DO

In the model, the concentration of DO is generated based on temperature, depth and salinity (Appendix A.3.4). Thus, we need to ensure that under similar temperature and salinity inputs to both systems, and with similar quantities of clam and background prey biomass, DO behaves similarly in the two systems. Validation of the DO, clam and background prey models was done iteratively since these models depend on each other. The particular behaviors focused on for validation include: seasonal component to DO variation; lower DOs in summer ranging from 0 to 8 and 0 to 12 (mg/L) in the winter; percent of estuary experiencing hypoxia at any time should be within $\approx$ 0% to $\approx$ 50% during July through October, depending on the year; the length of time areas of moderate depth remain continuously hypoxic should last anywhere from one day to two weeks.

The duration, extent and seasonal fluctuations in hypoxia agreed with those reported for the Neuse  (Selberg et al. 2001, Table 3) (Buzzelli et al. 2002, Table 2). Shallow portions of the model estuary (depth $<$ 2 m) were never hypoxic and the median time the deepest areas remained hypoxic was $\approx$ 9 days (Fig. B1). The empirically measured durations of hypoxia at stations LT9 and LT11 in the Neuse over years 1989 to 1999 are given in Fig. B2. However, the large amount of missing data in these time series biases the durations, making them appear shorter than they actually are. We conclude that the ranges displayed in both Figs. B1 and B2 are in general agreement. The maximum percentage of the estuary hypoxic at any given time during summer ranged between 40% and 60% (Fig. B3), agreeing with empirical measurements. For the model estuary, DO in summer ranged from 0 to 8 and 0 to 12 (mg/L) in the winter and mirrored changes in temperature.