Dust flux estimates for the Last Glacial Period in East Central Europe based on terrestrial records of loess deposits: a review more

Quaternary Science Reviews 29 (2010) 3157e3166 Contents lists available at ScienceDirect Quaternary Science Reviews journal homepage: www.elsevier.com/locate/quascirev Dust flux estimates for the Last Glacial Period in East Central Europe based on terrestrial records of loess deposits: a review Gábor Újvári a, *, János Kovács b, György Varga b, Béla Raucsik c, Slobodan B. Markovi d c a Geodetic and Geophysical Research Institute of the Hungarian Academy of Sciences, Csatkai Endre u. 6-8., H-9400 Sopron, Hungary Department of Geology, University of Pécs, Ifjúság u. 6., H-7624 Pécs, Hungary c Department of Earth and Environmental Sciences, University of Pannonia, Egyetem u. 10., H-8200 Veszprém, Hungary d Department of Physical Geography, Faculty of Sciences, University of Novi Sad, Trg Dositeja Obradovia 3, 21000 Novi Sad, Serbia c b a r t i c l e i n f o Article history: Received 26 March 2010 Received in revised form 12 June 2010 Accepted 2 July 2010 a b s t r a c t Loessepaleosol sequences are significant records of the mineral dust cycle of glacialeinterglacial periods. As dust particles give rise to direct and indirect radiative forcing, obtaining a reliable picture of the global and regional patterns of mineral dust fluxes during glacial periods can lead to a better understanding of the contribution of mineral dust to past climate changes. Recent progress in absolute dating of loess deposits in the Carpathian Basin in East Central Europe made it possible to provide correct aeolian flux estimates for the Last Glacial period, marine isotope stage (MIS) 2. Mass accumulation rates (MARs) from chronological data of 33 loess sites exhibited a wide range of values, from 150 to 1422 g mÀ2 aÀ1, centered around median and mean values of 338 and 417 g mÀ2 aÀ1. MARPM10 and MARPM2 estimates have been also calculated using grain size measurements of many loess samples and loess MARs in order to facilitate comparison with models, and since particles larger than 10 mm have a negligible radiative effect. Here we show that some previous model simulations of the dust cycle at the Last Glacial Maximum (LGM) significantly underestimated the real aeolian flux (ranges of our estimates: MARPM10 ¼ 34 À 324 and MARPM2 ¼ 9:3 À 88:2 g mÀ2 aÀ1 ) in East Central Europe. For this reason, some simulations of dustinduced direct radiative forcing of the LGM climate failed to yield reliable results for this mid-latitude region as they have been based on three-dimensional dust field models that are not capable of estimating the real aeolian fluxes in Central Europe. A recent global model of top of the atmosphere (TOA) radiative forcing by mineral aerosols at the LGM that has been based on more realistic parameterization of dust sources, transport, and deposition revealed zonally averaged surface cooling of À2  C for the latitudes of our study area. This surface cooling and TOA radiative forcing (À2 to À3 W mÀ2) are greater than recognized in other models and draws our attention to the importance of further modeling the impact of mineral dust on LGM climate in order to gain insight into the spatial pattern of radiative forcing and better understand resulting climate response in mid-latitude loess regions such as East Central Europe. Ó 2010 Elsevier Ltd. All rights reserved. 1. Introduction Mineral dust particles exert direct and indirect influence on global climate through diverse physical and biogeochemical processes and thus dust is considered to be an active component of the climate system (e.g. Arimoto, 2001; Harrison et al., 2001; Tegen, 2003). Dust affects the radiation budget through scattering and absorption of incoming solar and outgoing infrared radiation (e.g. Coakley et al., 1983; Tegen and Lacis, 1996; Liao and Seinfeld, 1998; Buseck and Pósfai, 1999; Satheesh and Krishna Moorthy, 2005), and * Corresponding author. Fax: þ36 99 508 355. E-mail address: ujvari@ggki.hu (G. Újvári). 0277-3791/$ e see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.quascirev.2010.07.005 by altering cloud optical properties, amounts and lifetimes through acting upon the number and size of cloud condensation nuclei (CCN) and ice nuclei (IN) (e.g. Albrecht, 1989; Rosenfeld et al., 2001; Yin et al., 2002; Fan et al., 2005; Andreae and Rosenfeld, 2008). Airborne mineral dust provides micronutrients (e.g. iron, silica) to marine and terrestrial ecosystems as well (e.g. Martin and Fitzwater, 1988; Martin et al., 1990; Duce and Tindale, 1991; Swap et al., 1992; Coale et al., 1996; Falkowski et al., 1998; Chadwick et al., 1999), thereby affecting productivity, influencing the carbon cycle, and eventually atmospheric greenhouse gas content (e.g. Martin, 1990; Archer et al., 1998; Broecker and Henderson, 1998; Mahowald et al., 2005; Maher et al., 2010). Clearly, greenhouse gases and their concentrations in the atmosphere exert an overriding influence on radiative forcing and thus climate. Mineral dust 3158 G. Újvári et al. / Quaternary Science Reviews 29 (2010) 3157e3166 also contributes to the radiative forcing, but to calculate whether it will cause surface warming or cooling is complicated as it depends on several factors such as dust particle size, mineralogy and mixing, its concentration and vertical distribution, the temperature and albedo of the underlying surface and the presence or absence of clouds (Tegen and Lacis, 1996; Tegen et al., 1996; Liao and Seinfeld, 1998; Claquin et al., 1999; Sokolik and Toon, 1999; Quijano et al., 2000; Andreae and Rosenfeld, 2008). In addition, since aerosol properties show large regional variations, the regional impact can be very different (Satheesh and Krishna Moorthy, 2005). Mineral dust loading during glacial periods was significantly higher than today. Increased dust is recognized in ice cores (Thompson and Mosley-Thompson, 1981; Petit et al., 1990; Steffensen, 1997; Lambert et al., 2008) and in loess deposits on continents (e.g. Porter and An, 1995; Kohfeld and Harrison, 2001; Derbyshire, 2003). Possible reasons for enhanced dust flux include increased wind speeds, reduced strength of the hydrological cycle, expansion of dust source areas, and the physiological effects of low atmospheric CO2 on terrestrial plant productivity (e.g. Yung et al., 1996; Andersen et al., 1998; Mahowald et al., 1999; Harrison et al., 2001; Tegen, 2003). Information on mass accumulation rates (MARs) calculated from accurately dated loess profiles can significantly contribute to the understanding of global and regional dust flux patterns during glacialeinterglacial periods. These calculations can also provide geological data to validate and refine paleo-dust cycle models (Kohfeld and Harrison, 2001; Derbyshire, 2003), and can help refine calculations of radiative forcing by ice-age atmospheric dust (Claquin et al., 2003). Previously, only three loess sections in the Carpathian Basin had calculated MARs and published (Frechen et al., 2003). As there has been considerable progress in absolute dating of loess deposits in Hungary, Croatia and Serbia in recent years, it is now possible to obtain more accurate calculations of MARs. In this paper, we provide a review of previously published MARs in the Carpathian Basin and release more accurate dust flux estimates for the Last Glacial period in the Carpathian Basin based on our own calculations (Fig. 1). We also present a short summary of provenance and mineralogy of loess deposits in the studied region. Finally, all of these data are discussed in the context of model simulation results to foster subsequent paleoclimate modeling efforts. 