Data Set Citation:
When using this data, please cite the data package:
NCEAS 2017: Prince: Global Primary Production Data Initiative , National Center for Ecological Analysis and Synthesis , and Esser G.
Osnabruck NPP Dataset: Worldwide Estimates and Bibliography of Net Primary Productivity derived from Pre-1982 Publications
nceas.154.10 (
General Information:
Title:Osnabruck NPP Dataset: Worldwide Estimates and Bibliography of Net Primary Productivity derived from Pre-1982 Publications
Alternate Identifier:ORNL-1997-2017-Osnabruck
An extensive compilation of field data on net primary productivity (NPP) of natural and agricultural ecosystems worldwide was synthesized in the 1970s and early 1980s by Prof. H. Lieth, Dr. G. Esser and others. Much of this work was carried out at the University of Osnabrueck, Germany. More than 700 single point estimates of NPP or biomass were extracted from the scientific literature, each with a geographical reference (latitude/longitude). The literature cited dates from 1869 to 1982, with the majority of references from the 1960s and 1970s. Although this data set has not been updated since the 1980s, it represents a wealth of information for use in model development and validation. In the early 1970s, a subset of these NPP data was used by Lieth, Esser and co-workers to develop and test a series of statistical-correlative models of NPP as a function of mean annual temperature and precipitation. The later versions of these models included modifications for soil, seasonality, agriculture, and other human influences ("Osnabrück Biosphere Mode,""High Resolution Biosphere Model," etc.). Most of the 720 unique NPP records (632, or 88 percent) have been matched to a bibliography of 356 references from the primary literature. The original form of this bibliography contained many more references than records, including multiple sources for the same author and study, as well as additional references to data on standing biomass, soils, and so forth. Since this is a useful resource in its own right, an edited and corrected compilation of these 858 references is available here with the cross-references to the NPP records highlighted. Of the 720 unique NPP records, about two-thirds have above-ground NPP estimates that range between 1 and 8530 g/m2/year (dry matter) -- or 2923 g/m2/year, excluding doubtful values, wetlands, and crops/pastures and other likely managed systems. Total NPP, for which more than half of the sites have estimates, ranges from 3 to 9320 g/m2/year (dry matter) -- or 3580 g/m2/year, excluding doubtful values, wetlands, and crops/pastures and other likely managed systems. Each record includes a site identifier, latitude, longitude, author, country, NPP estimates, vegetation type, and other variables. The vegetation-type field begins with a generalized biome type (including tundra, forest, Mediterranean, savanna, grassland, desert, wetland, and a number of managed vegetation types) and is followed by more specific vegetation terminology derived from the original data. Caution is advised in using these biome/vegetation types because they were not defined consistently within the original data set and nearly 200 sites lack any vegetation designation. To achieve completeness in a single synthesis file, a single NPP value (NPP_C) is included for each site that represents the sum of above-ground (ANPP) and below-ground (BNPP) components, expressed in grams of carbon per square meter per year (g C/m2/year). Where BNPP was not reported, it was assumed to be equal to ANPP. A ratio of 0.475 was used to convert dry biomass weight to carbon content. Total NPP was estimated as TNPP (where available), or as the sum of ANPP and BNPP (or from ANPP x 2, if BNPP was not estimated), and then converted to g C/m2/year.
  • net primary productivity NPP (theme)
  • biomass (theme)
  • agricultural ecosystems (theme)
  • wetland (theme)
  • savanna (theme)
  • Mediterranean (place)
  • tundra (theme)
  • forest (theme)
  • grassland (theme)
  • desert (theme)

Involved Parties

Data Set Creators:
Organization:NCEAS 2017: Prince: Global Primary Production Data Initiative
Organization:National Center for Ecological Analysis and Synthesis
Individual: G. Esser
Institute of Plant Ecology Justus-Liebig University Heinrich-Buff Ring 38,
Giessen, D-35392 Germany
49 (641) 99 35310 (voice)
49 (641) 99 35309 (Fax)
Email Address:
Data Set Contacts:
Individual: J. M. O. Scurlock
Visiting Research Fellow, The Open University,
Milton Keynes, MK7 6AA UK
+44 (7986) 094193 (voice)
Email Address:
Associated Parties:
Individual: H. Lieth
Individual: J. M. O. Scurlock
Individual: R. J. Olson
Metadata Providers:
Individual: Andrea Chadden

