RAPID AND SLOW COMMUNICATION OF OVERTURNING

1
RAPID AND SLOW COMMUNICATION OF OVERTURNING
CHANGES OVER THE NORTH ATLANTIC V. Roussenov and
R. Williams Oceanography Laboratories,
University of Liverpool, U.K.

• 1. Motivation
• We study the different mechanisms by
  which overturning changes are communicated over
  the basin
• rapid response wave propagation Kelvin waves,
  coastal trapped waves (Huthnance, 1978 Johnson
  Marshall, 2002)
• slower response advection along western
  boundary, local recirculation and evolution of
  dense water masses
• The study links to the UK NERC RAPID
  monitoring programme

2.Model experiments Isopycnic model (
MICOM ) is integrated over realistic
topography and driven by annual mean wind
stress and buoyancy forcing
(relaxation to Levitus data at high
latitudes). 100 years spin-up Twin experiment (10
years) standard vs. increased high latitude
buoyancy forcing Advection is monitored by a
transient model tracer, released in Labrador
Sea when the perturbed buoyancy forcing
starts.
coarse resolution 1.4o 0.7o fine
resolution 0.28o 0.14o
Depth m
Sea-surface height cm
Perturbed experiment Increased buoyancy forcing
simulated by shallowing of deep interfaces by
120m/10 days
layer interface depth along 35oN
3. Slow advective response Dense fluid is formed
in the Labrador basin and spreads equatorwards
along a deep western boundary current, as
revealed by releasing an idealised transient
tracer for 10 years.

• 4. Rapid response
• In the perturbed case, the increased buoyancy
  loss in the high latitudes leads to
• initially a pressure signal rapidly
  propagating along western boundary on a
  timescale of several weeks to months.
• an intermediate response involving changes in
  local circulation and layer thickness,
  excited by the waves (timescale of
  several months to years).
• eventually, the arrival of a transient tracer
  signal, after the increase in layer
  thickness and changes in sea-surface height.

year 0.5
year 1.5
year 2.5
Anomalies (perturbed-reference) in layer
thickness and
Labrador Sea sea-surface height


tracer
Layer 27.7 thickness m and transport (contours
in Sv) Labrador transient tracer
Time series of layer thickness (solid lines) for
the perturbed and reference case Transient
tracer (dashed lines)
N-S Hovmuller diagram of deep layer interface
depth anomalies (perturbed-reference), N-S
section following the western slope
Labrador Sea transient tracer - contours
W-E Hovmuller diagrams of deep layer interface
depth anomalies (perturbed-reference)

• 4.Conclusions
• Overturning changes via 3 phases
• rapid topographic waves
• local advective changes
• slower far-field advection
• The communication of overturning signals will be
  examined within the UK NERC RAPID monitoring
  programme by deploying bottom pressure recorders
  and vertical profilers along the western boundary
  of the N. Atlantic involving C. Hughes, P. Foden
  (POL), D. Marshall (Reading) and R. Williams
  (Liverpool)

References Huthnance J.M.,1978 On coastal
trapped waves analysis and numerical
calculations by inverse iteration. JPO,8,
74-92 Johnson H.M. and D.P. Marshall, 2002A
theory for the surface Atlantic response to
thermohaline variability. JPO, 32,
1121-1132 RAPID climate change thematic programme
http//rapid.nerc.ac.uk Contact
v.roussenov_at_liv.ac.uk ric_at_liv.ac.uk