There is considerable concern over the potential ecotoxicity to soil ecosystems posed,by zero-valent iron nanoparticles (Fe-0 NPs) released from in situ environmental remediation. However, a lack of quantitative risk assessment has hampered the development of appropriate testing methods used in environmental applications. Here we present a novel, empirical approach to assess Fe-0 NPs-associated soil ecosystems health risk using the nematode Caenorhabditis elegans as a model organism. A Hill-based dose-response model describing the concentration-fertility inhibition relationships was constructed. A Weibull model was used to estimate thresholds as a guideline to protect C. elegans from infertility when exposed to waterborne or foodborne Fe NPs. Finally, the risk metrics, exceedance risk (ER) and risk quotient (RQ) of Fe-0 NPs in various depths and distances from remediation sites can then be predicted. We showed that under 50% risk probability (ER = 0.5), upper soil layer had the highest infertility risk (95% confidence interval: 13.18-57.40%). The margins of safety and acceptable criteria for soil ecosystems health for using Fe-0 NPs in field scale applications were also recommended. Results showed that RQs are larger than I in all soil layers when setting a stricter threshold of similar to 1.02 mg L-1 of Fe NPs. This C. elegans biomarker-based risk model affords new insights into the links between widespread use of Fe-0 NPs and environmental risk assessment and offers potential environmental implications of metal-based NPs for in situ remediation. (C) 2016 Elsevier B.V. All rights reserved.