Check soil carbon pools and soil turnover times

[luyssaert]

The soil carbon pool in r6910 is on low side. The same was observed in Tag 2.1 and ORCHIDEE 3.0. In ORCHIDEE 3.0 a new set of soil parameter was proposed to improve on this issue. Those parameter values were copied to the trunk (r6916). Comparing r6910 and r6930 (first time the new parameters were tested) shows that the soil carbon pools increased as intended. The turnover time increases as well. The turnover time in r6930 was considerably higher than in r6910 and was also higher than the (very uncertain) data we have to compare this variable against. The proposed parameter changes resulted in substantial decrease in LAI (almost by 50% over the Amazon for PFT2).

In short: we would like to have the turnover and LAI of r6910 but the soil pools of r6930. This is possible but will require some iterations because the changes in LAI imply changes in the litter input. As we are working towards milestone 4.1 (matching the performance of tag 2.1) we will revert the parameter values to r6910 (values were reverted in r6946).

Task: once milestone 4.1 has been reached. Heterotrophic respiration from litter could be lowered such that more carbon enters the soil (to enhance the stocks) without affecting the turnover too much. If this results in too strong N-limitations, CTE_BACT and FRACN_RUNOFF could be tuned.

Figure: ul = the code of r6930 but Tag 2.1 parameters for the soil = CTRL
ur = CTRL + som_turnover from ORCHIDEE 3.0
ll = CTRL + som_turnover + bact from ORCHIDEE 3.0
lr = CTRL + som_turnover + frcn_leaching from ORCHIDEE 3.0

Tag 2.1 parameters
SOM_TURN_ISLOW = 0.14
SOM_TURN_IPASSIVE = 0.0045
CTE_BACT = 0.00005
FRACN_RUNOFF = 0.3

ORCHIDEE 3.0 parameters
SOM_TURN_ISLOW = 0.1
SOM_TURN_IPASSIVE = 0.00225
CTE_BACT = 0.00015
FRACN_RUNOFF = 1.0