# where to download NASA-NEX daily downscaled datasets?

I'm looking to download a small geographic area of the CMIP5 models part of NASA-NEX dataset. Previously I relied on the Planet OS OpenNEX GDDP website which had a great tool for selecting geographic areas, models and time periods, saving in this way a large amount of downloading time and disk space. The problem is that OpenNEX is gone and I haven't been able to find a similar tool.

Is there another tool/website/script that can be used similarly?

## 2 Answers

Google Earth Engine's NASA NEX dataset (or this) might be of interest to you. It is an online repository, where you can do your analysis using the GEE platform without having the need to download to local machine.

Besides the Google Earth application suggested by shahryar I'm not aware of an online tool that could do the job you are looking for.

What I would do is to download the files and operate with a NCO or CDO to select the domain you are after (so that you can get rid of the "big" file immediately after and save disk space).

If you are working on a windows machine and cannot use NCO or CDO, a way to automatize at least the downloading procedure would be a script in R. Something like this:

library(RCurl)
opts = curlOptions(proxy='my_proxy', userpwd = "mypassword", netrc = TRUE) ##you need to get the credentials beforehand

var = c("pr","tasmax","tasmin") # put variables here
rcp = c("rcp45", "rcp85") #put the Climate scenarios here
mod= c("CCSM4", "GFDL-CM3", "IPSL-CM5A-MR", "MIROC5",  "inmcm4") #put the GCM models you want here
year=c(seq(2036,2050,1), seq(2086,2100,1)) ##here you can select the years you are looking for

for (v in var) {
for (r in rcp) {
url<- paste0( 'ftp://ftp.nccs.nasa.gov/BCSD/', r, '/day/atmos/', v, '/r1i1p1/v1.0/', sep='')
for (m in mod) {
for (y in year) {
nfile<- paste0(v,'_day_BCSD_',r,"_r1i1p1_",m,'_',y,'.nc', sep='')
url1<- paste0(url,nfile, sep='')
destfile<-paste0('your_destination_path_NASA_NEX/',r,'/',v,'/',nfile, sep='')
bin <- getBinaryURL(url1, .opts = opts)
writeBin(bin, destfile)
Sys.sleep(1)
gc()
}
}
}
}


In the middle of the loop, using ncdf4 (give a look here) you can nest few lines to pick the portion of the data covering your domain, store it, and get rid of the original file you just downloaded. This is time consuming though. Surely it would be more efficient to use a combination of wget (as explained here by NASA) and then NCO or CDO.