A survey of computational methods for fossil data analysis

Aim: (1) Survey and organize computational approaches to fossil data analysis into a methodological framework. (2) Highlight the kinds of research questions about evolutionary and environmental change that can be answered by applying computational algorithms to mammal fossil data to better understand past ecosystems and climates. Questions: What models have been used for what research questions? What is their scope of application? What are their potential limitations? Search methods:- Our search of the literature was based on personal knowledge in combination with keyword-based searches. Papers were considered relevant if data-driven computational methods were used to analyse relationships between organisms and their environments at evolutionary time scales. Conclusions: We demonstrate that different research questions may be answered with the same computational algorithm, and different algorithms may be needed to answer the same question in different contexts. We believe that in order to move forward, we need to match knowledge of methods with knowledge of the fossil record in a research question-driven way. Figure 2 presents a proposed workflow. Following this framework, we survey existing work and highlight what research questions can potentially be answered with which methods, some of which may not have been reported in the evolutionary palaeontology literature to date. The "outcome of this survey is a proposal for a research agenda in computational fossil data analysis.