Czech Quaternary Palynological Database ��� PALYCZ: review and basic statistics of the data ��esk�� kvart��rn�� pylov�� datab��ze ��� PALYCZ: p��ehled a z��kladn�� statistika Petr K u n e ��1,2, Vojt��ch A b r a h a m1, Oleg K o v �� �� �� k3, Martin K o p e c k ��1,4 & PA LY C Z c o n t r i b u t o r s55 1Department of Botany, Faculty of Science, Charles University in Prague, Ben��tsk�� 2, CZ-128 01 Praha 2, Czech Republic, e-mail: petr@kunes.net 2Department of Earth Sci- ences, University of Aarhus, C. F. M��llers All�� 4, 8000 Aarhus C, Denmark 3Department of Computer Science and Engineering, Faculty of Electrical Engineering, Czech Tech- nical University in Prague, Karlovo n��m��st�� 13, CZ-121 35 Prague 2, Czech Republic 4Department of Vegetation Ecology, Institute of Botany, Academy of Sciences of the Czech Republic, Po�������� 3b, CZ-603 00, Brno, Czech Republic Kune�� P., Abraham V., Kov������k O., Kopeck�� M. & PALYCZ contributors (2009): Czech Quaternary Palynological Database ��� PALYCZ: review and basic statistics of the data. ��� Preslia 81: 209���238. This paper reviews the data on quaternary palynological sequences collected in the Czech Republic, attempts to store them in the Czech Quaternary Palynological Database (PALYCZ) and outlines a pos- sible use for regional syntheses. Work on pollen stratigraphies done over the last hundred years has yielded a very large amount of data for this region. These data can be used globally for various types of environmental reconstructions and are of local importance, especially when combined with local data- bases. For data to be included in PALYCZ it has to meet certain criteria, the determination of the pol- len of herbaceous plants must be well resolved and radiocarbon dated. As of 31 December 2008, we had reviewed 177 pollen profiles. Data from 152 sequences are already stored in PostgreSQL�� in re- lational tables, which allow a broad range of queries to be addressed using the html protocol. The data collected since 1959 by 15 authors contain raw pollen counts together with 14 C dates and various metadata on locality. All the pollen samples were ordered using non-metric multidimensional scaling. Display of the ordination diagram incorporating the appropriate millennial time slices revealed a com- mon pattern in all data. The quality of data is also discussed in the context of the history of the research and methods used. Database access can be found at http://botany.natur.cuni.cz/palycz. Preslia 81: 209���238, 2009 209 5 Eva B����zov��, Czech Geological Survey, Kl��rov 131/3, CZ-118 21 Praha 1, Czech Republic, email: eva.brizova@geology.cz Lydie Dudov��, Institute of Botany, Academy of Sciences of the Czech Republic, Po�������� 3b, CZ-603 00 Brno, Czech Republic, e-mail: cvilinek@gmail.com Vlasta Jankovsk��, Institute of Botany, Academy of Sciences of the Czech Republic, Po�������� 3b, CZ-603 00 Brno, Czech Republic, e-mail: jankovska@brno.cas.cz Maria Knipping, Institute of Botany, University of Hohenheim, DE-70593 Stuttgart, Germany, e-mail: knipping@uni-hohenheim.de Radka Koz��kov��, Institute of Archaeology, Academy of Sciences of the Czech Republic, Letensk�� 4, CZ-118 01 Praha 1, Czech Republic, e-mail: r.kozakova@seznam.cz Kate��ina Nov��kov��, Laboratory of Archeobotany and Palaeoecology, University of South Bohemia, Brani��ovsk�� 31, CZ-370 05 ��esk�� Bud��jovice, Czech Republic, e-mail: kacenka@prf.jcu.cz Libor Petr, Department of Botany, Faculty of Science, Charles University in Prague, Ben��tsk�� 2, CZ-128 01 Praha 2, Czech Republic, e-mail: liborpetr@atlas.cz Petr Pokorn��, Institute of Archaeology, Academy of Sciences of the Czech Republic, Letensk�� 4, CZ-118 01 Praha 1, Czech Republic, e-mail: pokorny@arup.cas.cz Alena Roszkov��, Department of Geological Sciences, Masaryk University, Kotl����sk�� 2, CZ-611 37 Brno, Czech Republic, e-mail: roszkova@email.cz Eli��ka Rybn����kov��, Institute of Botany, Academy of Sciences of the Czech Republic, Po�������� 3b, CZ-603 00 Brno, Czech Republic, e-mail: e.rybnickova@seznam.cz Helena Svobodov��-Svitavsk��, Institute of Botany, Academy of Sciences of the Czech Republic, CZ-252 43 Pr��honice, Czech Republic, e-mail: svitavska@ibot.cas.cz Agnieszka Wacnik, W. Szafer Institute of Botany, Polish Academy of Sciences, Lubicz 46, PL-31-512 Krak��w, Poland, e-mail: a.wacnik@botany.pl

