Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
40 changes: 36 additions & 4 deletions docs/ert/getting_started/howto/distance_localization.rst
Original file line number Diff line number Diff line change
Expand Up @@ -9,6 +9,34 @@ methods when the ensemble size is limited.

For theoretical background, see :ref:`distance_based_localization`.

Briefly how it works

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

there is already a section How it works

---------------------

In distance-based localization, lateral distance (not including vertical distance)
is measured by Euclidean distance by using the (x, y) coordinates of the

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

In distance-based localization, lateral (Euclidean) distance is estimated between the observation coordinate and field parameter location. This distance neglects vertical distance and uses purely horizontal one.

The implemented method ...

observations and location of field parameters. The implemented method is
based on the publication of Emerick (2016). A matrix, called RHO,
or localization matrix, has one element RHO(i,j) per pair of observation (index j)
and field parameter value (index i). If the distance between the observation
and the field parameter is 0, the RHO value, which is a scaling weight for
the Kalman Gain matrix, will be 1 and it will decrease with distance to 0
at two times the specified radius of influence of the observation.
The scaling function with normalized distance used here is the Gaspari-Cohn correlation
function. In ERT, each observation with an associated position is associated with an
influence radius in current version. When calculating updated field parameter,
the observations that have the field parameter within its influence range
(two times the specified radius with the Gaspari-Cohn function) will contribute
to the calculation of the updated field parameter.
A figure like the one below illustrates this.
Each of the four wells have their own radius of influence.
The grid cell colored with blue is within the range of observation located
at A and C, but not within the range of B and D. The grid cell colored
with red is within the range of observation in location B, C and D and
will be influenced by those in the update.

.. image:: illustrating_influence_range.png


Enabling distance based localization
-------------------------------------

Expand Down Expand Up @@ -113,7 +141,11 @@ Configuration file vs GUI
--------------------------

Distance based localization can only be enabled through the configuration
file.

For adaptive localization, the GUI provides a way to set the correlation threshold and truncation parameters.
Note that the values set in the GUI will override any values set in the configuration file.
file. The GUI analysis module dialog exposes only adaptive localization
settings.

If distance localization is configured for a parameter type, it is used for
that type even when adaptive localization is enabled in the GUI. The GUI's
adaptive localization settings are used only for parameters whose configured
strategy is ``ADAPTIVE``; they do not replace a configured ``DISTANCE``
strategy.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading