Share
Water management

# New Technology Reduces High Water Production in Heavy-Oil Field in Oman

## Field N is a complex heavy-oil field in the north of the Sultanate of Oman. The dynamic behavior of Field N is characterized by strong aquifer and is dominated by bottomwater drive.

The subject field is one of the largest mature oil accumulations in the South Oman Salt Basin. Production and integrity issues have been challenging in recent years, manifesting in abnormal water cut with time. A multidisciplinary study concluded that this water-cut behavior was the result of naturally occurring fractures that are limited in length and not vertically extensive. The proposed solutions included detecting fractures early using borehole imaging techniques and using elastomeric zonal isolation packers (EZIPs) in sealing as many fractures as possible. This completion method resulted in delayed water production and increased net present value by $2 to$3 million per well.

## Introduction

Field N is a complex heavy-oil field in the north of the Sultanate of Oman. It produces heavy oil from more than 1,000 wells, 80% of which are horizontal; all wells penetrate sandstone reservoirs. Production from these reservoirs started from vertical wells in 1985, but the field-development plan recommended a strategy of horizontal wells targeting the topmost part of the sand reservoirs to avoid the predicted oil/water contact to deter possible early water breakthrough. The dynamic behavior of Field N is characterized by strong aquifer and is dominated by bottomwater drive.

The study focused on the westernmost accumulation of Field N, called Field NG. From the time Field NG first came on production in 1984 until 2004, the water cut of the field gradually increased to 30% with an average of 20% initial water cut in this 20-year period. However, after 2004, the average initial basic sediment and water (BSW), or water cut, of new wells in the field jumped to 85%, and even wells that started with initially low BSW soon developed a very high water cut in 2–3 months. Through numerical simulation models calibrated with production history from Field N, it was observed that, if matrix conditions are assumed, water is unable to break through immediately at the start of production. Instead, for matrix reservoirs, 1–2 years are required before the water cut can reach 80%. High initial water cuts could only be explained by conductive features such as high-permeability streaks, fractures, and faults. The simulation models investigating the effect of fracture orientation on water development indicated that vertically nonextensive fractures had no influence on water production early on, thus demonstrating matrix-like behavior in the beginning. However, as cones developed and the water table rose with time, water was able to reach these short fractures, and therefore very high initial BSW was observed in new wells.

To prove this concept, an extensive literature search—many authors have studied fractures as a possible cause of early water breakthrough—and data-mining study was conducted, which suggested the presence of fractures in the field.

## Field Description

The area is bounded by a major fault in the north and there are salt withdrawal pods in the south. Many fractures are located close to the major faults. Fig. 1 shows the fault positions in the Field N area. The NG field looks to be heavily faulted, especially in the southwest. Consequently, the production behavior of wells in the southwest displays a development of high BSW.

The net daily oil production rate dropped steeply in 2012. The majority of wells drilled in 2012, 2013, and 2014 came on stream below target. Reviews of these wells indicated that high initial water cut was the main reason for poor performance.

## Study Finding

• Two main fracture groups are identified in the field: fracture corridors and megafractures. Fracture corridors are highly clustered around faults, while megafractures have low correlation to faults.
• Faults, fracture corridors, and dispersed fractures have no correlation to BSW.
• Unlike faults and fracture corridors, megafractures appear to have a correlation to BSW increase.
• Wells with high mud losses always produce high BSW or have rapid water breakthrough, although mud losses are only reported on a small fraction of wells.
• Underbalanced drilling (UBD) indicated that fractures do indeed play a role in water production.
• Spud date correlates to the rate of BSW rise, with recent wells being more prone to rapid water breakthrough.

## Field Performance: Production Data and Water-Breakthrough Behavior

For the purposes of this study, the NG reservoir was divided into six zones: northeast (NE), east flank north (EFN), east flank south (EFS), middle P5 valley (P5V), weathered Amin zone (WZ), and west. Overall, NG initial oil rate is declining with time while estimated ultimate recovery diminishes. Of approximately 160 wells, 100 have an initial oil rate of 50 m3/d or less.

