Taxonomic Harmonization of Diatom Counts From Lake Sediment Cores of Northeastern United States
On this page:
We collated five datasets of lake sediment diatoms from the Northeastern US generated by State and Federal agencies from 1991 to 2018 using the following harmonization process. First, after updating 2172 taxa names used in 1326 lake sediment samples to currently accepted diatom names, we obtained a “nomenclaturally” harmonized dataset. Second, we revised name usage in subsets of data generated by different laboratories by identifying known taxonomic synonyms and inspecting diatom slides and images associated with these data. We also used Indicator Species Analysis to identify names that were inconsistently applied among data subsets and that caused major disagreements among these subsets. Each harmonization step led to a progressive increase of the amount of variation in species data explained by environmental variables and a parallel decrease of the variation attributable to taxonomic inconsistency as estimated by partial constrained ordination analyses. However, the harmonization was achieved at the expense of taxonomic resolution, which is important to preserve for investigations of diatom species distribution patterns and identification of assemblage attributes that can be used as environmental indicators. We, therefore, constructed a dataset of original and harmonized counts, name translation schemes and specimen museum location data to enable further studies of the data and underlying physical specimens. A Voucher Flora set of images representing diatom specimens from Northeastern USA lakes is linked to the harmonized dataset as means of promoting taxonomic consistency of diatom count data generated in future.