Thursday 05/10/2017

Heat redistribution and the warming slowdown

Fig. 1: Deseasonalized global area mean surface temperature (T_s) anomaly time series. Data reference period is from 2001-2008.

Two periods are defined:
• T_s rising period 1985-2001.
• T_s slowdown (decrease) period 2002-2012.

Fig. 2: Trend over two periods (1985-2001 and 2002-2012) and their differences (2002-2012 minus 1985-2001) for T_s, ocean hean content (OHC) and mixing layer depth (MLD)

• The trends are significantly different between two periods
• T_s is from ERAINT, and OHC and MLD from ECMWF ocean reanalysis ORAS4.
• T_s trend increases from south to north in second period, but OHC and MLD show increase in opposite direction.

Fig. 3: Differences of zonal means from two periods for T_s, OHC and MLD

• T_s increases from south to north, with substantial increases at high latitudes of northern hemisphere
• Note: the differences for OHC and MLD are 1985-2001 minus 2002-2012 for comparison.
• Three curves are more or less consistent, implying the relationship between them.

The following mechanism may be proposed
• When there is more OHC at 50^oS over 2002-2012, means more heat into the ocean, implying the MLD will be deeper, so cooling T_s
• When there is less OHC at the high latitudes of northern hemisphere over 2002-2012, means more heat stayed near the surface, implying the MLD will be shallow, so warming the surface.

Fig. 4: Zonal mean net surface heat flux (F_T and F_s) differences (2002-2012 minus 1985-2001).

• Less heat to the tropical area in the second period, more heat into the ocean at other latitudes.
• Less heat to the tropical area, more heat into the ocean at other latitudes.
• This distribution cannot explain the T_s
distribution.

Fig. 5: Basin contributions to the zonal mean differences of T_s, OHC and MLD.

• Please note: the differences for OHC and MLD are 1985-2001 minus 2002-2012 for comparison.
• As expected, the Atlantic change dominates the meridional variability.
• It may be true the ocean reanalysis near the pole is not reliable.

Summary

It looks like the air-sea interaction is responsible for the T_s variability. Heat spatial redistribution in the ocean mixing layer may be the key for the warming slowdown, rather than what happened in the atmosphere.