A profile function to predict surface subsidence for down-cut slicing panels into inclined coal seam of the Barapukuria coal mine, Bangladesh

There are five discovered coal basins in the northwestern portion of the country. Most of the discovered coal basins in Bangladesh are characterized as thick to ultra-thick and multi-layered coal seams. The Barapukuria is the only coal-producing mine in the country among the all-discovered coal basins. In the Barapukuria coal basin, there are six coal seams, with coal seam VI being the ultra-thick seam, averaging 36 meters in thickness, and suitable for mining. The coal extraction process at the mine utilizes the longwall caving mining method.

In the longwall caving method, a continuous downward movement of the overburden strata is observed during and after the extraction of the resources from the ground. The downward movement of the subsurface strata developed a subsidence trough on the surface leading to various damages to the environment and physical structure of the society. A precise and validated subsidence prediction model is unavoidable to protect and mitigate the environment and structures from upcoming damages. The geometric shape of a subsidence trough varies based on the inclination of the coal seam. It (subsidence trough) is symmetric for a horizontal coal seam and asymmetric while the coal seam is inclined (Figure 1). The location of the peak subsidence point (Smax) resulting from inclined and deep coal seam mining moves towards the dipping side of the panel, as depicted in Figure 2 [1], [2].

A static subsidence is the final product of a subsidence trough that persists long after the completion of the mining operation while a dynamic or progressive subsidence forms during face movement. In longwall coal mining, creep subsidence typically occurs over a relatively short time frame (4 to 12 months) and amounts to 3% to 4% of maximum subsidence. This duration decreases even more with shallower depths [2]. In longwall coal extraction, subsidence is inevitable and continues for several months to several years after completing mining activities. Moreover, the thickness of the coal seam cover influences the surface subsidence duration. It is also observed that the thicker the coal seam cover, the longer the duration of surface subsidence, which is contingent upon individual characteristics such as the depth of excavation, the thickness of the extracted coal seam, and the behavior of surrounding rock masses.

Subsidence troughs can be characterized as subcritical, critical, and supercritical depending upon the width-to-depth ratio (W/H) of an opening. In the subcritical stage, while the opening width is less than 1.2x depth, subsidence is in dynamic conditions and does not reach its maximum possible subsidence (Smax). On the other hand, when the opening width is equal to its depth, the subsidence trough reaches its maximum value called the critical stage. Afterward, the maximum subsidence (Smax) does not increase on further development of the opening; rather, it spreads horizontally across an area. [1], [2].

A surface subsidence model is essential to study problems related to rock movement. An asymmetric subsidence trough is developed while mining an inclined seam, and the trough extends along the dip of rock occurrence. Moreover, the subsidence angle is greater on the inclined side of the mine working [3]. The Barapukuria coal basin is characterized by gentle to steep dipping seams. At the same time, the maximum dip is observed at the central part and the slope angle increases to the eastern side of the basin.

Maximum coal has been extracted from the northern district of the Barapukuria coal mine. 20 panels to date have developed and produced coal from the northern district of the mine. The coal is being extracted slices-wise from top to bottom (1st slice at the top and 4th slice at the bottom). Although the mine offers economic advantages, it also presents environmental concerns, notably in the aspect of land subsidence [4]-[7]. Many research works have been carried out on the Barapukuria underground coal mine. For instance, Islam [8] conducted research on remedial measures of subsidence and modeling for strength assessment of backfill materials to reduce surface subsidence. Islam & Shinjo [9]-[10] focused on mining-induced fault reactivation and stress distribution with a displacement surrounding the entry roadway for igneous intrusion as well as to study possible sources of gas emission from the seam, Islam et al. [11] analyzed the complications of the multi-slice longwall mining, Islam [12] performed another research on the orientation of maximum horizontal stress, Islam and Kamruzzaman [13] conducted a further study on geochemical and techno-environmental behavior of Gondwana coals, furthermore, Islam and Islam [14] pointed on the water inrush hazard of the coal mine. Islam and Hayashi [15] conducted an exploratory study when the subsidence began. However, the focus of the study was on coal bed methane rather than surface subsidence. Wide-ranging research work was performed by Howladar [16], [4] and Howladar and Hasan [17], focusing on the impact on water and the environment. Hossain et al. [18] analyzed the subsidence trend and annual rate of subsidence and subsurface condition.

It's worth mentioning that all these studies concentrated on assessing the effects of subsidence on water and the nearby environment, neglecting to address the geotechnical-geomechanical issues and the influence function that closely relates to the dynamism of subsidence. Apart from that the subsided lake at Barapukuria has been created for coal extraction from a series of multi-slice panels. This study targeted to fill this gap by analyzing the subsidence trough using the profile function parameters considering an inclined coal seam.

Different approaches are available for predicting subsidence troughs in both horizontal and inclined seams mining. These methods encompass theoretical and graphical techniques, physical and numerical modeling, as well as profiling and influence functions [2]. Subsidence prediction methodologies fall into categories such as empirical methods, physical models, and numerical methods [1], [22], [23].

The subsidence is a dynamic process depending on the extraction rate. Developing from the goaf the upward deformation movement reached the surface which created a subsidence trough. The curve of a subsidence trough is being considered by the subsidence prediction method called profile function. The coal seam VI of the Barapukuria coal mine dips to the middle-eastern side of the basin (Figure 3). Consequently, all the panels of the central district are dipping to the west-southern direction. The profile function coefficient of the Negin coal mine [25] is used in the calculation as the characteristics of the stratigraphic column over the panel of the Barapukuria coal mine are similar. A series of panels and slices perform coal extraction from the Barapukuria coal mine. A single subsidence trough and goaf is considered adding coal extraction thickness of all the individual slices and panels. In that case, there are two panels (1204 & 1104) at panel 1204; three panels (1306, 1206 & 1106) at panel 1306; three panels (1308, 1208 & 1108) at panel 1308; four panels (1310, 1210, 1210D & 1110) at panel 1310; four panels (1412, 1312, 1212 & 1112) at panel 1412; three panels (1314, 1214 & 1114) at panel 1314; and one panel at 1116. The series of panels have been developed and extracted coal from all those panels to meet the production target of the mine.

Various charts and graphs are developed for specific conditions and geometrical shapes in the empirical method. Sometimes these resources can quantify subsidence amounts. Among all, the National Coal Board (NCB) has proposed one of the most well-known graphs for subsidence prediction. For instance, Figure 7a-b illustrates a graph for predicting surface subsidence in horizontal panels.

Subsidence is a continuous process that started while starting coal extraction. To quantify the profile function, a series of profile functions must be considered. The study has considered two profile function curves for each panel position. The first one is along the west-east direction while using the width of the panels and the second one is along the north-south direction while using the length of the panels to develop the profile function curves. The newly developed profile function of the Barapukuria coal mine depends on all the mining and profile function parameters of the west-direction oriented panels listed in Table 1 & Table 2 and the profile functions are shown in Table 3.

Table 1. Mining operation parameters for W-E oriented central district panels of BCM.

Table 2. Profile function parameters for W-E oriented central district panels of BCM.

Table 3. Newly developed profile functions for west-east oriented panels of central districts of the Barapukuria coal mine.

At the same time, the mining operation and profile function parameters of the north-south oriented panels are listed in Table 4 & Table 5, and newly developed profile functions are shown in Table 6.

Table 4. Mining operation parameters for N-S oriented central district panels of BCM.

Table 5. Profile function parameters for N-S oriented central district panels of BCM.

Table 6. Newly developed profile functions for north-south oriented panels of central districts of the Barapukuria coal mine.

A subsidence curve of the west-east direction panels has been created based on the developed profile function, shown in Figure 8; and subsidence curves of the north-south direction have been created and shown in Figure 9a-f.

An ultra-thick coal seam of the Gondwana age has been discovered in the Barapukuria coal basin. The average thickness of the coal seam VI is 36m. It is observed from the feasibility study data that the coal seam is inclined and asymmetric having shallow depth at the periphery. A north-south aligned major fault boundary has controlled the eastern boundary of the coal basin. Consequently, the thickness of coal seam VI is increased and found deeper in depth toward the eastern portion of the central district of the Barapukuria coal basin.

