R packages

• MetricGraph: Statistical analysis of data on metric graphs, such as street or river networks. Available on CRAN. See the package homepage for more information. 
• rSPDE: Implements rational approximation of fractional SPDEs, available on CRAN. See the package homepage for more information.
• excursions: Implements Excursion sets and contour credible regions for latent Gaussian models, available on CRAN. See the development page for more information.
• ngme: Methods for estimation and prediction using non-Gaussian random fields and longitudinal models. 

Implementations for papers

• R package and code to replicate the results in the paper Efficient methods for Gaussian Markov random fields under sparse linear constraints can be found here.
• Matlab code implementing the methods described in the paper "Spatially adaptive covariance tapering" can be downloaded here.
• Matlab and C-code for estimation of spatially dependent vegetation trends, as described in the paper Fast estimation of spatially dependent temporal vegetation trends using Gaussian Markov random fields can be downloaded here.
• Matlab and R code for simulation and estimation of nested SPDE models, as described in the paper Spatial models generated by nested stochastic partial differential equations, with an application to global ozone mapping can be downloaded here.