Summary The need to increase productivity and to reduce drilling formation damage favors the use of underbalanced drilling technology. The main idea is to drill with equivalent circulating densities (ECD) that are less than the formation pore pressure and to avoid contact between the drilling fluid and the formation. In highly depleted reservoirs, pore pressures can be very low. Therefore, extremely low-density fluids, such as foams, are used to achieve circulating densities lower than the pore pressure. In such cases, the induced modification of the in-situ stresses has to be supported mainly by the rock, with little contribution from the drilling fluid pressure. In that sense, the application of underbalanced drilling depends on the mechanical stability of the drilled formation, among other factors. In general, poorly consolidated, depleted formations are not suited for that technology. This paper presents the wellbore stability simulation performed to establish the feasibility of using underbalanced drilling in highly depleted reservoirs in western Venezuela. The in-situ stress field and the mechanical properties of the formation were obtained. Pore pressure as low as 800 psi at 5,500 ft (2.7 lb/gal equivalent fluid density) was measured. The finite difference method and an elastoplastic constitutive model was used to obtain the new stress, deformation, and pore pressure distribution. The undrained condition (immediately after the wellbore is drilled) as well as the drained condition were analyzed. The analysis showed that horizontal wells could be drilled in an underbalanced condition with low instability risk. Following the recommendations, four horizontal wells were drilled in underbalanced conditions. Values as low as 2.0 lb/gal ECD were used to drill the wells, and no wellbore instability problems were reported. Production tests showed an enormous increase in the well productivity index in comparison with conventional overbalanced drilling. Introduction Years of crude exploitation can lead to enormous reservoir pressure decline in mature fields, leaving a huge amount of oil still in place that can only be exploited with new technologies. Once a mature field has reached an extremely low pore pressure, the target formation should be drilled in an underbalanced condition (with a drilling fluid pressure of less than the formation pore pressure) to reduce the risk of lost circulation, and most importantly, to reduce formation damage and increase productivity. Pore pressures as low as 800 psi at 5,500 ft true vertical depth (TVD) have been measured in matured fields. This is equivalent to a hydrostatic column of drilling fluid of 2.7 lb/gal density. To reach the underbalanced drilling condition in such a depleted reservoir, the ECD of the drilling fluid must be less than 2.7 lb/gal. Conventional drilling fluids are out of this range; therefore, the use of low-density drilling fluids, such as foams, is needed. Drilling fluids must provide good cuttings transport, among other things. In addition, when the hole is drilled, the in-situ original stresses change near the wellbore, and the drilling fluid pressure (or density) must somehow "replace" the support lost by removing the original volume of rock. In conventional overbalanced drilling, the drilling fluid pressure (or density) is usually high enough to provide this support. On the other hand, it is necessary to establish whether low-density fluids used for underbalanced drilling in highly depleted reservoirs will provide the pressure needed by the formation to keep it stable. If not, underbalanced drilling is not technically possible, and the minimum drilling fluid density (and pressure) will be established by the formation minimum collapse pressure obtained with rock-mechanics approaches. The main scope of this paper is to present the wellbore stability simulations performed in a highly depleted reservoir in Lake Maracaibo, Venezuela, to decide the feasibility of using underbalanced drilling technology. The field behavior of four wells drilled in underbalanced conditions is summarized. Benefits such as increase of rate of penetration, no drilling time lost because of lost circulation or stuck pipe, and most importantly, huge productivity increases were obtained in comparison with conventional overbalanced drilling. Background and Technological Challenge Many horizontal wells in a highly depleted reservoir in western Venezuela were drilled in an overbalanced and near-balanced condition, at pressures greater than the in-situ reservoir pore pressure. Lost-circulation problems and stuck pipes raised drilling costs, reducing the project's profitability. In addition, the expected productivity increases were seldom reached. Therefore, real underbalanced drilling technology was proposed to increase reservoir productivity. The challenge for the rock-mechanics team was to define the risk of wellbore instability when drilling in an underbalanced condition with ECDs as low as 2.0 lb/gal. In such a condition, the rock must be sufficiently strong to resist most of the redistributed stresses. Technology implementation was decided based on wellbore stability calculations and results. Rock Mechanics Data Wellbore stability in sandstone depends on many factors, such as wellbore geometry (azimuth, inclination, and diameter), current formation pore pressure, original in-situ field stress (magnitude and direction), geomechanical properties of the rock (including its geological particularities), and the ECD. Well geometry and drilling fluid density may be selected to reduce wellbore stability problems, such as high deformations, wellbore breakouts, washouts, and lost circulation caused by induced drilling fractures. Acoustic image logs, multipolar sonic logs, six arm-oriented caliper logs, density logs, minifrac tests, and rock mechanical laboratory tests on rock cores were used to obtain the required information for stability analysis.

