1.5.11. Running with anomaly forcing

Because performing fully coupled climate simulations is computationally expensive, an alternate method of running land-only simulations forced by future climate projections was developed for CTSM called ‘anomaly forcing’. The anomaly forcing method uses a previously completed fully coupled simulation to create monthly anomalies, relative to the present day, of near-surface atmospheric states and fluxes. These anomalies, representing the evolution of future climate projections, are applied to a repeating cycle of present day atmospheric forcing data, either as an additive (for states) or multiplicative (for fluxes) quantity. Thus, high-frequency variability is obtained from the present day atmospheric forcing data, while the long-term evolution of the climate is determined by the anomaly forcing dataset.

To enable anomaly forcing in a CTSM simulation, the following namelist variable can be added to the user_nl_datm file:

anomaly_forcing = ‘Anomaly.Forcing.Precip’,’Anomaly.Forcing.Temperature’,’Anomaly.Forcing.Pressure’,’Anomaly.Forcing.Humidity’,’Anomaly.Forcing.Uwind’,’Anomaly.Forcing.Vwind’,’Anomaly.Forcing.Shortwave’,’Anomaly.Forcing.Longwave’

Any combination or subset of forcing variables can be used, e.g. to modify only a single atmospheric forcing variable, one could use:

anomaly_forcing = ‘Anomaly.Forcing.Temperature’

which will only adjust the temperature (TBOT).

After the namelist has been created, the run directory will be populated with files such as these:

datm.streams.txt.Anomaly.Forcing.Temperature

which will contain the location of the default anomaly forcing datasets. To use alternative data, copy these files to the case directory with the ‘user_’ prefix, and modify the ‘user_*’ files accordingly, e.g.:

user_datm.streams.txt.Anomaly.Forcing.Temperature

For example, one could use the user_datm.streams.txt.Anomaly.Forcing.* files to point to these SSP-derived anomaly forcing datasets:

/glade/p/cesmdata/cseg/inputdata/atm/datm7/anomaly_forcing/CMIP6-SSP3-7.0

af.huss.cesm2.SSP3-7.0.2015-2100_c20200329.nc af.pr.cesm2.SSP3-7.0.2015-2100_c20200329.nc af.ps.cesm2.SSP3-7.0.2015-2100_c20200329.nc af.rlds.cesm2.SSP3-7.0.2015-2100_c20200329.nc af.rsds.cesm2.SSP3-7.0.2015-2100_c20200329.nc af.tas.cesm2.SSP3-7.0.2015-2100_c20200329.nc

Users may wish to also update files such as the landuse_timeseries and aerosol and Ndepostion files to correspond to the appropriate SSP.

For single point simulations, the global anomaly forcing files can be used, but the map_algo namelist variable should be appended with nearest neighbor values for each of the anomaly forcing fields, e.g.

mapalgo = ‘nn’,’nn’,’nn’,’nn’,’nn’,’nn’,’nn’,’nn’,’nn’,’nn’,’nn’,’nn’,’nn’ (the number of ‘nn’ values will depend on the number of original streams plus the number of anomaly forcing streams)

The cycling of the present-day (base) climate is controlled through the DATM_YR_START and DATM_YR_END variables in env_run.xml.