statistical analysis of the links between blocking and nor
TRANSCRIPT
1. Separate)storms)by)path,)randomly)select)100)events,)2. Composite)blocking)during)life)cycle)of)the)storms)
Can$we$see$blocking$affec0ng$storm$path?$
Bo8om:$sta0s0cally$significant$anomalies$WRT$a$Monte$Carlo$sampled$set$of$composites$
block frequency
block anomaly
Block frequency
block anomaly
100 storms taking nor’easter path
100 storms taking SW to NE path
Statistical Analysis of the links between Blocking and Nor’easters James F. Booth, City University of New York, City College, Stephan Pfahl, ETH Zurich
DISCUSSION
West Atlantic Cyclone’s Track Density
cyc
lon
es p
er w
inte
r
Blocking Frequency
blo
ck fre
qu
en
cy
Using ERA-Interim Reanalysis, 1979-2013, Oct. – Apr. 1. Track cyclones: using MCMS; Bauer and Del Genio, 2006. Some of the work will focus on US East Coast Cyclones: (shown at right, top) defined following Hirsch et al. 2001. 2. Identify blocks: (shown at right, bottom):
follow method of Schwierz et al. 2004 3. Track blocks, loosely based on “wind footprinting” algorithm of Leckebusch et al. 2008.
Motivated by recent studies showing or suggesting that blocking can influence extreme precipitation events, slow cyclones to increase storm surge, and impact the NAO. This study uses numerical algorithms to track extratropical cyclones and atmospheric blocks, and analyzes the interactions between the two. This poster will focus on interactions on the west side of the North Atlantic Ocean.
INTRODUCTION & METHODS PART 1: Can we identify a role for blocking in affecting storms’ paths in the Western North Atlantic Ocean?
Cyclone Speed vs.
Surface Wind Strength
Given that part of the motivation for this work is the influence of blocking on cyclone impacts, in part through the s lowing of storms, we also examined the relationship between cyclone propagation speed and cyclone windspeed (at right). There is a relationship, but it is weak. This shows that slow cyclones can indeed create strong wind hazards.
APPENDIX
JFB is par*ally funded through: ROSES-‐2012 NASA grant NNX14AD48G
Bauer M. and A. D. Del Genio, 2006: Composite analysis of winter cyclones in a GCM: Influence on climatological humidity. J. Climate, 19, 1652-‐1672.
Leckebusch, G. C., D. Renggli, U. Ulbrich, 2008: Development and applica*on of an objec*ve storm severity measure for the Northeast Atlan*c region. Meteorol. Z. 17, 575-‐587
Pfahl, S. and Wernli, H. Quan*fying the relevance of atmospheric blocking for co-‐located temperature extremes in the Northern Hemisphere on (sub-‐)daily *me scales, 2012: Geophys. Res. Le7. 39, L12807.
Pinto, J. G., S. Zacharias, S., A. H. Fink, G. C. Leckebusch, U. Ulbrich, 2009: Factors contribu*ng to the development of extreme North Atlan*c cyclones and their rela*onship with the NAO. Clim. Dyn., 32, 711–737.
Schwierz, C., Croci-‐Maspoli, M. and Davies, H. C., 2004: Perspicacious indicators of atmospheric blocking. Geophys. Res. Le7. 31, L06125.
125 oW 100 oW
75oW 50oW 25o W
0o
24 oN
36 oN
48 oN
60 oN
72 oN
125 oW 100 oW
75oW 50oW 25o W
0o
24 oN
36 oN
48 oN
60 oN
72 oN
Storms with lowest speedmatch valuescount: 50
125 oW
100 oW 75oW 50oW
25o W
0o
24 oN
36 oN
48 oN
60 oN
72 oN
125 oW
100 oW 75oW 50oW
25o W
0o
24 oN
36 oN
48 oN
60 oN
72 oN
Storms with highest speedmatch valuescount: 50
BLOCK FREQUENCY COMPOSITE
V2: CYCLONES CONCURRENT WITH BLOCKS
BLOCK FREQUENCY ANOMALIES
IDENTIFY CYCLONES OCCURRING CONCURRENT WITH BLOCKS
1. Retain cyclone/block (C/B) pairs with temporal overlap >= 2.5days
2. From this set, retain the C/B pairs with distance between any point within cluster and cyclone center <= 1500km for at least one time step.
