P225 PRODUCTION INDUCED MICROSEISMICITY: THE GEOMECHANICAL SIGNIFICANCE Introduction 1 Throughout the life of a reservoir the hydrocarbon production process induces pressure changes within the reservoir. These changes result in perturbations to the in situ stress conditions that propagate through the reservoir. One of the key drivers for the development of 4D (3D time-lapse) reflection seismic has been the need for time-lapse images of where and when these changes are taking place and also their magnitude and impact on production. Geomechanical modelling is an important tool in this characterisation process providing insight into the changes occurring within the reservoir due to

The determination of rock mass properties for engineering design is considered from two perspectives. These are in-situ measurement, including classification-based methods, and the limitations of the classification approach.Several measurement methods are available which will give useful results, if used appropriately. The choice of different methods will depend on the nature of the rock mass. The required accuracy should be considered realistically and, in many cases, high levels of accuracy are not, in fact, necessary. Examples are included, which show how the mass strengths of mine pillars were determined with acceptable accuracy using well known rock mass classification schemes, modified as necessary to accommodate local conditions.Rock mass classification is a widely used, economical and extremely useful basis for determining properties, but there are dangers in uncritical application. The classification methodology is critically examined, and the use of multivariate statistics applied in a multi-dimensional space is considered to optimize the usefulness of the measured data.

Abstract The flow characteristics of fractures are significantly affected by the stresses acting upon them. Reservoirs where the majority of the fluid flow is through fractures are very sensitive to the stress regime and its perturbations. Activities, such as production from a reservoir or injection into a reservoir, cause changes in the fluid pressure. This will alter the effective stress, which in turn will affect the flow of hydrocarbons through the fracture network. In order to understand how hydrocarbons flow within a fractured reservoir it is necessary to determine the influence of both stress and fluid pressure in a fracture network. There have been many experimental studies showing the influence of stress on the flow of fluid through single fractures from which a relationship between fracture permeability and effective stress can be developed. Effective stress acting on a fracture can be established by considering the orientation of the fracture relative to the in-situ stress regime and the fluid pressure within the fracture. Once the effective stress acting on a fracture has been found the corresponding permeability can be assigned within a discrete fracture network (DFN) model. This approach then allows examination of stress-sensitive flow within fractured reservoirs. This paper presents a methodology for incorporating in-situ stresses and stress changes due to production and injection in DFN modelling. The approach is significantly different from published methodologies on this subject, which are based on continuum porous medium modelling. By simulating and interpreting well tests in the fracture network during reservoir appraisal and after significant production, the presented approach would help in determining the influence of stress on hydrocarbon flow within a fracture network and to optimise future well locations and orientations.

Hot Dry Rock (HDR) technology started from an idea to help fulfill future energy needs as the availability of cheap fossil and other known fuels slowly reduces. The HDR concept itself is very simple but the development of the associated technology has taken significantly longer than anticipated. Anyone with experience of natural materials such as rocks knows that there are always imponderables that have not been really understood and indeed cannot at present be dealt with in a fully satisfactory manner. Furthermore, geology always has a habit of presenting us with new problems. Under these auspices and considering the limited funds that have been made available, it is encouraging to note that at long last light is visible at the end of the long tunnel of uncertainty. The results of the 1997 circulation test at Soultz-sous-Forets (France) certainly seem to show that we may have come up with a type of concept and an appropriate set of background site conditions to advance the technology.The concept of an HDR reservoir has evolved from that of a single penny-shaped fracture borrowed from the oil industry to the present graben or HWR (Hot Wet Rock) concept. International co-operation has been a key issue so far, and the expensive nature of this research demands that co-operation of this type continues to break new ground in the future. The necessary supporting technology has also evolved and the time appears to be ripe for taking advantage of this new and exciting development. Not all the answers are known, but at least we know now which questions to ask. It is worth remembering that there is still no commercial HDR plant in existence to provide real data on building, operating and maintenance costs for planning a new unit. This should not be regarded as an insuperable problem. If such were the case, then we would not have any aircraft, steel, shipping, telecommunication or nuclear industries.Many people fed that HDR technology will be needed sooner or later and the important question now is how quickly it can be put into practice when the need does arise! Recent moves to form a consortium for this next step at the Soultz site are very encouraging and show the promise and confidence that commercial and industrial interests have in its future. (C) 1999 CNR. Published by Elsevier Science Ltd. All rights reserved.

