Model-based assessment of estuary ecosystem health using the latent health factor index, with application to the richibucto estuary

• Published in: PloS one Vol. 8; no. 6; p. e65697
• Main Authors: Chiu, Grace S, Wu, Margaret A, Lu, Lin
• Format: Journal Article
• Language: English
• Published: United States Public Library of Science 13.06.2013; Public Library of Science (PLoS)
• Subjects: Algorithms; Aquaculture; Aquaculture industry; Aquatic ecosystems; Biodiversity; Bioindicators; Biological monitoring; Biology; Biomonitoring; Biota; Clay; Data processing; Ecological monitoring; Ecosystem; Ecosystem assessment; Ecosystems; Environment; Environmental Health; Estuaries; Estuarine ecosystems; Estuarine environments; Freshwater ecosystems; Health; Mariculture; Mathematical models; Mathematics; Methods; Models, Statistical; New Brunswick; Protection and preservation; Qualitative analysis; Salinity; Silt; Statistical inference; Statistical models; Studies; System effectiveness; New Brunswick; Europe; New Brunswick Canada
• ISSN: 1932-6203; 1932-6203
• DOI: 10.1371/journal.pone.0065697

The ability to quantitatively assess ecological health is of great interest to those tasked with monitoring and conserving ecosystems. For decades, biomonitoring research and policies have relied on multimetric health indices of various forms. Although indices are numbers, many are constructed based on qualitative procedures, thus limiting the quantitative rigor of the practical interpretations of such indices. The statistical modeling approach to construct the latent health factor index (LHFI) was recently developed. With ecological data that otherwise are used to construct conventional multimetric indices, the LHFI framework expresses such data in a rigorous quantitative model, integrating qualitative features of ecosystem health and preconceived ecological relationships among such features. This hierarchical modeling approach allows unified statistical inference of health for observed sites (along with prediction of health for partially observed sites, if desired) and of the relevance of ecological drivers, all accompanied by formal uncertainty statements from a single, integrated analysis. Thus far, the LHFI approach has been demonstrated and validated in a freshwater context. We adapt this approach to modeling estuarine health, and illustrate it on the previously unassessed system in Richibucto in New Brunswick, Canada, where active oyster farming is a potential stressor through its effects on sediment properties. Field data correspond to health metrics that constitute the popular AZTI marine biotic index and the infaunal trophic index, as well as abiotic predictors preconceived to influence biota. Our paper is the first to construct a scientifically sensible model that rigorously identifies the collective explanatory capacity of salinity, distance downstream, channel depth, and silt-clay content-all regarded a priori as qualitatively important abiotic drivers-towards site health in the Richibucto ecosystem. This suggests the potential effectiveness of the LHFI approach for assessing not only freshwater systems but aquatic ecosystems in general.