On Sat, 29 Dec 2012 09:58:00 +0100 Sergio Vignali <[email protected]> wrote:
> Hi all, > > I have a problem with the LSCV kernel density on adehabitatHR: > "Warning message: > In kernelUD(xy, id, h = "LSCV") : The algorithm did not converge > within the specified range of hlim: try to increase it" > I've tried to search on google and I've saw that it's an old problem, > but I've not found a solution. I have several points of birds take > with back mounted GPS device, and more of these points have the same > location, but I can't remove all of these because some points > account foraging areas. Can someone help me, please? Hello Sergio, the problem you are encountering is actually a feature of LSCV. When points are scattered in some particular patterns, the calculation is impossible, therefore there is no output. This is by no means a suggestion to move/eliminate locations to make the method work! This failure is rather a suggestion to explore other methods and consider the ecological reasons of the distribution of points, as you already did by identifying foraging areas. I have no experience in analysing bird data, but can suggest NNCH method, that is a variant of MCP. See adehabitat manual for more details and some literature on it. Let us know if this helps. Kind regards and happy new year :) Anne -- http://wiki.osgeo.org/wiki/User:Aghisla
signature.asc
Description: PGP signature
_______________________________________________ AniMov mailing list [email protected] http://lists.faunalia.it/cgi-bin/mailman/listinfo/animov
