Masterprosjekt V19 KGS

Masterprosjekt vår 2019 ved GEO - Andreas Born

Coupled climate-ice sheet dynamics

Main content

Project description

The understanding of earth's climate on time scales longer than 10,000 years critically depends on the build-up and demise of large continental ice sheets. Over the past million years, its number alternated between the two that are present today on Greenland and Antarctica and four, with two additional masses of ice over North America and Eurasia. Among other consequences, this caused
sea level to change in excess of 130m. Most of the time it stood below the current level, but occasionally several meters above.

Although the basic components of climate and ice sheets interactions are easily understood, the full picture is extremely complex due to the nonlinear nature of the underlying processes and the large difference in relevant time scales. While ice sheets respond on the scale of centuries to millennia, these changes are due to variations in climate that require a faithful description of the seasonal cycle.

One of the prime examples of this limited understanding is the so-called '100-kiloyear-problem' (Paillard, 1998; Raymo and Nisancioglu, 2003). It formulates the difficulties in relating the periodicity of 100,000 years in global ice volume to the apparent lack of forcing on those time scales. Studies of this problem must reconcile the need to run very long ice sheet simulations with the limited
computational speed of climate models (Abe-Ouchi et al., 2013). We recently developed a new method to compile synthetic transient climate model data that is consistent with proxy reconstructions (Jensen et al., 2017). It identifies fields of climate analogs from
the climate model data that best resemble the geological proxy record. Importantly, this method takes advantage of existing climate simulations which makes it computationally efficient enough to create climate fields spanning several glacial cycles. These can then be used to force an ice sheet model.

References:
Paillard, D. (1998): The timing of Pleistocene glaciations from a simple multi-state climate model, Nature 391, 378–381 Raymo, M. E. and K. H. Nisancioglu (2003): The 41 kyr world: Milankovitch's other unsolved mystery, Paleoceanography 18, 1011
Jensen et al. (2017): A spatio-temporal reconstruction of sea-surface temperatures in the North Atlantic during Dansgaard-Oeschger events 5–8, Climate of the Past Discussions
Abe-Ouchi et al., (2013): Insolation-driven 100,000-year glacial cycles and hysteresis of ice-sheet volume, Nature 500, 190–194

Hypothesis (scientific problem):
1) Synthetic climate reconstructions based on climate analogs are robust enough to reliably force an ice sheet model over multiple glacial cycles.
2) The dynamic interaction of climate and ice sheets gives rise to nonlinear phenomena such as hysteresis (memory) and bifurcation points.

Test (work):
The work consists of two major parts, the implementation of the analog method for which existing python code may be adapted, and the scientific evaluation of the results which involves forcing an ice sheet model with the synthetic climate data. The robustness of the climate analogs will be tested by means of a relatively straightforward Monte Carlo approach and a subsequent statistical evaluation. Special attention will be paid to key periods such as glacial inceptions and terminations. The second part involves preparing the climate data for use in a simple ice sheet model written in FORTRAN, running the model, and analyzing and visualizing its output.

Prerequisites

Candidates should have some previous knowledge of computer programing with languages such as python, matlab, R, IDE or similar. Please contact the supervisor before choosing this topic.