Skip to Main content Skip to Navigation

Graphical modeling of metal bioavailability to earthworm

Abstract : Soil organisms like earthworms are exposed to trace metals in contaminated soils. Only a fraction of total metal content in the environment can interact with soil organisms and lead to toxicological effects. Risk assessment of contaminated soils needs to measure the extent of this interaction between metals and organisms (i.e. bioavailability). Bioavailability is how- ever a complex process that cannot be measured directly. A three-step definition had been given: 1) environmental availability designates the metal pools in soil that can potentially be uptake by the organism, 2) environmental bioavailability reflects the process of uptake of the contaminant by the organism and 3) toxicological bioavailability indicates the effects of the contaminant on the organism. A number of chemical and biological methods exist to measure each of these steps but none of them was proven to be fully generic in realistic envi- ronmental contexts. In situ, soils are often lowly contaminated by multiple contaminants and the causal relationships between chemical and biological indicators in such conditions have to be established. In this thesis we propose a graphical model (structural equation model: SEM) to address the causal hypotheses implied by the three-step definition of bioavailability. A laboratory exposure experiment was designed to test a SEM of metal bioavailability to earthworm. We chose a target earthworm species, A. caliginosa, commonly found in temper- ate soils. An analysis of literature data was conducted to verify metal bioaccumulation in A. caliginosa was representative of other earthworm species. We further selected a wide panel of soils to explore a realistic gradient of metal exposure. Specific chemical and biological methods were selected to reflect each step of metal bioavailability in the model. Environ- mental availability was assessed by experimental (extractions) and theoretical (modeling) procedures. Metal quantifications in earthworms and in three subcellular compartments were conducted to constitute observed variables of environmental bioavailability. Finally, for toxicological bioavailability, biomarkers reflecting different processes by which metals exert their effects on earthworms were quantified. The strength of the relationships between single observed variables was investigated over the panel of soils. The results showed indicator- and metal-specific relationships and further highlighted the challenges of relating chemical and biological indicators when considering low levels of bioavailability. The conceptual SEM was then confronted to our empirical observations. The results identified sets of chemical and biological indicators verifying the causal assumptions of the three-step definition of bioavail- ability in field-contaminated soils. This study shows the relevance of the SEM approach, that goes beyond the imperfections of single observed variables to reflect complex theoretical en- tities, and represents a powerful and comprehensive framework to investigate bioavailability in realistic environmental context.
Document type :
Complete list of metadata

Cited literature [231 references]  Display  Hide  Download
Contributor : Migration Prodinra <>
Submitted on : Friday, August 7, 2020 - 1:08:47 PM
Last modification on : Friday, October 9, 2020 - 3:08:19 AM
Long-term archiving on: : Monday, November 30, 2020 - 4:20:38 PM


Files produced by the author(s)


  • HAL Id : tel-02796794, version 1
  • PRODINRA : 273509



Léa Beaumelle. Graphical modeling of metal bioavailability to earthworm. Life Sciences [q-bio]. ABIES Doctoral School; AgroParisTech, 2014. English. ⟨tel-02796794⟩



Record views


Files downloads