Doing some brainstorming to encourage discussion . . .
1) Should the raster be converted to a a point layer? Although the points would all be at equal distances, could the heterogeneity of the values (i.e. bird abundance) still be quantified?
2) Whether using raster or vector tools: how is spatial heterogeneity defined using GIS tools? Are some more reflective of ecological processes?
3) Within a specific polygon size (eg watershed scale), can patches of non-significant differences in values (i.e. similar bird abundance) be created? If so, number, location and size of patches of similar bird abundance within each polygon (i.e. watershed) would be interesting.
4) What tool gives average distance of significant difference between points (or cells) within a polygon? If patches (i.e. polygons) are created from the point/cells, is there a tool that defines the amount of edges (i.e. there would be a lot more edges in a highly heterogeneous landscape).
5) . . .
