[R-sig-Geo] Problem with Sequential Gaussian Simulation
Micha Silver
t@v|b@r @end|ng |rom gm@||@com
Mon Jan 28 23:19:47 CET 2019
I've run into a problem trying to produce a number of SGS realizations
using gstat::krige. (this was definitely working two weeks ago, and now
I'm getting a seg fault, and I can't figure out why. Any help would be
appreciated.
Here are the details:
My input "locations":
-------------------------
> str(gauges_spdf)
Formal class 'SpatialPointsDataFrame' [package "sp"] with 5 slots
..@ data :'data.frame': 24 obs. of 1 variable:
.. ..$ precip: num [1:24] 13.21 2.66 12.67 8.97 18.69 ...
..@ coords.nrs : num(0)
..@ coords : num [1:24, 1:2] -37.5 11.5 -9.5 20.5 -27.5 -3.5
-12.5 33.5 -4.5 34.5 ...
.. ..- attr(*, "dimnames")=List of 2
.. .. ..$ : NULL
.. .. ..$ : chr [1:2] "x" "y"
..@ bbox : num [1:2, 1:2] -45.5 -49.5 46.5 49.5
.. ..- attr(*, "dimnames")=List of 2
.. .. ..$ : chr [1:2] "x" "y"
.. .. ..$ : chr [1:2] "min" "max"
..@ proj4string:Formal class 'CRS' [package "sp"] with 1 slot
.. .. ..@ projargs: chr NA
My target "newdata" grid
-------------------------
> str(grd)
Formal class 'SpatialPixels' [package "sp"] with 5 slots
..@ grid :Formal class 'GridTopology' [package "sp"] with 3 slots
.. .. ..@ cellcentre.offset: Named num [1:2] -49.5 -49.5
.. .. .. ..- attr(*, "names")= chr [1:2] "x" "y"
.. .. ..@ cellsize : Named num [1:2] 1 1
.. .. .. ..- attr(*, "names")= chr [1:2] "x" "y"
.. .. ..@ cells.dim : Named int [1:2] 100 100
.. .. .. ..- attr(*, "names")= chr [1:2] "x" "y"
..@ grid.index : int [1:10000] 1 2 3 4 5 6 7 8 9 10 ...
..@ coords : num [1:10000, 1:2] -49.5 -48.5 -47.5 -46.5 -45.5
-44.5 -43.5 -42.5 -41.5 -40.5 ...
.. ..- attr(*, "dimnames")=List of 2
.. .. ..$ : chr [1:10000] "1" "2" "3" "4" ...
.. .. ..$ : chr [1:2] "x" "y"
..@ bbox : num [1:2, 1:2] -50 -50 50 50
.. ..- attr(*, "dimnames")=List of 2
.. .. ..$ : chr [1:2] "x" "y"
.. .. ..$ : chr [1:2] "min" "max"
..@ proj4string:Formal class 'CRS' [package "sp"] with 1 slot
.. .. ..@ projargs: chr NA
My variogram
-------------------------
> precip_fit
model psill range
1 Nug 0.6391787 0.00000
2 Exp 12.0158808 17.39194
The command that I am running
(any value of nsim causes the crash below. But oridinary kriging
*without* the nsim param works fine)
-------------------------
precip_SGS = krige(formula = precip~1,
locations = gauges_spdf,
newdata = grd,
model = precip_fit,
nsim = num_sims,
nmax = 4)
The error output:
-------------------------
drawing 10 GLS realisations of beta...
[using conditional Gaussian simulation]
*** caught segfault ***
address (nil), cause 'memory not mapped'
Traceback:
1: predict.gstat(g, newdata = newdata, block = block, nsim = nsim,
indicators = indicators, na.action = na.action, debug.level = debug.level)
2: predict(g, newdata = newdata, block = block, nsim = nsim,
indicators = indicators, na.action = na.action, debug.level =
debug.level)
3: .local(formula, locations, ...)
4: krige(formula = precip ~ 1, locations = gauges_spdf, newdata =
grd, model = precip_fit, nsim = num_sims, nmax = 4)
5: krige(formula = precip ~ 1, locations = gauges_spdf, newdata =
grd, model = precip_fit, nsim = num_sims, nmax = 4)
6: CreateRainRealizations(gauges_spdf, num_sims)
7: eval(ei, envir)
8: eval(ei, envir)
9: withVisible(eval(ei, envir))
10: source("~/Studies/Research/synthetic/code/run_synthetic.R")
My seesion Info:
-------------------------
> sessionInfo()
R version 3.5.2 (2018-12-20)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Debian GNU/Linux buster/sid
Matrix products: default
BLAS: /usr/lib/x86_64-linux-gnu/atlas/libblas.so.3.10.3
LAPACK: /usr/lib/x86_64-linux-gnu/atlas/liblapack.so.3.10.3
locale:
[1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C
[3] LC_TIME=en_US.UTF-8 LC_COLLATE=en_US.UTF-8
[5] LC_MONETARY=en_US.UTF-8 LC_MESSAGES=en_US.UTF-8
[7] LC_PAPER=en_US.UTF-8 LC_NAME=C
[9] LC_ADDRESS=C LC_TELEPHONE=C
[11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C
attached base packages:
[1] stats graphics grDevices utils datasets methods base
loaded via a namespace (and not attached):
[1] compiler_3.5.2
Many thanks,
Micha
More information about the R-sig-Geo
mailing list