Discharge and AET calibration suggestions

[Ignacio Fuentes]

Hi all,

I'm currently working on modelling a relatively large snowy catchment. To achieve this, I'm utilising discharge data from two points within the catchment (2 reaches) and MODIS AET data to calibrate actual evapotranspiration (AET) in two subbasins. Despite my efforts, I'm encountering challenges in obtaining satisfactory results for AET (the AET modelled signal seems lagged and too small).

In my calibration approach, I'm employing TLAPS and PLAPS parameters at the start, followed by subbasin snow parameters. After that, I'm fine-tuning other parameters. I've also opted to segregate the snowy subbasins from the rest during the calibration process. Additionally, I'm facing difficulties in accurately modelling the discharge behaviour during the dry years that have characterised the last period. I've attached the latest iteration of the calibration history for your reference. 

Have you encountered a similar difficulty before? Any advice you could provide to enhance the results would be greatly appreciated.

Regards

[Ignacio Fuentes]

Hi all,

I tried to progress in this issue and decided to plot PET to understand the predicted AET delay.

Basically, I'm assuming that if PET shows the same behaviour as the "observed AET" (MODIS AET), but predicted AET shows a delay then there is an issue with the vegetation (land cover chosen).

Effectively, PET follows the "observed AET" pattern from MODIS. Then, my conclusion is that the vegetation growth associated with the land cover (predominantly agricultural) is misrepresented (maybe because my study is in the South Hemisphere?).

Do you guys know what I could do to improve the results?

Regards

[erick...@gmail.com]

Try to calibrate LAI firt