Ordination and clustering methods are widely applied to ecological data that are non-negative (like species abundances or biomasses). These methods rely on a measure of multivariate proximity that quantifies differences between the sampling units (e.g. individuals, stations, time points), leading to results such as:
Ordinations of the units, where interpoint distances optimally display the measured differences
Clustering the units into homogeneous clusters
Assessing differences between pre-specified groups of units (e.g. regions, periods, treatment–control groups)
In this video, Michael Greenacre introduces his new article, ‘‘Size’ and ‘Shape’ in the Measurement of Multivariate Proximity’, published in Methods in Ecology and Evolution, May 2017. In the context of species abundances, for example, he explains how much a chosen proximity measure captures the difference in “size” between two samples, i.e. difference in overall abundances, and differences in “shape”, i.e. differences in compositions or relative abundances. He shows that the popular Bray-Curtis dissimilarity inevitably includes a part of the “size” difference in its measurement of multivariate proximity.
Francesco de Bello describes the main elements of the method he has recently published in Methods in Ecology and Evolution. The method aims at decoupling and combining functional trait and phylogenetic dissimilarities between organisms. This allows for a more effective combination of non-overlapping information between phylogeny and functional traits. Decoupling trait and phylogenetic information can also uncover otherwise hidden signals underlying species coexistence and turnover, by revealing the importance of functional differentiation between phylogenetically related species.
In the video Francesco visually represents what the authors think their tool is doing with the data so you can see its potential. This method can provide an avenue for connecting macro-evolutionary and local factors affecting coexistence and for understanding how complex species differences affect multiple ecosystem functions.
“‘Why is this plant growing here?’ Tackling this question has led me through wetlands, forests, deserts and grasslands. I’ve poked at this question from the scale of plant traits all the way up to satellite imagery. I employ tools that include multivariate analysis, community and landscape diversity metrics, simulation modelling, and spatial classification. My current focus is on agricultural decision support tools for pasture and rangeland.”
Sarah will be handling Applications articles for the journal. Applications papers describe new software, equipment or other practical tools, with the intention of promoting and maximising the uptake of these new approaches. All of our published Applications articles are freely available to everyone. Continue reading →
Matt is an Associate Editor for Methods in Ecology and Evolution. He was the principle organiser of this year’s SEEM conference. His research interests include Bayesian inference and hierarchical modelling, computational methodology, ecological statistics and much more. Matt is based at the University of Otago.
A photo taken during a lunch break at the conference
The Statistics in Ecology and Environmental Monitoring (SEEM) conference was held in Queenstown, New Zealand on June 22-26, 2015. Queenstown is a resort town in the Southern Alps of New Zealand that looks out on Lake Wakatipu, surrounded by snow-capped mountains. The venue gave a chance to explore some of the natural beauty of New Zealand, with excursions to local ski fields, wineries and various hiking trails.
SEEM conferences have been organized by members of the Statistics group at the University of Otago since 1993. The first SEEM conference was held in Dunedin, New Zealand and conferences were then held regularly (every 3 years) until 2002. The last SEEM conference, in 2007, also served as the EURING (European Union for Bird Ringing) technical meeting. With nearly ten years passing since 2007, we had a smaller conference of around 50 attendees this year. There was an engaging atmosphere during the meeting and productive discussion followed each of the 40 talks. The SEEM 2015 meeting maintained the tradition of previous SEEM conferences with delegates from across a broad spectrum of statistical ecology coming together to discuss research. Continue reading →
McGill et al. (2006) argued that community ecology had lost its way. Shipley (2010) accused community ecologists of acting like a bunch of demented accountants. Strong words – so what’s the issue exactly? And what can we do about it?
Their beef was that when studying groups of species and their environmental association, ecologists often were not thinking enough about the reasons for variation across species. (In this post we’ll focus on variation in abundance or in environmental response of abundance across species. We’re interpreting “abundance” loosely – counts, biomass, 1-0, whatever.) While alternative methods are more readily available nowadays, “accountancy” is still common.
Warton and colleagues’ article has recently been highlighted on Faculty of 1000, a platform providing post-publication peer-review and selecting only the most important articles in biology and medicine. Just 2% of published articles are highlighted on Faculty of 1000 each month.
We strongly recommend reading this article because, beyond technical issues, it stimulates reflections on our consciousness of limits of statistical tools, which is often overwhelmed by our addiction to their routine application.
The article is also accompanied by a free application and a really great video:
In case you haven’t seen them, this month we have published a new podcast and video so far.
In our latest video, David Warton, The University of New South Wales, Australia, presents his ‘mvabund’ package on multivariate analysis. What makes this software different from other ones on multivariate analysis, is that it’s all about models that you can fit to your data. David explains how to look at the properties of your data and the common pitfalls in modelling multivariate data. He also goes through how to fit generalised linear models to your data. Do check David’s dancing!