Trying to update version 3.0 model from 2002 to current version 3.6 -- slight differences

Hello, I am updating a version 3.0 model from 2002 to version 3.6. It is of a mining pit lake with a surface inflow and surface outflow. To keep it simple the only constituent used is TDS. I am using the exact same boundary conditions (bathymetry file, meteorology file, inflow, outflow, initial conditions, T and TDS of inflow, etc.). I have gone through the control file over and over and believe I have converted everything to the version 3.6 input format, keeping the same inputs from the original model. When I run the model, my results track the 2002 results fairly well in terms of trends, but they are not identical. Should they be identical? There seems to be some differences in vertical mixing. Have any changes been made in the algorithms related to vertical mixing since 2002? I have plotted up vertical profiles of T and TDS vs. depth, monthly for one year of simulation. I could send this along if it would help to see the differences. Basically I need to model current and possible future conditions, but am reluctant to proceed until I am confident that the version 3.6 model is as correct as I can get it, matching the original model results. Do you have any ideas? Thanks, Brian Park