Four more papers have gone online this month and we are close to being able to put together the first issue of the journal!
In the first new paper to be published Adrian Barnett and colleagues present a comparison of methods for selecting the correct variance structure for longitudinal data. This is likely to be of considerable interest as it is a paper about how to fit the most appropriate description of non-independence in ecological time series data. The approach is illustrated with an analysis of the effects of forest fragementation in a 15-year data set on bird species richness.
Can Hui et al. look at the problem of fitting and predicting allometric relationships. They introduce a new way to measure bias in prediction and show that this is effective in improving predictions.
A new model by Thomas Brey can be used to estimate rates of respiration by invertebrates. The aim of this approach is to allow rapid and easy estimation of estimates of respiration rates. To facilitate the use of this model, a simple to use Excel spreadsheet can be downloaded here to accompany the paper.
Finally for now, Claudio Signer and colleagues describe a new Telemetry system for estimating heart rate, body temperature and locomotor activity in ruminants. This system has been used on Alpine Ibex to provide continuous measurements for up to 2 years.
Title of articles are posted here as soon as they are accepted. Updates on journal developments and new papers are also published on Twitter and Facebook, do join us!
Four new papers have been published online this month. These cover a range of topics including ecological modelling, measuring diversity, detecting range shifts and physiological ecology.
In the first paper, Gideon Gal and William Anderson outline a new method for detecting regime shift in ecosystems. Regime shifts occur when the state of an ecosystem changes markedly and rapidly, usually with a dramatic shift in species composition. Such shifts can be difficult to identify, particularly if the system in question is very noisy. The new method borrows techniques from statistics and econometrics and has the advantage that it does not rely on any pre-determined threshold value. The technique is used to show a regime shift in the zooplankton assemblage of a lake ecosystem.
Jan Beck and Wolfgang Schwanghart look at the problem of estimating species diversity from inventories. They specifically deal with the issue of undersampling, that is when inventories are incomplete owing to lack of coverage. In their study they simulate data with known levels of undersampling, and ask which estimates of diversity give the least biased estimates of true diversity.
A modelling paper by Clive McMahon, Barry Brook, Neil Collier and Corey Bradshaw describes a spreadsheet-based tool for exploring the strategic management of invasive species. Designed for predicting how different culling strategies affect the densities of invasive ungulates, the tool is aimed at managers and those with little familiarity with theory and modelling. The approach is applied to the control of feral pigs, buffalo and horses in Kakadu National Park Australia – the general framework could easily be applied to any similar system.
Finally for this update, Elizabeth Freeman and colleagues describe a new enzyme immunoassay for monitoring progestagens in elephants. This assay is a step forward as it is relatively easy and cheap, does not require expensive equipment and can be performed in the field. Hormone monitoring is an important conservation tool, allowing for example the reproductive states of animals to be monitored.
This is an exciting, diverse and high quality set of papers – more are on there way, the list of the latest papers to be accepted can be found here.
p.s. The first paper published in the journal by Alain Zuur and colleagues was downloaded over 1100 times in the first month after publication!
A belated happy new year! Here is this month’s round-up of methods papers published in the last month. Do let me know if there are any papers that I have missed that could be featured.
In Systematic Biology Brian O’Meara presents new heuristics for joint species delimitation and tree inference. A new comparative method for logistic regression controlling for phylogeny is outlined by Ives & Garland, and Wertheim et al. publish an analysis of the use of relaxed clocks in phylogenetic inference.
Marc Cadotte and colleagues outline in the latest issue of Ecology Letters new metrics for measuring phylogenetic diversity in ecological communities.
In the Journal of Applied Ecology Len Thomas and colleagues present a review of distance sampling and its use in estimating population size; Marc Kéry and co-workers illustrate a method for estimating trends from replicated count data when detection is imperfect; Deanna Dawson and Murray Efford demonstrate a new method for estimating bird densities from acoustic data. William Kendall and Gary White have a cautionary note on substituting spatial subnits for temporally replicated sampling in estimating site occupancy.
A ‘how to’ paper in Journal of Animal Ecology by Alastair Wilson et al. presents a review and guide to using the ‘animal model’ in quantitative genetics.
