Pilot point interpolation exceeding parameter bounds in PEST–MODFLOW calibration

[Sergio Fernando Poveda Salazar]

Hi,

I’m calibrating a MODFLOW-2005 model using PEST through ModelMuse.

In the locations of the pilot points, the estimated hydraulic conductivity values respect the specified lower and upper bounds. However, in areas farther from pilot points, the interpolated K values become unrealistically high — completely outside the defined bounds.

I was expecting PEST/MODFLOW to enforce these bounds automatically during interpolation, but it seems this is not the case.

Could someone please clarify whether this behavior is expected, and what options I have to constrain or clip the interpolated values (for example, within ModelMuse or by modifying the pilot point setup)? I'm a biit concerned about the impact this could have on the historical matching.

Thanks in advance for any guidance.

Sergio Poveda