work on documentation

Author: b8raoult

docs/examples.rst

@@ -4,7 +4,6 @@
  Examples
 ##########
 
-
 .. toctree::
    :maxdepth: 2
    :glob:

docs/using/combining.rst

@@ -87,7 +87,24 @@ Please note that you can join more than two ``zarr`` files.
 
    from anemoi.datasets import open_dataset
 
-   ds = open_dataset(grids=[dataset1, dataset2, ...], method=...)
+   ds = open_dataset(grids=[dataset1, dataset2, ...], mode=...)
+
+The values for ``mode`` are:
+
+concatenate
+===========
+
+All the grid points are concatenated, in the order they are given. The
+`latitudes` and `longitudes` are also concatenated.
+
+cutout
+======
+
+The `cutout` mode only supports two datasets. The first dataset is the
+considered to be a limited area model (LAM), while the second one is
+considered to be a global model or boundary conditions. It is therefore
+expected that the bounding box of the first dataset is contained within
+the bounding box of the second dataset.
 
 .. image:: cutout-1.png
    :width: 75%

docs/using/grids.rst

@@ -23,20 +23,20 @@ So for 2D fields, the thinning will apply to both the latitude and
 longitude dimensions. For 1D fields, such as reduced Gaussian grids, the
 thinning will apply to the only dimension.
 
-The following example shows the effect of thinning a dataset with a 1 degree
-resolution:
+The following example shows the effect of thinning a dataset with a 1
+degree resolution:
 
 .. image:: thinning-before.png
    :width: 75%
    :align: center
 
-Thinning the dataset with ``thinning=4`` will result in the following dataset:
+Thinning the dataset with ``thinning=4`` will result in the following
+dataset:
 
 .. image:: thinning-after.png
    :width: 75%
    :align: center
 
-
 ******
  area
 ******