Data Quality, Abstraction and Aggregation in the LEDA Traitbase

Abstract

In providing an open European-wide web-based database of plant traits relevant for the conservation and sustainable use of biodiversity in changing European landscapes, the LEDA project aims to support different user groups by providing information about plant traits related to the key features of plant dynamics. The trait database (Traitbase) is currently built from scattered national database initiatives, literature sources and new measurements contributed by persons from the ecological and botanical research community. Here, we discuss the problems of data processing and output customisation arising from the different information needs of three user groups and the heterogeneous data quality in the Traitbase due to the different types of data sources. To this end, we first identify the user groups and their needs. Then, we introduce the terms aggregation and abstraction for the given context. After this, we discuss issues of data quality and finally we go into details of the aggregation methods preferred in generating output for the different user groups.