Budget enclosure problem

[ANIKET GUPTA]

Hi ParFLOW group,

I am facing a problem with the budget enclosure in ParFLOW-CLM. I force ParFLOW-CLM with half hourly forcing and write PF output at an half-hourly time step and CLM at hourly time step because I do not know how to write CLM at half-hourly when keeping hours as implicit units in PF. I therefore compute the budget from PF outputs (pressure, from which I compute overland flow from KWE post processing codes, storage) and from CLM outputs using evap_tot and SWE. I reckon that I can not technically close my budget using hourly evap_tot and SWE. However, I am also worried about the overland flow, because I see a few time-splitting during the peak snowmelt period. Ultimately, I see the imbalance in my budget calculation with runoff and evapotranspiration components overshooting the input precipitation (~30 mm).

As I spin up the model for 15 years the subsurface storage change is ~33 mm. And I don't see much difference in the evapotranspiration calculation (cross-checked from CLM's 'evap_tot' and ParFLOW's 'evaptranssum'). I wonder if a time-splitting simulation permits to compare the calculated  runoff against the simulation without any time-splitting?