Climate driven Holocene erosion in central Ukraine

1 Climate driven history of Holocene erosion in Eastern Europe- the example of a catchment at a giant 2 Chalcolithic settlement at Maidanetske, central Ukraine 3 4 5 Stefan Dreibrodt*, Institute for Ecosystem Research, CRC 1266, University of Kiel, Germany, [email protected] 6 7 Robert Hofmann, Institute for Pre- and Protohistoric Archaeology, CRC 1266, University of Kiel, Germany, [email protected] 8 9 György Sipos, Department of Physical Geography and Geoinformatics, University of Szeged, Hungary, [email protected] 10 11 Lorenz Schwark, Institute of Geoscience, CRC 1266, University of Kiel, Germany, [email protected] 12 13 Michail Videiko, Laboratory of Archaeology, Borys Grinchenko Kyiv University, Ukraine, [email protected] 14 15 Liudmyla Shatilo, Institute for Pre- and Protohistoric Archaeology, CRC 1266, University of Kiel, Germany, [email protected] 16 17 Sarah Martini, Institute for Pre- and Protohistoric Archaeology, University of Kiel, Germany, [email protected] 18 19 Philipp Saggau, Department of Geography, CRC 1266, University of Kiel, Germany, [email protected] 20 21 Rainer Duttmann, Department of Geography, CRC 1266, University of Kiel, Germany, [email protected] 22 23 Hans-Rudolf Bork, Institute for Ecosystem Research, CRC 1266, University of Kiel, Germany, [email protected] 24 25 Wiebke Kirleis, Institute for Pre- and Protohistoric Archaeology, CRC 1266, University of Kiel, Germany, [email protected] 26 27 Johannes Müller, Institute for Pre- and Protohistoric Archaeology, CRC 1266, University of Kiel, Germany, [email protected] 28 29 The paper version presented here is a non-peer reviewed version of a paper handed in for peer- 30 review to the journal Geomorphology. 31 *Corresponding author 32 1 33 Abstract 34 The younger Quaternary erosion history was reconstructed in a catchment close to the Chalcolithic 35 giant settlement Maidanetske, central Ukraine based on dated sediment sequences. Four trenches and 36 a long percussion drill-core were analyzed in a valley grading from a Loess covered plateau towards 37 the Talianky River. The sediments were dated via a combination of radiocarbon dating, optical 38 stimulated luminescence (OSL) and embedded artefacts. A suspicious non-coincidence between 39 phases of soil erosion and the settlement history at the site over long periods of the Holocene is 40 noticeable and suggests a climatically driven erosion at the site. The detected phases of erosion during 41 the past >20,000 years coincide with global (cal 27.6 +/- 1.3 kyrs BP, 12.0 +/- 0.4 kyrs BP), northern 42 hemispheric (cal 8.5 ± 0.3 kyrs BP), Mediterranean (cal 3.93 ± 0.1 kyrs BP) as well as western to central 43 European (2,700 to 2,000 cal BP) climate anomalies. For these anomalies, characterized by colder than 44 usual conditions in western and central Europe and dry conditions in the eastern Mediterranean and 45 the research area, a common trigger process seems possible. Increased occurrences of heavy 46 precipitation events, probably during phases of a weakened vegetation cover, could explain the 47 observed record. 48 A comparison of the Ukrainian record with other European erosion records raises the question again 49 about the contribution of climate variability on Holocene erosion processes. Whereas climatic 50 influence might be easier detectable in Eastern Europe, with a comparatively late onset of intensive 51 agricultural land use, in southern, central and western Europe the impact of climate variability might 52 be masked to a part according to the long history of intensive agricultural land use. 53 The composition of the sediments implies changes of the slope-channel connectivity during the 54 deposition history. Whereas the periglacial to early Holocene sediments were derived from the whole 55 catchment area, since the mid-Holocene a tendency to lower slope storage of colluvial material and 56 valley incision is noticeable. 57 Keywords: Holocene Erosion, climate and land-use, Ukraine, connectivity 2 58 1. Introduction 59 Based on numerous geomorphological investigations in southern and central Europe soil erosion has 60 been identified as one of the major and most serious impacts of humanity on the environment (e.g. 61 van Andel et al., 1990, Bork and Lang 2003, Butzer, 2005, Dotterweich, 2008, Thornes, 2009, Dreibrodt 62 et al., 2010a). Within the research region, few data about the younger Quaternary and Holocene 63 geomorphological processes at the slope scale are available. Without giving information about the land 64 use history of the catchment area Belyaev et al. (2004) report phases of gully activity in small 65 catchments in western Russia at ca. cal BP 1090-970 and 880-570. Similarly, without information about 66 Holocene land use history, Belyaev et al. (2005) report gully activity at two additional sites in western 67 Russia at ca. cal BP 8,950-8,480, 4,100-3,400, 3,140-2,870, 2,310-2,170, 1,590-1,031, and 640-490. 68 Panin et al. (2009) found a pre-Holocene origin of 15 of 19 studied gully systems in western Russia. 69 During the Holocene, these authors detected longer phases of erosion and gully activity from ca. 4,800 70 to 2,800 cal BP and 1,200 cal BP until today. Shorter periods of intenisve erosion were reconstructed 71 for the intervals ca. 4,800- 4,600, 3,900-3,600, 3,800- 2,800, 2,300- 2,100, 1,600-1,800, 1,000-800, and 72 700-500 cal BP. The phases of erosion were explained mainly by climate variability. Sycheva (2006) and 73 Sycheva et al. (2003) report a quasi-cyclicity of erosion and soil formation at the Russian part of the 74 East European Plain based on a compilation of radiocarbon dates form soils and slope deposits. The 75 observed cyclicity is ascribed to periodical climatic changes throughout the Holocene. Intervals of 76 intensive soil erosion were dated to ca. 10,200-9,500, 8,100-7,700, 6,600-6,300, 4,700-4,200, 2,700- 77 2,300, and 950-450 cal BP. Whereas researchers from southern and central Europe underline the role 78 of agricultural land use on soil erosion histories of the respective landscapes, eastern European 79 scholars rather see climatic variability and their effects on vegetation as the main drivers of Holocene 80 relief change. Thus, a comparison of the land use history known from intensive archaeological research 81 with the detectable phases of soil erosion at the research site is one focus of this paper. 82 83 3 84 2. Material and methods 85 2.1 The research site 86 The investigated catchment area is located at Majdanetskoe, district of Talne, central Ukraine 87 (48°48′N, 30°38′E) (Fig. 1). The close by archaeological site of Madanestske is a giant settlement of 88 the Tripyllia C1-period (Müller et al., 2013, 2016, Hofmann et al., 2019). Archaeological sites of this 89 type are unique because of their extremely large dimensions. At Maidanetske, on an area of 200 ha 90 approximately 3,000 houses arranged in a series of oval structures around an unbuilt central space 91 were inhabited approximately from 3,990 to 3,640 BCE (Müller et al., 2016, Ohlrau, 2018, Pickartz et 92 al., 2019). Surveys of the many potshards present on the recent surface, magnetic surveys, 93 excavations and exhaustive dating campaigns revealed a maximum number of ca. 1,500 houses was 94 inhabited contemporaneously by probably more than 10,000 people (Ohlrau, 2018, Pickartz et al., 95 2019). The climate in the region is humid continental (Dfb) today, with hot summers and cold wet 96 winters. The potential natural vegetation of the region belongs to the climate sensitive forest-steppe 97 transition zone. Where there is no agricultural land use, deciduous forests are present in the 98 landscape today. A mosaic of loess-covered plateaus dissected by small valleys characterizes the 99 recent topography. The surface soils are classified as particularly thick Chernozems in the research 100 area (Atlas of soils of the Ukrainian SSR, 1979). The studied catchment area covers ca. 6.3 km² and 101 grades from a Loess plateau towards the valley of the Talianky River spanning a relief gradient from 102 ca. 210 to 150 m a.s.l. Ditches and a small pond subdivide the valley nowadays. Meadows and shrubs 103 cover parts of the valley. The catchment area is used for large agricultural fields, subdivided by wind- 104 breaking tree lines, ditches and unpaved roads. 