How to interpret the negative (actual) evapotranspiration values?

Question

I am exploring FLUXNET sites for actual evapotranspiration (ETP) data (I convert the energy balance closure-corrected latent heat to ETP).

I am interested in the half-hourly data. During the night we see negative values. I would like to know the reason or physical process behind this.

Can physically-based hydrological models handle this situation? (: estimating negative half-hourly ETPs)

There are some discussion in web about how to handle the negative values of potential ETP, but not easy to find a solid discussion about the actual ETP in this case.

Answer 1

Negative values for the ET, both potential and actual, usually refer to condensation. See also "ERA5 mean evaporation rates data - Why is that negative?"

Regarding the ability of the physically-based hydrological models to handle condensation, I'd say that it's difficult to give you a general answer: it will depend on the specific math behind the model, but my general guess would be that they do. Even a simple Bucket Model, would integrate condensation in the water balance equation. While a more sophisticate Earth System Model would probably process it in the soil moisture and/or canopy component, in order to better represent these dynamics in the water-energy balance of the specific system. See for instance Longo et al 2019