Fig. 4: Weather@home 2010 ensemble mean.
From: Drivers behind the summer 2010 wave train leading to Russian heatwave and Pakistan flooding
Panel a: ensemble mean of SAT anomaly for the period 24.07–08.08 for Glob2010. Note that the apparent date mismatch relative to observations stems from the fact that weather@home uses a 360 days calendar. Panel b: PDF for standardized WRussia SAT index for GlobClim (solid black line) and for Glob2010 (solid orange line). Using standardized indices helps to correct for model biases for a better comparison with observation. Panel c: Same as for panel a but for Rainfall anomalies over the Indian subcontinent for Glob2010. Panel d: PDF for standardized Pakistan Rainfall index for GlobClim (solid gray line) and Glob2010 (solid blue line). Panels e and f: Same as panels c and d for Rainfall anomalies in Reg2010 and RegClim. Vertical lines in panels b, d, and f show the climatological 90th percentile threshold (black line) and the 2010 threshold (red line) together with the percentage of ensemble members exceeding each threshold in the climatology (upper value) and in the 2010 ensemble (lower value). Panel g: ensemble mean for Z300 anomalies from Glob2010. Panel h: Same as for Panel g but for V300 anomalies from Glob2010. Anomalies are calculated as the deviation from GlobClim and RegClim ensemble means. The stippling represents significant grid points at α = 0.05 and is calculated applying a Student-t Test. The smoothing of the curve is done using a Gaussian kernel to produce continuous density estimates.
