Our new collaborative paper (with researchers from five different countries) summarizing methods to detect early warning signals of ecological transitions using spatial patterns has been published in Plos ONE this March. Click here to read the paper.
The paper also comes with an R package to use the methods. Its available from the github: https://github.com/earlywarningtoolbox/spatial_warnings
If you would like to use this toolbox and face any problems, please feel free to contact me:
Story behind the paper: Vasilis Dakos, Marten Scheffer and Stephen R Carpenter organized a fantastic workshop in Santa Fe Institute (New Mexico, USA) and invited a bunch of people who had worked on developing early warning signals. It was two and a half days of extremely interesting discussion and we decided to summarize the methods of early warning signals in two papers.
One paper was to focus exclusively on methods to detect early warning signals in time series data. Vasilis Dakos took the lead and we published a paper in Plos ONE that can be found here. Together with the paper, we also provided an R package so that researchers can use the methods without having to develop their own code. The methods and package can be obtained from the website: www.early-warning-signals.org
A second paper was to focus on the methods for analyzing spatial patterns to detect early warning signals. Sonia Kefi and myself took the lead, and we had to face the challenge of summarizing results for analyzing spatial patterns where the methods have not been not so well developed. Finally, the second paper has also been now published! We now need to make the codes more easily accessible to users.