Question on Lavenne regularization airGR implementation
Currently airGR applied the GAPX function using transformed parameters:
But the article also refers to "drift functions" that should be minimsed "according to GAPX criterion" (section 3.1 and Table 3, section 3.4):
... which suggests that should not be used for X_4
and C
.
This is not very clear for me. @guillaume.thirel, what's your opinion on it?
Lavenne, Alban de, Vazken Andréassian, Guillaume Thirel, Maria-Helena Ramos, et Charles Perrin. 2019. « A Regularization Approach to Improve the Sequential Calibration of a Semidistributed Hydrological Model ». Water Resources Research 55 (11): 8821‑39. https://doi.org/10.1029/2018WR024266.