Following Nature’s Clues with High-Performance Computing

How do environmental conditions shape habitats and influence the evolution of species? At the University of Bayreuth, the high-performance computing system “Festus” is helping researchers explore these questions in unprecedented detail. The work of Christie Phillip demonstrates how billions of data points can be transformed into new insights on biodiversity and environmental change. Whether in physics, artificial intelligence or ecology, modern research generates vast quantities of data that can no longer be processed efficiently on conventional desktop computers. To enable researchers to tackle complex scientific questions within a practical timeframe, the University of Bayreuth operates the high-performance computing system “Festus” through the Bayreuth Center for High Performance Computing (BZHPC). Around 300 users from a wide range of disciplines rely on this central research infrastructure.

The importance of this computing power becomes particularly clear in the work of Christie Phillip, a doctoral researcher in sports ecology. Her research investigates how environmental conditions and environmental change influence the adaptation of species. To do so, she analyses highly resolved raster data with a spatial resolution of up to one square metre. Using specialised terrain-analysis software, she derives parameters such as the Topographic Wetness Index, which indicates how strongly moisture accumulates at a given location. Combined with information on slope, aspect, precipitation and temperature, these data provide a detailed picture of local environmental conditions and microhabitats.

Such fine-scale environmental information forms the basis for modelling ecological niches and understanding more precisely how species respond to their surroundings. This is particularly important when investigating speciation processes in sympatry, situations in which parental and hybrid populations occur in the same area at the same time. The greatest challenge is the sheer volume of data involved. Even relatively small study areas contain billions of raster cells at this level of resolution. The computing capabilities of “Festus” make it possible to process these datasets efficiently, calculate terrain parameters and run complex ecological models within a realistic timeframe. The resulting findings help researchers better understand biodiversity in relation to landscape structures and environmental change. They also provide valuable foundations for targeted conservation measures, including those aimed at protecting endemic whitebeam species in the Franconian Jura. In addition, potential shifts in species ranges and hybrid zones under climate change scenarios can be assessed more reliably. 

For Prof. Dr. Stephan Kümmel, Director of the Bayreuth Center for High Performance Computing, this example highlights the broader value of the infrastructure for the university as a whole: “The University of Bayreuth offers a powerful infrastructure in the field of high-performance computing. The fact that this infrastructure is made available through the BZHPC to researchers from a wide range of disciplines and contributes to socially relevant findings such as these is a real strength of the University of Bayreuth. The BZHPC is therefore a successful example of the interdisciplinary networking that is actively practised at our university.” 

However diverse the research questions may be, cutting-edge science increasingly depends on powerful digital tools. The work of Christie Phillip illustrates how high-performance computing can help researchers uncover the complex patterns of the natural world and gain a deeper understanding of ecological change.