“…In this particular case, this may be due to the peculiar convergence of meteorological features (snow cover feedback, temperature inversion, etc., see U., 1884; Angot, 1881) that may hardly be captured by 20CR and/or our downscaling method. Extremely hot seasons are generally well simulated, especially recent summers like 1976 (Brochet, 1977) or 2003 (Trigo et al, 2005), but also much older events like spring 1893 (Plumandon, 1893). Figure 15 shows the corresponding evolution of precipitation for Paris Montsouris.…”