G'day list, I'm calculating cross-variograms (variables below are sumso2="sulphur dioxide emissions" and intensity="rain intensity" - I have 2500 sample points) using GSTAT in R:
g <- gstat(NULL, id = "sumso2", form = sumso2 ~ 1, data=d1) g <- gstat(g, id = "intensity", form = intensity ~ 1, data=d1) d1xv.so2Vin <- variogram(g, cutoff=0.3, width=0.01) but - just looking at the variogram for sumso2 alone after this process (d1so.png): [cid:image002.png@01CD6345.4D6377B0] The sumso2 semi-variogram has a trend - and I need to remove it. If I calculate the variance of each point from the overall mean squared - and use that as sumso2 in the same gstat process - I get the sumso2 semi-variogram as (d1sores.png): [cid:image003.png@01CD6343.D40EC9B0] This is a better semi-variogram - not ideal yet - but better. My question is - is that calculation using the mean noted above suitable for removing the trend sufficient for ensuring a suitable cross-variogram? Note - I don't need to fit models or go to co-kriging - I'm just examining the form of the spatial relationships - I just want to describe the variable relationships as shown in the cross-variogram (a geography project). Now - if this is not suitable for removing the trend - then I understand I need to fit a polynomial to the variable sumso2 - could someone describe how I can do that it gstat please? My substantial thanks, ________________________________________________________ Michael Hewson PhD candidate Climate Research Group<http://www.gpem.uq.edu.au/crg> | Centre for Spatial Environmental Research<http://www.gpem.uq.edu.au/cser> The University of Queensland | Brisbane Q 4072 | Australia m.hew...@uq.edu.au | +61 (0)408 379 373 [Description: ClimateResearchGroupbanner]
<<inline: image001.jpg>>
<<inline: image003.png>>
<<inline: image002.png>>
<<attachment: d1sores.png>>
<<attachment: d1so.png>>
_______________________________________________ R-sig-Geo mailing list R-sig-Geo@r-project.org https://stat.ethz.ch/mailman/listinfo/r-sig-geo