105 106 107 108 4 109 2.2 Methods 110 2.2.1 Field methods 111 Five trenches were dug at the lower slopes of the catchment area of the investigated valley (Fig. 1). 112 Additionally, a sediment sequence was extracted from a long (5m) percussion-drilling core situated on 113 the colluvial fan of the investigated valley close to its outlet into the larger valley of the Talianky River. 114 The sequences of soils and sediments were documented in scaled drawings and described according 115 to field instructions (AG Boden, 2005). Sediments are termed as slope deposits (abbr. S) respectively 116 colluvial layers (abbr. M), if they are of pre-Holocene respectively Holocene age and numbered in the 117 order of their genesis. Samples were taken for dating and standard laboratory analyses. 118 2.2.2 Laboratory analysis 119 Dating 120 Dating of the soils and sediments was achieved through radiocarbon measurements, optical 121 stimulated luminescence (OSL) and typological analysis of embedded artifacts. Given the scarcity of 122 datable bioremains, radiocarbon dating of bulk samples soil organic matter samples was performed 123 after removal of carbonates. The results were calibrated using OxCal v4.2.3 (Bronk Ramsey and Lee, 124 2013) with the IntCal13 atmospheric calibration curve (Reimer et al., 2013) and are presented in cal 125 years BP (2 Sigma). OSL dating was carried out on unexposed samples taken in small tubes in 126 exposure 2 and from segments of a parallel core from drilling point 1. A RISO TL/OSL DA-15 127 luminescence reader equipped with a calibrated 90Sr/90Y source was used for measurements. 128 Stimulation was carried out using blue (470 nm) or IR (870 nm) LEDs, depending on the applied 129 mineral fraction. Detection was made through either a U-340 filter (quartz) or the combination of 130 BG39 and CN-7-59 filters (feldspar). Throughout the measurements different types of the Single 131 Aliquot Regeneration (SAR) protocol was used (Murray and Wintle, 2000, 2003, Wintle and Murray, 132 2006, Thiel et al., 2011, Buylaert et al., 2012). Prior to the measurement of the equivalent dose (De) 133 tests were carried out to determine optimal temperature parameters and the reproducibility of the 5 134 SAR procedure (combined preheat and dose recovery test). The equivalent dose was determined on 135 several aliquots in case of each sample. Only those aliquots were considered for De calculation which 136 passed the following rejection criteria (recycling ratio: 1.00±0.10; maximum dose error: 10%; 137 maximum recuperation: 5%, maximum IR/OSL depletion ratio: 5%). Sample De was determined on 138 the basis of each accepted aliquot De, using different statistical techniques (Galbraith et al., 1999). 139 Decision was made on the basis of over dispersion, skewness and kurtosis values. Environmental 140 dose rate D* was determined using high resolution, extended range gamma spectrometer (Canberra 141 XtRa Coaxial HpGe detector). Dry dose rates were calculated using the conversion factors of Liritzis et 142 al. (2013). Wet dose rates were assessed on the basis of in situ water contents. The dose rate 143 provided by cosmic radiation was determined on the basis of the geographical position and depth of 144 the samples below ground level, using the equation of Prescott and Hutton (1994). All OSL ages given 145 in the text and figures of this paper are given in cal years BP (1 Sigma). Artifacts embedded in soil or 146 sediments were dated according to prevailing typochronologies by the archaeologists. All radiometric 147 age data are given completely in Table 1a and 1b. 148 Geophysical and geochemical analysis 149 Soil and sediment samples were air dried (35°C), carefully disintegrated with mortar and pestle and 150 sieved through a 2 mm mesh sieve. 151 Grain size distribution analysis was carried out for profiles 2, 3, and the sediment core 1. After removal 152 of soil organic matter (H2O2, 70 °C) and carbonates (acetic acid buffer, 70°C, pH 4.8) a laser particle 153 sizer (Malvern Mastersizer 2000) was used to measure the grain size distribution (core1, profiles 154 2 and 3). Each sample was measured for at least 45 seconds, and the measurement was repeated 155 at least 10 times, and finally averaged. The magnetic susceptibility was measured on 10 ml samples 156 (< 2 mm fraction) using a Bartington MS2B susceptibility meter (resolution 2*10-6 SI, measuring range 157 1-9999*10-5 SI, systematic error 10 %). Measurements were carried out at low (0.465 kHz) and high 158 (4.65 kHz) frequency. A 1 % Fe3O4 (magnetite) was measured regularly to check for drift and calibrate 159 the results. Mass-specific susceptibilities and frequency-dependent magnetic susceptibility (χfd) were 6 160 calculated (Dearing, 1999). The color of the samples was measured using a Voltcraft Plus RGB-2000 161 Color Analyzer set to display in a 10-bit RGB color space within a spectral range of 400 to 700 nm 162 (Rabenhorst et al., 2014, Sanmartin et al., 2014). Loss on Ignition (LOI) values were measured as 163 estimates of the organic matter and carbonate content of the sediments (Dean, 1974). After drying the 164 samples at 105°C overnight, the weight loss of the samples was determined after heating times of 2 h 165 at 550 °C and 940 °C each. For selected profiles, some additional analysis was carried out. The total 166 carbon (TOC), total nitrogen (TN) were determined with an Elementar Vario EL-III CNS analyser 167 following standard procedures. Sulfanic acid (S= 18.5 weight %) was used for instrument calibration 168 and an analytical error of ± 0.01 % was determined. On selected samples from the soil and sediment 169 sequence of core 1 a lipid analysis was carried out to infer about the catchment vegetation. Lipids were 170 extracted using pressurized liquid Extraction (DIONEX ASE200) using a solvent mixture of 171 hexane/dichloromethane (9/1; v/v) and separated into non-polar and polar compound classes by 172 automated SPE (LC-Tech Freestyle) on 2 grams of pre-extracted and activated silica. Non-polar 173 compounds were eluted with hexane/dichloromethane (9/1; v/v) and subjected to gas 174 chromatography-mass spectrometry (GC-MS) using an Agilent 7890A GC equipped with a Phenomenex 175 Zebron ZB-5 column (30m × 0.25mm i.d.; 0.25 µm film thickness) and coupled to an Agilent 5975B mass 176 chromatograph. The injection temperature was held at 60°C for 4 min, after which the oven 177 temperature was raised to 140°C at 10°C/min and subsequently to 320 °C at 3°C/min, at which it was 178 held for 8 min. The MS was operated at an electron energy of 70 eV and an ion source temperature of 179 250°C. The homologues series of n-alkanes was detected via the m/z 85 mass chromatograms and peak 180 areas used for calculation of relative abundance ratios. 181 182 183 184 7 185 3. Results 186 Deposition history 187 Sequences of sediments deposited during the younger Quaternary and soils that had formed within 188 these sediments during phases of slope stability were detected at the different exposures (Fig. 1) and 189 at the drilling point (Fig. 2). 190 Sediment core 1 191 At the drilling point on the colluvial fan of the investigated valley, the thickest sediment sequence (ca. 192 5m) was recovered (Fig. 2). The base layer S1 (4.4- > 5.0 m) comprises of a larger amount of gravel (ca. 193 4.7- > 5.0 m) and sand of a light greyish color and dates to the LGM according to an OSL datum. Above, 194 a layer of Loess was deposited (S2, ca. 4.0- 4.4 m). This pale yellowish layer is composed mainly of silt 195 with some sand and clay admixed. It is unclear so far, whether S2 originated from aeolian deposition 196 or is a fluvial redeposition. S2 dates to a period between the LGM and the YD. A YD fluvial sediment 197 was detected above (S3, 3.3- 4.0 m). Its dark brown color and silty texture (finer than the lying Loess) 198 points to an Allerød soil within the catchment as the source of the sediment. An OSL age, backed by a 199 radiocarbon age of the soil organic matter, pointing to a deposition of S3 at ca. 12.0 +/- 0.4 ka BP. S3 200 was buried by an early Holocene deposit M1 (3.0- 3.3m). Although the texture of M1 again is comprised 201 mainly of silt, a significant switch towards finer silt particles implies a change in the depositional 202 conditions. The still dark brownish color indicates that the source of M1 was an early Holocene soil 203 that covered the catchment area. According to an OSL age, the deposition of M1 occurred at 8.5 +/- 204 0.3 ka BP. A radiocarbon age of soil organic matter from the layer is slightly younger (ca. 8.160- 7880 205 cal BP, 2 Sigma). Additional radiocarbon ages from the upper part of M1 imply that a soil has formed 206 after the deposition of the sediment. The numerical data suggest that this soil formation started by ca. 207 5,900 cal BP (2 Sigma). M1 was buried by M2 at ca. 3.93 +/- 0.3 ka BP according to an OSL age (backed 208 by a radiocarbon age of soil organic matter). M2 (1.95- 3.0 m) has a slightly paler color (dark grayish 209 brown), and, while still dominated by silt, a significant increase in sand (coarse and middle sand). In 8 210 the upper part of M2 another soil has formed from ca. 2,750 cal BP until it became buried by M3. 