In Global Ecology and Biogeography Cabral & Schurr have a paper illustrating a method for linking range dynamics and demographic models in the Fynbos. Andrés Baselga has a paper in the same issue illustrating a method for disentangling the contributions of spatial turnover and nestedness to beta diversity.
Tommaso Zillion and Fangliang He publish in Oikos a new method for linking species abundance distributions across scales.
In Journal of Ecology Damgaard & Fayolle present a new method for estimating the influence of competition in plants.
A mini-review by Gavin Stewart and colleagues in Conservation Letters reviews the design of temperate marine reserves from a analytic perspective. In the same issue Kyle Van Houtan et al. look at the effectiveness of translocations in conserving endangered species.
A round-up of methods papers published in the last month. If there are any papers that you think should be featured, email me or leave a comment and I will add them.
Liam Revell has a paper in Evolution on size correction and principal components analysis of phylogenetic comparative data. Olivier Gimenez and colleagues also have a paper in the same issue on generating fitness landscapes using mark-recapture data.
Systematic Biology has a number of papers with interesting methods: Campbell & Lapointe have a paper on the use and validity of composite taxa in phylogenetic analysis; Fitzjohn et al. have a nice paper on estimating trait-dependent speciation and extinction rates in phylogenies that are not complete; Bui Quang Minh and colleages present an algorithm for efficiently estimating phylogenetic diversity; Michael D. Pirie, Aelys M. Humphreys, Nigel P. Barker, and H. Peter Linder present an approach for dealing with implications of conflicting gene trees on inferences of evolutionary history above the species level.
In Conservation Biology, Angelia Vanderlaan and Christopher Tagaart describe how a voluntary scheme for ships to avoid cetain areas has worked in preventing lethal strikes on right whales.
In Ecological Applications, Cang Hui and colleagues compare approaches for extrapolating population sizes from abundance-occupancy relationships. Matthew Etterson et al. look at the problem of estimating population trends when there is detection heterogeneity and overdipsersion in the data. Paul Beier and co-workers use a case study to examine the use of least-cost modelling to design wildlife corridors.
Oscar Puebla and colleagues describe in Ecology a study that estimates dispersal using genetic distances in a coral reef fish. Sean Connolly et al. have a new bootstrap approach for testing species abudance models in the same issue. Andy Royle et al. present Bayesian method for estimating population sizes using camera trap data. David G. Angeler, Olga Viedma, and JoséM. Moreno present a critique of time lag analysis in time series modelling. David Carslake et al. have a paper presenting useful review of constraints and rules for elasticity analysis in matrix modelling. Finally in that same issue Paul Stapp and Daniel J. Salkeld look at the use of stable isotopes in studying host-parasite interactions.
Finally for this month in Animal Conservation, Heidy Kikillus et al. look at minimising false negatives in predicting distributions of invasive species. (Thanks to Andrew Tyre for pointing this one out).
Phylogenetic comparative methods are always an area of hot discussion and lots of methodological development. So I thought it would be useful to highlight some recent papers that have developed new methods in the past year. Please email me or leave a comment if there is anything I have omitted or if something new comes out.
Thomas Hansen and colleagues have introduced a new method for studying adaptation using comparative methods. Their approach is a generalisation of the Ornstein-Uhlenbeck model that allows for adaptive constraints and phylogenetic intertia. They have an R-package SLOUCH which can be used to fit the model.
In Evolution Liam Revell has developed a new approach for data reduction and size correction using phylogenetic approaches – this is often done wrongly as the transformation is commonly applied before phylogenetic analysis, however it should be correctly done at the same time.
A new method in Functional Ecology allows one to test for phylogenetic dependence in complex multivariate data that also incorporate measurement error.
What will prove, I think, to be a very popular method is a new approach for testing for phylogenetic signal and analysing correlates of binary traits, basically a phylogenetic logistic regression by Anthony Ives and Ted Garland. The approach will allow linear modelling of correlates of a binary traits, which has been difficult before.
In a related area, another very important development is in the analysis of speciation and extinction rates when these are affected by a binary trait. FitzJohn et al. have shown how the BiSSE model, developed to do this, can be applied when phylogenies are incompletely resolved.
