The Water Resources NSDI Node is an online centralized repository of digital metadata that conforms to the Federal Geographic Data Committee (FGDC) standards.

Questions and Answers about Guidelines for the Documentation, Review, and Release of Digital Spatial Data

Contents

1. What is a spatial data set?
2. What spatial data sets are required to be released to the public?
3. Can digital spatial data sets contain interpretive information?
4. How do I produce "FGDC-Compliant" Metadata?
5. If I have many data sets for my project, must I release each one of these separately or can they be grouped together?
6. What colleague reviews are required?
7. After colleague review, how is the data set approved?
8. Once my data set is approved, how do I release it?
9. Can I make the data available from my site's World-Wide Web pages rather than placing it in a node of Clearinghouse? How about a cooperating agency's pages?
10. Can the metadata serve as a report?
11. Should prior released versions of spatial data sets be saved?
12. In what format must I distribute my Arc/INFO (or any GIS) data set?
13. Who is the "author" of a spatial data set?
14. Can I release a data set in parts?
15. How should spatial data sets be referenced in publications?

1. What is a spatial data set?

Any data set that references it's data to a geographic location is a spatial data set. A common example of a spatial data set is an Arc/INFO coverage, but Triangular Irregular Networks (TIN's), GRIDS, ArcView shape files, images, satellite or aerial images, or similar data sets from other GIS's also are spatial data sets.

2. What spatial data sets are required to be released to the public?

The USGS interpretation of Executive Order (EO) 12906 requires that all spatial data sets that are finished products be released to the public. In addition USGS Fundamental Science Practices Instructional Memo IM OSQI 2015-03 Review and Approval of Scientific Data for Release imposes additional requirements such as the necessity for a Digital Object Identifier for data sets. The February 22, 2013, Office of Science and Technology Policy (OSTP) memorandum on “Increasing Access to the Results of Federally Funded Scientific Research,” requires public access to digital datasets resulting from federally funded research, including datasets used to support scholarly publications. This includes spatial data sets that are "deliverables" for a project. It also includes data sets used to make calculations for a report, such as GIS input spatial data sets for a ground-water model: if the data sets are necessary to reproduce the calculations, they should be released. Spatial data sets that result from digitizing interpretive analog (printed) maps and are available for the first time in digital form should be released.

Temporary or working data sets - actually, the vast majority of data sets on our computers -- are specifically excluded from the mandatory release requirement; however, based on circumstance these may be released, refer to: Guidance for Public Release of Preliminary or Provisional Data and Interpretive Information. Similarly, data sets that are copies or minor modifications (e.g., reprojection, map joining, or reselection) of data sets published elsewhere -- for example, 1:100,000-scale Digital Line Graphs - need not be released again. Even though not released changes in datasets must be documented.

According to EO12906, spatial data sets produced prior to January 11, 1995 are exempt from this requirement, but should be released if resources permit. Data sets associated with scholarly publications a.k.a. interpretive information products (USGS Series Reports, journal articles, etc.) published prior to February, 15, 2015 are exempt from requirements of the USGS FSP Instructional Memos on Data Release. Data sets of a proprietary nature or classified for national security reasons also are exempt.

3. Can digital spatial data sets contain interpretive information?

Yes. Scientific interpretation is the process of applying knowledge of scientific principles, judgment, and skill to reach a reasoned conclusion based upon data. Digitized data of scientific interpretation of mappable properties are considered interpretive.

Spatial data sets that result from routine cartographic operations, such as digitizing, clipping, etc., are not interpretive.

4. How do I produce "FGDC-Compliant" Metadata?

The FGDC content standards appear formidable, but automated processes can expedite your work considerably. The recommended practice is to produce a metadata file in an indented text (.xml) format that can be parsed by the "Geospatial Metadata Validation Service" program. This service checks your FGDC-compatible geospatial metadata record using mp (metadata parser). Provide a metadata record that is formatted as indented text, XML, or SGML. Results will show discrepancies between the structure of your metadata record and the expectation of the FGDC metadata standard (FGDC-STD-001-1998) and will show several potentially useful re-expressions of the metadata record itself. More information on the latest techniques can be found on the USGS WMA NSDI Tools and Guidance page.

5. If I have many data sets for my project, must I release each one of these separately or can they be grouped together?

Most of the time, each data set must stand on its own. That is, the data set and its metadata must be treated as a single logical entity; one data set corresponds to one metadata file.

