A dataset containing 5947873 records in GBIF matching the query: SELECT kingdom, kingdomkey, phylum, phylumkey, class, classkey, "order", orderkey, family, familykey, genus, genuskey, species, specieskey, "year", IF(ISNULL(kingdomkey), NULL, SUM(COUNT(*)) OVER ( PARTITION BY kingdomkey, "year")) kingdomcount, IF(ISNULL(phylumkey), NULL, SUM(COUNT(*)) OVER ( PARTITION BY phylumkey, "year")) phylumcount, IF(ISNULL(classkey), NULL, SUM(COUNT(*)) OVER ( PARTITION BY classkey, "year")) classcount, IF(ISNULL(orderkey), NULL, SUM(COUNT(*)) OVER ( PARTITION BY orderkey, "year")) ordercount, IF(ISNULL(familykey), NULL, SUM(COUNT(*)) OVER ( PARTITION BY familykey, "year")) familycount, IF(ISNULL(genuskey), NULL, SUM(COUNT(*)) OVER ( PARTITION BY genuskey, "year")) genuscount, COUNT(*) occurrences, MIN(GBIF_TEMPORALUNCERTAINTY(eventdate, eventtime)) mintemporaluncertainty, MIN(COALESCE(coordinateuncertaintyinmeters, 1000)) mincoordinateuncertaintyinmeters FROM occurrence WHERE occurrence.basisofrecord IN ('MACHINE_OBSERVATION', 'OBSERVATION', 'HUMAN_OBSERVATION', 'MATERIAL_SAMPLE', 'MATERIAL_CITATION', 'PRESERVED_SPECIMEN', 'FOSSIL_SPECIMEN', 'LIVING_SPECIMEN', 'OCCURRENCE') AND occurrence.countrycode IN ('US', 'CA', 'FR', 'GB', 'AU', 'SE', 'NL', 'ES', 'DE', 'DK', 'IN') AND occurrence.hascoordinate = TRUE AND occurrence.hasgeospatialissues = FALSE AND occurrence.occurrencestatus = 'PRESENT' AND (occurrence."year" >= 1970 AND occurrence."year" <= 2025) AND NOT GBIF_STRINGARRAYCONTAINS(occurrence.issue, 'TAXON_MATCH_FUZZY', TRUE) AND (occurrence.distancefromcentroidinmeters >= 2000.0 OR occurrence.distancefromcentroidinmeters IS NULL) AND (occurrence.specieskey IS NOT NULL AND occurrence."year" IS NOT NULL) GROUP BY occurrence.kingdom, occurrence.kingdomkey, occurrence.phylum, occurrence.phylumkey, occurrence.class, occurrence.classkey, occurrence."order", occurrence.orderkey, occurrence.family, occurrence.familykey, occurrence.genus, occurrence.genuskey, occurrence.species, occurrence.specieskey, occurrence."year" The dataset's 5947873 records were derived from 34514 constituent datasets; see https://api.gbif.org/v1/occurrence/download/0065953-251120083545085/datasets/export for details. Data from some individual datasets included in this download may be licensed under less restrictive terms.