The beauty of large collaborations: how many seabird scientists does it take to make a manuscript?
With rising temperatures, many species of plants, insects, birds and mammals now breed earlier than they used to, meaning spring generally arrives earlier than in previous decades. This can be important for productivity, and so changes in timing of breeding are widely studied as a consequence of climate change. However, while we have quite a bit of information about these trends in terrestrial systems, we know much less about phenological trends of organisms living in marine environments. It’s an important knowledge gap to fill so that we can gain a better understanding of how marine organisms respond to environmental change.
Our paper in Nature Climate Change, Global phenological insensitivity to shifting ocean temperatures among seabirds, can be found here.
When we started this project it was unclear how marine top predators like seabirds had responded to environmental change globally. Overall had breeding phenology become earlier or later over time, or in response to environmental change? Were there any species groups that were responding very differently to others? Did any regions of the world stand out as having populations with particularly unusual trends? Finally, did any life history traits, such as feeding or migration strategy make a species more or less likely to adjust its timing of breeding? Together with a team of researchers from the Centre for Ecology & Hydrology, British Antarctic Survey and the University of Edinburgh, the first chapter of my PhD aimed to answer these questions by bringing together phenological trends from seabird populations around the world, using a formal phylogenetic meta-analysis to analyse them.
The traditional way to do a meta-analysis is to take data from previously published studies and combine them in a new analysis. However, we quickly realised that to have more flexibility with the analyses and to find enough of the right kind of data for the study, raw data would be much more useful. We came up with a different way to compile the data: contact researchers directly and invite them to contribute their data and collaborate with us. And so the next few months became all about emailing scientists from all over the world. One of my supervisors recommended that I only email ten people in a day, and I soon knew why… every group of emails I sent created more questions, suggestions and replies to follow up. But, every question somebody asked and concern they raised allowed us to refine our questions and methods, and ultimately make our models better. Imagine having over 80 reviewers every step of the way! I felt privileged to be able to contact world renowned seabird researchers and marine biologists every day and have discussions on how to make our science more accurate. After sending many hundreds of emails, presenting the idea for the project at several conferences and even using Twitter to invite researchers to join the project, we had amassed quite a dataset. For the final manuscript we used breeding phenology data from 145 populations of 61 species of seabird that came from 64 locations across all seven continents.
After spending over a year collating datasets, getting to grips with the statistical techniques, building and refining the statistical models and writing the manuscript, we were ready to send a first draft of the manuscript to over 80 co-authors for the first round of comments. Then came the biggest challenge. What do you do with 50+ versions of the same manuscript with comments from over 80 people, and how do you incorporate all these suggestions into a manuscript with quite a limited word count? It was pretty overwhelming, but we got there. Six months later, several tweaks, and with greyer hair than I started with, we had a manuscript that 88 people were happy to submit!
We found that overall marine birds have not adjusted their breeding seasons over time, or in response to changing sea surface temperatures over the last 60 years. However, we did find that variance in timing of breeding between years was highly variable across phylogenetic groups (albatrosses & petrels are the least variable; cormorants, gannets and boobies the most), and that resident species are generally more variable between years than migrants, meaning that some groups may potentially be less vulnerable to environmental change than others. Our next step is to test this by looking at the consequences of these different responses for breeding success. Using data from all over the world allows us to draw powerful conclusions about a group of species that are experiencing many threats globally.
Keogan, K., Daunt, F., Wanless, S., Phillips, R. A., Walling, C. A., Agnew, P., Ainley, D. G., Anker-Nilssen, T., Ballard, G., Barrett, R. T., Barton, K. J., Bech, C., Becker, P., Berglund, P.-A., Bollache, L., Bond, A. L., Bouwhuis, S., Bradley, R. W., Burr, Z. M., Camphuysen, K., Catry, P., Chiaradia, A., Christensen-Dalsgaard, S., Cuthbert, R., Dehnhard, N., Descamps, S., Diamond, T., Divoky, G., Drummond, H., Dugger, K. M., Dunn, M. J., Emmerson, L., Erikstad, K. E., Fort, J., Fraser, W., Genovart, M., Gilg, O., González-Solís, J., Granadeiro, J. P., Grémillet, D., Hansen, J., Hanssen, S. A., Harris, M., Hedd, A., Hinke, J., Igual, J. M., Jahncke, J., Jones, I., Kappes, P. J., Lang, J., Langset, M., Lescroël, A., Lorentsen, S.-H., Lyver, P. O. B., Mallory, M., Moe, B., Montevecchi, W. A., Monticelli, D., Mostello, C., Newell, M., Nicholson, L., Nisbet, I., Olsson, O., Oro, D., Pattison, V., Poisbleau, M., Pyk, T., Quintana, F., Ramos, J. A., Ramos, R., Reiertsen, T. K., Rodríguez, C., Ryan, P., Sanz-Aguilar, A., Schmidt, N. M., Shannon, P., Sittler, B., Southwell, C., Surman, C., Svagelj, W. S., Trivelpiece, W., Warzybok, P., Watanuki, Y., Weimerskirch, H., Wilson, P. R., Wood, A. G., Phillimore, A. B. & Lewis, S. 2018. Global phenological insensitivity to shifting ocean temperatures among seabirds. Nature Climate Change 8, 313-318; doi: 10.1038/s41558-018-0115-z