2. Methods Loess profiles at 33 sites in Hungary, Croatia and Serbia have been documented using published data (Table 1) and classified according to their geomorphological settings as specified in DIRTMAP 3.0 database documentation (Maher and Kohfeld, 2009). This is because diverse aeolian accumulation and sedimentation are expected under different physiographic situations. Four main geomorphological settings have been distinguished: loess plateau (12 sites out of 33), river terrace/river bank (5 sites), hill slope/crest (12 sites) and plain (4 sites). The focus of the present study was on sections comprising typical loess and sandy loess layers (primary aeolian loess), while alluvial (infusion) loess sequences have been omitted from the study. Of the 33 individual loess records, 21 sections had at least two or more ages from radiocarbon and/or optical (TL-OSL-IRSL) dating. MAR estimates based on these profiles were considered as highly reliable. Other sections had only one published age from radiocarbon or optical dating or they were correlated based on amino acid racemization (AAR), magnetic susceptibility (MS) and/or pedostratigraphyical results. MAR estimates from these profiles have been designated as moderate or low reliability. Age models of loess records having multiple absolute age data have been derived assuming linear sedimentation between tie-points (Fig. 2). Dates associated with an age reversal have been rejected. As the present study aimed at calculating dust flux for MIS 2 (Martinson et al., 1987), MAR values have been computed between 12 and 28 ka (Thompson and Goldstein, 2006). For moderate to low reliability sections, 28 and 12 ka were assigned to the initiation and termination of L1L1 loess formation, respectively. It must be mentioned, however, that loess accumulation began on some spots before 28 ka in the Carpathian Basin according to previous dating frameworks (e.g. Frechen et al., 1997; Sümegi, 2005; Antoine et al., 2009) suggesting a potential bias in estimating MARs for MIS 2 in the case of poorly dated sections. Another bias is associated with the assumption of linear sedimentation between tie-points. To create an age model which closely approximates the real progress of sedimentation one would require an unrealistic number of absolute age data leading to the inevitable practice of presuming linear sedimentation between tie-points. This approach averages out the effects of periods characterized by high dust input (dust storm events) or phases of low sedimentation implying that only an approximation of real loess MARs can be achieved. Before creating age models, uncalibrated radiocarbon dates were converted to calendar years (cal BP) using CALIB 6.0.1 software and IntCal09 calibration curve (Stuiver and Reimer, 1993; } Reimer et al., 2009). Radiocarbon dates from the Sütto section (site 1) have already been published as calendar years, converted by CalPal (CalPal2007_Hulu calibration curve) (Novothny et al., 2009) and these original data are used herein for generating an age model. Aeolian mass accumulation rates (MARs, g mÀ2 aÀ1) have been computed for each site using the equation MAR ¼ LSR  rdry  feol where LSR is the linear sedimentation rate (m aÀ1), rdry is the dry bulk density (g mÀ3), and feol is the sediment fraction that is aeolian in origin (Kohfeld and Harrison, 2001). If we assume that loess is entirely aeolian, then feol ¼ 1 for all of the calculations. As for dry bulk density, a wide range of data exists in the literature ranging from ca 1.3 to 1.7 g cmÀ3. Frechen et al. (2003) used 1.65 g cmÀ3 for European loess deposits, a value that originates from Pye (1987) and seems to be too high in the light of average bulk density data from the Chinese Loess Plateau (CLP, 1.48 g cmÀ3, Kohfeld and Harrison, 2001) or North America (1.45 g cmÀ3, Bettis et al., 2003; Muhs et al., 2003). Recent measurements of bulk density of loess deposits from the Last Glacial period at Dunaföldvár (20 km north of the Paks section, Hungary; Fig. 1), carried out in the course of a gravity study (Papp, 2009) using two different techniques (3 samples  2 methods, n ¼ 6 tests), yielded a mean rdry value of 1.497 Æ 0.079 g cmÀ3. Therefore we used a dry bulk density of 1.5 g cmÀ3 for our MAR calculations. As the radiative effect of particles larger than 10 mm in diameter can be considered as negligible in the atmosphere (Tegen and Lacis, 1996) and for the purpose of model-paleodata comparison different fine fractions of MAR (<10 and <2 mm) have been calculated as MARPMx ¼ LSR  rdry  fx where fx is the x ¼ <10 or <2 mm fraction of the total aeolian mass. Particle size analysis of 51 Late Glacial loess samples collected across Hungary (along a 150 km transect at Mende, Paks, and at Majs 10 km north of the Zmajevac section) using Malvern Mastersizer S (n ¼ 43) and Fritsch Analysette 22 (n ¼ 8) laser analyzers revealed that the average proportion of the PM10 and PM2 fractions are 22.80 and 6.19%, respectively. These are the values used in the calculation of MARPM10 and MARPM2 estimates. 3. Results Sedimentation rates (SRs) varied between 0.10 and 0.95 mm aÀ1 with a median ð~Þ and mean ðxÞ of 0.23 and 0.28, respectively x (Table 2). Calculated mass accumulation rates (MARs) range from G. Újvári et al. / Quaternary Science Reviews 29 (2010) 3157e3166 3159 Fig. 1. Location of loess profiles in the Carpathian Basin considered in the paper (map modified from Csontos et al., 2002). Black arrows show inferred paleowind directions after Jámbor (2002), Markovi et al. (2008) and Bradák (2009). White and gray arrows denote simulated mean 10 m wind vectors in summer and winter for the Last Glacial Maximum c } (LGM) (Meyer and Kottmeier, 1989; Renssen et al., 2007). 1 ¼ Sütto, 2 ¼ Basaharc, 3 ¼ Mende, 4 ¼ Tápiósüly, 5 ¼ Albertirsa, 6 ¼ Bodrogkeresztúr I, 7 ¼ Csorgókút I, 8 ¼ Csorgókút II, 9 ¼ Tokaj (Patkó-quarry), 10 ¼ Tokaj (Kereszt Hill), 11 ¼ Debrecen (Alföldi brickyard), 12 ¼ Látókép, 13 ¼ Lakitelek, 14 ¼ Szeged-Öthalom, 15 ¼ Madaras, 16 ¼ Katymár, } 17 ¼ Dunaújváros, 18 ¼ Ságvár, 19 ¼ Paks, 20 ¼ Üveghuta, 21 ¼ Dunaszekcso, 22 ¼ Zmajevac, 23 ¼ Erdut, 24 ¼ Crvenka, 25 ¼ Susek, 26 ¼ Petrovaradin, 27 ¼ Irig, 28 ¼ Ruma, 29 ¼ Moorin, 30 ¼ Titel, 31 ¼ Stari Slankamen, 32 ¼ Surduk, and 33 ¼ Batajnica. D3 and D4 are loess regions of Smalley and Leach (1978). s 150 to 1422 g mÀ2 aÀ1 (~ ¼ 338 and x ¼ 417 g mÀ2 aÀ1 ). The dust x flux estimates determined in this study for the Paks, Albertirsa and Mende sites in Hungary are different from those published by Frechen et al. (2003), probably due to differences in methodology. In general, lower MARs are correlated with loess deposits accumulated on hill slopes and plains (e.g. Great Hungarian Plain) (~ ¼ 219 and x 246 g mÀ2 aÀ1, respectively), while higher MARs are correlated with plateau and river terrace/bank sites (~ ¼ 435 and 707 g mÀ2 aÀ1) x (Fig. 3). Lower dust fluxes (around 100e300 g mÀ2 aÀ1) can be detected along and beyond the Tisza River on the Great Hungarian Plain, on the Middle Bácska (Baka) region and in the surroundings c of the Fruka Gora Mountain in Serbia (Fig. 4). Several dust flux hot s spots, characterized by the highest aeolian flux, are located along or } close to the River Danube in Hungary (Sütto: 584, Mende: 761, } Dunaújváros: 1238, Paks: 1422, Dunaszekcso: 707 g mÀ2 aÀ1) and Croatia (Zmajevac: 437 g mÀ2 aÀ1) as well as on the confluence area of the Danube and Tisza rivers in Serbia (Surduk: 434 and Titel: 510 g mÀ2 aÀ1). MARPM10 estimates varied between 34 and 324 g mÀ2 aÀ1 with ~ a median ðxÞ and mean ðxÞ of 77 and 95 g mÀ2 aÀ1, respectively (Table 2; Fig. 4). MARPM2 values ranged from 9.3 to 88.2 g mÀ2 aÀ1, clustering around ~ ¼ 20:9 and x ¼ 25:9 g mÀ2 aÀ1 , respectively. x Obviously, these estimates showed an identical pattern in the Carpathian Basin with calculated bulk loess MARs as the fluxes of fine-grained fractions have been computed from bulk loess MAR estimates. 4. Discussion 4.1. Mineralogy and provenance of loess in the Carpathian Basin Constituents of the mineral assemblage of the Pleistocene “Young Loess Series” (MIS2-9) in Hungary are quartz (ca 20e50%), carbonate minerals (ca 10e40%), micas (10e15%), feldspars (5e15%), clay minerals (5e25%) and a minor proportion of heavy minerals such as magnetite, garnet, hornblende, rutile, zoisite, epidote, titanite, staurolite, kyanite, zircon, tourmaline, hyperstene and apatite (Codarcea, 1977; Szebényi, 1979; Pécsi-Donáth, 1985; Hum and Fényes, 1995; Hum, 2002). A similar mineralogical composition has been demonstrated by Raucsik (2009) from loess samples of the Last Glacial period at Paks in Hungary and by Kosti c and Proti (2000) from Serbian loess deposits. The clayemineral c assemblage of loess in Hungary and Serbia consists predominantly of illite and chlorite with minor proportions of smectite, kaolinite, mixed-layer illite/smectite, and rare mixed-layer chlorite/smectite (Gerei et al., 1979; Pécsi-Donáth, 1979, 1985; Hum and Fényes, 1995; Kosti and Proti, 2000; Markovi et al., 2004; Raucsik, 2009). c c c Generally, the dolomite content of Hungarian loess deposits is higher than their calcite content and the calcite/dolomite ratio is 1:1.5 or 1:2 (Gerei et al., 1979; Hum and Fényes, 1995) contrary to Serbian loess samples where the calcite/dolomite ratio is 2:1 or 3:1 (Kosti and Proti, 2000). Dolomite is found to be enriched in the c c 10e50 mm fraction of loess at Paks (Nemecz et al., 2000) and Hum and Fényes (1995) suggested that it is primary and thus can be traced back to the special composition of source rocks. Smalley and Leach (1978) did pioneer work in identifying the main sources of loess deposits in the Carpathian Basin. They distinguished two regions within the three countries (Hungary, Croatia and Serbia), the D3 and D4 regions. The quartz silt of loess in Hungary is thought to be derived from three main sources: (1) glacial