strange eddy viscosity / number of digits

I am modeling two lakes using CE-QUAL-W2 3.6. The model systematically underestimates vertical mixing. Eddy viscosity is simulated using the W2N formulation. I wanted to have a look at the simulated eddy viscosities in the snapshot file in order to check for potential reasons for the problem. There are two points I would like to be discussed: - first, the output is only three digits after the decimal point. Since most of the eddy viscosities in lakes are below 10^-3 m2 s-1, most of the values in the snapshot file are 0.000. Is there any way to change this? - second, during times of high eddy viscosity in the well mixed surface layer, often high viscosity in one layer alternates with low viscosity in the next one over a large number of layers, see example below. This seems quite strange to me. Did anyone ever observe something similar, and what could be the reason for this? Thanks, Martin ------------------- extract from snapshot file showing alternating eddy viscosities. April 29, 1988 Julian day = 2676 days 0.00 hours Vertical eddy viscosity [AZ], m^2/s Layer Depth 34 4 1.18 0.002 5 3.36 0.074 6 5.36 0.000 7 7.36 0.060 8 9.36 0.000 9 11.36 0.041 10 13.36 0.000 11 15.36 0.031 12 17.36 0.000 13 19.36 0.024 14 21.36 0.000 15 23.36 0.015 16 25.36 0.001