Role of Arctic Sea Ice Variability in Climate Models

• Main Author: Dammann, Dyre O.
• Other Authors: Bhatt, Uma, Polyakov, Igor, Zhang, Xiang
• Format: Thesis
• Language: English
• Published: 2011
• Subjects: arctic; sea; ice; Arctic; Fairbanks; albedo; Sea ice; Alaska
• Online Access: http://hdl.handle.net/11122/1879

Thesis (M.S.) University of Alaska Fairbanks, 2011 Arctic sea ice plays an important role in climate by influencing surface heat fluxes and albedo, so must be accurately represented in climate models. This study finds that the fully coupled ice-ocean-atmosphere-land Community Climate System Model (CCSM3.0) underestimates day-to-day ice variability compared to observations and employs the Community Atmosphere Model (CAM3.0) to investigate the atmospheric sensitivity to sea ice variability. Three 100-ensemble experiments are forced with climatological, daily-varying, and smoothly-varying sea ice conditions from an anomalously low ice period (September 2006-February 2007). Daily ice variability has a large local impact on the atmosphere when ice undergoes rapid changes, leading to local cooling and subsequent circulation changes. The most notable example of a large-scale atmospheric response occurs over Northern Europe during fall where daily ice variability forces reductions in the number and strength of cyclones, leading to positive sea level pressure anomalies, surface warming, and reduced cloud cover. Signature Page.i Title Page.ii Abstract.iii Table of Contents.iv List of Figures.vi List of Tables.x List of Appendices.xi 1 Introduction.14 1.1 Introduction – Arctic Sea Ice Evolution.14 1.2 Introduction – Impact of Sea Ice on Atmosphere.17 1.3 Model Versus Observed Arctic Sea Ice.21 1.3.1 Data Properties.21 1.3.2 Mean Ice Conditions.22 1.3.3 Standard Deviation.24 1.3.4 Day-to-day Change of Ice Concentration.27 Model, Data and Methods.30 2.1 Model Description.30 2.2 Observations.31 2.2.1 Data Used as Boundary Conditions in CTRL, DAILY and SMTH Experiments.32 2.3 Methods.41 2.3.1 Number of Ensembles.41 2.3.2 Storm Track Algorithm.42 2.3.3 Bandpassed Filtering.43 2 3 4 5 v Results and Discussion.44 3.1 Fall (September – October) Response.44 3.2 Winter (November – February) Response.55 3.3 Stormtrack Response in the Midlatitudes and the Arctic.59 3.4 Ensemble Analysis.62 Summary and Conclusions.65 References.68