Species-habitat associations are present if species are specialized to small-scale environmental conditions (Tilman & Pacala, 1993) and are more common within suitable habitats (Comita et al., 2007; Harms et al., 2001). Following, species-habitat associations show the importance of abiotic processes on the spatial patterning of plant populations (Garzon-Lopez et al., 2014).
There are mainly two methods that can be found in the literature to analyse such small-scale species-habitat associations, namely the gamma-test (Plotkin et al., 2000) and the torus-translation test (Harms et al., 2001). Both methods require data on the locations of plants (as spatstat::ppp point pattern) and on the environmental conditions (classified into discrete habitats as raster::raster). To show significance species-habitat associations, both methods randomize one of the components. Because the locations of plants as well as the habitats are most likely auto-correlated, the spatial structure must be kept while randomizing the data (Wiegand & Moloney, 2014).
All methods compare the abundance within a habitat between the observed data and the randomized null model data. If the count is below or above a pre-set threshold (e.g. the 2.5th and 97.5th quantiles), negative or positive associations, respectively, are present.
The following two methods randomize the environmental data, i.e. the RasterLayer data, while keeping the point pattern data fixed.
The torus-translation test (Harms et al. 2001) shifts the habitat map in all four cardinal directions around a torus. This is only possible for square rasters. To use this method in shar use translate_raster().
The randomze-habitats procedure (Harms et al. 2001) is also possible for non-square raster and randomizes the habitats using a random-walk algorithm. To use this method in shar use randomize_raster().
Contrastingly to the two methods described above, the following two methods randomize the point pattern data, while keeping the environmental data fixed.
The gamma-test (Plotkin et al. 2000) randomizes the data by fitting a point process model to the observed data and simulation n random point patterns using the fitted point process model. However, the method only works for point patterns that can be described by a theoretical point process model. To use this method in shar use fit_point_process().
Pattern reconstruction (Tscheschel & Stoyan 2006) randomizes the point pattern using simulated annealing (Kirkpatrick et al. 1983). This allows to randomize also complex point patterns without a theoretical point process model. To use this method in shar use reconstruct_pattern().
Comita, L. S., Condit, R., & Hubbell, S. P. (2007). Developmental changes in habitat associations of tropical trees. Journal of Ecology, 95(3), 482–492.
Garzon-Lopez, C. X., Jansen, P. A., Bohlman, S. A., Ordonez, A., & Olff, H. (2014). Effects of sampling scale on patterns of habitat association in tropical trees. Journal of Vegetation Science, 25(2), 349–362.
Harms, K. E., Condit, R., Hubbell, S. P., & Foster, R. B. (2001). Habitat associations of trees and shrubs in a 50-ha neotropical forest plot. Journal of Ecology, 89(6), 947–959.
Kirkpatrick, S., Gelatt, C. D. J., & Vecchi, M. P. (1983). Optimization by simulated annealing. Science, 220(4598), 671–680.
Plotkin, J. B., Potts, M. D., Leslie, N., Manokaran, N., LaFrankie, J. V., & Ashton, P. S. (2000). Species-area curves, spatial aggregation, and habitat specialization in tropical forests. Journal of Theoretical Biology, 207(1), 81–99.
Tilman, D., & Pacala, S. W. (1993). The maintenance of species richness in plant communities. In R. E. Ricklefs & D. Schluter (Eds.), Species Diversity in Ecological Communities (pp. 13–25). Chicago: University of Chicago Press.
Tscheschel, A., & Stoyan, D. (2006). Statistical reconstruction of random point patterns. Computational Statistics and Data Analysis, 51(2), 859–871.
Wiegand, T., & Moloney, K. A. (2014). Handbook of spatial point-pattern analysis in ecology. Boca Raton: Chapman and Hall/CRC Press.