International Journal of Mining Science and Technology 30 (2020) 63–67

Contents lists available at ScienceDirect

International Journal of Mining Science and Technology

Development of a multiple level underground limestone mine

a r t i c l e i n f o a b s t r a c t

Article history: The development of a multiple level underground mine is a complex task in which geology, engineering,
Received 9 April 2019 ground control, and unit operations are integrated into a single mine design. The components are inter-
Received in revised form 16 September dependent and must function cohesively to ensure a profitable underground mining operation. To opti-
2019
mize reserve recovery, mine planning should begin from the lowest level and progress up. This limits any
Accepted 12 October 2019
Available online 10 February 2020
misjudgments or oversights of a given level affecting the underlying levels and ensures the ability to
maximize recovery from each level. Mine planning should start with the exploration and characterization
of the geologic setting. Once the reserve geology and quality is well understood, then mine planning can
Keywords:
Limestone
begin with respect to the following: (1) orientation of mine works with respect to horizontal stress; (2)
Mine planning access to the reserve; (3) determination of opening widths; (4) selection of back, floor horizons and pillar
Geology centers; (5) selection of development and secondary mining heights; (6) appropriate inter-burden thick-
Multiple level nesses; and (7) examining the stability of the multiple level mine through numerical modeling. The mul-
Rock mechanics tiple level mine design process and decisions are presented through a case history example. The theme is
that there is one opportunity to ‘‘get it right” and many chances to overlook a small aspect within the
design that will plague the mine throughout all levels and through the entirety of its operating life.
Ó 2020 Published by Elsevier B.V. on behalf of China University of Mining & Technology. This is an open
access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction mine plan will increase the overall recovery of the reserve, greater
overburden depths, multiple level geometry, and the presence of
As surface quarries reach their economic stripping limit, and discontinuities all require an emphasis on ground control for the
encroach on mineral and/or reserve boundaries, they approach planning and development of a safe and productive multiple level
the end of surface mineable reserves. Furthermore, modern urban underground stone operation.
sprawl has placed previously remote surface quarries within sub- Modern underground stone mining operations are commonly
urban development and neighborhoods leading to significant designed with respect to the chemical and physical rock properties
increases in noise, dust, and vibration grievances by the commu- which sell for a premium. While this approach allows for immedi-
nity and regulators [1]. Many mining operations have begun ate and short-term gains, it can significantly limit the production of
year-round production of underground reserves to meet growing future levels and isolate large portions of the underground reserve.
market demands. This does not permit the typical sequence of In designing a multiple level underground stone mining operation,
stripping overburden in the winter to provide limestone to mine the stability of the main mine back will determine the geometry of
during the construction season. Underground stone mining opera- the current and subsequent levels as well as the size of under-
tions in the United States commonly utilize a room-and-pillar mine ground mining equipment and haulage. The main back should be
layout with drill-and-blast production methodology for the devel- defined such that access to the mining reserve and subsequent
opment of the reserve. Within the Midwest and Eastern regions, levels is optimized. Once a stable main back has been defined
thick and flat-lying limestone formations lend themselves to the within the lithology of a given reserve, underground pillars can
utilization of multiple level mining layouts. While a multiple level be appropriately sized to support both overburden load and multi-
ple level stress interactions while remaining stable during drill-
and-blast operations. In obtaining pillar dimensions which ade-
⇑ Corresponding author. quately support applied pillar loading, ground analyses of the
E-mail address: [email protected] (C. Newman). immediate and main floor will determine the impact of pillar

https://doi.org/10.1016/j.ijmst.2019.12.007
2095-2686/Ó 2020 Published by Elsevier B.V. on behalf of China University of Mining & Technology.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
64 C. Newman et al. / International Journal of Mining Science and Technology 30 (2020) 63–67