material carried through the Moravian Depression by glacial floodwater (D3 region), (2) glacial material from the Alpine region entrained by the Danube (D3 region), and (3) weathering products of the Carpathian flysch carried by the Tisza River (D4 region). The role of fluvial entrainment and alluvial plains as a source of silt has been emphasized. Loess deposits along the Danube in Croatia 3160 Table 1 Site information. Site name Albertirsa Basaharc Bodrogkeresztúr I Csorgókút I Csorgókút II Debrecen (Alföldi brickyard) } Dunaszekcso Dunaújváros Katymár Lakitelek I Látókép Madaras Mende Paks Ságvár } Sütto Szeged-Öthalom I Tápiósüly Tokaj (Kereszt Hill II) Tokaj (Patkó-quarry) Üveghuta-2 borehole Erdut Zmajevac Batajnica Crvenka Irig Moorin s Petrovaradin Ruma Stari Slankamen Surduk Susek Titel Country Hungary Hungary Hungary Hungary Hungary Hungary Hungary Hungary Hungary Hungary Hungary Hungary Hungary Hungary Hungary Hungary Hungary Hungary Hungary Hungary Hungary Croatia Croatia Serbia Serbia Serbia Serbia Serbia Serbia Serbia Serbia Serbia Serbia G. Újvári et al. / Quaternary Science Reviews 29 (2010) 3157e3166 Latitude ( ) 47 150 4400 N 47 480 1400 N 48 070 3700 N 48 070 3100 N 48 070 3200 N 47 310 5000 N 46 040 1200 N 46 580 1600 N 46 010 0300 N 46 520 5500 N 47 330 3700 N 46 020 1400 N 45 250 3200 N 46 380 2500 N 46 490 4200 N 47 440 2400 N 46 160 5800 N 47 270 N 48 070 3300 N 48 070 2900 N 46 110 5900 N 45 310 3600 N 45 480 4400 N 44 550 2900 N 45 390 42”N 45 050 0400 N 45 170 N 45 160 N 45 000 3800 N 45 070 3300 N 45 040 N 45 130 N 45 120 N Longitude ( ) 19 380 0600 E 18 500 3400 E 21 230 5800 E 21 240 0700 E 21 240 1200 E 21 340 1300 E 18 450 0300 E 18 560 2100 E 19 110 4300 E 20 010 2700 E 21 290 4000 E 19 170 1500 E 19 260 5100 E 18 520 3300 E 18 050 1200 E 18 260 4800 E 20 050 4800 E 19 310 E 21 240 0500 E 21 240 1400 E 18 360 2500 E 19 030 5300 E 18 490 1300 E 20 190 1100 E 19 280 43”E 19 510 5800 E 20 140 E 19 520 E 19 510 1500 E 20 150 5700 E 20 200 E 19 320 E 20 170 E Elevation (m) 146 158 109 150 132 n/a 106 w145 w107 w88 w123 132 w180 147 w170 256 85 w150 122 130 280 w130 95 113 105 185 123 129 123 140 111 130 119 Geomorphological setting Loess plateau River terrace Hill slope Hill slope Hill slope Plain River bank Terrace Loess plateau Alluvial fan Plain Loess plateau Loess plateau River terrace Gentle hillslope River terrace Loess on sand dune Loess plateau Hill slope Hill slope Hill crest Hill slope/terrace Loess plateau Loess plateau Loess plateau Gentle hillslope Loess plateau Gentle hillslope Gentle hillslope Loess plateau Loess plateau Gentle hillslope Loess plateau Dating method IRSL 14 C, TL, IRSL, AAR 14 C 14 C 14 C 14 C 14 C 14 C 14 C 14 C 14 C 14 C 14 C, TL, IRSL, AAR TL, IRSL, AAR 14 C AMS 14C, IRSL, AAR 14 C 14 C, TL 14 C 14 C MS IRSL IRSL MS MS MS, IRSL MS MS, AAR MS, AAR MS, IRSL-OSL AMS 14C, IRSL-OSL MS MS, IRSL Stratigraphic/dating framework 21 5, 23, 32, 34, 36 30, 31, 32 30, 31, 32 30, 31, 32 30, 32 24, 27 24, 27 9, 14, 30, 32, 34 30, 32, 34 30, 32 4, 10, 30, 32, 34 5,8, 23, 26, 29, 35, 37 5,23, 25, 35 7, 8, 12, 33 22, 23 13, 30, 32 8, 35 30, 32 30, 31, 32 11, 20 6 6 3, 18, 19 18 17, 18 18 15 16 18, 28 1 18 1, 2, 18 Age model 21 5, 32, 34 30, 31, 32 30, 31, 32 30, 31, 32 30, 32 24, 27 24, 27 30, 32, 34 30,32,34 30,32 10 5, 8, 26, 29 5 7, 8, 12 22 13, 30, 32 8, 35 30, 32 30, 31, 32 11, 20 6 6 3, 18, 19 18 17, 18 18 15 16 28 1 18 2 Abbreviations: nd ¼ no data; 14C ¼ conventional radiocarbon dating; AMS 14C ¼ Accelerator Mass Spectrometry radiocarbon dating; TL ¼ thermoluminescence; OSL ¼ optically stimulated luminescence; IRSL ¼ infrared optically stimulated luminescence; AAR ¼ amino acid racemization; and MS ¼ magnetic susceptibility. 1 ¼ Antoine et al. (2009); 2 ¼ Bokhorst et al. (2009); 3 ¼ Buggle et al. (2009); 4 ¼ Dobosi (1967); 5 ¼ Frechen et al. (1997); 6 ¼ Galovi et al. (2009); 7 ¼ Gábori-Csánk (1960); c 8 ¼ Geyh et al. (1969); 9 ¼ Hupuczi et al. (2006); 10 ¼ Hupuczi and Sümegi (2010); 11 ¼ Koloszár and Marsi (2005); 12 ¼ Krolopp and Sümegi (2002); 13 ¼ Krolopp et al. (1996); 14 ¼ Lócskai et al. (2006); 15 ¼ Markovi et al. (2005); 16 ¼ Markovi et al. (2006); 17 ¼ Markovi et al. (2007); 18 ¼ Markovi et al. (2008); 19 ¼ Markovi et al. c c c c c (2009); 20 ¼ Marsi et al. (2004); 21 ¼ Novothny et al. (2002); 22 ¼ Novothny et al. (2009); 23 ¼ Oches and McCoy (1995); 24 ¼ Pécsi and Pevzner (1974); 25 ¼ Pécsi (1979); 26 ¼ Pécsi et al. (1979); 27 ¼ Pécsi (1985); 28 ¼ Schmidt et al. (2009); 29 ¼ Seppäla (1971); 30 ¼ Sümegi (2005); 31 ¼ Sümegi and Hertelendi (1998); 32 ¼ Sümegi et al. (2007); 33 ¼ Vogel and Waterbolk (1964); 34 ¼ Willis et al. (2000); 35 ¼ Wintle and Packman (1988); 36 ¼ Zöller et al. (1994); and 37 ¼ Zöller and Wagner (1990). belong to the D3 region, while the D4 region includes the Serbian loess deposits. This implies that glacial material carried through the Moravian Depression may have been the major source of loess in Croatia, while weathering products of the Carpathian flysch and Carpathian Basin detritus carried by the Tisza River provided siltsized material to loess in Serbia. In addition, well-mixed sediments entrained by the Sava River are considered to be additional source material for loess in Serbia (Smalley and Leach, 1978). In the case of loess in Hungary, Smith et al. (1991) inferred a dominantly local source, suggesting that Pannonian sands (ca 1.8e12 Ma age) could have theoretically contributed to loess deposits in Hungary. More recently, Buggle et al. (2008) and Újvári et al. (2008) have independently addressed the question of provenance of loess deposits in Hungary and Serbia by analyzing their geochemical variability. These geochemical approaches revealed that fluvial action must have been critical in the formation of loess in the Carpathian Basin. This idea is also emphasized by Smalley et al. (2009) in a recent paper. The contribution of glaciofluvial material transported through the Moravian Depression to alluvial source material of loess in Serbia is assumed to be minimal by Buggle et al. (2008), while they considered the Carpathians and Eastern Alps as major sources of silt material. However, neither Újvári et al. (2008) nor Buggle et al. (2008) could undoubtedly prove the link between the supposed provenance areas (Smalley and Leach, 1978) and the loesses within the Carpathian Basin. According to the findings of the latest study of Újvári et al. (2010), strontium isotopic signatures and detrital zircon UePb ages of potential source rocks and loesses in Hungary demonstrate that the main sources of loesses must have been Variscan terrains and granitoids located on the catchment of the River Danube like the Bohemian Massif or intra-Alpine and intra-Carpathian Variscides. The River Danube and its tributaries acted as major suppliers of large amounts of silt-sized and fine-grained material into the Carpathian Basin during the Last Glacial period. This material was subsequently picked up by winds from alluvial fans and deposited to form thick loess plateaux. Simulations of paleowind patterns for the LGM showed northenortheasterly (NNE) winds in summer and westesouthwesterly (WSW) winds in winter in the Carpathian Basin (Fig. 1; Meyer and Kottmeier, 1989; Renssen et al., 2007), but with low wind velocities (0e4 m sÀ1; van Huissteden and Pollard, 2003). Jámbor (2002) identified northwestesoutheast directed Pleistocene wind tunnels by virtue of ventifact occurrences in Hungary which have been subsequently supported for Transdanubia by Bradák (2009) based on data of anisotropy of magnetic susceptibility. In Serbia, Markovi et al. c (2008) inferred northenorthwesterly and southeasterly wind regimes in North Serbia from loess isopach mapping, while for the Danube-Tisza confluence they have reported southwesterly winds. Large amounts of dust may have been entrained from alluvial fans by these paleowinds and after a short (50e300 km) G. Újvári et al. / Quaternary Science Reviews 29 (2010) 3157e3166 3161 Table 2 Average sedimentation and mass accumulation rates in the Carpathian Basin for marine isotope stage 2 (12e28 ka). Site name Albertirsab Basaharc Bodrogkeresztúr I Csorgókút I Csorgókút II Debrecen (Alföldi brickyard) } Dunaszekcso Dunaújváros Katymárc Lakitelek Id Látókép Madaras Mende Pakse Ságvárf } Sütto Szeged-Öthalom I Tápiósüly Tokaj (Kereszt Hill II) Tokaj (Patkó-quarry) Üveghuta-2 borehole Erdut Zmajevac Batajnica Crvenka Irig Moorin s Petrovaradin Ruma Stari Slankamen Surduk Susek Titel Statistics Min First quartile (Q1) Median (Q2) Third quartile (Q3) Max Mean SD (s) Standard errorg 0.10 0.14 0.23 0.34 0.95 0.28 0.19 0.03 150 215 338 504 1422 417 287 50.01 34 49 77 115 324 95 66 11.40 9.3 13.3 20.9 31.2 88.2 25.9 17.8 3.10 MAR MARPM10 MARPM2 Reliabilitya SR (mm aÀ1) (g mÀ2 aÀ1) (g mÀ2 aÀ1) (g mÀ2 