mariegosme / SSM.R

translation of SSM model by Thomas R. Sinclair and Afshin Soltani into R
GNU General Public License v3.0
1 stars 1 forks source link

Wheat LAI abnormally high (while reasonable for chickpea) (it should never exceed 10, and be around 5 on average) #19

Open achgls opened 3 years ago

achgls commented 3 years ago

After fixing LAI rising exponentially to 15 000 + and fixing wheat plant density back to 280 because I had too low of a LAI, I now have abnormally high LAI again for wheat, about 2 to 10 times too high, depending on simulation cases.

sLAI_still_too_high

Corresponding PLA (total leaf area by individual plant in cm-2) and MSNN (main stem node number for one individual plant):

PLA

PLA

MSNN

MSNN

Corresponding cumulated thermal units / °Cd:

sTU

NOTE

Values seem nominal for chickpea. Wheat phyllocron (leaf appeareance on mainstem / °Cd) = 118, currently (in crops.xlsx) Differents colors on plots are different locations / cases / simulations, that occur over the same timeperiod. They are: Meknes, Morocco 🟢 Turgutlu, Turkey 🔴 Sidi Kacem, Morocco 🔵 Mauguio, France 🟡 Bizerte, Tunisia 🟣 Relevant to note: overall, Turgutlu is the simulation where water is the most abundant, and thus the less effect of water-stress on the crop.

achgls commented 3 years ago

These high values could be linked to the fact that wheat phenological growth is halted at the beginning of the simulation and early stages where LAI does increase last way longer than they should. LAI value would go on increasing for longer than it normally would. Hence why it would reach these high values. Photoperiod function for wheat could be bugged, hence why the complete stop in phenological growth. Might need to look into it. Either the function is bugged, or the "ppsen" parameter in crops.xlsx is irrelevant.

Setting filter "photoperiod_on_BD" = FALSE (just to check...) gives more reasonable output: LAI_wheat_rainfed_10yrs LAI for rainfed wheat over the course of 10 years

Although, we do observe high fluctuations and extreme values. Changing water regime from RAINFED to AUTOMATIC or FIXED irrigation has little to NO effect on these high values:

LAI_wheat_irrigated_10yrs LAI for fixed irrigated wheat over the course of 10 years

It could have been water stress that created these extreme outlier values, hence why the change of water management.

achgls commented 3 years ago

max_LAI_10yrs LAI at its peak for each year and simulation case, highlighted values being too high

When cancelling the effect of photoperiod on phenological growth, and its effect on making the corp halt in its early stages, we observe reasonable values for the most part. There still remains a lot of outliers tho.

achgls commented 3 years ago

TURNING PP EFFECT ON FOR THERMAL UNITS

CTU_ppeffect_TRUEforTUandBD

Being set to false by default, when turning ON photoperiod effect for thermal unit, and setting it equal to its value for BD, we get more reasonable values of maximum LAI and LAI senescence seems to occur properly... However, we notice that LAI is close to 0 for the first ~120 days of the crop's life...

LAI_ppeffect_TRUEforTUandBD