Likely to be of interest to many using comparative methods is a paper by Richard Smith on the use and misuse of Reduced Major Axis line fitting. He discusses the assumptions of this method, which are not widely appreciated.
In the American Naturalist Marc Lajeunesse has developed methods linking comparative analysis and meta-analysis, basically allowing meta-analysis to be corrected for phylogenetic non-independence.
In Proc B a new method for integrating spatial and phylogenetic dependence has been presented, and in JEB there has been a review of the ‘deadly sins of comparative analysis’ (apologies for self-promotion!).
Just to end with here is another method for explaining adaptation.
The first paper accepted in Methods in Ecology & Evolution is now available online!
This paper is by Alain Zuur, Elena Ieno and Chris Elphick – Alain is well known in the ecological and R community for his books Analysing Ecological Data, Mixed Effects Models and Extensions in Ecology with R and A Beginner’s Guide to R and for the courses he runs on statistics. Elena Ieno is senior marine biologist at Highland Statistics Ltd. In 2004 she left academia to work full time in statistical consultancy. She now teaches statistics to ecologists and has shown she can bridge the gap between the two disciplines and dispel the dread of statistics shown by many biologists. Chris Elphick is a conservation biologist, based at the University of Connecticut, who primarily works on avian ecology in agricultural habitats and salt marshes.
In this paper Alain and colleagues draw on their experience of teaching and advising to highlight problems in the analysis of data and how they can be avoided by appropriate exploratory analysis. Specifically they look at:
- Detecting outliers
- Heterogeneity of variance
- Dependence of observations
- Interactions
- Zeros in multivariate data
- Zero inflation
- Identifying correct relationships between X and Y
They provide examples and advice in the paper. They also include the R code and datasets as a supplement, so readers can work through the examples themselves, which is a great help.
This paper should be of interest to anyone actively involved in the analysis of ecological data.
Here is a round-up of interesting methods and methods-related papers published in the past few weeks. Please do pass on any interesting-looking papers / links and I will include them.
In Heredity Blanya et al. review the use of Drosophila subobscura as a tool in research on the microevolutionary consequences of climate change.
There are several methods-relevant papers in the latest issue of Evolution: Céline Becquet and Molly Przeworski look at the problem of estimating the mode of speciation, specifically whether speciation occurs with gene flow; Calsbeek & Goodnight criticially examine methods for estimating G-test statistics; Craig White et al. revisit the issue of the scaling of BMR with body size in mammals, and show that the estimate of the scaling coefficient depends on the method.
A really nice paper by FitzJohn et al. has just appeared online in Systematic Biology that deals with the tricky problem of estimating trait-dependent speciation and extinction rates in the face of phylogenetic uncertainty. Another really interesting early online publication is the study by Kress et al. in PNAS who use barcoding methods for the first time to construct a phylogeny for a whole community.
A study by Guoke Chen and colleagues in Journal of Ecology presents an analysis of the factors that might influence detection probability in plant surveys.
In Journal of Applied Ecology Fieborg et al. present an overview of regression modelling of correlated data; Chetkiewicz & Boyce apply resource selection functions to identify conservation corridors; Lemke et al. revisit the use of fluorescent markers in studying seed dispersal. Satu Ramula and colleages present a comparison of integral projection models and matrix population models, and show that the former perform better with small datasets. On a related note, in Ecological Monographs Mark Rees and Steve Ellner present integral projection models for populations in temporally varying environments.
Thomas Cornulier and colleagues present in Ecology Letters a method for estimating the number of annual breeding attempts, and apply this to look at reasons for the decline of yellowhammers.
In the latest issue of Conservation Letters, Bruce Kendall looks at the use of diffusion approximations in Population Viability Analysis.
In a paper in Oikos Scott Forbes applies investment theory to understand how birds manage risk.
Finally for this month, for matrix population modellers Peter Zuidema et al. introudce in the American Naturalist a new tool for analysing how faster growing individuals contribute to population growth rates.
Here is a round-up of some interesting methods papers published in the past few weeks. If you see any more papers that you would like to see flagged up, leave a comment below or email me.
