Flux Inversion#

FluxProblem#

Constructor#

FluxProblem(obs, prior, forward_operator, ...)

Atmospheric flux inversion problem.

Attributes#

FluxProblem.prior

Prior state vector.

FluxProblem.obs

Observation vector.

FluxProblem.estimator

Return the fitted estimator (raises if not solved).

Methods#

FluxProblem.solve([estimator])

Solve the flux inversion problem using specified estimator.

FluxProblem.get_block(component, block[, ...])

Get block from a component (Vector or Matrix).

FluxInversionPipeline#

Constructor#

FluxInversionPipeline(config)

Abstract pipeline for atmospheric flux inversions.

Attributes#

FluxInversionPipeline.problem

FluxInversionPipeline.estimator

The estimator to use for solving the inverse problem, either as a class or a string identifier.

Methods#

FluxInversionPipeline.filter_state_space(...)

Filter state space by removing intervals with insufficient observations.

FluxInversionPipeline.run([estimator_kwargs])

Run the flux inversion pipeline.

FluxInversionPipeline.summarize()

Print a comprehensive statistical summary of the inversion results.

FluxPlotter#

Constructor#

FluxPlotter(inversion)

Plotting interface for FluxInversion results.

Attributes#

FluxPlotter.fluxes([time, truth, x_dim, ...])

Plot prior and posterior flux maps.

FluxPlotter.concentrations([location, ...])

Plot observed, prior, and posterior concentrations.

Transport Backends#

Warning

The STILT transport backend is experimental and not production-ready.

This module depends on the stilt package, which is:

  • Not available on PyPI

  • Only available on GitHub at jmineau/stilt

  • Subject to rapid changes and breaking updates

  • Not guaranteed to be stable or backwards-compatible

Use this module at your own risk. The API may change without notice.

The STILT transport backend provides the JacobianBuilder class for constructing forward operators from STILT atmospheric transport simulations. See the source code for implementation details.