Experiments

Experimental protocol for the LUCID-CMIP5 simulations

Every participating group performs several additional CMIP5 experiments with the following preferences:

  1. Simulation with RCP8.5 scenario is compulsory (core), RCP2.6 is voluntarily (optional)
  2. It is up to each group whether to perform ensemble runs (preferable) or a single simulation, depending on available computation resources
  3. Simulations could be done in emissions-driven mode (preferable, option L1) or concentration-driven mode (L2), see Table 1.

Option L1.  

  • Core simulations. Two simulations for ESMs capable to perform emission-driven scenarios allow a proper separation of biogeochemical and biogeophysical effects of landuse for the RCP8.5 scenario.
    1. The L1A85 simulation is the model run without landuse change (after yr 2005) and with atmospheric CO2 concentration prescribed from the CMIP run 5.3. This would provide biogeophysical effect of the landuse change by a difference in climate between the L1A85 and 5.3 runs.
    2. The L1B85 simulation is the model run without landuse change (after yr 2005) and with emission-driven atmospheric CO2 concentration from the RCP8.5 scenario. This would provide biogeochemical effect of landuse as a difference in atmospheric CO2 concentration between the 5.3 and L1B85 runs.
  • Optional simulations. Three simulations for ESMs capable to perform emission-driven scenarios allow a proper separation of biogeochemical and biogeophysical effects of landuse for the RCP2.6 scenario.
    1. The L1A26 simulation is the model run without landuse change (after yr 2005) and with atmospheric CO2 concentration prescribed from the run L1C26 listed below. This would provide biogeophysical effect of the landuse change by a difference in climate between the L1A26 and L1C26 runs.
    2. The L1B26 simulation is the model run without landuse change (after yr 2005) and with emission-driven atmospheric CO2 concentration. This would provide biogeochemical effect of landuse as a difference in atmospheric CO2 between L1C26 and L1B26.
    3. The L1C26 simulation is the model run with landuse change (after yr 2005) and with emission-driven atmospheric CO2 concentration.

Option L2.

Two concentration-driven model simulations allow calculation of (i) biogeophysical effects of landuse and (ii) landuse-associated CO2 emissions for models which include an output from the land carbon cycle.

  • Core simulation. The L2A85 simulation is the model run without landuse change (after yr 2005) and with atmospheric CO2 concentration prescribed from the RCP8.5 scenario. This would provide biogeophysical effect of the landuse change by a difference in climate between the L2A85 and CMIP5 4.2 runs. The landuse emissions could be figured out as a difference in land-atmosphere CO2 fluxes between the L2A85 and CMIP5 4.2 runs.
  • Optional simulation. The L2A26 simulation is the model run without landuse change (after yr 2005) and with atmospheric CO2 concentration prescribed from the RCP2.6 scenario. This would provide biogeophysical effect of the landuse change by a difference in climate between the L2A26 and CMIP5 4.3 runs. The landuse emissions could be figured out as a difference in land-atmosphere CO2 fluxes between the L2A26 and CMIP5 4.3 runs.