France
June 14, 2017
Numerous experiments have been performed throughout the world to compare the yields of grain legumes. INRA scientists have now developed an open-access database which can be used to compare the performance of different species.
Because of their notably high protein content, grain legumes are an important component in the diets of both humans and livestock. They also offer significant ecosystem services, particularly because of their ability to fix atmospheric nitrogen. Although there are very numerous legume species, nearly three-quarters of the land sown to legumes is occupied by two or three species, mainly pea in Europe and soybean in America and Asia.
Recent decades have seen a large number of experiments designed to study and compare different legume species in various parts of the world. The results of these experiments have produced valuable information and enabled the ranking of legume species as a function of their yields and nitrogen balances, but these data are disparate and have never been the subject of a global synthesis.
8581 combinations in 41 countries
Based on data extracted from 173 scientific articles, INRA scientists in Versailles-Grignon have created a global database that covers 360 experimental sites spread between 18 climatic zones (according to the Köppen-Geiger classification) and 41 countries. This database includes 8581 combinations of crops, sites, years and treatments. The information recorded describes climatic conditions, soil types, yields, plant nitrogen contents, soil N reserves and cultivation practices. When they were available, data collected on subsequent crops were also collated.
An open-access database
These data are now archived in a relational, open-access database that is accessible via the Dryad Digital Repository, which is a resource that enables the discovery, reuse and citation of data underpinning scientific publications. Dryad supplies a polyvalent framework for a wide variety of data.
These data can be analysed using statistical methods to compare different legume species according to several criteria (e.g. grain yield, protein yield, energy yield, nitrogen balance). They can be used to estimate how the cultivation of legumes affected the productivity of subsequent crops. The data can also be analysed by means of standard meta-analysis methods in order to rank the performance of a large number of grain legume species, and to identify those of the most interest according to several criteria.
More generally, this database offers a unique opportunity to scientists as it enables the analysis and comparison of legume performance under diverse environmental conditions.
Reference
Cernay C., Pelzer E., Makowski D. 2016. A global experimental dataset for assessing grain legume production. Scientific data 3:160084. doi:10.1038/sdata.2016.84
Species
