Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

The unexpectedly short Holocene Humid Period in Northern Arabia


The early to middle Holocene Humid Period led to a greening of today’s arid Saharo-Arabian desert belt. While this phase is well defined in North Africa and the Southern Arabian Peninsula, robust evidence from Northern Arabia is lacking. Here we fill this gap with unprecedented annually to sub-decadally resolved proxy data from Tayma, the only known varved lake sediments in Northern Arabia. Based on stable isotopes, micro-facies analyses and varve and radiocarbon dating, we distinguish five phases of lake development and show that the wet phase in Northern Arabia from 8800–7900 years BP is considerably shorter than the commonly defined Holocene Humid Period (~11,000–5500 years BP). Moreover, we find a two century-long peak humidity at times when a centennial-scale dry anomaly around 8200 years BP interrupted the Holocene Humid Period in adjacent regions. The short humid phase possibly favoured Neolithic migrations into Northern Arabia representing a strong human response to environmental changes.


Past millennial-scale pluvial periods driven by precession-forced intensification of summer monsoons and northward migration of associated rainfall are thought to have facilitated human dispersal out of Africa1,2,3 by providing ‘green corridors’ through today’s arid Saharo-Arabian desert belt4,5,6. Research on human–climate interaction on the Arabian Peninsula has intensified only recently, even though the region demonstrates high ecological sensitivity to climatic changes and represents the geographic nexus between Africa and Asia1,2,3,7,8. This recent wave of research in Arabia has already fundamentally transformed our perception of Arabian prehistory, including discoveries of Middle Palaeolithic (MIS 5 or even older) sites in Central Arabia9 or traces of Homo sapiens in the Nefud desert at approximately 87 ka8, i.e. phases associated with conditions more humid than today2,6. One of the emerging topics is the role of the early to middle Holocene Humid Period (HHP) in Neolithic migrations and cultural progress10,11.

Climate models suggest that the African Summer Monsoon (ASM) was the dominant moisture source on the Arabian Peninsula during pluvials1,12. Yet, this remains a matter of debate for the Northern Arabian desert13,14,15, as stronger insolation intensified and extended both ASM1,2,12,16,17,18 and, possibly, Mediterranean winter rains3,19. The latter system is the main source of moisture in this region today. In addition, tropical plumes (TPs), i.e. synoptic disturbances conveying water vapour as continuous mid-upper tropospheric cloud bands from the Intertropical Convergence Zone (ITCZ) to >15°N, are known to affect Northern Arabia during winter and spring14,20,21. Higher frequency of such patterns during past pluvials was suggested to have contributed increased rainfall to the Saharo-Arabian desert14,22,23, even though their past role as a moisture source remains poorly understood.

The rich archaeological heritage of Arabia is currently being investigated by major research initiatives7,11,24. “Potentially thousands of water bodies” have been reconstructed for past pluvials25, but it is still unknown how these water bodies and human habitats exactly looked like and for how long they existed14,15. Only a few climate records are available from speleothems in the wider region, i.e. the Levant26,27,28 and Southern Arabia29,30. The entire lack of high-resolution palaeoclimate data from Northern Arabia leads to an inconsistent picture about the timing and magnitude of the HHP for this culturally important corridor to the Middle East, where some lower-resolution lacustrine records have pointed to more humid conditions during the last interglacial (MIS 5) and the early to mid-Holocene10,17,31,32,33.

The Tayma palaeolake record4,34 is the only known high-resolution archive of the HHP in Northern Arabia providing insights into the early to mid-Holocene hydroclimate variability in unprecedented detail. The 20 km2-sized inland sabkha of Tayma has a 660 km2 hydrological catchment (Fig. 1; Supplementary Figs. 1 and 2) and is located at 27°40′N within the arid desert’s interior. It receives only scarce rains (on average 45 mm a−1) from Mediterranean winter storms, occasional cross-Saharan TPs or the Red Sea cyclones between autumn and spring14 (Fig. 1). Previous investigations of shoreline deposits (Supplementary Figs. 35) and sediment cores from the sabkha basin indicate the existence of a >17 m deep, perennial groundwater-supported lake15 and the spread of grassland4 during the early Holocene. The catchment-lake ratio (Fig. 1b; Supplementary Fig. 1b) and the short duration of this peak lake phase4,15,34 exclude the influence of tectonics on lake-level changes, emphasising the significance of the lake as a palaeoclimate archive that is mainly controlled by rainfall, groundwater inflow and evaporation (see also Supplementary Note 1 for details on controlling processes of the sedimentary archive). Yet, a precise determination of the timing of the lake phases was still missing, preventing a detailed view of the evolution of the palaeolake and the palaeoclimatological implications.

Fig. 1: Regional setting of Tayma.
figure 1

a Overview of the Arabian Peninsula and adjacent areas with the key palaeoclimate sites of Jeita Cave28, Soreq Cave26,27, GeoB 5844-2 in the Red Sea13 and Qunf Cave29, mean annual rainfall 1970–2000 (WorldClim 2 dataset77), average positions of the Intertropical Convergence Zone (ITCZ) in summer and the Subtropical Jet (STJ) in winter78, and atmospheric sources of regional precipitation (MWR Mediterranean winter rains, Winter TP winter tropical plumes, ASM African Summer Monsoon, ISM (SJ) Indian Summer Monsoon (Somali Jet), ARST Active Red Sea Trough)14. Isohyetes of 200 mm and 500 mm are adapted from the wettest simulation of the COSMOS climate model at 6000 BCE in ref. 12. b Reconstructed extent of the Tayma palaeolake during its peak phase, today’s surface catchment and groundwater catchment79. The topography is based on GTOPO30 data80.

In this paper, we decipher five phases of lake development at Tayma and show that the HHP in Northern Arabia from 8800 to 7900 years BP is shorter than commonly defined (~11,000–5500 years BP). Interestingly, we identify a two century-long peak humidity at Tayma overlapping with a centennial-scale dry anomaly around 8200 years BP that interrupted the HHP in adjacent regions. The results clearly demonstrate that regional patterns in palaeoclimate were more complex and spatially small-scale than previously assumed.


Chronology of the Tayma palaeolake record

The Tayma palaeolake record partly contains annually laminated sediments that were counted under the microscope (see Methods, Fig. 2, Supplementary Fig. 9). The new high-resolution age-depth model integrates AMS radiocarbon ages of pollen concentrates, microscopic varve counting and the independent age of a cryptotephra35 in a Bayesian model (see Methods, Supplementary Data 1, Supplementary Fig. 7). The floating varve chronology comprising 650 ± 40 couplets is anchored to the radiocarbon age scale and constrains the varved lake phase at Tayma to 8550–7900 ± 40 cal varve yr BP (±90 cal yr BP including the 14C calibration error). A robust time marker is provided by the identification of the central Anatolian ‘S1’ tephra in the lower part of the record, dated to 8983 ± 83 cal yr BP in the Dead Sea record34. The lacustrine and wetland sediments in the Tayma basin were deposited from ca. 9250 to ca. 4200 cal yr BP (Supplementary Fig. 7).

Fig. 2: Palaeolake evolution at Tayma between 9250 and 7500 cal yr BP.
figure 2

a Radiocarbon ages (triangles) and fix points (stars; see Methods); b δ18Ocarb/arag and δ13Ccarb/arag measured on bulk carbonates (solid lines) and on single aragonite layers (blue and green diamonds); c δDwax of n-alkanes nC31 and nC29 (δDnC29, nC31); d Tayma lake phases I–V and wetland phase VI; e varve sublayers expressed as % of varve thickness for the varve chronology 8550–7900 ± 40 cal varve yr BP, and microscope photographs of thin sections highlighting different micro-facies of lake phases II–V.

Changes in precipitation and evaporation signals in the Tayma record

Compound-specific hydrogen isotope compositions of plant-wax n-alkanes (δDnC29, nC31), as well as porewater, rainwater and groundwater isotopes (δ18Owater and δDwater) trace variations in moisture supply and rainfall amount (Fig. 2, Methods and Supplementary Fig. 8). Stable oxygen and carbon isotope compositions of single primary aragonite laminae from the varved core section (δ18Oarag and δ13Carag) and of bulk carbonates from the entire core (δ18Ocarb and δ13Ccarb) indicate changing groundwater and surface-water inflow, lake-water evaporation and lake-internal productivity (Fig. 2). The most striking finding was terrestrial plant wax δDnC29, nC31 values of about −100‰ during the shallow lake or wetland phase and much lighter values down to −150‰ during the palaeolake phase. These data reflect higher rainfall between 8800 and 7950 cal yr BP due to increased precipitation and a probable amount effect (Methods and Supplementary Fig. 8).

Evolution of the Tayma palaeolake

The evolution of the lake can be separated into phases I–V, followed by phases VI (wetland) and VII (sabkha). A basal zone of brownish-grey carbonate mud with irregular, coarse carbonate lamination from 9250 to 8800 cal yr BP (lake phase I) (Supplementary Fig. 6) represents a shallow lake initiated by increasing rainfall and recharge of the local Saq aquifer when clastic sediments were deposited in a deflated endorheic basin from a prevailing desert environment4. Carbonates precipitated with very high δ18Ocarb values of around +11‰ and low δ13Ccarb values of around -8‰.

At ca. 8800 cal yr BP (lake phase II), a sharp decrease of δ18Ocarb to +8‰, increasing δ13Ccarb (Fig. 2) and the in situ deposition of the brackish-water ostracod Cyprideis torosa (Fig. 2e) indicate reduced lake-water evaporation and the initial establishment of a shallow, but perennial and increasingly productive water body36 as a response to wetter conditions. The δDnC29, C31 values show a strong variability between −140‰ and −90‰.

