Fig. 4: Observed and dynamical and thermodynamical temperature trends. | Nature Communications

Fig. 4: Observed and dynamical and thermodynamical temperature trends.

From: Heat extremes in Western Europe increasing faster than simulated due to atmospheric circulation trends

Fig. 4

Dynamical (a) and thermodynamical (b) contributions to the summer TXx (summer maximum of maximal daily temperature) trends from ERA5 ECMWF Reanalysis (red line), E-OBS observation (orange line), and the 170 CMIP6 model simulations (names in ordinate) that were available (black dots) averaged over Western Europe. The thermodynamical contributions are simply calculated as residual by subtracting the dynamical trend from the total trend (Fig. 3). For reference, the red bar at the bottom of a stands for the 95% confidence interval of the estimate of the ERA5 TXx dynamical trend, estimated with a Gaussian assumption, i. e. the interval is calculated as plus or minus 2* the standard deviation (STD) of the error estimate on the trend coefficient. This confidence range describes the uncertainty related to the internal variability. This shows that this confidence range, calculated with the single realization of the observation, is consistent with the uncertainty range calculated from simulation members (respective standard deviations for observed trend and simulated trends of 0.28 and 0.25).

Back to article page