Site-specific weather data is required in order to perform historical simulation of power plants and similar systems. A long record of information is needed to adequately capture the range of operating conditions. An excellent source of such data is the Global Surface Summary of the Day (GSOD) database maintained by the National Climate Data Center (NCDC) operated by the National Oceanographic and Atmospheric Administration (NOAA). Data from thousands of meteorological stations around the world are packaged in "tar balls" (LINUX zip files), one for each year, available at their site ftp.ncdc.noaa.gov/pub/data/gsod/ While these files provide daily values, there is enough information to infer hourly behavior using the method of Waichler and Wigmosta described in, "Development of Hourly Meteorological Values From Daily Data," Proceedings of the American Meteorological Society, 2003.
This is just one example. Waichler and Wigmosta's equations are easily implemented in an Excel spreadsheet. By adding psychrometric functions (that I have provided elsewhere on RG), you can add relative humidity and other moist air properties too.
Thanks for your information.
However, daily data from Global Summary of the Day, have missing data, while some data for some stations are not available.
More importantly, data from other sources can also be acquired and used for research.
Though daily data from GSOD correspond with the conventional data obtained from meteorological agencies in each of the countries.
Daily data from GPCC have 99% accuracy with data collected from meteorological agencies.
I prefer to use raw data directly with no third party involved. I am always suspicious when it comes to climate data.
Why don’t you use datasets from various other sites such as ECMWF and Giovanni?
Several reasons: 1) the GSOD data goes back to 1929; 2) the whole world in a single tar file for each year; 3) always the same file format; 4) I trust NOAA; 5) it's free forever; 6) the files contain a variety of other related measurements, although not as complete coverage for anything but temperature and barometric pressure. For me, the choice is clear.
Dear Dudley J Benton , it seems that the website (ftp.ncdc.noaa.gov/pub/data/gsod/) is not accessible. Can you send us an accessible link?
Regards.
ftp://ftp.ncdc.noaa.gov/pub/data/gsod/
I just checked it. There's a folder for each year starting in 1929. The list of stations and countries is in the root folder along with a description of what the files contain. If you can't access the site, then something is blocking you, perhaps antivirus or perhaps internet provider or some international restriction based on your IP address. Can you access the NOAA site? https://www.noaa.gov/
I can access to https://www.noaa.gov/ link but, I can't access to ftp://ftp.ncdc.noaa.gov/pub/data/gsod/
Then there must be something blocking your connection. Let's hope is just the ftp:// vs. https:// and your web browser and not some access restriction. I don't know how else to access this information and it's much too large to send via email. You might try contacting NOAA through their website to see if there are any special requirements for access. The NCDC server is physically located in North Carolina.
When you (or others) do get the data, you will need the "untar" and "gzip" utilities to first unpack the UNIX tar balls and then expand the individual *.gz files within. Both are available on the Web for LINUX or Windows. I also have the source code plus a library that lets me read the files from within a program without having to actually unpack them at once. There are about 10,000 files per year, so having them all in one folder at the same time is a mess to navigate and clean up. It sounds like a lot of work but it's well worth the effort to have so much information readily available.
Dudley J Benton , thank for your suggestions. I will get back to you if the need arises. Regards.
- Badges
- Science topic