Abstract

This paper brings the discussion on brine mixing initiated at the 1999 SPE Symposium on Oilfield Scale full circle, and suggests that different scaling regimes may exist in any given reservoir that should impact the scale management strategy. The initial paper by White et al. (1999) identified lower than expected barium levels in many wells in the Alba field, and raised the question of where scale deposition is occurring. A follow-up paper at the same meeting the following year by Mackay and Sorbie (2000) identified from a theoretical standpoint where brine mixing should be expected, and suggested that significant scale deposition may occur deep within the reservoir, particularly where a large proportion of injected water sweeps through the aquifer. The current paper seeks to apply that theoretical analysis to the Alba field, on a well-by-well basis, to establish whether diagnostic tests can identify which wells should be treated conventionally, and which require special attention.

Firstly, production data and produced brine compositions are analysed to identify any recurring patterns. Wells are then classified and grouped according to various types based on this analysis. Well properties such as location within the reservoir and orientation to the flood front are then compared to identify whether they can be used to give a similar grouping of wells. Three principal zones are identified, with position relative to the injection wells and the aquifer identified as key parameters. Secondly, the flow patterns around each well are studied using the existing reservoir flow model. Modelling the dynamic mixing patterns throws further light on the differences between the zones, with scale dropout predicted in the aquifer in some areas, and in the oil-leg in others.

Recommendations are made regarding the treatment of existing wells and the optimum positioning, from a scale prevention perspective, of any new wells.

Introduction

Sulphate scale deposition is a common problem in reservoirs where injected seawater mixes with aquifer brines1,2. For reasons that will be explained in the next section, the problem is most severe in and around the production well bores, and can cause considerable disruption to hydrocarbon production after water breakthrough. To deal with the problem two general tasks must be performed. Firstly, the quantity and type of scale must be identified, together with the location and timing of the deposition. Secondly, a suitable removal and/or prevention strategy must be designed and implemented. Both of these tasks require reservoir and laboratory data, and field experience is also a vital component in ensuring successful treatment. Both tasks are routinely assisted by application of appropriate modelling tools3. The use of reservoir simulators to predict the timing and location of water breakthrough, and the application of scale prediction codes to identify the scaling regimes, have both been discussed previously4–9. In addition, models are routinely used in chemical scale inhibitor selection studies and to optimize inhibitor squeeze treatments10.

The topic of this paper is not so much the design of scale prevention treatments, although the findings have a bearing on this, but on identifying scale deposition regimes throughout an actual reservoir system. To do this produced brine chemistry data and field fluid flow modelling have been analysed on a well-by-well and field-wide basis for the Alba Field in the North Sea.

The evidence of both the produced brine chemistries and the flow calculations is that scale is depositing deep within the reservoir. However, the extent and impact of this deposition varies throughout the reservoir. Three regimes are identified, where the nature of the scaling problem at individual production wells can be related to the flow characteristics in that locality. Important factors are shown to be the proximity of injection wells and the proximity and extent of the aquifer.

This content is only available via PDF.
You can access this article if you purchase or spend a download.