punching and/or floor heave on pillar stability of the current as where t is the roof beam thickness; re the allowable tensile stress;
well as adjacent levels. and w the uniform load per unit length per unit width of the beam.
As stone mining operations continue to develop at deeper Academic investigations and industry experience indicate that
depths and in more geometric and geologically complex mining back conditions are directly related to the composition and thick-
scenarios, the integration of and focus on ground control within ness of the immediate roof beam. Therefore, the allowable tensile
the planning process provides mining operations with more effi- stress (re) of the roof beam should be defined with respect to the
cient methods of reserve recovery while maintaining stable ground average splitting tensile strength (ASTM D3967) conducted on
conditions through a brief literature review and a case history. 0.3 to 1.5 m increments of the immediate roof.
For multiple level mining conditions, the immediate roof beam
2. Case history background determines local ground conditions while the stability of the inter-
burden between adjacent levels determines the viability of the
Residential development and city expansion limited the expan- mining layout. The interburden between adjacent levels should
sion of the surface quarry at Mine A. Three exploratory holes were be defined such that multiple level stress interactions are limited.
drilled for the purpose of determining the quality and location of Commonly, the interburden thickness between adjacent levels
the Hermitage, Carters, Lebanon, Ridley, and Murfreesboro lime- should be defined with respect to the total excavation height of
stone formations. Initially, three back horizons have been identi- the overlying level. The stability of the interburden is a function
fied within the Hermitage and the Carter formations. Level 1 and of the lithology strengths and stratum thicknesses. The compe-
Level 2 are accessed through the existing surface quarry pit. The tency of the interburden should be determined with respect to
proposed Level 3 will be accessed by a future decline from Level the unconfined compressive strength (ASTM D7012) of the main
2. A 12 m header width has been defined for all levels. Secondary strata as well as geological observations within the core-hole as
recovery was considered for Level 1 and Level 2 with a total exca- well as underground. Due to the complexity of stress distributions
vation height of 18 m. Level 3 was limited to a total excavation between adjacent levels, the stability of the interburden should be
height of 12 m due to multiple seam stress influences. Analytical evaluated with respect to numerical modeling methodologies
and numerical analyses of Mine A will be used to support multiple (Fig. 1). Numerical modeling was conducted using RS3, developed
level design methodologies outlined throughout this paper for the and distributed by RocScience, assuming non-elastic rock materials
determination of appropriate back locations, interburden thick- and no discontinuities.
ness, and pillar dimensions. At Mine A, three levels have been identified for room-and-pillar
development with secondary floor recovery. The interburden
3. Stability of the immediate and main back thickness between Level 1 and Level 2 was determined with
respect to the total mining height of Level 1 (18.28 m) as well as
In determining the back horizon of a given level within a mul- core–hole observations. A thin shale parting was proposed as the
tiple level mine layout, mine planning considers product quality back of Level 2 resulting in an interburden thickness of 19 m. The
(chemical and/or physical), geologic features, and rock mechanics. Level 2–Level 3 interburden thickness was defined with respect
The selection of an appropriate and stable main back sets the to rock mechanics testing results (Table 1). Given the strength of
precedence for roof stability on the current level as well as ground the rock within the Level 2–Level 3 interburden (approximately
conditions on adjacent levels. A detailed summary on the selection 172.37 MPa), the interburden thickness was reduced from an ini-
of a main back with respect to site-specific geology and conditions tial thickness of 18 0.28–12.5 m. Strength and material properties
is provided by Newman and King [2]. Back elevation(s) should be defined within the numerical model (Fig. 1) were determined with
selected with respect to prominent horizontal geological features, respect to laboratory results for the unconfined compressive
such as thin shale partings or stylolite. These geologic discontinu- strength and modulus of the rock sample. The average intact com-
ities within the mining reserve provide a natural plane along which pressive strength of the rock, as implemented within the RocLab
the rock can ‘‘peel” away after a blast leaving a smooth back stra- program, was determined with respect to unconfined compressive
tum and limiting back break. While this provides an athletically strength test results. From the RocLab calculated intact compres-
pleasing back condition for the current level, the back stability sive strength of the rock, user defined geological strength index
within multiple level mining conditions is ultimately a function and intact rock parameter were estimated in regard to core condi-
of the competence of the roof beam. tion and behavior during sample documentation, preparation, and
Based on 34 single level site investigations conducted by NIOSH, testing in the determination of the Hoek-Brown failure criterion.
it was observed that mining operations maintaining a competent Numerical results indicate no significant stress communications
roof beam thickness of greater than 2 m did not require the instal- between adjacent levels.
lation of patterned roof support. Within the database, mining oper-
ations which maintained a roof beam thickness of <1.2 m had 3.1. Impact of discontinuities on back stability
unstable back conditions requiring the installation of additional
ground support [3]. As a ‘‘rule-of-thumb”, a stable roof requires a While localized roof instability has been correlated to the com-
massive, homogenous beam where the thickness is 10% of the petence and thickness of the roof beam, steeply dipping disconti-
span. As the roof span exceeds the span determined by the roof nuities can significantly impact the continuity of the roof beam
beam stability equation (Eq. (1)), roof deflections can become resulting in roof failures. The presence of faults, joints, calcite
excessive and may result in the development of tensile fracturing veins, clay veins, etc. provides a natural slip plane within the
along the pillar-back interface as well as the center of the header reserve. Therefore, discontinuities should be documented and
or crosscut [4]. Therefore, it is suggested that the header and mapped as they are encountered by core drilling, highwall
crosscut width are determined with respect to the stability of the observations and/or pit development, and underground pillar
roof beam span. Here the unsupported roof span (L) is calculated development. When discontinuities are observed underground,
as; the site-specific roof beam stability should be reevaluated. If
discontinuities within the immediate back are not appropriately
sffiffiffiffiffiffiffiffiffiffiffiffi supported with the installation of ground support or localized
4r e t 2 modifications to the mining layout, failure of the unsupported roof
L¼ ð1Þ
3w beam is likely and can significantly impact the underground
 C. Newman et al. / International Journal of Mining Science and Technology 30 (2020) 63–67 65

