Tempo, mode and causality in nature

Contact person: Lee Hsiang Liow
Keywords: Geology, Paleontology, Evolution, Environmental change
Research groups: PHAB (GEO), Integreat (MI), Natural History Museum (EPA)
Researchers: Trond Torsvik (GEO/PHAB), Lee Hsiang Liow (PHAB/NHM), Ingrid Kristine Glad (MI/Integreat), Trond Reitan (GEO/PHAB), Kjetil Lysne Voje (NHM)
Department of Geosciences (GEO), Department of Mathematics (MI), Natural History Museum (NHM)
 

The dynamics in biological and geological systems are key in revealing underlying short- and long-term natural processes in the geobiosphere. Recent databasing efforts continue to contribute to an exponential growth of publicly available timeseries data including those of morphologies/phenotypes, species and functional compositions, and abiotic measures including temperature, sea-level and climate. The last few years have also seen an acceleration of the development of computationally intensive timeseries approaches, including machine-learning and Bayesian causal inference, that are amendable to temporal data that are patchy, unequally sampled and noisy, characteristic of biological, geological, and other environmental variables. Research proposals could harness these approaches to understand processes approximated by biological and/or environmental data on decadal to millions of years timescales, further develop available methods and/or suggest new approaches for compiling (e.g. using AI in rapid phenotyping) such timeseries data, and for understanding the dynamic systems that produced such data.

Example of topics from natural sciences:

  • Tempo and mode of biological and geological evolution
  • Time-dependency in empirical rates of change in biology and geology
  • Drivers of long-term geobiosphere changes

Example of topics from methodological research:

  • Ornstein-Uhlenbeck and other stochastic differential equations as timeseries tools
  • Causal inference from timeseries data
  • Accounting for temporal and spatial uncertainty in statistical inference
  • Change point and anomaly detection

External partners:

  • Mentoring and internship will be offered by a relevant external partner.