Frances Metcalfe | Cambridge Consultants
Water is a significant by-product of the production of oil and gas – in some cases, accounting for up to 95% of produced fluids. The resulting reduction in revenues, coupled with the costs of water treatment and safe disposal, can render a well uneconomic – and even stop production altogether. But conventional techniques for monitoring the production profile of a well to enable troubleshooting are less effective in the increasing number of horizontal wells. Gaining insights into water production and location of inflows requires a number of trade-offs to be considered spanning accuracy, technology selection, measurement location and of course, cost. So what options are currently available to operators of oil and gas wells and what challenges remain?
The balance of measurement requirements of water production is important. Whilst an accuracy at the wellhead of 0.5% for overall water cut (the % of water in produced fluids) is needed for financial purposes (as the value of oil is much greater than that of water), operators of an oil or gas well need positional information on which reservoir zone is producing water so that mitigating action can be taken, but can trade this off against an accuracy of 5% to 10% water cut from downhole sensors.
So, where to start? To make an informed choice we need to consider the physical constraints and the available technologies.
Just making measurements in this environment presents many challenges requiring a multidisciplinary approach to engineer a good measurement – fluidics, sensor physics and electronics, and mechanical design are all key to allow sufficient sensitivity but also permit a sensor to operate over an extended period (10 years or more for a permanent sensor) in a harsh environment. Sensors must operate under high pressure, high temperature and offer chemical resistance (to H2S for example). Once that challenge is met, we still need to add data analysis and mathematics to extract actionable information from the sensor data. Add to this resilience; failure cannot interfere with production and redundancy will be important for key measurements, requiring rigorous application of reliability engineering.
Wellhead water cut meters are well established and are available from a number of suppliers but don’t provide information about the location of water production, and often represent measurements of co-mingled flows from several wells.
Downhole sensors can offer improved information on the location of a water inflow but introduce additional considerations: size (needing to fit in the confined annular space for example), power and communications requirements, and restrictions of number of and type of cable that can be run downhole and through packers. This is not a significant issue for a single well or lateral, but is much more of a challenge for multi-laterals. Solutions vary from permanent installations to intervention techniques and systems (e.g. wireline conveyed instruments). Ideally, these should offer a measurement of a representative sample of the zone, work for the lifetime of the well, and shouldn’t interfere with production e.g. by intruding into the flow itself.
Sensing and measurement techniques employed can be classified either as direct or indirect measures of water.
Direct measurements involve measuring the material properties of the produced fluid using physical techniques such as optical absorption and other EM or electrical methods (such as conductivity or capacitance). The performance of these can be limited by factors such as optical scattering or distortion of electrical properties due to variations in water salinity.
Indirect methods infer the water content and flow from measures such as downhole temperature and pressure, for example analyzing the outputs from downhole gauges or distributed temperature sensing (DTS) which monitor thermal profiles in a well, and comparing thermal models with the results to infer water inflow from deviations from the expected natural geothermal gradient. This technique becomes less effective as wells approach 90° deviation (horizontal), as there is no longer a geothermal gradient along the sensor path and temperature deviations can be smaller than the resolution of the DTS instrument. Emerging technologies employ chemical tracers that are selectively absorbed by water and oil in the flow; their concentrations in the produced fluids are then measured offline in samples taken at the surface. These have the advantages of being passive, without need for downhole power and communications, although their lifetime is finite and the techniques are not currently real-time since offline analysis is required.
In conclusion, we can see that downhole water detection is still an evolving area; there is no ‘one-size fits all’ technique, and in particular, multilateral (horizontal) wells present some new challenges. There are opportunities to develop new technologies such as RF measurements to measure differences in physical properties of hydrocarbons and water to determine the composition of fluid flowing past a non-intrusive sensor. In addition, the use of analytical techniques to fuse results from emerging and existing measurements offers the potential to address these important challenges and enable operators to take well-informed and early action to mitigate excess water production.