logic to choose parameters - also a documentation space for things have tried

frank0434 / Master

repo for the master project - water stress for APSIMX lucerne model

Creative Commons Zero v1.0 Universal

1 stars 1 forks source link

logic to choose parameters - also a documentation space for things have tried #6

Closed frank0434 closed 3 years ago

frank0434 commented 3 years ago

sensitivity analysis to rank the parameters
only calibrate the high sensitivity ones
try report uncertainty.

frank0434 commented 3 years ago

https://github.com/frank0434/Master/projects/1#card-49286505

frank0434 commented 3 years ago

BD is the most influential parameter in year one for SD2 in layer 1 for 5 paths

LL domaniant the influence in year 2011

LL RFV DUL and BD started to interfere in the year 2012

Notes Annual aggregation does not make sense. Need to do a farm year report.

BD out weighted all others when looking at the whole experiment Cautions!!! variances are larger than mustar

frank0434 commented 3 years ago

Seasonal output can be aggregated after simulation complete.

BD is more influential in season one.

frank0434 commented 3 years ago

start to add things into plan

frank0434 commented 3 years ago

Had a chat with Rog about the stone content parameter in apsimx. Currently, the parameter is a place holder. Also, it might be not necessary to set it up since the measurements are done in the field, which already a full soil profile.

For bulk density values, one or two sdev up/down from the mean values should be good enough to define the ranges for bulk density. The key factor is those values must be physically sensible to the real soils.

frank0434 commented 3 years ago

Found the accumulative thermal time didn't restart in the second season.