aÀ1) 0.39 0.23 0.25 0.19 0.30 0.16 0.47 0.83 0.42 0.17 0.14 0.25 0.51 0.95 0.12 0.39 0.22 0.34 0.15 0.22 0.23 0.14 0.29 0.22 0.13 0.13 0.26 0.12 0.13 0.11 0.29 0.10 0.34 587 348 381 284 453 237 707 1238 632 254 212 375 761 1422 176 584 332 504 222 332 338 215 437 329 197 192 395 174 192 168 434 150 510 134 79 87 65 103 54 161 282 144 58 48 86 173 324 40 133 76 115 51 76 77 49 100 75 45 44 90 40 44 38 99 34 116 36.4 21.6 23.6 17.6 28.1 14.7 43.8 76.7 39.2 15.7 13.1 23.3 47.2 88.2 10.9 36.2 20.6 31.2 13.8 20.6 20.9 13.3 27.1 20.4 12.2 11.9 24.5 10.8 11.9 10.4 26.9 9.3 31.6 High High High High High High Moderate Moderate High High Moderate Low High High High High High High High High Low Moderate High Low Low Moderate Low Moderate Moderate High High Low High Fig. 2. Ageedepth model of the Tokaj (Patkó-quarry) profile and the approach of MAR calculations. Profile description and uncalibrated radiocarbon ages are from Sümegi (2005) and Sümegi et al. (2007). Uncalibrated radiocarbon dates were converted to calendar years (cal BP) using CALIB 6.0.1 software and IntCal09 calibration curve (Stuiver and Reimer, 1993; Reimer et al., 2009) and these calibrated dates were used to create the age model. Abbreviations: P ¼ paleosol, Ch ¼ charchoal fragments, L ¼ loess, and Rs ¼ recent soil. transport the majority of the coarse silt fraction must have been removed from the atmosphere mainly by dry and subordinately by wet deposition. The fine to very fine fractions of loess material may originate from local and distant sources as well. Varga (in press-a, in press-b) claims that the fine-grained fractions of loesses in Hungary, showing up as secondary maximums in the grain size distributions, likely originated from a background dust load of the atmosphere and have been transported by upper level flow and deposited far from their source areas. Indeed, Stuut et al. (2009) also suggest that small dust (fine to very fine silt: 2e8 mm), originating from North Africa, could have provided a silty admixture to the soil/loess system in this region. 4.2. Dust flux estimates and model-paleodata comparison ~ Loess MARs in the Carpathian Basin have median ðxÞ and mean ðxÞ values of 338 and 417 g mÀ2 aÀ1, respectively (range: 150e1422 g mÀ2 aÀ1), which are lower than those in some regions of Europe (from 800 to 3200 g mÀ2 aÀ1 around the River Rhine; Frechen et al., 2003) and North America (>1500 up to 17,500 g mÀ2 aÀ1 in Nebraska, Kansas, Mississippi Valley; Bettis et al., 2003; Alaska: x ¼ 433 g mÀ2 aÀ1 ; Muhs et al., 2003), and slightly higher than that calculated for the Chinese Loess Plateau (~ ¼ 310 g mÀ2 aÀ1 ; Kohfeld and Harrison, 2003). Aeolian flux x calculated for the world’s oceans (Kohfeld and Harrison, 2001) are lower than loess MAR estimates in Europe as stated by Frechen et al. (2003) as well, but if one takes our MARPM2 estimates into account, which approximate much better the characteristic grain size distribution of marine sediments (around 1e5 mm; Rea, 1994), then these estimates overlap. Lower MARs were detected in plain and hill slope settings (Fig. 3), mainly around and beyond the Tisza River on the Great Hungarian Plain, on the Middle Bácska (Baka) region and c in the surroundings of the Fruka Gora Mountain in Serbia (Fig. 4). s This observation is likely attributable to 4 factors such as (1) weaker paleowinds around and beyond the Tisza River, in some cases caused by topographic barriers (e.g. Tokaj Hill, Fruka Gora Mountain) (2) s a Reliability based on age dating. Low: chronology based on only magnetic susceptibility and/or pedostratigraphyical data; Moderate: age model based on one 14 C or TL-OSL-IRSL datum and/or amino acid racemization data; High: dating based on multiple 14C and/or TL-IRSL-OSL data. b Albertirsa: data are given for the 20.6e28 ka period. c Katymár: data are given for the 17,000e28,000 cal BP period. d Lakitelek I: data are given for the 12,000e26,700 cal BP period. e Paks: data are given for the 13e19 ka period. f Ságvár: data are given ffiffiffiffi the 21,030e22,700 cal BP period. p for g Standard error ¼ s= N , where N ¼ number of independent MAR estimates (33). lower supply of silt-sized material by the Tisza River compared to the Danube, (3) temporarily unfavorable conditions for dust accretion owing to sheet wash on hill slopes, and (4) influence of paleovegetation (Sümegi et al., 2007). Higher loess MARs in terrace and plateau settings along or close to the River Danube and on the confluence area of the Danube and Tisza rivers (Fig. 4), are probably due to the proximity to the sediment source. Model simulations of the mineral dust cycle at the LGM provided diverse dust flux estimates using different, but mostly overlapping particle size classes for simulated aerosol transport (Andersen et al., 1998; Mahowald et al., 1999; Reader et al., 1999; Werner et al., 2002; Mahowald et al., 2006a). Andersen et al. (1998) have not published dust flux maps containing comparable 3162 G. Újvári et al. / Quaternary Science Reviews 29 (2010) 3157e3166 Fig. 3. Box-whisker plots of MARs derived from loess profiles located in different geomorphological settings in the Carpathian Basin for the Last Glacial period. The gray box represents the interquartile range of the data, the line in the box shows the median and the lines above and below the box are maximum and minimum values. dust deposition data for the Carpathian Basin. Mahowald et al. (1999) reported dust deposition fluxes ranging from 5 to 20 g mÀ2 aÀ1 calculated with a lognormal aerosol size distribution, centered at 2.5 mm mass median diameter. Taking our calculated x MARPM10 values into account (~ ¼ 77 and x ¼ 95 g mÀ2 aÀ1 ), the dust flux estimate of Mahowald et al. (1999) appears to be an underestimation of the real dust deposition (Fig. 5). Aeolian fluxes of 8e32 g mÀ2 aÀ1 presented by Reader et al. (1999) are generally in x accordance with MARPM2 results from this study (~ ¼ 20:9 and x ¼ 25:0 g mÀ2 aÀ1 ), while estimates of Werner et al. (2002) (1e5 g mÀ2 aÀ1) can be considered as a significant underestimation of the real flux. In the most recent modeling study, Mahowald et al. (2006a) realized the weaknesses of previous models, such as the large underestimation of dust deposition in continental regions like Europe and North America and additional source areas over Europe, North and South America have been included in their simulation. Inclusion of these glaciogenic sources resulted in significant increases of simulated aeolian fluxes over Central Europe, with modeled dust deposition in the Carpathian Basin reaching values of 20e200 g mÀ2 aÀ1 (TUNE1-LGM) as compared to MARs of 10e20 g mÀ2 aÀ1 for the case BASECO2-LGM simulated without glaciogenic sources, based solely on sources determined by terrestrial vegetation modeling by BIOME3. The range of this increased dust flux is very similar to the interquartile range of our MARPM10 observations (20e 200 versus 49e115 g mÀ2 aÀ1) (Fig. 5) implying that indeed extensive mineral dust sources must have existed in East Central Europe during MIS 2. 4.3. Possible radiative effects An assessment of the radiative effects of dust over East Central Europe during the Last Glacial period is especially complicated because it depends on many factors: dust particle size, mineralogy and mixing, concentration and vertical distribution, temperature and albedo of the underlying surface, and the presence or absence of clouds (Tegen and Lacis, 1996; Tegen et al., 1996; Liao and Seinfeld, 1998; Claquin et al., 1999; Sokolik and Toon, 1999; Quijano et al., 2000; Andreae and Rosenfeld, 2008). Although calculation of radiative forcing by dust or modeling is beyond the scope of the present study, some aspects of the mentioned factors are worth discussing. Glaciogenic or mountain sources of loess must have been significant sources, while loess deposits in the Carpathian Basin might have been sinks and sources of mineral dust during the Last Glacial period. The fine-grained (clay) fraction of mineral dust must have had a direct radiative effect on the climate of East Central Europe during glacial periods. Quartz is the dominant mineral in loess deposits in the Carpathian Basin and has strong absorption in the infrared (IR) band (Sokolik and Toon, 1999), but it is less abundant in the fine fraction (ca 5% in the <5 mm fraction; Nemecz et al., 2000). Clays such as illite, montmorillonite, and kaolinite are the most abundant minerals in the fine fraction of loess in the Carpathian Basin. These clays have strong absorption in the IR and absorb at solar wavelengths as well. Calcite, which is quite abundant in the fine fraction, shows almost no absorption at UV and visible wavelengths, but has strong IR absorption with band positions that are quite different from those of quartz and clays (Sokolik