Final Comments on Model Validation Report, E. Adams, July 2006

Sorption kinetics The model assumes that three chemical phases of PCBs are in equilibrium. As EPA points out, this is a common modeling assumption, but I believe it is only reasonable within a stationary sediment bed, and not within the water column during sediment resuspension. Following Wu and Gschwend (1986), and focusing only on sorption and desorption among two phases (sorbed PCB with concentration C s and dissolved PCB with concentration C d) d p s d s d p s C k K C k dt dC C k C K k dt dC 2 2 1 1 / − = − = (1) where k 1 and k 2 are rate constants constrained by p K k k ρ / 2 1 = (2) where ρ is the solid-water phase ratio φ ρ φ ρ /) 1 (s − = , ρ s is the solid density, φ is porosity and K p is the partition coefficient. For a stationary sediment bed, ρ is about 1, so for hydrophobic contaminants (large K p), most of the contaminant is sorbed; hence the processes of sorption and desorption cause C d to vary in the range p s d K C C / 0 ≤ ≤ , whereas C s remains nearly constant. The characteristic time for sorption/desorption for highly particle-reactive species in a stationary sediment bed (ρK p >> 1) is thus k 2-1. Wu and Gschwend (1986) describe the sorption/desorption process as one of molecular diffusion through particles (or particle aggregates). For large K p and an assumed intra-aggregate porosity, their effective (retarded) diffusion coefficient is given by s p m eff K D D ρ 2. 0 ≅ (3) where D m is the molecular diffusivity of PCBs. Wu and Gschwend (1988) fit a first-order sorption kinetics model to the radial diffusion model over the (initial) time during