Residual Oil Saturation Inside the Steam Chamber During SAGD

Abstract

The oil saturation inside the steam chamber during steamassisted gravity drainage (SAGD) has an important impact on economics and conservation. However, the SAGD residual oil saturation is difficult to determine and model because it involves long-term thermal effects and three-phase flow. SAGD has been widely studied and piloted, but improved understanding of longterm drainage effects represents a fundamental issue that requires improved understanding.

In the first part of the paper, we represent a sensitivity test done on the shapes and the endpoints of the two-phase relative permeability curves. We find that the water relative permeability and oil relative permeability in the gas-oil system are the main factors that determine the magnitude and shape of the oil saturation curve as a function of time.

Secondly, we demonstrate how to adjust the krog relative permeability curve to match a theoretically determined residual oil saturation, which is supported by laboratory data. We propose that, during SAGD, the flow of oil can be split into two regimes. In the first regime, close to the edge of the steam chamber, oil drains quickly in a short period of time. In the second regime, oil drains slowly within the steam chamber for a longer period of time, as it is produced by "film drainage." To capture these flow regimes, the oil relative permeability curve in the gas-oil system, krog, is split into two. At higher liquid saturations, the first flow regime is represented and, at lower liquid saturations, the second regime is modelled.

Thirdly, the krog curve was adjusted so that the decrease in oil saturation with respect to time closely matched the theoretical curve while maintaining oil production rates expected for SAGD. Using this new curve at different pressures, we show that the residual oil saturation increases at lower SAGD operating pressures.

Introduction

In a steam-assisted gravity drainage (SAGD) process, bitumen drainage occurs mainly along the transition zone, which is the mobile liquid region at the boundary of the steam chamber. Bitumen drainage also continues to occur within the body of the steam chamber over a long period of time, such that the residual oil saturation gradually falls.

In this paper, the term residual oil saturation refers to the average remaining oil saturation within the steam chamber where the steam chamber includes all grid blocks in a numerical model containing any amount of steam. It is difficult to obtain residual oil saturation data as a function of time in the field so, in this study, we relied on theoretical and laboratory data. We calibrated reservoir simulation results to the theoretical and laboratory data, and then used the simulator to investigate the effects of SAGD operating pressures on the residual oil saturation.

Relative Permeability Formulation

Multiphase flow in a porous medium can be described using Darcy's Law as follows:

An integrated risk model of a drinking-water-borne cryptosporidiosis outbreak

A dynamic risk model is developed to track the occurrence and evolution of a drinking-water-borne cryptosporidiosis outbreak. The model characterizes and integrates the various environmental, medical, institutional, and behavioral factors that determine outbreak development and outcome. These include contaminant delivery and detection, water treatment efficiency, the timing of interventions, and the choices that people make when confronted with a known or suspected risk. The model is used to evaluate the efficacy of alternative strategies for improving risk management during an outbreak, and to identify priorities for improvements in the public health system. Modeling results indicate that the greatest opportunity for curtailing a large outbreak is realized by minimizing delays in identifying and correcting a drinking-water problem. If these delays cannot be reduced, then the effectiveness of risk communication in preemptively reaching and persuading target populations to avoid exposure becomes important.

Previous clinical diagnosis of chlamydia helps patients predict outcome of new chlamydia clinical test

Background: As part of an ongoing clinical trial evaluating the effectiveness of an intervention aimed at decreasing STDs in adolescent females, a host of different background measures were collected at baseline. These measures include self-reported behaviors, risk and probability estimates concerning STDs, self-reported prior medical diagnoses, and a Chlamydia trachomatis (Ct) infection assay, among other measures. This analysis examines participants' accuracy in estimating their own chance of chlamydia infection.Methods: Participants were 131 sexually active adolescent females, 79% black, 13% white, mean age 16 (range 13-19), who had been recruited from an adolescent health service. At the time of this study, participants were not primarily being seen for a clinical visit. Participants answered two questions that are examined here: The first asked, "What is the percent chance that you have chlamydia right now," accompanied by a visual scale ranging from 0% (no chance) to 100% (certainly). The second question asked, "In your life, have you ever been told by a doctor or nurse that you had chlamydia?" Following all questions a vaginal swab was collected by the participant, which was analyzed by polymerase chain reaction (PCR) for Ct. A multiple regression was performed, predicting the outcome of the chlamydia swab test using participants' estimates that they had chlamydia, their self-reported prior diagnosis with chlamydia (dichotomous), and the interaction between these two variables.Results: The regression F(3, 127) = 10.95, p <.01, revealed that participants' estimates of the chance that they had chlamydia significantly predicted outcome of the clinical test, t(127) = 3.34, p <.01. Previous diagnosis with chlamydia alone did not predict outcome of the clinical test, t(127) = -0.84, ns, but the interaction between the two variables significantly predicted clinical outcome, t(127) = -2.07, p <.05. A previous chlamydia diagnosis made participants' own estimates highly predictive of the clinical outcome; the estimates of those who had never been previously diagnosed with chlamydia had very little predictive power. Conclusions: Those participants who had prior experience with a chlamydia diagnosis were much more able to interpret their own risks and symptoms, and arrive at a very accurate estimate of the chance that they had a chlamydia infection. It is important to note that these participants were not seeking care at the time of this study, but that they had enough information to predict clinical outcome of a chlamydia test, especially those who had been diagnosed with chlamydia in the past. Arguably, these are they very patients who are most at risk for long-term sequelae of chlamydia infection, and asking (or having the patient ask) the relevant questions may increase the likelihood that they are identified and treated promptly, and early in the course of infection.