printed by

www.postersession.comThis study was conducted with support of project NATO Science Program no. EST-CLG 979505.

DROUGHT EVENT PROBABILITY IN THE CZECH REPUBLIC UNDER THE

PRESENT AND CHANGED CLIMATIC CONDITIONSMiroslav Trnka1), Daniela Semerádová1), Martin Dubrovský2), Zdenek Žalud1), Mark Svoboda3), Mike Hayes3), Donald Wilhite3)

1Institute for Agrosystems and Bioclimatology, Mendel University of Agriculture and Forestry Brno, Czech Republic, mirek_trnka@yahoo.com2Institute of Atmospheric Physics, Academy of Sciences of the Czech Republic, Prague, Czech Republic3National Drought Mitigation Center, School of Natural Resources, University of Nebraska, Lincoln, USA

References Dubrovský,M.1997. Creating Daily Weather Series With Use of the Weather Generator. Environmetrics 8, 409-424

Harvey LDD, Gregory J, Hoffert M, Jain A, Lal M, Leemans R, Raper SBC, Wigley TML, de Wolde J (1997) Anintroduction to simple climate models used in the IPCC Second Assessment Report: IPCC Technical Paper 2 (edsJT Houghton, LG Meira Filho, DJ Griggs, M Noguer), Intergovernmental Panel on Climate Change, Geneva,Switzerland, pp.50

Hulme M, Wigley TML, Barrow EM, Raper SCB, Centella, Smith S, Chipanshi A.C. (2000) Using a climatescenario generator for vulnerability and adaptation assessments: MAGICC and SCENGEN Version 2.4 Workbook.Climatic Research Unit, Norwich, UK, 2000, 52 pp.

Santer BD, Wigley TML, Schlesinger ME, Mitchell JFB (1990) Developing climate scenarios from equilibrium GCMresults. Max Planck Institute für Meteorologie, Report No.47, Hamburg, Germany

Soil Survey Staff. 1975. Soil Taxonomy. A basic system of soil classification for making and interpreting soilsurveys. USDA Soil Conservation Service, Agric. Handbook No. 436. US Gov’t Printing Office, Washington, D.C.Svoboda, M.D., D. LeComte, M.J. Hayes, R. Heim, K. Gleason, J. Angel, B. Rippey, R. Tinker, M. Palecki, D.Stooksbury, D. Miskus, and S. Stevens, 2002. The Drought Monitor. Bulletin of the Amer. Meteorol. Society 83: 8,1181-1190

Van Wambeke, A. 1981. Calculated soil moisture and temperature regimes of South America. Soil ManagementSupport Services Technical Monograph No. 2, USDA-SCS, Washington, D.C.

Van Wambeke, A., P. Hastings, and M. Tolomeo. 1992. Newhall Simulation Model—A BASIC Program for the IBMPC (DOS 2.0 or later). Dept. of Agronomy, Cornell University, Ithaca, NY.

Waltman, W.J., Goddard, S. Reichenbach, S.E. Svoboda, M.D. Hayes, M.J., Peake, J.S. Patterns and trends ofsoil climate regimes and drought events in the northern Great Plains, 2003

Wilhite, D.A. (ed.), 2000. Drought: A Global Assessment (2 volumes, 51 chapters, 700 pages). Hazards andDisasters: A Series of Definitive Major Works (7 volume series), edited by A.Z. Keller. Routledge Publishers.

Figure 1: The spatial distribution of the fourmain categories of the maximum soil waterholding capacity in the Digital LandscapeModel.

Figure 2: The spatial distribution of 50 meteorol.stations used as the source of weather data inthe presented study and representation ofaltitude data available to the DLM.

Figure 3: The spatial representation of the probability of occurrence of the aridic and xeric eventsover the area of the Czech Republic under present climatic conditions (1961-2000).

Figure 4: The spatial representation of the probability of occurrence of the aridic and xericevents over the area of the Czech Republic under expected climatic conditions (2050) assumingB1-SRES emission scenario.

Figure 5: The spatial representation of the probability of occurrence of the aridic and xeric eventsover the area of the Czech Republic under expected climatic conditions (2050) assuming A2-SRES emission scenario.

Drought is the most complex and least understood of all natural hazards, affecting more people than anyother hazard (Wilhite Ed., 2000). Therefore, identification of drought risk associated with the present aswell as with expected climatic conditions remains an important issue. There are a number of parametersthat can be used for describing drought events over various time scales e.g. Standardized Precipitationindex or the Palmer Drought Severity Index. The use of soil moisture as one of the indicators of thedrought (especially in relation to projected agricultural impacts) is another essential part of many droughtmonitoring systems (e.g. Svoboda et al., 2002). One of the tools capable of describing soil moistureregimes and identifying soil climate categories in accordance with the USDA soil taxonomy classificationscheme (USDA, 1999) is the Newhall simulation model (Van Wambeke et al., 1992). We applied the modelin the presented study in order to quantify the potential impact of expected climate change on theprobability of drought occurrence in the Czech Republic and to identify the most threatened regions andtheir spatial extent.

• The Figure 3 documents that under present climatic conditions the majority of the country has a chance ofexperiencing an arid or xeric event at least once per century. On the other hand, only a fraction of theterritory is situated within the “high” risk area (aridic or xeric events probability over 45 %). The high droughtrisk is confined to the well-known regions of the country (i.e. South Moravia and the rain shadow of the“Krušné hory” mountain chain). Under present conditions weak and typic aridic regimes represent 15% to23% of the annual soil moisture regimes in these areas.

