Mean ecological indicator values: use EIVE but no cover-weighting

作者：Grzegorz Ostrowski, Severin Aicher, Agnieszka Mankiewicz, Olha Chusova, Iwona Dembicz, Stefan Widmer, Jürgen Dengler · 发表于：Vegetation Classification and Survey · 年份：2025 · DOI：10.3897/vcs.134800 · 被引用次数：3 · 研究领域：Sustainable Agricultural Systems Analysis、Environmental Conservation and Management、Botany and Plant Ecology Studies

Aims : To test the predictive power of mean ecological indicator values (EIVs) based on different EIV systems and weighting approaches. Study area : Preda in Grisons, Switzerland. Methods : We used three regional datasets of vegetation plots accompanied with measured soil pH values or mean annual near soil air temperature. We calculated mean EIVs for each plot with four EIV systems that cover the region, namely “Ellenberg”, “Landolt”, “Tichý” and the Ecological Indicator Values for Europe (EIVE), combined with four weighting approaches (unweighted, cover-weighted, square-root cover-weighted, inverse niche width weighted). We correlated the mean EIVs of each combination with the measured environmental variables and compared the mean Pearson r values. Results : No cover-weighting (0.78) was slightly better than square-root cover-weighting (0.75) and clearly better than full cover-weighting (0.68). In the two EIV systems providing a niche width measure (EIVE and “Landolt”), inverse niche weighting gave similar results than no weighting. Mean EIVE values (0.76) had significantly stronger correlations than “Tichý” (0.73), while the differences to “Ellenberg” (0.75) and “Landolt” (0.71) were not significant. Conclusions : The results suggest that even within the definition areas of two long-established EIV systems (“Ellenberg”, “Landolt”), EIVE gives at least as good predictions as these and significantly better than “Tichý”. Likewise, any type of cover-weighting reduces the predic...