Macro-Micro Response Characteristics of Surrounding Rock and Overlying Strata towards the Transition from Open-Pit to Underground Mining

The macro-micro mining response of the surrounding rock and overlying strata towards the transformation from open-pit to underground mining is examined in the present study, based on the engineering background of the Jinning phosphate mine (Yunnan Phosphate Chemical Group Co., Ltd.) via simulations involving similar materials, digital photographic measurement technology, and numerical simulation. The mining deformation of the surrounding rock underground, and of the overlying strata, is shown to develop in three stages, namely: (1) small and local deformation, (2) continuous linear increase, and (3) the violent nonlinear collapse of the entire system. The internal distribution of stress in the surrounding rock and adjacent overlying strata of the inclined mined-out area is complicated. The degrees of pressure increase and pressure relief have an important relationship with the size of the mining space. The pressure relief is more complete close to the mined area, and the stress reduction decreases with increasing distance. The cracks propagate in arc shapes and have a tendency to penetrate into the upper and lower ends of the stope. The size of the excavation space plays a key role in the generation, propagation, and penetration of the cracks. Due to the disturbance of the first mining level and the increase in excavation depth, the rate of damage to the surrounding and overlying rock increases in the second mining level. This process generates more cracks, which accelerate the instability of the surrounding rock and overlying strata.


Introduction
Most of China's open-pit mines were built in the 1950s. After several decades of continuous high-intensity mining, the vast majority of open-pit mines have entered the deep open-pit mining stage and even underground mining. The transition from open-pit to underground mining is a complex systems engineering problem, and the resulting deformation mechanism of the rock mass is highly complicated due to the effects of numerous stress fields. The process presents typical sudden and nonlinear characteristics and poses a serious challenge to the open-pit to underground mining project [1][2][3][4][5].
Research on the transition from open-pit to underground mining in China began around the 1990s, with many researchers conducting bottom friction modeling experiments [6,7], physical modeling tests [8][9][10], numerical simulations [11][12][13][14], combined numerical and physical simulations [15][16][17][18], and field measurements [19][20][21][22]. Recently, a mathematical model was established to describe the attenuation of peak particle velocity (PPV) in the open-pit slope, which is used to evaluate the influence of underground mine blasting on the slope stability [23]. Cheng et al. used the Universal Distinct Element Code (UDEC) numerical method to simulate the movement of strata in the footwall caused by underground mining [24]. In addition, the discontinuous deformation analysis (DDA) method has been used to study the process of slope instability induced by the caving method in rock structures of various mass during mining from openpit to underground [25][26][27]. Regassa et al. used the equivalent discontinuity modeling method (EDMM) to simulate the rock movement and failure caused by mining under the end slope of the Western open pit of the Yanqianshan iron mine [28]. With advances in technology, researchers have used micro seismic monitoring, in situ monitoring, and true triaxial modeling [29][30][31][32] to investigate the deformation characteristics and failure mechanisms of the slope rock mass and the surrounding rock of the underground stope after the transition from open-pit to underground mining.
Nevertheless, few studies have examined the changes in the surrounding rock and overlying strata of the underground stope due to influences in slope and mining coupled. Due to major differences in the conditions, mining techniques, mining methods, and complexity of the open-pit slope and underground mining environments, it is necessary to study the macro-micro response characteristics of the surrounding rock and overlying strata to the transition from open-pit to underground mining. Therefore, taking the open-pit to underground mining project of the Jinning phosphate mine belonging to the Yunlin Group as an example, experiments on similar materials, numerical simulations, and theoretical analyses are used in the present study to investigate the mining response characteristics of the surrounding rock and overlying strata during such a transition.

Engineering Background
After more than 30 years of mining, part of the no. 2 pit in the Jinning phosphate mine belonging to the Yunnan Phosphate Chemical Group Co., Ltd. has formed a high and steep slope. Figure 1(a) is the second pithead of the Jinning phosphate mine. The mining area is low in the north and high in the south and inclined to the east and west. The highest point is 2320 m, and the general elevation is between +2200 and +2320 m. The average dip angle of the ore body is 36°. The thickness of the ore body is generally 3-15 m, with an average thickness of 6.8 m, and the firmness coefficient of the deposit is 7-9. The roof is composed of dolomite with argillaceous rock (firmness coefficient: 6-10), and the floor is composed of argillaceous dolomite with a thin layer of chert (firmness coefficient: [14][15][16]. There are few weak interlayers in the rock layer of the open-pit slope; thus, the influence of faults and joints is small. At present, the total mining depth of pit 2 is over 120 m. The top slope angle is 45°, and the bottom slope angle is equal to the dip angle. Since only a small amount of geological reserves (above +2270 m) are available for openpit mining, the levels below this will be transferred to the underground mining stage. The representative section of exploration line 59 has been selected as the test section for the present study, and Figure 1(b) is the engineering geological section of exploration line 59. According to the actual mining process, one mining level per 50 m is selected for the study. The underground ore body is mined at a level of 10 m in each mining level.

