geost.base.Collection.add_model_data#

Collection.add_model_data(model: xr.Dataset | xr.DataArray, *, data_vars: str | list[str] = None, aggregate_vars: dict[str, str] = None, suffix: str = None) → Collection[source]#

Add information from one or more variables from a voxelmodel or layermodel as columns to the data table.

For each survey, this determines the vertical model stack at the survey’s location and identifies the layer boundaries within that stack based on changes in the selected model variables. Survey intervals are split at these model layer boundaries, and the corresponding model information is backfilled to the resulting intervals. The resulting data are sorted by depth. This is illustrated in the example below.

Survey data:
   nr  top  bottom lith
0   A  0.0    10.0    Z

Layer boundaries derived from the model data for "variable" at the location of the
survey:
   top  bottom  variable
0  0.0     8.0         1
1  8.0    11.0         2

Result:
   nr  top  bottom lith  variable
0   A  0.0     8.0    Z       1.0
1   A  8.0    10.0    Z       2.0
Parameters:

  • model (xr.Dataset | xr.DataArray) – Xarray Dataset or DataArray containing the model data.

  • data_vars (str | list[str], optional) – Variable or variables from the model to add. These variables are used to determine the layer boundaries. A layer boundary is defined where the value of a variable changes with depth; consecutive equal values are considered part of the same layer. If multiple variables are given, they are treated jointly when determining the layer boundaries. If None, all model variables are added and jointly considered for determining the layer boundaries.

  • aggregate_vars (dict[str, str], optional) – Optional dictionary specifying how to aggregate additional model variables, not given in data_vars, over the layers derived from data_vars. The keys are the model variable names and the values are the aggregation functions (e.g., “mean” or “sum”).

  • suffix (str, optional) – Suffix to append to the added model variable columns to avoid name conflicts with existing columns in the survey data. If None, no suffix is added. In case of a name conflict, the existing column is overwritten.

Returns:

New Collection containing the added information from the model in the data table.

Return type:

Collection

Examples

Add all model variables to the collection.

>>> result = collection.add_model_data(model)

To add specific model variables to the collection, use the data_vars parameter.

>>> result = collection.add_model_data(model, data_vars="variable") # This treats consecutive values in "variable" as a layer
>>> result = collection.add_model_data(
...     model, data_vars=["variable1", "variable2"]  # This treats jointly consecutive values in "variable1" and "variable2" as layers
... )

To add specific model variables and aggregate additional ones over the derived layers, use the aggregate_vars parameter.

>>> result = collection.add_model_data(
...     model, data_vars="variable", aggregate_vars={"other_variable": "mean"}
... )