Droughts and Drought Management in the Czech Republic in a Changing Climate
Droughts have important negative impacts on human society and many of its key activities. Several studies warned of a growing risk of droughts in central Europe in past decades (see Brazdil et al. 2015a; Trnka et al. 2016 for discussions of these studies). The problem of droughts and their negative impacts may become more severe in the context of future climate change because of enhanced anthropogenic forcings (Eitzinger et al. 2013; Trnka et al. 2011, 2014).
Past and Present Droughts
Different data can be used to characterize past droughts over the territory of the Czech lands (known as the Czech Republic since 1993). Based on meteorological measurements for the 1961-2014 period (chosen because it is best covered by meteorological stations), we noted a trend for increased drought occurrence at most stations. These tendencies have been documented through a series of drought indices (Figure 22.1) as well as in estimated soil moisture anomalies (Figure 22.2). Particularly during the period from April to June, a marked increase in the number of days with insufficient water content in the main rooting zone was identified. This increase can be attributed mainly to rising air temperatures, global radiation, and vapor pressure deficits, combined with little change in total precipitation (Trnka et al. 2015). The progressive decrease of water reserves in the soil in May and June, which are the critical months for agricultural and forest production, is alarming. As Figure 22.2 illustrates, these changes are widespread and well pronounced. The increased depletion of soil moisture reserves accumulated over winter months that has been observed during the April to June period also explains the increased variability of soil moisture content in the summer months (Trnka et al. 2015). As a result of the generally lower soil water at the end of June, the July-September soil moisture becomes more dependent on summer rainfall, which is highly variable. Therefore, interannual variability of the soil moisture content has increased as well.
Applying data from several secular meteorological stations, Czech temperature and precipitation series can be used to calculate a monthly series of several drought indices since 1805, namely the Standardized Precipitation Evapotranspiration Index for 1 and 12 months (SPEI-1 and SPEI-12), Palmer Z-index (Z-index), and Palmer Drought Severity Index (PDSI). All these indices showed a significant trend toward increased dryness in spring. Indices representing the long-term anomaly of water balance (SPEI-12 and PDSI)
Number of months within the summer half-year (April-September) with positive (light gray- toward more wet conditions) and negative (dark gray-toward more dry conditions) significant trends (a = 0.05) over the territory of the Czech and Slovak republics and northern Austria for (a) self-calibrated PDSI and (b) 12-month scSPEI. Evaluation of trends/tendencies was carried out individually for each month in the 1961-2014 period.
Number of days with reduced soil moisture availability in 0-0.4 m topsoil layer in April-June, the critical period for vegetation, over the territory of the Czech Republic. Mean number of days for the (a) 1961-1990 and (b) 1991-2014 periods is presented together with the (c) difference between the periods.
showed the same tendency for the whole year and summer, and to a lesser extent for autumn. No conclusive drying trends were identified during the last two centuries for winter, and in some regions a trend toward increased wetness was found (Brazdil et al. 2015a). The drying trends in the April-June period have been driven particularly by major temperature increases, leading to higher potential evapotranspiration.
Instrumental observations before 1800 are scarce, but knowledge of droughts for the preinstrumental period can be obtained from documentary evidence (Brazdil et al. 2005, 2010). Extensive documentary evidence related to droughts and their impacts in the Czech lands allowed analysis of the occurrence and severity of dry episodes on an annual timescale.
This information can be combined with drought indices calculated for the instrumental period to create long-term decadal frequencies of droughts from 1501 to 2012 (Brazdil et al. 2013). Despite great interdecadal variability (Figure 22.3a), the highest frequency of years with dry episodes during 50-year periods occurred between 1951 and 2000 (26 years), followed by 1751-1800 (25), 1701-1750 (24), and 1801-1850 (24). The lowest rate of dry years was recorded for 1651-1700 (16) and 1551-1600 (19). More detailed evidence of long-term drought fluctuations in the Czech lands can be obtained from a series of four drought indices (SPI, SPEI, Z-index, and PDSI) derived on the seasonal, half-year, and annual levels from documentary and instrumental data for the 1501-2015 period (Brazdil et al. 2016a). As shown in Figure 22.3b and c, fluctuations in annual SPEI-12 and PDSI and demonstrates great interannual and interdecadal variability.
The available documentary data also provide convincing evidence about several extraordinary episodes of droughts, such as those in 1534, 1536, 1540 (classified as a year of unprecedented European heat and drought by Wetter et al. 2014), 1590, 1616, 1718, 1719, 1726, 1746, and 1790. Their list can be extended using instrumental records for droughts in 1808, 1809, 1811, 1826, 1834, 1842, 1863, 1868, 1904, 1911, 1917, 1921, 1947, 1953(-1954), 1959, 1992, 2000, 2003, 2007, 2012, and 2015. Reported dry episodes had significant impacts on the daily life of the population and, in many cases, led to significant increases in food prices, followed by the adoption of various emergency measures. The broad extent of various impacts, including economic, social, and political, was documented by Brazdil et al. (2016b) in an analysis of the disastrous drought in 1947, which also had a broader European context.
Overall results of multiple studies (e.g., Brazdil et al. 2015a, 2016a) conclusively show that despite relatively strong variability in drought frequency, a trend toward increasing drought intensity could be identified in recent decades (see Figure 22.3), which is also closely linked with changed frequency of the drought conductive circulation types over central Europe (e.g., Brazdil et al. 2015a; Trnka et al. 2009).
Because documentary data from the Czech lands before 1500 are rather sporadic, some potential for drought reconstruction is offered by tree-ring width (TRW) data, represented by annually resolved oak TRW chronology, covering 761-2010 (Dobrovolny et al. 2015). Despite the complicated relationship of TRW to drought in the central European scale, minimal TRW values may identify the occurrence of dry seasons. The existing chronology showed greater frequency of years with minimal increments of wood (sign of a growth depression potentially caused by drought) at the end of the ninth century, the turn of the twelfth and thirteenth centuries, the mid-seventeenth century, and the beginning of the nineteenth century. Conversely, a smaller number of years with low growth was typical for the end of the eleventh century, the second half of the fourteenth century, and the first half of the eighteenth century.
Long-term fluctuations of droughts in the Czech lands since 1501 based on documentary and instrumental data: (a) decadal frequencies of years with detected drought episodes, (b) annual SPEI-12, and (c) annual PDSI. Series in (b) and (c) are smoothed by Gaussian filter for 20 years. (Compiled from Brazdil et al. 2013, 2016a.)