Fig. 1. Strength factor results from numerical analysis of Mine A multiple level layout.

Table 1
Summary of rock strength and material properties.

Level ID Limestone formation Sample location Laboratory testing result Hoek-Brown criterion
UCS Elastic modulus mb s a
(MPa) (MPa)
Level 1 Hermitage Pillar 114.24 9686.94 1.2623 0.002218 0.508086
Carter Interburden A 194.13 11611.34 1.2623 0.002218 0.508086
Level 2 Carter Pillar 170.07 22914.06 1.2623 0.002218 0.508086
Carter Interburden B 172.38 13430.94 1.2623 0.002218 0.508086
Level 3 Carter Pillar 80.26 11611.34 1.2623 0.002218 0.508086

working environment as well as production. Therefore, rehabilita- mine projections indicate that the projected fault will only be
tion efforts were taken by mine management to support the crossed as needed for haulage and ventilation. Future crossing of
exposed fault area with tension roof supports installed on pattern. the fault on Level 2 will be oriented perpendicular to the fault with
This area of the mining reserve has been abandoned. reduced header widths.
A large fault was observed during the development mining in
Level 1 and correlated to changes in the elevation of the immediate
roof stratum within the core holes and the highwall. Initially man- 4. Pillar dimensioning and layout
agement restricted the exposure of the fault by orienting produc-
tion headers near-perpendicular to the fault and reducing the With the appropriate selection of a back horizon and header
header width from 12.2 m as shown along 6 W between 4 N and width, pillar centers can be determined with respect to the pillar
3 N in Fig. 2. As development of Level 1 progressed, the unsup- strength and pillar loading. The room-and-pillar mining method
ported room beam located at the initial fault crossings remained with secondary floor recovery utilized within the stone mining
stable. Mining production decisions resulted in the exposure of industry results in slender pillar geometry. As the limestone pillar
the fault across a significantly larger span of 22.8 m at the intersec- is loaded, the brittle nature of the rock lends itself to a significant
tion of 7 W and 4 N. Although the roof remained stable following strength reduction post-yield [5]. The behavior of the rock pillar
excavation of the area, a significant roof fall event occurred at coupled with slender geometry results in pillar spalling, referred
the intersection of 7 W and 4 N that directly impacted under- to as ‘‘hour-glassing”. Therefore, in addition to supporting the over-
ground stone production and access to the remainder of the Level lying strata, limestone pillar designs should be evaluated with
1 reserves. respect to local rib stability.
Strike and dip observations of the fault made on Level 1 were The S-Pillar program, developed and distributed by NIOSH, pro-
projected onto Level 2 and Level 3 mine projections. While the vides an empirical approach to stone pillar design that has been
fault has not been encountered yet during Level 2 development, validated with respect to field observations and numerical model-
ing of single level mining operations. From a NIOSH pillar perfor-
mance survey of 34 single level underground stone mining
operations in the Eastern and Midwestern United States, under-
ground observations as well as analytical and numerical analyses,
a pillar strength (S) equation for stone pillars was developed [6];

w0:3
S ¼ 0:65  UCS  LDF  0:59
ð2Þ
h
where UCS is the uniaxial compressive strength of the intact
rock; LDF the large discontinuity factor; w the pillar width; and h
the pillar height.It should be noted that industry observations
and academic research suggest that rectangular pillars do not pro-
vide any significant pillar strength benefit when the pillar width-
to-height ratio is greater than 0.8 [7]. A width-to-height factor of
0.8 is the minimum pillar dimensions as recommended by S-
pillar. However, rectangular pillars are commonly utilized to limit
the exposed roof in the direction of maximum horizontal stress as
Fig. 2. Plan view of fault location at the intersection of 7 W and 4 N. well as providing a means of increasing underground ventilation
66 C. Newman et al. / International Journal of Mining Science and Technology 30 (2020) 63–67

