The Rapid Change Of Arctic Sea Ice Over The Past Three Decades

Simple item page
dc.contributor.advisorWang, Bin
dc.contributor.authorZhou, Xiao
dc.contributor.departmentAtmospheric Sciences
dc.date.accessioned2023-02-23T23:56:54Z
dc.date.available2023-02-23T23:56:54Z
dc.date.issued2022
dc.description.abstractThe rapid change of Arctic sea ice since the end of the 20th century has drawn much attention as an indicator of local and global climate. General global warming and amplified Arctic temperature rise have made the sea ice more vulnerable to natural fluctuation in the atmospheric and oceanic forcing. The area extent of the ice cover has been decreasing at an accelerating rate from 1998 to 2007, exhibiting more than seven times faster than from 1988 to 1997. However, the relative contribution of external forcing to the long-term decline of sea ice is far from clear. The reason for accelerated sea ice loss still remains unresolved. In addition, coupled global climate models are the primary objective tools to investigate the underlying mechanisms of sea ice response and provide future projections based on physical laws. However, substantial uncertainty persists in models’ simulated sea ice loss rate. Understanding the uncertainty source and selecting realistic models to make a reliable projection with reduced inter-model spread remains a frontline challenge.It is well recognized that the rapid change of sea ice has both external drivers and internal variabilities, but their relative importance is poorly known. A set of numerical experiments are conducted to detect the impact of external forcing on the sea ice. The numerical results reveal that the Arctic near-surface air temperature and its related longwave radiative and turbulent sensible heat play a principal role in melting Arctic sea ice and capture about 90% long-term decline trend. The sea ice thickness (SIT) also regulates the seasonal sea ice evolution in response to the Arctic warming. The observation result proves that the April SIT determines the sea ice area reduction rate in the melting season and the ice-albedo feedback further amplifies the sea ice retreat. The abrupt warming in the freezing season (DJFM) since 1998 significantly reduced the sea ice growth and further drove a rapid thinning of sea ice in April. The thinned sea ice favors a faster ice area reduction and accelerates the sea ice retreat in the melting season. Therefore, the September sea ice area (SIA) has dramatically declined since 1998. Extensive attention has been given to the performance of the sea ice model. Forty updated CMIP6 sea ice simulations are examined in this study. As with CMIP3 and CMIP5, a wide inter-model spread persists in the simulated September sea ice area. This huge uncertainty makes the all-model ensemble mean less meaningful. Therefore, selecting realistic models to create a reliable projection with reduced uncertainty is imperative. The existing SIA-based evaluation metrics are deficient due to observational uncertainty, prominent internal variability, and indirect Arctic response to global forcing. The result reveals that the model uncertainty may derive from the underestimation of near-surface air temperature and the inadequate intensity of ice-thickness adjustment in the melting season. Considering the dominant role of sea ice thickness in determining Arctic ice variation throughout the seasonal cycle, two SIT-based metrics are proposed, the April mean SIT and summer SIA response to April SIT, to assess climate models’ capability to reproduce the historical change of the Arctic sea ice area. The selected 11 good models reduce the uncertainty in the projected first ice-free Arctic by 70% relative to the 11 poor models. The chosen models’ ensemble mean projects the first ice-free year in 2049 (2043) under the SSP2-4.5 (SSP5-8.5) scenario with one standard deviation of the inter-model spread of 12.0 (8.9) years. Understanding the dominant role of sea ice thickness (SIT) could be an essential step in figuring out the sea ice seasonal evolution and the long-term decline in response to Arctic warming. SIT is determined by the near-surface air temperature in the winter, and it affects the sea ice melting rate in the following summer. The April SIT acts as a bridge, linking the winter freezing and summer melting processes together. Furthermore, the uncertainty of the sea ice model may also derive from the incorrect simulation of sea ice thickness. Given the notable spread in the CMIP6 models’ simulation, modelers should pay specific attention to improving sea ice thickness simulation, including the winter growth and summer thinning processes. Although some progress has been made, some critical issues still remain unresolved. The summer sea ice melting significantly slowed down after 2008 and showed a shallow mean state with strong interannual oscillation until the present. The reason is still unclear. In addition, the abrupt Arctic warming in the freezing season since 1998 remains a problem that currently receives extensive attention in the Arctic community. Both of these obstacles need to be addressed. A deep understanding of the sea ice behavior in response to Arctic warming could, to some degree, prepare us for a potentially ice-free Arctic in the future.
dc.description.degreePh.D.
dc.identifier.urihttps://hdl.handle.net/10125/104628
dc.languageeng
dc.publisherUniversity of Hawaii at Manoa
dc.subjectAtmospheric sciences
dc.titleThe Rapid Change Of Arctic Sea Ice Over The Past Three Decades
dc.typeThesis
dc.type.dcmiText
local.identifier.alturihttp://dissertations.umi.com/hawii:11570

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Zhou_hawii_0085A_11570.pdf
Size:
12.14 MB
Format:
Adobe Portable Document Format
Download

Collections

Ph.D. - Atmospheric Sciences