The subsided area observed at the Barapukuria coal mine is attributable to the extraction of coal using the longwall caving mining method. The subsidence lakes located above the coal mine are the cumulative result of a series of coal panels. In the central district of the mine, there are 7 north-south aligned panels arranged side-by-side. The longwall caving mining was planned by downward cutting, with 9 slices considered for each corresponding panel. Finally, four slices is to be operated in one panel, and three slices are to be applied in most of the panels to date (Figure 8).

The subsidence prediction for the Barapukuria Longwall mine is a special event that must be considered a cumulative function of all the panels. Moreover, the configuration (rectangular) of the panels maintained a west-east and north-south trend. Considering the panel configuration the subsidence prediction has been calculated based on the profile function. To find out the possible maximum subsidence for each panel, two subsidence factors have been considered as the lowest (0.25) and highest (0.75) limit of the maximum subsidence, shown in Tables 1 and 4.

Maximum subsidence has been calculated and the calculated subsidences are- above panel 1308 (lowest 2.92 m., highest 8.77 m.); panel 1412 (lowest 3.41 m., highest 10.23 m.), and panel 1314 (lowest 3.17 m., highest 9.5 m.). The maximum water depth (10.5 meters) of the lakes developed due to the subsidence has been measured by HawkEye DT1H Handheld depth finder. The measured values have been plotted according to coordination and correlated with the newly developed profile function data. The result obtained from the calculation using the profile function curves shows that the predicted values closely matched the measured data of panels 1308, 1412, and 1314 shown in Figure 10.

The correlation coefficients between the predicted and observed values of the mentioned panels are 0.9875, 0.9993, and 0.9752 respectively, indicating strong agreement within the datasets.

Mining-induced land subsidence is a continually evolving process that begins post-ore extraction and extends for several years beyond the cessation of mining activities. This ongoing process results in the gradual development of the subsidence trough, influenced by factors such as the mining method, extraction rate, shape and size of extraction panels, characteristics of the immediate roof rock, and geotechnical properties of the overburden. A giant subsidence trough of around 1.25 square kilometers has been developed in the central district of the mining area. Subsidence prediction in such a complex environment is one of the toughest exercises in mining industries. In this study, the newly proposed profile functions for all the rectangular panels have been adjusted by field investigation in the Barapukuria coal mine.

Utilizing the new model, which relies on the straightforward profile function, shortens the time needed to calculate surface subsidence predictions and enhances the precision of subsidence forecasts. The proposed model demonstrates a strong correlation between measured and predicted subsidence by utilizing the measured water depth instead of directly measuring surface subsidence in the subsidence lake at the Barapukuria coal mine. The correlation coefficient ranges from 0.9752 to 0.9993, indicating a highly satisfactory level of correlation.

This work is an original contribution to the field and has not been published in any form or language elsewhere. The results are presented clearly and honestly, without any fabrication, falsification, or inappropriate manipulation of data (including image-based manipulation). The authors follow discipline-specific rules for data acquisition, selection, and processing.

The submitted work is experimental work performed in the field and laboratory. No human subject or living organism/tissue is involved in this research.

No consent to publish is to be shared.

All authors contributed to the research work. Material preparation, data collection, and analysis were performed by Md. Abdul Malek. Moreover, the first draft of the manuscript was written by Md. Abdul Malek and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript

The authors stated that no financial support was received during the preparation of the manuscript. The investigation has not been financed by any organization or sources.

First and foremost, we would like to express our gratitude to the Bangladesh University Grants Commission (UGC) for awarding the prestigious PhD fellowship, which greatly facilitated our research endeavors in the country. We extend our heartfelt thanks to the Managing Director Md. Saiful Islam Sarkar, and all the officers & staff of the Barapukuria Coal Mining Company Limited (BCMCL) for their support, providing access to various published and unpublished documents, and assistance during fieldwork in and around the mine area.