You may write a separate report (DS, OFR, PP, SIR, SIM, etc., ) that treats the data set as part of a collection. You may even want to have the report and the data sets reviewed together. Each data set must have a unique title (for example, "ok_aqbound" or "ok_recharge"), so that referrers may readily distinguish among the individual data sets. A metadata file may indicate that a data set is part of a collection and refer back to the report. An example of such a collection is Open-File Report 97-574. You may release groups of data sets (collections) through the mechanism of a USGS Data Release and locate the description of these data on USGS Science Base. Refer to https://www.sciencebase.gov/catalog/ .You may write a separate report that describes the collection but the report itself might not contain the data. If you intend to release the data sets and report describing these data at the same time then everything should be reviewed together. Guidance is available here: https://internal.usgs.gov/publishing/authorsguidetodatareleases.html . Each data set must have a unique title (for example, "Oklahoma_Aquifer_Boundary or "Oklahoma_Recharge_Boundary "), so that referrers may understand and readily distinguish among the individual data sets. Each dataset's metadata file must use the larger work citation to indicate that the data set is part of a larger collection. The collection of data must have a Digital Object Identifier (DOI)  from the USGS DOI agent. Refer to https://www.usgs.gov/datamanagement/preserve/persistentIDs.php for information on USG DOIs and here for the USGS data DOI creation tool https://www1.usgs.gov/csas/doi/ .

6. What data and metadata reviews are required?

Data and metadata review requirements are identified in USGS Fundamental Science Practices Instructional Memo IM OSQI 2015-03 Review and Approval of Scientific Data for Release and IM OSQI 2015-02 Metadata for Scientific Data, Software, and Other Information Products. Additional information is available from FAQs on the FSP public website. For data release FAQs refer to https://www.usgs.gov/fsp/faqs_datarelease.asp and for meatadata FAQs refer to https://www.usgs.gov/fsp/faqs_metadata_for_scientific_data.asp.

7. After colleague review, how is the data set approved?

Refer to Data Release FAQ https://www.usgs.gov/fsp/faqs_datarelease.asp#faq1 .

Once the author has responded to the reviewers' comments, release may be approved by the District or equivalent Office Chief. Criteria for approval are:

• Has the report been adequately reviewed and has the author addressed the reviewers' comments?
• Is the data set (including transfer files, metadata, etc.) ready for delivery?
• Is the data set in conformance with all USGS policies?

Approval procedures are determined by the approving official. The name of the approving official must be given when submitting materials to the WRD Node for distribution.

8. Once my data set is approved, how do I release it?

The easiest way to release a spatial data set is to submit both the data set and its metadata to the WRD National Spatial Data Infrastructure (NSDI) Node. Instructions for submitting a spatial data set are available on the WWW. Releasing a data set on the WRD NSDI Node does not preclude distribution by other means.

The Water Resources Division Node operates as part of the FGDC National Spatial Data Infrastructure. The WRD Node stores metadata in a format suitable for searching under a standard specified by the Federal Geographic Data Committee (FGDC) and is registered with FGDC as a source for searchable data. In simple terms, this means that, if your metadata is stored on the WRD Node, anyone using the FGDC's search engine (data.gov, geoplatform, sciencebase, geospatial one stop, etc.,) will finding it.

In addition to serving metadata, the WRD Node also is designed to serve spatial data sets. The WRD Node is located on one of USGS's Internet gateways so transmitting data to outside users does not tie up internal USGS network resources. When a user asks to retrieve a data set, the WRD Node uses a script called "getspatial" to present a page that briefly describes the data set, highlights a link to the metadata, lists all the available files, and explains how to retrieve each format. For example, https://water.usgs.gov/lookup/getspatial?huc2m will use getspatial to retrieve the "huc2m" spatial data set.

Finally, the WRD Node provides archiving capability for data sets, even when they are revised. All data sets on the WRD Node remain a permanent part of the scientific record.

You can also release your data within the USGS through the Science Base site. Regardless of where your data and metadata reside, OMB requires the data be cataloged in the USGS Science Data Catalog (SDC). Refer to https://data.usgs.gov . If the data are housed in USGS Science Base or the WRD National Spatial Data Infrastructure (NSDI) Node the data is automated registered and harvested by the FGDC search clent.

