Evaluation of Horizontal Drilling

Modified from: in press, P.M. Gerlach, S. Bhattacharya, T.R. Carr, Cost Effective Techniques for the Independent Producer to Evaluate Horizontal Drilling Candidates in Mature Areas: AAPG Hedberg Conference, International Horizontal Well Symposium: Focus on the Reservoir, http://www.kgs.ku.edu/PRS/AAPG/papers/gerlach.html.

Horizontal wells are a cost efficient tool for reservoir management that has not been widely adopted by small independent operators of mature oil fields. Horizontal drilling has been extensively applied as an exploitation and exploration tool in relatively under-exploited reservoirs such as the Austin Chalk and in structurally complex reservoirs. In recent years horizontal technology has been extended to incremental oil recovery in the mature oil fields of southeast Saskatchewan. Though the technological needs in many mature onshore reservoirs are unique, the overall reservoir management objectives and requirements for commercial success are similar to those elsewhere. Application of horizontal drilling in Kansas has been limited to 28 wells. In Kansas results have been mixed with a few significant successes (Figure 2.11). Operator concerns for appropriate economic return, and difficulty in identifying candidate reservoirs have been the principal factors restricting application of horizontal drilling technology. Recent declines in cost factors have brought horizontal drilling technology within the economic reach of small independent producers. The remaining barrier to wider application of horizontal technology by the small independent is cost-effective approaches to target a horizontal well. We present several low-cost approaches that can be used to evaluate a potential horizontal well. These cost-effective screening techniques apply at the field scale, the lease level, and the well level. The techniques discussed enable the small independent producer to quickly and efficiently evaluate reservoir candidates, and predict performance of horizontal well application.

Kansas is a mature petroleum producing province with many marginal oil and gas fields operated by over 3,000 independent oil producers. As a result of operational and depositional-diagenetic heterogeneities most of these fields have recovery efficiencies of less than 30% original oil in place (OOIP). This low recovery efficiency results in significant remaining oil in place (ROIP). Operators can use horizontal technology to add new reserves by exploiting the ROIP in their existing fields, and to more efficiently recover known oil and gas reserves.

Operational heterogeneities are inherent in field development practices and results in significant ROIP. Examples of operational heterogeneities include inadequate drainage due to excessive well spacing, openhole/partial completions, bypassed attic oil, thin pays, and water coning. Depositional-diagenetic reservoir heterogeneities due to vertical and lateral variability of petrophysical properties create compartments in the reservoir. These types of heterogeneities are a function of original depositional architecture and the subsequent diagenetic overprint. Examples include highly variable pore geometry of carbonate rocks, anisotropic permeability in fractured reservoirs, and stratified flow units.

Cost-effective screening tools. The primary screening tool for identifying candidate reservoirs is "quick-look volumetric" calculations. This method uses only one well per unit area (e.g., quarter section) to identify pay height, porosity, and saturation to compute OOIP. These reservoir properties can be estimated from public domain data and computed using simple log analysis programs. PfEFFER, a low cost integrated log analysis tool developed by the Kansas Geological Survey, is used to identify well flow units, associated petrophysical and reservoir properties, and potential for production. Cumulative production per quarter section is then divided by OOIP to calculate recovery efficiency. The mapping of recovery efficiency across the field identifies target areas for further study. Regions with low recovery efficiency are those most likely to yield additional or incremental hydrocarbon reserves.

Detailed volumetric calculations at the lease level can be used to further evaluate potential target areas. Information from all wells on a lease is used to calculate and compare recovery efficiency between adjacent leases and adjacent wells. Mapping well production, normalized by petrophysical parameters, can approximate sweep efficiency between wells. Well production is normalized by dividing cumulative production by the product of payheight, hydrocarbon saturation, porosity, and horizontal permeability. NMR measurements on selected core plugs in reservoirs with significant micro-porosity are used as a cost-effective approach to separate total porosity from effective porosity. Permeability data can be estimated from well tests or by using porosity-permeability crossplots developed from core plug studies. Areas with low normalized production values suggest high reservoir heterogeneity and less effective sweep.

Following volumetric screening, the next step in the candidate selection process is identifying of the cause of poor recovery efficiency. The causes can be many-fold, but in Kansas the most prevalent are stratified thin pays, attic oil, excessive well spacing, coning due to strong water drive, and fractured reservoirs. ROIP in stratified thin pays can be identified by comparing initial rates of production and cumulative production between wells with similar payzone properties but different completion procedures. Attic oil is a common result of well spacing or lease boundaries coinciding with the structural axis of a payzone. First derivative maps show the change in the structural dip and can be used to identify the attics of a structure with undrained reserves. A simple method to determine excessive well spacing is to compare estimated ultimate production between primary vertical wells and infill vertical wells in an analog field. Analysis of lease total fluid production through time is a quick and cost effective method to suggest water break-through as a result of coning.

