Dissolved Oxygen

I'm modeling DO throughout the water column in a lake 10 m deep with 1 inlet, 1 outlet and 1 distributed tributary. I'm only using the zero order compartment for SOD and I've used values ranging from 0.2 to 1.0 g/m2/day which I believe are representative of lakes in this region (northern Alberta). I have BOD, LDOM, RDOM, LPOM and RPOM as active constituents with values assigned for initial lake concentrations and the two inputs two the lake. DO levels seem to react as expected when I change the Zero order SOD values (increases when SOD is decreased) but DO reacts opposite to what I expected when I change BOD values (decreases when BOD is decreased) or turn off the inlet DOM/POM compartments (leaving them as active constituents). I'm modelling from september to may and the values of interest occur late in the winter (mid to late march), under approximately 1 m ice since December, with max lake depth of 10 m. I'm using the Degray settings for BOD, nutrients and algae. Any thoughts on why the model is reacting like this and how to correct it would be appreciated.