A two-dimensional numerical model of coupled fluid flow, heat transfer and rock mechanics in naturally fractured rock is developed. The model is applicable to assessments of hot dry rock (HDR) geothermal reservoir characterisation experiments, and to the study of hydraulic stimulations and the heat extraction potential of HDR reservoirs. Modelling assumptions are based on the characteristics of the experimental HDR reservoir in the Carnmenellis granite in Cornwall, S. W. England. In particular, the model assumes that fluid flow is horizontal and is confined to the most hydraulically conductive fractures in two orthogonal and vertical fracture sets. The mathematical model representing the hydro-mechanical interactions that are induced during reservoir creation is described. Furthermore, the solution method for calculating thermally induced stresses and strains in the reservoir due to heat extraction is discussed. The resultant numerical model, HOTGRID, is used to investigate the effects of stress perturbations on the fluid flow distribution in an HDR geothermal reservoir.

To date, it has not been possible to take accurate in-situ measurements of the stress rise in the goaf as a result of longwall mining, due to the difficulty in maintaining contact between in-situ equipment and data loggers. For this reason, research into stress distributions within the goaf, loaded by overlying strata, has been addressed through numerical modelling, with a strain stiffening constitutive law for the goaf. The modelling results have been compared with existing methods and significant differences noted.

Probabilistic distributions have been fitted to joint mapping data obtained from South Crofty tin mine. This allowed a simple deterministic model of rock jointing to be developed. Using this model, key block analysis of the 380 Fm level footwall drive, of the No. 8 lode was carried out. The inherent limitations present in this simplified analysis are noted and the need for a probabilistic approach to key block formation is shown. The limitations of existing probabilistic methods of analysis are also examined. A different form of probabilistic analysis for key block determination has been developed using the computer code B3LHS. This method does not suffer from the inadequacies associated with other methods of probabilistic analysis.

At the hot dry rock geothermal energy project in Cornwall, England, run by Camborne School of Mines, a number of hydraulic injections have been completed with the aim of creating zones of enhanced permeability. The fluids used in the Phase 2 system, which is located at a depth of more than 2 km in granite, have been water, and high and low viscosity gels. A large volume, low viscosity gel injection, which relies upon controlled shear stimulation appears the most promising in the context of significant horizontal in situ stress anisotropy which applies at the Cornwall site.

The development of HDR reservoirs in the Carnmenellis granite has involved HFSM to depths of up to 2.6 km in several test series. The results of these and other stress measurements, and the operational experience gained, are reviewed in terms of proposed developments at 6 km. Potential borehole tangential stresses are examined in detail. HFSM will not be able to provide complete stress information because of geological and equipment limitations. However, in conjunction with other techniques, notably borehole breakout analysis, sufficiently comprehensive results appear possible to the extent that HDR systems can be designed and modelled.

