IEEE Geoscience and Remote Sensing Magazine - March 2020 - 72

To assist readers at different levels in better understanding and accessing recent advances, this article presents a
comprehensive review on the use of InSAR in mining deformation monitoring, parameter inversion, and forward
prediction. More specifically, we first briefly examine the
typical SAR techniques for surface deformation monitoring. We then comprehensively review the recent important
contributions made in SAR/InSAR-based 1D/2D mining
deformation monitoring, 3D mining displacement reconstruction, mechanical parameter inversion, and forward
prediction. In addition, we analyze the pros and cons of
these methods to provide general guidance for choosing
a suitable approach. Finally, we discuss some challenging
issues and possible future trends.
InSAR
More than 80% of the world's energy supply and over 90%
of industrial materials come from natural resource mining
(e.g., crude oil, gas, coal, and minerals) [1]. However, the
mining activities commonly lead to a series of geohazards,
such as landslides, infrastructure damage, land subsidence,
and so forth [2]. The direct cause of most mining-related
geohazards is ground surface deformation (e.g., subsidence,
horizontal motion, and strain) [3]. Consequently, mining
displacement observations with high spatiotemporal resolution play an essential role in mining deformation mechanism interpretation, deformation modeling and prediction,
and geohazard assessment.
Traditionally, a few sparse observation points are first
set up over the mining area of interest. Then, ground-based
geodetic surveying techniques [i.e., Global Positioning
System (GPS), precise leveling, and total station] are used
to repeatedly monitor the mining displacement. Although
this traditional approach can achieve a high level of accuracy, it is characterized by considerable cost, low efficiency,
a small working area, and coarse spatial resolution, which
dramatically hinder the performance of routine mining
deformation monitoring in most mining areas. As a consequence, mining deformation studies and geohazard assessments in these mining areas can be difficult to carry
out, due to the lack of in situ deformation observations.
InSAR is a microwave remote sensing technique that can
measure surface mining deformation over a very large area
with low cost, high spatial resolution, and high efficiency
that traditional geodetic surveying techniques cannot
achieve [4]. For example, theoretically, InSAR can generate
a ground surface deformation map with a swath width of
250 km, a spatial resolution of 5 m × 20 m, and an accuracy
level from centimeters to millimeters using Sentinel-1 SAR
images (interferometric wide mode) [5]. This implies that
mining-induced displacements with a high spatial resolution over a wide area can be monitored using InSAR. As a
result, InSAR can provide a large number of deformation
observations in many mining areas, offering great potential
for promoting or, to some extent, even revolutionizing traditional mining deformation studies, such as deformation
72

mechanism interpretation, parameter inversion, and deformation prediction.
Although the first InSAR-based ground surface deformation map associated with the expansion of water-absorbing clays was generated by Gabriel et al. [6] in 1989,
the first InSAR-based deformation map associated with underground mining was created by Carnec et al. [7] in 1996.
Fortunately, tremendous research progress has been made
in the monitoring of mining deformation using InSAR over
the past two decades. However, limited by the side-looking
imaging configuration of SAR sensors, most works have
been dedicated to the use of SAR or InSAR techniques in
accurately measuring 1D/2D mining displacements along
the radar line of sight (LOS) and/or azimuth directions for
a long period of time. As mining deformation occurs in
3D space, with high nonlinearity in both the spatial and
temporal domains [8], it is difficult to discern the real patterns of mining deformation from InSAR-derived 1D/2D
displacement observations. This dramatically limits the potential of InSAR in mining deformation studies.
To circumvent this, in recent years, significant efforts
have been made to reconstruct 3D mining displacements
from InSAR-derived 1D/2D displacement observations. In
addition, several methods have been developed for inversing model parameters and forward predicting 3D mining
displacements based on InSAR displacement measurements. These contributions have aggressively promoted
the use of SAR/InSAR in mining deformation studies and,
to some extent, have gradually revolutionized traditional
mining deformation studies. Unfortunately, to the best of
our knowledge, no systematic review of recent advances in
this field has been produced.
OVERVIEW OF TYPICAL SAR METHODS
FOR DEFORMATION MONITORING
SAR is an active microwave remote sensing technique that
records amplitude and phase information about the microwave backscatter characterization of scatterers within a
resolution cell using a complex number [4]. Since the first
InSAR-based deformation map was generated in 1989,
many methods have been developed for ground surface
deformation monitoring using recorded amplitude and/
or phase information (referred to as SAR/InSAR methods,
respectively). We classify these methods roughly into two
groups: 1) SAR/InSAR methods for deformation monitoring and 2) SAR/InSAR methods for time-series (TS) deformation monitoring. The following section presents overview of some classical methods in these two groups.
SAR/InSAR METHODS FOR DEFORMATION MAPPING
DIFFERENTIAL InSAR
Differential InSAR (DInSAR) is a classical technique used
for detecting ground surface deformation based on the
interferometric-phase information of two coregistered
SAR images (forming a single InSAR pair) over the same
IEEE GEOSCIENCE AND REMOTE SENSING MAGAZINE

MARCH 2020

IEEE Geoscience and Remote Sensing Magazine - March 2020

Table of Contents for the Digital Edition of IEEE Geoscience and Remote Sensing Magazine - March 2020