In the Cantarell field in Mexico oil is produced primarily from naturally fractured carbonate reservoirs. These types of reservoirs are among the most difficult to develop given the uncertainties associated with detection and characterization, both locally in the well and regionally in the field, of the fracture networks that predominantly control fluid flow. Additional complexities result from difficulties understanding matrix-fracture interactions.In the last few years, the oil-water contact in the Cantarell field has been advancing, and this has considerably reduced the oil window zone. One likely explanation is that the natural fracture network (which provides most of the permeability in the field) favors production of water and gas over oil. This new challenge has forced Petróleos Mexicanos (PEMEX E&P) to make new efforts to characterize their naturally fractured carbonate reservoirs and improve their reservoir models. Using the critically-stressed-fault hypothesis, which assumes that the faults and fractures that are hydrologically conductive today are those that are critically stressed (active) in the current stress field, GeoMechanics International (GMI) analyzed image logs of over 20 wells in detail, and constructed a geomechanical model of the Cantarell field, to define the role of the fracture network in fluid flow. In a previous study, GMI used all available data, including seismic data, caliper logs, image logs, electric logs, drilling reports, sedimentological analyses of cores, and regional tectonic studies, to develop a comprehensive geomechanical model of the field. We then used and complement this model in conjunction with all available image logs to predict the orientations of the most and least critically stressed (hence permeable and impermeable) fractures. The results have been effectively used to improve well trajectories both to minimize costs and to optimize production throughout the life of the reservoir. The results have been particularly useful in horizontal wells, where preventing water production requires avoiding steeply dipping fractures that are critically stressed.

TX 75083-3836, U.S.A., fax 01-972-952-9435. AbstractPetróleos Mexicanos (PEMEX E&P) and GeoMechanics International Inc. (GMI) worked together to define the potential for fault leakage in the Sihil and Frontal faults of the Cantarell and Sihil Fields, Mexico. Water production with a low salinity content in the northern area of the Akal block in the Field seems to be derived from neighboring blocks of Tertiary age and not from the water leg of the reservoir. Pore pressures in the reservoirs from the Akal block have depleted significantly since production started in the late seventies. Nitrogen injection started in the late nineties and has contributed to pressure maintenance since then. Overpressures in the shale layers above the reservoirs have been identified. A number of studies have shown that shear failure along reservoir-bounding faults can increase fault permeability and compromise fault trap integrity. The orientation of a fault and the magnitudes of the present-day stresses and pore pressure acting on the fault will determine whether the fault has the potential for shear failure and therefore a potential for leaking. The analysis of fault leakage potential was based on a geomechanical model covering the northern area of the Cantarell Field and its changes with time due to reservoir depletion and nitrogen injection. Information from wireline logs, downhole tests and drilling events from eight wells drilled between 1979 and 2005 were used to build the geomechanical models that change with time. The principal effective stress model (the magnitudes and orientation of the principal stresses), and the mechanical properties model were defined and validated with image logs, caliper logs and drilling experiences before their use for fault leakage potential analysis. Specific sectors of the two faults appear to be tectonically active during the production history and therefore had potential to leak. Fault leakage potential may explain water production with low salinity content in the northern area of the Akal Block in the Cantarell field.

