Negative signs for metdata

[aniursa]

Some metadata—particularly from tower data contributors—include negative values for certain measurements. While these values may not make physical sense, they are commonly accepted by the eddy covariance (EC) community, with the understanding that users are expected to interpret the physical meaning of each variable appropriately.

Are we planning to carry this approach forward for PSInet? I’ve noticed negative values in data from some sites (e.g., Tonzi), and I wanted to flag this for discussion. We might add a note on this.

[jessicaguo]

Negative values are flagged but retained. It's in one of the vignette that users need to consider the flags carefully and not apply them as a blanket cut off. Do you have a good idea of where to document this?