The data quality topic is getting some traction within the biodiversity community. As a data aggregator and one of the primary source for biodiversity data, GBIF needs to adapt, enhance and expand its current data quality activities. For users unfamiliar with nature and characteristics of the globally aggregated data, data quality remains one of the major concerns and the barriers for use. For expert users, data modifications to ensure fitness for use remain a time and effort consuming activity. Data quality and credit should be the primary concerns of data publishers worldwide, but the practices vary.

The presentation will cover the current state of the GBIF parsing and interpretation with a focus on the current data quality flags that are applied, and on how to make use of data quality flags. In addition, the recent development on the GBIF Data Validator enables a dataset to be parsed and interpreted before its publication online. By using Data Validator, some errors and enhancement possibilities can be detected and, possibly, fixed before the publication of the data through GBIF.org.

Major effort has been put into data quality documentation and solutions over the last years among different aggregators, institutions and contributors within the community. Based on the contributions from TDWG Biodiversity Data Quality (BDQ) Interest Group, especially from Task Group 2: Data Quality Tests and Assertions as well as the implementations available in the biodiversity informatics community, we will present an initial plan to merge different informatic code bases together in order to make it available to the community into a more lightweight form.

GBIF

Christian Gendreau