INTRODUCTION
Hydraulic fracturing stimulations are routinely performed for commercial recovery of methane from coal seams. Both low (conventional) and abnormally high treatment pressures have been reported. This paper evaluates the fracturing pressure decline after shut-in to detect the mechanisms responsible for the high pressures. Based on the pressure decline behavior, fracturing treatment data, from over 75 coal hydraulic stimulations carried out in the Black Warrior Basin (Alabama), are classified into generic treatment categories.
Coal formations are characterized by a natural microfracture network of face and butt cleats, usually oriented parallel to the maximum and minimum principal in-situ stresses, respectively. Bedding planes are well defined and are generally horizontal. Young's modulus and fracture toughness values in these friable materials tend to be an order of magnitude smaller than for sandstones. As the mechanical properties of coal differ from most conventional reservoirs, new approaches to stimulation design and fracturing analysis are required. As a first approach to this, the following sections use the pressure decline behavior, after shut-in following hydraulic fracture treatments, to classify fractures into generic treatment categories. Pressure data from wells in the Black Warrior Basin (Alabama), tabulated in the COMPAS (Coalbed Methane Production and Stimulation) Database (Bell, et al, 1987), were used for this purpose. For the reservoirs considered, the minimum horizontal stress gradients (0.70 to 0.85 psi/ft), obtained from in-situ stress measurement (Jones & Gajewsky, 1988), are smaller than the overburden gradient (0.95 to 1.1 psi/ft). Fracture mechanics theory predicts that fluid treatment pressures slightly higher than the minimum in-situ stress axe sufficient to propagate a fracture. On this basis, fluid treating pressure gradients greater than i psi/ft axe considered abnormally high for vertical fractures. However, for the Black Warrior Basin and elsewhere (Palmer, et al, 1989), such high pressures appear to be observed in greater number of treatments. Numerous fracturing mechanisms have been hypothesized to explain these high pressures. These are: tip plugging/fracture blocking (Jones, et al, 1989), parallel fractures (Jeffrey, et al, 1989), increase of fracture toughness (Shlypobersky 8z Wong, 1988), complex fracture branches interacting with the natural fracture network (Palmer, et al, 1989), poroelasticity (Palmer, et al, 1989; Jeffrey, et al, 1989), tip plasticity, etc. However, controversy remains in distinguishing, with certainty, the actual mechanisms active under field conditions. Nolte (1979), have demonstrated that important parameters quantifying a fracture and a fracturing process can be?
