Global Network of Isotopes in Precipitation (GNIP)
Welcome to the web page of the Global Network of Isotopes in Precipitation (GNIP). The IAEA's Water Resources Programme and the World Meteorological Organization (WMO) have been surveying the stable hydrogen and oxygen isotope and tritium composition in precipitation around the globe since 1961.
Isotopic data collected as part of the GNIP network serves as an indispensable database to inform a range of scientific disciplines, including, but not limited to hydrology, meteorology and climatology, oceanography and limnology, and in studies related to the Earth's water cycle and climate. Recently, GNIP has played an important new role in facilitating novel isotope research in ecological investigations, food authentication and traceability, and forensics.
A Short History of GNIP
The Global Network of Isotopes in Precipitation was initiated in 1958 by IAEA and WMO, and became operational in 1961. The objective was a systematic collection of basic spatial data on the isotope content of precipitation across global scales to determine temporal and spatial variations of both environmental stable isotopes and tritium in precipitation. While the initial driver, was to monitor atmospheric thermonuclear test fallout through the determination of the radioactive hydrogen isotope Tritium, since the 1970s the focus changed to a become observation network of stable hydrogen and oxygen isotope data for hydrologic studies.
For over 50 years, GNIP has provided global isotope data for the use in hydrological investigations water resources investigation, planning, conservation and development. Although these primary objectives remain as an important feature, over the past decade other scientific disciplines have increasing made use of this invaluable and unparalleled isotope database, including:
(i) verifying and improving atmospheric circulation models,
(ii) studying regional, global and temporal climate ,
(iii) studying the interactions of water between atmosphere and biosphere, and
(iv) providing baseline information for the authentication of commodities, such as food, plants, and for tracking migratory species (birds, fish, butterflies etc.), and for forensic purposes.
Mode of Operation
Precipitation samples are collected in cooperation with WMO, national meteorological services, national authorities and scientific volunteers. Since 1961, over 1,000 meteorological stations in more than 125 countries and territories have collected monthly precipitation samples as part of the Global Network of Isotopes in Precipitation. Stations participating in the GNIP collect monthly composites of rainfall for deuterium and oxygen-18 analysis (selected stations also sample for Tritium). The samples are mainly analysed in the IAEA´s Isotope Hydrology Laboratory in Vienna, but are also measured in cooperating laboratories. The technical procedures for GNIP sampling and shipping are available as leaflets in PDF format:
Ancillary data (e.g. temperature, rainfall, etc.) collected by GNIP stations that belonging to national networks are also provided to the IAEA, and are included in the database. Due to different publishing policies, delays may occur in the posting of national network data to the GNIP.
GNIP Station Information
The system of station index numbers is based on that of the WMO for identifying observation stations, but the numbers, names and locations used in the IAEA network may sometimes differ from those employed by WMO. This station number has the general format II iii jj, i.e. the block number (II), the station number (iii) and the substation number (jj). In addition, each station has a substation number (jj) for which the WMO index system does not have an equivalent. When a sampling station corresponds to the WMO station, the substation number will be 00. When a sampling station is not located at the WMO station location, it is assigned the block code and station number of the nearest WMO station with a substation code number 01, 02, 03, etc. Station index numbers for samples from the Antarctic were allocated under a special WMO scheme.
Official WMO station names are used for all stations hosted at a WMO meteorological station. Otherwise, the best known geographic name or the name used by the laboratory submitting the isotopic and climatic data is used.
Latitude, longitude and elevation
Latitude and longitude are given in decimal degrees. Station locations were carefully reviewed in 2012; nevertheless the coordinate accuracy of some of the 'traditional' stations does not reflect current technological capabilities. Altitude is given in metres above (below) mean sea level. For WMO stations it is the elevation of the average level of the terrain in the immediate vicinity of the sampling point.
For easy grouping of stations with similar climatic characteristics, each GNIP station was classified according to the Köppen-Geiger climate classification scheme (Kottek et al 2006). This information was derived from the WMO stations' monthly climatological normals or the NOAA Global Historical Climate Network (GHCN, Petersen & Vose 1997), from the GNIP data themselves, or from a nearby WMO station in case neither of the above is available.
