Arctic Change and Possible Influence on Midlatitude Climate and Weather. Workshop Summary

Transcription

1 Arctic Change and Possible Influence on Midlatitude Climate and Weather Workshop Summary J. Cohen, X. Zhang, J. Francis, T. Jung, R. Kwok and J. Overland July 20, 2017

2 ARCTIC AMPLIFICATION

3 Sea Ice Decline

4 Sea Ice and Snow Cover Decline

5 Annual Cycle of Arctic Temperatures Courtesy of Fred Laliberte/Lawrence Mudryk

6 Sea Ice loss and full AA Sea ice loss is not the biggest contributor to AA Pithan and Mauritsen (2014)

7 Sensible heat flux and downwelling longwave radiation Courtesy of Tingting Gong (Units: W m - 2 yr - 1 )

8 WARM ARCTIC-COLD CONTINENTS/EURASIA

9 Arctic Warmth reaches to the Stratosphere

10 Arctic Amplification (a) (b) DJF Area-Averaged Temperature Anomalies (c) Cohen et al Review paper

11 Arctic Amplification - Jet Stream Francis and Vavrus 2012

12 Extensive Snow Forced Cold Signal Stratospheric Polar Vortex Weakens warming 4 Background Westerlies 3 Upward Energy Flux Downward propagation of High pressure and southward displacement of jet. 5 1 Regional Perturbation over Siberia/Urals 2 Increased Eurasian snow cover Negative Arctic Oscillation High Pressure over the Arctic and frequent cold air outbreaks 6 Sept Oct Nov Dec Jan Feb

13 Reduced Sea Ice Forced Cold Signal observations model Some model runs forced with low sea ice have been able to simulate atmospheric response as observed. Kim et al. 2014

14 Synthesis of Sea Ice and Snow Cover Cohen et al Review paper 13

15 Challenges with Data and Models Scarcity of observations in the Arctic Short time series in observations since AA Model deficiencies Uncoordinated modeling studies Biases and uncertainties in metrics for quantitative analysis The climate system is complicated

16 Scarcity of Arctic Observation Stations Courtesy of Wendy Ermold, University of Washington

17 Mid-latitude Weather is Complicated NH&CRYOSPHERE&CHANGES&! Summer/Early&Fall&ArcAc&Sea&Ice&Loss&! Fall&Eurasian&Snow&Cover&Increase&! Late&Fall/Winter&ArcAc&Sea&Ice&Loss& POLAR&VORTEX& L ARCTIC&AMPLIFICATION& CHANGES&IN:&&! Storm&Tracks&! Jet&Stream&! Planetary&Waves& NATURAL&VARIABILITY&! Internal&Climate&Modes&! Solar&Cycle&! Volcanic&ErupAons& GLOBAL&CLIMATE&& CHANGE& NH&MID2LATITUDE& WEATHER& Cohen et al. 2014

18 Modeling Studies on Linkage Between Arctic Change and Mid-latitude Climate and Weather - Progresses and Challenges Xiangdong Zhang and Judah Cohen IARPC, May 25, 2017

19 b ART2 SLP Arctic warming forced changes in SAT a 0.2 ART1 (PCORR = 0.9) b 0 ART2 (PCORR = 0.9) d ART2 Z Kug et al ation anomalies linked to Arctic n of sea-level pressure (Pa) (a,b) and m) (c,d) with respect to de-trended monthly b,d) during December February for the 14. Shading denotes significant values at c d EXP MME OBS EXP MME OBS REGR. ART1 and EA CMIP5 REGR. ART2 and NA CMIP5

20 However, other model simulations show diversified results a Loss Growth b Cooling Warming c AOGCM NSIDC AOGCM AOGCM inferred GISTEMP AGCM control AGCM perturbation Density Change in BKS sea-ice concentration (%) Change in CEUR SAT ( C) CEUR SAT ( C) McCusker et al. 2016

21 Atmospheric dynamics linking Arctic sea ice retreat/warming to midlatitude climate and weather Kim et al SEPTEMBER 2014 M A T S U M U R A E T A L Arctic Oscillation (AO) Matsumura et al J. Wallace, UW + - Eurasian cold winter FIG. 3. Zonal-mean zonal wind (shaded; m s 21 ), EP flux (vectors; kg s 22 ), and the EP flux divergence (contour interval is 0.3 m s 21 day 21 with the zero contour omitted) regressed onto standardized June Eurasian SCE for NASA (a) April, (b) May, (c) July, and (d) August. Convergence (divergence) is indicated by dashed (solid) contours. Vertical length of the vector is 200 times larger than its horizontal length. White contours indicate statistical significance at the 95% level for zonal-mean zonal wind.

22 Non-robust AO/NAO responses (Doug Smith et al., US CLIVAR Workshop) Negative NAO (DJF, mslp, hpa) Deser et al 2016;; Honda et al 2009;; Seierstad and Bader 2009;; Mori et al 2014;; Kim et al 2014;; Peings and Magnusdottir 2014;; Nakamura et al 2015 Little NAO response Screen et al. 2013;; Petrie et al 2015;; Blackport and Kushner 2016 Positive NAO Screen et al 2014;; Singarayer et al 2006;; Strey et al 2010;; Orsolini et al 2012;; Rinke et al 2013;; Cassano et al 2014 NAO response that depends on the forcing Alexander et al 2004;; Petoukhov and Semenov 2010;; Sun et al. 2015;; Pedersen et al 2016;; Chen et al

23 Does AO/NAO really play a role in linking Arctic and midlatitudes? +AO No Amplified Arctic Warming AO-driven temperature changes do not capture the Arctic amplification, or warm Arctic-cold Eurasia. -AO

24 Atmospheric circulation dynamics: A spatial pattern shift and the Arctic Rapid change Pattern (ARP) The rapidly changed Arctic from the mid-1990s to the early 2000s provide an opportunity to detect this circulation change signal. Zhang et al In the mid-1990s AO ARP

25 An increase in frequency of occurrence of negative ARP during recent years Warm Arctic Cold Eurasia 2015/16 Winter Enhanced transport of warm and moist air into the central Arctic Ocean and cold air to the Eurasian midlatitude in the extremely negative ARP phase.

26 Emergence of the ARP pattern in the fully coupled model experiment: CESM1 RCP 8.5 forcing experiment Sea Ice Loss Related Responses SLP 2 m Air Temperature Blackport and Kushner 2017

27 Polar Amplification Multi-model Intercomparison Project (PA-MIP) - D. Smith et al., partially supported by the H2020 APLICATE Experiment Time Slice Control Forcing Present-day Climatological SST and Sea Ice (SIC) 1. AMIP SST Arctic SIC pi Pre-industry SST 2 degree Future 2 degree warming SST pi Pre-industry SIC 2 degree Future 2 degree warming SIC Antarctic SIC pi Pre-industry SIC 2 degree Future 2 degree warming SIC 2. Coupled Control Arctic SIC Antarctic SIC pi Constrained by Present-day Climatological SIC Constrained by Pre-industry SIC 2 degree Constrained by Future 2 degree warming SIC pi Constrained by Pre-industry SIC 2 degree Constrained by Future 2 degree warming SIC

28 Summary No consensus has been reached among the modeling studies;; Dynamic process linking Arctic and midlatitude has not been well understood, impacting selection of metrics to evaluate model performance;; Uncertainties exist in defining and prescribing forcing in AGCM or CGCM simulations;; Impacts of model systematic biases have not been well investigated;; Influence or modulation by tropical and midlatitude forcing remains unclear. Proposed effort Coordinated modeling experiments and analysis same design, forcing, and analysis metrics but different models. - PA-MIP: A great component.