As it has been reported by Shiogama et al. (, hereafter S16), the first half of 100 member ALL-forcing runs of Shiogama et al. (, hereafter S13) have the well mixed greenhouse gas concentrations of the year 1850 instead of that in the year 2006–2010 by an error. We have redone these experiments with the correct GHGs concentrations (S16). This error correction made little changes in the results and did not affect the conclusions of S13. The corrected Figures and are very similar to that in S13. The probabilities of drought events of ALL more at least as severe as the 2010 record is the same as S13 (1%) (see the first paragraph of the results section in page 172). With the bias corrections of ANSG and ENSO, the probabilities of drought events of ALL more at least as severe as the 2010 record are 14%. <3%, >29%, while those were 15% <5%, >32% in S13 (see the first paragraph of the right column of page 173). The ratios of ALL/NAT and noCS/ALL are 5.0 < 2.9, 8.0 > (6.3 < 3.2, 8.7 > in S13) and 1.1 < 0.9, 1.3 > (1.0 < 0.8, 1.1 > in S13), respectively (see the first paragraph of the right column of page 173). This error correction did not affect the other figures. 2 (Figure presented.) The black and 10 gray curves are probability density functions (PDF) of the Jul-Oct precipitation anomalies (%, relative to the 1979–2010 mean) during the 1979–2010 period for the observations and ALL-Long, respectively. The PDF is calculated as a normalized histogram of samples within half-overlapped 6%-width bins. The black vertical line is the 2010 observation. The green, blue, and red curves indicate the PDFs of the ALL, NAT, and noCS runs, respectively. Note that these PDFs were calculated from different numbers of samples: 32 for the observations and ALL-Long, and 100 for ALL, NAT and noCS. 4 (Figure presented.) The dashed green, blue, and red curves are the PDFs of the July-October precipitation anomalies (%) for the ALL, NAT, and noCS runs, respectively, as shown in Figure 1. The solid curves indicate the PDFs with the bias corrections (with the best estimates of the scaling factors) according to the precipitation responses to (a) ANSG, (b) ENSO, and (c) ANSG + ENSO. (d) Solid curves show the PDF estimates with the neutral ANSG + ENSO. Acknowledgements We thank Oliver Angélil who found the errors in the GHGs concentrations. This work was supported by the Program for Risk Information on Climate Change (SOUSEI program) and Grant-in-Aid 26281013 from the Ministry of Education, Culture, Sports, Science and Technology of Japan. Earth Simulator and NEC SX (NIES) were utilized for the simulations. We have no conflict of interest.
!!!All Science Journal Classification (ASJC) codes