Macro Mining Response Characteristics: Similar Material Experiments
3.1. Simulated Excavation. Experiment is the most traditional method in the field of geotechnical engineering [33][34][35][36][37][38][39][40][41][42][43][44][45][46][47][48]. In order to simulate the excavation from the open pit to underground, the two middle sections were selected; Figure 2 shows the two middle sections. In the actual mine, mining extended only to the +2120 m level, and the maximum surface elevation was +2320 m (see Figure 1(b)). In the present study, the simulated maximum mining depth and width are 200 and 300 m, respectively. The overlying strata in the monitoring scheme area is uniformly divided into sixteen displacement observation lines at horizontal intervals of 0.1 m, with the initial observation line positioned at +2270 m at a horizontal distance of 0.116 m from the slope. In the vertical direction, the lowest observation point in each line is 0.1 m from the ore body, and the height of successive observation points increased in steps of 0.1 m. In total, 128 deformation observation points are arranged in this experiment, Figure 2 is the simulation monitoring scheme.
Mining is performed via the sublevel caving method, Figure 3 shows the mining method and steps. The ore blocks are arranged along the strike and divided into two middle sections, each with a height of 0.5 m. Both of the two middle sections are mined in 5 steps from bottom to top, with subsection heights of 0.1 m. And the mining process is completed in 10 steps.

Model
Design. The ore body is chiefly phosphate rock, with a footwall consisting primarily of boundary phosphate rock, mud-bearing dolomite, and layered dolomite. The upper section consists mainly of mudstone-bearing quartz sandstone, dolomite, and a quaternary clay layer. Table 1 shows the rock mechanical parameters measured by laboratory physical tests after field sampling, which reflects the mechanical properties of rock, and the similar physical model experiment material ratio is also based on the mechanical parameters.
Based on the principle of similar material simulation, the similarity parameters for each simulation test are as follows: geometry similarity ratio = 1 : 100, bulk density similarity ratio = 0:81 : 1, stress similarity ratio = 1 : 123:50, and time similarity ratio = 1 : 10. A plane stress model was used in the present work to simulate similar materials along the dip of the phosphate rock. The main components of the model materials were sand, gypsum, calcium carbonate, mica powder, soft glue, engine oil, fine wood chips, and water. The appropriate material ratios were selected via the orthogonal test method and the use of a material testing machine, and the dimensions of the plane model were 3:00 × 0:30 × 2:00 m (length × width × height).  3 Geofluids structure occurs during the final excavation stages (steps 9-10, see Figure 4(f)).
In detail, an examination of Figure 4(a) indicates that only 10 observation points on lines 1-3 in the vicinity of the mined-out area are deformed, with a maximum subsidence of 5 mm in the central area of the stope. After steps 1-3 (see Figure 4(b)), small degrees of separation and caving have occurred in the rock mass, while it still remains intact and stable. As the mining process is continued (see Figure 4(c)), the underground mining space increases. The effects extend to line 5 on the right-hand side of the underground stope, and the disturbance becomes more serious. At this time, the maximum subsidence increases to 20.5 mm, and the deformation of the surrounding rock and overlying strata of the stope becomes intense in the area of stopes 2 and 3. The maximum vertical displacement has increased to 25.5 mm. After the eighth excavation step (see Figure 4(e)), the affected area extends to observation line 6, and the severe deformation is mainly distributed along observation lines 1-2 on the upper left and lines 3-4 on the upper right. At this time, the maximum vertical subsidence has increased to 31.5 mm. After the tenth excavation step (see Figure 4(f)), the mined area of the two middle sections is connected, the old mined area is "activated," and large numbers of macro through-cracks and large-scale pull-through microcracks are produced. At this time, the area affected by mining activity is seen to have increased sharply to 130 cm, the maximum displacement has increased to 33.2 mm, and the surface of the local area has collapsed.
In total, a gently inclined and thin-to medium-thick phosphate deposit was shown to transfer from the open pit to underground by the sublevel caving method, and the corresponding global subsidence curve (see Figure 4, the measure line 10 m and 20 m away from the stope) evolves from the irregular shape of a ladle to the final shape of a halfbowl. The results of the similar material model indicate that,   4 Geofluids near the mined-out area, the surrounding rock and overlying strata of a gently inclined, thin-to medium-thick phosphate deposit begin to deform and break after 1-10 excavation steps. During this process, the rock will bend and sink under the influence of the gravity and mining stress, to finally collapse when the internal stress exceeds the limiting strength of the rock stratum. The failure modes of sinking, bending, and breaking are identical for the upper and lower strata, and failure of the surrounding and overlying rock gradually progresses from the bottom to the top.

