No abstract

K field is a faulted anticline structure lying in a turbidite environmental setting, which consists of two main sand bodies of 50-100ft gross thickness. Two main challenges during development stages were, 1). The narrow thickness of the fault block requiring accurate landing, well placement and characterization. 2). The low vertical permeability inside the sand which requires the straddling and precision of horizontal placement relative to reservoir boundary. Geosteering was proposed to mitigate those challenges, which is then translated into a directional drilling plan. Hybrid LWD combination of Seismic, Reservoir Mapping, RT-Image and Formation Pressure While Drilling technology were used overcome those objectives and challenges. Seismic was used to update target uncertainty in both depth and lateral coordinates, hence accurate landing inside the target fault block was the main driver. Reservoir Mapping Technology and borehole images were used to update the presence of fault and plan for an accurate placement of the horizontal section. Pressure While Drilling was used to update the reservoir pressure and fault/ connectivity between fault/compartments. Post job subsurface modeling was updated using those measurement for accurate interpretation, which was used for future field development plan. The drilling workflow was extensively discussed among the stakeholders to make sure it was fit for purpose and could achieve the objectives. Horizontal well landing procedures using Seismic While Drilling technology helped the well penetrate into fault block A at 100m before fault edge. Reservoir Mapping While Drilling technology enabled successful Geosteering inside the desired target zone. Another important application regarding the Reservoir Mapping Technology is the capability of resolving the internal sedimentary bedding feature (higher dip feature) and the delineating of multiple faults including sub seismic fault block B, which helped the team define well TD at the desired fault block position. An integrated interpretation between Reservoir Mapping While Drilling, High Resolution Images and Formation Pressure While Drilling to detect pressure continuity between one block to the other proved to be very helpful for production management and completion optimization. At the end, two horizontal wells were successfully drilled in the desired fault block. Everything worked, ie the technology, the people and the process, resulting in an accurate and well controlled execution of a complex and high-profile project. The novel approach was successfully demonstrated to ensure well objectives are achieved in an integrated manner. The flawless execution allowed the team to avpid drilling any side tracks which would have been costly.

The X field is a mature oil field producing with water injection in place. As part of a phased development, in Phase Two, two infill wells were planned and drilled to extract incremental recovery from a discovered undeveloped reservoir. This includes planning two horizontal producer wells requiring active real time geosteering utilizing deep resistivity tool technology. The wells’ objectives are to ensure the well placement was in an optimal location and to maintain the trajectory within a reservoir that is approximately 100ft thick and was believed to be homogenous field wide. The main challenge to this feat is the faulted nature of the field and the uncertainty in reservoir thickness and extend due to limited well penetrations at this reservoir level. During the planning phase, it was identified early that a deep resistivity tool would be beneficial in geosteering the wells. Prior to drilling, an integrated pre-job model was designed to test multiple tool settings and subsurface scenarios to strategize an execution plan identifying key points where there is a need for real time trajectory adjustments and to pre-plan alternative trajectories based on subsurface scenarios to enable efficient turnaround time to react to real-time results. Conventional navigation tools yield only a shallow to medium depth of measurement (~15ft) which would not have met the objectives of the well given the geological complexities (high fault offsets, laminated reservoirs) and well design (high angle to horizontal). The ultra-deep resistivity (UDR) tool was employed instead to enable trajectory optimization with up to ~100ft depth of investigation (DOI), using a multi-frequency, multi-spaced antenna design from medium and long spaced transmitter receiver spacings providing up to 9 vector components. In real time, the 1D inversion (using 5 of the vector components) was used for early sand and fluid contact detection. During execution, the same integrated team was monitoring the well and close interaction between the subsurface, geosteering and directional drilling team was a key requirement to ensure drilling of the well was safely and objectively executed, especially with the challenges posed with virtual working through a pandemic. As is when dealing with subsurface uncertainties, there were numerous surprises encountered during the drilling of the horizontal wells. Particularly in the matter of fault throw uncertainty and sand distribution. The initial 1D real-time UDR results were able to assist in real-time trajectory adjustments and to provide some geological understandings with regards to fault throw and location of possible faults along the well bore which were then confirmed with borehole image logs. Additionally, 3D inversion images were processed post drilling, and further geological insights were discovered with regards to the depositional trends on the reservoir. In a reservoir that was initially thought to be sand-rich and homogenous, 3D inversion suggests evidence of possible channels. This revelation could explain the varying thickness of the reservoir that was observed during drilling on the 1D UDR canvass. There are plans for future work to incorporate the observations and the analysis of the UDR products for deeper reservoir understanding of the field. Studies to include full integration with seismic data and production data would prove beneficial in well and reservoir management. Additionally, insights gleaned from the optimized selection of tool frequency for real time use and calibration with azimuthal dips and images proved invaluable especially in resolving unexpected structural and depositional complexities. The challenges in delineating fluid contacts in a structurally complex reservoir was also apparent with multiple realizations (and associated probabilities) of contacts seen from the real time results, which proved valuable in re-affirming the difficulties in characterizing the uncertainties in the field

