PLAVCOVÁ E. and KYSELÝ J. Evaluation of daily temperatures in Central Europe and their links to large-scale circulation in an ensemble of regional climate models. Tellus 63A, 763–781.
Due to a computational error, circulation types classified using the comparison of flow strength and flow vorticity in a given day (i.e. the straight, cyclonic, anticyclonic and hybrid types) were determined incorrectly. Obviously, this error changes results of the relative frequencies of these types over the studied period (part of Table 5) and the mean temperature anomalies on days falling into the circulation types (parts of Table 6 and Figs 11 and 12). Results for the directional types as well as the evaluation of temperatures simulated in RCMs stay unchanged.
Revised Table 5 shows that after the correction, there are more/fewer days with straight flow in winter/summer, fewer days with cyclonic and anticyclonic types in winter, and an increase of anticyclonic types in summer in the ERA-40 data and all RCMs. However, the general patterns that RCMs underestimate straight flow in both seasons and underestimate/overestimate anticyclonic/cyclonic types in winter compared to ERA-40 remain unchanged. Since relative comparisons of the frequencies in the observed data and RCMs have been primarily discussed in the text, conclusions are unchanged.
- (Revised Fig. 11. )
[ Mean daily Tmax and Tmin anomalies on days falling into individual circulation types over 1961–1990 calculated for the CL region and airflow indices from the ERA-40 sea level pressure data (triangles) and for 8 RCMs (shown as boxplots). S – straight, C – cyclonic, A – anticyclonic, h – hybrid, NW – northwesterly, SW – southwesterly, SE – southeasterly, NE – northeasterly flow. ]
Revised Table 6 and Figs 11–12, however, indicate that there are changes in mean anomalies of Tmax and Tmin linked to these circulation types. The changes are in magnitude and significance of the anomaly, and in some cases in winter also in the sign of the anomaly. The most important change is for anticyclonic types in winter, when both mean Tmax and Tmin anomalies are significantly negative (in contrast to prior results when mean Tmax anomaly was positive). However, the pattern that Tmax is relatively warmer in comparison to Tmin for anticyclonic types is still true. So, all the discussions in the text that the underestimation of anticyclonic types contributes to the underestimation of diurnal temperature range are valid for the correct data, too.
- (Revised Fig. 12. )
[ Overview of temperature anomalies (represented by colours) on days falling into individual circulation types and biases in frequencies of circulation types, for observed and RCMs data, in the CL region over 1961–1990. Total temperature biases (in °C) in given seasons and for given variables are shown in the right columns. Relative frequencies of types (in %) for observed data (ERA-40) are shown in the first row of panels. Symbols + and – indicate the ratio between simulated and observed frequencies of the types (see legend). ]
In Discussion (Section 5.3), the finding that temperature-circulation links are generally more pronounced for Tmax than Tmin stays true only for summer. No changes are needed in the Abstract and Conclusions of the paper.