Toxicity bioassays for ecological risk assessment in arid and semiarid ecosystems.

Substantial tracts of land in the southwestern and western U.S. are undergoing or will require ERA. Toxicity bioassays employed in baseline ERAs are, for the most part. representative of mesic systems, and highly standardized test species (e.g., lettuce, earthworm) are generally not relevant to arid system toxicity testing. Conversely, relevant test species are often poorly characterized with regard to toxicant sensitivity and culture conditions. The applicability of toxicity bioassays to ecological risk assessment in arid and semiarid ecosystems was reviewed for bacteria and fungi, plants, terrestrial invertebrates, and terrestrial vertebrates. Bacteria and fungi are critical to soil processes, and understanding their ecology is important to understanding the ecological relevance of bioassays targeting either group. Terrestrial bacteria require a water film around soil particles to be active, while soil fungi can remain active in extremely dry soils. It is therefore expected that fungi will be of greater importance to arid and semiarid systems (Whitford 1989). If microbial processes are to be measured in soils of arid environments, it is recommended that bioassays target fungi. Regardless of the taxa studied, problems are associated with the standardization and interpretability of microbial tests, and regulatory acceptance may hinder widespread incorporation of microbial toxicity bioassays in arid system risk assessments. Plant toxicity bioassays are gaining recognition as sensitive indicators of soil conditions because they can provide a cost-effective and relatively rapid assessment of soil quality for both pre- and postremediation efforts. Phytotoxicity evaluations commonly target germination because environmental stressors have the greatest potential for exerting adverse effects in the early stages of growth. In arid systems, seeds respond rapidly to precipitation events, and it is typically after germination has occurred that plants must cope with water stress. Consequently, seedling emergence studies should be conducted under nonlimiting moisture conditions characteristic of mesic plant testing. Further ecological realism can be incorporated into advanced growth stages by creating moisture conditions representative of the arid system study site. Although the choices of suitable plant species for assessing mesic system soils are numerous, the choices for arid system soils are limited. Guidance is provided for evaluating plant species with regard to their suitability for serving as representative arid system flora. Terrestrial invertebrates can survive and flourish in extremely dry conditions. They play key roles in ecosystem functioning in arid environments. Perhaps the biggest drawback to using terrestrial invertebrates for toxicity bioassays involves uncertainties associated with choosing appropriate test species. Several examples of standard species exist for mesic soils (e.g., the earthworm Eisenia foetida and the collembolan Folsomia candida), whereas no analogous organisms are available for testing arid and semiarid soils. The aid of an expert taxonomist and some basic research are prerequisite to using ecologically relevant invertebrates. The use of birds for ecotoxicity testing in arid and semiarid environments is not recommended. On the other hand, mammals, especially rodents, are well represented in arid ecosystems. Much of the ecotoxicity testing performed on rodents is generally applicable to arid-adapted species; few demonstrations of rodent ecotoxicity testing for dry environments exist. Relative to other organisms discussed, such as soil invertebrates, the use of mammals in toxicity bioassays faces several obstacles. Terrestrial plants and soil invertebrates appear to be the most appropriate and feasible organisms for ecotoxicity testing in arid and semiarid environments. Potentially relevant test species for arid system testing are often poorly characterized with regard to toxicant sensitivity and culture conditions. Table 6 presents examples of standard and nonstandard species with these considerations in mind, and the best estimate of regulatory acceptance for each of the organisms is suggested. If currently accepted bioassays are not appropriate for evaluating risks in arid and semiarid ecosystems, their use in conducting ERAs in such environments may result in inadequate expenditure of time and money to develop data that accurately characterize risks. The inapplicability of this technical tool will thus hamper the risk management decision-making process and result in flawed decisions.