Date of Sampling
The samples are typically composites from the total precipitation at a station during one calendar month. The time of collection is identified under the columns year and month. For any given month, the day entered is 15 and the date accuracy is noted as 'monthly'. For daily or event-based samples, the sampling day is shown.
Basic meteorological information is given for each station or location for which isotope data are reported, if available. Other basic climatic information is taken from WMO publications [1, 2, 3] or received in answer to enquiries made to WMO or co-operating laboratories submitting the isotope data. This meteorological information may include the following:
Precipitation in millimetres (mm)
Total monthly precipitation values are rounded off to the nearest unit for data before 2000. Therefore, trace amounts of precipitation in a given month appear as 0 if they are less than 0.5 mm. After 2000, precipitation data is given in mm and tenths of mm.
Type of Precipitation:
The type of precipitation falling during the month is given when known, and recorded as 'sample medium':
• Precipitation – Rain
• Precipitation – Rain and Snow ("mixed")
• Precipitation – Snow
• Precipitation – Dry (for months during which no precipitation was recorded, but for which a sample was created for the sake of consistency, storing temperature and vapour pressure information.
given in degrees Celsius and tenths.
Water vapour pressure is given in millibars (mb). The vapour pressure values are calculated values obtained from temperature and relative humidity. For details on this calculation, please refer to [link to WMO]
Deuterium and Oxygen-18
Deuterium and oxygen-18 contents of the precipitation samples are measured by traditional isotope-ratio mass spectrometry and/or by laser absorption spectroscopy as ratios of hydrogen-2/1 and oxygen-18/16. They are expressed as the per mille (‰) deviation of the isotope ratio from the primary measurement standard, expressed in the typical delta notation (δ2H and δ18O). The primary reference measurement standard is VSMOW (Vienna Standard Mean Ocean Water) as defined by Craig  and Gonfiantini . The measurements reported in GNIP have a long term precision of about ±0.1‰ for oxygen-18 and ± 0.8‰ for deuterium, at one standard deviation.
The tritium content in each precipitation sample is given in tritium units (TU) and is corrected for decay to the middle of the month. One TU is defined as one tritium atom per 1018 atoms of hydrogen, equivalent to 0.118 Becquerel (Bq) per litre of water. The tritium half-life has been fixed as 12.32 years  for data since 2000. The analytical error for tritium analysis is given in TU when known. In several cases, the tritium values reported for a given station substantially exceed the expected levels, indicating local sources of technigenic tritium and/or contamination during collection of the samples (e.g. Bern, Switzerland or St. Petersburg, Russia).. Not all of the GNIP stations are used for tritium data, only a selected subset of representative locations.
Quality and availability of the Data and statistics
To date, more than 115,000 GNIP precipitation samples have been logged into the database. The quality of the stored isotopic information is periodically reviewed. When entering data into the database, obvious errors (typing mistakes, reversed signs, etc.) are eliminated or corrected in the first screening. The second part of the data quality control consists of the rejection of clear outliers, identified according to the chosen statistical criteria . The GNIP database undergoes frequent updating: new data, corrections or revisions received from participating laboratories are introduced at periodic intervals, striving to achieve public availability as soon as possible. Nevertheless, holding periods may apply at the request of some national networks or some externally contributed data.
The statistical evaluation of isotope and meteorological data is conducted as published in the IAEA Technical Reports Series No. 331 'Statistical Treatment of Data on Environmental Isotopes in Precipitation'. A dynamic computation of these statistical parameters is provided via the web portal WISER (Water Isotope System for Data Analysis, Visualization and Electronic Retrieval).
The basic features of the spatial and temporal distribution of deuterium and oxygen-18 in global precipitation, as derived from the IAEA/WMO database, were discussed in a recent review paper .
