High Performance Computing-enabled Sea-Level Rise Hazard Analysis

dc.creatorLuo, Xiao
dc.date.accessioned2023-12-13T16:15:47Z
dc.date.available2023-12-13T16:15:47Z
dc.date.issued2023-08
dc.description.abstractSea level rise, as a result of climate change, is expected to drive coastal hazards that bring significant damages to coastal communities in the future. There are mainly three factors that contribute to the global sea level rise: ocean thermal expansion, ice melting from Greenland and Antarctica ice sheets, and glacier melting. High uncertainties remain in the projections of sea level rise arising from sea level rise prediction models and possible climate trajectories. This dissertation aims at better quantification of sea level rise hazard from explore its source, e.g., the ice-ocean interactions, and performing Probabilistic Sea Level Rise Hazard Analyses (PSLRHA, Lin 2012; Thomas and Lin 2020) using the most updated datasets and experimental protocols in the community. Ocean warming induced melting of ice sheets largely contributes to the uncertainties of sea level rise contribution from ice sheets. To effectively quantify the ocean warming effects in future, a semi-empirical framework is proposed to derive the melting projections of both Greenland and Antarctica ice sheets under oceanic forcing, allowing for sampling uncertainties from a larger ensemble of projections given the input ocean warming is available. An ice plume simulator is used to model Greenland tidewater glaciers and an ocean box model is utilized to represent ice-ocean activities within ice shelves’ cavities in Antarctica. Response functions are generated from unit temperature rise experiments. The future projections of ocean warming induced melting are obtained by convoluting the responses with future ocean warming time series. This framework provides a novel and alternative perspective to simulate ocean-induced melting in a more efficient manner without loss of accuracy. Quantification and integration of the sea level rise uncertainties, arising from both models and climate scenarios, are essential to better inform coastal planning and decision making for climate adaptation. We prepare the simulation data based on the current generation of models and protocols from the climate science community to better portray the future climate and project sea level rise. The aggregation process produces the probability of exceeding a specific sea level rise threshold at a certain location and facilitates the creation of the global sea level rise hazard map. The relative importance of each climate scenario and sea level rise contributing models are demonstrated via the deaggregation process. We identify the models that have most contribution to extreme sea level rise thresholds, with large fluctuations in the high thresholds among ice sheet models. Finally, we show the practical implementation of PSLRHA results via compound flooding analyses using Houston as an illustrative example. More accurate PSLRHA analyses could benefit from updated model weights instead of equal model weights. In this dissertation, we also utilize different approaches to evaluate the performance of the process-based models, namely GCMs and ice sheet models, via comparison against observations of different variables. For the first time, we provide model weights for the ice sheet model simulations based on 1) comparison with observations and 2) interdependency among the models. The model weights are used to update the PSLRHA analyses. We find minimal differences of sea level rise hazard global maps. More significant differences are found in the hazard deaggregations of ice sheet models under low to medium thresholds. We expect the PSLRHA analyses enabled by observations and weighted inputs would produce more meaningful results for sea level rise hazard analyses of coastal regions. This dissertation improves the sea level rise hazard analyses from both exploring the ice-ocean interactions and performing probabilistic sea level rise hazard analyses with PSLRHA framework using the most updated datasets and model weights.
dc.description.abstractEmbargo status: Restricted until 09/2024. To request the author grant access, click on the PDF link to the left.
dc.identifier.urihttps://hdl.handle.net/2346/97157
dc.language.isoen
dc.subjectSea level rise
dc.subjecthigh performance computing
dc.titleHigh Performance Computing-enabled Sea-Level Rise Hazard Analysis

Files

Original bundle

Now showing 1 - 1 of 1
No Thumbnail Available
Name:
LUO-DISSERTATION-2023.pdf
Size:
20.94 MB
Format:
Adobe Portable Document Format

License bundle

Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
1.57 KB
Format:
Item-specific license agreed upon to submission
Description: