Master Thesis Defense by Johanna Felicia Carlborg

Title: 100,000 Years of Global Volcanism from Ice Cores

Volcanoes are famous for their climatic impact. Large volcanic eruptions are known to cause short-term cooling years after an eruption. Due to the unpredictability of eruptions, the frequency of large eruptions is of importance to estimate future climate change. However, during the Last Glacial Period, the frequency of large eruptions is less known. Sulfate aerosols from eruptions deposit on the ice sheets in Antarctica and Greenland. Hence, ice core studies reveal volcanism in a continuous record. If the eruption is large enough, sulfate from the same eruption can deposit in both hemispheres. These bipolar eruptions can be used to synchronize ice cores for improved dating of the cores. The difficulty with the deeper part of the ice core records, is the diffusion of the elements, and the thinning of the annual layers due to gravity.
Therefore, this thesis aims to examine the possibility of determining sulfate deposition in this part of the ice cores. A previous study has quantified the sulfate deposition for bipolar eruptions between 9-60 ka (b2k). Here, we try to quantify 253 predefined bipolar match points between 12-108 ka (b2k). The ice cores used in this study are NGRIP, GISP2, NEEM (Greenland), EDC and EDML (Antarctica). This thesis successfully quantified 234 of these match points in the sulfate records. The NGRIP, EDC, and EDML records all worked well to quantify eruptions in the deeper part due to their high resolution. However, GISP2 and NEEM were hard to analyse because of their poor resolution.
The conclusion is that it is possible to determine sulfate depositions in this part of the ice core, but continuous, or nearcontinuous analysis is required when processing the sulfate data in the lab. Latitudinal origin and stratospheric aerosol loading were also examined for the eruptions. 170 eruptions were predicted to have a latitudinal origin below 40°N, and 64 eruptions were predicted to have a latitudinal origin above 40°N. Further, the frequency of bipolar supereruptions was found to be 0.1 per millennium, the frequency of bipolar eruptions with ∼VEI 7 was found as 1.47 per millennium, and the frequency of bipolar eruptions with ≤ VEI 7 is 0.8 per millennium.
Lastly, the volcanic impact on Dansgaard-Oeschger (DO)-events was investigated. From this we see that between 26-39% of DO-events are preceded by a volcanic eruption. This is quite significant and shows that volcanic eruptions could have a role in the onset of DO-events.


Supervisors: Anders Svensson og Helle Astrid Kjær
Censor: Jens Olaf Pepke Pedersen (DTU)