211 Whether M3 was deposited during Iron Age or Medieval Times is not clear due to sparse numerical 212 age information. Data from the other exposures within the catchment area point to the former. 213 Changes in the sediment composition could be used to subdivide M3. A change in sediment color 214 (darker), grain size (little sand), and the C:N ratio of the sediment indicates a former soil surface (A- 215 horizon, soil formation) in a depth of ca. 1.5 m, coinciding with a radiocarbon age of ca. 910- 730 cal 216 BP (Medieval Times). Another noticeable change of the sediment properties is visible in ca. 1.0 m 217 depth. Similarly, few sand, additionally higher clay content, a switch to darker sediment colors and 218 wider C:N ratios indicate another former surface horizon (A-horizon, soil formation). Thus, although 219 not dated numerically the deposition of an Iron Age colluvium followed by two subsequent colluvial 220 layers could be derived from the sediment properties. 221 The nC27/(nC27+31) plant wax alkane ratio of the sediment indicates increasing amount of tree leaves 222 within the soil organic matter comparing the Late Glacial to mid-Holocene sediment record. It is the 223 smallest in one YD sample, increases in the samples of the early Holocene layer, and further to a more 224 tree-dominated value in the mid-Holocene samples. 225 Trenches at the lower slopes 226 At the lower slopes that incline towards the studied valley (trenches 2, 3, 5, 6), varying but smaller 227 thicknesses of sediments of water erosion were exposed (Fig. 1, 2; between 1-2 m). All sediments are 228 composed of silt, clay, and fine sand, and containing no significant amount of coarser particles. There 229 are different occurrences of Late Glacial to early Holocene sediments (trenches 2, 3). In one trench, a 230 thin Early Bronze Age colluvium was detected (trench 3). All trenches contain a colluvial layer that 231 dates to ca. 4,000 cal BP. In two trenches, the presence of a sediment deposited ca. 2,700- 2,300 yrs 232 cal BP (trenches 2, 5) is proven. In all trenches, spurs of buried soils are present. At the base of the 233 trenches, remnants of a buried Bw-horizon (Cambisol) indicate the presence of a wooded landscape 234 prior to the nowadays-widespread Chernozems. Additionally, pronounced A-horizons subdivide the 235 sediment sequences indicating a succession of alternating phases of slope stability and erosion 9 236 throughout the younger Quaternary. Within the YD sediment deposited at trench 2, a humic surface 237 soil horizon has formed dating to ca. 5,900- 5,650 yrs cal BP. In trench 3, similar phases of soil formation 238 are indicated. These occurred in the upper part of the early Holocene colluvial layer at ca. 7,800-7,600 239 yrs cal BP until burying at ca. 5,000- 4,900 yrs cal BP and in the colluvial layer suspicious to have been 240 deposited at ca. 4,000 yrs BP at ca. 3,900-3,700 yrs cal BP until burying at ca. 3,000- 2,900 yrs cal BP. 241 In general, the sediments and soils exposed at the lower slopes resemble the chronostratigraphy 242 detected in the long percussion-drilling core at the colluvial fan. Fig. 2 b and c illustrate properties of 243 the deposited sediments and soils in the trenches 2 and 3. Noteworthy is the comparable similar grain 244 size distribution (mainly silt with some clay) in trench 2 and 3. This might be explained by their 245 delivering sediment sources comprising of Loess at the investigated slopes. While there are similar 246 trends in LOI, magnetic susceptibility and colors of the sediment sequences in trench 2 and 3, there is 247 an obvious difference at the base of the Holocene part of the sequences. All, the LOI 940 values, the 248 magnetic susceptibility and the colors in trench 2 show an abrupt step at this chronostratigraphical 249 border whereas there is a gradual transition in trench 3. This indicates an erosional discordance in 250 trench 2 between the Late Glacial and the mid-Holocene. Erosion of parts of the soil developed in the 251 Late Glacial deposit immediately before the onset of soil formation (ca. 5,900- 5,650 yrs cal BP) seems 252 the most probable reason for the observed data. 253 An additional exposure was studied in a small quarry ca. 3 km southwest of the investigated catchment 254 area (trench 4). Whereas the start of erosion was found to have happened ca. 3,700- 3,500 yrs cal BP, 255 the subsequent colluvial layer dates to ca. 2,700- 2,400 yrs cal BP, resembling an erosional phase 256 detected in the investigated valley. A pronounced buried Bw-horizon is present at the base of the 257 sequence. 258 259 260 10 261 4. Discussion 262 A comparison of the reconstructed phases of erosion and soil formation with the well known 263 settlement history of the region and Holocene erosion histories from the Russian Plain and Germany 264 is given in Fig. 3. The data from the investigated trenches and the percussion-drilling core indicate that 265 the younger Quaternary erosion at the sites occurred in discrete phases. Slight deviations between 266 datings can be ascribed to uncertainties in using bulk samples for radiocarbon dating. A comparison 267 with the settlement history, thoroughly investigated through extensive archaeological surveys and 268 excavations near the research area shows a conspicuous non-coincidence between land-use and 269 erosion history. The only noticeable exception is the last millennium, where we do not have numerical 270 age information about the sediment deposition. No traces of erosion were found to be related with 271 the phases with the largest number of prehistoric settlements in the area (20 km radius) at ca. 6,450- 272 5,350 yrs cal BP (Tripyllia culture) or at ca. 1,700-1,500 yrs cal BP (Late Roman Iron Age). This 273 strengthens the opinion of a group of eastern European geomorphologists that Holocene erosion in 274 Eastern Europe was mainly driven by climate variability (Sycheva et al., 2003, Belyaev et al. 2004, 2005, 275 Sycheva 2006, Panin et al. 2009). A comparison of the numerical ages of the detected erosion phases 276 reveals a weak accordance between the results from central Ukraine and the Russian Plain for some 277 erosion phases. Whereas the records from Russia show no pronounced consistence viewed by itself, 278 the erosion phases at ca. 8.0 kyrs BP, ca. 4,000 yrs cal BP, at ca. 2,700-2,300 yrs cal BP and during the 279 last millennium detected in central Ukraine are also visible in the Russian record. 280 Considering them separately, all erosion phases detected at Maidanetske coincide with periods of 281 known extreme climatic conditions or rapid climate variability. 282 An in generally cooler and drier than today environment has been reconstructed for the LGM (e.g. 283 Lowe et al., 2008). Large regions of the non-glaciated forelands were characterized by permafrost (e.g. 284 Renssen and Vandenberghe, 2003), leading to increased amounts of runoff during summer thawing or 285 precipitation events (Panin et al., 2009). This resulted in widespread increased erosion processes as 286 described for the Mediterranean (Rossato and Mozzi, 2016) or Russia (Panin et al. 2009). Of 19 gullies 11 287 studied by Panin et al. (2009) in central Russia 15 were incised initially already during the Pleistocene. 