As with the initial oil rate, the initial BSW of wells was investigated as a function of the spud date. The initial water cut of more-recent wells has been higher than that of older wells. Northern zones seem to have a wider spread in the current BSW data, although there are a few wells in the west that indicate a lower initial BSW.

In addition to initial water cut, the time to reach 80% water cut was also investigated. These data also show inverse relationship with time. As the initial water cut increases, the time to reach 80% water cut decreases in newer wells.

Finally, initial oil rate vs. initial BSW was plotted. The initial oil rate drops with increasing water cut in all zones except the weathered Amin, which does not show any trend. Thus, high water cut is responsible for the low initial oil rates.

## Reservoir Simulation Modeling Study

A numerical sector model was created to explain these findings. A simple 3D rectangular box model was created in a black-oil simulation software. A simple model rather than a sector model was preferred because the intention was to differentiate the effect of matrix behavior from that of fracture behavior without considering heterogeneity or well-trajectory differences. The model was compared with real field examples to calibrate and study reservoir behavior. History matching was not a goal of this study. This modeling allowed the detailed examination of the effect of a wide range of parameters on the water-cut behavior.

The results show that during oil production the water/oil interface may rise and deform into a conical shape near the well because of a localized decrease in pressure. This phenomenon is known as water coning. At the time of water breakthrough, the cone is observed to be narrower than in more-advanced stages when the water cut has risen to higher levels. This phenomenon is described in detail in the complete paper, along with the results of fracture-sensitivity studies.

## Solutions to Early Water Breakthrough

Different options were investigated to stop water encroachment into the wells. These options can be classified into two categories: smart EZIP placement and smart well-completion techniques. EZIPs are water-swellable packers that provide segmentation in the wellbore for future water-shutoff options. Smart EZIP placement refers to obtaining information about the location and number of fractures in the wellbore and then placing EZIPs or blanks to seal these water-conducting features. This information can be obtained in various ways (e.g., UBD and borehole-imaging technologies). UBD was used to provide dynamic data to allocate active fractures and to seal them out using EZIPs. Fractures can also be discovered using a sonic tool, which can give more indication of the zonal permeability of the unit and show the aperture of the fractures. The sonic tool can be run together with formation microimaging (FMI) to acquire more information regarding the fractures. After interpretation of the FMI data and identification of the conductive fractures, EZIPs and blanks were used in the completion to isolate the fractures. The team used productivity data to optimize the number of EZIPs and length of blanks to ensure the maximum oil rate could be achieved.

New water-shutoff technologies were also researched. A suitable commercially available option is the autonomous inflow-control device (AICD). AICD technology consists of mechanical devices installed with the sandface completion, which react in real time to the local fluid dynamics, managing (decreasing) the water influx and, therefore, equalizing the drawdown along the horizontal section of the well. No cables are required, and once the devices are installed and the completion is in the ground, no interpretations, actions, or decisions are necessary. Also, AICD technology is of considerably lower cost when compared with automated inflow-control valves.

AICD Field Examples. Two wells, NMx1 and NMx2, were completed with AICDs in 2018 and indicated significant reduction in water cut as a result. The wells were compared with their neighbors to evaluate the outcome. In the first example, Well NMx1 presented a water-cut rate six times lower than that of its neighbor. Even though the first 3-month average oil rates were comparable, the new well produced less water. In the second example, Well NMx2 presented a ­water-cut rate four times lower than that of its neighbor Moreover, the first 3-month average oil rate was three times higher than that of the neighboring well, with a significantly decreased gross rate.

This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper SPE 193658, “Tackling High Water Production in Oman South Fields With New Technology,” by Ayca Sivrikoz, Maria Jimenez Chavez, and Salim Buwaiqi, Petroleum Development Oman, prepared for the 2018 SPE International Heavy Oil Conference and Exhibition, Kuwait City, Kuwait, 10–12 December. The paper has not been peer reviewed.