efficiency [8]. The stability of the rock pillar is determined as the While thorough geological investigations of the mining reserve
quotient of pillar strength by pillar loading as defined by the Tribu- should identify the presences of joints, faults, and major disconti-
tary Area Theory. Analysis of the S-pillar empirical database indi- nuities, the majority of discontinuities observed within the stone
cates a minimum pillar stability factor of 1.8 and a minimum pillar are localized discontinuities. These discontinuities are com-
width-to-height ratio of 0.8 (Fig. 3). monly exposed during development mining or during secondary
Although S-pillar does not consider multiple level stress distri- recovery of the floor when pillar loading has increased. Under-
butions, the design methodology implemented can be applied to ground observations and numerical investigations indicate that
multiple level mining through the consideration of additional pillar steeply dipping discontinuities (30–60°) within the pillar will
loading with respect to the multiple level stress interactions impact the overall pillar strength. Within the S-pillar empirical
(Fig. 3). Given the geometric and geologic complexities commonly database, large angular discontinuities were present in 65% of the
observed in underground stone mining operations, it is recom- site-observations [6]. To mitigate the impact of large angular dis-
mended that pillar dimensions be finalized with respect to numer- continuities on pillar stability, it is recommended that the pillar
ical modeling. At Mine A, Level 1 pillars were developed on 27.4 m layout be modified such that angular discontinuities within the
by 27.4 m centers, 12.2 m header and crosscut widths, develop- pillar propagate into the floor and/or roof and are not exposed
ment height of 7.62 m and total excavation height after benching within the excavated entry by restricting movement along the
of 18.28 m. To minimize multiple level stress interactions, Level discontinuity-rock interface.
2 and Level 3 pillars were columnized with respect to those previ- Similarly, the presence of near-horizontal discontinuities such
ously developed on Level 1. While Level 2 and Level 3 maintain as thin shale partings, calcite veins, clay seams, etc. can contribute
similar pillar spacing (center-to-center), the final excavation height to pillar failure. Numerical investigations and underground obser-
for Level 2 is 18.28 m while the final excavation height for Level 3 vations indicate failure within the pillar can be caused by tensile
is 12.2 m. By columnizing pillars on adjacent levels, vertical load- failure attributed to the presence of a weak bedding plane [9]. As
ing is transferred from the pillar core of the upper level to the pillar vertical loading of the pillar increases, the weaker material begins
core of the lower level as previously illustrated in Fig. 1. to fail and is extruded from the stone pillar. As vertical loading con-
Numerical modeling results indicate that pillars on Levels 1 and 2 tinues to increase, tensile stress develops within the pillar further
show the potential for ‘hour-glassing’ and pillar deterioration when contributing to progressive spalling.
the floor is taken. The safety factors at mid-pillar are lower than
those along the back and floor line where the confinement is higher. 5. Stability of the immediate floor
In an attempt to increase rock production, an additional floor bench
of 6.1 m was considered for Level 3. As shown in Fig. 4, the develop- The stability of the immediate floor material has generally been
ment of slender pillars with reduced width-to-height ratios on Level in mine planning and design. Given the high uniaxial compressive
3 results in significant pillar slough age and pillar instability. There- strength of the limestone, pillar punching due to vertical loading
fore, the final excavation height of 12.2 m was determined to be and the resulting floor heave are not common ground control
appropriate for Level 3 pillars (Fig. 1). Currently, mining production issues for multiple level stone mining operations. While floor sta-
in Level 1 has ceased with all development mining occurring in Level bility issues were not observed within the S-pillar empirical data-
2. Underground observations of benched pillars in Level 1 indicate base, field observations have indicated unfavorable back, pillar,
that the pillar support system is performing as designed. and floor conditions in the presence of a weak immediate floor
[10]. Similar to the weak horizontal discontinuities within the pil-
4.1. Impact of discontinuities on pillar strength lar, as vertical loading of the pillar increases the weak floor mate-
rial begins to fail and is displaced resulting in the development of
As indicated by the LDF in Eq. (2), the presence of discontinu- tensile stress within the stone pillar contributing to pillar spalling.
ities within the pillar can significantly reduce the overall strength. Furthermore, displacement of the floor material due to vertical

Fig. 3. S-pillar pillar stability analysis for single level stone mine.
 C. Newman et al. / International Journal of Mining Science and Technology 30 (2020) 63–67 67

Fig. 4. Reduction in Level 3 pillar stability due to increased benching height.

