Overcoming real-time LWD limitations of longest horizontal drilling: A unique experience from Abu Dhabi

NASHAT ABBAS, ADNOC; and CHANDRAMANI SHRIVASTAVA, SLB

INTRODUCTION

Due to the ever-present need to reduce operational costs and increase performance while reaching deeper formation targets, complex well profiles are often necessary. Real-time data have become critical, with operations getting more sophisticated and autonomous. High-fidelity real-time data enable real-time decisions that can alter the course of operations. The data needed ranges from simple measurement-while-drilling (MWD) to advanced logging-while-drilling (LWD), depending on the operation's complexity.

To develop a large carbonate field offshore Abu Dhabi, four environmentally friendly islands were constructed in shallow waters. To place multilateral wells in 18-ft zones, 250 m apart (Fig. 1), the pay zone's geological characterization is essential, as well as of the thick argillaceous sequences above and below it. The target is the central portion with the best porosity, avoiding the unproductive upper and lower layers marked by stylolites. Most wells required extended reach drilling (ERD), due to the vast areal extent of producing layers. This article presents the case study from the longest well drilled in the UAE.

Fig. 1. Well location with 250-m spacing within the study area.

Planning such an ERD well with maximum reservoir contact begins well before execution. Placing and steering the lateral section in a multilayered anticlinal reservoir with significant geological uncertainties posed major challenges, due to the complex carbonate facies heterogeneities, alongside the difficulty of transmitting signals to the surface.

Real-time GR, density/porosity, and azimuthal logs helped steer the wells and detect vertical changes. Borehole image logs helped avoid severe undulation when discontinuities were encountered.

A detailed study of the reservoir behavior through nearby pre-existing wells was conducted, along with understanding the geological structure of the area, since the target zone comprised an 18-ft limestone layer overlying a 3-ft stylolite layer with very poor permeability. An extended reach, maximum reservoir contact (MRC) well was drilled in the low-permeability, western region of this giant offshore oil field. The objective was to develop the western region's reservoir productivity and injectivity capabilities

With a horizontal section exceeding 28.5K ft, significant challenges related to the capability of the drilling tool and maintaining a smooth well path emerged. Critical success factors included navigating within good reservoir quality over long intervals without penetrating low-quality rock, where LWD services were essential to stay in the sweet spot of the reservoir. Additionally, to account for potential lateral facies changes, the sensors needed to be positioned as close to the drill bit as possible, to satisfy the need for real-time image-log data to guide the trajectory through the best zones, using dips from real-time images.

The drilling team faced concerns about using nuclear source density tools to navigate without creating high dogleg severity, which could potentially lead to fishing operations, especially in fractured areas. The well was drilled to a record depth of more than 52,000 ft, with a horizontal section of 28,670 ft within the reservoir. However, such long wells face limited telemetry on mud-pulse, due to data-heavy logging tools on the drilling bottomhole assembly (BHA), signal attenuation, and various drilling-related events. Factors like higher mud weight, hole cleaning efficiency, rig pumps, doppler echoes, and others affect the signal. Legacy telemetry platforms often fail to provide the quality and fidelity of LWD logs needed for these extreme drilling conditions, as was realized in the previous section drilled.

GEOLOGY OF THE STUDY AREA

Fig. 2. Stratigraphy of the study area (after Gamal, 2000).

The study area lies in a prolific oil field, offshore Abu Dhabi. It’s a large anticlinal trap, with reservoir properties varying along the flanks. The field is a large, low-relief anticline that runs east-west, and is characterized by gentle northerly dips in lower Cretaceous carbonate reservoirs. The field is comprised of three major zones (I, II, and III) overlying each other. These zones are further subdivided into many producing reservoirs, Fig. 2.

Typical response of petrophysical properties (Fig. 3) through the reservoir layers is illustrated to understand the subsurface characterization. The thick (>100 ft) limestone reservoir of interest in this study is subdivided into six porous layers, labeled A to F, each separated by a thin stylolite, low porosity layer.