288 The deposition of a sediment in the sequence of Maidanetske, rich in stones and sand, at 26.5 +/- 0.7 289 ka cal BP could have been related to an intense runoff event on partly frozen ground. Its coarse texture 290 might reflect high runoff energy and resulting incision of gullies/ channels into the bedrock. Loess 291 contributed, if even, only a small amount to the sediment. 292 The YD climate oscillation is well studied in a large number of palaeoenvironmental archives (e.g. Bar- 293 Matthews et al., 1997, Brauer et al., 2001, Andersen et al., 2004, Dykoski et al., 2005, Staubwasser and 294 Weiss, 2006, Bordon et al., 2009) and characterized as a cold and dry phase across Europe. Slope 295 instability associated with abrupt climate change has been reported from various sites in Europe (e.g. 296 Andres et al., 2001, Dotterweich et al., 2013) or Anatolia (e.g. Dreibrodt et al., 2014). Regardless if 297 permafrost processes affected the research region during the YD, the vegetation cover and thus the 298 shelter of the surface soil was very probably affected by climate change. These conditions could explain 299 the observed erosion phase in central Ukraine by runoff events produced during water rich snow-melts 300 or intensive precipitation events on unsheltered surface soils. The layers detected at two points in the 301 sedimentation area contain a large amount of silt, indicating the presence of a Loess cover in the 302 catchment area that was not cut through by the erosion processes. 303 The detection of a slope instability phase at ca. 8,000 yrs cal BP coincides with another well-known 304 climate oscillation phase (e.g. Alley and Ágústsdóttir, 2005). Response to this phase of rapid climate 305 change has been reported widespread from different types of palaeoenvironmental archives, such as 306 lakes (e.g. Migowski et al., 2006, Prasad et al., 2007, Bordon et al., 2009), tree rings (e.g. Spurk et al., 307 2002), or speleothems (e.g. Bar-Matthews et al., 1997, Bar-Matthews and Ayalon, 2011). While it is 308 accepted that the 8 ka BP phase was related to cold conditions in the northern mid-latitudes its 309 hydrologic impact is less clear. In spite of few evidence for flooding (e.g. Macklin et al., 2006) most 310 researchers interpret the occurrence of slope instability as a result of wetter conditions (e.g. Zolitschka 311 and Negendank, 1998). However, dry spells, which led to a destruction of the vegetation cover 312 (wildfires), might provide an alternative reason for slope instability (e.g. Dreibrodt et al. 2010b). Since 12 313 lake level highstands were used as an additional argument for wetter conditions across western and 314 central Europe (e.g. Magny et al., 2003) it might be considered that both, colder temperatures and a 315 sparser vegetation cover in the lakes catchment might also result in lake level increases. From the 316 eastern Mediterranean, there is indication for drier climate conditions at around 8,000 cal BP (e.g. Bar- 317 Matthews et al., 1997, Migowski et al., 2006, Bar-Matthews and Ayalon, 2011). Some scholars even 318 argued about a close relationship between the climate anomaly and early societal evolution in the 319 Mediterranean (Weninger et al., 2006). Investigations on slope deposits have revealed a pronounced 320 phase of slope instability at this interval reported from sites as distant as western and central Europe 321 (e.g. Dreibrodt et al., 2010b, Vincent et al., 2010, Lubos et al., 2011, Schumacher et al., 2018) or 322 Anatolia (Dreibrodt et al., 2014). The 8.0 ka climate oscillation is considered to have been of smaller 323 amplitudes in temperature and moisture changes as well as duration compared with the YD phase. 324 Effects of permafrost or enduring changes of the vegetation cover are less probable to explain the 325 observed erosion in central Ukraine. A weakened vegetation cover could have well played a role, but 326 an accentuation of patterns of precipitation events is also quite possible. 327 The erosion phase at ca. 4,000 yrs cal BP coincides with a climate anomaly reported from different 328 sites across Eurasia. Whereas northern Europe and the Alps experienced a colder than usual phase 329 (e.g. Bakke et al., 2010, Le Roy et al., 2017) from southern Europe and the Mediterranean the climate 330 oscillation is rather known because of prominent drought phases (e.g. Weiss and Bradley, 2001, 331 Staubwasser and Weiss, 2006, Migowski et al., 2006, Cheng et al., 2010, Schirrmacher et al., 2019). A 332 prominent dry phase was also reconstructed from the lake level of Lake Balqash (Kremenetski, 1997) 333 and through pollen studies for the research region in the period from ca. 4,300 to 3,600 yrs cal BP 334 (Gerasimenko, 1997). Intensive erosion during the period was detected in Greece (e.g. van Andel et 335 al., 1990) or Anatolia (Dusar et al., 2014). Thus, accentuated precipitation events during an in general 336 drier than usual phase with a weakened vegetation cover, could explain the erosion phase detected at 337 Maidanetske. 13 338 Between ca. 2,700 and 2,300 yrs cal BP another erosion phase occurred at Maidanetske. This coincides 339 with a climatic deterioration phase recorded across western and central Europe (e.g. van Geel et al., 340 1996). Prominent dry conditions were reconstructed for ca. 3,000- 2,000 cal BP from marine sediments 341 of the eastern Mediterranean (Schilman et al., 2001) and for the period between ca. 2,700- 2,000 cal 342 BP from the lake level of Lake Balqash (Kremenetski, 1997). Pollen studies from the research region 343 indicate a drier than usual phase from ca. 3,000 to 2,400 yrs cal BP (Gerasimenko, 1997). In central 344 Europe, frequent erosion has been reported from a large number of sites during this period (e.g. Lang, 345 2003, Dreibrodt et al., 2010a), including phases of gullying (Dreibrodt and Wiethold, 2015). Note the 346 presence of a high number of colluvial layers deposited in Germany in the period between 2,700 to 347 2,300 yrs cal BP (Fig. 3). Erosion is reported during the period from Anatolia (Kaniewskie et al., 2008, 348 Dreibrodt et al., 2014, Dusar et al., 2014) and Greece (van Andel et al., 1990, Fuchs, 2007), additionally. 349 Thus, accentuated precipitation events during a generally drier than usual phase with a weakened 350 vegetation cover, could explain the erosion phase detected at Maidanetske. 351 Since we do not have numerical age information about the erosion processes that were in action during 352 the past millennium at Maidanestke, we can only state that this phase was the strongest influenced by 353 intensive agricultural land use at the research site. Maxima of erosion are reported from central Europe 354 (e.g. Bork and Lang, 2003, Dotterweich, 2008, Dreibrodt et al., 2010a) and Russia (Panin et al., 2009) 355 to have happened during this period. If we consider the record at the colluvial fan in core 1 we could 356 deduce that about 150 cm of the Holocene record was deposited during the last 1,000 years 357 (representing ca. 42 % of the Holocene sediment). That underlines again the crucial importance of 358 intensive agricultural land use on Holocene soil erosion processes. Additionally, it implies that the 359 intensity of prehistoric land use was below a critical threshold, thus no or very little soil erosion was 360 triggered by their subsistence systems. 361 Summarizing the discussion of the long-term Younger Quaternary erosion history at Maidanetske 362 (LGM- 1,000 yrs BP) there is a non-coincidence of erosion with the local and regional settlement history 363 but an obvious pattern of coincidence of erosion at the site with well-known phases of climate 14 364 anomalies. The latter reflect anomalies reported from western and central Europe and the 365 Mediterranean climate system. Their visibility in central Ukraine might reflect the convergence of the 366 two climate systems in that part of Eastern Europe. As the climate anomalies conspicuous to have 367 resulted in the observed erosion were characterized by similar conditions (colder than usual in central 368 and western Europe and drier than usual in the eastern Mediterranean and the research area) a 369 common trigger of the observed erosion phases might be possible. Episodic occurrences of more 370 intensive than usual precipitation events in the research area one a perhaps weakened vegetation 371 could explain the observed record. This is corroborated by the accordance of dating of sediment layers 372 at the different investigation points that implies discrete phases of Holocene erosion. A response of 373 the local vegetation cover to slight climatic changes seems probable considering the position of the 374 site in the sensitive ecotone of the forest-steppe transition. If occurrences of heavy precipitation 375 events coinciding with the climate anomalies were triggered by short response mechanisms of the 376 climate system as occurrences of meridional transfer of heat and water from the eastern 377 Mediterranean towards the interior of Eurasia remains speculative and is a matter of ongoing research. 