PMISER is a parallel version of a previously developed state of the art serial environmental remediation simulator, MISER. This simulator is the product of an interdisciplinary effort between computer scientists and application engineers and is designed to utilize powerful parallel architectures to model the physical, chemical, and biological interactions in field scale soil vapor extraction (SVE) and bioventing (BV) systems in two space dimensions. MISER is based on standard Galerkin finite element techniques with linear triangular elements. The partial differential equations solved describe multiphase flow, multicomponent advective diffusive transport, and bioreaction. These equations are typically highly nonlinear and strongly coupled due to the nature of the constitutive relationships, material properties, and bioreaction terms. Rate limited mass exchange between phases is modeled with linear driving force expressions. The resulting sets of flow, transport, and biodegradation equations are solved sequentially using a set-iterative approach. The set-iterative approach substantially reduces the dimensions of solution matrices and provides increased model flexibility. PMISER was developed using the Message Passing Interface (MPI). Parallelization was accomplished with a domain decomposition approach by partitioning the global domain into blocks, each assigned to an individual processor. A parallel solver package, Aztec, developed at Sandia National Laboratories is used to solve the resulting partitioned set of equations. The motivating factor behind the development of PMISER was the need for larger simulation domains with more refined numerical grids, and the inclusion of more complex and complete descriptions of the subsurface environment (i.e. more components, processes, and/or phases). These issues of scale and model complexity are critical to the accurate simulation of subsurface remediation. Examples will be presented which highlight the complex nature of the modeled processes and demonstrate the performance gains achieved by parallelization. Acknowledgment: This work was funded as part of the Applications Thrust Area in Engineering supported by the National Partnership for Advanced Computational Infrastructure (NPACI). Cooperating institutions are the Center for Subsurface Modeling at the University of Texas and the Center for Parallel Computing at the University of Michigan.