IEEE Geoscience and Remote Sensing Magazine - March 2020 - 81

InSAR-BASED 3D MINING DISPLACEMENT
PREDICTION AND DAMAGE ASSESSMENT
3D MINING DISPLACEMENT PREDICTION
It is essential to predict the ground surface 3D displacements caused by future or planned underground mining
activities so as to assess mining-related geohazards and further make controlling measures, because mining-induced
deformation is the direct cause of most mining-related
geohazards [8], [124], [125]. In fact, the core tasks for 3D
mining displacement prediction are selecting the deformation model and inverting the related model parameters
described previously. Once the deformation model and parameters are determined, the 3D ground surface deformation can be readily predicted [3], [8].
In 2014, Fan et al. [113] presented the first InSARbased 3D maps showing displacement caused by underground coal mining using the PIM and its InSAR-based
model parameter estimates. The model parameter related
to horizontal motions (i.e., the horizontal motion constant) is empirically designated because it is absent in the
constructed function model (see the "Function Model
Construction" section). Such a strategy would cause errors in the predicted 3D displacements, especially in horizontal motions.
In 2016, a method was proposed to predict 3D mining
displacement using the PIM and its full model parameters
inverted from InSAR measurements [111]. The test over the
Qianyingzi coal mining area of China (see Figure 4) demonstrated that the predicted results have good agreement
with the in situ 3D displacements, and relative errors of
roughly 5 and 11% as they relate to the maximum vertical
subsidence and horizontal motion, respectively. Such an
accuracy is acceptable for practical applications [111]. The
PIM is normally used to predict static 3D mining displacements after the whole extraction, rather than the TS caused
by the entire process of mining.
To overcome this limitation, the dynamic (or TS) 3D
displacements over the Qianyingzi coal mining area of
China during the entire process of mining were predicted
based on two spatiotemporal deformation models (i.e.,
temporal and generalized PIM, respectively) and their InSAR-based estimates of model parameters [119], [126]. The
results suggest that the accuracy of the predicted dynamic

3D displacements during the entire process of underground
extraction (i.e., for subcritical, critical, and supercritical extraction) can meet the practical requirements. This, to some
extent, overcomes the poor performance of the PIM for dynamic deformation prediction, but more model parameters
(e.g., parameters related to the Knothe function in the temporal PIM) are introduced accordingly, thus increasing the
difficulty of model parameter inversion with InSAR measurements. Consequently, for critical/supercritical extraction, prediction methods based on the PIM are preferable.
For subcritical extraction or dynamic deformation prediction, however, the methods based on temporal or generalized PIMs are recommended, although more parameters
will need to be inversed.
MINING-INDUCED DAMAGE ASSESSMENT
Mining-induced damage to ground surface infrastructures
is a very common mining-related geohazard. Hence, mining-induced damage assessment is crucial for implementing
damage-control measures. The common method for damage assessment is first predicting the mining-induced slopes,
curvatures, and horizontal strains, because infrastructures
are sensitive to these deformation indicators. The damage
categories can then be defined according to the predicted
deformation indicators and the resistance of the different infrastructures to these three deformations [125]. T
- able 1 lists
a damage category classification method for brick buildings
damaged by underground mining activities.
To reduce the cost and improve the efficiency of traditional infrastructure damage assessment, some InSARbased damage assessment methods have been developed.
Yang et al. [111] proposed a scheme for mining-induced
building damage assessment in which the ground surface
slopes, curvatures, and horizontal strains are predicted by
the PIM and its InSAR-based model parameter estimates.
In 2018, Yang et al. [126] proposed a new method for progressive building damage assessment during the entire
mining operation: the deformation indicators listed in
Table 1 are predicted by the temporal PIM and its InSARbased model parameter estimates. Figure 5 shows the assessed building damage caused by extracting five planned
working panels numbered one to five (marked by the black
dashed rectangles) in the Qianyingzi coal mining area
of China using the temporal PIM and its InSAR-derived

TABLE 1. THE CLASSIFICATION OF MINING-RELATED STRUCTURAL DAMAGE FOR BRICK BUILDINGS [8].
MAGNITUDE OF EXPECTED DEFORMATION

DAMAGE
CATEGORY

POSSIBLE DEGREE OF DAMAGE

1
2

EMAX (10 −3/m)

IMAX (mm/m)

KMAX (mm/m)

A small number of cracks in the walls and ceilings, but fewer than 4 cm.

0 1 I max # 3

0 1 K max # 0.2

0 1 E max # 2

Cracks in the walls and ceilings enlarged to roughly 15 cm, and windows and
doors jammed.

3 1 I max # 6

0.2 1 K max # 0.4

2 1 E max # 4

Cracks enlarged to roughly 30 cm, and windows and doors showing serious
deformation.

6 1 I max # 10

0.4 1 K max # 0.6

4 1 E max # 6

Inclination serious for buildings, and bad jamming in the walls and ceilings.

I max 210

K max 2 0.6

E max 2 6

MARCH 2020

IEEE GEOSCIENCE AND REMOTE SENSING MAGAZINE

IEEE Geoscience and Remote Sensing Magazine - March 2020

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