67The PDO represents the temporal evolution of specific patterns of sea level pres-
sure and temperature of the Pacific Ocean poleward of 20°N (Zhang et al. 1997 ),
which is caused by the response of the ocean to spatially coherent atmospheric forc-
ing (Saravanan and McWilliams 1998 ; Wu and Liu 2003 ). The PDO is of consider-
able interest because variations correlate with the productivity of the fishing industry
in the Pacific (Chavez et al. 2003 ). An index based on analysis of the patterns of
SST conducted by the University of Washington^14 is used.
The IOD index^15 represents the temperature gradient between the Western and
Southeastern portions of the equatorial Indian Ocean (Saji et al. 1999 ). The IOD
index is used so that all three major ocean basins are represented. Variations in the
IOD have important regional effects, including rainfall in Australia (Cai et al. 2011 ).
However, global effects are small, most likely due to the small size of the Indian
Ocean relative to the Atlantic and Pacific oceans.
The increase in the RF of climate due to human activity causes a rise in tempera-
ture of both the atmosphere and the water column of the world’s oceans (Raper et al.
2002 ; Hansen et al. 2011 ; Schwartz 2012 ). The oceanographic community has used
measurements of temperature throughout the water column, obtained by a variety of
sensor systems and data assimilation techniques, to estimate the time variation of
the heat content of the world’s oceans (OHC, or Ocean Heat Content) (Carton and
Santorelli 2008 ). Generally the focus has been on the upper 700 m of the oceans.
Considerable uncertainty exists in OHC. Figure 2.8 shows estimates of OHC in
the upper 700 m of the world’s oceans from six studies: Ishii and Kimoto ( 2009 ),
Carton and Giese ( 2008 ), Balmaseda et al. ( 2013 ), Levitus et al. ( 2012 ), Church
et al. ( 2011 ), Gouretski and Reseghetti ( 2010 ) as well as the average of the data
from these six studies. Ostensibly, all of the studies make use of similar (if not the
same) measurements from expendable bathy-thermograph (XBT) devices and the
more accurate conductivity temperature depth (CTD) probes. Use of CTDs began in
the 1980s, and expanded considerably in 2001 based on the deployment of thou-
sands of drifting floats under the Argo program (Riser et al. 2016 ). Alas, the ocean
is vast and much is not sampled. The differences in OHC shown in Fig. 2.8 pub-
lished by various groups represent different methods to fill in regions not sampled
by CTDs, as well as various assumptions regarding the calibration (including fall
rate correction) of data returned by XBTs.
The QOCEAN i term in Eq. 2.3 is the EM-GC representation of OHE in units of W
m−^2 : i.e., OHE is heat flux. The quantity OHC represents the energy content of the
upper 700 m of the world’s oceans. To relate OHC and OHE, several computational
steps are necessary. First, the OHC values shown in Fig. 2.8 are multiplied by 1.42
(which equals 1/0.7) to account for the estimate that 70 % of the rise in OHC of the
(^14) The PDO index is at http://research.jisao.washington.edu/pdo/PDO. This record begins in year
- Prior to 1900 we assume PDOi is equal to 0.
(^15) The index for IOD from 1982 to present is based on this record provided by the Observing
System Monitoring Center of NOAA http://stateoftheocean.osmc.noaa.gov/sur/data/dmi.nc
From 1860 to 1981, IOD is based on data provided by the Japan Agency for Marine-Earth Science
and Technology at http://www.jamstec.go.jp/frcgc/research/d1/iod/kaplan_sst_dmi_new.txt
2.2 Empirical Model of Global Climate