K e y w o r d s : data archive, Czech Republic, Holocene, multivariate analysis, pollen analysis, Pleistocene Introduction Research on pollen stratigraphies had quite a long tradition in the former Czechoslovakia. The adoption of a more rigorous determination of time and taxonomy in pollen analysis resulted in the data being frequently used for reconstructing changes in the postglacial en- vironment. Recently, more scientists addressing specific questions have requested com- parisons of pollen analytical data from several stratigraphies (Pokorn�� 2002b, Pokorn�� 2004, Kune�� et al. 2008). At a continental scale, information on past vegetation has been used to answer questions about past climate change (Davis et al. 2003), the spread and distribution of woody species (e.g., Magri 2008) and potential future conservation of the environment (Anderson et al. 2006). Many studies could benefit from the European Pollen Database (EPD), where around 40 datapoints from the Czech Republic are already archived. These datapoints originated mainly from pollen sequences published in the 1970s and 1980s and some of them are core localities with well-established chronologies (e.g., Jankovsk�� 1987, Rybn����kov�� & Rybn����ek 1988a). However, during the last 15 years, Czech palynologists have analyzed and dated many new pollen sequences of high importance. Some of these sequences were pub- lished in international journals and therefore are well-known and readily available to authors (Pokorn�� 2002a, Svobodov�� et al. 2002, Pokorn�� et al. 2006, Rybn����kov�� & Rybn����ek 2006). Others are, unfortunately, published in local journals and sometimes in local lan- guages (Jankovsk�� 1998, Svobodov�� 2004) others remain unpublished (Appendix 1). Although global questions require integrated datasets, the existence of local and re- gional databases has advantages: database managers can benefit from their familiarity with the area of the Czech Republic they know most of the researchers personally, the his- torical background and taxonomic concepts utilized by individual researchers. This auto- matically results in a high-level of accuracy of the data, which can be immediately checked, and database managers can easily track current research and encourage authors to submit their data, with communication occuring at a personal level. Electronic databases enhance knowledge by providing large collections of information, which can be used in wider syntheses of data. In the region of Central Europe, there are several examples, ALPADABA (Bern), Polish Pollen Database (Ralska-Jasiewiczowa et al. 2004) and the Czech National Phytosociological Database (Chytr�� & Rafajov�� 2003). This encouraged us to compile a computer-based database of pollen stratigraphical data, which will provide: (i) an archive of raw data (pollen counts) and metadata (ii) statis- tics for regional syntheses (iii) support for the EPD and (iv) a possible link with other lo- cal databases (archaeological database, modern pollen database, archaeobotanical data- base of plant macrofossils, phytosociological database), all of which would be a great con- tribution to future research (Fig. 1). The aim of this paper is to review data from the Czech Quaternary Palynological Data- base (PALYCZ), which consists of pollen sequences analysed during the last 50 years in the Czech Republic, and indicate the basic statistical outputs and possibilities for further analyses. In addition, data analysed by Czech and Polish palynologists in Slovakia and a few sites near the border in Germany are also included into PALYCZ. Many researchers will benefit from this data in the future. 210 Preslia 81: 209���238, 2009

