Reserves, however estimated, rely on the integrity, skill, and judgment of the evaluator and are affected by the geological complexity, stage of development, degree of depletion of the reservoirs, and amount of available data. This article presents a discussion of the various reserve estimation methodologies as contained in SPE’s website (www.spe.org). The examples provided for each of the methodologies are from my own work. The field that was used is an incised valley sequence with minimal fault complications. There are a total of 18 penetrations in the field and several additional wells outside of the incised valley to constrain the valley limits. The pay area is amplitude-supported, and the observed water level matches the RFT-predicted water-level, so there is minimal uncertainty.
The acceptable methods for estimating reserves include (i) the volumetric method; (ii) evaluation of the performance history, which evaluation may include an analysis and projection of producing ranges, reservoir pressures, oil-water ratios, gas-oil ratios and gas-liquid ratios; (iii) development of a mathematical model through consideration of material balance and computer simulation techniques; (iv) analogy to other reservoirs if geographic location, formation characteristics or similar factors render such analogy appropriate. In estimating reserves, Reserve Estimators should utilize the particular methods, and the number of methods, which in their professional judgment are most appropriate.
Volumetric Methodologies
Estimating reserves in accordance with the volumetric method involves estimation of oil or gas in place based upon review and analysis of such documents and information as (i) ownership and development maps; (ii) geophysical and geologic maps, including structure and TVT net pay; (iii) electric logs and formation tests;(iv) relevant reservoir and core data; and (v) information regarding the completion of oil and gas wells and any production performance thereof. Appropriately-estimated recovery efficiency must be applied to the resulting in place volume in order to derive estimated reserves.
In deterministic volumetric methodologies a single best estimate of reserves is made based on known geological, engineer-ing, and economic data. This is an acceptable methodology when there is sufficient data available to minimize the uncertainty.
The method of volumetric estimation is probabilistic when the known geological, engineering, and economic data are used to generate a range of estimates and their associated probabilities. Probabilistic methodologies can either be Monte Carlo (Figure 2) using multiple random calculations to determine the probability distribution, or a true probability method that utilizes probability statistics to define the mean. (Figure 3).
Performance Data Analysis
For reservoirs for which performance has disclosed reliable production trends, reserves may be estimated by analysis of performance histories and projections of such trends. These estimates may be primarily predicated on an analysis of the rates of decline in production (Figure 4) and on appropriate consideration of other performance parameters such as reservoir pressures, oil-water ratios, gas-oil ratios and gas-liquid ratios. In the example shown in Figure 4, the Decline Curve Analysis yields an estimated ultimate recovery of 25.5 MMstb. With this field’s typical recovery efficiencies of 30 to 35%, the STOIIP calculated by this method ranges from 73 to 85 MMstb.
Mathematical Models
Reserves and future production performance can be estimated through a combination of detailed geologic and reservoir engineering studies and mathematical (Material Balance) or computer simulation models (3D Dynamic Modeling). The validity of the mathematical simulation models is enhanced by the degree to which the calculated history matches the performance history. Where performance history is unavailable, special consideration should be given to determining the sensitivity of the calculated ultimate recoveries to the data that is the most uncertain. After making such sensitivity determination, the ultimate recovery should be based on computed results using a combination of input parameters appropriate for the class of reserves assigned. In the example shown in Figure 5, the material balance analysis resulted in a STOIIP of 88 MMstb. The geomodel (Figure 6) yielded a STOIIP of 79 MMstb.
Summary
As one can see in the Table below, there is a range of 14% from the lowest volume estimate (DCA) to the highest volume estimate (MBAL), although in general, the various methods yielded similar volume ranges for the field, with the true probabilistic method capturing the widest range of uncertainty.
The differences in the volume estimation resulting from the various methods reflect the inherent uncertainty in estimating volumes, even with reasonably plentiful data. It should therefore be considered good practice to conduct volume estimates with sever-al methodologies.