and Toon, 1999). It must be mentioned, however, that the majority of calcite minerals in loess are not primary detrital, but secondary minerals formed during diagenetic and pedogenetic processes. Hematite, which is the strongest absorber in the UV and visible (Sokolik and Toon,1999; Claquin et al., 1999), is not abundant in loess in the Carpathian Basin. Almost all of the loess minerals mentioned above absorb mainly in the IR band, except clays which are the most abundant minerals in the fine fraction of loess and absorb at solar wavelengths as well. This suggests strong IR absorption and somewhat weaker absorption at UV and solar wavelengths caused by mixtures of these dust forming minerals in the Last Glacial atmosphere over the Carpathian Basin. The magnitude and even the sign of radiative forcing by mineral dust depends on the underlying surface albedo as well (e.g. Tegen et al., 1996), which must have been low in the Carpathian Basin during the Last Glacial period as it was covered by boreal forests and grasslands (Willis et al., 2000; Rudner and Sümegi, 2001). Therefore, increased dustiness of the atmosphere over this region could probably lead to surface cooling. A model simulation of Overpeck et al. (1996), however, found that high glacial dust loadings might have caused significant regional warming (ca 5  C) during the LGM, but this study does not provide a realistic estimate of the magnitude and spatial distribution of this effect (for further discussion see Harrison et al., 2001; Claquin et al., 2003). According to a subsequent modeling study (Claquin et al., 2003), the annual average change (LGM minus modern) in direct radiative forcing at the top of the atmosphere (TOA) for the Carpathian Basin amounted to ca À1 to þ1 W mÀ2. As Claquin et al. (2003) have provided LGM climate sensitivities (surface temperature change per unit radiative forcing, K W mÀ2) only for the tropics we are unable to translate these values to surface temperature change in the Carpathian Basin based on this study. However, considering modeled radiative forcing (À1 to þ1 W mÀ2) by Claquin et al. (2003) and climate sensitivities published in other papers (e.g. Kiehl et al., 2006; Mahowald et al., 2006b; Schneider von Deimling et al., 2006; Chylek and Lohmann, 2008) resulting surface temperature change can be roughly quantified. This surface cooling or warming attributable to high dust loadings during the Last Glacial period is expected to be low in the Carpathian Basin, based at least on the model of Claquin et al. (2003) who used the three-dimensional dust fields of Mahowald et al. (1999) to compute total column optical depth estimates. The problem with the Mahowald et al. (1999) model is that it significantly underestimates dust loading and deposition in East Central Europe. As a result, surface temperature change during the Last Glacial period owing to higher dust loadings in the Carpathian Basin may have been larger than was formerly assumed. This idea is strongly supported by a global simulation of Mahowald et al. (2006b) that has been parameterized using the findings of the more realistic mineral dust cycle model of Mahowald et al. (2006a). In this study, dealing with G. Újvári et al. / Quaternary Science Reviews 29 (2010) 3157e3166 3163 Fig. 4. Aeolian MAR (a) and MARPM10 (b) estimates at different loess sites in the Carpathian Basin for marine isotope stage (MIS) 2. climatic impacts of mineral aerosols at the LGM, the authors reported zonally averaged values of TOA direct radiative forcing for the 44e48 latitudes of ca À2 to À2.5 W mÀ2 and a resulting surface cooling of about À2  C. These figures are likely much more accurate given the MARs calculated in this study, which suggest higher mineral dust content of the atmosphere over East Central Europe. Unfortunately, not all researchers continue to use this higher mineral dust amount in their models. For example, the radiative effects of dust over East Central Europe at the LGM appear to be underestimated in a recent study of Takemura et al. (2009), because of a mineral dust cycle model that provided unrealistic aeolian flux for this region. Beside the direct radiative effects on the radiation budget, mineral dust particles can act as cloud condensation nuclei (CCN) or ice nuclei (IN) in some atmospheric situations thereby causing an indirect radiative effect (Lohmann and Feichter, 2005; Andreae and Rosenfeld, 2008). Dust particles can either suppress or enhance precipitation formation under different conditions (Kelly et al., 3164 G. Újvári et al. / Quaternary Science Reviews 29 (2010) 3157e3166 Fig. 5. Comparison between modeled and observed dust depositions in the study area for the Last Glacial period. Horizontal and vertical bars represent the ranges of observed and modeled dust deposition fluxes. Filled circles show the mean values of loess MARs (MARPM10 : 95, and MARPM2 : 25.9 g mÀ2 aÀ1) and the averages of modeled deposition. For further details see Section 4.2. 2007). Cloud droplets cannot nucleate on entirely insoluble mineral dust particles, but for some conditions, particles larger than 2 mm can activate regardless of their composition (Kelly et al., 2007). With atmospheric ageing, the ability of dust particles such as calcite, a frequently occurring mineral in loess, to nucleate increases significantly (Gibson et al., 2006; Koehler et al., 2009; Sullivan et al., 2009). In the clay fraction of loess deposits illite, montmorillonite and kaolinite are the most abundant minerals which have high ice nucleation efficiency contrary to quartz and calcite which proved to be poor ice nuclei under laboratory conditions (Eastwood et al., 2008; Zimmermann et al., 2008). These processes exert an influence on cloud formation, and therefore have an effect on radiative forcing depending on clear-sky or cloudy atmospheric conditions and the position of the dust layer (Liao and Seinfeld, 1998; Quijano et al., 2000). Indeed, simulations have highlighted the fact that the indirect effect of mineral dust particles is mainly caused by their properties to act as ice nuclei (Takemura et al., 2009). This indirect effect may have contributed to the cold LGM climate as well. Taking the several aforementioned factors (particle size, mineralogy, mixing, vertical distribution, surface albedo, presence or absence of clouds) into account, calculation of the net radiative forcing is inherently difficult and it remains a challenge for future modeling studies. Nevertheless, a more accurate reconstruction for mid-latitude loess regions such as East Central Europe is necessary to gain a better understanding of climate during the LGM. The main dust deposition hot spots are located along or close to the River Danube. MARPM10 and MARPM2 estimates calculated from loess MARs varied between 34e324 and 9.3e88.2 g mÀ2 aÀ1, respectively. These estimated mineral dust fluxes are lower than those in some regions of Europe and the USA, while larger than those in China (Bettis et al., 2003; Frechen et al., 2003; Kohfeld and Harrison, 2003; Muhs et al., 2003). The magnitude of the aeolian fluxes calculated for the oceans on the northern hemisphere was of the same order as our fine-grained (<2 mm) MAR estimates (Kohfeld and Harrison, 2001). Comparison of model and paleodata showed that LGM dust cycle models tend to significantly underestimate mid-latitude continental dust deposition (Mahowald et al., 1999; Werner et al., 2002). The inclusion of glaciogenic and non-glaciogenic mineral dust sources into these models, as Mahowald et al. (2006a) have done is indispensable. These mineral dust sources must have been located in front of the Scandinavian ice sheet and along the Rhine and Danube rivers in Europe. Silt-sized material carried by the Danube and its tributaries from the Alps and Carpathians accumulated on alluvial fans and was subject to subsequent aeolian erosion, transport and deposition (Buggle et al., 2008; Újvári et al., 2008). Since the mineralogical composition of loess deposits closely approximates the mineralogy of their sources and because loess regions might have been sources and not just sinks of Last Glacial mineral dust, mineralogical data on loess can serve as fundamental input data for model simulations of the climatic impact of high glacial dust loadings over mid-latitude loess regions. Such simulations are challenging owing to the complexity of calculations of direct and indirect radiative effects of dust. Future modeling studies of dust-induced radiative forcing at the LGM should include dust field models that better approximate mid-latitude terrestrial aeolian fluxes in order to gain insight into the real climatic effects of last glacial dust loadings over East Central Europe. Acknowledgments We are indebted to Szabolcs Czigány for loess grain size measurements in the Laboratory of Pedology and Quaternary Studies at the Washington State University, Pullman, USA. Thanks are extended to the two anonymous reviewers whose comments helped to improve the clarity of the manuscript. References Albrecht, B.A., 1989. Aerosols, cloud microphysics, and fractional cloudiness. Science 245, 1227e1230. Andersen, K.K., Armengaud, A., Genthon, C., 1998. Atmospheric dust under glacial and interglacial conditions. Geophysical Research Letters 25, 2281e2284. Andreae, M.O., Rosenfeld, D., 2008. Aerosolecloudeprecipitation interactions. Part 1. The nature and sources of cloud-active aerosols. Earth-Science Reviews 89, 13e41. Antoine, P., Rousseau, D.