PALYCZ from a historical perspective Collecting data from palynological sequences had a long tradition in the former Czecho- slovakia. Although the first investigation analysing an inventory of peat-bogs and a few macrofossils was published by Franti��ek Ladislav Sitensk�� (1885, 1886, 1891), the real beginning of quaternary palynology is connected with Karl Rudolph (born 11. 4. 1881 in Teplice), who worked at the German University in Prague. He was inspired by attending the ���IV. Internationale Pflanzengeographische Exkursion��� in 1916 in Scandinavia, where he explored northern-European vegetation and met L. von Post, the leading pollen analyst at that time. Investigations first led K. Rudolph to the T��ebo�� Basin (S Bohemia) and the first publications of data for ��irok�� blato, P����braz and Mirochov (Rudolph 1917). Franz Firbas, Rudolph���s first co-worker, focused on the Plou��nice region (Polzengebiet) in N Bohemia (Firbas 1927 see Fig. 2), where he analysed 25 sites. His students and colleagues continued research in the Jizersk�� hory Mts (Plail 1927), Cheb Basin (Funeck 1931) and Orlick�� hory Mts (M��ller 1929) so intensively that by 1929 Rudolph could publish a re- view article summarizing these results (Rudolph 1929). Researchers from Rudolph���s school also collected data outside the borders of Czechoslovakia ��� in the Pannonian Basin (Kinzler 1936), N Tatra Mts (Peterschilka 1927) and other areas (see Firbas 1949, 1952). The palynological workgroup educated many good students. Nevertheless, the outbreak of World War II inhibited further expansion of the group. Some of the students were killed (like Karl Preis 1941 in Russia), while others were expelled from Czechoslovakia in 1945. Franz Firbas re-established his group at the University in G��ttingen. In 1988, Hans Schmeidl, the last student of K. Rudolph, was still lecturing on vegetation history in Mu- nich. After World War II, Hubert Losert (who worked at Komo��ansk�� lezero lake and in the Elbe Basin) and Hugo Salaschek (who worked on Moravo-Silesian peat-bogs) did not return to palynology but both became secondary school teachers. However, they analysed and published an outstanding number of profiles, which are still a great inspiration for modern palaeoecology (Fig. 2). Kune�� et al.: Czech Quaternary Palynological Database 211 Fig. 1. ��� Schematic diagram of possible interactions between PALYCZ and other databases or datasets. Dashed indicates in progress.

To this generation of German palynologists we can also add a few scientists that worked at the Czech Agricultural University (Kle��ka 1926a, b) and the Czech part of Charles University (Puchmajerov�� 1929, ��t��p��nov�� 1930). The last-mentioned author was active until 1950. Pacltov�� (1957), Kriesl (1959) and Kri��o (1958) studied pollen analyses relevant to forestry, while Kneblov�� (1956) focused on geological questions. Opravil (1959) began with palynological studies in Keprn��k-Jesen��ky, but later switched to archaeobotanical macrofossil analyses. A comprehensive overview of all the data from this early period of research in the Czech Republic is illustrated in Fig. 2. Even though much of the primary data from this pe- riod are available in publications they are not included in the PALYCZ for reasons de- scribed below. 212 Preslia 81: 209���238, 2009 Fig. 2. ��� Map of palynological profiles not meeting the required standard of data quality. Black points indicate sites for which the results were included in the database only after the sites were revisited. References to numbers: 1 ��� Fahl 1926 2 ��� Fejfar et al. 1955 3 ��� Firbas 1927 4 ��� Firbas 1929 5 ��� Firbas & Losert 1949 6 ��� Funeck 1931 7 ��� Gough 1992 8 ��� Granzner 1936 9 ��� M. Kaplan, unpublished 10 ��� Kern 1939���1940 11 ��� Kle��ka 1926a 12 ��� Kle��ka 1926b 13 ��� Kle��ka 1928 14 ��� Kneblov��-Vodi��kov�� 1966a 15 ��� Kneblov�� 1956a 16 ��� Koz��kov�� & Kaplan 2006 17 ��� Kral 1979 18 ��� Kriesl 1959 19 ��� Kri��o 1958 20 ��� Losert 1940a 21 ��� Losert 1940b 22 ��� Losert 1940c 23 ��� Mr��z & Pacltov�� 1956 24 ��� M��ller 1927 25 ��� M��ller 1929 26 ��� N��mejc & Pacltov�� 1956 27 ��� Opravil 1959 28 ��� Opravil 1962 29 ��� Pacltov�� 1957 30 ��� Pacltov�� & Huben�� 1994 31 ��� Pacltov�� & ��pinar 1958 32 ��� Plail 1927 33 ��� Puchmajerov�� 1929 34 ��� Puchmajerov�� 1936 35 ��� Puchmajerov�� 1943 36 ��� Puchmajerov�� 1944 37 ��� Puchmajerov�� 1945 38 ��� Puchmajerov�� 1947a 39 ��� Puchmajerov�� 1947b 40 ��� Puchmajerov�� 1950 41 ��� Puchmajerov�� & Jankovsk�� 1978 42 ��� Purkyn�� & Rudolph 1925 43 ��� Purkyn�� & Rudolph 1927, 44 ��� K. Rudolph, unpublished 45 ��� Rudolph 1917 46 ��� Rudolph 1926 47 ��� Rudolph 1931 48 ��� Rudolph & Firbas 1924 49 ��� Rudolph & Firbas 1927 50 ��� Salaschek 1936 51 ��� Schmeidl 1940 52 ��� Sl��dkov��-Hynkov�� 1974 53 ��� Stark & Overbeck 1929 54 ��� ��t��p��nov�� 1930 55 ��� Vodi��kov�� 1981 56 ��� W��nsch 1935 57 ��� W��nsch 1939 58 ��� ��ebera 1957.

