SGSDist.jl Documentation

Overview

SGSDist.jl is a Julia package that implements the stochastically generated skewed (SGS) distribution, a probability density function (pdf) obtained from the correlated additive and multiplicative (CAM) noise stochastic climate model developed by Sura and Sardeshmukh (2008) and Sardeshmukh and Sura (2009).

For more information on the form of the SGS distribution used in this package, see Sardeshmukh et al. (2015).

Installation

Note

SGSDist.jl is not yet distributed through the Julia package manager, so one has to install the package locally.

SGSDist.jl may be installed using the Julia package manager. From the Julia REPL, type ] to enter the Pkg REPL mode and run:

add SGSDist

Local installation

If you want to install the package locally, first decide what directory will house the SGSDist.jl repository. There, clone the repository to your local machine:

git clone git@github.com:brwst/SGSDist.jl.git

Then, add the parent directory that contains the repository to the Julia LOAD_PATH by editing the startup.jl file located in ~/.julia/config:

push!(LOAD_PATH, "/path/to/dir/")

Reloading any Julia instances should make the SGSDist.jl module available for import.

References

SGSDist.jl is based largely on the form of the SGS distribution and its associated Markov process outlined in Sardeshmukh et al. (2015):

Sardeshmukh, P. D., G. P. Compo, and C. Penland, 2015: Need for Caution in Interpreting Extreme Weather Statistics. Journal of Climate, 28, 9166–9187, https://doi.org/10.1175/JCLI-D-15-0020.1.

For the theoretical development of the CAM noise stochastic climate model and the SGS distribution, see:

Sura, P., and P. D. Sardeshmukh, 2008: A Global View of Non-Gaussian SST Variability. Journal of Physical Oceanography, 38, 639–647, https://doi.org/10.1175/2007JPO3761.1.
Sardeshmukh, P. D., and P. Sura, 2009: Reconciling Non-Gaussian Climate Statistics with Linear Dynamics. Journal of Climate, 22, 1193–1207, https://doi.org/10.1175/2008JCLI2358.1.