Abstract type to build subtypes from

EmpiricalOrthogonalFunction(dataset; center=true, ddof=1)

Create an Empirical Orthogonal Function object. The EOF solution is computed at initialization time. Method calls are used to retrieve or update computed quantities.

Truncate the EOF data structure to trim unneccesay modes

Empirical orthogonal functions (EOFs) expressed as the correlation between the principal component time series (PCs) and the time series of the eof input dataset at each grid point.

Empirical orthogonal functions (EOFs) expressed as the covariance between the principal component time series (PCs) and the time series of the eof input dataset at each grid point.

Extract the eigenvalues (decreasing variances) associated with each EOF.

Extract the empirical orthogonal functions (EOFs)

The method of North et al. (1982) is used to compute the typical error for each eigenvalue. It is assumed that the number of times in the input data set is the same as the number of independent realizations. If this assumption is not valid then the result may be inappropriate.

Inplace orthorotation that will override the data structure of the input EOF object

Apply orthogonal rotation to EOF data. After rotation the original dataset will be projected on the rotated EOF to create new PCs. Additionally new EOFs and PCs are ordered in decreasing variance

Extract the principal component time series (PCs)

Project a field onto the EOFs. Given a data set, projects it onto the EOFs to generate a corresponding set of pseudo-PCs.

Reconstructed input data field based on a subset of EOFs.

Total variance associated with the field of anomalies (the sum of the eigenvalues).

Apply truncate inplace on EOF data structure

Fractional EOF mode variances. The fraction of the total variance explained by each EOF mode, values between 0 and 1 inclusive.