At ca. 8550 cal yr BP the formation of varves started (Fig. 2; Supplementary Figs. 6 and 9). Fine lamination over a period of 650 ± 40 varve years (lake phases III and IV) reflects the onset of a deep (>17 m15, Supplementary Figs. 35) and stratified permanent lake, similar and coeval to, e.g. the Awafi palaeolake at Ras Al-Khaimah, United Arab Emirates37. From ca. 8550 to 8250 cal yr BP (lake phase III), variable but continuously decreasing plant wax δDnC29, nC31 values between −150 and −100‰ indicate a humid period with enhanced seasonality38,39. The alternating deposition of dark clastic, clay- and organic-rich laminae and white, primary aragonite laminae reflects pronounced wet and dry seasons. Suspension grading observed in some of the clastic varves indicates wadi activation and fluvial sediment input into a standing water body during the wet season. The aragonite laminae represent the arid season of enhanced evaporation, reduced or absent surface-water input, lake contraction and lake-level fall, as well as increased concentration of soluble matter40. The highest concentrations of brine occur near the water surface, where aragonite crystals form and settle through the water column as pelagic rain41,42. The δ18Ocarb values generally decrease from +8‰ to +6‰ simultaneously with progressively increasing δ13Ccarb values from about −6 up to +2‰ toward enhanced lake productivity43,44. The positive excursion of δ18Ocarb to >+10‰ centred at ca. 8400 cal yr BP reflects a decadal- to centennial-scale drawback to even stronger dry-season evaporation. This intensified dry-season evaporation was compensated by enhanced humidity during the rainy season and groundwater inflow, indicated by the decreasing trend of δDnC29, nC31 at that time, data which mostly reflects the wet season of leaf growth39. This rainy-season moisture surplus was sufficient to sustain a high lake level and varve formation.

The phase between ca. 8250 and 7950 cal yr BP (lake phase IV) shows the highest production rate of organic matter in the lake and annual blooms of planktonic diatoms (mainly Cyclotella cf. choctawhatcheeana) (Supplementary Figs. 6 and 9). In combination with the greatest abundances of foraminifera, the lowest δDnC29, nC31 values down to −155‰ and weakest dry-season evaporation with lowest δ18Ocarb values of +4‰, they indicate the highest lake stand and most humid period at Tayma during the Holocene. This is supported by a distinct change in varve composition from evaporation-driven aragonite varves to pronounced productivity-driven diatom-aragonite varves and total organic carbon (TOC) contents of up to 5%. The highest ratio between δ13Ccarb and δ18Ocarb in this part of the core is related to phytoplankton bloom controlled by 12C depletion due to photosynthetic uptake of CO2, and higher precipitation–evaporation balance43. From about 8200 cal yr BP the δDnC29, nC31 values again start to vary between −140‰ and −100‰, and the δ18Ocarb values increase from +4 to +8‰ (Fig. 2; Supplementary Fig. 6).

At ca. 7950 cal yr BP, ceasing diatom and aragonite laminae and more abundant clastic quartz grains of aeolian origin, as well as the first appearance of gypsum (Supplementary Figs. 6 and 9), show a rapidly declining lake level (lake phase V). The increased mobility of sand grains indicates a regional decline of vegetation cover, the retreat of grasslands and the establishment of drought-resistant steppe vegetation4. The occurrence of gypsum is related to the enrichment of Ca2+ ions in the contracting water body and sulphate dissolved in the reduced groundwater and surface water inflow45. This led to the disappearance of varves within a few decades, accompanied by a sharp reduction in TOC content and a decline of δ13Ccarb back to a level comparable to the early shallow-lake phase II, prior to 8550 cal yr BP. Progressively enriched δDnC29, nC31 values of up to −60‰ and δ18Ocarb values toward +12‰ reflect the decrease in surface-water and groundwater inflow and a strongly increasing evaporation, indicating a gradual end of the humid phase over 100–150 years until ca. 7800 cal yr BP.

Further increasing gypsum and Mg–carbonate precipitation and δ18Ocarb rising to +12‰ (Fig. 2; Supplementary Fig. 6) point to a shrinking lake and concentration of brine under an increasingly arid climate between 7800 and ca. 6800 cal yr BP. At that time, wetland conditions set in with TOC levels close to 0 and further increasing aeolian influx (phase VI). Around ca. 4200 cal yr BP, greyish mud is replaced by reddish-brown, oxidised, mainly aeolian clastics mixed with gypsum (phase VII) (Supplementary Figs. 6 and 7). This pattern reflects temporary desiccation and a further aridification pulse correlating with a dry event recorded at several sites in the Eastern Mediterranean/Middle East, e.g. the Northern Red Sea46.


Our data support existing low-resolution Northern Arabian palaeoenvironmental records10,11,31, but they also show that the HHP in Northern Arabia was remarkably short, its peak lasting only ca. 650 years from 8550 to 7900 cal yr BP. A temporary intensification and northeastward shift of the ASM to provide increased rainfall in northwestern Arabia around Tayma during the short HHP is supported by the δ18Owater and δDwater data. The evaporation line calculated from modern water samples around Tayma and the palaeolake data best fits the isotope data from δD-depleted ASM precipitation at Khartoum, Sudan47 (Supplementary Fig. 8). The ASM fuelling the HHP in Northern Arabia has also been suggested by several proxy-based17,32 and climate modelling1,12,16 studies. We propose a moderate influence of the ASM during the HHP, as Tayma is located at the fringes of the ASM influence at that time12.