3. This gives a “1-1” cyclones/blocks set: - 279 cyclones, 224 blocks
RESULT: in this case, downstream blocking is present for both the fast and slow storms.
BLOCK FREQUENCY COMPOSITE
BLOCK FREQUENCY ANOMALIES
PART 2: What is the role of blocking in affecting storm propagation speed?
Here we separate the cyclones based on their propagation speed and then analyze the block composites. This is done first for all storms and then repeated using only storms that occur concurrent with a block.
RESULT: blocking near the track genesis is present for the slow storms and absent for the fast storms.
V1: CONSIDER ALL CYCLONES
4 Days Prior to Cyclogenesis
4 Days Post Cyclogenesis
LEAD/LAG Composites for C/B pairs
DIRECTION 1: FIND CYCLONES, LOOK FOR BLOCKS Subjectively separate cyclones based on path. Subsample each path set and analyze block composites.
1-DIRECTION: For Nor’easter path, the block location suggests a steering of the storm west of Greenland. For the SW to NE case, lack of blocking consistent with storms to pass through. DIRECTION-2: We identify the strong block events in the region, and then find the associated cyclone tracks, and the path is consistent with 1-DIRECTION.
DIRECTION 2: FIND BLOCKS, LOOK FOR CYCLONES
For a fixed region, area-average the blocking flag (region shown in red box below) and then identify the top 50 events that remain strong for at least 4 days. For the dates of the peak of these events plus or minus 1 day (yeilding 112.5 total days in consideration), determine if there are any cyclones within a 1000 km radius of the region.
Count of Storms found: 31 PRELUDE: Cyclone
speed vs. NAO For each cyclone in the North Atlantic, determine the phase of the NAO during the time that the cyclone tracks o v e r t h e o c e a n . T h e n calculate the average speed o f p ro p a g a t i o n o f t h e cyclone during the time it is over the ocean. Pinto et al. (2009) examined extreme cyclone strength, and found NAO+ à stronger events. Here we find NAO+ à faster events. Sampling the events for which NAO is strongly positive or negative suggests a difference, however the NAO+/- distributions do not pass the Kolmogorov-Smirnov test. The correlation between the two variables is: 0.3
0 5 10 15 20 25 300
0.05
0.1
0.15
0.2
0.25
0.3
0.35
Forward Speed (m/s)
Prob
abili
ty
null true for Kolmologoz−Smirnov Test
NEGPOSNEUTRAL
−3 −2 −1 0 1 2 30
5
10
15
20
25
30
NAO val (avg. during CYC in ATLC)
Forw
ard
Spee
d (m
/s)
Scatter plot for each cyclone track PROPAGATION SPEED NAO < -1 vs. NAO > 1 vs. -1 < NAO < 1
In part 1, we found the connection between block location cyclones’ paths that most would expect based on anecdotal evidence: blocks steer the Northeasters and are not present during the storms that move SW to NE.
In part 2, the examination of cyclone speed we also found the expected to results: slower cyclones were associated with more blocking. However, this only holds true if we include all cyclones, regardless of the presence of a block. If, instead we only consider the cyclones that occur during blocking, we find that faster storms are associated with more blocking. In this case, the anticyclonic circulation around the block appears to work in tandem with the cyclone motion to advect the storm northward at a greater propagation speed. These results are being tested in more detail using case-studies and the Pacific ocean basin. The take home message of this poster is: the interactions between blocks and cyclones is not always as expected, and in some cases, blocks can occur in the presence of storms with fast cyclone propagation speeds.