Opening doors to Open Science in ecology and evolution

On 7 December, Amsterdam's famous Artis Zoo will be the backdrop for a symposium on Open Science, followed by a day of workshops at Amsterdam Science Park. The aim is to speed up the adoption of Open Science standards in the fields of ecology and evolution. The organiser of the symposium is NIOO-KNAW postdoc Antica Culina.

"I always saw myself as an ornithologist", laughs Antica Culina. But after moving from Zagreb to Oxford to do a PhD, she found that she was gradually turning into an evolutionary ecologist: "It was more about the topic than about the organism." What she describes as a "long journey" towards Open Science advocacy had begun.

In Oxford, she studied pair bonds in birds. "I looked at why some pairs stay together while other pairs divorce or break up." One of the methods she used was meta-analysis: using results from other studies to be able to draw more general conclusions and/or increase sample size. But that turned out to be problematic.

Loss of data

"I realised that in a lot of studies, when you read them carefully there is some contradictory information, or some things are not reported - for example the sample size. So then you have to e-mail the authors. When you e-mail them, more than half don't reply and of those who do reply, more than half don't have the data to check what went wrong."

When important data stops being available to researchers, it's effectively lost. "People collect years of information on breeding populations, for example. They spend lots of time, money and energy. But after maybe thirty or forty years, there's nothing to show for that apart from one or two published papers."

She began to wonder: how can we improve this? And how do we improve reporting standards: can we trust what's reported? This brought her to what's become known as 'Open Science'. "There's now this shift towards the idea that all research actually has to be open. So everybody can come and understand what you have done."

New standards

Open Science practices such as open access to publications, open data and open (source) code are on their way to becoming the new mainstream in science, and more and more funders and journals actually require them. The advantages for the kind of ecological research Culina is doing are also obvious.

"Let's say that in ten years'  time, somebody were to take a meta-analysis I've done and re-run it with twenty more studies that appeared in the meantime. If I did not submit my data, this person would have to do everything again. And if the outcome of their analysis was different, they'd have to find out if the methodology was slightly different, or if maybe something new had started happening in the meantime."

Yet ecology and evolutionary biology are fields where there's still a lot of work to be done. "The thing is that research in ecology is very diverse", says Culina. "It's a very heterogeneous field. Compare that to astrophysics, for example: they started from a small community so they had standards from the very beginning, which were easy to expand later."

From within

The key to speeding up the adoption of Open Science in ecology and evolutionary research, believes Culina, is taking the broader standards that already exist and adapting or "slightly changing" them for these specific fields. And the only people who can really do this are themselves ecologists and evolutionary biologists.

"So far it's been more of a top-down approach: somebody else who doesn't know much about our field imposing some standards that may not fit the field. We are the ones who should influence these standards, because we are the ones working with the data."

Bringing these two worlds together was the idea behind organising a meeting: "I'd go to conferences on Open Science and I'd be the only ecologist out of 300-400 participants. But if you go to ecological conferences it's the same: none of these people are working on Open Science on a political or policy level. They don't come to our conferences."

Community

What was supposed to be a series of small workshops took on more ambitious proportions when DANS, the Royal Academy of Arts and Sciences' data institute, came on board. The University of Amsterdam's Institute for Biodiversity and Ecosystem Dynamics, IBED, and the Netherlands Organisation for Scientific Research, NWO, offered support.

"The audience is still ecologists and evolutionary biologists", says Culina. "What we're hoping to learn from the workshops is what researchers need and what they know already, how big the gap is. And at the symposium, our plenary speakers are ecologists who are advocates of Open Science and who know the pitfalls and the benefits."

For the organisers, says Culina, it's an excellent opportunity to lay out some future goals and discuss the next step. "A group of us are planning to meet the day after and maybe draught a little white paper: this is our community, and this is what our community needs."
 

• Open Science Tools, Data & Technologies for Efficient Ecological & Evolutionary Research: symposium and workshops to be held in Amsterdam on 7 & 8 December. For programme details and more information, please click here.