Exploitation of the reservoirs is based on segregated production and injection; commingling between reservoirs is prohibited to optimize recovery in producers and sweep efficiency in injectors. Horizontal drilling was introduced in 1989, and an extensive horizontal drilling application has been implemented, using steerable drilling. With time, extended reach wells of tens of thousands of lateral feet were started drilling, while trying to keep the well trajectory through the best of the reservoir facies.

Fig. 3. Reservoir layers A, B, C, D, E & F are separated by thin, lower-porosity stylolite.

Extended reach wells have their own sets of challenges through these heterogeneous carbonates. This study analyzes how availability of high-fidelity LWD data at greatest recorded measured depths, enabled by advances in a drilling data acquisition platform, can help achieve the well objectives.

CHALLENGE

Geosteering of the longest horizontal well required innovative solutions and best practices to overcome several technical and operational challenges: 

• Directional control: Ensuring that the wellbore stayed within the target zone.  
• Geological hazards: Avoiding several geological hazards that could jeopardize the drilling operation. 
• Drilling challenges: Maintaining parameters within rig capabilities whilst transferring sufficient energy to the bit to optimize drilling performance, to achieve the best penetration rate to minimize the drilling time and cost. 
• Friction and torque: Managing the friction and torque that resulted from the long horizontal section and the high inclination angle.  
• Wellbore stability: Ensuring the wellbore stability

DRILLING OPERATIONS INFORMATION

The study well’s objective was to accurately place it within the target zone by using real-time LWD data and geosteering techniques. The correct well placement was essential to achieving optimal production performance, by maximizing the reservoir exposure and drainage area and by minimizing the water breakthrough. The well placement also had to account for the reservoir heterogeneity and anisotropy, which could affect the well’s performance. To overcome the challenges and achieve the objectives of drilling the world’s longest horizontal well, several innovative solutions and best practices were deployed.

Delivering consistent performance with reduction of operational costs while reaching deeper formation targets leads to complex well profiles. The importance of real-time data is becoming even more critical, as drilling operations are getting more sophisticated and autonomous. Availability of high-fidelity real-time data is of paramount importance to enable real-time decisions, often changing and improving the course of operations while avoiding remedial activities and more consequential costs.

We used a comprehensive geological and geophysical (G&G) study, which included seismic, well log, and core data, to identify and map the geological hazards along the well path. We also used a pressure-while-drilling (PWD) system, which allowed us to control the annular pressure and the mud weight to prevent fluid losses and blowouts. 

Historically, extended reach wells experience difficult drilling-related events, such as torque and drag, pressure management, mud weight changes, and hole cleaning efficiency, with ultra-long trajectories requiring delicate management of tortuosity. There is even more complexity when the ultra-long trajectory requires a perfect understanding of the reservoir to be able to take decisions on the spot; in all those conditions, it is key to have a reliable real-time data transmission.

Currently, legacy and new telemetry options (prior to this study) cannot provide reliable, high-quality uninterrupted LWD logs (including borehole images) that are needed to efficiently drill the well to TD under these extreme drilling conditions. There exists the need for reliable measurements enhancing efficient and accurate drilling operations under all possible conditions. In addition, there is the need for an intelligent system that can transfer required data at faster rates, especially in harsh drilling environments, to enable optimal placement of the well trajectory through intended geological targets.

Fig. 4. Drilling BHA used in the final TD section of the well.

At the same time, continuity and integrity of data transmission needs to be maintained to avoid any potential pulling out-of-hole (POOH) scenarios.  To address this need and gap in technology, a new telemetry platform was deployed, built on intelligent optimization of signal processing that utilizes machine learning algorithms to enable real-time data in challenging drilling conditions.  

A comprehensive drilling BHA was prepared with required drilling tools and LWD services, Fig. 4. A multi-function, compact formation, evaluation tool (EcoScope™) was used to provide the petrophysical data of gamma ray, neutron porosity, density porosity, phase and attenuation array resistivities for different depths of investigation, density images and spectroscopy data. The real-time data transmission frames were prepared with enough data points (d-points) of gamma ray, neutron/ density and some resistivity channels (both phase/ attenuation) including the density images for geosteering. The planning phase included the simulation of data quality with high ROP at the required depths under the provided mud-conditions.

