More than 2000 emails later

Like Comment
Read the paper

To get some insights about what happened behind the scenes of the paper ‚Adaptive responses of animals to climate change are most likely insufficient’ (publ. at Nature Communications), a co-author pretended to be a journalist and interviewed his colleague Viktoriia Radchuk, leading author of the paper. Here is their conversation:

Q: Who are you?

A: I am a quantitative ecologist, working as a post-doc at the Leibniz Institute for Zoo and Wildlife Research (IZW, Berlin, Germany). I am Ukrainian by nationality, and I was lucky enough to obtain a MATRA scholarship for a Master programme at Wageningen University (The Netherlands). I then did my PhD at the Université Catholique de Louvain (Belgium) with a focus on population dynamics models for two vulnerable butterfly species. Since finishing my PhD I have undertaken a few post-docs and extended my research agenda from a single species perspective focusing on rather short ecological time scales to a multi-species, community perspective adopting an evolutionary stance. My personal life led me to Germany. And, at the moment I am pregnant and looking forward to this new, exciting, but also challenging period of my life – with the baby.

Q: How would you summarize the results of this study?

A: Bird species respond to warming temperatures by advancing their laying dates. Such earlier phenological responses are adaptive as evidenced by fitness benefits, for example in terms of higher reproductive output. However, these adaptive responses are not sufficient to ensure long-term species persistence. We show that, despite the adaptive responses, the populations are still at risk of extinction.

A Common Guillemot (Uria aalge) at the moment that the large single egg is laid. The timing of breeding of this species was followed for 19 consecutive years on the Isle of May, southeast Scotland. Photo credit: M P Harris
A Common Guillemot (Uria aalge) at the moment that the large single egg is laid. The timing of breeding of this species was followed for 19 consecutive years on the Isle of May, southeast Scotland. Photo credit: M P Harris

Q: Do the findings reflect your expectations? Was there something surprising in the results?

A: I did not expect that phenological advancement would be adaptive on average. That came as a bit of a surprise. Even though our findings are a little ambivalent, with a p-value very close to 0.05, they were still surprising. In contrast, the fact that most populations of the studied species have low population viability was, unfortunately, what we expected. 

Q: How long did the whole process last, i.e. from the conception of the idea to the acceptance of the paper?

A: The idea to conduct such a study started forming in my head during the SEED meeting ‘Species adaptations to global change - a comprehensive risk assessment’ in Potsdam (7-8th of December 2014), led by co-author Kirsten Thonicke. That winter I conducted a literature search and started skimming abstracts of potential papers. The study was finished by the end of 2017 and we first submitted the paper in winter 2018. However, it took us a few rejections and over a year of revisions to finally get it published. The paper was officially accepted in mid-May 2019. So, it took roughly 4.5 years from the very beginning to having the paper accepted.

Q: What was the least exciting task within this study?

A: It is difficult to identify a single task. Two actually come to mind, screening the abstracts returned by our literature search and answering hundreds of emails. These tasks aren’t that exciting for quite different reasons. Screening abstracts is interesting at first, but after five days spent mainly on this I became bored and it was difficult to push myself to still speed-read them. Unfortunately, I could not just stop or switch to the next step in the project: I had to finish abstract screening to get a rough estimate of how many studies could be useful to our project, to ensure that the meta-analysis was feasible and worth conducting.

Email exchange is different in that emails were arriving all the time  and answering them would distract me from focusing on other tasks. Yet, I knew that it is best to answer them without delay to keep people interested and collaborative. I think this is the first project of mine for which I spent so much time on correspondence. Correspondence of different character: contacting people asking them to share their data for our meta-analysis, clarifying the co-authorship rules we agreed about, discussing the results… A quick look into my email archives suggests there were at least 2,230 incoming emails related to this project!

Q: What was the most exciting task within this study?

A: The most exciting task was developing the R package to implement the whole workflow and reproduce the results presented in the paper. I did it together with Alexandre Courtiol, a co-author who is also based here at the IZW in Berlin. We spent quite a few days (weeks?) together, working on the package. This was my first R package and I have learnt a lot from Alex. Some of the tasks we tackled were quite challenging, but then succeeding at solving them felt very fulfilling.

Q: How would you rank the tasks according to how time-consuming they were?

A: I think the most time-consuming task was the email correspondence, closely followed by two tasks: 1) reading the abstracts and papers; and 2) performing the analyses while developing the R package. It is difficult to objectively say which of these was more time consuming. The task of writing and editing the manuscript, although it did take quite some time, was the fastest. Overall, although this project did take quite long time, the time invested into it is still negligible compared to the time spent by the researchers to collect the primary data and keep their long-term studies going. Without such studies our project would not have been possible.

Adult red-billed gull (Larus novaehollandiae scopulinus) with chick. The birds are part of a 54 year study in the Kaikoura Peninsula, New Zealand. Photo credit: D A Mills
Adult red-billed gull (Larus novaehollandiae scopulinus) with chick. The birds are part of a 54 year study in the Kaikoura Peninsula, New Zealand. Photo credit: D A Mills

Q: What was the least exciting period during the whole project?

A: After I finally finished skimming the abstracts and first tried to extract the data from the identified papers by digitizing the figures I realized that some of the data were not reported in the papers. The only way to get those data was to contact the authors. I was afraid they would not answer and all the efforts until that point would have been in vain – the weeks spent skimming through the abstracts…. At first, actually, my fears appeared to be true: the response rate was very low during the first 3-4 weeks. But it did get better: for some reasons, people working with birds seem to be much more responsive than the researchers working on mammals and reptiles. What a relief! 

