Rechercher

Sur ce site


le jeudi 10 septembre à 13:00

Tomislav Hengl (Post-doc, Universiteit van Amsterdam)

salle de formation du PRABI

par Vincent Daubin - 10 septembre 2009

Creating dynamic maps of biodiversity : examples in R + Google Earth

The seminar will focus on application of open source tools for
analysis of biodiversity data. The focus is put on R statistical
computing environment and its various Environometrics and Spatial
packages. There are three main opportunities for using R : (1) vitality
and high speed of development of R, (2) academic openness of developers
and their willingness to collaborate, and (3) increasing sympathy for
spatial data analysis and visualization. Spatial statisticians generally
believe that geostatistical techniques are suited only for modeling of
features that are inherently continuous (spatial fields) ; discrete
objects (points, lines, polygons) should be analyzed using point pattern
analysis and similar methods. A method is suggested that tries to bridge
this gap by combining the strong points of various packages in R with
the geographical analysis capabilities of the open source GIS. The
presenter will demonstrates how to combine geostatistical techniques
with conceptually different techniques --- point pattern analysis and
Niche analysis --- to allow prediction of species’ distributions using
regression-kriging.

References :
1. Hengl, T., Sierdsema, H., Radovic, A., Dilo, A., 2009. Spatial
prediction of species’ distributions from occurrence-only records :
combining point pattern analysis, ENFA and regression-kriging.
Ecological Modelling, in press.
2. Hengl, T., 2007. A Practical Guide to Geostatistical Mapping of
Environmental Variables. EUR 22904 EN Scientific and Technical Research
series, Office for Official Publications of the European Communities,
Luxemburg, 143 pp. ISBN : 978-92-79-06904-8
3. Hengl T., Heuvelink G.B.M., Rossiter D.G., 2007. About
regression-kriging : from equations to case studies. Computers and
Geosciences, 33(10) : 1301-1315