[R-sig-Geo] questions on spatial interpolation exercise(NL)

Tue Nov 25 23:15:13 CET 2014

On 25-11-2014 21:41, Mengxi Yang wrote:
> attached file is monv_201301.Please have a look.
>

Janurary has 31 days hence you need to adjust the number of rows in that 
table:

 > tmp <- readLines(paste(getwd(), "/", "monv_201301.txt", sep=""))
 > tmp <- gsub(pattern=' . ', replacement=' 0.0 ', tmp, fixed = TRUE)
 > tmp <- gsub(pattern='[[:space:]]{2,7}', replacement=',', tmp)
 > tmp[8]
[1] "# 
NR,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,I,NORM,II,NORM,III,NORM,MND,NORM"
 > write.table(t(matrix(unlist(strsplit(tmp, ",")[-c(1:8)]), nrow=42)), 
"tmp.txt", col.names=FALSE, row.names=FALSE)
 > monv_201301 <- read.table("tmp.txt", col.names=c("empty1", "NR", 
paste("D", 1:31, sep=""), "I", "NORM", "II", "NORM", "III", "NORM", 
"MND", "NORM", "Locatie"))
 > monv_201301$empty1 <- NULL
 > monv_201301$Locatie <- as.factor(gsub(pattern='[[:space:]]+$', 
replacement='', monv_201301$Locatie))
 > str(monv_201301)
'data.frame':   325 obs. of  41 variables:
  $ NR     : int  760 458 680 572 89 664 678 560 910 441 ...
  $ D1     : num  7.2 17.2 10 9.6 12 11.9 10.2 14.3 13.6 15.5 ...
  $ D2     : num  0.4 2.8 4.7 2.7 3.2 4.5 3 2.8 0.8 3.2 ...
  $ D3     : num  0.9 0.9 1.7 0.5 0.5 1.4 1 0.5 0.9 0.2 ...
...

 From here on you can continue running the code from 
http://spatial-analyst.net/wiki/index.php?title=Spatial_interpolation_exercises_(NL)

> by the way, I want to do interpolation in different spatial resolution.
> I download the grid file from Worldclim project. But it is different
> from the exercise you provided.
> In the exercise, the prec6.asc is predicted long-term precipitation for
> June based on the Worldclim project page.
> could you please tell me how to do this?

You need to download some long term rainfall map (climatic) for January 
(I've used the one from http://worldclim.org/current) and then resample 
it to the Dutch projection system. I am sure if you search a bit you can 
find even more up-to-date maps of long term precipitation (climatic 
images) for Europe.

all the best with your work,

T. Hengl

>
> Thanks very much.
>
> On Mon, Nov 24, 2014 at 7:24 PM, Tomislav Hengl
> <hengl at spatial-analyst.net <mailto:hengl at spatial-analyst.net>> wrote:
>
>
>     You need to provide access to "monv_201310.txt" so that we can see
>     what is the structure of that file. The original file with rainfall
>     data from that URL is simply a text file from the KNMI website
>     (without any tabular structure) so that e.g.:
>
>     ...
>     write.table(t(matrix(unlist(__strsplit(tmp, ",")[-c(1:8)]),
>     nrow=41)), "tmp.txt", col.names=FALSE, row.names=FALSE)
>     ...
>     monv_200906 <- read.table("tmp.txt", col.names=c("empty1", "NR",
>     paste("D", 1:30, sep=""), "I", "NORM", "II", "NORM", "III", "NORM",
>     "MND", "NORM", "Locatie"))
>
>     creates structure based on what is in the text object (hence I first
>     open the text file and then figure out the code to read the values).
>     Of course, the structure of the raw KNMI files could have changed
>     from 2009.
>
>     If you are planning to interpolate monthly rainfall data take also a
>     look at the two geostatistical packages that elaborate geostatistics
>     with skewed values / variables with skewed distributions (I have not
>     updated the
>     http://spatial-analyst.net/__wiki/index.php?title=Spatial___interpolation_exercises_(NL)
>     <http://spatial-analyst.net/wiki/index.php?title=Spatial_interpolation_exercises_(NL)>
>     pages for a long time; sorry for that):
>
>     http://cran.r-project.org/__package=georob
>     <http://cran.r-project.org/package=georob>
>     http://cran.r-project.org/__package=geostatsp
>     <http://cran.r-project.org/package=geostatsp>
>     described in -> http://pbrown.ca/geostatsp/__document-rev.pdf
>     <http://pbrown.ca/geostatsp/document-rev.pdf>
>
>     HTH,
>
>     T. Hengl
>
>
>     On 23-11-2014 23:51, Mengxi Yang wrote:
>
>         Dear list,
>         I have a question on write.table. I found an nice exercise on
>         spatial
>         interpolation from
>         http://spatial-analyst.net/__wiki/index.php?title=Spatial___interpolation_exercises_(NL)#__Spatial_prediction_of_daily___precipitation_in_the___Netherlands
>         <http://spatial-analyst.net/wiki/index.php?title=Spatial_interpolation_exercises_(NL)#Spatial_prediction_of_daily_precipitation_in_the_Netherlands>
>
>         In this exercise, the date is 200906. but when I change to
>         201310. there is
>         a wrong message like:
>
>             tmp <- readLines(paste(getwd(), "/", "monv_201310.txt", sep=""))
>             tmp <- gsub(pattern=' . ', replacement=' 0.0 ', tmp, fixed =
>             TRUE)
>             tmp <- gsub(pattern='[[:space:]]{2,7}__', replacement=',', tmp)
>             write.table(t(matrix(unlist(__strsplit(tmp, ",")[-c(1:8)]),
>             nrow=41)),
>
>         "tmp.txt", col.names=FALSE, row.names=FALSE)
>         Warning message:
>         In matrix(unlist(strsplit(tmp, ",")[-c(1:8)]), nrow = 41) :
>             data length [13625] is not a sub-multiple or multiple of the
>         number of
>         rows [41]
>
>
>         I checked several month, some success but some not.
>         Does anyone know what is the wrong with it? should I change
>         something?
>
>         Thank you very much.
>
>         Mengxi Yang
>
>                  [[alternative HTML version deleted]]
>
>         _________________________________________________
>         R-sig-Geo mailing list
>         R-sig-Geo at r-project.org <mailto:R-sig-Geo at r-project.org>
>         https://stat.ethz.ch/mailman/__listinfo/r-sig-geo
>         <https://stat.ethz.ch/mailman/listinfo/r-sig-geo>
>
>