Supplementary code for analyses of Barner et al. 2018, "Fundamental contradictions among observational and experimental estimates of non-trophic species interactions"

Barner et al. (2018, Ecology) compares the estimation of pairwise species interactions from observational spatial co-occurrence data against species interactions estimated experimentally.

Here we provide code to:

1. Replicate the only co-occurrence method not previously implemented in R ('frequency distribution' or the COOC algorithm of Sfenthourakis et al. 2006 doi: 10.1111/j.1466-822X.2005.00192.x). All other pairwise co-occurrence analyses use methods implemented in existing R packages, detailed in the main text and Appendix S2.

   COOC_code_example.md illustrates an example implementation of this algorithm
   COOC_algorithm.R is the raw R code needed to execute the algorithm

2. Replicate the aggregation of spatially gridded community data, described in Appendix S1.

   spatial_aggregate.R aggregates gridded presence/absence data to increasingly large grain size, using functions written by Rich FitzJohn

3. Replicate the analysis that accounts for habitat filtering in species pairwise co-occurrences, described in Appendix S2.

   habitat_filter.md illustrates an example implementation of several environmental-correction methods (doi: 10.1111/ecog.00779)