Boundaries have a somewhat complicated set of options, superficially similar to those of points and lines, but dependent on both special properties of boundaries and the specific type of boundary data (point elements vs. line elements).
Additional options for boundaries.
If the boundary elements consist of points (i.e., boundary elements were generated from continuous wombling), one has the option of plotting the elements themselves or the sub-boundary connections.
Boundary Style window. The specific available options depend on whether the boundaries consist of point elements or line elements.
(A)
(B)
Boundary elements for oral cancer mortality in males in Western Europe, generated from continous wombling (top 20% of elements are shown). (A) Basic point boundary element plot; (B) the connected subboundaries rather than the individual elements.
When plotting the elements (but not the connections), there are additional options available. Instead of simply showing the location of the elements, one can draw the elements as vectors indicating not only the location of the element but the direction of change associated with each element.
Boundary elements for oral cancer mortality in males in Western Europe drawn as vectors rather than as points. The arrow indicates the direction of change associated with each element.
Additionally, one can use color, style, and size to indicate the magnitude of change associated with each element. The color and style can be used with both point and vector displays, but not with connections because the values are associated with the point elements themselves and not the connections between them.
(A)
(B)
Boundary elements for oral cancer mortality in males in Western Europe. Color is used to indicate the magnitude of change with aqua indicating low values and purple high values. (A) Point element plot, using size as well as color to indicate the magnitude of change; (B) Vector plot.
If the boundary elements consist of line segments (i.e., generated from categorical or distance-based wombling), the options are somewhat reduced. Only the individual elements can be displayed (since the elements already consist of lines) and no vector display is possible since the vector of change is, by definition, perpendicular to each segment (and for categorical data, there is not necessarily a high-to-low direction). One can still color indicate the magnitude of change using color and style; in this case the lines themselves are colored since the lines are the boundary elements.
(A)
(B)
Boundary elements for oral cancer mortality in males in Western Europe, generated from categorical/distance wombling rather than continuous wombling as in previous examples. (A) Basic plot of the boundary elements; (B) Color and line thickness are used to indicate the magnitude of change with aqua indicating low values and purple high values.