find number of consecutive days in NC files by R

Hi jim,



I have a new question. I have 71 years raster data from 1949 to 2019 and ? am trying to convert these 71 yeears raster an array to calculate a liner regration.





library(raster)

r<-raster("C:/Teaching/MSCprojects/2020/Ahmet/soilm/max_consecutive_days/max_consecutive_days_1949.tif")

a<-array(NA,dim=c(dim(r)[1:2],70))

i <- 1

for (year in 1949:2019) {

  fi<-paste0("C:/Teaching/MSCprojects/2020/Ahmet/soilm/max_consecutive_days_",year,".tif")

  r<-raster(fi)

  a[,,i]<-getValues(r,format="matrix")

            i<-i+1

}

Best wishes,

Windows 10 i?in Posta ile g?nderildi



Kimden: Jim Lemon
G?nderilme: 10 A?ustos 2020 Pazartesi 06:28
Kime: ahmet varl?
Konu: Re: [R] find number of consecutive days in NC files by R



That's right. If you want to get the result for all cells with valid
readings, step through the cells, texting for valid data. The result
will be quite large, so I suggest sending the output to a file:

sink("soil_moisture_result.txt")
for(i in 1:nrows(soil_moist)) {
 for(j in 1:ncols(soil_moist)) {
 if(sum(!is.na(soil_moist[i,j,]) > 0) {
  # process the cell here and print out the result
 }
}
sink()

caution: untested

Jim

On Mon, Aug 10, 2020 at 2:02 PM ahmet varl? <varli61 at windowslive.com> wrote:

Hi Jim,



I would like to find out how many consecutive days each cell is under the specific value for a certain date range. ?f I am right your code is for just one cell





Kimden: Jim Lemon
G?nderilme: 10 A?ustos 2020 Pazartesi 04:23
Kime: ahmet varl?; r-help mailing list
Konu: Re: [R] find number of consecutive days in NC files by R



Hi Ahmet,
An easy way is this:

library(ncdf4)
soilm<-nc_open("soilw.0-10cm.gauss.1949.nc")
soil_moist<-ncvar_get(soilm)
smdim<-dim(soil_moist)
# identify NA grid cells
sm_NA_count<-matrix(NA,nrow=smdim[1],ncol=smdim[2])
for(i in 1:smdim[1]) {
 for(j in 1:smdim[2]) {
  sm_NA_count[i,j]<-sum(!is.na(soil_moist[i,j,]))
 }
}

The resulting matrix contains the counts of valid (not NA) values in
each 365 day series in the array. It looks to me as though there are
5914 complete series and the rest are all NA. This does not tell you
why some files (the third dimension) are all NA. Probably the best
guess is that the soil moisture content is not measurable for some
reason. Here is the explanation from NOAA:

Missing Data:

There is no missing data though the ocean has 0's. There is a file
with the percent of the grid that is land. Another file has simply 1
and 0's for land/ocean. Grids where the percent of land is zero are
"missing".

You can get a feel for the geographic coverage like this (white cells
are not NA):

library(maps)
library(plotrix)
color2D.matplot(t(sm_NA_count))

Jim

On Mon, Aug 10, 2020 at 4:09 AM ahmet varl? <varli61 at windowslive.com> wrote:

Hi Jim,



Could you help me to remove NA values which are water values ?





Kimden: Jim Lemon
G?nderilme: 7 A?ustos 2020 Cuma 22:53
Kime: ahmet varl?
Konu: Re: [R] find number of consecutive days in NC files by R



There are 17848 grid cells in the file I downloaded for 1949. Many of
them only contain NA values, probably because they are from a
geographic grid that is covered by water. In the code there is a
section that prints out a list of the grid cells that contain minimum
values less than 0.3. Since I don't know which grid cell you are
using, I had to find one that would produce interpretable results for
the problem you are trying to solve.

Jim

On Fri, Aug 7, 2020 at 11:03 PM ahmet varl? <varli61 at windowslive.com> wrote:

I am greatfull for your helps and ? just want to ask why did you use cell 159,66