species diversity gradient problem

[Don McAllister]

Mike Crisp wrote:

> Currently I am dealing with a similar problem to that of Peter Schuchart
> (below).  I am trying to relate species richness and endemism of plants in
> Australia to climatic variables, latitude, etc.  Our data are scored in 1
> degree grid cells, and we too have a problem of sampling artifacts, eg the
> deserts are not well explored.......

Dr. Frederick W. Schueler and I also used 1 degree latitude-longitude grid
cells to plot fish species diversity, climatic and other variables for
freshwater fishes of United States and southern Canada (see chapter in
Zoogeography of Freshwater Fishes of North America edited by C. Hocutt and E.
Wiley).  Later we developed a global equal-area grid (see chapter in Mapping
the Diversity of Nature edited by Ronald I Miller) to get rid of the shrinkage
of 1 degree cells towards the poles.

We discussed the problem of uneven sampling.  We concluded that with sparse
data you would have to increase the sizes of the cells in the grid (until
samples were sufficient).  One could measure the number of species per unit
area with different sizes of cells - or depth ranges in the case of species x
depth curves,  and see where the curve tended to flatten out.

We also calculated, using a couple of methods, the geographic sizes of ranges
of species and prepared frequency graphs showing how many had ranges of a given
size.

The other depth-wise approach is to find a single location where sampling is
thorough.  For fishes, since I am an ichthyologist, I would first check William
Beebe's data for the offing of Bermuda.  That might provide a gauge of the
variables.

don
Don McAllister