378 The sensitivity of the central Ukrainian landscape we claim here is probably related to two 379 preconditions. The first is the late onset of intensive agricultural land use in the region, similar as 380 pointed out for Russia (Panin et al., 2009). This is visible in the thick layer of colluvial material deposited 381 during the last millennium in our long percussion-core. The second precondition is related to the 382 location of the area in the forest-steppe borderland zone, considered to be sensitive to slight climatic 383 changes and, additionally located in a position where western and southern European climate systems 384 converge. 385 Considering the erosion processes in action during the Younger Quaternary deposition history an 386 additional observation could be made. The sediment deposited during the periglacial to early Holocene 387 erosion processes show properties that resemble the Loess cover deposited over the whole catchment 388 area (Fig. 2). Since the 4,000 yrs cal BP erosion phase, the sediment on the colluvial fan contains more 389 sand in general. This is not visible in the trenches at the lower slopes, where the Loess cover was 15 390 nowhere found to have been cut through completely. This hints to the start of a stronger incision in 391 the valley itself and aggradation of colluvial material at the lower slopes. Additionally, the biomarker 392 signal of increasing amounts of tree leave organic matter in the valley sediments points to erosion and 393 redeposition of soil in the valley bottom, because the valley bottom is the most probable place for the 394 growth of gallery forests throughout the Holocene. Thus, a change in the overall geomorphic 395 connectivity within the investigated catchment area occurred at the mid-Holocene (since 4,000 yrs cal 396 BP). This could reflect changes in the intensity of the reconstructed erosional events in an order (from 397 stronger to weaker): LGM > YD > early Holocene >> mid-Holocene. 398 399 5. Conclusions 400 A long-term Younger Quaternary erosion history mainly driven by climate variability was reconstructed 401 at a central Ukrainian site. This is in accordance with observations from neighboring regions. It might 402 reflect the late onset of intensive agricultural land use in the region and the position of the site in an 403 environment sensitive to slight climatic shifts where the western and southern European climate 404 systems converge. Additionally, in western, central and southern European records of Holocene 405 erosion response to climate variability might be present but masked by the anthropogenically 406 intensified erosion of early intensive land use. 407 Changes in the properties of the sediment deposited at a colluvial fan indicate a change from a stronger 408 connectivity of erosion processes during the glacial to early Holocene erosion phases towards a 409 weakened connectivity since the mid-Holocene (4,000 yrs cal BP). 410 411 Acknowledgements 412 We are grateful to the DFG for funding in the frame of the CRC 1266 “Scales of transformation”. Many 413 thanks to the landowners for allowance of the fieldwork, and Imke Meyer, Manfred Beckers, Tine 414 Karck, Katie Lehnen, and a group of students of the University Kiel for help in the laboratory. 16 415 References 416 AG Boden, 2005. Bodenkundliche Kartieranleitung. 5. verbesserte und erweiterte Aufl. 417 Schweizerbart'sche Verlagsbuchhandlung, Stuttgart. 418 Alley, R.B., Ágústsdóttir, A.-M., 2005. The 8k event: cause and consequences of a major Holocene 419 abrupt climate change. Quaternary Science Reviews 24, 1123–1149. 420 Andersen, K.K., Azuma, N., Barnola, J.M., Bigler, M., Biscaye, P., Caillon, N., et al., High-resolution 421 record of Northern Hemisphere climate extending into the last interglacial period. Nature 431, 147- 422 151. 423 Andres, W., Bos, J.A.A., Houben, P., Kalis, A., Nolte, S., Rittweger, H., Wunderlich, J., 2001. 424 Environmental change and fluviual activity during the Younger Dryas in central Germany. Quaternary 425 International 79, 89–100. 426 Atlas of Soils of the Ukrainian Socialistic Soviet Republic, 1979. Harvest, Kiew. 427 Bakke, J., Dahl, S.O., Paasche, O., Riis Simonsen, J., Kvisvik, B., Bakke, K., Nesje, A., 2010. A complete 428 record of Holocene glacier variability at Austre Okstindbreen, northern Norway: An integrated 429 approach. Quaternary Science Reviews 19, 1246-1262. 430 Bar-Matthews, M., Ayalon, A., 2011. Mid-Holocene climate variations revealed by high-resolution 431 speleothem records from Soreq Cave, Israel and their correlation with cultural changes. The Holocene 432 21(1), 163–171. 433 Bar-Matthews, M., Ayalon, A., Kaufman, A., 1997. Late Quaternary palaeoclimate in the eastern 434 Mediterranean region from stable isotope analysis of speleothems at Soreq Cave, Israel. Quaternary 435 Research 47: 155–168. 436 Belayev, Y.R., Panin, A.,V., Belayev, V.,R., 2004. Climate-induced and local-scale erosion and 437 sedimentation features in small catchments: Holocene history of two small valleys in Central Russia. 438 Sediment Transfer through the Fluvial System, IAHS Publ. 288, 3-12. 17 439 Belayev, V.R., Eremenko, E.A., Panin, A.V., Belayev, Y.R., 2005. Stages of Late Holocene gully 440 development in the central Russian plain. International Journal of Sediment Research 20, 224-232. 441 Bordon, A., Peyron, O., Lézine, A.-M., Brewer, S., Fouache, E., Pollen-inferred Lat-Glacial and 442 Holocene climate in southern Balkans (Lake Maliq). Quaternary International 200, 19-30. 443 Bork, H.-R., Lang, A., 2003. Quantification of past soil erosion and land use/ land cover changes in 444 Germany. In: Lang, A., Hennrich, K., Dikau, R. (Eds.), Long Term Hillslope and Fluvial System 445 Modelling: Concepts and Case Studies from the Rhine River Catchment. Springer, Heidelberg, pp. 446 231-239. 447 Brauer, A., Litt, T., Negendank, J.F.W., Zolitschka, B., 2001. Lateglacial varve chronology and 448 biostratigraphy of lakes Holzmaar and Meerfelder Maar, Germany. Boreas 30, 83–88. 449 Bronk Ramsey, C., Lee, S., 2013. Recent and planned developments of the program OxCal. 450 Radiocarbon 55, 3–4. 451 Butzer, K.W., 2005. Environmental history in the Mediterranean world: cross-disciplinary 452 investigation of cause-and-effect for degradation and soil erosion. Journal of Archaeological Science 453 32, 1773-1800. 454 Buylaert et al., 2012. A robust feldspar luminescence dating method for Middle and Late Pleistocene 455 sediments. Boreas 41, 435–451. 456 Cheng, H., Sinha, A., Verheyden, S., Nader, F.H., Li, X., Peng, Y.B., Rao, Z.H., Ning, Y.F., Edwards, R.L., 457 2015. The climate variability in northern Levant over the past 20,000 years. Geophysical Research 458 Letters 22, 8641-8650. 459 Dean, W.E., 1974. Determination of Carbonate and Organic Matter in Calcareous Sediments and 460 Sedimentary Rocks by Loss on Ignition: Comparison with Other Methods. Journal of Sedimentary 461 Petrology 44, 242–48. 462 Dearing, J., 1999. Environmental Magnetic Susceptibility: Using the Bartington MS2 System. 2nd ed. 463 (Bartington Instruments Ltd.). 18 464 Dotterweich, M., 2008. The history of soil erosion and fluvial deposits in small catchments of central 465 Europe: deciphering the long term interaction between human and environment – a review. 466 Geomorphology 101, 192–208. 467 Dotterweich, M., Kühn, P., Tolksdorf, J.F., Müller, S., Nelle, O., 2013. Late Pleistocene to Early 468 Holocene natural and human influenced sediment dynamics and soil formation in a 0-order 469 catchment in SW-Germany (Palatinate Forest), Quaternary International 306, 42-59. 