5.4. Resource forcing
5.4.1. Osmose <= 4.0.0
In Osmose versions <= 4.0.0, there were several ways to read the resource forcing files, depending on the value of the ltl.java.classname parameter.
5.4.1.1. ECO3M
If the class name is set equal to fr.ird.osmose.ltl.LTLForcingECO3M or
fr.ird.osmose.ltl.LTLFastForcingECO3M, data values were read as follows.
ltl.nstep |
Number of total time steps in the LTL files |
plankton.conversion2tons.plk# |
Conversion factor in tons |
ltl.netcdf.var.plankton.plk# |
Names of the Netcdf resource names |
ltl.netcdf.file.t# |
Names of the NetCDF files to read. |
ltl.netcdf.var.zlevel |
Name of the 3D depth variable |
grid.stride# |
Number of input cells that will be aggregated in an Osmose cell. |
ltl.integration.depth |
Maximum depth where to perform vertical integration |
5.4.1.2. BFM
If the class name is set equal to fr.ird.osmose.ltl.LTLForcingBFM or
fr.ird.osmose.ltl.LTLFastForcingBFM, data values were read as follows.
ltl.nstep |
Number of time steps within a file. |
plankton.conversion2tons.plk# |
Conversion factor in tons |
ltl.netcdf.var.plankton.plk# |
Names of the Netcdf resource names |
ltl.netcdf.file.t# |
Names of the NetCDF files to read. |
ltl.netcdf.dim.ntime |
Number of time steps within a file |
grid.netcdf.file |
Name of the bathymetry file |
ltl.netcdf.var.zlevel |
Name of the 1D depth variable |
ltl.netcdf.var.bathy |
Name of the bathymetry variable |
ltl.integration.depth |
Maximum depth where to perform vertical integration |
5.4.1.3. ROMS/PISCES
If the class name is set equal to fr.ird.osmose.ltl.LTLForcingRomsPisces or
fr.ird.osmose.ltl.LTLFastForcingRomsPisces, data values were read as follows.
ltl.nstep |
Number of time steps within a file. |
plankton.conversion2tons.plk# |
Conversion factor in tons |
ltl.netcdf.var.plankton.plk# |
Names of the Netcdf resource names |
ltl.netcdf.file.t# |
Names of the NetCDF files to read. |
ltl.netcdf.grid.file |
Name of the grid file |
ltl.netcdf.var.lon |
Name of the longitude variable |
ltl.netcdf.var.lat |
Name of the latitude variable |
ltl.netcdf.var.bathy |
Name of the bathymetry variable |
ltl.netcdf.var.csr |
Name of the CSR variable |
ltl.netcdf.var.hc |
Name of the Hc variable |
ltl.integration.depth |
Maximum depth where to perform vertical integration |
5.4.1.4. Fast forcing
In the above configurations, Osmose processes the file to convert them into the right format. However, there is also the possibility to use a file already in the right format using
the fr.ird.osmose.ltl.LTLFastForcing class.
ltl.nstep |
Number of time steps within a file. |
ltl.netcdf.file |
Name of the resource NetCDF file |
plankton.biomass.total.plk# |
Total biomass within the domain. If not found, reads value from NetCDF |
plankton.multiplier.plk# |
Multiplier of input biomass (default 1, used to run sensitivity experiments). |
The netcdf file must contain a 4d variable ltl_biomass, whose dimensions are (time, ltl, lat, lon)
Danger
The order along the ltl dimension must be consistent with the order of definition of
the plankton parameters.
Danger
With the fast forcing, the plankton.conversion2tons.plk# parameter is not used since data must be provided
in tons!
5.4.2. Osmose 4.1.0 - 4.2.0
In versions 4.1.0 and 4.2.0, only the FastForcing method has been kept. However, it has been slightly improved. Now
the model can take one NetCDF file per resource variable, and the variable in the NetCDF must match the plankton.name.plk#
parameter. In this way, there is no more dependency on the order.
plankton.biomass.total.plk# |
Total biomass within the domain. If not found, reads value from NetCDF |
ltl.nstep |
Number of time steps within a file. |
ltl.netcdf.file.plk# |
Name of the resource NetCDF file (one for each resource) |
plankton.multiplier.plk# |
Multiplier of input biomass (default 1, used to run sensitivity experiments). |
Danger
Resource species for which plankton.biomass.total.plk# is used should be defined last, i.e.
their index # should be greater than those of the other species.
Danger
The resource files must be provided in tons! **The plankton.conversion2tons.plk# parameter is not used.
5.4.3. Osmose >= 4.3.0
In the 4.3.0 version, the resource forcing remains similar to versions 4.1-4.2, with some changes in parameters.
species.biomass.total.sp# |
Total biomass for the given ressource (will be distributed over the whole domain) |
species.file.sp# |
Regular expression defining the input files. Can be a file name. |
species.file.caching.sp# |
Resource caching method. Must be |
If the species.biomass.total.sp# is not found, then the value will be read from the NetCDF file.
The species.file.sp# parameter can now take as an input regular expressions, which will allow to loop over all the files. The regular expressions must be defined in Java mode
(see for instance XXX).
The species.file.caching.sp# defines how the input data will be read:
In
allmode, all the dataset is read at the first time step and stored into memory. Should be used for climatological forcings for instance.In
incrementalmode, each time a new time-step is read from file, it is stored in memory. Previous time-steps are kept in memory.In
nonemode, the data is read from file at each time-step. This mode is costly in term of runtime but light in memory since only one time-step is stored.
Note
In version 4.3.0, the resource forcing parameter also applies to background species.