Wednesday 26 July 2017
Evaluation of extreme climate indices

Besides the general behaviour of the climate, characterization of climate extremes is an important part of the assessment of climate models, and consequently, for the exploration of climate change. In order to realise this aspect, various climate indices were defined by number of international programmes and projects dealing with climate change (for instance, CCL/CLIVAR/JCOMM - Commission for Climatology of WMO/Climate Variability and Predictability Research Programme/Joint WMO-IOC Technical Commission for Oceanography and Marine Meteorology -, ECA&D project, and some climate research teams from meteorological services, like KNMI from the Netherlands or Hadley Centre from the United Kingdom). The definitions of extreme indices are available at CLIVAR homepage.

Using regional climate model results of the Hungarian Meteorological Service 11 precipitation and 9 temperature-related indices and a complex index were analysed so far. In the future some additional, less common indices are planned to be computed. The applied indices are listed in Table 1.


Table 1: Investigated extreme climate indices

  Name Definition Unit
FD Frost days Tmin < 0 °C days
CFD Annual maximum number of consecutive frost days days when Tmin < 0 °C days
SU Summer days Tmax > 25 °C days
TN-10LT Extremely cold days Tmin < -10 °C days
TN20GT Tropical nights Tmin > 20 °C days
TX0LT Winter days Tmax < 0 °C days
TX30GE Hot days Tmax >= 30 °C days
TX35GE Extremely hot days Tmax >= 35 °C days
HEAT Heat waves (1st, 2nd, 3rd degree) Tavg >= 25 °C for 1 day / 3 days or >= 27 °C for 1 day / 3 days days
CCD Annual maximum number of consecutive dry days days when Rday < 1 mm days
CWD Annual maximum number of consecutive wet days days when Rday >= 1 mm days
RR0.1 Precipitation above threshold 0.1 mm Rday >= 0,1 mm days
RR1 Precipitation above threshold 1 mm Rday >= 1 mm days
RR5 Precipitation above threshold 5 mm Rday >= 5 mm days
RR10 Heavy precipitation days Rday >= 10 mm days
RR20 Extremely heavy precipitation. Rday >= 20 mm days
RX1 Annual maximum 1-day precipitation Max (Rday) in a year mm
RX5 Annual maximum 5-day precipitation Max (Rday i, i+1, i+2, i+3, i+4) mm
SDII Simple Daily Intensity Index Precipitation amount/rainy days (Rday >= 1 mm) mm/days
SPI Standardized Precipitation Index On annual or seasonal basis (R-Ravg)/spread numbers
CEI Climate Extreme Index Complex: it quantifies how an area is affected by extremes  


We have analysed various daily-resolution, gridded databases. For the Carpathian Basin the following data sets were examined: HUGRID, the latest ECA&D dataset and the two regional climate model (ALADIN-Climate and REMO) results for different periods. (Detailed description of these databases and models can be found in the related parts of this web page.) Similarly to the mean climate behaviour, the significance of changes (or significance of errors in the reference period) is also investigated for extreme indices (using different statistical tests, like Welch or t-test).

The Hungarian climate impact studies mainly focused on the extreme indices listed in Table 3, however, sometimes special needs should be met and fulfilled with the calculation of additional derived indices (since more and more impact assessments are computed on the basis of the RCM results).

In Fig. 1 and 2 some examples can be seen, for instance, Fig. 1 indicates the occurrence of extremely hot days (TX35GE), which affects at least 10% of the Hungarian territory as for the ALADIN-Climate model for different periods. This clearly shows that while for the control period (end of the 20th century) the TX35GE episodes over some parts of Hungary are rare (mostly in July and August), it becomes much more frequent for the first and second half of the 21st century (might occur any time from May to September). The amplitude of change between the past and near-future periods is much larger than it is the case for the near- and far-future ones. Fig. 2 represents the expected change (in %) of the Simple Daily Intensity Index (SDII) simulated by REMO for 2021-2050 with respect to 1961-1990 (areas with significant changes are indicated by dots). This result supports earlier finding of European projects based on coarser (50 km and lower) resolution model outputs: the daily precipitation intensity index mainly increases over Europe. According to the simulation results this means if rainfall occurs in the future its intensity will be higher than in the past. This intensification is around 10% for Hungary, however, it is only significant in the Southern part of the Great Hungarian Plains.

1. ábra

Figure 1: The frequency of extremely hot days affecting at least 10% of the Hungarian territory by the ALADIN-Climate
model for different periods (1961-1990, 2021-2050,
and 2071-2100, respectively)
The variability can be seen, however, the precise
annual frequencies cannot be read from the figure

2. ábra

Figure 2: Expected change (%) of Simple Daily Intensity Index simulated by REMO
for 2021-2050 with respect to 1961-1990 (dots indicate areas of significant change)