470 Dreibrodt, S., Wiethold, 2015. Lake Belau and its catchment (northern Germany): A key archive of 471 environmental history in northern central Europe since the onset of agriculture. The Holocene 25(2), 472 296–322. 473 Dreibrodt, S., Lubos, C., Terhorst, B., Damm, B., Bork, H.-R., 2010a. Historical soil erosion by water in 474 Germany: Scales and archives, chronology, research perspectives. Quaternary International 222, 80– 475 95. 476 Dreibrodt, S., Lomax, J., Nelle, O., Lubos, C., Fischer, P., Mitusov, A., Reiss, S., Radtke, U., Nadeau, M, 477 Grootes, P.M., Bork, H.-R., 2010b. Are mid-latitude slopes sensitive to climatic oscillations? 478 Implications from an Early Holocene sequence of slope deposits and buried soils from eastern 479 Germany. Geomorphology 122, 351–369. 480 Dreibrodt, S., Lubos, C., Lomax, J., Sipos, G., Schroedter, T., Nelle, O., 2014. Holocene landscape 481 dynamics at the tell Arslantepe, Malatya, Turkey – Soil erosion, buried soils and settlement layers, 482 slope and river activity in a middle Euphrates catchment. The Holocene 24, 1351–1368. 483 Dusar, B., Verstraeten, G., D´Haen, K., Bakker, J., Kapijn, E., Waelkens, M., 2012. Sensitivity of the 484 Eastern Mediterranean geomorphic system iowards environmentla change during the Late Holocene: 485 a chronological perspective. Journal of Quaternary Science 27, 371-382. 486 Dykoski, C.A., Edwards, R.L., Cheng, H., Yuan, D., Cai, Y., Zhang, M., Lin, Y., Qing, J., An, Z., Revenaugh, 487 J., 2005. A high-resolution, absolute dated Holocene and deglacial Asian monsoon record from 488 Dongge cave, China. Earth and Planetary Science Letters 233, 71-86. 489 Fuchs, M., 2007. An assessment of human versus climatic impacts on Holocene soil erosion in NE 490 Peloponnese, Greece. Quaternary Research 67, 349–356. 19 491 Galbraith R.F., Roberts R.G., Laslett G.M., Yoshida H., Olley J.M., 1999. Optical dating of single and 492 multiple grains of quartz from Jinmium Rock Shelter, northern Australia: Part 1, experimental design 493 and statistical models. Archaeometry 41, 339-364. 494 Gerasimenko, N., 1997. Environmental and Climatic Chnages between 3 and 5 ka BP in Southeastern 495 Ukraine. In: Dalfes, H.N., Kukla, G., Weiss, H. (Eds.), Third Millennium BC Collapse and Old Word 496 Change. Springer, Berlin, pp. 371-399. 497 Hofmann R, Müller J, Shatilo L, Videiko, M, Ohlrau R, Rud V, Burdo, N. DalCorso, M., Dreibrodt, S., 498 Kirleis, W., 2019. Governing Tripolye: Integrative architecture in Tripolye settlements. PLoS ONE 499 14(9): e0222243 500 Ivanova, 2016. С. В. Иванова. Курганы vs поселения: скотоводы vs земледельцы// Культурные 501 взаимодействия. Динамика и смыслы. Сборник статей в честь 60-летия И.В. Манзуры. Кишинёв: 502 Stratum Plus, 273-291. 503 ІУМІС, 1972. Історія міст і сіл Української РСР: В 26 т. Черкаська область. Том 24. / Ред. кол. тому: 504 Стешенко О. Л. (гол. редкол.), Гольцев Є. М., Горкун А. І., Дудник О. М., Зайцев М. С., Звєрєв С. 505 М., Зудіна Г. М., Коваленко В. Я., Кузнецов С. М., Курносов Ю. О., Непийвода Ф. М., Степаненко 506 А. О., Тканко О. В. (заст. гол. редкол.), Храбан Г. Ю., Червінський О. А. (відп. секр. редкол.), Шпак 507 В. Т. АН УРСР. Інститут історії. – К.: Голов. ред. УРЕ АН УРСР, 1972. – 788 с. 508 Kaniewski, D., Paulissen, E., De Laet, V., Waelkens, M., 2008. Late Holocene fire impact and post-fire 509 regeneration from Bereket basin, Taurus Mountains, southwest Turkey. Quaternary Research 70, 228- 510 239. 511 Kremenetski, C.V., 1997. The Late Holocene Environmental Shift in Russia and Surrounding lands. In: 512 Dalfes, H.N., Kukla, G., Weiss, H. (Eds.), Third Millennium BC Collapse and Old Word Change. Springer, 513 Berlin, pp. 351-370. 20 514 Kruts et al., 1981. Круц В.А., Рыжов С.Н. Отчет о работе Тальянковского отряда Трипольской 515 экспедиции в 1981 г. // НА ИА НАНУ, 11б. 516 Kushtan, 2013. Куштан Д.П. Периодизация и хронология памятников эпохи поздней бронзы 517 Центральной Украины // Проблемы периодизации и хронологии в археологии эпохи раннего 518 металла Восточной Европы: Матер. тематич. науч. конференц. Санкт-Петербург, 4-6 декабря 519 2013 г. — СПб.: «Скифия-принт», 80-85. 520 Lang, A., 2003. Phases of soil erosion-derived colluviation in the loess hills of South Germany. Catena 521 51, 209-221. 522 Le Roy, M., Deline, P., Carcaillet, J., Schimmelpfennig, I., Ermini, M., 2017. 10Be exposure dating of 523 the timing of Neoglacial glacier advances in the Ecrins-Pelvoux massif, southern French Alps. 524 Quaternary Science Reviews 178, 118-138. 525 Liritzis, I., Stamoulis, K., Papachristodoulou, C., Ioannides, K., 2013. A re-evaluation of radiation dose- 526 rate conversion factors. Mediterranean Archaeology and Archaeometry 13, 1-15. 527 Lowe, J.J., Rasmussen, S.O., Björck, S., Hoek, W.Z., Steffensen, J.P., Walker, M.J.C., Yu, Z.C., INTIMATE 528 group, 2008. Synchronisation of palaeoenvironmental events in the North Atlantic region during the 529 Last Termination: a revised protocol recommended by the INTIMATE group. Quaternary Science 530 Reviews 27, 6-17. 531 Lubos, C.C.M., Dreibrodt, S., Nelle, O., Klamm, M., Friederich, S., Meller, H., Nadeau, J.M., Grootes, 532 P.M., Fuchs, M., Bork, H.-R., 2011. A multi-layered prehistoric settlement structure (tell?) at 533 Niederröblingen, Germany and its implications. Journal of Archaeological Science 38, 1101-1110. 534 Macklin, M., Benito, G., Gregory, K.J., Johnstone, E., Lewin, J., Michczynska, D.J., Soja, R., Starkel, L., 535 Thorndycraft, V.R., 2006. Past hydrological events reflected in the Holocene fluvial record of Europe. 536 Catena 66, 145-154. 21 537 Magny, M., Bégeot, C., Guiot, J., Peyron, O., 2003. Contrasting patterns of hydrological changes in 538 Europe in response to Holocene climate cooling phases. Quaternary Science Reviews 22, 1589-1596. 539 Magomedov, B.V., Didenko, S.V., 2009. The Cemetery of Chernyakhov Culture by village Legedzino. 540 Excavation in 2008-2009. in: Kruts, V.A., Korvin-Piotrovskiy, A.G., Menotti, F., Ryzhov, S.N., Tolochko, 541 D.V., Chabanyuk, V.V. (Eds.), Talianki - settlement-giant of the Tripolian culture. Investigations in 2009. 542 Kiev, 2009, pp. 56-92. 543 Migowski, C., Stein, M., Prasad, S., Negendank, J.F.W., Agnon, A., 2006. Holocene climate variability 544 and cultural evolution in the Near East from the Dead Sea sedimentary record. Quaternary Research 545 66, 421-431. 546 Murray A.S., Wintle A.G., 2000. Luminescence dating using an improved single-aliquot regenerative- 547 dose protocol. Radiation Measurements 32, 57-73. 548 Murray A.S., Wintle A.G., 2003. The single aliquot regenerative dose protocol: Potential for 549 improvements in reliability. Radiation Measurements 37, 377-381. 550 Müller, J., Hofmann, R., Kierleis, W., Dreibrodt, S., Ohlrau, R., Brandtstätter, L., DalCorso, M., Out, W., 551 Rassmann, K., Burdo, N., Videiko, M., 2013. Maidanestke 2013. New excavations at a Tripyllia Mega- 552 site. Studien zur Archäologie in Ostmitteleuropa 16, Habelt, Bonn. 553 Müller, J., Rassmann, K., Videiko, M., 2016.Trypilla Mega-Sites and European Prehistory 4100–3400 554 BCE. Routledge Taylor and Francis Group, London. 555 Neradenko, 2011. Нераденко Т.М. Археологія Черкащини: Посібник-довідник. – Черкаси: Вид. 556 Чабаненко Ю.А., 307. 557 Ohlrau, R., 2018. Maidanets’ke. Development and decline of a Trypillian mega-site in Central Ukraine. 558 PhD Thesis, University of Kiel, Kiel. 559 Panin, A.V., Fuzeina, J.N., Belyaev, V.R., 2009. Long-term development of Holocene and Pleistocene 560 gullies in the Protva River basin, Central Russia. Geomorphology 108, 71-91. 22 561 Pickartz, N., Hofmann, R., Dreibrodt, S., Rassmann, K., Shatilo, L., Ohlrau, R., Wilken, D., Rabbel, W., 562 2019. Deciphering archeological contexts from the magnetic map: Determination of daub distribution 563 and mass of Chalcolithic house remains. The Holocene 29, 1637-1652. 564 Prasad, S., Brauer, A., Rein, B., Negendank, J.F.W., 2007. Rapid climate change during the early 565 Holocene in Western Europe and Greenland. The Holocene 16 (2), 153–158. 566 Prescott, J.R., Hutton, J.T., 1994. Cosmic ray contributions to dose rates for luminescence and ESR 567 dating: large depths and long term variations. Radiation Measurements, 23, 497-500. 568 Rabenhorst, M.C., Schmehling, A., Thompson, J.A., Hirmas, D.R., Graham, R.C., Rossi, A.M., 2014. 