4D inversion of borehole gravity data for monitoring fluid fronts

Abstract

Monitoring fluid movement is an important component in enhanced oil recovery (EOR) and CO2 sequestration. The newly available slim-hole gravimeter operating at high temperature offers a new avenue for such monitoring efforts because of the direct sensitivity to the change in the density distribution. We present a time-lapse gravity inversion algorithm for recovering the front of injected fluid using borehole gravity measurements. We assume that the horizontal extent of the fluid can be represented by a polygon with known but variable thickness and density contrast due to fluid substitution. We represent the evolution of the front as a 4D function of the spatial position and time since the initiation of the injection. The inversion can be carried out either independently at discrete time points or as a single inversion simultaneously over all time points. We demonstrate that the latter approach is superior in that it is more stable and offers improved capability in detecting break-through events at later times. In this paper, we will describe the details of the two inversion approaches, including two different model objective functions in polar coordinates and the nonlinear solution strategies. We will illustrate the advantages and drawbacks of independent and simultaneous 4D inversions using numerical examples