Fig. 7: Soil moisture in June.
From: Drivers behind the summer 2010 wave train leading to Russian heatwave and Pakistan flooding
Panel a: composites of soil moisture for ensemble members selected with soil moisture signal over WRussia signal below the 10th percentile (65 ensemble members out of 649) during June (Glob2010 | soilM). Panel b: composites of SAT for Glob2010 | soilM and also exceeding the 90th WRussia SAT index (GlobClim) percentile (black solid line in panel c). Panel c: PDF of WRussia SAT index for Glob2010 (orange solid line) and for Glob2010 | soilM (red solid line). The rug shows the exact position of the data. Panel d: composites of regional Rainfall for Reg2010 | soilM and also exceeding the 90th Pakistan regional Rainfall index (RegClim) percentile (black solid line in Panel e). Panel d: PDF for Pakistan Rainfall index in Reg2010 (solid light blue line) and for Reg2010 | soilM (blue solid line). Both in panels c and e vertical solid black line show the 90th percentile calculated from the climatological PDF and vertical solid red lines show the 2010 threshold (expressed in units of s.d.) together with the percentage of events that exceed this threshold in the soilM sub-selection. Anomalies are calculated based on the Glob2010 (or Reg2010 for Rainfall) ensemble mean to differentiate from the impact of 2010 global SST. 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 a continuous density estimates.
