Data Distribution and Documenting Uncertainty Information for the Earth System Data Record: The Global Terrestrial Water Cycle

A new balanced global terrestrial water cycle dataset is being created for the NASAs' Making Earth Science Data Records for use in Research Environments (MEaSURE) project. This dataset will be comprised of multiple remotely-sensed datasets and model generated data, and will be merged into a single unified multi-decade, high spatial resolution, climate consistent Earth Science Data Record (ESDR). In addition to the final unified ESDR, the seven remotely-sensed and model generated input datasets will also be available as ESDRs. These input datasets are: (1) VIC model derived water cycle variables, (3 hourly, from 1948 - 2010, on a 0.25 degree grid); (2) Satellite derived precipitation data from two separate sources: (a) GPCC data, (monthly, from 1983 - 1998, on a 0.5 degree grid), and (b) TRMM TMPA data, (3 hourly, from 1998 - 2010, on a 0.25 degree grid); (3) Satellite derived evapotranspiration data, based on the SRB/ISCCP radiation forcings, ( 3 hourly, from 1983 - 2007, on a 0.5 degree spatial resolution; (4) Satellite based soil moisture is derived from multiple satellite sensors, but, primarily the TRMM TMI and AMSR-E, (daily, from 1998 - 2011, on a 0.25 degree grid); (5) Satellite derived water management variables (monthly, from 1992 - 2010, on a variable degree grid); (6) Satellite derived surface radiations, based on ISCCP-DX data, (3 hourly, from 1983 - 2007, on a 0.5 degree grid) (7) Model derived Surface Meteorological Forcing Fields (3 hourly, from 1948 - 2010, on a 0.25 degree grid). Along with these ESDRs, it is also necessary to distribute uncertainty information about the data. Including this information will assist the data users with how much error there is in the estimations in the data, the limitations in the data, as well as knowing how the final merged components of the water balance equation was weighted in order to obtain a balanced equation. This uncertainty information was generated from either the producers of the individual datatsets and/or the water imbalance corrections required to to produce the final unified dataset. Because of the multiple producers of this information, it is necessary to document these sources and methodologies. At the end of the project, these datasets will be available through the NASA Goddard Earth Sciences (GES) Data and Information Services Center (DISC), however, as input datasets become available for input processing, they are placed onto a dedicated project server, at George Mason University, and made available to the public, using such tools as FTP and the GrADS-DODS data server. This server provides the web infrastructure to communicate and distribute the large amount of data and documentations/metadata to the community. Along with these distribution tools, metrics were put in place to track user information, such as origins and number of visits and datasets downloaded. This poster illustrates the methodology used to generate the uncertainty information for a few of the input ESDRs. Because there are multiple methods for calculating this uncertainty, it was necessary to include as much information about how these estimates were obtained, as well as other relevant information within the metadata. To address this issue, standard metadata constructs were used to describe the data, and pass information along to the user.