| 1 | = News from the !AniMove project = |
| 2 | |
| 3 | == New GRASS modules == |
| 4 | |
| 5 | Clément Calenge programmed three new add-ons for the GRASS software, and thanks to the kind help of Markus Neteler, they are now available on the GRASS Addons repository. These addons allow to estimate the home range of animals monitored using radio-tracking, GPS, Argos, etc. They implement three main home range estimation methods provided by the package adehabitat for the R software: |
| 6 | |
| 7 | * '''v.adehabitat.clusthr''': this addon allows the estimation of the home range by single linkage cluster analysis, using the methodology of Kenward R.E. et al. (2001; Ecology). This method has been "translated" for GRASS from the R function "clusthr" in the R package adehabitat. |
| 8 | * '''v.adehabitat.mcp''': this addon estimates the minimum convex polygon home range (Mohr, C.O. 1947. The American Midland Naturalist). This function is just a slight modification of the module v.hull of Andrea Aime. |
| 9 | * '''v.adehabitat.kernelUD''': this addon allows the estimation of the kernel utilization distribution and home ranges, following the guidelines provided in Worton (1995; Journal of Wildlife Management). This module heavily relies on the code written by Stefano Menegon for his module v.kernel. It has just been extended to allow the LSCV and ad hoc estimation of the smoothing parameter, and the computation of the volume under the UD for home range estimation. The only difference with the function kernelUD from the R package adehabitat is related to the LSCV method: because the minimization of the MISE requires numerical methods, the smoothing parameter found by v.adehabitat.kernelUD differ slightly from those returned by the function kernelUD in adehabitat (the correlation between the two is often greater than 0.99, pers.obs.). |
| 10 | |
| 11 | == Adehabitat News == |
| 12 | A new version of adehabitat R package is now available through the official CRAN archives. |
| 13 | |
| 14 | Significant changes are listed below: |
| 15 | |
| 16 | * The package has been reorganized into four parts (see ?adehabitat-package for a description): |
| 17 | * management of raster maps |
| 18 | * habitat selection / ecological niche analysis |
| 19 | * home range analysis |
| 20 | * analysis of animals trajects. |
| 21 | * The package contains several demo files to allow an overview of these parts : demo(rastermaps), demo(homerange), demo(managltraj), demo(analysisltraj), demo(nichehs). |
| 22 | * the package now contains a new function allowing the exploration of the ecological niche, which generalizes several factor analyses (ENFA, MADIFA, ...) and is closely related to several methods (Mahalanobis distances, selection ratios, etc.), named gnesfa() (see the examples of the help page for the properties of this analysis). |
| 23 | * The class ltraj now distinguishes two types of trajects: type I (time not recorded, e.g. tracks of animals in the snow) and type II (time recorded, e.g. GPS monitoring). Trajects of type II may either be "regular" (constant time lag between relocations) or not. |
| 24 | * Numerous example datasets have been added to the package to illustrate the analysis of animals trajects: 4 porpoises, 6 albatross, 1 hooded seal, 1 whale, 1 brown bear, two roe deer, two chamois, 4 ibex, 1 mouflon, 3 wild boar |
| 25 | * Many functions have been added to allow the management of animals trajects within R: Some functions allow to handle the attributes or the storage of the trajects (typeII2typeI, typeI2typeII, sett0, cutltr, is.regular, is.sd, mindistkeep, offsetdate, set.limits), other allow to manage missing values and test their random distribution in the traject (setNA, summaryNAltraj, plotNAltraj, runsNAltraj), other allow a graphical exploration of the properties of the trajects (hist.ltraj, plot.ltraj, plotltr, sliwinltr). |
| 26 | * Several functions now allow to test the independence of the descriptive parameters in the trajects (indmove and wawotest for dx, dy and dist, testang.ltraj for rel.angle and abs.angle) |
| 27 | * Several functions allow to simulate common models of trajects: the correlated random walk (simm.crw), the brownian motion (simm.brown), the arithmetic brownian motion (simm.mba), the Ornstein Uhlenbeck process (simm.mou), the brownian bridge (simm.bb) and the Levy process (simm.levy). |
| 28 | * The function explore.kasc() provides a Tk interface for the exploration of a multi-layer raster map f class "kasc" |
| 29 | * A partitioning algorithm (still under research) is also available to partition a traject into segments with homogeneous properties (see the help page of modpartltraj) |
| 30 | * The bugs in redisltraj and mcp.area have been corrected |