Soil moisture loss after changing soil texture

[yan Chen]

I am running a NEON site simulation and noticed a behavior that I do not fully understand.

In a transient simulation, after modifying the soil texture to make the soil more sandy, I found that the soil moisture in the first year decreased quite noticeably. To trace where the water went, I checked several output flux variables, including ET, QOVER, QDRAI, QDRAI_PERCH, and QFLX_SNOW_DRAIN, but I did not see any obvious increase in these fluxes that could explain the magnitude of the soil moisture decline.

I am therefore wondering where this “missing” water may have gone. Are there other important variables or water balance terms that I may have overlooked and should examine?

 

[slevis]

@yan Chen I agree:
If the column's total water volume per unit area TOTSOILICE + TOTSOILLIQ (kg/m2) declines, I expect an increase in one or more of the fluxes out of the soil. Is it possible that the transition in soil water occurs quickly (say in a few timesteps)? Consider looking for the signal in timestep-level output for the first day or so after implementing the change. In other words try zooming in to see the signal in case this helps you understand the model behavior.