OCEANIC PRECIPITATION VARIABILITY ASSOCIATED WITH THE NORTH ATLANTIC OSCILLATION

1
OCEANIC PRECIPITATION VARIABILITYASSOCIATED WITH
THE NORTH ATLANTIC OSCILLATION

• Phillip A. Arkin, Heidi Cullen and Pingping
  XieUniversity of Maryland, ESSIC, NCAR/ESIG and
  Climate Prediction Center, NOAA

2
OUTLINE

• Background and questions to be addressed
• Data
• Around the Atlantic
• Boreal Winter
• Boreal Summer
• Elsewhere
• Conclusions/Points for further study

3
The North Atlantic Oscillation (NAO)

• Dominant mode of climate variability in the
  Atlantic in winter (van Loon Rogers, 1972)
• Seesaw of atmospheric mass between subtropical
  high and subpolar low (Walker and Bliss, 1932)
• Controls the path and intensity of storm track
  (Hurrell, 1995)
• Spectral density of NAO weakly exists at 2-3
  years (QBO), 7-10 years, also an increasing trend
  (Hurrell and van Loon, 1997)
• Significant impact on marine and terrestrial
  ecosystems

Images courtesy Martin Visbeck
4
Objectives/Questions

• Describe oceanic precipitation anomalies
  associated with the NAO
• What do they look like? In the Atlantic, tropical
  oceans?
• What is the relationship to anomalies determined
  from gauge data?
• Are they reasonable can CMAP et al. be used
  outside the tropics?
• How do the observed anomalies relate to
  circulation and storm track anomalies?
• Are the features we find sensitive to the dataset
  (CMAP all/observation only GPCP) used?
• Are they sensitive to the time scale
  (monthly/pentad) examined?

5
Data Used

• Indices from CPC web site
• NAO (700 hpa Z), AO (1000 hpa Z)
• Precipitation from CMAP
• CMAP/O uses a combination of IR and microwave
  satellite estimates and gauge observations
• CMAP/A uses reanalysis precipitation as well in
  regions without other data
• Xie and Arkin (BAMS, 1997)
• GPCP pentad
• Same inputs as CMAP/O
• Constrained to match GPCP monthly analysis
• Xie et al. (J. Climate, accepted pending minor
  revisions)
• Circulation from NCEP/NCAR Reanalysis
• 1000 and 500hpa winds and geopotential heights
• Kalnay et al. (BAMS, 1996)

6
Precipitation Climatology (CMAP)
DJFM
JJAS
7
Correlations with NAO Index (DJFM)
CMAP precipitation
REOF of 700mb Z
January
8
Correlations with NAO Index (DJFM)
1000 hpa height
500 hpa height
New CPC NAO index based on 500 hpa eof
9
DJFM Composites based on NAO Index (using monthly
anomalies)
CMAP precipitation High NAO quartile minus Low
NAO quartile
10
How much difference do using the reanalysis
precipitation make in these composites ?
CMAP/O
CMAP/A
Not much maybe provides a little continuity in
high latitudes
11
How robust are these results?
Low NAO
High NAO
Normalized CMAP composites
Each month normalized by standard deviation of
1000 composites based on random time series
12
Conclusions - DJFM

• CMAP et al. appear to be usable in extratropics
• Winter precipitation signal is consistent with
  earlier results
• Positive NAO associated with wet conditions in
  North Atlantic/northern Europe
• Negative NAO associated with wet conditions in
  Atlantic between 30-40N extending across the
  Mediterranean into the Middle East
• Oceanic precipitation anomalies strongest near
  eastern end of the main storm track as much as
  50 of the mean
• Signal appears to extend into the tropics

13
Correlations with NAO Index (JJAS)
CMAP precipitation
REOF of 700hpa Z
1000 hpa height
500 hpa height
14
JJAS Composites based on NAO Index (using monthly
anomalies)
15
How significant is the NAO signal in Boreal
summer?
High NAO
Low NAO
Normalized CMAP composites
16
Conclusions - JJAS

• Positive/negative NAO associated with dry/wet
  conditions in northern Europe with opposite
  anomalies farther north
• Oceanic anomalies less prominent than during DJFM
• Positive NAO associated with positive rainfall
  anomalies in western Atlantic between 15-30N
  probably because of the positive SLP anomalies
  just to the north and their influence on tropical
  systems
• Appears to be substantial dependence of the
  details of the results on the definition of the
  index

17
Does the NAO have manifestations outside the
North Atlantic Ocean?

• Hoerling et al. (Science, 2001) found that
  increasing trend in tropical precipitation from
  1950-2000 was related to similar trend in NAO
• Our record too short to compare directly, but
  maybe periods of high/low NAO index are
  characterized by coherent anomaly patterns away
  from the Atlantic Ocean

Implication high NAO index associated with
greater Indian/Pacific Ocean tropical
precipitation?
18
Composite DJFM Precipitation AnomaliesHigh NAO
Low NAO
CMAP/A
GPCP Pentad
19
CMAP normalized composite anomalies
Based on NAO index
DJFM
Based on Arctic Oscillation index
20
Conclusions

• Merged precipitation datasets good enough for
  descriptive diagnostic studies
• CMAP, GPCP pentad give similar results
• High latitude results look pretty reasonable
• The NAO has robust manifestations in Atlantic
  Ocean precipitation and circulation during Boreal
  winter
• Boreal summer signal present, but weaker and
  displaced northward
• Coherent precipitation/500hpa signal extends deep
  into tropics (maybe into SH for precip)
• Some (equivocal) evidence of a tropical signal in
  Indian and Pacific Oceans