Long-term changes in drought indices in the Eastern Europe Jaak Jaagus Department of Geography, University of Tartu, Estonia Lisbon, 18 June 2019

Objective To analyse trends in long time series of drought indices in the Eastern European countries (in the meridional belt between the Baltic and Black Seas) during 1949-2018 based on station data To prepare a joint article on that topic (for publication in the International Journal of Climatology)

Initial data Monthly precipitation, mean wind speed, relative humidity, solar radiation, sea-level pressure, and monthly mean daily maximum and minimum temperatures SPI and SPEI using the Hargreaves and Penman methods for calculation of PET SPI1, SPI3, SPI6, SPEI1, SPEI3, SPEI6 Monthly, annual and seasonal mean values

Stations from different countries Estonia: 14 stations (7 with the Penman method) Lithuania: 15 stations (4 with the Penman method) Belarus: 13 stations (8 with the Penman method) Ukraine: 8 stations (from the ECAD dataset) Moldova: 5 stations Poland: 36 stations (11 with the Penman method) Czech Republic: 12 stations (all with the Penman method) Latvia: 11 stations (4 with the Penman method)

Partners and co-authors Estonia: Anto Aasa, Kiira Mõisja Lithuania: Egidijus Rimkus, Edvinas Stonevičius Belarus: Irina Danilovich, Viktar Melnik Ukraine: Inna Semenova Moldova: Vera Potopova, Boris Boincean Poland: Joanna Wibig Czech Republic: Ladislava Rezničkova, Mirek Trnka, Petr Stepanek Latvia: Agrita Briede, Svetlana Aniškeviča

Partners and co-authors Spain: Sergio Vicente-Serrano, Fernando Domínguez Castro Slovakia: Livia Labudkova Hungary: Rita Pongracz Romania: Roxana Bojariu

Methods Calculating of monthly SPI and SPEI values using the special program in R SPEI was found using two methods for the calculation of PET: Penman-Monteith and Hargraves Trend analysis of the time series using linear regression and Student’s t-test Mann-Kendall test for trends, Sen’s method for calculating the slope Maps of trend values Grouping of stations with similar variations and trends using PCA (regionalization)

Results Generally no large changes in SPI and SPEI In the northern part of the region (Estonia, Latvia?) upward trends during the cold half-year and in June Weak drying tendency in April and September In Lithuania, similar trends in winter, no change in June, increase in some stations in July Upward trends in winter and downward tendency in April, May and August in Belarus Increase more in northern and decrease in southern Belarus Stronger negative trends in case of SPEI than SPI

SPI1 in Ristna in June

SPEI1 (Hargreaves) in Kybartai in April

SPEI1 (Hargreaves) in August

Results In Ukraine, few significant trends were detected Downward tendency, i.e. drying is prevailing in July and especially in August. Same pattern in Moldova Large variety oftrends in drought indices in Poland Mostly upward trends in March, and downward trends in April and August Prevailing of moistening tendencies in Northern Poland (Elblag, Bialystok, Olsztyn) and of drying tendencies in Southern Poland (Legnica, Opole, Bielsko Biala) Different trends in mountains (Sniezka, Zakopane, Kasprowy Wiech)

SPEI3 in Legnica for June

SPEI3 (Penman) for August

Results Drying tendencies have dominated in the Czech Republic from April til August, especially in the cases of SPI3, SPI6, SPEI3 and SPEI 6 There have been the strongest negative trends like in southern Poland

SPEI3 (Penman) for September in Hradec Kralove

Conclusions Drought indices are, first of all, determined by precipitation amounts, trends in precipitation are reflected in drought indices Time series of SPEI values show stronger decrease than that of SPI values due to taking into account the increase in PET that is caused by higher temperatures In the northern part of the study area an upward tendency in the indices revealed in the cold half-year In the southern part, a downward tendency is prevailing during the warm half-year Trend values vary remarkably between the stations