Multiplying by 2.5 in channel alpha?

[jensdebruijn]

channelAlpha is multiplied by 2.5 here, but it is not clear (to me ;-)), why this is done.

CWatM/cwatm/hydrological_modules/routing_reservoirs/routing_kinematic.py

Line 276 in 31ed493

self.var.channelAlpha = alpTermChan * (self.var.chanWettedPerimeterAlpha ** self.var.alpPower) *2.5

It seems different from the CWatM paper (eq.21) : https://gmd.copernicus.org/articles/13/3267/2020/

And also different from the PCR-GLOBWB implementation

https://github.com/UU-Hydro/PCR-GLOBWB_model/blob/2e3a5d85b0a1c264c7cb22dbc864bce87d115053/model/routing.py#L1243

[dof1985]

Interesting, I have looked back, and it seems to be there since at least 5 years

https://github.com/iiasa/CWatM/blob/935286bdd6f7919ac98a90f4c098d0f43ba68a10/cwatm/hydrological_modules/routing_reservoirs/routing_kinematic.py

[jensdebruijn]

7 even! https://github.com/iiasa/CWatM/blame/31ed493131202c16483a02a7192411fc78d76342/cwatm/hydrological_modules/routing_reservoirs/routing_kinematic.py#L276

[PeterBurek]

Jens I guess you found a bug. The 2.5 does not make sense at all. I think it was in to test the sensitivity and never taken out.

We have no the problem that the change in the source code would require a change in all our calibrations or at least putting a
factor in for all calibration for the calibration parameter manningsN

self.var.chanMan = loadmap('chanMan') * manningsFactor
alpTermChan = (self.var.chanMan / (np.sqrt(self.var.chanGrad))) ** self.var.beta
self.var.channelAlpha = alpTermChan * (self.var.chanWettedPerimeterAlpha ** self.var.alpPower) *2.5

The calibration parameter manningsN compensates somehow for the 2.5 factor.
I guess we will remove the 2.5 in the following releases. But till then we leave it in.