Resources

Portal Spatial Quality Tab

The spatial quality tab provides an automated assessment of the quality of a data record. It does so by passing spatial and taxonomic information present in a record through a web service that examines this information for completeness, obvious mistakes, and inconsistencies. To find the Spatial Quality Tab, simply complete a search in the VertNet data portal and then click on any of the individual records returned (see screen capture below).

Overview

The goal of the tool is to help you identify if there any issues or errors in regard to the fitness for use of the record, not to provide an authoritative answer about overall data quality. The report provided in the Spatial Quality tab within the VertNet data portal in particular, rely on bringing multiple types of data to bear, including the individual record and country and species range map boundaries.

To help you understand how the results of the data quality assessment are produced, we’ve described each test below.

NOTE: The spatial quality tab performs the data quality check in real-time. It may take a moment to populate on your screen. VN Portal Spatial Quality Tab image

Spatial Quality Tab

Data Completeness

Data completeness assesses the presence and absence of certain fields within a record. The assessments simply reveal if a given field is present or not, or, in some cases, if the field contains a potentially correct value or not.

Are coordinates present?

This test checks whether or not the fields “decimalLatitude” and “decimalLongitude” contain content. It does not test for values in “verbatimLatitude” or “verbatimLongitude” since these are not interpreted fields. If the test determines that one or both fields are empty the record is flagged as having a problem. It is important to note that if coordinates are not present, many of the following tests cannot be run.

Is the country value present?

This tests checks whether or not the field “country” contains content. Certain values, such as “n/a” or “Not specified”, are reviewed as empty fields so the result of the assessment will return a negative result. A record with an empty field, or a field interpreted as being empty, will be flagged as having a problem.

Are both coordinates 0 (zero)?

This test assesses whether or not both “decimalLatitude” and “decimalLongitude” contain the value zero (0). In some cases, the correct coordinates for the record may be 0,0, but in the majority of cases, this value is a placeholder for unknown coordinates. Therefore, if this test determines both coordinates to be zero (0), the result is a warning flag.

Do coordinates have three or more decimal figures?

This checks whether or not both “decimalLatitude” and “decimalLongitude” fields are filled with a number with three or more figures after the decimal separator. Results that demonstrate that one or both fields contain values with fewer than three decimal places will flag the record with a warning. For more information on why this is important see Page 27, Table 4, of the BioGeomancer Guide to Best Practices for Georeferencing.

Do coordinates have datum?

This checks whether or not the field “geodeticDatum” contains content. The datum is an important piece of information for a correct understanding of the actual location of an organism. If a datum is not present the record is flagged as having a problem. It is wise to verify that the datum in which each datum has been recorded is valid. You can verify most datums and ellipsoids in Appendix B of NIMA’s Department of Defence WorldGeodetic System 1984: Its Definition and Relationships with Local Geodetic Systems (Updated, 2004).

Data Inconsistencies

The tests in this section seek to identify mismatches between different sources of data. All of the assessments are performed with the Map Of Life quality validation tool. It is important to note that an inconsistency does not necessarily equate with a mistake. The tool is only designed to flag potential issues for the user of the data to review.

Are coordinates within specified country?

Using the GADM Database of Global Administrative Areas, this test checks whether or not the coordinates in “decimalLatitude” and “decimalLongitude” fall inside the boundaries of the country specified in “country” field of the record. It is not possible currently to determine if a pair of coordinates that fall in the ocean are found within the territorial waters of the specified country, since the GADM database can be used to verify land cover. In cases such as marine fish, this assessment will therefore most likely provide misleading results.

Distance outside of specified country (in degrees)

If the coordinates from a record are found to be outside the boundaries of the specified country, this test will use the GADM Database of Global Administrative Areas to determine the distance between the coordinates in “decimalLatitude” and “decimalLongitude” and the closest point of the country specified in “country”. The measured distance is given in degrees. Like the previous test, it is not possible to determine if a pair of coordinates that fall in the ocean are found within the territorial waters of the specified country, thus the value of this assessment will be greater than zero (0) for points that fall in the ocean.

Distance outside of species range map (in degrees)

This test checks the distance between the coordinates in “decimalLatitude” and “decimalLongitude” and the closest point of the International Union for Conservation of Nature (IUCN) range map for the species. The measured distance is given in degrees. It might be possible that there is no available range map for a given species from IUCN. In these instances, the test will return as “Could not be assessed”. In other cases, if the “Are coordinates within specified country” test fails or returns a negative value, this test may also fail.

Data Errors

The tests in this section seek to identify data errors within specific fields of a record. Three of these assessments are performed with the Map Of Life quality validation tool.

Is latitude between 90 and -90?

If latitude values are present in the “decimalLatitude” field, this simply determines if the value is between 90 and -90. Instances in which the value is not between 90 and -90 will flag the record as containing an error.

Is longitude between 180 and -180?

If longitude values are present in the “decimalLongitude” field, this simply determines if the value is between 180 and -180. Instances in which the value is not between 180 and -180 will flag the record as containing an error.

Are coordinates transposed?

If errors or inconsistencies are found in previous tests, it may be possible that the values for latitude and longitude have been inverted. This test assesses the coordinates to see if an inversion has occurred by transposing the values to determine if the coordinates fall into the specified country.

Is latitude hemisphere correct?

If errors or inconsistencies are found in previous tests, it may be possible that the hemisphere for latitude has been recorded incorrectly (e.g., 40.0176 vs -40.0176). This test determines if this error has occurred by multiplying the “decimalLatitude” value by -1, and then checking if the negated coordinates fall into the specified country.

Is longitude hemisphere correct?

If errors or inconsistencies are found in previous tests, it may be possible that the hemisphere for longitude has been recorded incorrectly (e.g., -105.2797 vs 105.2797). This test determines if this error has occurred by multiplying the “decimalLongitude” value by -1, and then checking if the negated coordinates fall into the specified country.

If you have any questions about this document, please contact VertNet's support team.

Visit our Help page for more resources created for the VertNet project.


Orig Release, 05Sept2014 (Javier Otegui and Rob Guralnick)