Rock Physics and Time-Lapse Monitoring of Heavy-Oil Reservoirs D.R
Total Page:16
File Type:pdf, Size:1020Kb
CANADIAN HEAVY OIL ASSOCIATION SPE/PS-CIM/CHOA 98075 PS2005-435 Rock Physics and Time-Lapse Monitoring of Heavy-Oil Reservoirs D.R. Schmitt, U. of Alberta Copyright 2005, SPE/PS-CIM/CHOA International Thermal Operations and Heavy Oil Symposium overall success of a production strategy. Tracking the loss of This paper was prepared for presentation at the 2005 SPE International Thermal Operations injected fluids from the reservoir is critical not only to the and Heavy Oil Symposium held in Calgary, Alberta, Canada, 1–3 November 2005. overall direct operational costs but could be significant to This paper was selected for presentation by an SPE/PS-CIM/CHOA Program Committee following review of information contained in a proposal submitted by the author(s). Contents of longer term environmental liabilities should leaked fluids the paper, as presented, have not been reviewed by the Society of Petroleum Engineers, contaminate surrounding formations. Secondarily, having Petroleum Society–Canadian Institute of Mining, Metallurgy & Petroleum, or the Canadian Heavy Oil Association and are subject to correction by the author(s). The material, as timely detailed temporal and spatial constraints on in situ presented, does not necessarily reflect any position of the SPE/PS-CIM/CHOA, its officers, or members. Papers presented at SPE and PS-CIM/CHOA meetings are subject to publication conditions (e.g. pore fluid pressure, saturation, and review by Editorial Committees of the SPE and PS-CIM/CHOA. Electronic reproduction, temperature) within the reservoir could assist ongoing distribution, or storage of any part of this paper for commercial purposes without the written consent of the SPE or PS-CIM/CHOA is prohibited. Permission to reproduce in print is adaptation of the production process. restricted to a proposal of not more than 300 words; illustrations may not be copied. The proposal must contain conspicuous acknowledgment of where and by whom the paper was There have been substantial developments in the use of presented. Write Librarian, SPE, P.O. Box 833836, Richardson, TX 75083-3836, U.S.A., fax time-lapse (or 4-D) seismic techniques to assist in monitoring 01-972-952-9435. such reservoirs over the last decade particularly in the context of heavy oil production (e.g. Eastwood, 1993; Eastwood et al., Abstract 1994; Kalantzis et al., 1993; Li et al., 2001; Lumley, 1995; Production of hydrocarbons from heavy oil reservoirs often Macrides et al., 1988; Paulsson et al., 1994; Pullin et al., 1987; requires large investments both in capital and in operation and Schmitt, 1999; Siewert, 1994, Lines, 2002); most of this work despite the best efforts portions of the reservoir may be is published within the geophysical literature. Here, the basic bypassed or injected fluids may be lost. Time-lapse concepts of time-lapse geophysical surveying in the context of geophysical methods have shown promise in monitoring such heavy oil production is reviewed beginning with some reservoirs, particularly in the case of weakly consolidated elementary principles of reflection seismology. The sands. The seismic response can depend on a variety of production from heavy oil reservoirs results in large changes factors that include temperature, pore pressure, stress, fluid with time of temperature, stress, pore pressure, and fluid saturation state, and material damage. These effects are content. These variations influence, sometimes quite strongly, briefly reviewed in the context of Gassmann’s relation that the elastic compressibilities and densities of the subsurface predicts the compressibility, rigidity, and density of a formation that in turn cause the overall seismic responses to saturated porous rock. While Gassmann’s relation is widely change. These rock physics factors are considered in some used in fluid substitution studies, it is not yet clear whether it detail. While existing theories have been employed to model can be appropriately applied in the context of viscous heavy the seismic properties of heavy oil reservoirs, there still oil reservoirs. remains a great deal to accomplish in the understanding of this complex material. Introduction The continual increase in demand for hydrocarbon energy Background Information sources coupled with the depletion of conventional light oil Time-lapse seismology is predicated on observing changes in reservoirs around the world has placed a great deal of attention the seismic behaviour with time, these changes are produced on production from bituminous and heavy oils. The high by variations in the physical properties of the rocks at depth. viscosity of these fluids, however, complicates their This background section begins with some tutorial information production using normal techniques. A variety of methods are on reflection seismology in order to place