Groundwater balance and percolation mismatch for swat+ gwflow model

[veerendra gopi]

Dear all,

I am currently working with a SWAT+ model coupled with GWFLOW and have encountered an issue regarding the groundwater water balance.

After enabling GWFLOW, I observed that the annual groundwater balance output (gwflow_balance_gw_yearly.txt) shows groundwater recharge (rech) values that are much lower than the annual percolation leaving the soil profile reported in the SWAT+ water balance. As a result, there is a significant mismatch between soil percolation and groundwater recharge.

From my understanding, once GWFLOW is enabled, the percolation from the soil profile should largely become recharge to the aquifer (subject to any specified losses or delays). However, the reported recharge is considerably smaller than the percolation values, and I am unsure whether this is expected behavior or indicates an issue with my model setup.

I would greatly appreciate your guidance on the following questions:

1. Should the annual groundwater recharge (rech) in GWFLOW approximately match the annual percolation from the SWAT+ soil profile?
2. What processes or model components can cause recharge to be substantially lower than percolation?
3. Which input files or parameters should I examine to identify the source of this discrepancy?
4. Is there a recommended workflow to verify that recharge is being correctly transferred from the SWAT+ HRUs to the GWFLOW grid cells?
5. Is it possible to reduce this mismatch through model calibration? If so, which parameters are typically the most influential for improving the agreement between soil percolation and groundwater recharge?

Any suggestions or insights from those who have worked with SWAT+ GWFLOW would be greatly appreciated.

Thank you for your time and assistance.