l96_simulate: Simulate data from the lorenz96 model

In robertsy/lorenzr: Lorenz Models

Description

Simulate data from the lorenz96 model

Usage

l96_simulate(duration, freq, ndim = 40, Forcing = 8, sig = sqrt(0.5),
  obs_type = "odd", R_sig = sig, deltat = 1/1000, ...)

Arguments

duration	total time to integrate model (duration=freq*(nsteps-1) (nsteps = duration/freq + 1)
freq	frequency of time integration (freq=duration/(nsteps-1))
ndim	dimension of system
Forcing	external forcing on dynamics
sig	standard deviation of error measurments
obs_type	(all: all sites, odd: every other site, partial: regular observations at nobs sites, one: unique observation in the middle)
R_sig	standard deviation of error measurements for assimilation (default=sig)
deltat	time interval for model integration
...	additional arguments passed to lorenz_observe (e.g. nobs in case obs_type=partial)

Value

list with: state.ts matrix of nsteps x ndim y.ts list of nsteps lists, each of them containing y, R, H, d and y.loc (sites which are observed) f.propagate function to propagate the ensemble according to the model specification f.propagate <- function(state=initial state, nsteps=how many steps to propagate, ...) ndim, duration and freq to keep track of some parameters

Examples

l96_run <- l96_simulate(0.1*100, 0.1, ndim=40, Forcing=8, deltat=1/1000)
lorenz_plot(l96_run$state.ts[101,], l96_run$y.ts[[101]])

robertsy/lorenzr documentation built on May 27, 2019, 10:33 a.m.