Auxiliary materials are available in the HTML. doi:10.1029/[2012JD017804]

jgrd50053-sup-0001-supp_fig1.pdfPDF document549KSupplementary Figure 1. Annual average boundary conditions used for the ENSO-like (a, b) and LGM (c) experiments. SAT anomalies (°C) for ENSO+ (A) and ENSO- (B) simulations, relative to the preindustrial control, are shown. Average observed surface temperature anomalies [Reynolds and Smith, 1994] associated with strong (×3) El Niño (ENSO+) and La Niña (ENSO-) conditions are determined from the NINO3.4 index and applied to SAT fields for the 0k simulations as surface boundary conditions. Also shown are topographic differences (100 m) between the two ice cap reconstructions utilized in this study (the modified LIC model) [Licciardi et al., 1998] and ICE-5.2G reconstruction prescribed for PMIP2 simulations [Peltier, 2004]. Global mean anomalies are given at the top right of each panel.
jgrd50053-sup-0002-supp_fig2.pdfPDF document570KSupplementary Figure 2. Comparison of simulated and observed El Niño precipitation rate anomalies (mm/day). Observed NCEP/NCAR reanalysis anomalies are composites of 1965, 1972, 1982, 1991, 1993, 1994, 1997, 2002, and 2002 events, relative to 1981–2010 climatology (after the NOAA/ERSL Physical Sciences Division), and are shown for the boreal summer (a) and winter (b). Equivalent simulated precipitation rate anomalies, relative to the preindustrial, are shown for the JJA (c) and DJF (d) seasons as mean surface conditions from the ENSO+ experiment. This is an idealized experiment designed with a strong surface forcing in order to elicit a response in the hydrological cycle, although reflects the spatial precipitation variability observed during El Niño phases.
jgrd50053-sup-0003-supp_fig3.pdfPDF document368KSUPPORTING INFORMATION

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.