In addition, we observe an intriguing regional hydroclimatic diversity, since the aforementioned peak humidity in Northern Arabia from 8550 to 7900 cal yr BP coincides with a widespread, centennial-scale dry and cool anomaly centred around the 8.2 ka cold event at other low-latitude sites in the Northern Hemisphere such as the Eastern Mediterranean or Southern Arabia48 (Fig. 3). A low-latitude dry period between ca. 8500 and 7800 cal yr BP was the most pronounced hydroclimatic drawback of the HHP (Fig. 3a). It is evidenced, e.g. in the desiccation or temporary lowstands of North African lakes49 (Fig. 3f), slightly heavier δ18O values in Northern Red Sea planktonic foraminifera13 (Fig. 3e) and diminished runoff of the Nile River50, leading to re-oxygenation of the Eastern Mediterranean Sea and interruption of sapropel S151,52 (Fig. 3b). Drier conditions mostly resulted from reduced summer–monsoon rainfall29,30. However, speleothem records from the Levantine region26 (Jeita and Soreq caves; Fig. 3b, c) and marine records from the Eastern Mediterranean suggest that Mediterranean winter rains might have been reduced as well, as a result of temporary, meltwater-related deceleration of the North Atlantic thermohaline circulation48,53 (Fig. 3). The inference of moderately higher than average rainfall during the HHP from the Soreq speleothem record, mostly during the phase of excess winter rains between 8000 and 7000 cal yr BP, is based on the recalculation of the δ18O data in ref. 51 eliminating the source effect, i.e. the bias through changes of isotopic composition in the Eastern Mediterranean surface waters28.

Fig. 3: Oxygen isotopes recording humidity changes during the early to middle Holocene Humid Period (HHP) across the Eastern Mediterranean to Southern Arabia regions.
figure 3

a Summer insolation at 30°N81, duration of the HHP sensu ref. 18 (grey), the low-latitude dry anomaly48 (yellow), and the 8.2 ka cold event in Greenland ice cores82 (white bar); b speleothem δ18O from Jeita cave (Lebanon)28, and c Soreq cave (Israel)26,27, with the timing of sapropel formation in the Eastern Mediterranean Sea and excess winter rain in the Southern Levant51; d δ18Ocarb from the Tayma palaeolake (this study); e δ18OG. ruber reflecting temperature, and calculated salinity changes from the Northern Red Sea (GeoB 5844-2)13; f frequency histograms of lake records reflecting wet or moderately wet conditions in the East African Summer Monsoon domain >30°E18; g speleothem δ18O from Qunf cave (Oman)29. All δ18O scales are reversed to reflect higher humidity upwards. Palaeoclimate sites are shown in Fig. 1a.

These smaller-scale, regional discrepancies may be explained by the additional contribution of synoptic-scale patterns, which are scarce today. They may have played a more dominant role in delivering moisture to Northern Arabia between 8250 and 8000 cal yr BP resulting in a humidity peak at Tayma. In particular, more frequent TPs may have led to a regional moisture surplus in Northern Arabia. In contrast to short and localised convective cells of the Active Red Sea Trough pattern triggering flash floods in the Southern Levant, TPs promote long-lasting moderate rains and thus more effective moisture over a slightly larger region20. TP formation was likely favoured by ocean-atmosphere feedbacks during the ‘cool poles—dry tropics’ anomaly around 8.2 ka: lower sea-surface temperatures in the North Atlantic and the Mediterranean Sea promote deeper, southwards penetrating mid-latitude troughs and stronger sub-tropical anticyclones (i.e. drier air masses). This leads to an intensification of tropical moisture advection and the sub-tropical jet stream, inducing jet streaks that reach as far as northern tropical West Africa and convey moist air to Northern Arabia at mid-to-upper tropospheric levels21. The regionally and chronologically confined enhanced contribution of TPs may have compensated the weakened ASM around 8.2 ka and contributed to the strong seasonal pattern reflected by high-level, high-amplitude δ18Ocarb and δDnC29, nC31 data. Mediterranean westerlies as a source of additional winter rainfall, however, seem unlikely, as climate models indicate an even lower intensity during the HHP compared to today12.

There is multiple evidence that the observed moisture surplus of 8550–7900 cal yr BP identified in the Tayma record, in combination with charged aquifers, had distinct short-term impacts on the local environment and probably also on human migration. Vegetation resources4 and the abundance of prey animals12 increased and stimulated Neolithic migrations into Northern Arabia as indicated by abundant Levant-type Pre-Pottery Neolithic A and B assemblages identified in the Northern branch of the Nefud desert10,11.


Tayma sediment cores

Drilling on today’s sabkha of the Tayma palaeolake basin was performed in 2011 and 2013 using an Atlas Copco vibracoring device (Cobra mk1) fitted with closed steel auger heads and PVC liners with a diameter of 5 cm. Two series of ca. 6 m long sediment cores (Tay 220/221 and Tay 253/254/255/256) capturing the entire Holocene sequence and reaching down to Ordovician sandstone (Qasim Formation) were obtained in close vicinity (‘mastercores’ in Supplementary Fig. 1a). They each consist of two parallel, overlapping core sequences A and B with 1 m-long core sections. The cores were opened and photographically documented at the University of Cologne (Laboratory for Physical Geography) and GFZ Potsdam, Germany. The construction of composite profiles and correlation of the sediment cores is based on 24 macroscopic lithological marker layers (fixed marker horizons, FMH). Archive cores of the master site are stored and accessible in the core storage facility of the Institute of Geography, Heidelberg University, refrigerated at 4 °C.