In PLoS Biology Wayne Getz presents a thoughtful review of the models and modelling approaches that might be useful in predicting the consequences of multiple threats to ecosystems from a food web / ecosystem perspective.
Ecology has several interesting methods papers: Murray Efford and colleagues show how it is possible to use likelihood methods to estimate densities of animals from arrays of passive detectors (such as arrays of microphones). Michael Neubert et al. present a new method for estimating the rate of growth of perturbations in transient dynamics. Jessica Metcalf et al. apply integral projection models to the problem of estimating flux of individuals in tropical forests. And Grosbois et al. demonstrate a new approach for estimating individual survival / mortality rates from mult-population data.
In Conservation Biology Berlund et al. show how Bayesian methods can be used to understand habitat association of trees from presence records and environmental data. Finn et al. compare methods for estimating population size variability with a view to priortising populations that are more risk to extinction from variability. And Christopher Grouios & Lisa Manne ask whether occupancy or abundance data are more useful in predicting population persistence and how this impacts on reserve design.
Ecology Letters has a paper by Paul Murtaugh comparing model selection methods that is likely to be of general interest. Mosser et al argue that density may not be a generally good measure of habitat quality (in terms of food/ resources), particularly if low quality habitat provides a refuge for non-reproductive individuals.
Finally, in Systematic Biology Sennblad & Lagergren show how probabilistic orthology analysis can be used to overcome some of the problems in identifying orthologous genes and gene products. And there is some debate about the use of barcodes in taxonomy centring on the effects of sampling error on the model used to delimit species.
One of the big motivating factors in setting up Methods in Ecology and Evolution was the recognition that there are lots of ways to present research (without losing sight of the importance of peer review, rigour, and quality assurance).
However in terms of uptake and usage, the problem with the conventional paper is that it is not necessarily tailored to conveying research in a quick and convenient manner. For ‘methods’ papers a big problem is that users need to understand how new approaches and tools work in practice and even to see examples of the method in action. The printed paper or pdf is not always the best way to do this.
Some publications are trying out different ways of presenting research. For example the Journal of Visualized Experiments is placing the emphasis on video presentation of research. There are some examples from Evolution and Ecology. For example, here is some research on visual sensitivity in lizards:
http://www.jove.com/index/details.stp?ID=127
And here is something on applied ecology:
http://www.jove.com/index/details.stp?ID=227
(you need to register to see the whole thing, but the clip should give a good introduction).
More generally I think that in writing methods papers we could do a lot more to ‘sell’ our methods. For instance, more ‘tutorial’-style supplements to papers, there being a mechanism for authors and readers to talk to each other, and for users to share their experience. By viewing these elements as a key part of the publication process everyone will benefit: authors will attract more readers, and users will find it easier to use new techniques.
Finally, and a bit more tongue in cheek this video includes an amusing take on the conventional publication process and the way that scientists conduct their debates via the literature (plus more!):
http://www.thedailyshow.com/watch/wed-august-5-2009/human-s-closest-relative
There were a number of ideas that led us to launch a journal devoted to methods. These include:
- Methods papers need support for readers to implement new approaches. This can include code, tutorials and a platform to communicate with authors and other users.
- Methods are often generic – they can be used across ecology and evolution, and there is justification
- There was no existsing journal specialising in methods papers.
So we have started this new journal and papers are starting to come in. However we are in the early stages and keen to get feedback and suggestions. Some of the things we are already doing are:
- We have a blog to keep readers and authors updated on what is happening.
- We will have a discussion forum that will allow readers to post questions about methods and authors to reply. This will be a useful resource for anyone applying methods in the future
- We can support all sorts of supplementary material: so not just code etc. but video, audio etc.
- We are happy to publish “Applications” which are short descriptions of code / software: this gives a ‘citeable source’ for those developing such tools.
So what else should we be doing? We want to hear from you if you have suggestions. These might include:
- Who should be on our editorial board? We are looking to cover a wide range of subject areas.
- What subjects should we be covering?
- Submissions – maybe you can think of a possible reasearch or review article?
Graziella Iossa, the MEE Journal Coordinator will be at the BES meeting and glad to hear your suggestions! Or leave a comment below, or email methods@me.com before, after or during the meeting. Have a good time in Hatfield!
