Dear list,
I am going to build a spacetime::STFDF object where the n spatial
locations hold h different values (different fields) over the m times.
Along the documentation (spacetime vignette, ?stConstruct, ?STFDF, ?reshape)
I only found examples with layout of {n·m·1} observations, and not
Dear experts,
rather than a technical question, this is more about how to set up a model
for spatiotemporal kriging predictions.
With special regard to the fitting of a spatio-temporal model, whichever the
covariance model I want will use, I was wondering /how deep/ in time should
my `spacetime`
Dear Tomislav,
could you eventually manage to develop some R example for WCPS queries?
I am going to ingest some spatio-temporal raster collection into rasdaman,
and I was interested in a direct access from inside R to this data.
Thanks for any hint,
Piero
--
View this message in context:
Dear list,
after reading through Spatio-temporal prediction of daily temperatures
using time-series of MODIS LST images (Hengl et al.,
http://dx.doi.org/10.1007/s00704-011-0464-2), I wanted to try out these SAGA
*grid_tools* modules to fill the gaps in my MODIS AOT datasets.
I tried to follow
Dear list,
I used to plot points of variable size with spplot this way:
library(RColorBrewer)
data(meuse)
coordinates(meuse) - ~x+y
spplot(meuse, zcol=om,
col.regions=colorRampPalette(brewer.pal(7, PuBu)[-(1:1)])(20),
scales=list(draw=TRUE),
cex=.4*(1:5)
)
but now,
Dear list,
I have a SpatialGridDataFrame, whose SRS is EPSG:32632:
+init=epsg:32632 +proj=utm +zone=32 +ellps=WGS84 +datum=WGS84 +units=m
+no_defs +towgs84=0,0,0
I am willing to overlay these grid values over point locations defined in a
SpatialPixelsDataFrame with a different SRS using the
A quick and dirty answer would be to work with grids, and e.g. retain a
single value in each grid cell: that's what I would like to do, since my
points pattern is highly clustered.
I thought I could do something like:
overlay(points, grid, fn=mean)
but looking at methods?sp::overlay it seems
Hi,
Jan also wrote:
so i expect to get 36 s values and 480 t values. Instead for the first cell
and the first timestep the value is missing, the lm result is only 35 s
values and 479 t values ( + one intercept coefficient ). Is there a reason
behind this and some way to fix it?
Any clue about
Maybe this is how the combination is meant to be:
- apply regression model on the area of interest
- use the residuals at sampling location of the target variable to create
direct- and cross-variograms along with the non-ubiquitous predictors
- apply co-kriging by means of these variograms to
Dear list,
a slightly different scenario: what if one has a set of ubiquitous
predictors and e.g. one predictor which has a limited number of
observations?
Hengl in his book A practical guide to geostatistical mapping suggests
(probably) to combine RK with CK so that additional, more densely
If any of the participants is interested in finding a shared accommodation,
please contact me.
Cheers,
Piero
--
View this message in context:
http://r-sig-geo.2731867.n2.nabble.com/R-spatial-intensive-PhD-course-3-10-May-2011-tp5983951p6023564.html
Sent from the R-sig-geo mailing list archive
Ok sure.
I will hereby simulate the interpolation of a target variable points
(whose data is attached) via a single factor predictor classes created
randomly over the region of interest:
# libraries
library(sp)
library(gstat)
library(automap)
# Import the point pattern from the txt file and
That's it. Thank you!
Piero
PS:
To let the example work actually before calling krige() you should run this
command:
names(SPgrid) - CL
Moreover the right krige call is the following:
pm10.rk - krige(as.formula(lm$call$formula), points, SPgrid,
rvar$var_model)
--
View this message in context:
Hi list,
I'm applying a PCA to a georeferences set of variables.
As in A practical Guide to geostatistical mapping by Hengl, I'm using the
command prcomp in R.
To recover the georeference of the principal components (which could have
some NA) it is suggested to use the values available in the
This is more about kriging theory actually,
but I can't understand in which different ways the elevation information is
taken into account in the prediction process when e.g.:
- I apply ordinary cokriging on input data which include 3D (x,y,h) spatial
coordinates;
- I apply universal cokriging
Hi,
this is not a technical question, and maybe I shouldn't post it here.
Anyway, let's say you have several georeferenced datasets of a same variable
to be interpolated, each one representing different days, for example.
Is it better to compute offline a unique variogram model over the WHOLE
I would submit the same question.
Piero
(Pierre, did you manage somehow then?)
--
View this message in context:
http://r-sig-geo.2731867.n2.nabble.com/Continuous-colorbar-legend-for-a-SpatialPointsDataFrame-object-tp4805804p5855549.html
Sent from the R-sig-geo mailing list archive at
17 matches
Mail list logo