Model Processing

class pyfuse.pyFUSE(name, OPTIONS, PARS, RAIN, EVAPO, CONST, INITFRAC=0.1, oldfile=False)

Create pyFUSE hydrological model, a python version and extention of the FORTRAN FUSE model environment by Clarke, 2008 [1] This is not a wrapper of the Fortran implementation of Clark, but a complete rewrite of the original code in order to make further extensions easier.

Parameters:

name: str

A given name for the constructed model structure

OPTIONS: dict

Dictionary with the model options for construction

PARS: dict

Dictionary of parameters used for the model evaluation, coming from input parameter file

RAIN: array

numpy array containting the rainfall data

EVAPO: array

numpy array containting the potential evapotranspiration data, with same time frame and resolution as the rain

CONST: dict

dictionary of neede constant values for the model configuration and calculation

INITFRAC: float

fraction of maximum storage used as initial condition

References

[1] Clark, Martyn P., A. G. Slater, D. E. Rupp, R. A. Woods, Jasper A. Vrugt, H. V. Gupta, Thorsten Wagener, and L. E. Hay. Framework for Understanding Structural Errors (FUSE): A modular framework to diagnose differences between hydrological models. Water Resources Research 44 (2008): 14.

array_output(outname, outtype='FLUX', run_id='testrun')

Get output array of the selected outname; fluxes in mm/hour, states in mm and pars in the given units

Parameters:

outtype: str

one of these values: ‘FLUX’, ‘STATE’ or ‘PAR’

outname: str

the specific flux/stae of par needed

run_id: str

name of the run_id to get the flow from

Returns:

output: array

output of the specified model output

clean_outputs(todelete)

Delete the none interesting output groups and the related datasets

Parameters:

todelete: list

list of strings with the groups to delete

close_h5()

close the hdf5 file to stop the model analysis

get_all_runids()

Get list of all used run_ids saved in the hdf5 file

get_const_info(run_id='testrun1')

Get overview of the constant values used in the run

get_model_info()

Get overview of the structure options working with

load_new_pars(inff=None)

Load a parameter set from a input parameter file or dict and put in parameter dictionary or load them from the pars given in dict

Currently only from file is supported, directly passing a dictionary should also be possible and will be implemented

Parameters:

inff: str

name of the input parameter file

run(custom_period=None, new_pars=False, run_id='testrun')

Run the model!

Parameters:

custom_period:

calculate the model for a specific subperiod of the total data lenght

new_pars: str

textfile of the new parameters used to (re)run the model or a dictionary

run_id: str

used to identify the outputs of the specific modelrun, if nothing given, a testrun-group is added

total_outflow(run_id='testrun')

Outflow of the catchment in m3/s when hourly timestep

Parameters:

run_id: str

name of the run_id to get the flow from

Returns:

totalout: array

array of the flow output

Monte Carlo Runs

run_MC(nruns)

When the model is constructed with the parameters according to the defined ranges, Monte Carlo simulations can be performed using the run_MC(number of runs) command

Re-Load an existing model structure

pyfuse.Load_model(hdffile, parfile)

Load an old model run and set up the model

Parameters:

hdffile: HDF5 file

previous simulation outputs file to re-use

parfile: parameter textfile

parameterfile of the model under consideration

Returns:

pyFUSE_Model: model instance

loaded model to do calculations

See also

pyFUSE.Set_pars

Loading in parameter sets example of input parameter file