Q: What was the most exciting / memorable moment for you during this study?

A: The most memorable moment was when I contacted Prof. Steve Beissinger asking whether he would be willing to share his data for our meta-analysis. He answered saying that he would be in Berlin that whole year, based at WiKo (Wissenschaftskolleg zu Berlin) and would be happy to meet in person and discuss the project in detail. That was a big honour for me because Steve Beissinger is one of those people whose papers I had read during my PhD (I worked on Population Viability Analysis, PVA) and whose work I really admired. I used to put him in that category of people who  was not within reach of us, ‘young’ researchers. I was very excited and nervous when going to the meeting with Steve at WiKo. But, all my fears disappeared quickly: Steve was very friendly and supportive. He spent the whole day with me, discussing the project and commenting on things that may need particular attention. This meeting and interactions with the other co-authors taught me that, no matter their status, scientists are often keen to talk about their research and exchange ideas with young researchers. You just have to dare to ask!

Q: How many co-authors worked on this paper?

A: We are an international collective of 64 co-authors. 

Q: How many of these co-authors did you know personally prior to starting the project?

A: Not many, around 10 people, I would say.

Q: How many of the co-authors did you get to know better and collaborate closer since then?

A: I’ve got to know around 10 of my collaborators much better, with the majority of whom we then wrote a successful sDiv proposal for our new working group ‘sTraitChange’ (sDiv stands for Synthesis Centre for Biodiversity Research, based at iDiv, Leipzig). With these people I’m continuing to work on the questions that arose from this study published in Nature Communications. For more information, please see

Q: Have you visited any of the study populations?

A: No. But the next time I am in the Netherlands for a week or so, I will do my best to go and visit the great tit study population. It is actually located in the same area where I conducted the field work for my Master’s thesis – but at the time I was working on butterflies. 

A great tit (Parus major) in De Hoge Veluwe (The Netherlands). The population of this species was followed for 34 consecutive years. Photo credit: M E Visser
A great tit (Parus major) in De Hoge Veluwe (The Netherlands). The population of this species was followed for 34 consecutive years. Photo credit: M E Visser

Q: Was it your hardest project?

A: Not sure about the hardest because there are so many aspects to what makes something ‘hard’… But this project was definitely my longest one until now. Even my complete PhD project was shorter, although it included four seasons of field work, digitizing field data, data analyses and development of several population dynamics models…

Q: Now that you know how challenging and hard this project was, would you still do it if you were back to 2015 but with the more objective estimate of what it would take to accomplish it?

A: If I was the ‘old me’ of 2015 I would have still gone for it, even if I knew how much time and effort it would be. I knew from the beginning that this project was going to be a huge undertaking, although I must admit I underestimated the true amount of time and energy that would be required. Nevertheless, it felt good to do something that may be important and useful and to address an interesting question. Although there were a lot of costs associated with this project, there were also lots of benefits, irrespective of the publication outcome. Getting to know many researchers worldwide, learning a lot about populations of different species, and learning how to develop an R package. Since this project I have made it a practice to develop an R package for every project of mine that requires a substantial investment in terms of analyses.

Q: If this study would have been conducted in 20 years, how would it be different?

A: In 20 years we should have longer time series, and the power of analyses would likely be higher. I would expect our results would no longer be marginal and we would get more certainty in our conclusions. At the same time, machine learning algorithms are getting more and more popular, so perhaps future analyses would allow us to relax some of the assumptions of our study, such as linearity in relationships. It would be really great to be able to disentangle microevolutionary responses from phenotypic plasticity, but I am actually skeptical whether that would be possible in 20 years for enough of our time series…

Leading author of the paper. Photo credit: J Zwilling
Leading author of the paper. Photo credit: J Zwilling

Q: Do you have ideas of follow-up projects for other researchers?

A: Yes. It would be great to disentangle the effects of multiple environmental factors on species responses. Climate change is just one aspect of global change. In the Anthropocene, species responses (adaptive or not) are triggered by a set of environmental factors acting in concert: climatic conditions (the focus of the current study), habitat modification and fragmentation, invasive species, pollution and many more. It would be useful for the management of species and insightful in general to assess the relative contributions of these different environmental factors on the phenotypic responses of species, and, in turn, their population viability. Another avenue of research, as we mention in the discussion of this paper, could be to look at the impacts of climatic variation (and not just changes in mean climate) on the species. Another very important issue to address is how much of the observed responses are due to phenotypic plasticity vs microevolution. Answering this question is impeded by the fact that even more data would be needed per study. For example, a pedigree would be needed to disentangle genetic responses from phenotypic plasticity. Unfortunately there are only a few studies that currently have such data. 

Q: Do you have a last message for early career researchers?

A: This will sound very banal, but I would suggest to follow your heart. If you feel passionate about the topic and convinced that addressing it is the right thing to do, go for it, no matter how challenging it may seem! I think being inspired helps us overcome many issues, which otherwise would seem insurmountable. From a more practical point of view, undertaking a study like this gives an opportunity to develop your collaboration network and to learn from the experience of others. 

Viktoriia Radchuk

Scientist, Leibniz Institute for Zoo and Wildlife Research

I am a quantitative ecologist, broadly interested in factors affecting the dynamics of a system at different levels of organization: from an individual level (traits) to community level