pht3d - icbund question

[grk...@gmail.com]

Hi all, particularly PHT3D users,

I'm simulating transient flow and reactive transport (ion exchange + calcite dissolution/precipitation) in a large 3D domain. The MODFLOW model alone runs in about 15 minutes.

To reduce PHT3D runtime, I tried restricting the reactive transport domain by defining ICBUND > 0 only in my area of interest. It’s still a large zone, but significantly smaller than the full model. Between this active region and the upgradient eastern CHD boundary, there’s a large gap.

If I define this gap as ICBUND = 0 (inactive), the model runs fast — but pure water enters the domain from the east, causing unrealistic demineralization of the reactive zone.

To prevent this, I tried setting the gap to ICBUND = -1 with constant concentration values, so that inflowing water has realistic chemistry. This solves the chemical boundary issue — but the simulation becomes 5–10 times slower than even a fully active domain.

Has anyone dealt with this before?

Any suggestions on how to preserve chemically realistic inflow without killing performance?

Thanks!