View source: R/sl.ens.gcdist.rmse.R
sl.ens.gcdist.rmse | R Documentation |
This function can be used to calculate the root mean square of the great-circle distance errors of a trajectory ensemble, either compared to an observed trajectory or to each one of the ensemble members in a perfect-model approach.
sl.ens.gcdist.rmse(ensm.in, t.ind = 1:length(ensm.in[1,1,]), obs = NULL, mode = "pm", Rsphere = 1)
ensm.in |
a numeric array containing the trajectory positions (in lon-lat coordinates), with shape [N.mem, 2, N.time], with the number of members |
t.ind |
a numeric vector of maximal length N.time, containing the indices of the time steps that shall be considered. All time steps are used by default, i.e. |
obs |
a numeric array of shape [2,N.time] which contains the coordinates of an observed trajectory. Must be provided if |
mode |
a string, either 'pm' (perfect model) or 'obs' (observation), defining the type of RMSE that will be computed. See section 'Details' for more information. |
Rsphere |
a numerical scalar defining the radius of the sphere for which the great-circle distance of the positions is calculated. |
For mode='pm'
, the root mean square error is calculated using each ensemble member as the 'truth' once (see reference for further information). Otherwise, the observation is used to determine the position errors.
Returns a numeric vector of length length(t.ind)
with a time series of the RMS distance error.
Simon Reifenberg
Collins, M. (2003), Climate Predictability on Seasonal to Decadal Time Scales: The Initial Value Problem, Climate Dyamics, 19,
sl.gc.dist
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.