GIS-Based Decision Support Methodology for the Assessment of the Impacts of Mining Subsidence

The objectives of this project were to develop and demonstrate practical decision support methodology for the assessment of the impacts of mining subsidence on natural features. The decision support tools were developed within the flexibility of the Geographic Information System (GIS) environment and uses relevant case studies to demonstrate the usefulness of GIS tools. The use of GIS was prompted by the fact that the process of understanding and managing coalmine subsidence impacts is, to a large part, a spatial one and that many of the factors that are critical to the assessment of subsidence impacts have a strong spatial component. For instance, some of the most pertinent factors that govern subsidence susceptibility in a mining area include proximity to longwalls, terrain attributes, and the distribution of sensitive features. Five main roles for GIS in subsidence impact management were described:

• the storage and management of spatial data related to mine workings (planned and existing) and the associated environment;
• site characterisation and identification and quantification of susceptible features, such as cliff lines and watercourses;
• the assessment or prediction of the extent and magnitude of subsidence impacts;
• researching and understanding subsidence processes; and,
• 2D and 3D visualisation for communication and research purposes.

In particular, the authors recognise that GIS and spatial science can be used to aid site characterisation/feature identification or the assessment/prediction of subsidence impacts as outlined in Section 6 of the 2003 NSW DPI SMP Guidelines (page 14), which requires a process that “(1) Characterises the nature, extent and magnitude of the expected subsidence impacts due to the proposed mining, and (2) Identifies priority risks, highlighting the expected subsidence impacts with high risk levels and /or potentially severe consequences.”. Accordingly, this report has focussed on these roles for GIS in subsidence impact assessment, and presented relevant case studies for the Southern Coalfield in collaboration with the industry partner for this project, BHPB Illawarra Coal. It was originally intended that case studies from the Western Coalfields and other partners in the Southern Coalfield would be undertaken, though these case studies were not forthcoming largely because of data availability issues.

Site characterisation is required for (a) reporting, wherein the site characterisation process can be used to provide a general description of the area, and (b) decision-making, because site characterisation can direct attention to particular features of interest and guide subsequent field observations. A case study is presented in which high-resolution topographic data are used to perform a rapid, ‘desktop’ characterisation of the surface over the proposed Dendrobium Area 2 coalmine. Relevant natural features such as cliff and steep slopes, valley floors, and areas of high-erosion potential were identified.

Features identified using spatial data and GIS (or traditional field-based methods) can then be evaluated for potential impact susceptibility based on either knowledge-based or data-driven methodologies. Case studies are presented based on both methods, focussing on the cliffs of the Nepean River near the proposed BHPB Illawarra Coal Douglas mine. The knowledge-based case study employs a number of spatial data layers and the factors identified in the “Management Information Handbook – The Undermining  of Cliffs, Gorges, and River Systems” (produced by Waddington Kay and Associates in 2003) to produce an assessment of expected cliff impacts. The results of this case study demonstrate the advantages of a digital, spatial approach to impact assessment, which include the ability to undertake rapid, flexible assessments which yield an easy-to-interpret visual product (i.e., a digital or hardcopy map) as an output. Data-driven modelling was also undertaken for the cliffs along the Nepean River gorge using a GIS-based ‘weights-of-evidence’ method and six evidential themes (i.e, predictor data sets):

• Surface Slope
• Cliff Height
• Planform Curvature
• Profile Curvature
• Distance to Watercourses
• Distance to Workings

Case Studies are also presented that examine two ‘emerging’ methods for coalmine subsidence mapping. The first case study evaluated the feasibility of using satellite-based differential interferometry with synthetic aperture radar (SAR) data for coalmine subsidence mapping and monitoring. Historical data, collected in 1995 by the Japanese Earth Resources Satellite, was used to trial the capacity of L-band SAR data in mapping subsidence in the Southern Coalfield. Although a relatively high quality interferometric image was derived for the region, which included surface deformation ‘fringes’ over some active mining areas, it was evident that the technology is not yet at a stage suitable for operational use in coalmine subsidence mapping. Major limitations include the high degree of expertise required to process the data, high software and data costs, limited data availability, and coarse pixel sizes. Conversely, airborne laser scan (ALS) data, which was evaluated using surveys in rugged terrain over Dendrobium Longwall 1, showed a great deal of promise despite some current limitations. Two pre-mining surveys were compared in order to quantify the magnitude of expected errors intrinsic to the process of ALS surveying in rugged, heavily vegetated terrain. The results indicated that large errors (in excess of 1 m) in surface height change mapping occur in areas of steep terrain such as along cliff lines or within drainage channels. A similar pattern of errors was evident when a post-mining ALS-derived surface was compared to a pre-mining surface. However, despite these errors the results generally matched the expected pattern of subsidence including the detection of relatively higher subsidence over multi-seam workings associated with past mining.

The main recommendations to come from this report is that the coalmining industry as a whole should encourage and facilitate the development of subsidence impact databases, consisting of mapped and annotated impacts associated with past and current mining activities. The value of collecting accurate spatial records of subsidence impacts, such as surface and underground fracturing, upsidence, cliff falls and so on is demonstrated in numerous case studies throughout this report.

An e-newsletter has also been published for this project, highlighting its significance for the industry.