{"id":1893,"date":"2014-11-25T18:25:08","date_gmt":"2014-11-26T02:25:08","guid":{"rendered":"http:\/\/lukemiller.org\/?p=1893"},"modified":"2018-03-12T09:57:18","modified_gmt":"2018-03-12T16:57:18","slug":"extracting-noaa-sea-surface-temperatures-with-ncdf4","status":"publish","type":"post","link":"https:\/\/lukemiller.org\/index.php\/2014\/11\/extracting-noaa-sea-surface-temperatures-with-ncdf4\/","title":{"rendered":"Extracting NOAA sea surface temperatures with ncdf4"},"content":{"rendered":"

I’ve written previously about some example R scripts<\/a> I created to extract sea surface temperature data from NOAA’s Optimum Interpolated Sea Surface Temperature products. If you want daily global sea surface temperatures on a 0.25×0.25\u00b0 grid, they gather those into 1-year files available at http:\/\/www.esrl.noaa.gov\/psd\/data\/gridded\/data.noaa.oisst.v2.highres.html.<\/a> If you want weekly average SST values on a 1×1\u00b0 grid, you can get those back to 1981 at http:\/\/www.esrl.noaa.gov\/psd\/data\/gridded\/data.noaa.oisst.v2.html<\/a>. If you just want a smaller file for a single day on the 0.25×0.25\u00b0 grid, you can find those at http:\/\/www.ncdc.noaa.gov\/sst\/griddata.php<\/a>.<\/p>\n

\"Ocean<\/a>
Ocean temperatures around Africa for 2014-11-24.<\/figcaption><\/figure>\n

The high-resolution daily data currently appear to use a mixture of NetCDF3 formats for 2013 and earlier, while the 2014 data are NetCDF4. The older files can be opened by David Pierce’s ncdf\u00a0 <\/a>package, which is easily installed from CRAN on all platforms. The newer NetCDF4 files cannot be opened by the ncdf package, but David Pierce’s newer ncdf4 package<\/a> will handle both older and newer file formats. That package should be available for OSX and Linux users directly from CRAN, but it is not available for Windows users directly. Instead, you can currently visit David Pierce’s ncdf\/ncdf4 site<\/a> where there are Windows-compatible versions of the package available. To install that .zip file, simply save it to your computer. Open an instance of Rgui.exe. Go to the menu item “Packages>Install package(s) from local zip file” and point R to your copy of the zip file. If that works, you should be able to call library(ncdf4)<\/code> at the R command line and get access to all of the package’s functions. Fun! (Edit 2018: <\/strong>ncdf4<\/a><\/code> has been available on CRAN for some time now, so you should be able to just run install.packages('ncdf4')<\/code> from the R command line now instead of downloading and installing the zip files)<\/p>\n

With that caveat out of the way, I’ve created a new version of my R script<\/a> called NOAA_OISST_ncdf4.R<\/code> that contains three functions for extracting sets of SST data from the three types of OISST files (weekly, daily, single-day files), and a simple plotting function. The script is found at https:\/\/github.com\/millerlp\/Misc_R_scripts\/blob\/master\/NOAA_OISST_ncdf4.R<\/a>. Save that file into your R working directory as NOAA_OISST_ncdf4.R, then call source(\"NOAA_OISST_ncdf4.R\")<\/code> at the R command line to load up the four functions. The functions are<\/p>\n