The sdmTMB package as a whole is distributed under GPL-3 (GNU General Public
License Version 3).

The sdmTMB package includes other open source software components. The following
is a list of these components.

- VAST, https://github.com/James-Thorson-NOAA/VAST
- flexsurv, https://github.com/chjackson/flexsurv
- glmmTMB, https://github.com/glmmTMB/glmmTMB
- SPDE barrier implementation, https://github.com/skaug/tmb-case-studies
- mgcv, https://CRAN.R-project.org/package=mgcv
- brms, https://CRAN.R-project.org/package=brms
- TMB anisotropy R implementation, https://github.com/kaskr/adcomp/tree/master/tmb_examples


The following are licensed under GPL-3 as included for sdmTMB:

1. VAST for the general structure of the original TMB model code
   (heavily modified into src/sdmTMB.cpp), anisotropy plotting code (modified
   into R/plot.R plot_anisotropy() and plot_anisotropy2()), approach to
   bias correction (R/index.R get_generic()), and approach to optimization
   (modified into R/extra-optimization.R)

2. flexsurv for the cumulative distribution function for the generalised
   gamma distribution (modified into R/residuals.R pgengamma())

The following are licensed under AGPL-3:

1. glmmTMB for general approach to prediction (heavily modified into
   R/predict.R) selected parameterizations of the likelihood densities
   (src/sdmTMB.cpp 'family' section: truncated distributions, nbinom1, Beta)
   effects methods (R/methods.R Effect.sdmTMB()), and emmeans methods (emmeans.R).

The following are licensed under GPL-2:

1. SPDE barrier implementation (src/utils.h spde_barrier_t and R/mesh.R
   make_barrier_spde())

2. Function to aid prediction from smooths from mgcv documentation
   smooth2random() (R/smoothers.R s2rPred())

3. brms code snippets used to construct penalized smoother matrices with mgcv
   tools (modified into R/smoothers.R parse_smoothers())

4. TMB anisotropy R implementation from TMB examples folder (R/mesh.R
   make_anisotropy_spde()) and corresponding code in src/sdmTMB.cpp
