geo_findlocal: Search for likely simulated localities with geographic...
In rsh249/vegdistmod: Vegetation distribution modeling and spatial informatics

This wraps the search() function within a set of commands that randomly splits the distribution in to 4 quadrants 'n' number of times. For widespread, genetically diverse species this has the effect of applying more locally relevant probability functions to the data.

1
2
3

geo_findlocal(ext_ob, clim, type, maxiter = 10, searchrep = 1,
  manip = "condi", alpha = 0, divisions = 10, factor = 4,
  parallel = FALSE, nclus = 2, w = FALSE)

`ext_ob`	A data.frame of climate values (i.e., extracted from the climate raster object). Will be passed to multiv_likelihood().
`clim`	A raster object of climate data (matching x)
`type`	Designate either ".gauss" or ".kde".
`maxiter`	Maximum number of search iterations.
`searchrep`	How many times to search for simulated localities (per parent occurrence) per iteration. Recommend 1, but feel free to tune this parameter.
`manip`	Specify either 'reg', 'condi', or 'bayes. Default is 'condi'. See documentation for densform() for descriptions of these.
`alpha`	The value of alpha to be used to calculate the initial confidence interval for removing climatic outliers in the sample(s).
`divisions`	How many times should the data be split into quadrants? Default is 5 resulting in 20 (5x4quads) geographically oriented samples to be selected.
`factor`	To be passed to the extraction() function for post search thinning of data to limit overfitting. Set to 1 to ignore.
`parallel`	True or False to use parallel computing. This implements each division as embarassingly parallel processes. However, note that this option changes the meaning of the divisions object to the maximum number of division iterations. Fewer iterations may be returned if too few occurrences are selected.
`nclus`	If parallel is TRUE then set the maximum number of cores to use in the compute cluster. Default is 2.
`w`	To weight log-likelihoods by the coefficient of variation or not.

## Not run: 
data(abies);
head(abies)
ext.abies = extraction(abies, climondbioclim, 
 schema='flat', factor =4,  
 rm.outlier=FALSE);
sea <- geo_findlocal(ext.abies, climondbioclim, 
 type = '.kde', maxiter = 5, 
 searchrep = 1, manip = 'condi', 
 divisions = 8, parallel =TRUE, nclus = 4)
 plot_clim(sea, climondbioclim[[5]])
 
## End(Not run)