Electric logs and borehole imaging were used to build a geomechanical model to predict wellbore instability during drilling and to optimize casing designs and completions to control sand production in the Burgos Basin. This paper describes how geomechanical models can explain several problems in drilling and completion during the development of some of the fields in Northern Mexico. The total stress tensor model, the pore pressure model, and the mechanical properties model are discussed. Validation of the model is a critical step before it can be applied to design new wells. The geomechanical model was used to design a new well and to optimize drilling, completion, and production techniques for the reservoir. The model was used to optimize the placement and orientation of perforations, evaluate proposed hydraulic fracturing designs, and select the critical draw-down to produce a well. Actual drilling, completion, and production events for the new well were analyzed, and conclusions about how the geomechanical model supported the complete design are also included in this paper. Introduction Several studies worldwide have confirmed that the optimization of non productive time related to borehole instability while drilling, the determination of the preferential permeability in naturally fractured reservoirs, the integrity of seals in geological faults, sand production potential, reservoir compaction and casing damage, are controlled by the geostatic in situ stresses, the pressure of the fluids in the pore spaces and the mechanical properties of the rocks. A geomechanical model for the field was built based on information from conventional and electric imaging logs in addition to the drilling information of the first exploratory well, F-1, drilled in the field. The model was validated predicting breakouts in the F-1 well and comparing them with rock failures observed in the caliper log, breakouts observed in the electric imaging log and with drilling problems related to wellbore instabilities. The geomechanical model for the field predicts well the observations and problems identified in the F-1 well during the drilling, completion (hydraulic fracture) and initial production of the well. A wellbore stability analysis during drilling and completion was performed for the F-1 well to predict the mud weight requirements during drilling to prevent large breakouts/washouts or lost of circulation problems and the required borehole pressure to prevent sand or formation flowing during the production phase. Furthermore, the lessons learned from F-1 were successfully applied to drill, stimulate and produce a new well, F-101. Geomechanical Model The project started collecting all the information related to planning, drilling and completion of the F-1 well. This well was drilled as a directional J-shape well to reach the final objective in the Jackson Formation, as presented in Figure 1. General information of the Field was reviewed (including surface seismic and maps), the structural model, standard and advanced wireline logs including dipole sonic and images acquired in the F-1 well, drilling and completion information (including daily reports, mud weight reports, well design schematic, leak off tests and Minifrac tests), pore pressure measurements in the reservoir, well geometry and formation tops.

The need to increase productivity reducing drilling formation damage favours the use of underbalanced drilling technology. The main idea is to drill with equivalent circulating densities (ECD) below the formation pore pressure and to avoid the contact between the drilling fluid and the formation. In highly depleted reservoirs, pore pressures can be very low. Therefore, extremely low density fluids such as foams are used to achieve circulating densities below pore pressure. In such cases, the induced modification of the in situ stresses has to be supported mainly by the rock, with low contribution of the drilling fluid pressure. In that sense, the application of underbalanced drilling depends, among other factors, on the mechanical stability of the drilled formation. In general, poorly consolidated depleted formations are not suited for that technology. This paper presents the wellbore stability simulation performed in order to establish the feasibility of using underbalanced drilling in highly depleted reservoirs in western Venezuela. The in situ stress field and the mechanical properties of the formation were obtained. Pore pressure as low as 800 psi at 5500 ft (2.7 lb/gal equivalent fluid density) was measured. The finite difference method and an elastoplastic constitutive model was used to obtain the new stress, deformation and pore pressure distribution. The undrained condition (immediately after the wellbore is drilled) as well as the drained condition were analysed. The analysis showed that horizontal wells could be drilled in an underbalanced condition with low instability risk. Following the recommendations, four horizontal wells were drilled in underbalanced conditions. Values as low as 2.0 lb/gal ECD were used to drill the wells and no wellbore instability problems were reported. Production tests showed an enormous increment in well productivity index in comparison with conventional overbalanced drilling. Introduction Years of crude exploitation can lead to enormous reservoir pressure decline in mature fields, leaving huge amount of oil still in place that can only be exploited with new technologies. Once a mature field has reached an extremely low pore pressure, the target formation should be drilled in an underbalanced condition (using a drilling fluid pressure below the formation pore pressure) to reduce the risk of lost of circulation, but most important, to reduce formation damage and increase productivity. Pore pressures as low as 800 psi at 5500 ft (TVD) has been measured in matured fields. This is equivalent to an hydrostatic column of drilling fluid of 2.7 lb/gal density. To reach the underbalanced drilling condition in such a depleted reservoir, the equivalent circulating density (ECD) of the drilling fluid must be less than 2.7 lb/gal. Conventional drilling fluids are out of this range, therefore the use of low density drilling fluids such as foams is needed. Drilling fluids must, among others, provide good cuttings transport. In addition, when the hole is drilled, the in situ original stresses change near the wellbore and the drilling fluid pressure (or density) must "replace" somehow the support lost by removal the original volume of rock. In conventional overbalanced drilling, the drilling fluid pressure (or density) is usually high enough to provide this support. On the other hand, it is necessary to establish whether low density fluids used for underbalanced drilling in highly depleted reservoirs will provide the pressure needed by the formation to keep it stable. If not, underbalanced drilling is not technically possible and the minimum drilling fluid density (and pressure) will be established by the formation minimum collapse pressure obtained with rock mechanics approaches.