Model Geometry and Boundary
Conditions. In recent years, computer technology has developed rapidly, and numerical simulation has become an important research method . Based on the specific geological conditions and mechanical parameters of the deep inclined mediumthick phosphate ore body in the northern area of pit 2 (Jinning County phosphate mine), a particle flow code (PFC) numerical model was run with the same size as that of the similar material model, i.e., height × width = 300 cm × 200 cm; Figure 5(a) shows the particle flow code (PFC) model diagram, and Figure 5(b) shows the initial state of the model.

Physical and Mechanical
Parameters of the Rock. The mesomechanical parameters of the PFC model were based on the measured mechanical properties of rock samples obtained on site. They account for the size effect and similarity ratio via repeated adjustments to make the simulated settlement close to that of the similar material model. Figure 6(a) shows the comparative picture of the experiment and simulation, Figure 6(b) shows the subsidence curve of the experiment and simulation, and Figure 6 shows the total comparative results of the similar material model and the PFC numerical model, the results illustrated the effectiveness of this simulation method, and this method can be used to analyze the mechanical behavior of surrounding rock and overlying strata during the mining process. Table 2 shows the final PFC mesoparameters of the rock samples, and the PFC calculated results are consistent with the similar material model experiment.

Model Measuring Circle Layout.
According to the mine pressure theory, after the underground mining, the surrounding rock and overlying strata are affected by mining stress and mined-out areas, and the in situ stress will redistribute. Figure 7 shows the in situ stress distribution after excavation step 1, step 2, step 5, and step 10, which illustrate the pressure increase and pressure relief zone of the surrounding rock and overlying strata near the goaf.
To analyze the stress changes in the surrounding rock and overlying strata at various locations after the first and second excavation steps, measuring circles were laid out in the surrounding rock, roof corner, and floor corner of the mined-out area, respectively.        7 Geofluids stabilization. The larger the mining space, the longer the fluctuation time. In detail, the curves in Figure 9(a) reveal that the internal stress in the surrounding rocks at the upper right-hand corner of the roof (measurement circle 1, Figure 8(a)) and the lower left-hand corner of the floor (measurement circle 4, Figure 8(a)) decreases after the first excavation step, thus indicating a process of unloading. After the second excavation step (Figure 8(b)), the internal stress in the surrounding rock at the upper right-hand corner of the roof (measurement circle 11, Figure 8(b)) and the bottom left-hand corner (measurement circle 41, Figure 8(b)) continues to decrease, but the situation is no longer identical at the two locations. Thus, the amplitude of unloading of measurement circle 11 is 8 kPa, while the circle 41 is about 3 kPa. The final absolute stress value of measurement circle 11 after the second excavation step is about 0 kPa. This indicates that     9 Geofluids the surrounding rocks at this location are now in a state of stress balance. However, as the excavation space increases during subsequent excavation steps, the unloading area will continue to increase. Hence, due to gravity, this part of the surrounding rock could be placed under tension accompanied by internal bonds breaking and cracks appearing. Figures 10(a) and 10(b) show the stress evolution in the overlying strata on the right-hand side of the roof and the left-hand side of the floor after the first excavation step (measurement circles 2 and 5, respectively) and after the second excavation step (measurement circles 21 and 51, respectively). An examination of Figure 10(a) indicates that the internal stress of the surrounding rocks on the left and right sides is increased at the end of the first excavation step. In addition, the increase in stress on the left side of the floor (measurement circle 5, Figure 8(a)) is more pronounced with a maximum amplitude of~3 kPa, whereas that on the right (measurement circle 2, Figure 8(a)) increases slightly. As the mining space is increased during the second excavation step (Figure 10(b)), the pressure on measurement circle 21 continues to increase, whereas that on measurement circle 51 is significantly decreased, because the surrounding rock on the left side of the floor is continuously unloaded during the second excavation step. The absolute stress on the surrounding rocks on the right and left sides at the end of the second excavation step is about 8 kPa and 10 kPa, respectively, thus indicating that the rock mass was not damaged or moved. The above analysis indicates that the internal stress in the surrounding rocks on the left and right sides changes in a complicated manner as the excavation space is    10 Geofluids increased. Hence, the scope of the mining space is a key factor affecting the increase in internal stress or the relief of the surrounding rocks. Figures 11(a) and 11(b) show the stress evolution in the overlying strata at the upper left-hand corner of the roof and the lower right-hand corner of the floor after the first excavation step (measurement circles 3 and 6, respectively) and after the second excavation step (measurement circles 31 and 61, respectively). The results in Figure 11(a) show that the internal stress in the surrounding rock at the upper lefthand corner of the roof (measurement circle 3, Figure 8(a)) and the lower right-hand corner of the floor (measurement circle 6, Figure 8(a)) increases during the first excavation step, thus