search

tilbud / rCMEM

GNU General Public License v3.0
2 stars 4 forks source link

Benchmarking against VBA model #7

Closed ktoddbrown closed 5 years ago

ktoddbrown commented 5 years ago

I ingested the output that @tilbud sent out, added it to a data folder, and started to scope out a unit test in this commit: 2d1492a What follows is just looking at the third model output.

Things aren't really working however and it's hard to tell if it's an issue with the VBA code or the R code. I could use some help.

Based on the output graphs I think there might be a bug in the VBA. 1) First (and easiest to fix I imagine) a mineral concentration of 13000 g/cm2 seems several orders of magnitude off and inconsistent with the given mineral loading parameters. I assumed there was a conversion factor issue and below is actually in g/m2 despite what the label says. 2) The total root biomass of the profile is ~500 g and both the R and VBA model agree on this. However the VBA model seems to think that would end up with over 400 g/m2 past the root zone where as the R model puts this at around 100 g/m2. If 20% of the input is refractory (based on the parameters) then you should end up with the number around 100 g/m2 if I understand things correctly.

image

Thoughts?

ktoddbrown commented 5 years ago

Done! 001a293ddbcfccca9755db4d0b7d8b2bb18b565f

Issue was with a bad mass to volume conversion primarily.