Characterisation of Exposure – Bioavailability
The characterisation of chemical exposure involves the estimation of a chemical dose or concentration in contact with an organism. The assumption is often made that the entire amount of a chemical present in soil is available for uptake and assimilation by potential receptors. Numerous studies in the past two decades have shown that organic compounds and metals in soil are not entirely bioavailable. (3) For example, a large fraction of petroleum material may be irreversibly sorbed to or sequestered in soil. Therefore, studies have been taken to quantify the bioavailability of chemicals at petroleum-contaminated sites. Researchers at Lawrence Berkeley National Laboratory (LBNL) are developing models and using controlled chamber studies to estimate plant uptake of hydrocarbons. Preliminary data suggest that the ratios of plant concentrations to soil concentrations for polycyclic aromatic hydrocarbons and n-alkanes (C-20-C-30) are lower than those for other organic compounds. Regressions for plant uptake of metals from soil have been developed by ORNL, but the predictions are highly uncertain, and assessors would benefit from the incorporation of additional soil characteristics. (4) LBNL has been investigating the gastrointestinal bioavailability of petroleum hydrocarbons to improve human health risk assessment, but methods of investigation are applicable to risk assessments for wildlife as well. (5)
Figure 2: A Preliminary Ecological Framework for
Evaluating Terrestrial Vertebrate Populations at E&P Sites
Characterisation of Effects – Population Ecology
An ecological risk assessment for vertebrates should specify whether the management goal (which may be based on a regulation) is to protect all or a fraction of individuals, or to protect the population. For threatened or endangered species, the protection of the population entails the protection of individuals. However, in most cases, the goal of ecological risk assessment for vertebrates is to protect the population. Brine and petroleum spills may affect terrestrial vertebrates through loss of reproductive habitat or reduced food availability rather than direct toxicity. In risk assessments, spills may sometimes be treated like roads and wells (i.e. as physical stressors), rather than chemicals. (6) An ecological framework for evaluating impacts of spills has been proposed by ORNL in collaboration the Lawrence Livermore National Laboratory (LLNL). The framework recommends the use of individualbased models (IBMs) to assess risk to wildlife populations. IBMs can simulate mechanistic links between the physical environment, as modified by human activities, and biological processes that influence individual animals (e.g. mortality, reproduction, ageing and mate choice). Such models were used to simulate American badger (Taxidea taxus) and prairie vole (Microtus ochrogaster) populations at the Tallgrass Prairie Preserve E&P site in Osage County, Oklahoma, using both existing and hypothetical spill areas and spatial distributions. (7) The size of simulated populations decreased with increasing brine spill area (see Figure 2). The value of this modelling exercise would be increased by additional research aimed at field verification of the results. Future population modelling to investigate habitat fragmentation from spills and E&P infrastructure such as roads is planned on an E&P site in Utah with funding from the US Bureau of Land Management. Similarly, large spills and roads may interrupt ecological corridors. A tool called ‘pathway analysis through habitat’ has been developed to predict the location of favoured corridors of animal movement between patches of habitat within any map. (8) Virtual ‘walkers’ with habitat preferences of particular species are simulated using a parallel supercomputer. This tool requires a map of habitat suitability categories, a map of spatially contiguous patches of each habitat suitability category, and the habitat category or categories between which corridors are to be identified. In addition, species-specific data are needed – habitat preferences, as well as energy costs of movement, the likelihood of finding food and likelihood of mortality in each habitat type. Corridor results between meadows in Yellowstone National Park are depicted in Figure 3. Through the use of this tool, habitat corridors may be avoided during the siting of infrastructure, and spill restoration priorities can be identified.
Figure 3: Simulated Habitat Loss from
Hypothetical Brine Scars Decreased Final Number
of Badgers at the Tallgrass Prairie Preserve in Oklahoma, USA
Model and figure provided by H I Jager,
Oak Ridge National Laboratory, Tennessee, US
Category:
Environment
|
Download PDF
References