loading of the stone pillar results in floor heaving. It should be observations, benched pillars are performing as designed in Level
noted that the stability of the immediate floor is a function of mul- 1. Further observations will be made once benching begins on
tiple geometric and geo-mechanical indices. In the Eastern and Level 2 and development mining in Level 3. This paper discussed
Midwestern United States, floor instability is commonly a result the impact of pillar and back stability as well as discontinuity with
of horizontal stress magnitude and orientation. respect to multiple level mine designs. However, further ground
related issues such as horizontal stress, floor benching, steeply dip-
6. Summary and conclusions ping seams, and water have not been discussed but should all be
considered for mine planning and design purposes [11–13]. The
As more surface quarries begin the development of under- utilization of ground control techniques within mine planning
ground mining complexes, multiple level limestone mining opera- and design will provide mining operations with the means of
tions will allow for production in reserves previously unattainable increasing reserve recovery while maintaining safe and productive
from the surface. However, due to the increased overburden, geo- underground working environments.
metric and geological conditions, the presence of discontinuities,
and multiple level stress interactions, multiple level mine design References
for stone mining operations requires significant emphasis on
[1] Newman D. Roof control, pillar stability and ground control issues in
ground control. The geometric and geo-mechanical complexities underground stone mines. Min Eng 2017;69(8):53–8.
present in modern multiple level mine planning requires an itera- [2] Newman D, King B. The development and planning of multiple level
tive approach to mine design which considers the stability of the underground limestone mines. 35th international conference on ground
control in mining. Morgantown, WV; 2016.
back, pillar, floor, and interburden. Once a multiple level mine [3] Esterhuizen G, Dolinar D, Ellenberger J, Prosser L, Iannacchione A. Roof stability
design has been implemented, it is imperative that the mining issues in underground limestone mines in the United States. In: 36th
operations continually monitor roof, pillar, and floor stabilities international conference on ground control in mining. Morgantown, WV;
2007.
through daily visual inspection and/or rock instrumentation.
[4] Peng S. Coal mine ground control. New York: John Wiley and Sons; 1978. p.
The planning of a multiple level stone mining operations begins 450.
with a thorough understanding of the site-specific geology. The ini- [5] Esterhuizen G. An evaluation of the strength of slender pillars. Transactions
tial focus of a multiple level mine plan should be the location of society for mining, metallurgy, and exploration, vol. 320. Littleton, CO; 2006. p.
69-76.
back horizons for each level with respect to site-specific geology [6] Esterhuizen G, Dolinar D, Ellenberger J, Prosser L. Pillar and rood span design
and rock mechanics. Following back selection, each level should guidelines for underground stone mines. NIOSH Information Circular 9526;
maintain a competent and strong immediate roof beam with a 2011.
[7] Dolinar D, Esterhuizen G. Evaluation of the effect of length on the strength of
minimum thickness equal to 10% of the header width. Pillars slender pillars in limestone mines using numerical modeling. In: 26th
within a given level should be appropriately sized to adequately international conference on ground control in mining. Morgantown, WV;
support the overburden load as well as the additional multiple 2007.
[8] Iannacchione A, Marshall T, Burke L, Melville R, Litsenberger J. Safer mine
seam stress. Pillars on adjacent levels should be columnized with layouts for underground stone mines subject to excessive levels of horizontal
an initial interburden thickness equal to the excavation height of stress. Min Eng 2003;55(4):25–31.
the overlying seam. Discontinuities within the roof and/or pillar [9] Esterhuizen G, Ellenberger J. Effects of weak bands on pillar stability in stone
mines: field observations and numerical assessment. In: 26th international
should be evaluated with respect to site-specific conditions and conference on ground control in mining. Morgantown, WV; 2007.
loading. Utilizing numerical modeling methods, the stability of [10] Murphy M, Ellenberger J, Eserhuizen G, Miller T. Analysis of roof and pillar
the immediate floor should be evaluated with respect to pillar failure associated with weak floor at a limestone mine. Int J Mining Sci Technol
2016;26(3):471–6.
punching and floor heave.
[11] Newman D, Newman C. The effect and measurement of horizontal stress in an
While this paper outlined ground control considerations with Underground Limestone Mine. In: 38th international conference on ground
respect to multiple level mine planning and design, each mining control in mining. Morgantown, WV; 2019.
locations is unique and therefore may require further investiga- [12] Esterhuizen G, Dolinar D, Ellenberger J. Observations and evaluation of floor
benching effects on pillar stability in US limestone mines. Vancouver,
tions and analyses. The mine design methodology outlined within Canada: U.S. Rock Mechanics Symposium; 2007.
this paper has provided Mine A with a successful mine plan and [13] Sears M, Slaker B, Rashed G. Numerical model simulation of a development
rock production. While pillar slough age on Level 1 was predicted pillar at a steeply dipping underground limestone mine. In: 37th international
conference on ground control in mining. Morgantown, WV; 2018.
by the numerical model and confirmed through underground