Database > Hunt duplicates

Menu: Database | Hunt duplicates

Duplicates may occur when relevés from various sources (field books, literature, and computerised databases) are imported into Turboveg databases. Duplicates can be detected by calculating the floristic similarity between two relevés using the formula: i = c/(A + B - c) (= Jacquard-index), where c is the number of species shared, A the number of species unique to one relevé and B the total number of species of the other relevé. If i = 0 there is no floristic similarity between the relevés, if i = 1 the relevés are identical.

This formula can be extended in such a way that the cover abundance of the species is also taken into account: Srij = ∑kykiykj/(∑kyki2+ ∑kykj2 - ∑kykiykj) (Similarity ratio of Ball).

Compare with…: Select the database to compare with the already opened database. You can also hunt duplicates in the same database.

Restrict comparison: You can speed up the comparison by specifying Year and/or Locality. Then, only relevés with similar dates and/or locality will be compared.

Calculation: Calculation of the similarity-index can be based on the presence or absence of species (Qualitative) or on the cover abundance of the relevés (Quantitative).

A matrix is presented where for each relevé compared the four most alike relevés are shown. Double-click on one of the grid cells to see a comparison between two relevés.

Save: The matrix can be written to a file.