The final step in the candidate selection process is reservoir simulation to accurately identify ROIP on a grid cell by grid cell basis. Boast4, a freeware black oil simulator, was used to history match the performance of the Schaben Field, (Ness County, Kansas). Remaining hydrocarbon saturation-feet map (Figure 2.12) from this simulation was used to select areas with greatest potential for infill drilling. Boast VHS (vertical-horizontal-slant) simulation was used to predict and compare the performance of infill vertical and horizontal wells.

The Kansas Geological Survey has been working to develop and transfer cost-effective technologies to evaluate a potential horizontal target. We believe that small independents can successfully apply horizontal drilling technology to recover additional oil and gas in mature areas. Our approach recommends low-cost techniques to understand reservoir heterogeneity, to evaluate recovery potential at the field, lease and well scales, and to characterize and simulate candidate reservoirs.

Synopsis of PfEFFER

The petrophysical analysis and reservoir evaluation computer package (PfEFFER) was enhanced in conjunction with the Class 2 project. PfEFFER version 2.0 and PfEFFER Pro were released in February 1998. Prototype software was tested and successfully applied in Schaben Field. PfEFFER stands for "Petrofacies Evaluation of Formations For Engineering Reservoirs" (Doveton and others, 1995).

The minimum log data required by the spreadsheet-based software are a porosity and resistivity log. Old logs are well suited to this analysis once they are digitized or simply typed into the spreadsheet. Toolbars and menus perform most operations through the utilization of nearly 8,000 lines of Visual Basic code. PfEFFER v. 1 reads standard LAS log data files such as those obtained from a logging truck or permits manual entry, organizes digital data by well and zone, and creates a "Super Pickett" crossplot, depth plots, and lithology solutions (if sufficient logs are available (Figure 2.13).

The software is focused on interpreting and analyzing reservoir pore type, permeability trends, and variations in mineral composition. PfEFFER provides procedures for optimal estimation of bulk volume water and water saturation (including irreducible values) to better evaluate potential production, reservoir quality, and heterogeneity. Also capillary pressure data can be incorporated to further calibrate well log data with pore size or to assess depth to the free water level. The program will assemble zonal information from well workbooks into a project workbook and can automatically generate map and 3-D visualizations of key parameters as defined by the user. "Hot links" are maintained in the project workbook to each well workbook to aid in data management (Figure 2.14).

All of the standard EXCEL features continue to be available to users for independent analysis and data exploration. The simplicity of hardware and software requirements means that PfEFFER is an attractive option for companies of all sizes. The range and versatility of module capabilities makes them powerful tools for the analysis of both old log suites and the latest generation of logging measurements.

In addition to revising the appearance of the spreadsheet and refining the modules as described above, PfEFFER v. 2.0 contains new modules. The new features in PfEFFER v. 2.0 are:

1. Vshale (shale proportion) can be calculated, based on either the gamma ray or the neutron and density porosity logs.
2. Porosity can be calculated using density, neutron, density/neutron, or sonic with and without correction for shale volume.
3. Shaly sand models are available for Sw calculation. Sw model menu permit selection of Archie water saturation model (the default) and two shaly sand models, the Simandoux model and the dual-water model.
4. Hough Transform is included. The Hough transform is used for simultaneous solution of Archie equation constants and formation water resistivity.
5. Secondary porosity is calculated as the difference between the total porosity (from density or neutron porosity) minus sonic porosity.
6. Moveable hydrocarbons can be determined (Figure 2.15). Based on the assumption that the zone near the well is permeated with mud filtrate, the Archie equation is used to compute Sxo, the filtrate saturation of this flushed zone. Sxo is used to compute moveable and residual hydrocarbon saturations. PfEFFER generates a moveable oil plot consisting of bulk volume water, bulk volume fluid, and porosity. Difference between BVF and BVW represents the moveable hydrocarbon saturation.
7. Lithological analysis now includes two options in PfEFFER v. 2.0, one based on the RHOMAA-UMAA plot and the other allowing a more general selection of logs and system components. The general option allows user to compute up to six components based on up to five logs. Any log can be employed and the component selection is at the discretion of the user.
8. Depth-constrained multivariate cluster analysis can be employed to segment the entire spreadsheet into subintervals based on user-specified set of logs. A hierarchical cluster (Ward's method) is used to produce subintervals that are as homogeneous as possible and distinct as possible from each other, in terms of their log characteristics. Option is useful in evaluating flow units and can be used as a blocking function.
9. Forward modeling module implements equations to predict values of rx, capillary pressure, and hydrocarbon column height for a range of water saturation values based on specific values of permeability and porosity.
10. Pay flag cuttoff can be activated to add color to cells of selected variables used to determine pay (porosity, BVW, Sw, and Vsh) and to color cells in the pay column according to pay and non-pay intervals.

Three additional modules (add-ins) are available as PfEFFER Pro. These modules include color cross section generation, map coordinate conversion (longitude-latitude to UTM x-y), and software to help build an input file for a reservoir simulator based on the petrophysical characterization. DOE's freeware reservoir simulation software, BOAST 3, was used in the development and testing.