Scientific Area
Abstract Detail
Nº613/847 - Precision of molecular plant identification of plants in the coastal dunes in Thy National Park, Denmark
Format: ORAL
Authors
Conny B.A. Lange1, Henrik . Pedersen3, Else . Andersen4 and Anders, S. Barfod2
Affiliations
1 Department of Plant and Environmental Sciences, University of Copenhagen, Denmark
2 Department of Bioscience, University of Aarhus, Denmark
3 Select Nature, Denmark
4 Thy National Park, Denmark
Abstract
Coastal sand dunes are highly dynamic landscapes shaped by wind, water and salt. In Europe they are mainly distributed along the Atlantic coastlines. In Denmark sand dunes are widespread and are characterized by a rapid turnover of the constituent habitat types. In this project we address the need for novel and efficient ways of monitoring dunes dynamics. Traditionally, a trained botanist identify plants either by recognizing them or using a scientific flora. However, this procedure is constricted to the flowering season and often requires complete plant specimens. Molecular plant identification is a tool that may prolong the monitoring season, can identify plants based on fragments, and can add to the quality of plant identification. To test the precision of molecular plant identification we collected four individuals from each of 100 plant species common to or characteristic of the sand dunes in Thy National Park. For each individual we sampled a herbarium voucher specimen and leaf material dried and preserved in silica gel. From the leaf material the DNA barcode genes rbcL and matK were sequenced. We used rbcL sequences to probe into the ability of the international databases, Genbank and BOLD to retrieve the correct identifications. Less than 50% of the individuals were unambiguously identified to the correct species. However, a barcode gap existed between all species, except for a few hybrids of Juncus. A Neighbour Joining analysis clustered individuals of each species and was congruent with the current APGIV classification. We conclude that caution should be exercised when basing plant identification purely on BOLD and GenBank. The most important sources of error in the online databases were mismatch between the sequence uploaded and the underlying herbarium voucher specimen, wrong identification of herbarium specimens, and mismatch between the length of the sequence in the database and the sequence submitted.