Deep Electrical Images, Geosignal and Real Time Inversion Help Guide Steering Decisions

Abstract

Abstract Early production, as well as ultimate oil and gas recovery, from a reservoir often depend on the timeliness and the accuracy of geosteering decisions. Exiting the reservoir during drilling results in costly nonproductive intervals. Even staying within the reservoir but in a nonoptimal location eventually leads to early water breakthrough while leaving behind valuable oil. In recent years, azimuthal deep resistivity measurements have been recognized as beneficial to real-time steering decisions. Because of their deep investigation, azimuthal deep resistivity measurements anticipate exits from the reservoir well before such events occur. In addition, their azimuthal sensitivity clearly points to the direction of preferred evasive actions. Azimuthal wave resistivity measurements take on multiple embodiments with multiple characteristics and multiple depths of investigation. The best results are achieved by jointly interpreting several of these measurements according to workflows that are specific to the particular applications. In the simplest cases, the up-down resistivity curves exhibit an unexpected behavior that has proven valuable both to petrophysicists and to geosteering specialists. When approaching conductive overlaying shale, for example, the up-curve tends to read the resistivity of the reservoir and the down-curve exhibits amplified horns beneficial to reservoir navigation. Resistivity images feature bright spots whose progression with increasing depth of investigation facilitates the avoidance of unwanted boundaries. A new measurement, designated as a geosignal, features strong lateral sensitivity. The geosignal from the deepest spacing is best suited to provide an early indication of the approaching boundary, with a near-exponential dependence on the distance to boundary. Quantitative inversion based on a subset of the azimuthal resistivity logs and the use of limited local knowledge helps to quantify the distance to the reservoir boundaries and their rate of approach. This paper presents some of the most commonly used interpretation methods are demonstrated on computer models, and then applied to various wells, with applications varying from thick reservoirs to interbedded sand-shale sequences.