There are many variables that act together to determine how a run of JULES is set up and these are covered in detail in The JULES namelist files. Additionally, standard configurations and example runs illustrate suitable combinations of options. Here we highlight a few key switches that select broad areas of science, particularly for the benefit of new users.
The phenology model for natural vegetation can be enabled using l_phenol which uses the leaf turnover rate to calculate a time-varying Leaf Area Index (LAI).
To simulate carbon stocks in natural vegetation, the TRIFFID dynamic vegetation model can be enabled via the switch l_triffid. When TRIFFID is on, competition between tiles is switched on with l_veg_compete and the effect of nitrogen on vegetation growth is enabled via l_nitrogen.
The crop model, which is simulates phenology and carbon stocks in crops, can be switched on by setting the number of crop tiles ncpft to a non-zero value.
The crop model and TRIFFID cannot currently be used together. To simulate agricultural areas within TRIFFID, a fraction of the gridbox can be reserved for agricultural PFTS (as defined by crop_io > 0). Agricultural PFT competition and a representation of harvest carbon can be switched on with l_trif_crop.
If neither the phenology model nor the crop model are used, LAI for each vegetation tile can be set to a constant (lai_io) or a time series or seasonal cycle can be prescribed (JULES_PRESCRIBED).
To simulate carbon and nitrogen stocks in the soil, a model based on RothC should be selected by setting soil_bgc_model = 2. This option adds prognostic soil pools and must be used with the TRIFFID vegetation model. If TRIFFID is not used, prescribed soil pools must be invoked via soil_bgc_model = 1. Layered soil pools are used if l_layeredc = .TRUE..
A multi-layer snow model can be selected using nsmax. Parameterisations of surface and subsurface runoff generation are controlled using l_top and l_pdm, while the routing of water in rivers uses l_rivers.