Question about MLDAS SLA and Vcmax DA script

[cenlinhe]

@CharlesZheZhang
The following ensemble run for SLA and Vcmax parameter optimization using LAI and SIF does not seem right to me, since it does not use ensemble parameter values to drive each Noah-MP ensemble simulation. This does not treat SLA or Vcmax as state variable in DA:
https://github.com/NCAR/hrldas/blob/master/hrldas/run/examples/single_MLDAS/MP-Crop-MLDAS-draft-v1.0.py#L158-L172

Maybe I missed something.

[cenlinhe]

I was wrong. The strategy is correct. It is just the model ensemble does not account for the parameter uncertainty, which should still be OK.