Temperature variability; Bayesian methods; uncertainty; model-data comparison
PhD position, Spatial climate variability patterns reconstructed with Bayesian Hierarchical Learning - Potsdam, Germany
Location
Potsdam
Germany
Category
Post date
Monday, July 8, 2024 - 11:19
Logistics
This project is part of the HEIBRiDS programme (for more information see: https://www.heibrids.berlin/).
The place of employment will be Potsdam in the Earth System Diagnostics group at the Alfred Wegener Institute (https://www.awi.de/en/science/geosciences/polar-terrestrial-environmental-systems/research-foci/earth-system-diagnostics.html).
It is a full-time position, limited to 4 years. The salary will be paid in accordance with the Collective Agreement for the Public Service of the Federation (Tarifvertrag des öffentlichen Dienstes, TVöD Bund), up to salary level 13.
The place of employment will be Potsdam in the Earth System Diagnostics group at the Alfred Wegener Institute (https://www.awi.de/en/science/geosciences/polar-terrestrial-environmental-systems/research-foci/earth-system-diagnostics.html).
It is a full-time position, limited to 4 years. The salary will be paid in accordance with the Collective Agreement for the Public Service of the Federation (Tarifvertrag des öffentlichen Dienstes, TVöD Bund), up to salary level 13.
Description
Understanding natural climate variability is crucial for assessing the range of plausible future climate trajectories in the next centuries. Natural climate variability will impact future regional climate trends and risks, and is important to consider for climate change attribution and adaptation planning (Laepple et al., 2023). Previous studies suggest that climate models often fail to capture long-term variability recorded in paleoclimate data and that this discrepancy varies depending on the spatial scale (local vs. global) and between regions (Hébert et al., 2022; Laepple et al., 2023).
This project will, for the first time, map the time-scale-dependent temperature variability based on paleoclimate data to investigate the corresponding range of possible future climate trends. Ultimately, the obtained results will help to improve the understanding of the origins of this variability and the representation of variability in climate models.
The interdisciplinary and complementary expertise of the team, which includes specialists in climate variability, the use of paleoclimate data, Bayesian hierarchical modeling, and data science, will provide an optimal supervision basis for the project.
Selection of relevant publications:
Hébert, R., Herzschuh, U. & Laepple, T. (2022). Millennial-scale climate variability over land overprinted by ocean temperature fluctuations. Nature Geoscience, 1–7. https://doi.org/10.1038/s41561-022-01056-4.
Klein, N., Kneib, T. & Lang, S. (2015). Bayesian generalized additive models for location, scale, and shape for zero-inflated and overdispersed count data. Journal of the American Statistical Association, 110(509):405–419. doi:10.1080/01621459.2014.912955.
Laepple, T., Ziegler, E., Weitzel, N., Hébert, R., Ellerhoff, B., Schoch, P., ... & Rehfeld, K. (2023). Regional but not global temperature variability underestimated by climate models at supradecadal timescales. Nature Geoscience, 16(11), 958-966.
Senf, C., Pflugmacher, D., Heurich, M., & Krueger, T. (2017). A Bayesian hierarchical model for estimating spatial and temporal variation in vegetation phenology from Landsat time series. Remote Sensing of Environment, 194, 155-160.
This project will, for the first time, map the time-scale-dependent temperature variability based on paleoclimate data to investigate the corresponding range of possible future climate trends. Ultimately, the obtained results will help to improve the understanding of the origins of this variability and the representation of variability in climate models.
The interdisciplinary and complementary expertise of the team, which includes specialists in climate variability, the use of paleoclimate data, Bayesian hierarchical modeling, and data science, will provide an optimal supervision basis for the project.
Selection of relevant publications:
Hébert, R., Herzschuh, U. & Laepple, T. (2022). Millennial-scale climate variability over land overprinted by ocean temperature fluctuations. Nature Geoscience, 1–7. https://doi.org/10.1038/s41561-022-01056-4.
Klein, N., Kneib, T. & Lang, S. (2015). Bayesian generalized additive models for location, scale, and shape for zero-inflated and overdispersed count data. Journal of the American Statistical Association, 110(509):405–419. doi:10.1080/01621459.2014.912955.
Laepple, T., Ziegler, E., Weitzel, N., Hébert, R., Ellerhoff, B., Schoch, P., ... & Rehfeld, K. (2023). Regional but not global temperature variability underestimated by climate models at supradecadal timescales. Nature Geoscience, 16(11), 958-966.
Senf, C., Pflugmacher, D., Heurich, M., & Krueger, T. (2017). A Bayesian hierarchical model for estimating spatial and temporal variation in vegetation phenology from Landsat time series. Remote Sensing of Environment, 194, 155-160.
Tasks
The selected candidate will:
• Synthesise proxy-based estimates of variability from the ocean, land and ice sheets into a spatio-frequency map of temperature variability.
• Develop a Bayesian Hierarchical Model that incorporate various data, forward models, prior information on the spatial structure of variability from instrumental data and climate models and the associated uncertainties.
• Use hierarchical weighted spatial priors to account for recent propositions that the spatial degrees of freedom of temperature on centennial- to millennial timescales are more similar to the one on multi-decadal timescales contrary to what climate models suggest (Laepple et al., 2023).
• Characterise the spatial structure of climate variability.
• Study the implications of the results for future projections of variability e.g. for frequency of extreme events.
• Synthesise proxy-based estimates of variability from the ocean, land and ice sheets into a spatio-frequency map of temperature variability.
• Develop a Bayesian Hierarchical Model that incorporate various data, forward models, prior information on the spatial structure of variability from instrumental data and climate models and the associated uncertainties.
• Use hierarchical weighted spatial priors to account for recent propositions that the spatial degrees of freedom of temperature on centennial- to millennial timescales are more similar to the one on multi-decadal timescales contrary to what climate models suggest (Laepple et al., 2023).
• Characterise the spatial structure of climate variability.
• Study the implications of the results for future projections of variability e.g. for frequency of extreme events.
Requirements
• MSc or equivalent degree in physics, climate sciences, applied mathematics or a related field
• Strong analytical and statistical skills
• Proficiency in a programming language (preferably R, Matlab or Python)
• Excellent English language skills, both written and spoken
• Strong motivation and ability to carry out research both independently and as part of a team in an international environment
• Detailed knowledge on climate reconstructions and paleo-proxy data and is an
advantage
• Knowledge of spectral analysis and/or Bayesian methods is an advantage
• Experience with large datasets (e.g. model fields and paleoclimate proxy databases) is an advantage
• Strong analytical and statistical skills
• Proficiency in a programming language (preferably R, Matlab or Python)
• Excellent English language skills, both written and spoken
• Strong motivation and ability to carry out research both independently and as part of a team in an international environment
• Detailed knowledge on climate reconstructions and paleo-proxy data and is an
advantage
• Knowledge of spectral analysis and/or Bayesian methods is an advantage
• Experience with large datasets (e.g. model fields and paleoclimate proxy databases) is an advantage
Applications
For more information and to apply see the application portal under the following link: https://heibrids.mdc-berlin.de/site/index.php
Application deadline
Further information
For more information, please contact Prof. Dr. Thomas Laepple (Thomas.Laepple@awi.de, https://www.awi.de/ueber-uns/organisation/mitarbeiter/detailseite/thomas-laepple.html)
Contact email