Testing the CORMIX model using thermal plume data from four Maryland power plants

Abstract

Historical thermal plume studies from four Maryland power plants (Calvert Cliffs, Chalk Point, Dickerson, and Wagner) were used to test the realism of the CORnell MIXing Zone Expert System (cormix). Test data were from a wide range of challenging discharge environments, including a large freshwater river (Potomac), a narrow tidal estuary (Patuxent), a wide tidal estuary (Chesapeake Bay), and a wind-driven tidal estuary (Baltimore Harbor). Historical case studies were simulated, and results were compared qualitatively and quantitatively with historical measurements. Qualitative results show that the model performed optimally for simple discharges into large basins such as Chesapeake Bay. For complex discharges and complex ambient environments, the model often mixed plumes too rapidly, resulting in smaller modeled plumes that were cooler than the measured plumes. The mixing model also could not account for the re-entrainment of effluent from previous tidal cycles. Sensitivity results show that sensitivity is often dependent on model run time and discontinuities in the cormix flow classification scheme. Users of the cormix model need to be aware of these limitations in applying the model to complex situations. cormix results should be used with caution in evaluating the effects of a discharge and only in conjunction with information from the field.

Introduction

The Cornell Mixing Zone Expert System (cormix) is a software package with a series of modules for the analysis, prediction, and design of aqueous toxic or conventional pollutant discharges into diverse water bodies (Doneker and Jirka, 1990, Akar and Jirka, 1991, Jirka et al., 1996). It was developed under cooperative funding agreements from the U.S. Environmental Protection Agency, the Maryland Department of Natural Resources (MDNR) Power Plant Research Program (PPRP), and the Delaware Department of Natural Resources and Environmental Control. It is the recommended analysis tool in key guidance documents on the permitting of industrial discharges to receiving waters (USEPA, 1991a, USEPA, 1991b). The system's major emphasis is on predicting the geometry and dilution characteristics of the initial mixing zone to evaluate compliance with acute and chronic regulatory requirements. The purpose of this study was to test the cormix model using historical measurement of thermal plumes at selected Maryland power plants.

cormix was designed as a first-order, screening/design model. It does not carry out detailed hydrodynamic calculations using the exact geometry of the discharge location, nor does it explicitly handle dynamic ambient currents (i.e. tides), except as described below. It uses a simplified representation of the physical conditions at the discharge location to approximate the fundamental behavior of the plume.

cormix classifies discharge parameters into discrete categories which determine qualitative plume behavior. To predict an effluent discharge's dilution and plume trajectory, cormix typically breaks the prediction problem into several stages. A solution is calculated for the steady-state, simplified flow patterns that characterize each stage. These solutions are combined to provide a complete simulation from the outflow point to the distance limit set by the user.

Two formal attempts have been made to adapt cormix to tidal situations. The Delaware River Basin Commission (DRBC) undertook an effort to automate Version 2.10 of cormix for mixing zone evaluations in tidal waters, specifically the Delaware River Estuary, and to provide post-processing capabilities (DRBC, 1995). The fundamental approach was simply to run multiple cormix simulations at even time intervals throughout a tidal cycle, while the preprocessing program adjusted ambient water heights and velocity parameters automatically. This method produces the normal cormix prediction files; however, the user must interpret the data and apply it case by case to understand the plume dynamics. In some cases, the DRBC tidal version can be a useful screening tool, but for shallow surface discharges it forces assumptions about changes in water depth in the discharge channel that may not be realistic. In addition, the DRBC version tends to cause computer malfunctions for some simulations.

In a separate effort, PPRP funded the cormix team at Cornell to develop and incorporate a module to deal with tidal situations, particularly with respect to heated discharges of power plant cooling water (Nash and Jirka, 1995). This work, incorporated in cormix Version 3.20 used in the present study, developed a model of plume behavior during tidal current reversal. On the basis of laboratory experiments, a set of pseudo- steady-state variables were defined to allow the standard cormix solutions to be modified for some tidal effects. These modifications can improve the plume boundary predictions in the interval near slack tide but do not account for thermal build-up and re-entrainment over multiple tidal cycles.