Is there any institute out there collecting time series data of precipitation, and making them accessible online? I need data for a few different locations, but:

  • Low spatial resolution is fine. For example, if I need rainfall time series for France, a space-averaged value would be satisfactory.

  • Low accuracy is fine. For example, remote sensing rather than in situ, or rainfall measured in a few points and then interpolated, would be satisfactory as well.

  • 1
    $\begingroup$ The question is spot-on for this site. Welcome! $\endgroup$ Jan 10, 2015 at 18:00
  • $\begingroup$ Definitely agree with @milancurcic, I will be very interested to read the answer(s) to this and are posting a link to it on my Twitter account. $\endgroup$
    – user889
    Jan 10, 2015 at 18:16

5 Answers 5


Yes. Nearly global satellite and radar derived rainfall data are hosted by NASA Goddard Space Flight Center, namely:

  • Tropical Rainfall Measurement Mission (TRMM) aboard the TRMM Microwave Imager (TMI), a multi-channel, dual polarized, conical scanning passive microwave radiometer designed to measure rain rates over a wide swath under the TRMM satellite. TRMM was launched on November 27, 1997 into a semi-equatorial (+/- 35 deg) orbit carrying 5 instruments. Three levels of data processing are available on the NASA's server, where higher level indicates more processing, quality control, and merging of multiple raw datasets. For your purposes, use level 3 TRMM as it combines satellite data from multiple swaths with precipitation radar data and other sources, and provide 3-hourly data in the 60S-60N latitude band, at 0.25 degree horizontal resolution.

  • Global Precipitation Measurement mission was launched in early 2014, and provides full global coverage at 0.1 degree horizontal resolution at 30 minute intervals in time. This product is in its very early stages, and is still under development and testing.

In both cases, if you are interested in more precise rainfall estimates at a particular point in space and time, look at level 1 data. It is raw data that provides a time step for every single point along an orbital trajectory of the satellite. Use with caution.

In addition, ground validation data is available on the same server, and some or most of these data is assimilated (merged) into level 3 TRMM and GPM products.

Once you download the data, you can use their data viewer, THOR, or use your favorite programming language or analysis software to display it yourself.

  • $\begingroup$ we use TRMM as well. But is it going to be around for much longer or is that a separate question ? $\endgroup$
    – user1066
    Jan 11, 2015 at 2:24
  • 1
    $\begingroup$ I was hoping into some kind of a wider database, as these series can't be used yet for a long-term back analysis - less than 20y of data for the oldest. Anyway will check them out. Thank you. $\endgroup$
    – natario
    Jan 11, 2015 at 14:16
  • $\begingroup$ @gansub Regarding TRMM shutdown: TRMM observations will continue as the spacecraft descends to 335 km, at which time it will be shutdown prior to re-entry. This date is nominally estimated to be February 2016, but may change depending on solar activity. According to the NASA TRMM site. $\endgroup$
    – decvalts
    Jan 13, 2015 at 17:19

If you are looking for a dataset going back a bit further, but still sticking to satellite and radar data, GPCP (Global Precipitation Climatology Project) goes back to 1979. The limitation of certain TRMM datasets, in addition to the relatively short length of the project, is that they only cover the tropics and so you lack data in the mid-latitudes and higher for a proper global analysis. Some TRMM datasets extend up to 60 degrees latitude by incorporating radar and other sources into the final product.

If you want to go back even further (to 1901 globally) the GPCC (Global Precipitation Climatology Centre) is a good place to start. Anything before the launch of TRMM in 1997 will likely be a collation of various data sets, so the comparability between regions is occasionally questionable, although there are robust ways of correcting for this applied to the datasets. This is a truly global data set, according to their spatial coverage description (90.0N - 90.0S, 0.0E - 360.0E)

Beyond that, the datasets are patchy between different countries. You would have to collate your own global dataset from the various national meteorological archives. The UK precipitation archives (Armagh Observatory), for example, go back to about 1795, but I'm not aware of them being included in many (any?) global collations.

N.B. Almost all of this data will be in some form of Cartesian or geographic co-ordinate format, so you probably won't be able to look up individual places or countries, but will need to extract latitude and longitude bounded regions from the data.

Link to the GPCC download site: GPCC gridded data download

  • 3
    $\begingroup$ +1 for additional datasets. Worth noting is that GPCC climatology provides monthly accumulated rainfall, so it's rather coarse in time. Also note that TRMM level 3 has a coverage between 60S and 60N, as the radar and gauge data are added in. If monthly accumulated rainfall is satisfactory, it should be preferred over satellite data in my opinion, as it is more accurate. $\endgroup$ Jan 13, 2015 at 17:18
  • $\begingroup$ Good points, yes I forgot about the extended possibilities of TRMM up to 60N/S. $\endgroup$
    – decvalts
    Jan 13, 2015 at 17:22

WorldClim is an alternative. It is a set of climate layers with spatial resolution of 1 square kilometer. It provides current conditions, forecast and historical data.

Besides rainfall there are data available for Min. Temperature, Temperature, Mean Temperature, Bioclim and Altitude.

Data is divided in tiles and available in GeoTiff or ESRI grid format.

  • $\begingroup$ This is a great resource, the OP is asking for time series. WorldClim is not a time series but a mean climatology including values such as 'average monthly minimum temperature'. It does give these values for historic, current, and future climates but these are not 'time series'. $\endgroup$
    – Abe
    May 21, 2016 at 3:56

One option is NOAA's Climate Data Online service. From that page you can use the Mapping Tool, view Time-Related Maps, and select Hourly/Sub-hourly view. You can then check the Global Hourly dataset to see stations worldwide which have hourly* precipitation data.

*This data (Integrated Surface Data or ISD) tends to include less than hourly temporal resolution outside of the U.S. and some major international airports, but you'll still have finer-than-daily resolution in most cases. In terms of spatial resolution, there are many stations included in this service around the world as you'll see - over 20,000 stations with better coverage (and quality data) in some areas than others. Selecting one or more of these stations and downloading their weather data provides a time-series as a digital table. Some examples: text file; web form.

You can view metadata for the CDO's Global Hourly data here: https://gis.ncdc.noaa.gov/geoportal/catalog/search/resource/details.page?id=gov.noaa.ncdc:C00532 On that page it states the purpose of this dataset as:

Global Hourly Data provides a long-term record of hourly, sub-hourly and synoptic weather observations from station networks around the world. Observation sources include the World Meteorological Organization, Automated Surface Observing System, Automated Weather Observing Stations, US Climate Reference Network, and others. Weather elements observed include sky conditions, visibility, weather type, dry bulb, wet bulb, and dew point temperatures, relative humidity, wind speed and direction, wind gusts, station and sea level pressure, altimeter, and precipitation. Online data are typically delayed for 24 hours.


You can try CMORPH data set- CMORPH Description. As explained in the original reference - CMORPH publication the authors obtain half hourly global precipitation estimates derived from passive microwave satellite scans and they are propagated by motion vectors derived from geostationary IR data. So these are spatially and temporally complete microwave based precipitation estimates independent of the IR temperature field.

The reason why it is called a "MORPH" is because shape and intensity of the precipitation estimates are modified during the time between two successive microwave scans by a time weighted linear interpolation scheme. The flexibility of CMORPH indicates it can enhance the value Global Precipitation Mission Data by combining IR scans with GPM and hence increasing the temporal resolution of GPM to thirty minutes.

CMORPH can be downloaded from Download URL.

If you are only downloading via OPeNDAP you can get it from the RDA RDA CMORPH1 and RDA CMORPH2


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.