In a typical waterflooding philosophy, water injection is often constrained by the in-situ stress of the overlying geological seal, in order to avoid out-of-zone injection and ensure reservoir fluid containment. Direct measurement of in-situ stress from wireline Micro-frac testing is insightful as compared to a conventional leak-off test (LOT) conducted at the casing shoe, due to the smaller volume and lower injection rate used. Unlike the conventional LOT, Micro-frac tests can be performed at discrete and multiple intervals in one wireline run, away from the casing shoe. In 2017, Shell Malaysia Exploration and Production (SMEP) conducted the first wireline Micro-frac test in a caprock shale, in a deepwater injector well, offshore Malaysia. The job was successfully executed with an improved design of the flowback mechanism, using a combination of a small volume pump and a drawdown chamber. While the small volume pump provided the advantage of accurate volume/rate measurement, the modified drawdown chamber setup provided a constant flow rate, to aid in a more accurate fracture closure pressure interpretation. The improved setup was the first successful application of this device for the purpose of Micro-frac. Both the methodologies showed good and comparable results, thereby adding higher confidence to the interpretation of fracture closure pressure. A workflow has been developed to ensure successful execution of the logging job. Pre-job modeling was completed using offset well data to derive formation mechanical properties as well as in-situ stress profile. This information was then used to characterize formation breakdown pressure, selection of test intervals and optimum straddle packer positioning. Learnings from the test conducted at the first depth station were used to optimize the number of cycles for the subsequent stations. In addition, an operational decision tree was developed to assess the number of cycles required and contingency criteria. The paper presents an improved flowback design using the drawdown chamber as well as the operational challenges in a deepwater environment. The methodology provided a reliable estimate of fracture closure pressure in a non-permeable formation. Results have been used to update the water injection envelope, in an effort to avoid undesired out-of-zone injection.

A case study on the integration of 4D seismic with various multi-scale and multidimensional field data to understand dynamic behaviour of the reservoirs is presented. 4D seismic is a key dataset amongst others e.g.geochemical fingerprinting, well inflow tracers, injection logging tools/production logging tools, and multi-well pressure deconvolution) together withconventional field data, which is acquired since starting up the field in late-2016. 4D data proved to be an essential piece that complemented field observations and is integral for constraining the subsurface models in support of a rapid second pahse of development and WRFM decisions. The paper describes the approaches taken to integrate these distinct datasets in the dynamic model as well as the various challenges faced in assimilating them in a coherent manner. One key subsurface challenge is to understand the degree of compartmentalisation risk to make sound WRFM decisions and to plan for a robust phase 2 development. As a starting point, conventional field performance analysis (production & injection performance) indicated connectivity across the reservoirs, though more limited in certain areas. This was supplemented with other subsurface data to further validate and improve the dynamic models. The 4D signals provided an indication of pressure and fluid connectivity as well as an indication of water sweep direction. Updates to the dynamic fault seal were performed in line with observations from 4D seismic and various field data. Understanding the dynamic behavior of the M field is key in view of the various challenges faced in reservoir management, e.g. increasing GOR trends and lower WI performance, in parallel with developing plans for the Phase 2 development. The incorporation of data of different scales and dimensions (4D seismic, fluid chemistry, PLT, multi-well pressure deconvolution) added value to the process of updating the dynamic models.