• While under the present climate the only two small distinct drought prone areas might distinguished (orangeareas at Figure 3) the expected change of climate conditions during this century will likely lead to significantincrease of drought risk. Even if greenhouse gases emissions are kept low and subsequent climaticchanges are relatively minor a significant increase of areas experiencing high probability (more than 45 %)of aridic or xeric events is to be expected (Figure 4). The extent of the high risk area will grow more than 16-20 times compared to the present climate. At the same time about 0.1% of the Czech Republic mightexperience aridic or xeric soil moisture regime during 6 out of 10 years. Consequently the probability of theudic or wet tempustic soil moisture regimes is close to zero in such areas. It should also be noted that areaswith the chance of aridic or xeric events probability between 15-45% will also grow substantially.

• If greenhouse gas emissions continue to grow (as the scenario A2 indicates) then the changes to the soilclimate in the Czech Republic will be enormous (Figure 5). Depending on the GCM model applied, up to6.3-8.1% of the country area will face aridic or xeric events occurring up to six years out of ten

• The Newhall Simulation Model enables one to model the annual soil moisture regime and allows fordevelopment of frequencies and probabilities of soil moisture regimes. At the same time it enables us toidentify major drought and wet cycles both within the year or entire climate period. The most recent use andinnovation of the model has been done by (Waltman et al., 2003) who developed the Enhanced NewhallSimulation model (ENSM). One of the main advantages of the ENMS is its flexibility and the fact that itallows the user to include the influence of physical soil properties into the calculations. The soil profile isdescribed in terms of the maximum water holding capacity of the root zone (MWHC). In order to input thisparameter for the whole country and to interconnect the ENSM with GIS ArcMap software the DigitalLandscape Model (DLM) was developed at the Institute of Agrosystems and Bioclimatology (MZLU). TheDLM model uses a 1x1 km basic grid area and stores information (e.g. on the soil type, altitude, longitudeand latitude) for each grid. The DLM model was modified for the purposes of this study when the predefinedvalue of the MWHC was assigned to each grid. The rooting depth of the soil profile as well as its MWHCwere derived from a representative sample of over 1000 detailed soil pits performed during the ComplexSoil Survey of the Czech Republic. The overview of the soil grops that are distinguished in the presentDLM version and the assigned MWHC values in mm for each group is presented at Fig. 1. The probabilityof two of six major soil moisture regimes occurrence recognized by USDA soil classification (USDA, 1999)was considered in this study (i.e. Xeric and Aridic soil moisture regimes).

• The climate change scenarios applied in this paper are based on two GCMs (ECHAM4 and HadCM3) usingSRES-A2 emission scenario. The GCM output were taken from the IPCC-DDC database (http://ipcc-ddc.cru.uea.ac.uk). The climate change scenarios were constructed using a pattern scaling technique(Santer et al., 1990): the scenario is derived as a product of the standardized scenario and the change ofthe global mean temperature. Changes in the global mean temperature (∆TG) for the evaluation periodcentered around the year 2050 were calculated by a simple climate model MAGICC (Harvey et al. 1997,Hulme et al. 2000) assuming two combinations of an emission scenarios and climate sensitivity. The lowerestimate of ∆TG is based on a low climate sensitivity (∆TG,2×CO2 = 1.5 K) and SRES-B1 emission scenario.The upper estimate of ∆TG is based on a high climate sensitivity (∆TG,2×CO2 = 4.5 K) and SRES-A2 emissionscenario. The two versions of the climate change scenarios related to each GCM will be referred to as B1and A2.

• In order to estimate probability of occurrence of predefined soil moisture regimes over the Czech Republica database containing parameters derived from the weather generator Met&Roll (Dubrovský, 1997) on 50weather stations was used (Fig.2). These parameters have been derived from observed weather series at43 stations based on the 1961-2000 period while the remaining 7 stations a 1961-1990 record. In the nextstep we applied the ENSM at each of the 50 stations using the generated series of 100 years (for both thepresent and changed climatic conditions). The simulation was repeated for each of the 4 categories ofMWHC (i.e. 140, 180, 220 and 260 mm) at each station (Fig.1). Then we calculated the probability ofoccurrence of Xeric and Aridic years for all stations and for each value of MWHC. Multiple linear regressionmodels were consequently fitted to each data set containing altitude, northing and easting (as well as theirfirst order interactions and squares) as independent variables, with the probability of xeric and aridic annualsoil moisture regimes serving as the dependent variable.

The Newhall simulation model that allows for determining the probability of occurrence of a particulargroup of soil climate conditions was tested over the Czech Republic. It has been found that under thepresent climate only a fraction of the territory is situated within the “high” risk area with an aridic or xericevents probability over 45 %. The high drought risk areas are confined to two well-known dry regions ofthe country i.e. South Moravia and the rain shadow of “Krušné hory” mountain chain. When the modelwas run using the climatic data corresponding with the conditions expected under climate changearound 2050 (assuming B1 emission scenario) a significant increase of areas having a high probabilityof aridic or xeric events was found. At the same time, the probability of “wet” events is close to zero inthese areas, which may lead to changes in the soil processes and toward an increased risk ofsalinization. When the model was run using data assuming an increase of greenhouse gasesemissions according to the A2 scenario, shifts in the soil climate were found to be enormous.Approximately 6.3-8.1% of the country area will face aridic or xeric events in at least six out of ten yearsby 2050 with a high probability of weak aridic and even typic aridic events. It should also be stressedthat these areas belong to the prime agricultural regions of the country. At the same time, theprobability of aridic and xeric events is expected to increase significantly in the highlands, which aregenerally not adapted at all to such conditions. As the aridic or xeric soil climate regimes are closelyrelated to other drought impacts, such as decrease of crop yiselds, damage to forest stands, lowstream flow and reservoir levels etc. a significant increase of drought related losses is to be expectedunder the climate change scenarios.