DOC: review comment [ci doc]

jklymak · jklymak · commit 8af19acad051 · 2026-04-03T16:51:26.000-07:00
diff --git a/galleries/users_explain/plotting/gridded.py b/galleries/users_explain/plotting/gridded.py
@@ -1,8 +1,8 @@
 """
 .. _plot_gridded:
 
-Gridded plots
-=============
+Gridded data
+============
 
 This section discusses data that mapped onto a two-dimensional grid.  The data
 usually has structured coordinates defined by arrays of x and y values, and data
diff --git a/galleries/users_explain/plotting/index.rst b/galleries/users_explain/plotting/index.rst
@@ -1,6 +1,6 @@
-++++++++++++++++
-Plotting methods
-++++++++++++++++
++++++++++++++
+Plotting data
++++++++++++++
 
 .. _users-guide-plotting:
 
diff --git a/galleries/users_explain/plotting/threed.py b/galleries/users_explain/plotting/threed.py
@@ -7,12 +7,12 @@
 Matplotlib has support for three-dimensional plots, including surface plots,
 wireframes, and 3D scatter plots.
 
-Note, however, that Matplotlib does not do proper ray-tracing, and hence
-multiple surfaces in one visualization will not occlude one another correctly;
-the last surface drawn will cover previous surfaces, regardless of their
-distance from the viewer.  Either try to combine surfaces into one (which is
-possible for simple plots, see :ref:`intersecting_planes` and :ref:`box3d`), or
-for true 3D, consider using `Mayavi
+Note, however, that Matplotlib does not do true 3D rendering with depth
+buffering, and hence multiple surfaces in one visualization will not occlude
+one another correctly; the last surface drawn will cover previous surfaces,
+regardless of their distance from the viewer.  Either try to combine surfaces
+into one (which is possible for simple plots, see :ref:`intersecting_planes`
+and :ref:`box3d`), or for true 3D, consider using `Mayavi
 <https://docs.enthought.com/mayavi/mayavi/>`_.
 
 Plotting in 3D can be done using the `.mpl_toolkits.mplot3d` toolkit by
diff --git a/galleries/users_explain/plotting/unstructuredgrid.py b/galleries/users_explain/plotting/unstructuredgrid.py
@@ -4,21 +4,25 @@
 Unstructured data
 =================
 
-Unlike :ref:`structured grids <plot_gridded>`, three-dimensional data is often
-collected without the benefit of a grid, with :math:`z = f(x, y)` where
-:math:`x` and :math:`y` are one-dimensional arrays of the same length, and
-hence :math:`z` is also one-dimensional.  Sometimes it is useful to grid that
-data via appropriate interpolation schemes (see `scipy.interpolate.griddata`
-for some in python). However, if the data is spaced appropriately, it can sometimes
-be plotted directly in Matplotlib.
-
-Matplotlib accomplishes this via a suite of triangulation routines in the
-:mod:`matplotlib.tri` module.  The most basic triangulation routine is
-`~matplotlib.tri.Triangulation`, which is a simple wrapper around the popular
-`Qhull <http://www.qhull.org/>`_ library.  The `~matplotlib.tri.Triangulation`
-class is used to create a triangulation of a set of points, and this
-triangulation can then be used to create either faceted psuedocolor plots, or
-contour plots.
+Unlike :ref:`structured grids <plot_gridded>`, unstructured data are not
+sampled on a regular grid. In this case, :math:`x` and :math:`y` are
+one-dimensional arrays of the same length, and :math:`z = f(x, y)` is also
+one-dimensional. For example, consider a temperature field
+:math:`T(\\text{longitude}, \\text{latitude})` measured at 120 weather stations,
+with each station at a different latitude and longitude.
+
+You can interpolate this kind of data onto a regular grid (for example with
+`scipy.interpolate.griddata`), but if the sampling is suitable it can often be
+plotted directly in Matplotlib.
+
+Matplotlib supports this through triangulation routines in
+:mod:`matplotlib.tri`. The `Qhull <http://www.qhull.org/>`_ library is used to
+build a triangulation from the input points via
+`~matplotlib.tri.Triangulation`, which can then be used for faceted
+pseudocolor and contour plots.
+
+The triangulation treats the points :math:`(x, y)` as unordered, connecting
+points based on proximity.
 
 """
 
@@ -87,7 +91,7 @@ class is used to create a triangulation of a set of points, and this
 # --------------------------
 #
 # The `~.axes.Axes.tricontour` and `~.axes.Axes.tricontourf` methods can be
-# used to plot contours of unstractured data, where the underlying
+# used to plot contours of unstructured data, where the underlying
 # triangulation is used to interpolate the contour locations.  Here we plot
 # contours of the same data as above.
 