indicating a concentration of stress at the two locations. During the second excavation step (Figure 11(b)), the internal stress in the surrounding rock at measurement circle 31 and circle 61 continues to increase with implementation of the excavation space. However, the increase in pressure is   11 Geofluids more marked on the lower right, reaching about 6.5 kPa. In Figure 9(b), the unloading amplitude of the surrounding rock at the upper right-hand corner of the roof is greater during the second excavation. Hence, due to gravity, the roof has a tendency to move downwards and the stress concentrated in the upper left-hand corner. By contrast, the stress on the surrounding rock on the right-hand side is alleviated by the unloading of the overlying rock in the roof, thus limiting the increase in the internal stress at the lower right-hand corner of the floor. Figures 12(a) and 12(b) show the stress evolution in the overlying strata at vertical intervals of 0.09 m on the upper right-hand corner of the roof after the first excavation step (measurement circles 1, 7, and 8) and after the second excavation step (measurement circles 11, 71, and 81). During the first excavation step, the results in Figure 12   that the decrease in the internal stress on the overlying strata varies according to the vertical depth. Specifically, the degree of stress reduction deceases with increasing depth so that the original rock level of stress maintains at a certain depth. After the second excavation step (Figure 12(b)), the pattern of pressure relief with depth is like that at the end of the first excavation step. It is evident that the increase in the minedout area results in a longer fluctuation time for the internal stress in the overlying strata; i.e., the larger the mining space, the longer the overlying rock stress stabilizes.
In brief, the stress evolution curves of overlying strata after various excavation steps (see  demonstrate that a complicated pattern of internal stress distribution occurs in the adjacent overlying strata during the process of converting from open-pit to underground mining. Thus, both pressure-increasing and pressure-relieving zones exist and are exactly opposite for the roof and the floor. Moreover, the pressure-increasing and pressurereleasing zones in the roof are more severe than those at the floor. In addition, the surrounding rocks on each side form opposite pressure-increasing and pressure-relieving zones during the mining process, with that on the right side being more obvious. Thus, there is an important relationship between the size of the mining space and the pattern of pressure-increasing and pressure-relieving zones. The degree of stress reduction becomes weaken as the distance from the mined-out area increased, and, conversely, the pressure relief becomes more complete closer to the mined-out area, with the stress release being gradually completed towards the deeper parts.

Crack Propagation Behavior of the Surrounding Rock and Overlying Strata after Each Excavation
Step. To investigate the mesomechanical characteristics of the surrounding rock and adjacent overlying strata after stepwise excavation, mesoscale distribution of contact forces between particles of binding force F and patterns of crack propagation around the mined-out area are diagrammed in Figures 13-15. Figures 13(a)-13(c) show the mesoscale force field distribution and crack propagation during the first, second, and third excavation steps, respectively. During step 1, Figure 13(a) reveals the presence of concentrated areas of adhesion at the left-hand corner of the roof and the righthand of the floor, whereas areas of weak adhesion (i.e., pressure-relieving areas) are indicated at the other two corners of the stope. As the excavation space increases during step 2 (see Figure 13(b)), the increase in pressure on the surrounding rocks becomes more marked. At the same time, the pressure-relieving areas become larger and the pressure relief is greater. As mining continues into step 3 (see Figure 13(c)), the internal tensile stress eventually exceeds the bonding tensile strength. The bonds break and microcracks appear in the pressure-relieving area. This result is consistent with the analyses in Figures 8-11 (physical modeling). Note that a few microcracks also appear in the surrounding rock on the bottom right-hand side of the mined-out area due to gravity. Since the mining space is small, there are no large-scale internal cracks, and the surrounding rock and adjacent overlying rocks remain stable after pressure relief. 13 Geofluids sixth excavation steps, respectively. With further progress in mining and further increase in the mined-out area during steps 4-6, the cracks expand more noticeably (see Figure 14). During steps 4 and 5, the cracks become arcshaped and tend to penetrate and connect the upper and lower ends of the stope (labelled 1 in Figure 14). As the surrounding rock on the right ruptures and falls, the area of concentrated cohesive force at the bottom of the surrounding rock moves towards right (area 2 in Figure 14). As the roof of the excavation is close to the ground surface, unloading during mining causes the cracks in the roof, which firstly extend to the ground surface (the area labelled 3 in Figure 14(e)). After the lower part of the surrounding rock on the right-hand side falls and collapses, the cracks gradually expand upwards (area 4). When the next mining level is excavated during step 6 (see Figure 14(f)), the cracks between the upper and lower mining levels penetrated (area 5). Due to the greater depth of the next mining level, the original rock stress is correspondingly greater and, hence, more cracks form at the bottom of the surrounding rock on the right-hand side. Thus, it leads to faster destruction of the rock mass (area 6).

