# How to manipulate horizontal diffusion coefficient?

Hello.

I have a question about pollutant transport modeling of dye.
I’d like to manipulate diffusion of pollutant cloud.
In EFDC manual, AHO (horizontal eddy viscosity) is the same as mass diffusivity.
(And, I understand that mass diffusivity is horizontal diffusion coefficient.)
So, I changed AHO to check the variation of peak concentration of dye at a specific point.

I think if AHO is the same as mass diffusivity, peak concentration should be decreased when AHO is increasing.
However, peak concentration increased when AHO is increased.

So, I am curious how the AHO constant is related to the diffusion coefficient in what way.

There are two different ways in which mass diffusion can impact the model. If you add diffusivity to a system it essentially makes it more viscous and dampens the energy – the peak concentrations can go up as you have slowed the transport by using up energy in the cell. AHO represents the eddy viscosity which uses energy in a single cell – so when you add diffusivity then energy is reduced for the overall transport and you can get higher concentrations.

However, in EE it is not automatic that mass is adjusted to diffusivity values – in Hydrodynamics Tab>Turbulence Options>Modify>Turbulent Diffusion tab, you must set the diffusivity option to 2 (ie. Activate HMD with Smagorinsky and WC diffusion). In this case Wall Roughness cannot be set to zero. Only after setting this is diffusivity being applied to constituent transport as well as hydrodynamics.

Hi, could I ask in the coastal tide case, WHAT are the normal values or default values for the horizontal eddy viscosity coefficient (AHO) and Dimensionless Horizontal momentum diffusion coefficient (AHD)? I saw some test cases, different values are used in different cases, AHO=10, and AHD=0.05 in one case, while in another case, AHO=0.3, and AHD=0.25, which has a big difference. anyone knows if there is a standard rule to set these values, thanks very much!

Hi Zhou,

For AHD values may vary from 0.05 to 0.25 with typical values of 0.05 to 0.1

For AHO the value is specifically grid dependent – one quick way to get an approximate value is to find the area of the smallest cell in area of interest and divide it by 1000.