Obtaining and citing GNIP data
GNIP data and statistics can be obtained free of charge from the WISER application . To cite GNIP in report and publications, use the following parameters (a modification according to the particular journal's guideline may be necessary):
IAEA/WMO (current Year). Global Network of Isotopes in Precipitation. The GNIP Database. Accessible at: http://www.iaea.org/water
In addition, a number of stations may have a literature reference. If a substantial number of the GNIP data used comes from published stations, kindly cite also the associated publication. Please be aware that proper citation helps the researchers and institutes involved to affirm the importance of their work and contributions to GNIP and that failure to properly cite data may undermine voluntary contributions to GNIP.
Become involved in GNIP
GNIP is essentially a dynamic network of voluntary data and sample contributions. This accordingly results in different operating periods for individual stations as national and individuals project funding and resources allow. IAEA strives to assure sufficient spatial network coverage, nevertheless we are very happy to receive proposals for additional stations. The support IAEA may provide depends on the location and background of the institution hosting the station.
In addition, IAEA warmly welcomes your published and unpublished data contributions. If you have collected and/or published precipitation isotope data, and you wish to make your results available via the GNIP repository to the isotope community, please consider incorporating your data into the GNIP database. Provided that the sampling technique is compliant with the GNIP procedures, we shall be happy to assist you to integrate datasets into the GNIP-Monthly or GNIP-Events (for any other sampling interval) databases. If you wish to get involved or contribute data, please do not hesitate to contact the GNIP team via the contact point firstname.lastname@example.org.
- 1. WORLD METEOROLOGICAL ORGANIZATION, Climatological Normals (Clino) for climate and climate ship stations for the period 1932-1960, WMO/OMM No. 17, TP.52.
- 2. WORLD METEOROLOGICAL ORGANIZATION, Weather Reporting, Volume A: Observing Stations, WMO/OMM No. 9, TP.4.
- 3. WORLD METEOROLOGICAL ORGANIZATION in Co-operation with Environmental Science Services Administration USA, Monthly Climatic Data for the World, United States Government Printing Office, monthly editions.
- 4. LUCAS, L. L., UNTERWEGER, M. P., Comprehensible Review and Critical of the Half-life of Tritium. Journal of Research of the National Institute of Standards and Technology, 105-4 (2000), 541-549.
- 5. CRAIG, H., Standard for reporting concentration of deuterium and oxygen-18 in natural waters, Science 113 (1961) 1833.
- 6. GONFIANTINI, R., Standard for stable isotope measurements in natural compounds, Nature (London) 271 (1978) 534-536.
- 7. INTERNATIONAL ATOMIC ENERGY AGENCY, Statistical Treatment of Data on Environmental Isotopes in Precipitation, Technical Reports Series No. 331, IAEA, Vienna (1992) 781.
- 8. TAYLOR, C.B., Stable Isotope Composition of Monthly Precipitation Samples Collected in New Zealand and Rarotonga, Physical Science Rep. 3, Dept. of Scientific and Industrial Research, Lower Hutt, New Zealand (1990) 93.
- 9. ROZANSKI, K., ARAGUAS ARAGUAS, L., GONFIANTINI, R., Isotope Patterns in Modern Global Precipitation, Geophysical Monograph 78. In: Climate Change in Continental Isotope Records, American Geophysical Union (1993) 1-36.
- 10. INTERNATIONAL ATOMIC ENERGY AGENCY. Environmental Isotope Data No. 1-10: World Survey of Isotope Concentration in Precipitation.
- No. 1 (STI/DOC/10/96) 1953-1963, 421 pages, published in 1969
- No. 2 (STI/DOC/10/117) 1964-1965, 404 pages, published in 1970
- No. 3 (STI/DOC/10/129) 1966-1967, 421 pages, published in 1971
- No. 4 (STI/DOC/10/147) 1968-1969, 404 pages, published in 1973
- No. 5 (STI/DOC/10/165) 1970-1971, 421 pages, published in 1975
- No. 6 (STI/DOC/10/192) 1972-1975, 404 pages, published in 1979
- No. 7 (STI/DOC/10/226) 1976-1979, 421 pages, published in 1983
- No. 8 (STI/DOC/10/264) 1980-1983, 404 pages, published in 1986
- No. 9 (STI/DOC/10/311) 1984-1987, 188 pages, published in 1990
- No. 10 (STI/DOC/10/371) 1988-1991, 200 pages, published in 1994