-D., Fuchs, M., Hatté, C., Gauthier, C., Markovi, S.B., c Jovanovi, M., Gaudenyi, T., Moine, O., Rossignol, J., 2009. High-resolution c record of the last climatic cycle in the southern Carpathian Basin (Surduk, Vojvodina, Serbia). Quaternary International 198, 19e36. Archer, D., Winguth, A., Lea, D., Mahowald, N., 1998. What caused the glacialeinterglacial atmospheric pCO2 cycles? Reviews of Geophysics 38, 159e189. Arimoto, R., 2001. Eolian dust and climate: relationships to sources, tropospheric chemistry, transport and deposition. Earth Sciences Reviews 54, 29e42. Bettis III, E.A., Muhs, D.R., Roberts, H.M., Wintle, A.G., 2003. Last Glacial loess in the conterminous USA. Quaternary Science Reviews 22, 1907e1946. Bokhorst, M.P., Beets, C.J., Markovic, S.B., Gerasimenko, N.P., Matvishiina, Z.N., Frechen, M., 2009. Pedo-chemical climate proxies in late Pleistocene SerbianeUkranian loess sequences. Quaternary International 198, 113e123. Bradák, B., 2009. Application of anisotropy of magnetic susceptibility (AMS) for the determination of paleo-wind directions and paleo-environment during the accumulation period of Bag Tephra, Hungary. Quaternary International 198, 77e84. Broecker, W.S., Henderson, G.H., 1998. The sequence of events surrounding Termination II and their implications for the cause of glacialeinterglacial CO2 changes. Paleoceanography 13, 352e364. Buggle, B., Glaser, B., Zöller, L., Hambach, U., Markovi, S., Glaser, I., Gerasimenko, N., c 2008. Geochemical characterization and origin of Southeastern and Eastern 5. Conclusions Chronological data of 33 loess sites in Hungary, Croatia and Serbia have been compiled to calculate aeolian mass accumulation rates in the Carpathian Basin for the Last Glacial period (MIS 2, 12e28 ka). Dust fluxes were found to be higher in plateau and river terrace settings than in hill slope and alluvial plain settings. Aeolian fluxes over the Last Glacial period in the Carpathian Basin can be characterized by median ð~Þ and mean ðxÞ values of 338 and 417 g mÀ2 aÀ1. x G. Újvári et al. / Quaternary Science Reviews 29 (2010) 3157e3166 European loesses (Serbia, Romania, Ukraine). Quaternary Science Reviews 27, 1058e1075. Buggle, B., Hambach, U., Glaser, B., Gerasimenko, N., Markovi, S., Glaser, I., Zöller, L., c 2009. Stratigraphy, and spatial and temporal paleoclimatic trends in Southeastern/Eastern European loess paleosol sequences. Quaternary International 196, 86e106. Buseck, P.R., Pósfai, M., 1999. Airborne minerals and related aerosol particles: Effects on climate and the environment. Proceedings of the National Academy of Sciences of the United States of America 96, 3372e3379. Chadwick, L.A., Derry, O.A., Vitousek, P.M., Huebert, B.J., Hedin, L.O., 1999. Changing sources of nutrients during four million years of ecosystem development. Nature 397, 491e497. Chylek, P., Lohmann, U., 2008. Aerosol radiative forcing and climate sensitivity deduced from the Last Glacial Maximum to Holocene transition. Geophysical Research Letters 35, L04804. doi:10.1029/2007GL032759. Claquin, T., Schulz, M., Balkanski, Y.J., 1999. Modeling the mineralogy of atmospheric dust sources. Journal of Geophysical Research 104, 22243e22256. Claquin, T., Roelandt, C., Kohfeld, K.E., Harrison, S.P., Tegen, I., Prentice, I.C., Balkanski, Y., Bergametti, G., Hansson, M., Mahowald, N., Rodhe, H., Schulz, M., 2003. Radiative forcing of climate by ice-age atmospheric dust. Climate Dynamics 20, 193e202. Coakley Jr., J.A., Cess, R.D., Yurevich, F.B., 1983. The effect of tropospheric aerosols on the earth’s radiation budget: a parameterization for climate models. Journal of Atmospheric Science 40, 116e138. Coale, K.H., Johnson, K.S., Fitzwater, S.E., Gordon, R.M., Tanner, S., Chavez, F.P., Ferioli, L., Sakamoto, C., Rogers, P., Millero, F., Steinberg, P., Nightingale, P., Cooper, D., Cochlan, W.P., Landry, M.R., Constantinou, J., Rollwagen, G., Trasvina, A., Kudela, R., 1996. A massive phytoplankton bloom induced by an ecosystem-scale iron fertilization experiment in the equatorial Pacific Ocean. Nature 383, 495e501. Codarcea, V., 1977. Percentage distribution of heavy minerals in the loess profiles at Paks and Mohács. Földrajzi Közlemények 25, 138e143. Csontos, L., Benkovics, L., Bergerat, F., Mansy, J., Wórum, G., 2002. Tertiary deformation history from seismic section study and fault analysis in a former European Tethyan margin (the MecsekeVillány area, SW Hungary). Tectonophysics 357, 81e102. Derbyshire, E., 2003. Loess, and the Dust Indicators and Records of Terrestrial and Marine Palaeoenvironments (DIRTMAP) database. Quaternary Science Reviews 22, 1813e1819. } Dobosi, V., 1967. Új felso-paleolit telep az Alföldön. (New Upper Paleolithic settle} ment on the Great Hungarian Plain). Archeológiai Értesíto 94, 184e193. Duce, R.A., Tindale, N.W., 1991. Atmospheric transport of iron and its deposition in the ocean. Limnology and Oceanography 36, 1715e1726. Eastwood, M.L., Cremel, S., Gehrke, C., Girard, E., Bertram, A.K., 2008. Ice nucleation on mineral dust particles: onset conditions, nucleation rates and contact angles. Journal of Geophysical Research 113, D22203. doi:10.1029/2008JD010639. Falkowski, P.G., Barber, R.T., Smetacek, V., 1998. Biogeochemical controls and feedbacks on ocean primary production. Science 281, 200e206. Fan, S.M., Moxim, W.J., Levy, H., 2005. Implications of droplet nucleation to mineral dust aerosol deposition and transport. Geophysical Research Letters 32, L10805. Frechen, M., Horváth, E., Gábris, G., 1997. Geochronology of Middle and Upper Pleistocene loess sections in Hungary. Quaternary Research 48, 291e312. Frechen, M., Oches, E.A., Kohfeld, K.E., 2003. Loess in Europe e mass accumulation rates during the Last Glacial Period. Quaternary Science Reviews 22, 1835e1857. c c Galovi, L., Frechen, M., Halami, J., Durn, G., Romi, M., 2009. Loess chroc nostratigraphy in Eastern Croatia e a luminescence dating approach. Quaternary International 198, 85e97. Gábori-Csánk, V., 1960. A ságvári telep abszolút kormeghatározása. (Absolute age } determination of the settlement at Ságvár). Archeológiai Értesíto 87, 125e129. Gerei, L., Reményi, M., Pécsi-Donáth, É,1979. Mineralogical analysis of the borehole drilled } 1978 on the loess plateau of Dunakömlod. Acta Geologica Hungarica 22, 501e511. Geyh, M.A., Schweitzer, F., Vértes, L., Vogel, I.C., 1969. A magyarorszagi würmi eljegesedés új kronológiai adatai. (New chronological data of the Weichselian glaciation in Hungary). Földrajzi Közlemények 18, 5e18. Gibson, E.R., Hudson, P.K., Grassian, V.H., 2006. Aerosol chemistry and climate: laboratory studies of the carbonate component of mineral dust and its reaction products. Geophysical Research Letters 33, L13811. doi:10.1029/2006GL026386. Harrison, S.P., Kohfeld, K.E., Roelandt, C., Claquin, T., 2001. The role of dust in climate changes today, at the last glacial maximum and in the future. Earth Science Reviews 54, 43e80. Hum, L., 2002. Délkelet-dunántúli löszösszletek ásványos és geokémiai jellegei és ezek eredete. (Origins of the mineral and geochemical characteristics of the loess sediments in SE Transdanubia). Földtani Közlöny 132, 117e132. Hum, L., Fényes, J., 1995. The geochemical characteristics of loesses and paleosols in the South-Eastern Transdanube (Hungary). Acta Mineralogica-Petrographica (Szeged) 36, 89e100. Hupuczi, J., Lócskai, T., Hum, L., Sümegi, P., 2006. Heinrich események kimutatása hazai löszszelvény alapján. (The demonstration of Heinrich events in Hungarian loess profiles). Malakológiai Tájékoztató 24, 31e34. Hupuczi, J., Sümegi, P., 2010. The Late Pleistocene paleoenvironment and paleoclimate of the Madaras section (South Hungary), based on preliminary records from mollusks. Central European Journal of Geosciences 2, 64e70. } Jámbor, Á, 2002. A magyarországi pleisztocén éleskavics elofordulások és földtani } jelentoségük. (Occurrences and importance of the Pleistocene ventifacts in Hungary). Földtani Közlöny 132, 101e116. 