Skip to content

Documenting calibration - #494

Merged
theavuik merged 1 commit into
masterfrom
documentation_calibration
Jul 10, 2026
Merged

theavuik merged 1 commit into
masterfrom
documentation_calibration

Conversation

@theavuik

Copy link
Copy Markdown
Contributor

No description provided.

@theavuik
theavuik requested a review from nilsvanvelzen July 10, 2026 11:48

@nilsvanvelzen nilsvanvelzen left a comment

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I do approve but look at some minor things

Comment thread docs/source/calibration.rst Outdated

Calibration starts with the definition of a scalar cost function that measures
the misfit between model predictions and the available observations over the
assimilation interval. The cost function is typically defined as the sum of

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

weighted sum of squares

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Replaced by weigthed sum of squared differences.

Comment thread docs/source/calibration.rst Outdated
optimal initial condition is the one that minimizes :math:`J`), :math:`x(k)` is
the variable at time :math:`t_k`, :math:`H` is the observation operator,
:math:`y^o(k)` is the observation at time :math:`t_k`, and :math:`R` is the
representation covariance. For the second cost function, we additionally have

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Representation covariance? Not covariance of measurement errors?

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Replaced.


* :math:`J(x_o) = \sum_{k=1}^N (y^o(k)-Hx(k))R^{-1}(y^o(k)-Hx(k))^\top`; and

* :math:`J(x_o) = (x_o-x^b)^\top(P^b)^{-1}(x_o-x^b)+\sum_{k=1}^N (y^o(k)-Hx(k))R^{-1}(y^o(k)-Hx(k))^\top`.

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I suppose this is the weak constraint variant? Probably good idea to call it this way in th text.

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Added

Comment thread docs/source/calibration.rst Outdated
:math:`y^o(k)` is the observation at time :math:`t_k`, and :math:`R` is the
representation covariance. For the second cost function, we additionally have
:math:`x^b`, the background or initial estimate of :math:`x_o`, and
:math:`P^b`, its covariance. This additional component ensures that the

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

mention that P^b is a measure for how sure you are of your initial parameter values

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Added.

@theavuik
theavuik force-pushed the documentation_calibration branch from 0e02aa3 to 51098d2 Compare July 10, 2026 12:36
@theavuik
theavuik merged commit c8d8558 into master Jul 10, 2026
1 of 2 checks passed
@theavuik
theavuik deleted the documentation_calibration branch July 10, 2026 12:36
@sonarqubecloud

Copy link
Copy Markdown

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants