Sunday, 1 November 2015

Melodie McGeoch & Cang Hui - Estimating biodiversity turnover

The reading group will meet this Friday (6th November) at 1pm, We are reading and discussing a recent paper on biodiversity turnover by Melodie McGeoch & Cang Hui 
Melodie is an ecologists working on  the dynamics of biological invasions and the response of communities to changing environmentsCang is a mathematical ecologist interested in proposing models and theories for explaining emerging patterns in ecology. They have been working together on quantifying and estimating biodiversity turnover. 
We will read a joint paper of theirs in the American Naturalist titled Zeta Diversity as a Concept and Metric That UnifiesIncidence-Based Biodiversity Patterns

1 comment:

  1. This paper drew some enthusiastic responses, as a new and nice way to think about biodiversity. I guess the test of any new methodology is do we learn any new biolgical insights that we wouldn't see from existing approaches? Maybe that is still an open question?

    The nice neat straight lines on Fig 2a-b drew some complements - zeta diversity seemed to scale quite nicely with number of sites, although there is quite strong dependence between points so there was always a limit to how messy it could get.

    Andres Baselaga has done a fair bit of work on multiple site dissimilarity metrics which seems related, but which we didn't see referenced, maybe we are missing some of the subtleties here but we thought that would be related?

    There is little consideration of uncertainty - once you have a metric, how do you estimate the uncertainty around it due to sample variation? How to compare two zeta metrics to see if they are different? etc Maybe there are some opportunities there. This is a bit like the ver Hoef paper on stream networks, in the sense that this looks like it could be "the start of something"...