Multidimensional

HydroShare uses the Network Common Data Form (NetCDF) specification for multidimensional data. Multidimensional space-time data is commonly used for hydrological modeling (e.g., precipitation or surface air temperature that vary in space and time). 

NetCDF files contain self-describing metadata and support the creation of an array-oriented dataset, which may include three spatial dimensions (e.g., x, y, z), a time dimension, and other user defined dimensions. A HydroShare multidimensional (NetCDF) content aggregation is created from a single NetCDF file with a “.nc” file extension. HydroShare uses ncdump -h (a NetCDF utility function) to extract information from the file header, which is saved as a text file alongside the data file in the content aggregation folder.  

While any NetCDF file can be used in HydroShare, we encourage use of Climate and Forecast (CF) and ESIP Attribute Convention for Data Discovery 1-3 (ACDD) conventions. HydroShare matches metadata from the header file with CF and ACDD convention tags to identify and populate content level and resource level metadata (see the table below). When a NetCDF file with author, keyword, spatial and temporal reference, and other matched metadata is imported to HydroShare, there is no need for the user to enter this metadata as it will be extracted automatically.

Example

A NetCDF file is added to a HydroShare resource (e.g., by dragging it onto the content area of the file browser).

HydroShare recognizes this as a NetCDF file and automatically reads its metadata (if the file is defined by the CF or ACDD conventions - see here for the mapping of terms to HydroShare). As a result:

  • The header information is extracted into a text file (<filename>_header_info.txt) stored in the aggregation folder (along with the data file) depicted by the multidimensional symbol, where <filename> is the name of the NetCDF file. These can be viewed in the Content Metadata tab.
  • At the resource level, Title, Abstract, Authors, Keywords, and Coverage are set based on information extracted from the file unless they have already been set.
  • The content level metadata attributes Title, Keywords, and Coverage are also automatically populated.

A multidimensional file in the contents section called "SWE_time".The contents section, file tree at the top says "contents/SWE_time". The files in this folder are "SWE_time_header_info.txt" and "SWE_time.nc".

The automatically extracted metadata are displayed in the right panel of the file browser.

The same SWE_time file on the content section on the left, on the right the metadata entry field is open. In the Title entry field at the top is "EUB model simulation of snow water equivalent in Logan River watershed from 2008 to 2009". Below this, the keywords entry field has 3 keywords "Snow water equivalent", "EUB", and "Logan River".

The metadata panel also shows the content of the full NetCDF header file. 

SWE_time file on the left. On the left in the metadata entry field is the header "NetCDF Header Information". Below the header, above the text box is the file tree "data/contents/SWE_time_header_info.txt" In the text field is the extracted file data defining time, dimensions, and variables.

This resource holds the file used in this example.

Update Metadata in the NetCDF File

HydroShare enables interactive editing for the NetCDF file metadata and the resource specific metadata on the resource landing page within the metadata panel. If a user makes changes to the resource specific metadata, a warning will appear at the top of the metadata panel that says the “NetCDF file needs to be synced with metadata changes” in which case the user can click “Update NetCDF file” (shown below) to update the NetCDF file metadata.

SWE_time multidimensional file, same image as above but with a dialogue on the right that says "NetCDF file needs to be synced with Metadata changes" with a button underneath that says "Update NetCDF File".

OPeNDAP Web Services

An OPeNDAP web service enables users to access and subset NetCDF files stored in HydroShare through an OPeNDAP data access form in a web browser or using OPeNDAP client software for data visualization or analysis. You can create or visualize a subset of dataset from the original NetCDF file without downloading the file from HydroShare to local computers.

OPeNDAP services are automatically enabled for public resources that contain multidimensional NetCDF content. A resource that is private or discoverable may not be accessed by OPeNDAP.

A user can use OPeNDAP to access the multidimensional content by clicking on the “Open with” button at the top of the Resource Landing Page and selecting the OPeNDAP data service.

The open with button with the dropdown open. Options are: Hydroshare GIS, CUAHSI JupyterHub, and OpeNDAP. OpeNDAP is circled.

THREDDS Data Server

THREDDS data server allows users to access multidimensional content in OpenDap, through HTTPserver download, ISO, NCML, and UDDC. This allows users to view datasets from NetCDF files without downloading the file from Hydroshare to a local computer. To access the THREDDS data server, right click the file and select THREDDS Data Server. Alternatively, select the Open With... button at the top of the page to select a method of opening the file. 

 Multidimensional file right click drop-down. It includes Open, Download, Download Zipped, Get file URL, and THREDDS data server.