the context for the employed to lower the viscosity such that the oils will flow rock physics principles employed to understand a heavy oil with the more popular ones requiring the injection of heat (via reservoir. steam or hot water) or solvents (such as light hydrocarbons or Time-lapse Seismolgy carbon dioxide). Different rock types are characterized by variations in their The capital and operational expenses associated with such elastic properties and densities which can further be related to enhanced oil recovery technologies are substantial; and it is the seismic compressional VP and shear VS wave velocities important that as much information about the production according to: process be collected as possible. To first order, having knowledge of which sections of the reservoir are accessed and which have been inadvertently bypassed is crucial to the 2 SPE/PS-CIM/CHOA 98075 4 A number of the extrinsic conditions such as pressure and K + µ temperature change with time within a producing reservoir. 3 These will influence the rock properties, as will shortly be VP = ρ (1) seen. Consider the case of the illustration of the three layers of Fig. 1 with the middle layer 2 being the reservoir. Again for µ V = purposes of illustration, assume that for a variety of reasons, S ρ replacement of gas with water during injection for example, that both ρ2 and the seismic VP2 increase with time while the where K and µ are respectively the bulk and shear moduli of velocities and densities of layers 1 and 3 are static. This will the overall composite rock. Ignoring for the moment any result in variations in the absolute values of both of the temporal variations, in the simplest case of a layered reflection co-efficients from R12 → R12’ and R23 → R23’ as sedimentary geology each layer will be characterized by a shown in Fig. 2. Mapping of such amplitude and travel-time certain seismic velocity and density; and there is a contrast in changes form the basis for the field observational aspects of these properties at the interface between two layers. When an time-lapse seismology. Although it is beyond the scope here, elastic wave reaches such an interface the energy is partitioned these data can either be interpreted by forward modeling of the with, usually, a small amount reflected (echoed) back towards expected seismic response or by inversion to obtain measures the surface with the remainder continuting to propagate into of the changes in the elastic impedance Z = ρVP. the earth. For the simplest case of a reflection of a vertically propagating wave colliding with a horizontal geological interface, the strength of the reflected wave in terms of its amplitude is: r’(t) s’(t) ρ V − ρ V R = 2 P2 1 P1 (2) ρ2VP2 + ρ1VP1 w(t) where R is called the ‘reflection co-efficient’ and the subscripts 1 and 2 denote the physical properties of the upper R’12 and the lower layers across the interface, respectively. Examination of Eqn. (2) shows that the magnitude of R = increases with the simple difference ρ V – ρ V and that -1 2 P2 1 P1 R’23 * ≤ R ≤ 1 depending on the constrast in the properties between the layers. This change in sign defines the polarity of a reflection in the seismic trace. In the simple layered 1-D world, the seismic trace s(t) essentially consists of the convolution of a wavelet w(t) (i.e. Fig. 2. Illustration of the change in the reflectivity function due to the time plot of the behaviour of the particle motions or an increase in the velocity and density of layer 2. Dashed lines in particle velocities input to the earth by the seismic source) s’(t) delineate the original s(t) of Fig. 1. In this case the layer bounded by reflections R12 and R23 in Fig. 1 has increased in with the reflectivity r(t) (i.e. the time plot of the reflection co- velocity and density resulting in stronger reflections R12’ and R23’ efficients with depth into the earth with each reflection co- and a shift in the time upwards of R23’. efficient appearing as a simple spike with amplitude R at the time the reflection reaches back to the surface) as illustrated in The cartoon of Fig. 2 can hold for thick reservoirs. In the thin Fig. 1. Ensembles of such seismograms obtained over an area reservoirs typical in the Western Canada Sedimentary Basin or along a profile are used to create the seismic images for the responses are more complex as the two reflections will interpretation. Note that the cartoon of Fig. 1 is given in terms ‘tune’ together and there will often be little differential time of two-way traveltime instead of depth. shift. In this case, monitoring requires analysis of more subtle r(t) s(t) changes in the overall seismic amplitude supported by modeling (e.g., Zhang and Schmitt, 2004a,b). w(t) General Assumptions in Rock Physics The field data above can provide some indications of the R12 changes in velocity and density which may then be related = back to the elastic moduli K and µ using Eqns.