Hydraulic fracturing has been a key technology in the development of coalbed methane (CBM) resources worldwide. Obtaining adequate fracture length and conductivity has limited the ability to obtain adequate productivity improvement to further develop many small seams or stacked-seam reservoirs. Fracture complexity and growth out of the interval have frequently been cited as limiting factors in achieving optimal length in these types of intervals; however, the diagnostics to evaluate and model these effects have been limited. Finally, many of the past studies of hydraulic fracturing mechanics in coal have been focused on North American examples where normal faulting stress states are present, unlike many of the coal-producing basins worldwide and particularly in Eastern Australia. Using examples from the Scotia Field, we describe how past and present stress framework analyses and post-frac treatment diagnostics were integrated to better describe the in-situ stress state. Through analyses of these examples, we qualify the inter-relationships of productivity to in-situ stress, pre-existing fractures, and observations from the induced hydraulic fractures. Finally, we describe cases where the hydraulic fracturing complexity and in-situ stress conditions lead to wellbore complications and observable rock-mechanical failures. The end result is a more predictive model, which is being used to develop this CBM resource. Introduction Initially, most CBM projects focus on identifying contiguous coal sections with sufficient gas in place to support commercial production. Most often, basic geologic mapping and coal quality information are used to delineate and high-grade prospective areas. Ultimately, the reserves potential of these areas must be proven by direct measurement of key reservoir properties and production testing. Experience from analyses of coalbed methane projects worldwide indicate the more important reservoir parameters controlling gas productivity include:permeability (fracture/cleat system),gas content,well spacing,initial gas and water saturation in the fracture/cleat system, andthe relationship between reservoir pressure and flowing well pressure.1 Of these, permeability is the most important since it controls the flow rate of gas and water and can be affected by in-situ stress.In CBM projects in-situ stress can also affect the benefit derived from reservoir stimulation; often the well productivity and economics of a project can become highly dependent on the degree of stimulation effectiveness.2 Inter-relationship of Permeability to Stress. Since coals are a complex, fractured, dual-porosity system, a bulk permeability value is used to describe the system of low-permeability organic matrix components surrounded by cleats or natural fractures.3Because of the inherent variability in permeability and productivity, CBM exploration and development projects tend to be statistical plays, where large multi-well projects are used to explore and develop an area of reservoir.Ultimately, the average well productivity and the required well spacing to maximize gas recovery will govern the overall economics of the project. In many areas, proper characterization of the in-situ stress regime has played a pivotal role in the successful commercialization of CBM. Firstly, the aperture of fractures and cleats and hence their contribution to permeability tends to be extremely stress-sensitive.Sparks et al., identified how the permeability of fractures and cleats vary as a log function of in-situ stress (Fig. 1).4 Thus, low mean stress regimes support open and more conductive natural fractures. Inter-relationship of Permeability to Stress. Since coals are a complex, fractured, dual-porosity system, a bulk permeability value is used to describe the system of low-permeability organic matrix components surrounded by cleats or natural fractures.3Because of the inherent variability in permeability and productivity, CBM exploration and development projects tend to be statistical plays, where large multi-well projects are used to explore and develop an area of reservoir.Ultimately, the average well productivity and the required well spacing to maximize gas recovery will govern the overall economics of the project. In many areas, proper characterization of the in-situ stress regime has played a pivotal role in the successful commercialization of CBM. Firstly, the aperture of fractures and cleats and hence their contribution to permeability tends to be extremely stress-sensitive.Sparks et al., identified how the permeability of fractures and cleats vary as a log function of in-situ stress (Fig. 1).4 Thus, low mean stress regimes support open and more conductive natural fractures.

The contribution investigates the relationship between in situ stress regimes, natural fracture systems and the propagation of induced hydraulic fractures in APLNG's (Australia Pacific Liquid Natural Gas) acreage within the Jurassic to Cretaceous Surat Basin in southeast Queensland. On a regional scale the data suggest that large basement fault systems have significant influence on the lateral and vertical interplay between geomechanical components which ultimately control permeability distribution in the area. At a local scale we show several case studies of significant in-situ stress variations (changes in tectonic regime from reverse to strike-slip, changes in horizontal stress orientation as well as changes in differential horizontal stress magnitude) which are identified from wireline image log interpretations and geomechanical models constructed from wireline sonic and density data. These variations are reflected in hydraulic fracture propagation, which is monitored through microseismic monitoring, tiltmeter monitoring. Reverse stress regimes result in the propagation of horizontal fractures; in areas of higher differential stress linear hydraulic fracture orientations are common, whereas in regions of lower differential stress the orientation of hydraulic fractures appears influenced by both stress and pre-existing fractures. The paper is relevant for fracture simulation in areas with complex in-situ stress regimes. The major technical contribution of the study is the use of geomechanical modelling for predicting hydraulic fracture propagation styles.

