103We have modified the PICR value for direct radiative forcing of sulfate, using data
from Stern (2006a, b), and Smith et al. ( 2011 ), as described in our methods paper
(Canty et al. 2013 ), because the modified time series is deemed to be more accurate
than the RCP value, which was based on projections of sulfate emission reductions
conducted prior to the publication of Smith et al. ( 2011 ).
The estimates of direct ΔRF from the various aerosol types are then combined
into two time series: one for the aerosols that cool, the other for the aerosols that
heat. Next, these two time series are multiplied by scaling parameters that represent
the aerosol indirect effect^33 for aerosols that cool and for aerosols that warm. These
are the six curves shown using colors that correspond to aerosol type. The total
direct ΔRF of aerosols that warm, and aerosols that cool, are shown by the red and
blue lines, respectively. The line labeled Net is the sum of the total warming and
total cooling term, and reflects the time series of Aerosol ΔRF (^) i input to the EM-GC
(Eq. 2.2). Finally, the black open square marks AerRF 2011 = −0.9 W m−^2 along the
Net time series, which is the best estimate of total ΔRF due to anthropogenic tropo-
spheric aerosols given by IPCC ( 2013 ).
Canty et al. ( 2013 ) relied on scaling parameters that were tied to numerical esti-
mates of upper and lower limits of the aerosol indirect effect given by IPCC ( 2007 )
(their Fig. 4). Figure 2.21 is our new scaling parameter “road map”, updated to
reflect estimates of the aerosol indirect effect by IPCC ( 2013 ). The set of scaling
parameters used in Fig. 2.6 are given by the intersection of “Middle Road” with the
AerRF2011 = −0.9 W m−^2 line in Fig. 2.21: i.e., αHEAT = 2.19 and αCOOL = 2.43.
Further details of our approach for assessing a wide range of aerosol ΔRF scenarios
in a manner consistent with both CMIP5 and IPCC is given in Canty et al. ( 2013 ).
Figure 2.7 shows time series of Aerosol ΔRF (^) i found using scaling parameters
αHEAT and αCOOL, combined with estimates of direct ΔRF of climate found as
described above, for five values of AerRF 2011 : −0.1, −0.4, −0.9, −1.5, and −1.9 W
m−^2 (open squares). The highest and lowest values of AerRF 2011 are the upper and
lower limits of the possible range, the second highest and second lowest values are
the limits of the likely range, and the middle value is the best estimate, all from
IPCC ( 2013 ). Three curves are shown for each value of AerRF 2011 : the solid curve
uses values for scaling parameters αHEAT and αCOOL along the Middle Road of
Fig. 2.21, whereas the other lines use parameters along the High and Low Roads.
Figure 2.8 shows time series of ocean heat content for the upper 700 m of earth’s
oceans from six sources, as indicated. The data have all been normalized to a com-
mon value of zero, at the start of 1993. This normalization is done for visual conve-
nience; the EM-GC model simulates OHE, which is the time rate of change of
OHC. The time rate of change is the slope of each dataset, which is unaltered upon
application of an offset. The data sources are:
Balmaseda et al. ( 2013 ): http://www.cgd.ucar.edu/cas/catalog/ocean/OHC700m.tar.gz
Church et al. ( 2011 ): http://www.cmar.csiro.au/sealevel/TSL_OHC_20110926.html
(^33) The aerosol indirect effect is scientific nomenclature for changes in the radiative forcing of cli-
mate due to modifications to clouds caused by anthropogenic aerosols.
2.6 Methods