IEEE Geoscience and Remote Sensing Magazine - September 2022 - 7

limited in time and spatial coverage. Electromagnetic
waves do not penetrate deep into the ocean, so remote
sensing is limited to the ocean surface. This article introduces
a set of techniques, known as deep ocean remote sensing,
for extracting subsurface information from surface
remote sensing measurements through utilizing the physical
connection between them. Data-driven approaches are
also used to derive models directly from the observations
but sometimes lack some physical interpretation. Various
combinations of dynamics and data-driven methods,
such as data assimilation and the combination of ocean
models with deep learning, promise to incorporate the
advantages of both approaches. Future directions in this
field, as we enter the " big data " era, will include further
work on such integrated approaches and the use of techniques
such as data mining. More work needs to be done
on the physical interpretation of the results, as this is essential
to our understanding of the mechanisms governing
subsurface processes.
Our fourth feature, by Zhang et al., addresses the cryosphere
and looks at the use of SAR interferometry (InSAR)
to monitor permafrost, defined as soil and rock that remain
below freezing for two or more consecutive years.
Permafrost covers a quarter of Earth's land surface, largely
in the Arctic and on the Qinghai-Tibet Plateau. Climate
change is degrading permafrost, which can have important
ecological implications as well as threaten the stability of
engineering construction. InSAR is a very useful tool for
monitoring these changes. This article presents a comprehensive
review of the application of InSAR techniques to
the study of permafrost, introducing the current capabilities
and limitations of InSAR and some guidance for future
research directions. Starting with a review of the basic principles
of InSAR, such as coherence analysis, the article then
shows how these can be applied to measure the deformation
of permafrost and the infrastructure that it supports
and determine the active layer thickness. Challenges in
this application of InSAR include temporal decorrelation,
atmospheric delay, deformation models, soil moisture variability,
and phase unwrapping.
A specific engineering application, monitoring the stability
of permafrost for infrastructure construction, is introduced,
using the Qinghai-Tibet Railway (QTR) as an
example. Constructed in 1958, the QTR is the highestaltitude
longest-route plateau railway in the world. Climate
change-induced permafrost degradation may cause instability
of the railbed, posing a threat to operation and passenger
safety. TerraSAR-X and Sentinel-1A data were used to
detect this deformation. The authors conclude by introducing
some future research directions, including observation
of changes in lakes within the permafrost regions, monitoring
the freeze-thaw cycles, and wildfire-induced subsidence.
More accurate digital elevation maps, soil moisture,
vegetation, and porosity data would improve the retrieval
of permafrost from InSAR. Near-future deployment of advanced
SAR missions (such as NISAR) combined with deep
SEPTEMBER 2022 IEEE GEOSCIENCE AND REMOTE SENSING MAGAZINE
learning and supercomputing are expected to be significant
contributions to this application.
Keeping on the topic of SAR, we move on to two more
articles related to this very popular technique. The first one
looks at target detection using compact polarimetric (CP)
SAR. CP-SAR uses the transmission of a signal with a special
polarization (typically circular) and receives the echo
in a set of orthogonal polarizations (horizontal, vertical,
and right- and left-handed circular). CP-SAR can theoretically
recover much of the same information as full polarimetric
(FP) SAR but with simpler instrumentation. Comparative
studies suggest that the performance of CP and FP
are similar in practice. Target detection, particularly in a
marine environment, has been one of the key applications
of CP-SAR. An in-depth review of CP-SAR, starting from
the basic principles and developing the theory of various
detection methods, is presented in this article. Three promising
research directions are introduced in the article: feature
extraction through exploiting different characteristics
in the response of CP-SAR to different targets, better understanding
of target characteristics, and intelligent detection
techniques. Several nations
and institutions are launching
SAR missions with a CP
observation mode, so interest
in this technique is expected
to remain high.
Our final article on SAR
in this issue, by Liang et al.,
reviews phase synchronization
for bistatic and multistatic
systems, in which
the transmitter and receiver
are physically separated by
some distance. Experiments
using such configurations have been conducted with various
combinations of spaceborne, airborne, and groundbased
receivers and transmitters. Phase synchronization
between the transmitter and receiver is an important
problem in these systems. Synchronization is possible in
three ways: through simultaneous reception of a direct
signal from the transmitter, through use of a dedicated
synchronization link, and by autonomous estimation.
Whereas use of the direct signal or dedicated link can
achieve very low synchronization error, these increase
hardware cost and complexity. Autonomous estimation
techniques are similar to the methods used to reduce residual
motion error in traditional SAR. Two such methods,
autofocus and multisquint processing, are described
in the article. Future development of these methods includes
the MirrorSAR concept, in which each receiver
satellite is a simple transponder upconverting and transmitting
the received signal back to the transmitter for
synchronization. Advances in atomic clocks, of course,
could eventually provide high accuracy without the need
for phase synchronization.
CLIMATE CHANGE IS
DEGRADING PERMAFROST,
WHICH CAN HAVE
IMPORTANT ECOLOGICAL
IMPLICATIONS AS WELL AS
THREATEN THE STABILITY
OF ENGINEERING
CONSTRUCTION.
7
IEEE Geoscience and Remote Sensing Magazine - September 2022

Table of Contents for the Digital Edition of IEEE Geoscience and Remote Sensing Magazine - September 2022