The modern palynological approach, which uses the determination of herb palynomorphs and 14C dating, was founded at the Institute of Botany of the Academy of Sciences in Brno by E. Rybn����kov��. Since the 1960s, several palynologists have gone through this institute (M. Peichlov��, A. Kon��topsk��, H. Sl��dkov��-Hynkov��, H. H��tte- mann and many others), including V. Jankovsk�� (still active there) and H. Svobodov��- Svitavsk��, who entered in the 1980s and subsequently moved to the Institute of Botany at Pr��honice. Associated were quaternary palynologists among the geologists in Prague, namely V. Kneblov��-Vodi��kov�� and E. B����zov��, who is currently working at the Czech Geological Survey in Prague. During the 1970s and early 1980s the group established an internationally recognized palynological school in Central Europe, which is documented by a number of foreign exchanges and cooperative studies. For Slovakia E. Krippel pub- lished a comprehensive study of postglacial development of vegetation in that area (Krippel 1986). A summary of the major interactions and developments during the past century is shown in Fig. 3. The first attempts to establish a Holocene pollen database for former Czechoslovakia were made by E. Rybn����kov�� and K. Rybn����ek based on isopollen maps (Rybn����kov�� & Rybn����ek 1988b, Rybn����ek & Rybn����kov�� 1994) and profited from the wide synthesis published by Rybn����kov�� (1985). Then, an advanced pollen database was created by Pokorn�� (2002b, 2004) however, much of the data were extracted directly from pollen di- agrams by recalculating scanned pollen curves. Therefore, we decided to collect all the pollen data for the Czech Republic and store them in a unified database. Data collection, database structure and nomenclature in PALYCZ PALYCZ contains data from quaternary pollen sequences from the Czech Republic that were mainly analysed after the late 1950s. However, data are not included if: (i) the author is deceased and the data are lost, (ii) for some reason the data do not match pre-defined cri- teria (see the section Assessment of data quality) or (iii) they are is still undetermined for technical or other reasons proposed by the author. Additionally, we included into PALYCZ data from Slovakia collected by Czech and Polish palynologists and few sequences from bordering areas in Germany. All the pollen sequences were obtained directly from the au- thors or from original publications and where possible the raw pollen counts were stored with metadata from the locality. Metadata for each profile consists of author, a description of the locality (including geographic coordinates), type of sediment, radiocarbon dates, etc. A complete list of pollen profiles included in the database (as of 31. 12. 2008) is avail- able in Appendix 1 and their geographical distribution is shown in Fig. 4. The PostgreSQL�� database software was used to store the data. This is an open source application, which offers easy implementation into html protocol. The structure of the da- tabase follows the EPD structure. Data are stored in relational tables to allow for a very broad range of queries, which may provide answers to very specific scientific questions. PALYCZ maintains original taxa names used by each pollen analyst. These names are then linked to two alternative nomenclatures: PALYCZ taxa, which is based on ALPADABA and includes details of the taxa designated by the original authors, and Beug taxa (Beug 2004), which merge some groups. Kune�� et al.: Czech Quaternary Palynological Database 213