Tayma sediment cores were analysed for their sedimentology (XRF [X-ray fluorescence] core scanning, quantitative XRF on discrete samples, semi-quantitative X-ray diffraction, micro-facies analyses on thin sections), geochemistry (elemental analyses, stable isotopes, lipid biomarkers), palynology (vegetation reconstruction through pollen analysis) and micropalaeontology (assemblages of foraminifera, ostracods and diatoms). Here, we used stable isotopes of oxygen and carbon measured on aragonite laminae of the annually laminated (varved) core section (δ18Oarag, δ13Carag) as well as bulk carbonates from the entire core (δ18Ocarb, δ13Ccarb) in combination with micro-facies analyses of the varved sediments to trace the evolution of the early to mid-Holocene palaeolake at Tayma. Further proxy data have partially been published4,36,54 or will be presented in forthcoming publications.

In Supplementary Fig. 6, the lithological profile of the ca. 6.5 m-long composite core, TOC content54, δ18Ocarb and δ13Ccarb (see Methodological details further down), and statistical clustering results of the XRF core-scanning record are shown. The elemental composition of the sediment core was determined by non-destructive XRF core scanning on the split-core sediment surface using an ITRAX elemental scanner at GFZ Potsdam. Measurements were obtained every 0.2 mm using a Cr X-ray source, operated at 30 kV, 30 mA and 10 s, to capture intensities of the elements Si, S, Cl, K, Ca, Ti, Fe, Sr and Zr. A centred log-ratio (clr = ln [element intensity/geometric mean of all nine elements]) transform was performed for all elements of each measurement to eliminate the influences of physical properties, sample geometry and matrix effects55,56 and to enable robust statistical analyses57. Statistical clustering (Ward’s method) of XRF core-scanning results indicates four main sediment groups (Supplementary Fig. 6): Cluster 1 (light grey) is dominated by Si, Ti and Fe and describes the siliciclastic sediments and occurs predominantly in the upper part of the Tayma profile (VII—sabkha phase). Cluster 2 (green) does not show a clear preference, but is rather a mixture of all considered elements, describing clastic, carbonate and evaporitic ‘background’ sediments. Cluster 3 (blue) is dominated by Sr and Ca and describes aragonite, which occurs exclusively in the varved sediments of the Tayma core representing the deep-lake phases III and IV (Fig. 2e). Cluster 4 (orange) is dominated by the elements S and Ca and mainly describes gypsum that was deposited during the terminal lake phase (V) and thereafter, when wetlands occupied the Tayma basin (phase VI).

TOC indicates the production and preservation of organic matter in the lake and was measured on in situ decalcified samples using an elemental analyser (NC2500 Carlo Erba) at GFZ Potsdam, Germany. Ca. 3 mg of freeze-dried and powdered sample material was weighed into Ag-capsules, treated with 20% HCl, heated and dried for 3 h at 75 °C. The calibration of the data was carried out applying element standards (Acetanilide, Urea). It was verified using a soil reference sample (Boden3, HEKATECH). The reproducibility for replicate samples was ~0.2%54. The sediments deposited in the Tayma basin are mainly composed of clay, silt and sand, evaporites (sulphates), authigenic carbonates and in parts high amounts of diatoms (at the mastercore site in the centre of the basin), gastropod shells, ostracods and foraminifera (along the margins of the basin representing the palaeo-shoreline) (Supplementary Fig. 6). Clay- and silt-sized detritus is dominant in the core and was deposited as dark grey, mm- to cm-thick, occasionally graded layers. Coarser silt- to sand-sized minerals (mainly quartz) are scattered in the sediments or are concentrated in the uppermost part of the Tayma profile. Evaporites were mainly identified in the form of whitish-beige, finer-grained laminae of gypsum and other sulphates. Carbonates are present in the form of white, sub-mm thick primary aragonite laminae, and, in the upper part, Mg–carbonate layers. Biogenic carbonates (ostracods, foraminifera, and barnacle and gastropod shell fragments) are dominant along the shorelines of the palaeolake. Their concentration is much lower in the centre of the basin where the mastercore was taken.

Varve micro-facies analysis

We used changes in varve micro-facies, i.e. the composition of seasonal sublayers of the annual laminations, to infer changing seasonality and the interannual variability of lake-internal evaporation and productivity. The thickness and composition of varve sublayers were analysed under the microscope along with varve counting on petrographic thin sections. A total of eleven different sublayer types were grouped into five main sediment components (carbonate, organic, clastic, diatoms and gypsum). Data are given as relative contribution (in %) to the varve thickness (Fig. 2e). Raw data of micro-facies sub-layer thicknesses are presented in Supplementary Fig. 9.