MINIMIZING BOREHOLE UNDULATIONS

Fig. 5. Real-time LWD density image comparison, with legacy RT images on the left and a new telemetry-streamed RT image on the right. Both images are oriented to the top of hole, and the snapshot illustrates 100 ft of data at 1:200 scale. White arrows illustrate the high-definition feature picking on density images enabled by new telemetry system.

To facilitate running of the reperforated, 6 5/8-in. uncemented liner in the 8.5-in. lateral, a smooth borehole and less than 1^o directional change per 100 ft of hole length were desired. To achieve this, a state-of-the-art, point-the-bit, rotary steerable system (RSS), directional drilling tool was employed. The bottomhole drilling assembly rotates while performing accurate directional control. Compared to drilling motors, RSS systems deliver a smoother borehole, with minimum dogleg, as there are no sliding intervals. In addition, the RSS system employed has continuous directional and inclination (D&I) measurements 14 ft from the drill bit, facilitating prompt and agile control of inclination, as requested by the geosteering team with a 0.1^o angle adjustment accuracy.

The downhole drilling assembly does not immediately respond to commands, and there is a delay between the drill bit penetrating the rock and the geosteering team receiving measurements on the type of rock penetrated. The larger the distance between the rock lithology sensors and the drill bit, the longer the delay. It’s only the anticipatory skills of the geosteering team that prevent the borehole drifting off into undesirable facies, in this case the very-low-permeability limestones. The multifunction formation evaluation tool addresses the distance to sensor problems successfully in the intelligently programmed BHA.

The 6¾-in. robust M/LWD new-generation service was deployed successfully while drilling with OBM, and it was drilled in an 8 ½-in horizontal section

DISCUSSIONS AND EXAMPLES

Deployment of a new telemetry platform, built on intelligent optimization of signal processing with embedded artificial intelligence (AI), provides the required service by achieving richer data density at higher ROP. This results in high-quality well logs made available in real time with intelligent automation that uses machine learning to optimize the planning.

Fig. 6. Real-time LWD data representation, showing high quality of RT images and petrophysical well-logs. Track 1 = depth (4,000 ft at 1:4000), Track 2 = phase & attenuation resistivities, Track 3 = gamma ray, Track 4 = neutron/ density, Track 5 = density image (static), Track 6 = density image (dynamic) with dip sinusoids, Track 7 = dip track. Borehole images are oriented to the top of hole.

It achieved unprecedented high-quality data transmission capability in the longest-ever-drilled, extended reach well in Abu Dhabi (>52,000 ft, MD), using the conventional triple combo data and density images to steer the well. Optimal data density was delivered in real time, aided by smart planning to execution on the new telemetry platform, coupled with the advanced MWD system.

An example (Fig. 5) shows the comparison of legacy real-time density image with a new telemetry-streamed real-time density image. This clearly shows that confident and crisper features with higher-quality data are acquired with the help of new system. The features shown with white arrows clearly illustrate the difference; it is to be noted that the ambiguity in feature-picking can lead to big differences in picked dips, both for azimuth and magnitude.

An example (Fig. 6) shows the high-quality real-time data made available by the innovative advanced telemetry system. The data are shown at 1:4000 scale for around 4,000 ft towards the end of the total depth. Gamma-ray, resistivities and neutron/ density all show high-fidelity data for real-time interpretation. The most striking feature is the quality of density image, where images show the sub-horizontal facies boundaries picked with confident dips in real-time data. The frowning and smiley-faces sinusoids placed on the dynamically normalized images confidently pick the facies variation, where the well trajectory goes stratigraphically down into slightly dense-facies (picked aptly on the static RT density image and neutron-density) and then exits, going stratigraphically upwards.

Fig. 7. Comparison of real-time (RT) LWD data (left to the depth-track) and memory recorded mode (RM) data (right to the depth track), showing high quality of well-logs over 4,000 ft at vertical scale of 1:4000, with density images oriented to the top of hole.

Another example (Fig. 7) shows the validation of these real-time data with recorded mode memory data after the tool was out of the hole. The quality of real-time data stands out at these extreme conditions, compared to the memory-retrieved high-quality data. The data displayed left to the depth-track is real-time, and to the right is recorded mode. High fidelity of RT data is evident in comparison to the RM data for both image logs and petrophysical triple-combo curves (GR, density/ neutron, resistivity).

