Events and Meetings of Italian Statistical Society, Advances in Latent Variables - Methods, Models and Applications

Font Size: 
Bayesian analysis of the multivariate dependence of three transitional water ecosystem classifications
Alessio Pollice, Serena Arima, Giovanna Jona Lasinio, Alberto Basset, Ilaria Rosati

Last modified: 2013-06-14

Abstract


The Water Framework Directive (WFD) recognizes benthic macroinvertebrates as a good biological quality element for transitional waters as they are the most exposed to natural variability patterns characteristic of these ecosystems, due to their life cycles and space-use behavior. Here, we address the ecological status classification issue for three lagoons in Apulia, using benthic macroinvertebrates and three proposed multimetric indices (namely M-AMBI, BITS and ISS), likely to respond differently to different sources of stress and natural variability. Lagoon classification is based on discretization by standard classification boundaries with only partial consideration of the natural variability of ecosystem properties and possible inaccuracies of the classification procedures. In order to investigate the possible contrasting behavior of the three classifications, we propose Bayesian hierarchical models in which the multimetric indices and their discrete counterparts are jointly modeled as function of abiotic covariates, external anthropogenic pressures indicators and spatio-temporal effects.


Full Text: PDF