NEDAS.models.lorenz05 package

class NEDAS.models.lorenz05.Lorenz05Model(**kwargs)[source]

Bases: OdeModel

Lorenz (2005) Models II and III: Lorenz-96 with spatially smooth large-scale waves (Model II), plus superimposed small-scale activity coupled to them (Model III).

As in DART models/lorenz_04 (coded in 2004 by J. Hansen) and DAPPER mods/Lorenz05, whose vectorized formulation is used here. I=1 gives Model II, and also K=1 gives Lorenz-96.

Parameters:

  • nx (int) – state vector length

  • K (int) – width of the smoothing kernel in the advection term (larger = smoother, longer waves)

  • I (int) – half-width of the filter separating the large and small scales

  • space_time_scale (float) – scaling of the small-scale variability (b in the paper)

  • coupling (float) – coupling strength of the scales (c in the paper)

  • F (float) – forcing

cyclic: bool = True
nx: int
K: int
I: int
space_time_scale: float
coupling: float
F: float
property state_size: int

length of the state vector

x0() → ndarray[source]

a reference state, perturbed and spun up to start the truth and the ensemble

decompose(z: ndarray) → tuple[ndarray, ndarray][source]

the large-scale (x) and small-scale (y) parts of the state

dxdt(z: ndarray) → ndarray[source]

the time derivative of the state (for an ODE stepped by step)

Submodules