This week I went to an interesting seminar at UCL on spatial statistics in ecology by the interdisciplinary statistician Janine B. Illian. The seminar focused on using Cox point processes for species distribution modelling. Point processes can add value to SDM, but previously it was difficult to fit these models – these models can now be fitted relatively quickly using INLA (integrated nested Laplace approximations) which can be implemented in R using the R-INLA package. There was a recent paper on point processes by Renner et al. in Methods in Ecology and Evolution which reviews point process models for SDMs and outlines methods for fitting them through a worked example.
This week the @biotweeps twitter has been taken over by Bob O’Hara – have a look at the timeline for some thoughts on the directions in which SDMs should be going, as well as some general insights into the day-to-day working life of a statistical ecologist.
Swallow et al. have developed Bayesian hierarchical methods which deal with data with a spike at zero – namely, where the observer did not encounter the species during the study period. They have used this method to extract the signal from the BTO Garden Bird Feeding Survey to examine the relationship between songbirds and their predators, finding that increases in sparrowhawks may have led to a decrease in house sparrows.
Jobs Senior scientist in functional biodiversity research at the University of Leipzig, in the field of plant science. Funded PhD at Leiden University on fitness maximising approaches for evaluating vegetation functioning.