Summary A research hot dry rock (HDR) geothermal energy reservoir is currently under development in the Carnmenellis granite in Cornwall, U.K., at a depth of about 6,600 ft [2000 m]. This paper details the performance, analysis, and interpretation of hydraulic tests using low to medium pressures and flow rates. These tests allowed an evaluation pressures and flow rates. These tests allowed an evaluation of the in-situ hydraulic parameters relevant to initial water storage and long-term leakage. Tests were analyzed using both diffusion theory and an explicit, Coupled hydraulic/mechanical computer code. At low fluid overpressures, the permeabilities, in terms of uniform diffusion, were in the range of 1 to 10d, rising to about 60 d at overpressures of about 725 psi (5 MPa). These figures imply that long-term leakage psi (5 MPa). These figures imply that long-term leakage from an HDR reservoir is minor. The computer model was able to simulate several cycles of injections and shut-ins at progressively higher pressures and flow rates with a single set of variables. Introduction An HDR geothermal energy research project is being conducted by the Camborne School of Mines (CSM) at their test facility at Rosemanowes quarry in Cornwall (Fig. 1). The facility includes two deviated wells, RH11 and RH12, each drilled to a vertical depth of about 6,600 ft [2000 m] (Fig. 2). The holes are located within the Carnmenellis granite throughout their lengths, and are inclined at 30 from vertical in their lower sections to intersect the predominantly vertical natural jointing. The primary objective of the project is to link the two wells through the natural jointing by using a combination of explosive pretreatment and hydraulic stimulation to create a heat transfer zone (or reservoir) for the circulating water. However, it was first necessary to characterize the hydraulic behavior of the rock mass with modest flow rates and fluid overpressures before the natural joint structure was modified by major hydraulic injections. The resulting information then could give reference conditions for comparing the results of the high energy tests. The information is relevant for determining the leakage losses at the limits of the stimulated reservoir. It is also of interest to compare the results with those obtained at depths of 1,000 ft [300 m] in previous experiments at the same site to obtain a depth perspective for the hydraulic behavior. Since the test depth is unique for crystalline rocks in the U.K. and is unusual elsewhere, the results are of interest for other projects involving facilities at comparable depths (e.g., liquid waste disposal and deep-mined hydroelectric pumped storage). Geological Conditions The Carnmenellis granite pluton was emplaced about 285 million years ago, and subsequently has undergone a complex tectonic history. The pluton is exposed at the ground surface over an approximately circular area of about 50 sq miles [130 km]. The Rosemanowes quarry test site is near the center of the exposure (Fig. 1). Topography is gently undulating with a maximum relief of about 330 ft [100 m]. There is a network of dominant, near orthogonal, vertical joints throughout the granite, with typical spacings of 3 to 10 ft [1 to 3 m] near the ground surface, 10 to 30 ft [3 to 10 m] at 2,500-ft [750-m] depth, and similar to wider spacing at 6,600 ft [2000 m]. Nearly horizontal joints also occur but less frequently. All joints are typically tight, away from zones of stress relief. In-situ stresses have been measured in some detail. At a depth of 6,600 ft [2000 m], the vertical, maximum, and minimum horizontal stresses are approximately 7,500 psi [52 MPa], 8,700 to 10,200 psi (60 to 70 MPa) and 4,400 psi [30 MPa], respectively. The horizontal stress directions and major near-vertical joint set strike orientations are shown in Fig. 3. The fabric of the Carnmenellis granite consists of megacrysts of alkali feldspar set in a coarse ground mass of plagioclase, alkali feldspar, micas, and trace minerals. Grain size is variable, with megacrysts up to 0.75 in. [20 mm] long and crystals in the ground mass typically 0.08 to 0.2 in. [2 to 5 mm] in diameter. The permeability of the fabric is extremely low, with values permeability of the fabric is extremely low, with values of 10(-3) d obtained by laboratory water diffusion tests. The mechanical properties may be characterized by a uniaxial compressive strength of about 19,600 psi [135 MPa], with dimensionless triaxial strength parameters m and s of 21 and 1.0 per Hoek and Brown, parameters m and s of 21 and 1.0 per Hoek and Brown, Young's modulus of 9.43 × 10(6) psi [65 GPa], and Poisson's ratio of 0.2. Poisson's ratio of 0.2. JPT P. 1982

Large scale hydraulic injections into rock at significant depth are common in the exploitation of hydrocarbon formations, geothermal energy systems and for liquid waste disposal. One example is the hot dry rock geothermal energy project operated by the Camborne School of Mines in Cornwall, U.K., where a cumulative injected volume of over 300,000 m3 was pumped into 2 km deep boreholes in granite.

A programme of overcoring rock stress measurement was conducted at South Crofty mine, Camborne, at a depth of 790 m in the Carmenellis granite to obtain information required for the Camborne School of Mines (CSM) hot dry rock (HDR) Geothermal Energy Research Programme at Rosemanowes quary in Cornwall, U.K.

Hydrofracture stress measurments were conducted in the Carnmenellis granite at the Camborne School of Mines Hot Dry Rock Geothermal Energy Research site at Rosemanowes Quarry near Penryn, Cornwall, U.K.

The design of large underground excavations in rock is generally based on a combination of theoretical and empirical methods. Shallow caverns require a greater emphasis on the geological regime and its influence on cavern stability. Methods are discussed of obtaining, presenting and evaluating structural data in relation to the field stress conditions.