Forecasting the effects of global climate change on species distributions is a major challenge with direct relevance for conservation, management, and policy. Incorporating genetic information into these models may help improve these predictions by accounting for within-species variation in demographic history and adaptation to local environments. In a recent issue of Molecular Ecology Resources, Dr. Joaquín Martínez-Minaya and colleagues present a new approach to forecasting species distributions accounting for within-species genetic variation and spatial autocorrelation using the model species Arabidopsis thaliana. Here, co-author Dr. F. Xavier Picó gives us a behind-the-scenes look at the study.

Link to study: https://onlinelibrary.wiley.com/doi/10.1111/1755-0998.13024

What led to your interest in this topic / what was the motivation for this study? 
Climate change models ought to incorporate more realistic species’ attributes to better understand their response in predicted and probably inevitable warming scenarios for the near future. In other words, we need to model the demographic, ecological, genetic and evolutionary processes that account for changes in distribution range mediated by warming. However, these data are hard to obtain at large spatial scales even for a single species, which seriously limits our understanding of the impact of global climate change on biodiversity. Our motivation was to develop a model to overcome such limitations using long-term existing data for an annual plant that, interestingly, may be the result of the suite of processes mentioned above, such as genetic structure and spatial autocorrelation of data.

What difficulties did you run into along the way? 
This work is the result of a collaborative effort among scientists from different disciplines, including genetics, ecology and mathematics, that in some cases worked together for the very first time. We had to find the way to speak the same language to achieve a common objective from planning to execution. In a way, we taught each other as learned from each other throughout the development of this work. Although this can be regarded as a difficulty, it was also an extremely rewarding process to see how we all pulled off the project.

What is the biggest or most surprising innovation highlighted in this study? 
From a conceptual viewpoint, the heterogeneity that any species has – due to historical, demographic, ecological and genetic factors – can no longer be overlooked whatever the research question and goal. Molecular markers allow us to tackle such inherent heterogeneity beyond our full comprehension of the underlying forces accounting for it. From a technical viewpoint, spatial Bayesian models take spatial autocorrelation of data into account. Ignoring spatial autocorrelation is a huge problem when it comes to the correct interpretation of spatial models. Handling these two key elements at once represents the most remarkable innovation of this study.

Moving forward, what are the next steps for this research?
With no doubt, the most important next step is to include demographic and evolutionary processes explicitly into global climate change models. We need to expand our current Bayesian framework to include dispersal, establishment of new populations, and local adaptation in a context of rapid environmental and land-use changes fuelled by global climate change. These fundamental processes, which can be modelled and/or parameterized with empirical data, will confer realism and power to model predictions. Only realistic models will generate an array of likely global climate change scenarios upon which we will be able to pose working hypotheses and appropriate actions to mitigate the impacts of global climate change on biodiversity.

What would your message be for students about to start their first research projects in this topic? 
Learning in science is a tough process, but extremely rewarding. Try to learn from the best to acquire solid foundations. Nevertheless, do not forget that your particular view on a given problem may open up new paths to keep making progress on the discipline. Recall what all renowned artists normally do: they first copy the old masters to learn the techniques to end up innovating and developing their own artistic style. Innovation can only be done when one understands the potential and limitations that any technique has.

What have you learned about methods and resources development over the course of this project? 
Complex problems require imaginative solutions that, for common people, can only be addressed by gathering together professionals from different disciplines sharing similar interests. In addition, it is important to bear in mind that the biological knowledge of the study species is of paramount importance to assess the value and impact of new methodologies. Inevitably, we believe quite often that any new method developed by us might become a panacea to solve multiple problems. Although this enthusiasm is necessary to move forward, it is very important to clearly detect the caveats and limitations of the methods and resources developed. In our particular case, the knowledge of the study species across the region allowed us to identify when the model outcomes were interpretable and when they could be biased.

Describe the significance of this research for the general scientific community in one sentence.
Promising progress is being made to improve models that will allow us to figure out realistically the future of biodiversity in a context of warming, which seems to be inevitable.

Describe the significance of this research for your scientific community in one sentence.
Genetic heterogeneity and spatial autocorrelation, the result of multiple forces acting in concert, can be handled and interpreted to better understand the response of any species to global climate change.

Citation
Joaquín Martínez‐Minaya, David Conesa, Marie‐Josée Fortin, Carlos Alonso‐Blanco, F. Xavier Picó, & Arnald Marcer. (2019). A hierarchical Bayesian Beta regression approach to study the effects of geographical genetic structure and spatial autocorrelation on species distribution range shifts. Molecular Ecology Resources, 19(4), 929-943. https://onlinelibrary.wiley.com/doi/10.1111/1755-0998.13024