Agronomic data: advances in documentation and protocols for exchange and use

Data from agronomy experiments are typically collected and stored in a numb er of minimally documented computer files, with additional information bein g entered and archived in field books or diaries. Data manipulation is gene rally cumbersome and error-prone, and data loss is frequent. Modern databas e technology has the potential to resolve these issues. However, experience gained by an international network of experimenters and crop modellers (th e International Benchmark Sites Network for Agrotechnology Transfer; IBSNAT ) in using a database for agronomic experiments conducted by many workers a t different sites highlighted problems of data entry, quality control, and changing requirements for storage and output variables. In an attempt to mi nimize these problems, IBSNAT reduced its focus on a central database, but considerably enhanced its effort on the design and use of a set of simple, standard experiment documentation and results files that could be establish ed and edited easily, transferred directly among workers, used as inputs to analytical software and crop models, and read by database and spreadsheet software. The standard files which were developed, and which were used in a software package termed DSSAT V3, have recently been upgraded by a consort ium of experimenters and modellers (the International Consortium for Agricu ltural Systems Applications ICASA), These new files are described briefly h ere. The ICASA files constitute an advance in the potential for good docume ntation and storage of agronomic data, but only partly solve the problem of overall data management and use. There is still need for central and local databases that facilitate both the searching of information from different experiments, and the examination of relationships that may be apparent in a large array of data. A number of such databases have been developed for s pecific applications, and a few of these are briefly touched upon. In parti cular, recent work with one large database currently being developed by a n umber of international Agricultural Research Centers, National Research Org anizations, and Universities, (the International Crop Information System, [ CIS), is briefly described. (C) 2001 Elsevier Science Ltd. All rights reser ved.