3165 Kelly, J.T., Chuang, C.C., Wexler, A.S., 2007. Influence of dust composition on cloud droplet formation. Atmospheric Environment 41, 2904e2916. Kiehl, J.T., Shields, C.A., Hack, J.J., Collins, W.D., 2006. The climate sensitivity of the Community Climate System Model version 3 (CCSM3). Journal of Climate 19, 2584e2596. Koehler, K., Kreidenweis, S.M., DeMott, P.J., Petters, M.D., Prenni, A.J., Carrico, C.M., 2009. Hygroscopicity and cloud droplet activation of mineral dust aerosol. Geophysical Research Letters 36, L08805. doi:10.1029/2009GL037348. Kohfeld, K.E., Harrison, S.P., 2001. DIRTMAP: the geological record of dust. EarthScience Reviews 54, 81e114. Kohfeld, K.E., Harrison, S.P., 2003. Glacialeinterglacial changes in dust deposition on the Chinese Loess Plateau. Quaternary Science Reviews 22, 1859e1878. Koloszár, L., Marsi, I., 2005. Formations of Late Neogene and Pleistocene terrestrial sediments in the region of Mórágy Hill (Hungary). Acta Geologica Hungarica 48, 317e337. Kosti, N., Proti, N., 2000. Pedology and mineralogy of loess profiles at Kapelac c Batajnica and Stala, Serbia. Catena 41, 217e227. c Krolopp, E., Sümegi, P., 2002. A ságvári lösz-rétegsor csigafaunája. (The mollusc fauna of the Ságvár loess profile). Malakológiai Tájékoztató 20, 7e14. Krolopp, E., Sümegi, P., Kuti, L., Hertelendi, E., Kordos, L., 1996. Szeged környéki } löszképzodmények keletkezésének paleoökológiai rekonstrukciója. (Palaeoecological reconstruction of the formation of loess deposits in the environs of Szeged). Földtani Közlöny 125, 309e361. Lambert, F., Delmonte, B., Petit, J.R., Bigler, M., Kaufmann, P.R., Hutterli, M.A., Stocker, T.F., Ruth, U., Steffensen, J.P., Maggi, V., 2008. Dusteclimate couplings over the past 800,000 years from the EPICA Dome C ice core. Nature 452, 616e619. Liao, H., Seinfeld, J.H., 1998. Radiative forcing by mineral dust aerosols: sensitivity to key variables. Journal of Geophysical Research 103, 31637e31645. Lócskai, T., Hupuczi, J., Hum, L., Sümegi, P., 2006. Dansgaard-Oeschger ciklusok } kimutatása hazai löszszelvénybol. (The demonstration of Dansgaard-Oeshger cycles from Hungarian loess profiles). Malakológiai Tájékoztató 24, 35e39. Lohmann, U., Feichter, J., 2005. Global indirect aerosol effects: a review. Atmospheric Chemistry and Physics 5, 715e737. Maher, B.A., Kohfeld, K.E., 2009. DIRTMAP Version 3. LGM, Late Holocene, and Modern Eolian Fluxes from Ice Cores, Marine Sediments, Marine Sediment Traps, and Terrestrial Deposits. http://www.lec.lancs.ac.uk/dirtmap3. Maher, B.A, Prospero, J.M., Mackie, D., Gaiero, D., Hesse, P.P., Balkanski, Y., 2010. Global connections between aeolian dust, climate and ocean biogeochemistry at the present day and at the last glacial maximum. Earth Science Reviews 99, 61e97. Mahowald, N., Kohfeld, K.E., Hansson, M., Balkanski, Y., Harrison, S.P., Prentice, I.C., Schulz, M., Rodhe, H., 1999. Dust sources and deposition during the Last Glacial Maximum and current climate: a comparison of model results with paleodata from ice cores and marine sediments. Journal of Geophysical Research 104,15895e15916. Mahowald, N.M., Baker, A.R., Bergametti, G., Brooks, N., Duce, R.A., Jickells, T.D., Kubilay, N., Prospero, J.M., Tegen, I., 2005. Atmospheric global dust cycle and iron inputs to the ocean. Global Biogeochemical Cycles 19, GB4025. doi:10.1029/ 2004GB002402. Mahowald, N.M., Muhs, D.R., Levis, S., Rasch, P.J., Yoshioka, M., Zender, C.S., Luo, C., 2006a. Change in atmospheric mineral aerosols in response to climate: last glacial period, preindustrial, modern, and doubled carbon dioxide climates. Journal of Geophysical Research 111, D10202. doi:10.1029/2005JD006653. Mahowald, N.M., Yoshioka, M., Collins, W.D., Conley, A.J., Fillmore, D.W., Coleman, D.B., 2006b. Climate response and radiative forcing from mineral aerosols during the last glacial maximum, pre-industrial, current and doubledcarbon dioxide climates. Geophysical Research Letters 33, L20705. doi:10.1029/ 2006GL026126. Markovi, S.B., Kosti, N., Oches, E.A., 2004. Paleosols in the Ruma loess section. c c Revista Mexicana de Ciencias Geológicas 21, 79e87. Markovi, S.B., McCoy, W.D., Oches, E.A., Savi, S., Gaudenyi, T., Jovanovi, M., c c c c c Stevens, T., Walther, R., Ivanisevi, P., Gali, Z., 2005. Paleoclimate record in the upper Pleistocene loessepalaeosol sequence at Petrovaradin brickyard (Vojvodina, Serbia). Geologica Carpathica 56, 545e552. Markovi, S.B., Oches, E.A., Sümegi, P., Jovanovi, M., Gaudenyi, T., 2006. An introc c duction to the middle and upper Pleistocene loessepalaeosol sequence at Ruma brickyard, Vojvodina, Serbia. Quaternary International 149, 80e86. Markovi, S.B., Oches, E.A., McCoy, W.D., Frechen, M., Gaudenyi, T., 2007. Malacoc logical and sedimentological evidence for ‘warm’ glacial climate from the Irig loess sequence, Vojvodina, Serbia. Geochemistry Geophysics Geosystems 8, Q09008. doi:10.1029/2006GC001565. Markovi, S.B., Bokhorst, M.P., Vandenberghe, J., McCoy, W.D., Oches, E.A., c Hambach, U., Gaudenyi, T., Jovanovi, M., Stevens, T., Zöller, L., Machalett, B., c 2008. Late Pleistocene loessepalaeosol sequences in the Vojvodina region, North Serbia. Journal of Quaternary Science 23, 73e84. Markovi, S.B., Hambach, U., Catto, N., Jovanovi, M., Buggle, B., Machalett, B., c c Zöller, L., Glaser, B., Frechen, M., 2009. The middle and late Pleistocene loessepalaeosol sequences at Batajanica, Vojvodina, Serbia. Quaternary International 198, 255e266. Marsi, I., Don, Gy., Földvári, M., Koloszár, L., Kovács-Pálffy, P., Krolopp, E., Lantos, M., Nagy-Bodor, E., Zilahi-Sebess, L., 2004. Quaternary sediments of the northeastern Mórágy Block. Annual Report of the Geological Institute of Hungary for the Year 2003, pp. 343e359. Martin, J.H., 1990. Glacialeinterglacial CO2 change: the iron hypothesis. Palaeoceanography 5, 1e13. Martin, J.H., Fitzwater, S.E., 1988. Iron deficiency limits phytoplankton growth in the north-east Pacific subarctic. Nature 331, 341e343. 3166 G. Újvári et al. / Quaternary Science Reviews 29 (2010) 3157e3166 Smalley, I., O’Hara-Dhand, K., Wint, J., Machalett, B., Jary, Z., Jefferson, I., 2009. Rivers and loess: the significance of long river transportation in the complex eventsequence approach to loess deposit formation. Quaternary International 198, 7e18. Smith, B.J., Wright, J.S., Whalley, W.B., 1991. Simulated aeolian abrasion of Pannonian sands and its applications for the origins of the Hungarian loess. Earth Surface Processes and Landforms 16, 745e752. Sokolik, I.N., Toon, O.B., 1999. Incorporation of mineralogical composition into models of the radiative properties of mineral aerosol from UV to IR wavelengths. Journal of Geophysical Research 104, 9423e9444. Steffensen, J.P., 1997. The size distribution of microparticles from selected segments of the Greenland Ice Core Project ice core representing different climate periods. Journal of Geophysical Research 102, 26755e26763. Stuiver, M., Reimer, P.J., 1993. Extended 14C database and revised CALIB radiocarbon calibration program. Radiocarbon 35, 215e230. Stuut, J.-B., Smalley, I., O’Hara-Dhand, K., 2009. Aeolian dust in Europe: African sources and European deposits. Quaternary International 198, 234e245. Sullivan, R.C., Moore, M.J.K., Petters, M.D., Kreidenweis, S.M., Roberts, G.C., Prather, K.A., 2009. Effect of chemical mixing state on the hygroscopicity and cloud nucleation properties of calcium mineral dust particles. Atmospheric Chemistry and Physics 9, 3303e3316. Sümegi, P., 2005. Loess and Upper Paleolithic Environment in Hungary. Aurea Kiadó, Nagykovácsi, 312 pp. Sümegi, P., Hertelendi, E., 1998. Reconstruction of microenvironmental changes in Kopasz Hill loess area at Tokaj (Hungary) between 15,000-70,000 BP years. Radiocarbon 40, 855e863. Sümegi, P., Molnár, M., Svingor, É., Szántó, Zs, Hum, L., Gulyás, S., 2007. Results of radiocarbon analysis of Upper Weichselian loess sequences from Hungary. Radiocarbon 49, 1023e1030. Swap, R., Garstang, M., Greco, S., Talbot, R., Kallberg, P., 1992. Saharan dust in the Amazon Basin. Tellus 44B, 133e149. Szebényi, E., 1979. Sedimentological and micromineralogical analysis of the southern profile of the Paks loess exposure. Acta Geologica Hungarica 22, 427e432. Takemura, T., Egashira, M., Matsuzawa, K., Ichijo, H., O’ishi, R., Abe-Ouchi, A., 2009. A simulation of the global distribution and radiative forcing of soil dust aerosols at the Last Glacial Maximum. Atmospheric Chemistry and Physics 9, 3061e3073. Tegen, I., 2003. Modeling the mineral dust aerosol cycle in the climate system. Quaternary Science Reviews 22, 1821e1834. Tegen, I., Lacis, A.A., 1996. Modeling of particle size distribution and its influence ont he radiative properties of mineral dust aerosol. Journal of Geophysical Reserach 101, 19237e19244. Tegen, I., Lacis, A.A., Fung, I., 1996. The influence on climate forcing of mineral aerosols from disturbed soils. Nature 380, 419e422. Thompson, L.G., Mosley-Thompson, E., 1981. Microparticle concentration variations linked with climatic change: evidence from polar ice cores. Science 212, 812e815. Thompson, W.G., Goldstein, S.L., 2006. A radiometric calibration of the SPECMAP timescale. Quaternary Science Reviews 25, 3207e3215. Újvári, G., Varga, A., Balogh-Brunstad, Z., 2008. Origin, weatehring, and geochemical composition of loess in southwestern Hungary. Quaternary Research 69, 421e437. Újvári, G., Ramos, F.C., Dimond, C., Kovács, J., Varga, Gy, Varga, A., 2010. On possible sources of loess deposits in the Carpathian Basin: an isotopic approach. Geophysical Research Abstracts 12 EGU2010-3881. Yin, Y., Wurzler, S., Levin, Z., Reisin, T.G., 2002. Interactions of mineral dust particles and clouds: effects on precipitation and cloud optical properties. Journal of Geophysical Research 107, 4724. doi:10.1029/2001JD001544. Yung, Y.L., Lee, T., Wang, C.