The dataset can be reused for contintental-wide synthesis of the cover of Australian grasses. It consists of high quality, well-described plot-based data extracted from TERN repository on 13/3/2014. The data includes vegetation records for the Poaceae family from the following dataset: ABARES Ground Cover Reference Sites Database, Biological Survey of South Australia - Vegetation Survey, Biological Database of South Australia, Corveg (Queensland), TERN AusPlots Rangelands Survey Program, Biological Survey of the Ravensthorpe Range (Western Australia).The entire content of the portal was initially extracted using the portal's download feature to obtain the full extent of available data for the following all datasets. These data were loaded into a PostgreSQL database. Subsequently, a SQL query was built for each of the cited datasets which produced a flat table containing information about the survey name, site identifier, visit date, coordinates, species, abundance, biomass and/or cover class, filtering on species of the Poaceae family using a genus list obtained from the website of the Atlas of Living Australia (http://www.ala.org.au/).

The AEKOS Australian Vegetation sPlot dataset consists of high quality, well-described plot-based data extracted from the AEKOS (portal.aekos.org.au) on 11/11/2014. The data includes vegetation records for the following datasets: Australian Ground Cover Reference Sites Database, Biological Survey of South Australia - Vegetation Survey - Biological Database of South Australia, Atlas of NSW database: VIS flora survey module, Queensland CORVEG Database, TERN AusPlots Rangelands, Transect for Environmental Monitoring and Decision Making (TREND), AusCover Supersites SLATS Star Transects, Biological Survey of the Ravensthorpe Range (Western Australia).The portal's vegetation plot data was extracted using the portal's download feature to obtain the full extent of available data for the all datasets. In addition, an average cover value was calculated for each site using a slight modification of the ingestion scripts normally used to ingest the source data into AEKOS. The altitude values derived from a map layer using the site coordinates were obtained from the AEKOS index. Finally, land use and vegetation type were derived from map layers using the site coordinates. These data were loaded in different tables of a PostgreSQL database. Subsequently, two SQL queries were built to centralise the available data in two tables: table r_site containing the site specific data and table r_speciesobservations containing the individual data on observed specimen. A PostgreSQL backup file containing these two table was then built using the pg_dump tool. The dataset can be reused for contintental-wide or global synthesis of the cover of Australian vegetation.