Feature description
Auto local creation of synthetic cultivar for APSIM where the right cultivar is not in APSIM
Suggested solution
I’ve been thinking about how we can better represent local cultivar performance in APSIM and then benchmark the results against LAI time series and yield statistics for the region.
I propose focusing on four key parameters that directly affect yield and canopy development:
[Grain].NumberFunction.GrainNumber.GrainsPerGramOfStem.FixedValue – controls yield potential. Higher values mean more grains per unit stem biomass.
Range: 0–100 (step of 10)
[Phenology].Phyllochron.BasePhyllochron.FixedValue – controls LAI peak timing. Smaller values mean leaves appear faster, canopy closes earlier.
Range: 0–100 (step of 10)
[Leaf].ExtinctionCoeff.VegetativePhase.FixedValue – influences canopy light interception, hence the magnitude of LAI.
Range: 0–1 (step of 1)
[Phenology].CAMP.FLNparams.PpLN – affects final leaf number and thereby both LAI peak value and timing.
Range: 0–1 (step of 1)
If we treat these ranges systematically, we end up with:
11 possibilities for GrainNumber
11 possibilities for Phyllochron
2 possibilities for ExtinctionCoeff
2 possibilities for PpLN
That makes 11 × 11 × 2 × 2 = 484 combinations in total.
The idea is to generate synthetic cultivars across this parameter space, run them in APSIM, and then identify the parameter sets that best match observed LAI trajectories and yield statistics.
Additional context
No response
Feature description
Auto local creation of synthetic cultivar for APSIM where the right cultivar is not in APSIM
Suggested solution
I’ve been thinking about how we can better represent local cultivar performance in APSIM and then benchmark the results against LAI time series and yield statistics for the region.
I propose focusing on four key parameters that directly affect yield and canopy development:
[Grain].NumberFunction.GrainNumber.GrainsPerGramOfStem.FixedValue – controls yield potential. Higher values mean more grains per unit stem biomass.
Range: 0–100 (step of 10)
[Phenology].Phyllochron.BasePhyllochron.FixedValue – controls LAI peak timing. Smaller values mean leaves appear faster, canopy closes earlier.
Range: 0–100 (step of 10)
[Leaf].ExtinctionCoeff.VegetativePhase.FixedValue – influences canopy light interception, hence the magnitude of LAI.
Range: 0–1 (step of 1)
[Phenology].CAMP.FLNparams.PpLN – affects final leaf number and thereby both LAI peak value and timing.
Range: 0–1 (step of 1)
If we treat these ranges systematically, we end up with:
11 possibilities for GrainNumber
11 possibilities for Phyllochron
2 possibilities for ExtinctionCoeff
2 possibilities for PpLN
That makes 11 × 11 × 2 × 2 = 484 combinations in total.
The idea is to generate synthetic cultivars across this parameter space, run them in APSIM, and then identify the parameter sets that best match observed LAI trajectories and yield statistics.
Additional context
No response