-H., Shieh, Y.-T., 1996. Dust: a diagnostic of the hydrological cycle during the Last Glacial Maximum. Science 271, 962e963. van Huissteden, K., Pollard, D., 2003. Oxygen isotope stage 3 fluvial and eolian successions in Europe compared with climate model results. Quaternary Research 59, 223e233. Varga, Gy. A Kárpát-medence légköri ásványi porkoncentrációjának alakulása a pleisztocén során (Atmospheric mineral dust concentration during the Pleistocene in the Carpathian Basin). Légkör, in press-a. Varga, Gy. Similarities of the PlioePleistocene terrestrial aeolian dust deposits in the World and in the Carpathian Basin. Quaternary International, in press-b. Vogel, I.C., Waterbolk, H.T., 1964. Groningen radiocarbon dates. V. Radiocarbon 6, 349e369. Werner, M., Tegen, I., Harrison, S., Kohfeld, K., Prentice, I.C., Balkanski, Y., Rodhe, H., Roelandt, C., 2002. Seasonal and interannual variability of the mineral dust cycle under present and glacial climate conditions. Journal of Geophysical Research 107, 4744. doi:10.1029/2002JD002365. Willis, K.J., Rudner, Z.E., Sümegi, P., 2000. Full-glacial forests of Central and South Eastern Europe. Quaternary Research 53, 203e213. Wintle, A.G., Packman, S.C., 1988. Thermoluminescence ages for three sections in Hungary. Quaternary Science Reviews 7, 315e320. Zimmermann, F., Weinbruch, S., Schütz, L., Hofmann, H., Ebert, M., Kandler, K., Worringen, A., 2008. Ice nucleation properties of the most abundant mineral dust phases. Journal of Geophysical Research 113, D23204. doi:10.1029/ 2008JD010655. Zöller, L., Oches, E.A., McCoy, W.D., 1994. Towards a revised chronostratigraphy of loess in Austria with respect to key sections in the Czech Republic and in Hungary. Quaternary Geochronology/Quaternary Science Reviews 13, 465e472. Zöller, L., Wagner, G.A., 1990. Thermoluminescence dating of loess e recent developments. Quaternary International 7/8, 119e128. Martin, J.H., Gordon, R.M., Fitzwater, S.E., 1990. Iron in Antarctic waters. Nature 345, 156e158. Martinson, D.G., Pisias, N.G., Hays, J.D., Imbrie, J., Moore, T.C., Shackleton, N.J., 1987. Age dating and the orbital theory of the Ice Ages: development of a highresolution 0 to 300,000-year chronostratigraphy. Quaternary Research 27, 1e29. Meyer, H.-H., Kottmeier, Ch., 1989. The atmospheric circulation in Europe during the Weichselian Pleniglacial e as derived from palaeowind indicators and model simulations. Eiszeitalter und Gegenwart 39, 10e18. Muhs, D.R., Ager, T.A., Bettis III, E.A., McGeehin, J., Been, J.M., Begét, J.E., Pavich, M.J., Stafford Jr., T.W., Stevens, D.A.S.P., 2003. Stratigraphy and palaeoclimatic significance of Late Quaternary loessepalaeosol sequences of the Last InterglacialeGlacial cycle in central Alaska. Quaternary Science Reviews 22,1947e1986. Nemecz, E., Pécsi, M., Hartyáni, Zs., Horváth, T., 2000. The origin of the silt size quartz grains and minerals in loess. Quaternary International 68e71, 199e208. Novothny, Á., Horváth, E., Frechen, M., 2002. The loess profile at Albertirsa, Hungary e improvements in loess stratigraphy by luminescence dating. Quaternary International 95/96, 155e163. Novothny, Á., Frechen, M., Horváth, E., Bradák, B., Oches, E.A., McCoy, W.D., Stevens, T., 2009. Luminescence and amino acid racemization chronology of the } loessepaleosol sequence at Sütto, Hungary. Quaternary International 198, 62e76. Oches, E.A., McCoy, W.D., 1995. Aminostratigraphic evaluation of conflicting age estimates for the “Young Loess” of Hungary. Quaternary Research 44, 160e170. Overpeck, J.T., Rind, D., Lacis, A., Healy, R., 1996. Possible role of dust-induced regional warming in abrupt climate change during the Last Glacial period. Nature 384, 447e449. Papp, G., 2009. Simultaneous determination of terrain correction and local average topographic density. Acta Geodaetica and Geophysica Hungarica 44, 191e202. Pécsi, M., Pevzner, M.A., 1974. Paleomagnetic measurements in the loess sequences at Paks and Dunaföldvár, Hungary. Földrajzi Közlemények 22, 215e224. Pécsi, M., 1979. Lithostratigraphical subdivision of the loess profiles at Paks. Acta Geologica Hungarica 22, 409e418. Pécsi, M., Szebényi, E., Pevzner, M.A., 1979. Upper Pleistocene litho- and chronostratigraphical type profile from the exposure at Mende. Acta Geologica Hungarica 22, 371e389. Pécsi, M., 1985. Chronostratigraphy of Hungarian loesses and underlying subaerial formation. In: Pécsi, M. (Ed.), Loess and the Quaternary: Chinese and Hungarian Case Studies. Studies in Geography in Hungary, vol. 18, pp. 33e49. Pécsi-Donáth, É, 1979. Thermal investigation of the loesses and fossil soils of Paks. Acta Geologica Hungarica 22, 419e426. Pécsi-Donáth, É, 1985. On the mineralogical and petrological properties of the younger loess in Hungary. In: Pécsi, M. (Ed.), Loess and the Quaternary: Chinese and Hungarian Case Studies. Studies in Geography in Hungary, vol.18, pp. 93e104. Petit, J.R., Mounier, L., Jouzel, J., Korotkevich, Y.S., Kotlyakov, V.I., Lorius, C., 1990. Palaeoclimatological and chronological implications of the Vostok core dust record. Nature 343, 56e58. Porter, S.C., An, Z.S., 1995. Correlation between climate events in the North Atlantic and China during the last glaciation. Nature 375, 305e308. Pye, K., 1987. Aeolian Dust and Dust Deposits. Academic Press, London, 334 pp. Quijano, A.L., Sokolik, I.N., Toon, O.B., 2000. Radiative heating rates and direct radiative forcing by mineral dust in cloudy atmospheric conditions. Journal of Geophysical Research 105, 12207e12219. Raucsik, B., 2009. Preliminary Mineralogical Data from the Paks loess Profile, Hungary. Report, University of Pannonia, Veszprém. Rea, D.K., 1994. The paleoclimatic record provided by eolian deposition in the deep sea: the geologic history of wind. Reviews of Geophysics 32, 159e195. Reader, M.C., Fung, I., McFarlane, N., 1999. The mineral dust aerosol cycle during the Last Glacial Maximum. Journal of Geophysical Research 104, 9381e9398. Reimer, P.J., Baillie, M.G.L., Bard, E., Bayliss, A., Beck, J.W., Blackwell, P.G., Bronk Ramsey, C., Buck, C.E., Burr, G.S., Edwards, R.L., Friedrich, M., Grootes, P.M., Guilderson, T.P., Hajdas, I., Heaton, T.J., Hogg, A.G., Hughen, K.A., Kaiser, K.F., Kromer, B., McCormac, F.G., Manning, S.W., Reimer, R.W., Richards, D.A., Southon, J.R., Talamo, S., Turney, C.S.M., van der Plicht, J., Weyhenmeyer, C.E., 2009. IntCal09 and Marine09 radiocarbon age calibration curves, 0e50,000 years cal BP. Radiocarbon 51, 1111e1150. Renssen, H., Kasse, C., Vandenberghe, J., Lorenz, S.J., 2007. Weichselian Late Pleniglacial surface winds over northwest and central Europe: a model-data comparison. Journal of Quaternary Science 22, 281e293. Rosenfeld, D., Rudich, Y., Lahav, R., 2001. Desert dust suppressing precipitation: a possible desertification feedback loop. Proceedings of the National Academy of Sciences of the United States of America 98, 5975e5980. Rudner, Z.E., Sümegi, P., 2001. Recurring Taiga forest-steppe habitats in the Carpathian Basin in the Upper Weichselian. Quaternary International 76/77, 177e189. Satheesh, S.K., Krishna Moorthy, K., 2005. Radiative effects of natural aerosols: a review. Atmospheric Environment 39, 2089e2110. Schmidt, E.D., Machalett, B., Markovi, S.B., Tsukamoto, S., Frechen, M., 2009. c Luminescence chronology of the upper part of the Stari Slankamen loess sequence (Vojvodina, Serbia). Quaternary Geochronology, 1e6. Schneider von Deimling, T., Held, H., Ganopolski, A., Rahmstorf, S., 2006. Climate sensitivity estimated from ensemble simulations of glacial climate. Climate Dynamics 27, 149e163. Seppäla, M., 1971. Stratigraphy and material of the loess layers at Mende, Hungary. Bulletin of the Geological Society of Finland 43, 109e123. Smalley, I.J., Leach, J.A., 1978. The origin and distribution of loess in the Danube Basin and associated regions of East-Central Europe e a review. Sedimentary Geology 21, 1e26.