9. Can I make the data set available from my district's World-Wide Web site rather than placing it in a node of the Clearinghouse?

Yes, but you must be sure to meet the following conditions:

1. The data must meet all FSP related data release policies.
2. The server housing the data must meet data preservation requirements pursuant to IM OSQI 2015-04 Preservation Requirements for Digital Scientific Data , must be curated and meet trusted digital repository standards.
3. The data set and all its released revisions must be available to the public. (See question 11.)
4. The URL of the metadata must be persistent so that it can be referenced in printed reports. See question 15 for an example.
5. The persistent URL must also have a digital object identifier (DOI) associated with it. Refer to https://www1.usgs.gov/csas/doi/ to obtain DOI.
6. The metadata and data must both reside together on the Web site.
7. The data must be registered with the USGS Science Data Catalog

10. Can the metadata serve as the basis of a data release instead of a USGS series reporta report?

No. However, data and associated metadata can be released as described here: https://www.usgs.gov/usgs-manual/im/IM-OSQI-2015-03.html

11. Should prior released versions of spatial data sets be saved?

Yes, but they do not have to be online. Like any important files, prior versions should be saved on secure media or remain on the node. A contact name (a position, not a person) for an office should be given in the metadata for users who need to obtain these earlier versions. The WRD NSDI Node will automatically take care of this when a data set on the node is updated. Ideally updating the metadata with a <edition> tag describing the explanation of the update, the date of the update and a numeric value (Version 2) should be used. Each version of a spatial data set should be assigned an edition number under the Identification_Information:Citation:Edition tag of the metadata. A three-part system such as "1.2.3" is recommended. The version number must always be mentioned in citations and download areas (See sir2016_5022_usgsdatarelease.xml.)

12. In what format must I distribute my GIS (or any data) data set?

Spatial data can be distributed in the Spatial Data Transfer Standard (SDTS) format. You may, in addition, distribute them in "native application mode," Arc/INFO export format, shape, geodatabase, etc., USGS Policy states that data should be made available to the public in a non-proprietary format, if possible. Data sets in CAD or raster formats, can be made available in non-proprietary DXF or image format (for example, TIFF, BIL), respectively. TIN data can be exported as a point data set through SDTS. Grids can be made into a nonproprietary format with the Arc/INFO GRIDASCII command, and compressed.

13. Who is the "author" of a spatial data set?

At least one author must be named in the "Originator" field of the metadata. Authorship normally is restricted to participants who contributed substantially to creation of an original data set. For participants responsible for assembling a nonoriginal data set, they should be credited as compilers or editors.

It is not uncommon for spatial data sets to be revised and reissued, sometimes by persons other than the original author/compiler. The supervisor of the unit producing the revised data set shall decide authorship in such cases. In all of these situations, however, the original source or author(s) of the information must be cited.

Acknowledgment of other assistance may be included in the "Data_Set_Credit" field of the metadata.

14. Can I release a data set inparts?

Some users need only a part of the data set. Where it is logical to do so, dividing (reselecting) a spatial data set into parts can be helpful in distributing it because users can retrieve smaller files. There is one metadata file, and the parts are considered just an alternative distribution option.
The WRD Node is designed to offer users the option of retrieving data sets in parts, or of retrieving the entire data set.

15. How should spatial data sets be referenced in publications?

The reference to a spatial data set should include the author, title, and the URL as follows:

Powell, J.W. (1869), Digital Map of Streams in the Grand Canyon, Version 1.2.1, USGS data available on the World Wide Web at https://water.usgs.gov/lookup/getspatial?gc_stream.

or the newest refered method as the accepted reference method is available at https://www.usgs.gov/datamanagement/preserve/persistentIDs.php . Specifically the use of a Digital Object Identifier (DOI) is required to cite your data.

Note that the citations above give a persistent URL based on the assigned coverage name. The WRD NSDI Node will support this format for the forseeable future. The getspatial program, which operates on the WRD NSDI Node, does a lookup to find the current physical location of the gc_stream spatial data set files, and presents them to the user. A real example of getspatial for the 1:2,000,000 hydrologic units is https://water.usgs.gov/lookup/getspatial?huc2m.

    The Water Resources NSDI Node is an online centralized repository of digital metadata that conforms to the Federal Geographic Data Committee (FGDC) standards.