General

This program allows the addition of reserve estimates as probabilistic distributions with dependence.
Further facilities are the volumetric calculation of trap capacity for a single reservoir in a single trap (VOLCAL), the addition of a number of identical (hypothetical) traps (NIEC) and the merging of a number of expectation curves (MERGE).
A number of statistics as well as expectation curves form the output.

Input

The input screen for an AddRes option (addition) is shown below:

Addition - Summing of expectation curves of different reservoirs or traps

The individual prospect expectation curves are represented by a vector of 10,000 elements for the Monte Carlo addition. For each prospect a distribution type may be specified. The default is the normal distribution, but the following types are possible, and for each distribution the meaning of the Percentile inputs are different.

1. "N" for normal
   
   

   P90	P50	P10
   The P90 percentile of the normal distribution	The mean value, is equal to the Mode and the Median	The 90 percentile

2. "L" for lognormal
   
   The lognormal is used for positively skewed distributions. It is appropriate for cases where small values are more common than the larger ones. Typically, oil field reserve sizes are lognormally distributed.

   P90	P50	P10
   The P90 percentile of the lognormal distribution The lowest possible value is zero, but for practical reasons use a small positive number.	The median value. Note that this is not the "Most likely" value.	the 90 percentile

3. "T" for triangular
   
   

   P90	P50	P10
   The "Low" parameter	The "Mode", the most likely value	the High parameter

   
   
   
4. "R" for rectangular
   The rectangular ("uniform") distribution has only two parameters, and will use the first and the last of the three inputs. The rectangular is rarely used in practice.

   P90	P50	P10
   The Lower boundary	Not used, can be left blank	The Upper boundary

The parameters are used to simulate a distribution as a vector with the specified number of elements, and taking the Probability of success (POSg) into account.

The vectors are partially sorted to provide the right dependence, as given by the correlation coefficient r in the input.

The output is a result table that can be copied to the clipboard and pasted in reports or a spreadsheet for further editing.

On the left are the most important statistics, such as percentiles. These are a short version of the expectation curve percentiles in 5% steps that are shown in the rightmost column. These values are the result of the simulation of individual prospects with their cutoffs, and, on top of that the overall cutoff. The values are "unrisked" which means that the Probability of Success after cutoff (POSc) has to be taken into account for dicision making. For a check on the effect of the individual and overall cutoff, the third column shows the 5% step percentiles. These are the values when no cutoff is applied and also unrisked. The probability of success associated with this column is the POSg, which is always equal or greater than the POSc.

NIEC - Number of identical expectation curves

When a play is evaluated, but there is only a "conceptual" trap description, as well as an estimate of the uncertain number of such traps, the NIEC option can cope with that situation.

The input consists of

• a set of three numbers at the top of the input window, which are the Low, Mode and High estimates of the number of traps. A triangular distribution is assumed.
• Description of the "typical" trap on the first line of the input grid.

The output is the standard result window as for the other options.

The method is a MC calculation of the single trap volume resulting in a vector of Ncyc length. For this vector there is a choice for the distribution shape as mentioned above. Another vector is calculated as a discrete triangular distribution of the number of traps. Both vectors are in random order. It would be possible, in priciple, to multiply the volume vector with each element of the trap number vector and calculate the mean for each element of the result vector, but it is sufficient to calculate the the row-wise product vector. After sorting this is the expectation curve of the result:

Merging is required if the conceptual model of a prospect is not unique. Most uncertainty about a prospect can be described by the uncertainty of the input variables. However, in some situations there are two quite dissimilar models possible. Note that it may be difficult to make a single appraisal of such situation. An example is where a mound is detected by seismic on a line. it could be an intrusive, a salt dome or a shale diapir. In practice velocity and or gravity information would help to decide which is the most likely model. If not, one can assign degrees of belief to the three alternatives.

Merge will allow description of the volume distribution of the alternatives and then merge the three expectation curves into one result curve.

The input to a MERGE is a set of (up to 10) lines in the input grid, one for each alternative. Essential is to assign the degrees of belief as a percentage. The percentages should add up to 100%. The output is the standard result window.

VOLCAL - Volume calculation for a single reservoir in a trap

This is a separate program which can be called from within AddRes. The program calculates the volume of recoverable hydrocarbons in a single reservoir. The input screen is shown below

There are three options:

1. Trap filled to spillpoint
2. Length of HC column and the percentage gascap within this column.
3. Position of Oil/water contact (OWC) and any Gas/Oil/Contact (GOC), or in in the unassociated gas case the GWC.

The possible trap types include:

1. Simple geometric trap descriptions with Length, Width and Vertical closure.
2. Anticlinal structures with a list of contour depth at top reservoir and corresponding areas.
3. Stratigraphic traps with a list of depths and contour areas for the top and the base of the reservoir.

The output is in terms of PPP values and mean to be used as input in AddRes.

• Gaeapas - internal report, Geology And Energy Analysis
• Dixon, W.L. & F.J. Massey Jr. (1957) "Introduction to statistical analysis"
  McGraw-Hill Book Company, Inc. New York, 488 p.