Friday linksfest: Messiness and using Google Street View to map species

Anyone that knows me in real life will know that I’m I hate mess. I hate wires that tangle everything up in my flat, creating impenetrable black and off-white thickets. I hated my old office mate leaving his fieldwork kit strewn all across the office, grass festering away on the edges of his quadrat. I hate my sister’s inability to cook without the whole kitchen descending into culinary chaos. In summary, I hate mess.*

So it was with some concern that I saw the recent TED talk by Tim Harford, talking about how messiness can improve problem-solving flash up on my twitter feed. I clicked the link, despairing that I might have to scatter papers all over my desk in an effort to become a better scientist. Thankfully, that wasn’t what happened. Harford actually says that putting constraints or introducing randomness into how you solve problems can mean you produce better solutions. It’s very TED-y, linking disparate ideas from music and science, but I think Harford has something. Give it a look.


Something else that grabbed my attention this week was a paper which used Google Street View to build up a picture of where a plant species in Spain can be found. I like papers like this that just blow my mind. I would never have thought differently enough to come up with this idea.

I love using ggplot for analysis in R but up until know I have had to use a different package to do my diagnostic plots for models. Thankfully, now this has been solved with the ggfortify package so you never need to use ugly base R plots again….

And finally, if you are an early career researcher in the UK you might be interested in a joint BES & ZSL meeting aimed at helping you establish a career in conservation. Check it out, it has great speakers and I’m sure it will be as good as all other BES meetings I’ve been to.


* I do however, love all of the people I mention here. Even if I don’t like the mess you cause.

Friday Linkfest: Vagueness in ecology, and was I wrong about tropical forest recovery?

Those of you who read my latest post on defining resilience will know I just spent two days discussing what resilience is. It is somewhat ironic, then, that over on Dynamic Ecology on Jeremy Fox just put up a blog post discussing situations when vagueness in ecology can be useful. Considering that he has previously been a stickler for precision he is very open to vagueness, but only when it is productive. For example the concepts of ‘species,’ ‘ecosystem’ and, even ‘ecology’ are fairly vaguely defined – but very few people would argue that they aren’t useful…

Thompson-Reuters have just released their Impact Factors for the year. Exciting, right?  Methods in Ecology and Evolution have moved up to be the 9th highest ranked journal in ecology, and have a blog post about this. Worth reading if only for Bob O’Hara’s recognition that the whole system is a bit ridiculous.

A really nice looking new paper out this week shows that animals with bigger prey, that forage in three dimension tend to have bigger home ranges. Not usually my topic, but I was blown away by how good the figures looked in this paper.

In case you missed it, our paper on the response of different functional groups to forest recovery just came out a week or two ago, you can read a blog post on it here and see the open-access paper here.

For those of you into r and stats, here is a nice ggplot2 cheat sheet that is easily searchable (HT Dynamic Ecology).

A paper that challenges the conclusions of some of my work on forest regeneration has just come out in PNAS. The paper suggests that the characteristics of tropical forests recovering from clearance are highly idiosyncratic. One of our previous studies suggested that these recovery trajectories are relatively predictable, but I am open to the possibility we were wrong. I think this is one paper I need to have a proper read of and write a full post about…

Friday linksfest – Biodiversity hackathon, what difference does conservation make? & Should we be using trait analysis more?

Sooo, I decided it would be a good idea to steal an idea from my favourite ecology blog and start sticking up some content once a week covering all the things that I haven’t had time to write about properly (cue a chorus of people saying that none of my writing is ‘proper’ anyway…). So here goes.

There will be a biodiversity hackathon held in the fantastic setting of the Natural history museum in London on the 20th June. From the looks of it the hackathon will be working with data from the exciting Predicts project and the Living planet Index with the aim of “developing new tools and techniques to streamline data mining for biodiversity.” I’ll be going along, so give me a shout if you are too.

A new paper in Conservation Biology discusses what difference conservation makes to the extinction risk of 235 ungulate species, using redlist categories. The paper finds that if conservation actions for these species had suddenly ceased in 1996, by 2008  148 of the species would have deteriorated by one red list category. Quantifying the impacts of conservation is vital if we are to stop the extinction of species that are worth the effort of saving, but we need to get on and work out how to prioritise which species we should focus on saving. (HT to Paul Woodcock on twitter for highlighting that paper).

For those of you interested in research on logging in tropical forests, I just chanced across the NCEAS working group on Land-sharing/sparing in tropical logged forests – I’m intrigued to see what they will produce.

Also this week I noticed the Ask for evidence campaign run by Sense about Science, after the claims by UK minister Theresa May that EU policy on migrants would encourage more migrants to make the trip across the Mediterranean were challenged in an open letter. Given my love of all things evidence-based I encourage those of you that hear dodgy claims by your member of parliament to challenge these views. We need more evidence in politics, making the people we elect know that it is important is a start on this long road.

And finally David Warton put well thought out post on the Methods in Ecology blog, calling for ecologists to use species traits in our analyses more frequently. David makes the good point that we have spent a lot of time producing separate models to work out what influences a species’ abundance, rather than just producing one model with different trait values for each species. I largely agree with this as it allows us to get address the question of why species abundance varies across gradients of climate or human disturbance. I am now cursing myself that I didn’t think of this before submitting my most recent manuscript….