A comparative study on three EOF analysis techniques using decades of Arctic sea-ice concentration data

Abstract

Change in Arctic sea ice extent is one of the indicators of global climate changes. Spatio-temporal change and change patterns can be identified using various methods to facilitate human understanding global climate changes. Three empirical orthogonal function (EOF) techniques are discussed and applied to decades of sea-ice concentration (SIC) dataset in Arctic area for identifying independent patterns. It was found that: 1) discrepancies exist in magnitude and scope for each EOF pattern, however, the first two leading EOFs of variability possess high similarities in structure and shape; 2) Even though there are somewhat differences in amplitude of each PC mode, the first two leading PC modes maintain consistent in overall trend and periodicity; 3) There are significant discrepancies and inconsistencies in the third and fourth leading EOF and PC modes. The accuracies of three techniques are further validated in representing the physical phenomena of SIC anomaly patterns.

Keywords

empirical orthogonal functions / sea-ice concentration / comparative study / patterns / arctic

References

[1] Canadian ice service (CIS) MANICE manual of standard procedures for observing and reporting ice conditions [M]. 9th ed. Ottawa, 2002.

[2] WalshJ E, JohnsonC M. An analysis of arctic sea ice fluctuation, 1953–1977 [J]. Journal of Physical Oceanography, 1979, 9: 580-590

[3] DeserC, WalshJ E, TimlinM S. Arctic sea ice variability in the context of recent atmospheric circulation trends [J]. Journal of Climate, 2000, 13(3): 617-633

[4] WangJ, IkedaM. Arctic oscillation and Arctic sea-ice oscillation [J]. Geophysical Research Letters, 2000, 27(9): 1287-1290

[5] KwokR, SpreenG, PangS. Arctic sea ice circulation and drift speed: Decadal trends and ocean currents [J]. Journal of Geophysical Research: Oceans, 2013, 118(5): 2408-2425

[6] HendersonG R, BarrettB S, LafleurD M. Arctic sea ice and the madden-Julian oscillation (MJO) [J]. Climate Dynamics, 2014, 43(7/8): 2185-2196

[7] ZhangR-n, WuB-yi. The northern hemisphere atmospheric response to spring Arctic sea ice anomalies in CAM3.0 model [J]. Chinese Journal of Atmospheric Sciences, 2011, 35(5): 847-862

[8] LiF, WangH-j, LiuJ-ping. The strengthening relationship between Arctic oscillation and ENSO after the mid-1990s [J]. International Journal of Climatology, 2014, 34(7): 2515-2521

[9] LuZ-m, LiuC, MassinankeS, ZhangC-x, WangLe. Clustering method based on data division and partition [J]. Journal of Central South University, 2014, 21: 213-222

[10] SingaryerJ S, BamberJ L. EOF analysis of three records of sea-ice concentration spanning the last 30 years [J]. Geophysical Research Letters, 2003, 30(5): 335-343

[11] WangD W, YangX Q. Temporal and spatial patterns of arctic sea ice variations [J]. Acta Meteorologica Sinica, 2002, 60(2): 129-138

[12] StroeveJ, FreiA, MccreightJ, GhatakD. Arctic sea-ice variability revisited [J]. Annals of Glaciology, 2008, 48: 71-81

[13] DeserC, TengH-yan. Recent trends in Arctic sea ice and the evolving role of atmospheric circulation forcing, 1979–2007 [C]. Arctic Sea Ice Decline Observations Projections Mechanisms & Implications, 20137-26

[14] DommengetD, LatifM. A cautionary note on the interpretation of EOF [J]. Journal of Climate, 2002, 15(2): 216-222

[15] DommengetD. Evaluating EOF modes against a stochastic null hypothesis [J]. Climate Dynamics, 2007, 28(5): 517-531

[16] KimK, WuQ. A comparison study of EOF techniques: analysis of non-stationary data with periodic statistics [J]. Journal of Climate, 1999, 12(1): 185-199

[17] National SnowIce Data Center.Bootstrap sea ice concentrations from nimbus-7 SMMR and DMSP SSM/I-SSMIS, Version 2 [EB/OL], 2005

[18] CavalieriD C, ParkinsonP, ZwallyH JSea ice concentrations from Nimbus-7 SMMR and DMSP SSM/I passive microwave data [CD-ROM], 2002Boulder, ColoNatl Snow and Ice Data Cent

[19] ComisoJ CThe bootstrap algorithm [EB/OL], 1992

[20] BjornssonH, VenegasS A. A manual for EOF and SVD analyses of climatic data [J]. CCGCR Report, 1997, 97(1): 112-134

[21] FraedrichK, McbrideJ L, FrankW M, WangR. Extended EOF analysis of tropical disturbances: TOGA COARE [J]. Journal of Atmospheric Sciences, 1997, 54(19): 2363-2372

[22] HannachiA, JolliffeI T, StephnsonD B. Empirical orthogonal functions and related techniques in atmospheric science: A review [J]. International Journal of Climate, 2007, 27: 1119-1152

[23] OgiM, WallaceJ M. The role of summer surface wind anomalies in the summer Arctic sea ice extent in 2010 and 2011 [J]. Geophysical Research Letters, 2012, 39(9): 254-269

[24] CavalieriD J, ParkinsonC L. Arctic sea ice variability and trends, 1979–2010 [J]. The Cryosphere, 2012, 6(4): 881-889