Eclipse Syntax

An Eclipse data file is comprised of eight sections headed by a section header:

RUNSPEC (required)
GRID (required)
EDIT (optional)
PROPS (required)
REGIONS (optional)
SOLUTION (required)
SUMMARY (optional)
SCHEDULE (required)

These sections must come in the prescribed order, but the order of the keywords within each section is arbitrary (except the SCHEDULE section where time-dependency is handled in the order it is defined).

Data format

Eclipse data format

The different data items are identified by keywords, and most often followed by the associated data. Keywords are not case-sensitive.
Only standard ASCII characters are allowed.
The length of the keyword is limited to eight characters; all that is after the eighth character, is perceived as a comment. Therefore, the data following the keyword should be in the next line.
The file has a default width of 132 characters. All that is after the 130 character, is perceived as a comment.
The keyword data is always terminated by a slash (‘/’). In some cases data occur in groups, where each group is terminated by a slash.

Example

RUNSPEC            -- section header

OIL                -- keyword without data
WATER

DIMENS             -- keyword with data
24   25   15  / -- data associated to keyword DIMENS

A Grid property like porosity must be defined with one value for each grid cell.
The Eclipse data format expects exactly this: with a total of N = NX*NY*NZ cells, N porosity values must be given.
The order is according to the “book order”, line after line until a page has been read, then the next page until all data has been read. I.e. First the top layer (K=1) is read, line by line (begin with J=1, read I=1,...,NX, then J=2, etc.) Then repeat for K=2,...,NZ.

An example for porosity is then,

PORO
--(Totally N values)
0.32 0.24 0.24 0.3 0.31 0.31 ...
...
...0.26 0.25 0.25 0.25 0.28 /

Data repeats

When providing a data vector associated with a keyword, often the same value will be repeated. One example is that we often look at models where we assume that porosity is constant in each layer. Then the porosity keyword would be comprised of NX*NY equal values for layer 1, followed by another NX*NY equal values for layer 2 etc. This situation is so common that Eclipse fortunately allows for a shorthand notation: n*val means that the value val shall be repeated n times. So,

PORO
0.23 0.23 0.23 0.23 0.23 0.23 0.23 0.23 0.24 0.25 0.25 0.25 /

is equivalent to

PORO
8*0.23 0.24 3*0.25 /

Note that there must not be any blanks on any side of the asterisk (star).

Default values

In many cases required data items may be defaulted. Eclipse has defined default values for many values, and if we on input set these items to default, Eclipse will use the defined default values instead. A default value is specified by “1*”, and can be repeated as above, with n* (e.g. to specify that the next three items shall be defaulted use 3*).
If “the rest” of the data items are defaulted, this is not necessary to state explicitly.
As an example, the COMPDAT keyword requires 14 data items, but we will default the last five.
Either of the syntaxes below are valid.

COMPDAT<br />
'WP1' 10 2 1 2 'OPEN' 2* 0.5 5* /
'WP2' 10 3 1 2 'OPEN' 2* 0.5 /
/

Operations on the properties

In EDIT section the properties (such as PORO or PERMX, PERMY etc.) can be modified in the following manner:

1. To set the certain value of property firstly the BOX defined (keywords BOX, ENDBOX) and within this BOX the property is set:

BOX
--x1 x2   y1   y2   z1 z2
137 143  249  255   80 95 /  
/

MULTPV
--number of cells*multiplier
  900*30 /
/

ENDBOX/
/

2. The value can be multiplied:

MULTIPLY
--name  modifier  x1 x2   y1   y2   z1 z2
  PORV   30     137 143  249  255   80 95 /  
/

Comments

Any line in an Eclipse input file that starts with two dashes (--) is treated as a comment.
Comments are used to embed readable annotations for easier to understanding the input data.

Example

--This comment will be ignored by Eclipse

INCLUDE statements

Some of the data items can be huge, especially the GRID section data. It is possible to keep them outside the DATA file.
Eclipse allows for including other files in the data file, also on several levels (a file that is included may itself contain INCLUDE statements). Eclipse also knows about absolute and relative addressing of folders.

Includes example.

If the file to include resides in the same folder as the data file, INCLUDE statement will be:

INCLUDE
‘EXAMPLE.GRDECL’ /

In case the file EXAMPLE.GRDECL and the data filereside in different folders:

INCLUDE
‘../include/GRID/EXAMPLE.GRDECL’ /

Note: Each INCLUDE statement contains exactly one file to include. It is not possible to list several files in the same INCLUDE keyword.

Eclipse Syntax

Contents