (for each variable) is important for
all of the operational applications. A gap in some key variables might lead
to a significant degradation of the performance of various applications.
Replacement could be achieved by repositioning an on-orbit asset or by
a launching a new satellite.
Real-time operational applications require timely data.
30
IEEE GEOSCIENCE AND REMOTE SENSING MAGAZINE SEPTEMBER 2021
IEEE Geoscience and Remote Sensing Magazine - September 2021

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