Characterizing the relationship between different taxonomic groups is critical to identify potential surrogates for biodiversity. Previous studies have shown that cross-taxa relationships are generally weak and/or inconsistent. The difficulties in finding predictive patterns have often been attributed to the spatial and temporal scales of these studies and on the differences in the measure used to evaluate such relationships (species richness versus composition). However, the choice of the analytical approach used to evaluate cross-taxon congruence inevitably represents a major source of variation. Here, we described the use of a range of methods that can be used to comprehensively assess cross-taxa relationships. To do so, we used data for two taxonomic groups, wetland plants and water beetles, collected from 54 farmland ponds in Ireland. Specifically, we used the Pearson correlation and rarefaction curves to analyse patterns in species richness, while Mantel tests, Procrustes analysis, and co-correspondence analysis were used to evaluate congruence in species composition. We compared the results of these analyses and we described some of the potential pitfalls associated with the use of each of these statistical approaches. Cross-taxon congruence was moderate to strong, depending on the choice of the analytical approach, on the nature of the response variable, and on local and environmental conditions. Our findings indicate that multiple approaches and measures of community structure are required for a comprehensive assessment of cross-taxa relationships. In particular, we showed that selection of surrogate taxa in conservation planning should not be based on a single statistic expressing the degree of correlation in species richness or composition. Potential solutions to the analytical issues associated with the assessment of cross-taxon congruence are provided and the implications of our findings in the selection of surrogates for biodiversity are discussed
Evaluating and interpreting cross-taxon congruence: Potential pitfalls and solutions
BACARO, Giovanni;
2011-01-01
Abstract
Characterizing the relationship between different taxonomic groups is critical to identify potential surrogates for biodiversity. Previous studies have shown that cross-taxa relationships are generally weak and/or inconsistent. The difficulties in finding predictive patterns have often been attributed to the spatial and temporal scales of these studies and on the differences in the measure used to evaluate such relationships (species richness versus composition). However, the choice of the analytical approach used to evaluate cross-taxon congruence inevitably represents a major source of variation. Here, we described the use of a range of methods that can be used to comprehensively assess cross-taxa relationships. To do so, we used data for two taxonomic groups, wetland plants and water beetles, collected from 54 farmland ponds in Ireland. Specifically, we used the Pearson correlation and rarefaction curves to analyse patterns in species richness, while Mantel tests, Procrustes analysis, and co-correspondence analysis were used to evaluate congruence in species composition. We compared the results of these analyses and we described some of the potential pitfalls associated with the use of each of these statistical approaches. Cross-taxon congruence was moderate to strong, depending on the choice of the analytical approach, on the nature of the response variable, and on local and environmental conditions. Our findings indicate that multiple approaches and measures of community structure are required for a comprehensive assessment of cross-taxa relationships. In particular, we showed that selection of surrogate taxa in conservation planning should not be based on a single statistic expressing the degree of correlation in species richness or composition. Potential solutions to the analytical issues associated with the assessment of cross-taxon congruence are provided and the implications of our findings in the selection of surrogates for biodiversity are discussedPubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