Age model construction

Due to the absence of datable terrestrial macroscopic plant remains in Tayma cores and reported hard-water effects altering radiocarbon ages from gastropods, ostracods and Ruppia seeds for up to 1500 years4,15, preliminary age models4,54 were based on AMS radiocarbon dating of pollen grains, as these are unsusceptible to incorporating old carbon58,59. Pollen extraction from a total of 33 samples of 1–13 cm long sediment sections followed a combination of physical and chemical separation protocols58,59,60,61. Sample preparation included sieving (at 6, 20, 40 and 70 µm), treatment with heated HCl, KOH and H2SO4, and heavy-liquid density separation using CsCl and sodium polytungstate.

Varve counting was performed on 14 large-scale (10 cm × 1.5 cm) petrographic thin sections using a Leica DMLP petrographic microscope under semi-/fully polarised light and with 50× magnification. Thin sections were prepared following standard procedures for soft sediments62 including freeze-drying and impregnation with epoxy resin (Araldite 2020). Sawing and polishing were performed manually under dry conditions to avoid salt crystallisation. Multiple counting and the definition of correlation marker layers ensured a negligible subjective counting error. Counting uncertainty due to poor sublayer quality is ±40 varves (6.2%).

The age-depth model was constructed with Bacon v2.2 using flexible Bayesian modelling63 including implemented outlier analysis and the IntCal13 atmospheric calibration curve64. All 38 radiocarbon ages of pollen concentrate, other plant remains (Ruppia seeds, non-pollen palynomorphs, charred plant particles), two mollusc samples, as well as a tephrochronological anchor identified close to the base of the Tayma sediment record (the central Anatolian ‘S1’ tephra dated at 8983 ± 83 cal yr BP in the Dead Sea)35, were considered for age modelling (Supplementary Data 1). The floating varve chronology of 650 ± 40 varve years served to refine the Bayesian model within the varved section. The start of varve formation is defined by 14C dating to 8549 cal yr BP (8470–8605 cal yr BP for the 95.4% probability range). Based on this fixed point (F1), the varve age of a turbidite layer at 8138 ± 40 varve years BP and the end of varve formation at 7900 ± 40 varve years BP were used as further fix points (F2 and F3) in the adjusted Bayesian model (Supplementary Fig. 7).

Outlier analysis reliably discarded samples (n = 6) containing ≤50% pollen or hard-water-affected material (gastropod shells, Ruppia seeds), and 13 samples with ≥50% pollen unsuitable for the Bayesian age-depth model. All remaining 18 14C ages of pollen concentrate included in the final model contained high pollen concentrations of at least 50% (Supplementary Data 1, Supplementary Fig. 7).

Reconstruction of hydroclimatic conditions

The stable isotope composition of lake water (δ18Owater and δDwater) in closed lakes is mainly controlled by precipitation and evaporation and reflects hydrological changes and moisture sources43. The δ18Ocarb13Ccarb) values of bulk carbonates and δDnC29, nC31 from fossil leaf waxes in lake sediments are proxies for hydroclimatic conditions and were used to reconstruct the precipitation, lake-water evaporation and temperature during the early to mid-Holocene. To assess the hydrological balance of the Tayma palaeolake (8800–7950 cal yr BP), the wetland (7800–6800 cal yr BP), and the potential moisture sources during the HHP, we compared calculated δDp (precipitation) and δ18Owater (lake water) values with the isotopic characterisation of the main regional atmospheric systems, recent precipitation, as well as surface and groundwater isotope compositions (Supplementary Fig. 8).

Stable oxygen and carbon isotopes

Stable oxygen and carbon isotope measurements (δ13Ccarb and δ18Ocarb) were performed on the carbonate fraction of a total of 262 freeze-dried and ground samples taken in cm slices from core Tay 220. Bulk samples of ~0.4 mg were loaded into 10 ml Labco Exetainer vials automatically flushed with He and reacted in phosphoric acid (100%) at 75 °C for 60 min65. The stable isotope compositions were determined at GFZ Potsdam using a Finnigan GasBenchII with carbonate option coupled to a DELTAplusXL IRMS (isotope ratio mass spectrometer) (Thermo Fisher Scientific). For finer resolution and to exclude biogenic carbonates, we sampled pure primary aragonite of laminae from dried and impregnated sediment blocks by drilling. The aragonite samples of about 0.06 mg were measured for δ18Oarag and δ13Carag at GFZ Potsdam with an automated carbonate device (KIEL IV) coupled to a Finnigan MAT253 IRMS (Thermo Fisher Scientific) on cryogenically purified CO2 released by dissolution with 103% H3PO4 at 72 °C for 10 min. Oxygen and carbon isotope compositions are given relative to the VPDB (Vienna Peedee Belemnite) standard in conventional delta notation δ (‰). Calibration was performed using international reference standards (NBS18 and NBS19). For both methods, standard deviations (1σ) for reference and replicate analyses are better than 0.08‰ for both δ18O and δ13C.

In closed lakes, δ18Ocarb values mainly reflect hydrological changes and are used as a proxy for precipitation, groundwater influx and lake evaporation because: (i) seasonality and temperature have little effect on oxygen isotope fractionation of precipitation in low-latitude regions43,66; (ii) the lake-water oxygen isotopic composition in an endorheic basin is governed by evaporation under arid climate conditions resulting in increased δ18Owater; (iii) equilibrium oxygen isotope fractionation is assumed for inorganic carbonates; and (iv) primary inorganic carbonates precipitate during the spring–summer season induced by evaporation and/or phytoplankton bloom in the epilimnion. The latter is consistent with increasing δ13Ccarb values, indicating 12C depletion of the total dissolved inorganic carbon due to atmospheric release and/or preferential use of aquatic plants.