569 “Reliability of Soil Color Standards.” Soil Science Society of America Journal 79, 193–99. 570 Reimer, P.J., Baillie, M.G.L., Bard, E., Bayliss, A., Beck, J.W., Bertrand, C., Blackwell, P.G., Buck, C.E., 571 Burr, G., Cutler, C.B., Damon, P.E., Edwards, R.L., Fairbanks, R.G., Friedrich, M., Guilderson, T.P., 572 Hughen, K.A., Kromer, B., McCormack, F.G., Manning, S., Bronk Ramsey, C., Reimer, R.W., Remmele, 573 S., Southon, J.R., Stuiver, M., Talamo, S., Taylor, F.W., van der Pflicht, J., Weyhenmeyer, C., 2004. 574 IntCal04 terrestrial radiocarbon age calibration, 26–0 ka BP. Radiocarbon 46, 1029–1058. 575 Renssen, H., Vandenberghe, J., 2003. Investigation of the relationship between permafrost 576 distribution in NW Europe and extensive winter sea-ice cover in the North Atlantic Ocean during the 577 cold phases of the Last Glaciation. Quaternary Science Reviews 22, 209–223. 578 Rossato, S., Mozzi, P., 2016. Inferring LGM sedimentary and climatic changes in the southern Eastern 579 Alps foreland through the analysis of a 14C ages database (Brenta megafan, Italy). Quaternary Science 580 Reviews 148, 115-127. 581 Sanmartín, P., Chorro, E., Vázquez-Nion, D., Miguel Martínez-Verdú, F., Prieto, B., 2014. Conversion 582 of a Digital Camera into a Non-Contact Colorimeter for Use in Stone Cultural Heritage: The 583 Application Case to Spanish Granites. Measurement 56, 194–202. 584 Schilman, B., Bar-Matthews, M., Almogi-Labin, A., Luz, B., 2001. Global climate instability reflected by 585 Eastern Mediterranean marine records during the late Holocene. Palaeogeography, 586 Palaeoclimatology, Palaeoecology 176: 157–176. 23 587 Schirrmacher, J., Weinelt, M., Blanz, T., Andersen, N., Salgueiro, E., Schneider, R.R., 2019. Multi- 588 decadal atmospheric and marine climate variability in southern Iberia during the mid- to late- 589 Holocene. Climate of the Past 15, 617-634. 590 Schumacher, M., Dobos, A., Schier, W., Schütt, B., 2018. Holocene valley incision in the southern 591 Bükk foreland: Climate-human-environment interferences in northern Hungary. Quaternary 592 International, 463, 91-109. 593 Shidlovsky et al, 2004. Шидловський П.С., Пічкур Є.В., Чорновол Д.К. (2004) Археологічні 594 дослідження поблизу села Аполянка на Уманщині. В Археологічні відкриття в Україні 2002-2003, 595 рр 361-365. 596 Spurk, M., Leuschner, H.H., Baillie, M.G.L., Briffa, K.R., Friedrich, M., 2002. Depositional frequency of 597 German subfossil oaks: climatically and non-climatically induced fluctuations in the Holocene. The 598 Holocene 12 (6), 707–715. 599 Staubwasser, M., Weiss, H., 2006. Holocene climate and cultural evolution in late prehistoric-early 600 historic West Asia. Quaternary Research 66, 372-387. 601 Sycheva, S.A., 2006. Long-term pedolithogenic rhythms in the Holocene. Quaternary International 602 152-153, 181-191. 603 Sycheva, S., Glasko, M., Chichagova, O., 2003. Holocene rhythms of soil formation and sedimentation 604 in the Central Russian Upland. Quaternary International 106-107, 203-213. 605 Terenozhkin, 1961. Тереножкин А.И. Предскифский период на Днепровском Правобережье. — 606 Киев: Изд-во АН УССР, 1961, 248. 607 Thiel, C., Buylaert, P., Murray, A., Terhorst, B., Hofer, I., Tsukamoto, S., Frechen, M., 2011. 608 Luminescence dating of the Stratzing loess profile (Austria) Testing the potential of an elevated 609 temperature post-IR IRSL protocol. Quaternary International 234, 23–31. 610 Thornes, J., 2009. Land Degradation. In: Woodward, J. (Ed.), The Physical Geography of the 611 Mediterranean, Oxford University Press, Oxford, pp. 563-581. 24 612 Van Andel, T., Zangger, E., Demitrack, A., 1990. Land use and soil erosion in Prehistoric and Historical 613 Greece. Journal of Field Archaeology 17, 379-396. 614 van Geel, B., Buurman, J., Waterbolk, H.T., 1996. Archaeological and palaeoecological indications of 615 an abrupt climate change in the Netherlands and evidence for climatological teleconnections around 616 2650 BP. Journal of Quaternary Science 11, 451–60. 617 Vincent, P.J., Lord, T.C., Twelfer, M., Wilson, P., 2010. Early Holocene loessic colluviation in northwest 618 England: new evidence for the 8.2ka event in the terrestrial record? Boreas 40, 105–115. 619 Weiss, H., Bradley, R.S., 2001. What did drive societal collapse? Science 291, 609–610. 620 Weninger, B., Alram-Stern, E., Bauer, E., Clare, L., Danzeglocke, U., Jöris, O., Kubatzki, C., Rollefson, 621 G., Todorova, H., van Andel, T., 2006. Climate forcing due to the 8200 cal yr B.P. event observed at 622 Early Neolithic sites in the Eastern Mediterranean. Quaternary Research 66: 401–420. 623 Wintle A.G., Murray, A.S., 2006. A review of quartz optically stimulated luminescence characteristics 624 and their relevance in single-aliquot regeneration dating protocols. Radiation Measurements 41, 369- 625 391. 626 Zalizniak et al., 2005. Залізняк Л. Л., Товкайло М. Т., Кухарчук Ю. В. Дослідження стоянок біля с. 627 Добрянка на Черкащині археологічною експедицією НаУКМА у 2001, 2003, 2004 рр.// 628 Маґістеріум. Вип. 20. Археологічні студії / [упоряд.: Л. Л. Залізняк], 6-17. 629 Zolitschka, B., Negendank, J.F.W., 1998. A high resolution record of Holocene palaeohydrological 630 changes from Lake Holzmaar (Germany). In: Frenzel, B. (Ed.), Palaeohydrology as reflected in lake- 631 level changes as climatic evidence for Holocene times: European Palaeoclimate and Man, vol. 17, pp. 632 37–52. 633 634 635 636 25 637 Figure captions 638 Figure 1. Location of the investigation site a) in Eastern Europe, b) the investigation points in the 639 valley of the Talyanki River close to the Tripyllia Giant Settlement Maidanetske (plan of burned 640 houses indicated), and c) simplified chronostratigraphy of the investigated trenches (number on the 641 left side of the columns: MUNSELL color values); data of core 1: Fig. 2a). 642 Figure 2. Selected laboratory data from a) the long percussion-drilling core 1, b) trench 3 and c) 643 trench 2. Fig. 2 a) TOC- red line, C/N ratio- black line; Fig. 2 c) LOI 500- upper axis, LOI 940- lower axis. 644 Figure 3. Comparison of the detected Late Quaternary Erosion phases at Maidanetske with the 645 known settlement history, and records of Holocene soil erosion from Russia (Sycheva, 2006, Panin et 646 al., 2009) and Germany (histogram: orange- dated via embedded/ buried archaeological record, 647 green- dated via radiocarbon dating, blue- dated via OSL, Dreibrodt et al., 2010a). 648 649 Tables 650 Table 1 Radiocarbon data 651 Table 2 OSL data 652 Table 3 Settlement history of the site (5 km radius) and the region (20 km radius) 653 654 655 656 657 658 26 a) Lit. Poland b) Russia Belarus 5 6 Trypillian se lements Maidanetske 32 Ecozones Moldova Romania montane complex mixed -steppe steppe semi-desert forests Ukraine 1 4 Depth [m] 0 1 2 MUNSELL Colors Trench 2 10YR 3/2 10YR 4/3 10YR 5/3 10YR 5/4 2.5Y 7/4 10YR 6/6 10YR 7/6 MUNSELL Colors Trench 3 2.5Y 5/2 5Y 3/1 2678-2345 4234-3991 2.5Y 5/2 5893-5657 2.5Y 3/2 10.9 +/10YR 4/2 0.97 2.5Y 5/3 25.2 +/1.3 MUNSELL Colors Trench 5 3058-2867 3871-3701 5039-4857 7819-7621 10YR 3/1 10YR 3/2 10YR 4/2 10YR 5/3 10YR 5/4 10YR 7/6 Trench 6 10YR 3/1 2724-2384 4063-3781 6306-6208 10YR 3/2 10YR 5/2 10YR 5/4 10YR 7/6 Trench 4 0 10YR 3/1 3321-3062 4085-3889 6188-5944 10YR 3/2 2721-2381 3687-3480 1 10YR 4/4 8172-8001 10YR 7/4 2.5Y 6/4 2.5Y 7/4 large Bronze Age po ery shard (2,000-1,400 BCE) slope deposits/ colluvial layers Late Glacial to Holocene Age MUNSELL Colors 2.5Y 7/1 pronounced A horizon humic soil horizons/ sediments fossil Bw horizon calcareous Loess (C) radiocarbon age cal 2 Sigma BP OSL age 1 Sigma BP 2 Depth [m] c) MUNSELL Colors a) radiocarbon ages (cal 2 Sigma BP)/ OSL ages (1 Sigma ka BP) 0 Stra graphy Grain Size Distribu on (< 2 mm) (%) MUNSELL COLORS 0 100 2.5 Y 4/1 2.5 Y 5/2 2.5 Y 4/2 10 YR 4/2 2.5 Y 4/2 10 YR 4/2 2.5 Y 4/2 150 2.5 Y 4/1 10 YR 4/2 2.5 Y 4/2 2.5 Y 4/1 50 Medieval/ Modern M4 911- 735 Iron Age M3 fAh M2 X 2426- 2162 2751- 2506 X X 5731- 5608 X (charcoal) M2 fAh M1 EH M1 4438- 4257 X 5320- 5052 X 5918- 5750 X 8.5±0.3 X 8159- 7878 2 4 6 nC27/(nC27+31) 8 150 Iron Age soil BronzeIron Age 10 YR 4/1 10 YR 4/2 2.5 Y 5/2 200 200 4,0 ka BP 2.5 Y 6/2 3.93±0.1 4,0 ka BP TOC [% by weight] Color RGB 250 2.5 Y 7/2 2.5 Y 6/2 2.5 Y 7/2 300 10 YR 6/2 2.5 Y 5/2 10 YR 6/2 trees 250 soil TripolyeYamnaya 300 EH grasses 2.5 Y 7/2 S2 350 2.5 Y 7/2 2.5 Y 8/4 400 Cc (Löss) 10 YR 7/6 2.5 Y 8/3 2.5 Y 7/4 2.5 Y 8/4 2.5 Y 8/3 2.5 Y 8/2 5 Y 8/1 Gl 1 8/N 450 26.5±0.7 LGM 350 2.5 Y 6/3 12.0±0.4 Late Glacial aeolian YD N below detec on limits YD X 12030- 11407 S1 500 0 25 25 25 0 0.1 0.2 100 C/N (org. ma er) Dating Stratigraphy X Loess slope deposit/ A-horizon colluvium b) sand 0 0 20 40 clay silt <2000 <630 <200 <63 <20 <6.3 ... ... ... ... ... ... 