geost.models.model_array.ModelDataArray.slice_depth_interval#

ModelDataArray.slice_depth_interval(upper: int | float | ndarray | DataArray = None, lower: int | float | ndarray | DataArray = None, how: Literal['overlap', 'majority', 'inner'] = 'overlap', update_top_bottom: bool = True, drop: bool = True) → Dataset | DataArray#

Slice a specified depth interval from a voxelmodel or layermodel between upper and lower bounds.

Parameters:

  • upper (int | float | xr.DataArray, optional) – Upper and/or lower bound of the depth interval. This can be a single value or a 1D or 2D DataArray containing variable depths. In case of a DataArray, a 1D DataArray should contain either the “x” or “y” dimension and a 2D DataArray should contain both “x” and “y” dimensions. The names of the “x” and “y” dimensions of the DataArray must match those of the model. Otherwise broadcasting cannot be done correctly and the slicing cannot be done. In case of a Numpy array, the axis order is expected to be (y, x) for 2D arrays. The default is None.

  • lower (int | float | xr.DataArray, optional) – Upper and/or lower bound of the depth interval. This can be a single value or a 1D or 2D DataArray containing variable depths. In case of a DataArray, a 1D DataArray should contain either the “x” or “y” dimension and a 2D DataArray should contain both “x” and “y” dimensions. The names of the “x” and “y” dimensions of the DataArray must match those of the model. Otherwise broadcasting cannot be done correctly and the slicing cannot be done. In case of a Numpy array, the axis order is expected to be (y, x) for 2D arrays. The default is None.

  • how ({"overlap", "majority", "inner"}, optional) – Method to use for slicing. This parameter is only applicable to voxelmodels (i.e., model.gst.model_type is ModelType.VOXEL) and will be ignored for layermodels. The default is “overlap”. - “overlap”: Include voxels that at least partially overlap with the specified depth interval. - “majority”: Include voxels that have 50% or more of their volume within the specified depth interval. - “inner”: Include only voxels that are completely within the specified depth interval.

  • update_top_bottom (bool, optional) – If True, the “top” and “bottom” variables of a layermodel will be updated to reflect the new depth interval after slicing. If False, the “top” and “bottom” variables will remain unchanged. This parameter is only applicable to layermodels (i.e., model.gst.model_type is ModelType.LAYER) and will be ignored for voxelmodels. The default is True.

  • drop (bool, optional) – If True, depths where the result only contains missing values will be dropped from the slice result. If False, the original shape is kept. The default is True.

Returns:

A new xarray.Dataset or xarray.DataArray containing only the data within the specified depth interval.

Return type:

xr.Dataset | xr.DataArray

Examples

Slice a fixed depth interval between -10 and -20:

>>> sliced = model.gst.slice_depth_interval(upper=-10, lower=-20)

Slice a model of 2 rows and 2 columns between variable depth intervals using DataArrays.

>>> upper = xr.DataArray([[-10, -15], [-12, -18]], dims=("y", "x"))
>>> upper  # 2D array with different depth interval for each "x" and "y"
<xarray.DataArray (y: 2, x: 2)>
array([[-15, -20],
       [-17, -23]])
Dimensions without coordinates: y, x
>>> sliced = model.gst.slice_depth_interval(upper=upper, lower=upper - 5)

>>> upper = xr.DataArray([-10, -15], dims=("y",)) # slice along "y"
>>> upper  # 1D array with different depth interval for each "y" but the same for every "x"
<xarray.DataArray (y: 2)>
array([-15, -20])
Dimensions without coordinates: y
>>> sliced = model.gst.slice_depth_interval(upper=upper, lower=upper - 5)