Meeting the challenges of delivering site specific weather data/drivers: The Daymet example

Background/Question/Methods

Many researchers and scientific institutions are finding it necessary to archive and make available data products that are produced by funded research activities.  Some of these data sets have fundamental and wide-ranging applications within ecological research.  Daymet is an example of such a data set that was developed to provide forcing data for terrestrial ecosystem modelling.  Recently updated and reprocessed to expand its spatial and temporal extent, Daymet is a spatially referenced, gridded, fine-scale weather data set covering a large geographic region and a significant temporal period.  Ecological research applications benefitting from these types of data include site specific characterizations needing daily weather estimates or driver data for uncoupled model scenarios when other forcings are not available.  Once a data set like Daymet is produced, the challenge then becomes providing these data in such a way that it addresses the diverse needs of the education and research community. With the use of advanced Web-based technologies, coupled with standardized data formats, delivery tools and services were developed within the NASA funded Oak Ridge National Laboratory Distributed Active Archive Center (ORNL DAAC) that allow access to the Daymet data that fulfills the varied needs of the ecological community.

Results/Conclusions

With the ORNL DAAC release of the Daymet data set, it was the first time that the scientific community had convenient access to estimates of daily 1-km gridded weather data covering the conterminous United States, Mexico, and Southern Canada from 1980 to present.  Daymet is one of the highest downloaded data sets from those distributed by the ORNL DAAC.  Daymet data downloads and use are characterized by number and types of downloads and data set citation reference information from a Digital Object Identifier (DOI).  Every month, tens of thousands of users access Daymet, with 500 GB to 1TB of data downloaded for the same reporting period.  Since 2012, 17 peer-reviewed journal articles cite the Daymet data set; 82 other publications cite the Daymet model description publications.  The popularity of the Daymet data set results from many aspects, but at the forefront is the careful consideration of community needs.  Fundamental to this is following data archive and distribution best practices by providing open, standardized, and self-describing data.  From that, specialized tools and web services have been developed including a Web service based data query and download tool, a THREDDS Data Server with multiple access options, and FTP access.

http://daymet.ornl.gov