2 20 6.3 630 200 63 Grain Size Distribu on (<2mm) [%] (cal 2 Sigma BP) 60 80 <2 sus. mag. LF [10-5 m3/kg] LOI 550/ 940 [% by weight] 100 0 2 4 6 8 20 40 60 MUNSELL COLORS RGB Colors 0 200 400 600 2.5Y 5/2 M4 5Y3/1 3,058- 2,867 X fAh M3 4,0 ka BP radiocarbon datum OSL datum radiocarbon ages Modern to Iron Age Grain Size [µm] M3 3,871- 3,701 50 2.5Y 5/2 5Y3/1 X 2.5Y3/2 5,0 ka BP M2 5,039- 4,857 Loess 7,819- 7,621 8,172- 8,001 Iron Age 4,0 ka BP YD X X 200 Loess (redeposited?) radiocarbon ages (cal 2 Sigma BP)/ OSL ages (1 Sigma ka BP) 0 M3Ap M2 M1 2,678- 2,345 X 4,234- 3.991 X 3,950- 3,350 (potsherd) 5,893- 5,657 X S1 fAh 10.99 +/- 0.97 altered by fBw1 soil forma on fBw2 27.2 +/- 1.3 fBw2 Loess Periglacial 10YR5/3 2.5Y5/3 10YR5/3 10YR6/3 2.5Y6/4 2.5Y7/4 2.5Y6/3 2.5Y7/4 2.5Y7/3 10YR7/4 2.5Y6/4 150 fBw M1 c) Medieval/ Modern 100 10YR4/2 fAh M1 EH altered by soil forma on X Cc (Löss) Grain Size Distribution (< 2 mm) [%] 0 20 40 60 80 sus. mag LF [10-5 m³/kg] LOI 550/ 940 [% by weight] 100 0 4 2 8 6 20 40 MUNSELL COLORS 0 60 10YR 3/2 5 15 25 35 45 55 62 69 77 84 91 102 109 124 134 144 154 164 174 184 194 204 214 224 234 244 10YR 4/3 10YR 5/3 10YR 5/4 2.5Y 7/4 10YR 7/6 10YR 6/6 10YR 6/6 2.5Y 7/4 0 4 8 12 RGB Colors 200 400 600 0.3 Phases of Erosion at Maidanetske, central Ukraine Age yrs cal BP Core 1 Trench 2 Trench 3 Trench 5 Trench 6 Regional settlement history (20 km radius, dark grey: 5 km radius) Trench 4 Age yrs cal BP a) Number archaeological sites/ settlements 0 10 20 30 Erosion history Germany Erosion history Russian Plain 40 Age yrs cal BP b) rela ve probability 1 0 erosion soil forma on Age yrs cal BP 0 0 0 0 1 1 1 1 2 2 2 2 3 3 3 3 4 4 4 4 5 5 5 5 6 6 6 6 7 7 7 7 8 8 8 8 9 9 9 9 10 10 10 10 11 11 11 11 12 12 12 12 13 13 13 13 14 14 14 14 hiatus number of colluvial layers [n=306, 101 sites] 0 10 20 30 40 50 Table 1a Radiocarbon data Lab ID profile Kiel Kiel Kiel Beta Beta Beta Kiel Kiel Kiel 52670 53079 53078 529991 529992 529993 52669 53077 52667 Kiel 53076 1 1 1 1 1 1 1 1 1 1 Kiel 52668 Kiel Posznan Posznan Posznan Posznan Posznan Posznan Posznan Posznan 53075 62408 62410 62407 113975 113974 113973 113971 113970 Lab 1 1 2 2 2 3 3 3 3 3 Depth (cm) radiocarbon age BP 340-344 323-338 298-303 293-298 288-293 283-288 280-284 234-239 200-204 10130±55 6410±35 7175±55 5100±55 6710±30 2999±40 4550±40 3927±26 4949±27 194-199 2550±24 180-184 2310±40 144-149 95-100 65-70 45-50 150-155 120-125 95-100 60-65 40-45 895±30 5015±35 3755±30 2385±30 7260±40 6880±40 4370±30 3515±30 2840±30 cal 2 Sigma BP remarks 12030-11597(86.4%), 11561-11472(6.3%), 11454-11407(2.7%) 7420-7275(95.4%) 8159-8087(10.2%), 8069-7931(83.4%), 7893-7878(1.8%) 5918-5846(37.2%), 5831-5750(58.2%) 7653-7639(1.8%), 7624-7556(75.7%), 7545-7511(17.8%) 3336-3290(7.3%), 3261-3028(87.7%), 3014-3008(0.4%) 5320-5213(37.1%), 5193-5052(58.3%) 4438-4286(93.0%), 4273-4257(2.4%) 5731-5608(95.4 %) 2751-2698(67.2%), 2635-2617(8.2%), 2591-2537(15.9%), 2531-2506(4.1%) 2426-2392(2.1%), 2382-2302(76.9%), 2246-2178(15.8%), 2171-2162(0.7%) 911-735(95.4%) 5893-5805(38.9%), 5796-5781(2.5%), 5774-5657(54.0%) 4234-4198(10.4%), 4184-4070(68.6%), 4045-3991(16.3%) 2678-2667(1.3%), 2656-2644(1.6%), 2492-2345(92.5%) 8172-8001(95.4%) 7819-7814(0.6%), 7796-7621(94.8%) 5039-5005(9.7%), 4981-4857(85.7%) 3871-3701(95.4%) 3058-3049(1.5%), 3040-2867(93.9%) Sediment Sediment, oulier (krotowina?) Sediment Sediment, Soil formation Sediment, outlier, too few org. C Sediment, outlier, too few org. C Sediment, Soil formation Sediment Charcoal, outlier (redeposition?) Sediment, Soil formation Sediment Sediment Sediment, Soil formation Sediment Sediment Sediment Sediment, Soil formation Sediment Sediment, Soil formation Sediment Lab Lab ID profile Depth (cm) Posznan Posznan Posznan Posznan Posznan Posznan Posznan Posznan 113547 113546 114060 114059 114058 114064 114062 114061 4 4 5 5 5 6 6 6 80-90 40-60 70-80 50-60 30-40 100-110 70-80 50-60 radiocarbon age BP cal 2 Sigma BCE*/CE** remarks 3345±35 2475±30 5460±30 3595±35 2480±30 5290±40 3650±30 2980±30 3687-3665(5.2%), 3645-3480(90.2%) 2721-2427(93.6%), 2413-2406(0.6%), 2395-2381(1.3%) 6306-6208(95.4%) 4063-4051(1.0%), 3986-3829(93.9%), 3787-3781(0.5%) 2724-2432(94.6%), 2391-2384(0.5%) 6188-5986(89.4%), 5973-5944(6.0%) 4085-3889(95.4%) 3321-3309(1.1%), 3247-3062*(94.3%) Sediment Sediment Relict Bw-horizon Sediment Sediment Relict BW-horizon Sediment Sediment Table 1 b OSL data Lab Szeged Szeged Szeged Szeged Lab Gdynia Gdynia Lab ID profile Depth (cm) 1 1 1 1 465 380 295 210 1504 1505 1506 1507 Lab.-Nr. profile GdTL-1892 GdTL-1893 2 2 OSL Water grain content size (%) (µm) 20±5 11-20 19±5 11-20 17±5 11-20 19±5 11-20 Depth (cm) 180 100 U (ppm) Th (ppm) K (%) D* (Gy/ka) De (Gy) OSL age (ka) 2.77±0.02 10.03±0.15 1.63±0.04 2.61±0.06 69.47±0.81 26.5±0.7 2.98±0.03 9.23±0.16 1.89±0.06 2.86±0.07 34.68±0.67 12.0±0.4 2.98±0.03 9.76±0.16 1.87±0.05 2.94±0.07 25.30±0.41 8.5±0.3 2.95±0.03 10.09±0.15 1.68±0.04 2.76±0.06 10.84±0.09 3.93±0.1 238 U (ppm) 29.14±0.74 24.30±0.70 232 Th (ppm) 42.5±0.12 40.5±0.12 40 K (ppm) 498±33 576±40 D* (Gy/ka) 2.54±0.11 2.53±0.12 De (Gy), aliquots 69.2±1.4 28.0±2.0 OSL age (ka) 27.2±1.3 10.99±0.97 Period Numerical age (BCE*/CE**//BP) Archaeological sites in the micro-region bold = 5 km radius, black =20 km radius, grey = 20 km, no precise dating available “material-culture” Mesolithic Lower …//until 150,000 Middle …//until 35,000 Upper …//until 9,950 8,000– 6,000*// 9,950- 7,950 Gordashovka, Lashova Dobryanka 1 Kukrek Neolithic 6,000– 4,800*// 7,950- 6,750 Dobryanka 3 Buh-Dniester culture Chalcolithic Early (Tripolye A) 4,600 –4,500*// 6,550- 6,450 Middle (Tripolye B) 4,500-3,900*// 6,450- 5,850 Grebenukiv Yar, Romanovka Tripolye Onoprievka, Vesely Kut, Gordashovka 1, Hlybochok, Rozsohovatka, Kolodyste 1, Krivi kolina, Pischana, Sverdlikove, Nebelivka Maidanetske, Kobrinovo, Romanovka, Moshurov 1, Moshurov 2, Moshurov 3, Gordashovka 2, Talne 1, 2 and 3, Rohy, Talianki, Kamyaneche, Kolodyste Kurgans close to Maidanetske, Legedzyne, Dobrovody, settlement Maidanetske (Shirokiy bereg), Belashki “Oksanichev yar”, Vishnopil, Talne (3), Rohy, Moshurov - Tripolye Palaeolithic Late (Tripolye C) 3,900-3,400*// 5,850- 5,350 Bronze Age Early Iron Age Early Bronze Age 3,000– 2,500*// 4,950- 4,450 Middle Bronze Age 2,600– 2,200*// 4,550- 4,150 Transitional period 2,200– 1,700*// 4,150- 3,650 Late Bronze Age 1,700– 1,300*// 3,650- 3,250 Final Bronze Age 1,300– 900*// 3,250- 2,850 Pre Scythian time 9th– mid 7th c.*// 2,750- 2,600 Reference Shidlovsky et al., 2004: 364 Neradenko, 2011 Zalizniak et al., 2005 Zalizniak et al., 2005 Tripolye Yamnaya culture, kurgans Отчеты, Иванова, 2016: 273-290; Kruts et al., 1981: 4 Maidanetske (?) Legedzyne 2 No settlements Magomedov and Didenko, 2009: 56; Куштан, 2013: 84 Terenozkin, 1961 Scythian time mid 7th– 3rd c.*// 2,600- 2,350 Late Roman time Middle Ages Sarmat time 3rd- 2ndc.*– 4thc.**//2,350- 1,550 mid 3rd- first half 5th c.**// 1,700- 1,500 Early middle Age 5th-10th с.**// 1,450- 950 Kurgans close to Legedzyne, Kolodiste Belashki, Moshurov (settlements)- „early iron age“ Kurgan in Kolodiste Scythian, kurgans Kruts et al., 1981: 4. Maidanetske, Legedgzyne 1 and 2, Legedzyne graveyard, Sverdlikove (burials), Kobrinovo, Belashki (4), Glibochok 1 and 2, Vesely Kut, Potash, Papuzentci, Pavlivka 1, Zelenkiv, Gordashivka 1, 2 and 3, Vishnopil (2), Talne, Rohy, Oksanine 1 and 2, Kolodiste Moshurov, Pishana (Penkovska culture) Chernyakhov culture Magomedov and Didenko, 2009: 56; Kruts et al., 1981: 4 High Middle Ages 10thc.-1250**// 950- 750 - Late middle age 1250- 1500**// 750- 450 1/1 villages ІУМІС, 1972 Early modern period 1500- 1750**// 450- 200 1/33 villages ІУМІС, 1972 Late modern period since 1750**// since 200 1/41 villages At the end of the 19th c. a sugar factory was built in Maidanetske, in action until the end of the 20th century. Construction of cascade ponds. ІУМІС, 1972