Figures 15(a)-15(c)
show the mesoscale force field distribution and crack propagation during the seventh, eighth, and tenth excavation steps, respectively. During the process of mining the next level, the crack propagation behavior is roughly the same as that under the previous mining level; i.e., the cracks extend in arc shapes and penetrate the upper and lower ends of the stope (area 2 in Figure 15). Due to the increase in mining depth, the crack propagation rate is higher than that of the previous mining level such that the fissures around the two mining levels penetrate each other (area 1, see Figure 15) and subsequently propagate to the surface (areas 3 and 4, see Figure 15). At the end of the final excavation step, the relatively independent mined-out area has disappeared (see Figure 15(c)), causing the number of cracks to increase rapidly and propagate to the surface, thus resulting in overall instability of the mined-out area.
In brief, the crack propagation behavior is characterized by expansion in arc shapes and a tendency to penetrate the upper and lower ends of the stope. Moreover, the size of the excavation space is seen to play a key role in the generation, propagation, and penetration of the cracks. Thus, due to the disturbance of the first mining level and the increase in  14 Geofluids excavation depth, the damage rate of the surrounding rock and overlying rock in the second mining level is greater and more cracks are generated, thus accelerating the instability of the surrounding rock and overlying strata.

Conclusions
The open-pit end slope, the surrounding rock of the pit bottom, and the underground mine constitute a compound mining system after the transfer from open-pit mining to underground mining. During the period of open-pit mining, the rock around the end of slope and the pit bottom is disturbed and, on this basis, the underground excavation leads to secondary disturbance. With the advancement of the excavation space, the dynamic superposition of this secondary disturbance influences the state of the underground surrounding rock and the overlying strata. The present study has used similar material models and numerical simulation methods to analyze the settlement curves, stress changes, and crack propagation behavior of the surrounding rock and overlying strata during the process of converting from open-pit to underground mining. The fit of the settlement curve from the model experiment with the numerical calculation illustrated the effectiveness of this simulation method enabling an analysis of the internal stress changes and crack propagation behavior of the surrounding rock and overlying strata during the mining process, and the results can be concluded as follows: (1) When a gently inclined thin-to medium-thick phosphate deposit was transferred from open-pit to underground mining via a sublevel caving method, the surrounding rock and overlying strata gradually evolved from a stable state through a continuous linear failure state to a final large-scale nonlinear collapse state. The corresponding global subsidence curve was shown to evolve from an irregular ladle shape to a final half-bowl shape (2) The internal stress distribution of the surrounding rock and adjacent overlying rock in the inclined mined-out area during the process of converting from open-pit to underground mining was shown to be complicated. The degrees of pressure increase and pressure relief on the internal stress of the surrounding rock and the adjacent overlying strata in the mined-out area were shown to have an important relationship with the size of the mining space. Specifically, the degree of stress reduction was weakened with increasing distance, and, conversely, the relief of pressure and the release of stress became gradually more complete towards the deeper part closer to the mined-out area (3) The cracks were shown to expand in an arc shape and had a tendency to penetrate the upper and lower ends of the stope. The size of the excavation space was found to play a key role in the generation, propagation, and penetration of the cracks. Due to the disturbance of the first mining level and the increase in excavation depth, the rate of damage to the surrounding and overlying rock in the second mining level was greater and more cracks were generated, thus accelerating the instability of the surrounding rock and overlying strata (4) Many factors were shown to influence the evolution and dynamic characteristics of rock mass deformation and failure after the transition from open-pit to underground mining. It is therefore necessary to study the time-effect and space-time evolution of the mining response characteristics of the rock mass and the effects of various underground mining methods

Data Availability
Data are available from the authors upon reasonable request.

Conflicts of Interest
The authors declare that they have no conflict of interest.