An efficient way to assess the overall real-time data quality provided by the new telemetry system is to compare the RT density images acquired with the RM density images over the interval of logging. In this case, density images were streamed in real time for geosteering-related decisions. A comparative illustration of the real-time and recorded mode density images would be a good indicator of the confidence to be put in the new telemetry system. An example in Fig. 8 presents ~10,000 ft of density images through the horizontal section, compressed at 1:10,000 scale to cover the overall facies variations. Comparison of dynamically normalized density images of RT and RM across the depth track illustrates the obvious good-quality data made available in real time for steering the well.

Fig. 8. Real-time LWD data representation, showing high quality of RT images and RM images. Track definition from bottom to top: Track-1: statically normalized RT density image, Track-2: dynamically normalized RT density image, Track-3: depth track showing 10,000 ft data compressed at 1:10000 scale, Track-4: dynamically normalized RM density image, Track-5: statically normalized RM density image. All the image logs are oriented to the top of hole.

CONCLUSION

High-quality real-time LWD data were made available in a record-ERD well that was TD’d upwards of 52,000 ft. The real-time density image acquired with a multi-functional compact formation evaluation service provided facies boundary and their dips to steer with confidence. This was made possible with a recently deployed, new telemetry platform, coupled with MWD that provided high-quality data at those greater depths with high ROP. This opens avenues for further extending the limits of well operations, reducing time and operating cost. The well achieved an average 98 ft/hr., which was 30% higher than the previous wells in the field.

Geosteering and placement of the longest horizontal well in the world, enabled by innovative MWD and a new telemetry platform for LWD real-time data transmission, resulted in several benefits, which included:

• Increased productivity and recovery per well: The well achieved a record horizontal section of ~35,000 ft, including ~11,000 ft of last drilling run with innovative MWD and telemetry system. This resulted in increasing the reservoir contact and drainage area of the well by more than 35%, compared to previous wells drilled in the same reservoir. The resultant net uplift gained by well lateral extension is ~3,000 STBOPD obtained from the tightest part of the reservoir. Also, placing the well accurately within the target zone is expected to maximize vertical sweep efficiency and, therefore, increase ultimate well recovery by ~5%. The innovative successful geosteering of this well has unlocked many opportunities from a field development perspective, such as increasing and accelerating the recovery of oil-in-place (OIP) while reducing well count.
• Reduced environmental impact: Extending MRC laterals in tight carbonates leads to a more sustainable way of development by reducing footprint required and maximizing oil production per well, leading to reduction in CO₂ production associated with operational activities.

ACKNOWLEDGEMENTS

The authors acknowledge the support of ADNOC management and their approval to present this work. They also thank SLB for technical discussions leading to preparation of this article, which is derived from SPE paper 222798-MS, presented at ADIPEC, Abu Dhabi, UAE, November 2024.

REFERENCES

Gamal Ismail, A. S. Fada'q, S. Kikuchi,H. El Khatib, “Ten years’ experience in horizontal application & pushing the limits of well construction approach in Upper Zakum field (offshore Abu Dhabi),” Abu Dhabi International Petroleum Exhibition and Conference, 2000.

NASHAT ABBAS is a widely published, distinguished geologist with ADNOC Offshore. He is an expert in well operations and geosteering and has steered the world’s top five ultra-reach longest wells. Mr. Abbas holds degrees in geology and geophysics, with a master’s degree in Petrophysics. With over 35 years of experience, starting from assistant lecturer in Egypt to almost 30 years with ADNOC, Mr. Abbas is highly regarded in the technical community of the MENA region.

CHANDRAMANI SHRIVASTAVA is a geology advisor at SLB. He holds master’s degrees in applied geology and petroleum engineering. He has worked in India, the Middle East, Southeast Asia, West Africa and the U.S. in various subsurface technical roles. Mr. Shrivastava is widely published and has been an SPE Distinguished Lecturer. He also holds key positions in international professional societies. He is working on developing geology-while-drilling automation workflows with real-time answer products.