Negative biomass allocation

[luyssaert]

Running FG4 WARNING 10a starts to appear after 53 years (when running over a smaller domain than the default) with r6779 (and an unregistered parameter file). This warning signals negative leaf, sapwood and/or root allocation and is typically caused when the biomass suddenly changed due to coppicing, thinning or land cover changes. If the warning is given no allocation will happen and the allocatable biomass will be moved back into the reserve pool until the next time step. If the problem continues allocation is basically stopped.

[luyssaert]

• The imbalance is so large (100s of g of C on trees of 10e4 to 10e5 g C) that the problem is not getting resolved before the end of the simulation in 2010. This can be seen by plotting IND and/or DIAMETER. The dynamics in both variables are as expected until 1953. In subsequent years IND and DIAMETER no longer change.
• The warning appeared for Pinus pinaster, Quercus ilex/suber, Residual Temperate (evergreens), Pinus sylvestris, ... Looks like all PFTs for which this problem happens are evergreen PFTs.
• A sudden drop in biomass can indeed be seen between 1952 and 1953. Some of the problematic PFTs were under coppice management, others were under thin and fell management (unlikely that it is a management specific issue in sapiens_foresty).
• The drop happens within a single age class (unlikely the problem is in age_class_distribution)
• There is no sudden drop in IND (unlikely the problem is caused by management)
• There is no increase in veget_max (unlikely the problem is caused by land cover change)

=> First challenge: search what is causing the sudden drop diameter (sapwood and heartwood biomass)?