In order to directly test the effect of changing environments on a species, we need to reconstruct those changes. Climate reconstructions, using the large scale computer simulations also used to predict climate change, have been run for certain periods in the past. A major difficulty with having only a few reconstructed time points is that we do not know what happened in-between them. Did climate change gradually between those time points, or did it remain stable for a long time and suddenly changed? As an important first part of the project, we worked to build large scale datasets in which the same climate model was run at regular, short intervals, thus providing a continuous reconstruction of the environment through time and space. Highlights of this work was a complete atlas of how the ice sheets changed over the last million years (Batchelor et al. Nat. Comm.), a high quality reconstruction of changing global climate over the last 120k years (Beyer et al. Sc. Data), and an extension of these reconstructions to cover the last 800k years using a climate model emulator(Krapp et al. Sc. Data). We also produced free software to easily manipulate and use these large reconstructions for archaeological and ecological studies (in the form of the R package pastclim,
https://evolecolgroup.github.io/pastclim/(s’ouvre dans une nouvelle fenêtre)). We used these reconstructions in our work (see below), but these datasets are also being extensively used by the broader scientific community.
A further difficulty in assessing the effect of changing environment on evolution is the fact that, when the environment changes, species often modify their ranges (e.g. through expansions or contractions). Such demographic changes can have big impacts on the genetics of a species, potentially leaving signatures that can be confused with those left behind by selection. Thus, the next step in the project was to reconstruct the demographic history of our species, tracking range changes that resulted from changes in the environment. For this task, we needed high coverage ancient genomes for modelling. Together with collaborators, we were the first to sequence ancient whole-genomes from Africa (Gallego et al. Science) and East Asia (Siska et al. Science Advances), improving our understanding of migrations in these two understudied areas. We also investigated the mixing of ancestral population groups in the eastern parts of Eurasia (Jones et al. Curr. Biol). We detected a new, important ancestral group in the Caucasus which later contributed to the Bronze age migrations into Europe (Jones et al. Curr. Biol.).
Using these new, high quality genomic datasets, together with data published by other geneticists, we used spatially explicit models to reconstruct the effects of climatic change on the demographic history of our species. We were able to study the role of climate in the expansions out of Africa of our ancestors ( Beyer et al Nat. Comm.), showing how aridity in the Arabian Peninsula acted as a gate opening and closing passage to Eurasia. A major finding was that suitable periods seemed to correspond to contacts with Neanderthal, which might have acted as competitors that prevented the successful colonisation of the Eurasian continent until a relatively later exit. We were also able to make a full reconstruction of the population demography of hunter gatherers as they left Africa and colonised the rest of the world (Maisano Delser bioRxiv), showing how climate and topography shaped the divergence of the main genetic lineages in humans. Furthermore, we showed the importance of climate in mediating the admixture of hunter-gatherers and farmers during the Neolithic (Betti et al. Nature Human Behav.). We were also able to adapt our model to other organisms, and reconstructed the impact of the last glaciation on an American passerine bird (Miller et al, bioRxiv). This crossover of our models to other species is very exciting, as the decreasing cost of sequencing means that large genetic datasets are becoming increasing available for animals, thus allowing us to explicitly test the impact of climatic change on the world fauna.
Leveraging our climate and demographic reconstructions, we have then been able to start to formally test the role of climate on shaping our species. A major result was the formal testing of the impact of climate on brain and body size evolution (Wills et al. Nat. Comm.). Many theories had been suggested for the important changes that we see in our species over the last million years, but these had not been tested. Thanks to the resources that we developed during the early parts of our project, we were able to show that body size has been directly affected by temperature. For brain size, we found that climate only explained a limited amount of change, but that the need for cooperative hunting of large megafauna in open environments was a key driver of larger brains. There are also two further studies, one on migraine and one on malaria, that were not published during the project, but that will be coming out soon.
Over the whole duration of the grant, together with the postdoc employed by this project, we have published 30 scientific papers, many of which were covered by the media. We also developed a tabletop game that has been used in several countries to teach children about the impact of changing environments on animals.
