[ai-geostats] Spatio-temporal semivariograms
I'm looking for published case studies involving good discussions on structural analyses (in a geostatistical point of view, that is variography, etc.) of variables that have been referenced with a heterogeneous coordinate system, like
X = Distance (i.e.Latitude or Longitude)
Y = Time
Z = anything (in my case occurences of an event)
I'm not very familiar with spatio-temporal modelling but I assume that one has to work practically with directional semivariograms (0 degrees, that is West-East) with no tolerance (in order to avoid the use of future measurements for estimating the past), that the search strategy of neighbouring points during the kriging step also needs to be directional…
I also imagine that directional semivariograms that differ from 0 and 90 degrees are meaningless in a physical point of view.
Thank you very much in advance for any hints/tips/papers.
Gregoire Dubois (Ph.D.)
JRC - European Commission
IES - Emissions and Health Unit
Radioactivity Environmental Monitoring group
TP 441, Via Fermi 1
21020 Ispra (VA)
Tel. +39 (0)332 78 6360
Fax. +39 (0)332 78 5466
"The views expressed are purely those of the writer and may not in any circumstances be regarded as stating an official position of the European Commission."