Services on Demand
Journal
Article
Indicators
- Cited by SciELO
- Access statistics
Related links
- Similars in SciELO
Share
Boletín de la Sociedad Geológica Mexicana
Print version ISSN 1405-3322
Abstract
CORREA-METRIO, Alexander; DECHNIK, Yanus; LOZANO-GARCIA, Socorro and CABALLERO, Margarita. Detrended correspondence analysis: A useful tool to quantify ecological changes from fossil data sets. Bol. Soc. Geol. Mex [online]. 2014, vol.66, n.1, pp.135-143. ISSN 1405-3322.
Fossil assemblages reflect the parental communities that produced them, and are in turn associated with specific environmental conditions. Thus, climatic and environmental changes are associated with changes in both the biotic communities and the fossil assemblages they produce. As a consequence, the environmental reconstruction of the past relies on the interpretation of multivariate fossil sequences that are commonly analyzed through dimensional rescaling techniques. Detrended correspondence analysis (DCA) has proven as an excellent technique to summarize ecological changes through time with the advantage of few prior assumptions and results that can be directly interpreted in terms of ecological turnover. In this paper we elaborate on a brief description of the technique and the interpretation of results, using a worked example on pollen and diatom data sets from Lago Verde (Los Tuxtlas, Mexico). With this worked example, we highlight the three basic ways in which DCA can provide useful approaches for a clear and relatively easy interpretation of the fossil data: i) identification of the ecological space through the a priori interpretation of species ordination; ii) localization of time slices within the ecological space defined by species, and quantification of the ecological turnover among samples; and iii) calculation of ecological distances as a means for putting individual samples into the historical context provided by the time frame in question.
Keywords : Ecological distance; ecological turnover; ecological baseline; pollen; diatoms; detrended correspondence analysis (DCA).