The calculation of δ18OVSMOW palaeolake water from δ18OVPDB values using the re-expressed relationship of ref. 67 in the simplified Eq. (1) according to ref. 43

$$T=13.8-4.58(\delta c-\delta w)$$

under equilibrium water (w)-calcite (c) precipitation at a temperature (T in °C) of 21 °C (as the average temperature in spring) and an offset of +0.6‰ for aragonite and magnesium bearing calcite reveals comparable δ18O values between precipitated carbonates and hot water. The modelled mean δ18Owater for the palaeolake water is high with +8.4‰ and, thus, much lighter due to freshwater inflow of surface and groundwater than for the wetland with a calculated mean δ18Owater of +13.1‰ due to lower precipitation and high evaporation.

Stable hydrogen isotopes of leaf-wax n-alkanes (δDwax)

Stable hydrogen isotopes of leaf-wax n-alkanes (δDwax) were measured on 64 samples from core Tay 255. Samples were taken in 1 cm-slices, freeze-dried and ground for lipid biomarkers extraction at the Max Planck Institute for Biogeochemistry in Jena. 5–15 g of the sample was extracted using a 40 ml dichloromethane:methanol (9:1) mixture at 100 °C and 120 bar for 15 min in two consecutive cycles using a BÜCHI SpeedExtractor. The total lipid extract was separated into aliphatic, aromatic and alcohol/fatty acid fractions using solid-phase extraction on silica gel according to the method presented in ref. 68. The aliphatic hydrocarbon fraction was desulfurized using HCl-activated copper (15% HCl). Identification and quantification of n-alkanes were accomplished using a GC-MS (Agilent Technologies, 7890A GC-System; 220 Ion trap MS) by comparing peak areas and retention times with an external n-alkane standard mixture (nC16 to nC36). Compound-specific hydrogen isotope ratios (expressed as δD) of the n-alkanes were measured on a DELTA Vplus IRMS (Thermo Fisher Scientific) coupled to an Agilent 7890 GC (Agilent Technologies) at GFZ Potsdam. Every sample was measured in triplicates. The mean standard deviation of all measured samples was 3‰. The δD values were normalised to the Vienna Standard Mean Ocean Water (VSMOW).

The changes in δDwax of the lake records are interpreted as indicators for the variability in precipitation, humidity and vegetation type66,68,69. Hydrogen isotopes δDnC29, nC31 of terrestrial leaf wax nC29 and nC31 n-alkanes were used to calculate δDp between precipitation (p) according to refs. 38,39,69. The negative isotopic fractionation from δDp to δDwax due to the incorporation of hydrogen in leaf waxes has been calculated using Eq. (2)

$$\delta {{{{{\rm{D}}}}}}_{{{{{{\rm{p}}}}}}}=[(\delta {{{{{\rm{D}}}}}}_{{{{{{\rm{wax}}}}}}}+1000)/((\varepsilon /1000)+1)]-1000$$

with ε = −130 for nC29 and nC31 n-alkanes representing the mixture of C3/C4 plant waxes of grasses, shrubs and trees that can be expected for the region4 (Supplementary Fig. 8). Following ref. 5, we inferred the precipitation rate from δDp values. The relationship of precipitation and rainfall amount for the Sahara region described a non-linear dependence with a steep slope in δDp values below 100 mm yr−1 and a strong influence of the amount effect.

δ18Owater and δDwater isotopes of water samples

Filtered water samples of groundwater from the historical Bir Haddaj well of the Tayma oasis, from the Tay 255 borehole in the palaeolake, and evaporated rainwaters from small water pools south of the palaeolake taken shortly after a rain event in December 2015, were triple-measured at the GFZ for δ18Owater and δDwater relative to VSMOW using Cavity Ring-Down spectrometers (PICARRO L2120-i and L2130-i). Analytical precision of VSMOW and SLAP calibrated analyses was <1‰ for both δ18Owater and δDwater.

The isotopic fractionation of lake-water evaporation was calculated for the remaining lake water sampled in December 2015 (δ18Orw) using initial groundwater-supported lake water with δ18Oiw of −3‰ with simple Rayleigh distillation after Eq. (3)

$${\delta }^{18}{{{{{{\rm{O}}}}}}}_{{{{{{\rm{rw}}}}}}}={\delta }^{18}{{{{{{\rm{O}}}}}}}_{{{{{{\rm{iw}}}}}}}-1000({f}^{(a-1)}-1)$$

where α = fractionation factor between water and vapour at 21 °C70 and ƒ = fraction of remaining lake water.

