Ecological soil screening levels (Eco-SSLs) are soil concentrations of chemicals which, when not exceeded, will theoretically protect terrestrial ecosystems from unacceptable harmful effects due to chemicals. Eco-SSLs are derived using data generated from laboratory toxicity tests with many different test organisms that theoretically represent the vast array of naturally occurring ecological receptors. Even though research has shown that the bioavailability of chemicals to organisms in soil is substantially less than 100 percent, it is typically been assumed that all of the chemical in soil is bioavailable to soil invertebrates and plants during the derivation of Eco-SSLs. Bioavailability of chemicals to organisms in soil is modified by a vast array of physical, chemical, and biological modifying factors. In order to screen contaminated sites for the ecological risk assessment process, it was necessary to develop rapid, inexpensive, routine methods for measuring the bioavailability of chemicals for plants and soil invertebrates.

The objective of this project was to address how chemical bioavailability and toxicity to soil invertebrates and plants is modified by soil physical/chemical properties and how bioavailability can be measured in soil systems for compounds of concern to terrestrial ecosystems, including polycyclic aromatic hydrocarbons (PAHs), 2,4,6-trinitrotoluene (TNT), hexahydro-trinitro-triazine (RDX), cadmium (Cd), lead (Pb),arsenic (As), and zinc (Zn).

This project involved the spiking of chemicals to soils varying in physical/chemical parameters. Spiked soils were subject to wet/dry cycles and chemical availability was monitored over time to ensure that rapid soil/chemical interactions have occurred prior to conducting bioassays. Bioassays were conducted with earthworms, enchytraeids, Collembola, and a number of plant species. Endpoints monitored during the bioassays included reproduction for invertebrates and germination and early seedling growth for plants. Chemical exposure was expressed as total chemical concentration and as well as measures more reflective of the level of chemical bioavailable to test organisms. Solid-phase extraction techniques were used as measures of bioavailability for organic compounds, while sequential extraction techniques ranging from weak salt solutions to digestion with nitric acid were used for metals. Chemical bioavailability was also determined through residues in the organisms, both in terms of bioaccumulation (non-toxic levels) and residues associated with sublethal effects such as growth and reproduction (critical body residues). Multiple regression models were developed to describe empirical relationships between soil physical/chemical characteristics and toxicity and/or bioaccumulation in soil invertebrates and plants.

Soils studied included Teller and Sassafras sandy loams (TSL and SSL, respectively) and Richfield, Kirkland, and Webster clay loams (RCL, KCL, and WCL, respectively).

The project compiled and reviewed ecotoxicological benchmarks for terrestrial plants and soil invertebrates in order to develop scientifically defensible terrestrial plant-based and soil invertebrate-based draft Eco-SSLs for TNT and RDX. The draft Eco-SSL values were derived using the EC₂₀ (the 20% negative) level toxicity benchmarks for the effects of each explosive compound as a contaminant in soil and on plant growth or soil invertebrate reproduction endpoints, which were determined in standardized toxicity tests. Ecotoxicological testing was specifically designed and reviewed for these studies to meet the U.S. Environmental Protection Agency (U.S. EPA) criteria for Eco-SSL derivation. Following acceptance by the U.S. EPA, these draft Eco-SSL values will be used to screen site-soil data to identify the contaminants that are not of potential ecological concern and do not need to be considered in the baseline ecological risk assessment, which will result in significant cost-savings during site assessments. These Eco-SSLs will also provide an indispensable tool for installation managers to gauge the ecotoxicological impacts of military operations that involve the use of explosives, thus ultimately promoting the sustainable use of testing and training ranges.

Reproduction toxicity of TNT weathered-and-aged in soil, based on EC₅₀ values, correlated strongly with soil organic matter and clay contents. The order of soil toxicity (from greatest to least) was: TSL > SSL > KCL = RCL > WCL. RDX was less toxic to E. crypticus compared with TNT in the same soil types. RDX toxicity for E. crypticus was greater in sandy loam soils compared with clay loam soils.

Reproduction toxicity of TNT to E. fetida in freshly amended soils was in the order (greatest to least) of TSL > SSL = KCL = RCL > WCL. Weathering-and-aging of TNT in SSL, KCL, RCL, and WCL soils increased the toxicity to E. fetida compared with corresponding freshly amended treatments. Reproduction toxicity of RDX weathered-and-aged (W-A) in soil was comparable with that of TNT W-A in TSL, SSL, RCL, and WCL soils. No clear relationships were identified between TNT or RDX toxicities and the soil organic matter or clay contents or the pH levels.

Based on the effective concentrations for 50% of the population (EC₅₀ values), for TNT weathered-and-aged in soil, chronic toxicity to F. candida was in the order KCL > RCL > TSL > WCL > SSL, and for RDX weathered-and-aged in soil, the order was RCL > WCL > TSL > KCL > SSL. Organic matter was the dominant soil property that mitigated TNT toxicity for adult survival in freshly amended soil; soil pH correlated strongly with acute toxicity benchmarks (the effective concentrations for 20% of the population [EC₂₀ values] for adult survival) for RDX freshly amended in soil. These correlations were not sustained after weathering-and-aging of TNT or RDX in soil.

Reproduction was a more sensitive endpoint than adult survival for evaluation of exposure effects for E. crypticus, E. fetida and F. candida; therefore, reproduction endpoint-based toxicity benchmarks should be used to set screening criteria for TNT and RDX. This finding also supports the U.S. EPA Eco-SSL requirement of using reproduction endpoints for toxicity benchmark development.

Kuperman, R. G., R. T. Checkai, M. Simini, C. T. Phillips, J. E. Kolakowski, and C. W. Kurnas. 2005. Weathering and Aging of 2,4,6-Trinitrotoluene in Soil Increases Toxicity to Potworm Enchytraeus crypticus. Environmental Toxicology and Chemistry, 24(10):2509-2518.

Kuperman, R. G., R. T. Checkai, M. Simini, C. T. Phillips, J. E. Kolakowski, C. W. Kurnas, and G. I. Sunahara. 2003. Survival and Reproduction of Enchytraeus crypticus (Oligochaeta, Enchytraeidae) in a Natural Sandy Loam Soil Amended with the Nitro-heterocyclic Explosives RDX and HMX. Pedobiologia, 47:651-656.

Kuperman, R. G., R. T. Checkai, M. Simini, C. T. Phillips, J. E. Kolakowski, and R. Lanno. 2013. Soil Properties Affect the Toxicities of 2,4,6-Trinitrotoluene (TNT) and Hexahydro-1,3,5-Triazine (RDX) to the Enchytraeid Worm Enchytraeus crypticus. Environmental Toxicology and Chemistry, 32(11):2648-2659. 

Rocheleau, S., R. G. Kuperman, M. Martel, L. Paquet, G. Bardai, S. Wong, M. Sarrazin, S. Dodard, P. Gong, J. Hawari, R. T. Checkai, and G. I. Sunahara. 2005. Phytotoxicity of Nitroaromatic Energetic Compounds Freshly Amended or Weathered and Aged in Sandy Loam Soil. Chemosphere, 62:545-558.