Following Metadata Best Practices

HydroShare provides a standard set of metadata elements for describing the resources that you create. These elements follow the Dublin Core Metadata Initiative’s standard metadata terms, including title, abstract, subject keywords, spatial and temporal coverage, etc. HydroShare users should treat datasets and models as first class research products - along with the papers that are written about them. There is opportunity to maximize the value of these research products by practicing good data management and creating descriptive metadata. The information below provides some best practices to keep in mind while describing your resource with metadata. For more detailed best practices, refer to the HydroShare Guide for Data Authors and Publishers.


The Abstract should put the content of the resource in context, including the rationale for the data collection or modeling and should be a specific description of the content files included in the resource. Ensure that the results described in the Abstract correspond to the files contained in the resource. The Abstract should include a description of the resource content and organization to help a potential user navigate the various files included. Where the content and organization of a resource is complicated, the Abstract may reference a README file included as a content file in the resource that provides a more detailed description of the resource’s content files. Additional details may be added to the Abstract to describe the methods of data creation - i.e., what you did, why you did it, how you did it, when and where you did it, etc. If text is used from the Abstract of a journal article related to the resource, add additional details focused on describing the actual content of the resource.


Keywords facilitate data discovery and should be selected to be descriptive and thorough for the associated data. Consider what the data represent, the variables included in the dataset, how the data were generated, and the geographic area that they represent. Unique acronym-like keywords may also be used similar to the way hashtags are used in some social media. For example, the term “iUTAH” is unique to a specific Utah research project named “iUTAH.” Use of this as a keyword facilitates discovery of resources associated with the iUTAH project. 


File names should be descriptive and meaningful. Where many files are included in a resource, a README file that describes the content and/or organization of each file or the resource as a whole can be very useful in helping others understand the content of the resource.

Spatial and Temporal Coverage

These are optional metadata and should be used if the data or model in the resource has a geospatial “footprint” or location or a temporal component/time window. Including appropriate spatial and temporal coverage metadata can help others discover your data if they are searching within a specific geographic location or over a specific time period.

 Other Details

All abbreviations need to be defined somewhere within the resource. Even seemingly obvious cases should be spelled out (e.g., C=Carbon, N=Nitrogen). Unit symbols and abbreviations should also be defined (e.g., mg/L=milligrams per liter). Definitions should be handled wherever sub-file metadata is contained (e.g, in the resource Abstract, in a README file, or in file headers).