Optimization of phytoremediation of contaminated soil with heavy metals and petroleum hydrocarbons using SEM and MCDM techniques.

This research presents an evaluation system and model for optimizing phytoremediation on petroleum and heavy metal contaminated soils to simplify the methods, reduce the costs, and minimize the use of long-term tests in the phytoremediation process. The proposed evaluation system consists of Structural Equation Modeling (SEM) and the Analytic network process (ANP) and includes three criteria and twelve sub-criteria. The calculated weights indicated that plant species and soil texture were the most influential factors in phytoremediation, with loading factors of 0.910 and 0.876, respectively. Seven dominant plants from the study area were used as alternative plants for phytoremediation, and the highest weight of ranking alternative (0.201) suggested that Medicago sativa L (Alfalfa) could be a suitable plant for this purpose and that the suitable texture of the soil would be sandy clay loam for the phytoremediation measure. Sensitivity analysis confirmed that the combination of the ANP and SEM in this course was impartial and attainable. The application of structural equation modeling and multi-criteria decision methods were found to be effective since they could help diagnose not only the effects of soil texture on petroleum hydrocarbons reduction but also the impacts of selected plant species on heavy metals reduction as factors in phytoremediation inside the study area. Therefore, it was concluded that combined application of these two methods would provide an effective approach in monitoring contaminated sites for decision makers. [ABSTRACT FROM AUTHOR]