Diagnostics

Diagnostics are separate objects (instances of subclasses of RogerDiagnostic) responsible for handling I/O, restart mechanics, and monitoring of the numerical solution. All available diagnostics are instantiated and added to a dictionary attribute RogerState.diagnostics (with a key determined by their name attribute). Options for diagnostics may be set during the RogerSetup.set_diagnostics() method:

Output aggregated to daily:

class MyModelSetup(RogerSetup):
    ...
    def set_diagnostics(self, state):
        diagnostics = state.diagnostics
        diagnostics['rate'].output_variables = ['transp', 'q_ss']
        diagnostics['rate'].sampling_frequency = 1
        diagnostics['rate'].output_frequency =  24 * 60 * 60

Output aggregated to hourly:

class MyModelSetup(RogerSetup):
    ...
    def set_diagnostics(self, state):
        diagnostics = state.diagnostics
        diagnostics['rate'].output_variables = ['transp', 'q_ss']
        diagnostics['rate'].sampling_frequency = 1
        diagnostics['rate'].output_frequency =  60 * 60

Output aggregated to 10 minutes:

class MyModelSetup(RogerSetup):
    ...
    def set_diagnostics(self, state):
        diagnostics = state.diagnostics
        diagnostics['rate'].output_variables = ['transp', 'q_ss']
        diagnostics['rate'].sampling_frequency = 1
        diagnostics['rate'].output_frequency =  10 * 60

Please not, that within the adaptive-time stepping greater time steps than provided output_frequency will not be downscaled (e.g. from daily to hourly).

Base class

This class implements some common logic for all diagnostics. This makes it easy to write your own diagnostics: Just derive from this class, and implement the virtual functions.

class roger.diagnostics.base.RogerDiagnostic(state)

Bases: object

Base class for diagnostics. Provides an interface and wrappers for common I/O.

Any diagnostic needs to implement the 9 interface methods and set some attributes.

name = None

Name that identifies the current diagnostic

abstract initialize(state)

Called at the end of setup. Use this to process user settings and handle setup.

abstract diagnose(state)

Called with frequency sampling_frequency.

abstract output(state)

Called with frequency output_frequency.

Available diagnostics

Currently, the following diagnostics are implemented and added to RogerState.diagnostics:

Snapshot

class roger.diagnostics.snapshot.Snapshot(state)

Bases: RogerDiagnostic

Writes snapshots of the current solution. Also reads and writes the main restart data required for restarting a Roger simulation.

output_path = '{identifier}.snapshot.nc'

File to write to. May contain format strings that are replaced with Roger attributes.

name = 'snapshot'

output_frequency = None

Frequency (in seconds) in which output is written.

Average

class roger.diagnostics.average.Average(state)

Bases: RogerDiagnostic

Average output diagnostic.

All registered variables are summed up when diagnose() is called, and averaged and output upon calling output().

name = 'average'

output_path = '{identifier}.average.nc'

File to write to. May contain format strings that are replaced with Roger attributes.

output_frequency = None

Frequency (in seconds) in which output is written.

sampling_frequency = None

Frequency (in seconds) in which variables are accumulated.

output_variables = None

Iterable containing all variables to be averaged. Changes have no effect after initialize has been called.

Minimum

class roger.diagnostics.minimum.Minimum(state)

Bases: RogerDiagnostic

Minimum output diagnostic.

All registered variables are stacked when diagnose() is called, and returns minimum and output upon calling output().

name = 'minimum'

output_path = '{identifier}.minimum.nc'

File to write to. May contain format strings that are replaced with Roger attributes.

output_frequency = None

Frequency (in seconds) in which output is written.

sampling_frequency = None

Frequency (in seconds) in which variables are accumulated.

output_variables = None

Iterable containing all variables to be minimumd. Changes have no effect after initialize has been called.

Maximum

class roger.diagnostics.maximum.Maximum(state)

Bases: RogerDiagnostic

Maximum output diagnostic.

All registered variables are stacked when diagnose() is called, and returns maximum and output upon calling output().

name = 'maximum'

output_path = '{identifier}.maximum.nc'

File to write to. May contain format strings that are replaced with Roger attributes.

output_frequency = None

Frequency (in seconds) in which output is written.

sampling_frequency = None

Frequency (in seconds) in which variables are accumulated.

output_variables = None

Iterable containing all variables to be maximumd. Changes have no effect after initialize has been called.

Rate

class roger.diagnostics.rate.Rate(state)

Bases: RogerDiagnostic

Rate output diagnostic.

All registered variables are summed up when diagnose() is called, and output upon calling output().

name = 'rate'

output_frequency = None

Frequency (in hours) in which output is written.

sampling_frequency = None

Frequency (in hours) in which variables are accumulated.

output_variables = None

Iterable containing all variables to be averaged. Changes have no effect after initialize has been called.

Collect

class roger.diagnostics.collect.Collect(state)

Bases: RogerDiagnostic

Collect output diagnostic.

All registered variables are collected when diagnose() is called, and collected upon calling output().

name = 'collect'

output_path = '{identifier}.collect.nc'

File to write to. May contain format strings that are replaced with Roger attributes.

output_frequency = None

Frequency (in seconds) in which output is written.

sampling_frequency = None

Frequency (in seconds) in which variables are accumulated.

output_variables = None

Iterable containing all variables to be volumed. Changes have no effect after initialize has been called.

Constant

class roger.diagnostics.constant.Constant(state)

Bases: RogerDiagnostic

Constant output diagnostic.

All registered variables are collected when diagnose() is called, and written upon calling output() at beginning of simulation.

name = 'constant'

output_path = '{identifier}.constant.nc'

File to write to. May contain format strings that are replaced with Roger attributes.

output_frequency = None

Frequency (in seconds) in which output is written.

sampling_frequency = None

Frequency (in seconds) in which variables are accumulated.

output_variables = None

Iterable containing all variables to be volumed. Changes have no effect after initialize has been called.

Water monitor

class roger.diagnostics.water_monitor.WaterMonitor(state)

Bases: RogerDiagnostic

Diagnostic monitoring of water storage.

Writes output to stdout (no binary output).

name = 'water_monitor'

Name that identifies the current diagnostic

output_frequency = None

Tracer monitor

class roger.diagnostics.tracer_monitor.TracerMonitor(state)

Bases: RogerDiagnostic

Diagnostic monitoring tracer contents / fluxes.

Writes output to stdout (no binary output).

name = 'tracer_monitor'

Name that identifies the current diagnostic

output_frequency = None