Summary Modern hydraulic-fracture treatments are designed by use of fracture simulators that require formation-related inputs, such as in-situ stresses and rock mechanical properties, to optimize stimulation designs for targeted reservoir zones. Log-derived stress and mechanical properties that are properly calibrated with injection data provide critical descriptions of variations in different lithologies at varying depths. From a practical standpoint, however, most of the hydraulic-fracturing simulators that are currently used for treatment design use only a limited portion of a geologic-based rock-mechanical-property characterization, thus resulting in outputs that may not completely align with observed outcomes from a fracturing treatment. By use of examples from hydraulic-fracture stimulations of coals in a complex but well-characterized stress environment in Surat Basin of eastern Australia, we obtain the reservoir-rock-related input parameters that are important for the design of hydraulic fractures and also identify those that are not essential. To understand the effect on improving future fracture-stimulation designs, the authors present work flows for pressure-history matching of treatments and subsequent comparison of corresponding geometries with external measurements, such as microseismic (MS) surveys, to calibrate geomechanical models. The paper presents examples discussing synergies, discrepancies, and gaps that currently exist between “geologic” geomechanical concepts in contrast to the geomechanical descriptions and concepts that are used and implemented in hydraulic-fracturing stimulations. Ultimately it remains paramount to constrain as many critical variables as realistically and as uniquely as possible. Significant emphasis is placed on reservoir-specific pretreatment data acquisition and post-treatment analysis. Some of the obvious differences observed between the measured and fracture-model-derived geometries are also presented in the paper, highlighting the areas in fracture modeling where significant improvement is needed. The approach presented in this paper can be used to refine hydraulic-fracture-treatment designs in similar complex reservoirs worldwide.

Modern hydraulic fracture treatments are specifically designed to unlock reserves from particular rock types, especially in unconventional reservoirs. Progressive improvements in fracture design can be critically informed by post stimulation pressure analysis, yet this process is often overlooked. This paper documents the evolution of fracture designs by successively incorporating post-stimulation pressure analyses after major design changes that ultimately led to the design-optimization of fracture treatments in low permeability coals. The coals under context are the Walloons coal measures in Jurassic to Cretaceous aged rocks in the Surat Basin of southeast Queensland, Australia.Significant challenges are faced in stimulating the Walloons coal measures due to their thinbedded nature, that range from 0.2 to 3.0 m [0.66 to 9.8 ft] in thickness and, which are also inter-bedded with low permeability siltstones, minor sandstones and carbonaceous shales. Net coal thickness is 20 to 40 m [98.43 to 131.23 ft] in a gross sequence of 300 to 400 m [948.3 to 1,312.3 f] thickness. Reservoir complexity is further impacted by lateral continuity variations of coals, which generally have a high Poisson's ratio (Ͼ0.32). In particular where coal reservoirs display low permeability, understanding and implementing reservoir beneficial fracture treatments becomes pivotal to successful well performance.Modification of fracture designs during the fracture campaign included changing key parameters such as fluid types, pump rates, proppant loading and gel concentration. Both, the treatment and the calculated bottom-hole pressures, were evaluated using 3D fracture models, supplemented by an array of diagnostics such as surface tilt-meters, diagnostic fracture injection tests, micro-seismic monitoring and tracer logs as well as log derived stress models. The results of these diagnostics helped shape the design changes implemented throughout the campaign and has influenced designs for future trials also. Ultimately, it was observed that the treatments that were pumped using low gel loadings in conjunction with high proppant concentrations, and at relatively lower rates, resulted in better well performance.This paper presents the design and treatment evaluation process and also provides an insight into the progression of fracture design and subsequent treatments which were successful in overcoming reservoir complexities. The outlined approach can be used to refine hydraulic fracture treatment designs in similar complex reservoirs in Queensland, with worldwide applicability.