Data Set Characteristics

Geographic Region:
Geographic Description:Point locations in the data are worldwide; data synthesis carried out at the University of Osnabruck, Germany.
Bounding Coordinates:
West:  -180  degrees
East:  180  degrees
North:  90  degrees
South:  -90  degrees
Time Period:

Sampling, Processing and Quality Control Methods

Step by Step Procedures
Step 1:  
Data acquisition: Several different versions of the Osnabruck data set were exchanged between Esser, Scurlock and Olson between 1988 and 1995. Starting with an October 1995 version and eliminating duplicates, these have been condensed into a list of 683 unique records by Scurlock and Olson. Most of these records (about 90 percent) were matched to a list of references from the primary literature. However, this bibliography contains more references than records, including multiple sources for the same author and study, as well as additional references to data on standing biomass, soils, and so forth. Since this is a useful resource in its own right, an edited and corrected compilation of these 856 references is provided here as an accompanying file. The NPP study sites represent primarily natural ecosystems; however, unusual sites, such as very young or very old forest stands, crops, fertilized pastures, and plantations are also included (and have been flagged where possible). The data set includes NPP data based on measurements collected over the past 100 years by investigators using a variety of methods and algorithms to estimate NPP. The scientific literature considered for inclusion in the data set was selected through a review of collections such as Biological Abstracts, Chemical Abstracts, Agricola, and Current Contents, as well as of a number of textbooks and monographs. Lieth, Esser and co-workers extracted data predominantly from primary publications, that is, those papers that described the original work. The minimum requirements for data to be considered were the following: 1) at least a vague geographical reference to the site of measurement (data related to vegetation types only were not considered) and 2) the use of one of the commonly accepted methods of assessing terrestrial NPP (Whittaker and Marks, 1975). Where the geographical coordinates of the experimental site were not included in the original paper, Lieth, Esser and co-workers selected the coordinates from maps or based them on site descriptions. Operational navigation charts (Defense Mapping Agency Aerospace Center, St. Louis Air Force Base, Missouri 63118, USA) and other regional and local maps were used for this purpose.
Quality Control Step 1:  
QA/QC procedures: The quality assessment process performed recently on these data by Scurlock and Olson consisted of selecting those records for which consistent information was available on (1) NPP, (2) latitude/longitude (corresponding to known land masses), (3) biome or vegetation type, and, as far as possible, (4) at least one literature reference. The criteria for consistency included the use of common systems of names, units, and so on. Names of countries and other data categories were translated from a mixture of English and German to English only. By sorting and re-sorting the records in order of each variable, it was possible to check for out-of-range values and to cross-check many suspect records against the original primary literature (or at least against the titles of the primary literature references). In certain cases where the primary literature was readily available (e.g., Ecology, Journal of Ecology, Oecologia, etc.), data, vegetation type, and geographical coordinates were checked more thoroughly (see also below). Further duplicates or near-duplicates were eliminated at this stage. Sites were re-mapped using Geographical Information System software, and suspect sites that were located in oceans or other unlikely areas were identified. Suspect data were checked against original records and corrected in the data set where necessary. The bibliography of original-source references in the Appendix is provided as a separate file. The references can be linked to the site records by the Author combined with the Year; we have found that only the first four characters of Author are generally needed (the data file usually contains the name of the first author, whereas the file of source references contains the full list of authors). The list of 856 references has been edited by Scurlock and Olson for consistency of citation style, since it was originally compiled by a number of individuals using different forms of abbreviations. Journal and book titles were checked against citation indexes and major libraries and are given in full where available. Duplicates, errors, and spelling and typographical mistakes have been eliminated as far as possible, as have obvious mismatches between journal volume numbers and years. A number of missing references described in the original data set have been restored. Literature that is not directly cross-referenced to NPP records is marked with an asterisk (*).

Data Set Usage Rights

no restrictions
Access Control:
Auth System:knb
Metadata download: Ecological Metadata Language (EML) File