Reconstruction of palaeo-moisture source and lake-water evaporation

The few meteoric water samples from Tabuk (IAEA) plot with δ18Op ~−1‰ closely to the global meteoric water line (GMWL), except for one lighter sample tending more to the Eastern Mediterranean meteoric water line (EMMWL). Recent (12/2015) evaporated rainwater samples collected in water pools in a wadi SW of the Tayma palaeolake show slightly enriched δ18Owater and δDwater values of around −0.5‰. Using δDwax to estimate past precipitation rates as explained above reveals the lowest values of about −2‰ δDp for the wetland phase and values as low as −28‰ for the palaeolake (Supplementary Fig. 8), indicating higher rainfall amounts between 8800 and 7950 cal yr BP due to increased precipitation and a probable amount effect.

The stable isotopes of the historical Bir Haddaj well in Tayma, one of the largest and most prominent historical wells on the Arabian Peninsula71,72 that taps the uppermost level of the Saq aquifer, reflect subsurface groundwater with −3.5‰ δ18Owater and −24.6‰ δDwater, similar to the middle of the Saq aquifer further south73,74. The water from the palaeolake sampled in 1.5 m depth of the well Tay 255 in 2015 with +7.4‰ δ18Owater and +16.2‰ δDwater probably reflects pore-water isotope composition. The portion of surface-water evaporation calculated using Rayleigh fractionation between −3‰ δ18Owater for groundwater and +7.4‰ δ18Owater for the Tay 255 well reaches about 70% for the deep lake phase and >80% for the wetland phase. The isotopic difference between the deep palaeolake and the wetland water is mainly influenced by decreases in precipitation and increasing evaporation.

Although the three main atmospheric systems affecting the northwestern Arabian Peninsula (Indian Monsoon, Mediterranean Westerlies and African Monsoon) show isotopic fingerprints with more or less variation and deuterium excess, it is unreasonable to decipher the moisture sources during the time of palaeolake formation due to the determination of precipitation δDp using δDwax and δ18Ocarb being indirect, as well as associated fractionation effects. In general, the moisture source-related isotope fingerprints over the Arabian Peninsula are masked by strong evaporation, continental and altitude effects, sub-cloud evaporation, moisture recycling and the amount effect75.

Data availability

Data reported here are stored at GFZ Data Services ( and freely accessible under a Creative Commons Attribution 4.0 International License (CC BY 4.0)76. Further supplementary data and information are available as Supplementary Material and Supplementary Data in the online version of the paper. Correspondence and requests for materials should be addressed to M.E.

Code availability

This study does not use custom code or mathematical algorithm that is deemed central to the conclusions.


  1. Jennings, R. P. et al. The greening of Arabia: multiple opportunities for human occupation of the Arabian Peninsula during the Late Pleistocene inferred from an ensemble of climate model simulations. Quat. Int. 382, 181–199 (2015).

    Google Scholar 

  2. Parton, A. et al. Alluvial fan records from southeast Arabia reveal multiple windows for human dispersal. Geology 43, 295–298 (2015).

    Google Scholar 

  3. Breeze, P. S. et al. Palaeohydrological corridors for hominin dispersals in the Middle East 250–70,000 years ago. Quat. Sci. Rev. 144, 155–185 (2016).

    Google Scholar 

  4. Dinies, M., Plessen, B., Neef, R. & Kürschner, H. When the dessert was green: grassland expansion during the early Holocene in northwestern Arabia. Quat. Int. 382, 293–302 (2015).

    Google Scholar 

  5. Tierney, J. E., Pausata, F. S. R. & deMenocal, P. B. Rainfall regimes of the Green Sahara. Sci. Adv. 3, e1601503 (2017).

    Google Scholar 

  6. Nicholson, S. L., Hosfield, R., Groucutt, H. S., Pike, A. W. & Fleitmann, D. Beyond arrows on a map: the dynamics of Homo sapiens dispersal and occupation of Arabia during Marine Isotope Stage 5. J. Anthropol. Archaeol. 62, 101269 (2021).

    Google Scholar 

  7. Armitage, S. J. et al. The southern route “Out of Africa”: evidence for an early expansion of modern humans into Arabia. Science 331, 453–456 (2011).

    CAS  Google Scholar 

  8. Groucutt, H. S. et al. Homo sapiens in Arabia by 85,000 years ago. Nature Ecol. Evol. 2, 800–809 (2018).

    Google Scholar 

  9. Crassard, R., Hilbert, Y. H., Preusser, F., Wulf, G. & Schiettecatte, J. Middle Palaeolithic occupations in central Saudi Arabia during MIS 5 and MIS 7: new insights on the origins of the peopling of Arabia. Archaeol. Anthropol. Sci. 11, 3101–3120 (2019).

    Google Scholar 

  10. Crassard, R. et al. Beyond the Levant: First evidence of a Pre-Pottery Neolithic incursion into the Nefud Desert, Saudi Arabia. PLoS ONE 8, e68061 (2013).

    CAS  Google Scholar 

  11. Petraglia, M. D., Groucutt, H. S., Guagnin, M., Breeze, P. S. & Boivin, N. Human responses to climate and ecosystem change in ancient Arabia. Proc. Natl Acad. Sci. USA 117, 8263–8270 (2020).

    CAS  Google Scholar 

  12. Guagnin, M. et al. Rock art imagery as a proxy for Holocene environmental change: a view from Shuwaymis, NW Saudi Arabia. Holocene 26, 1822–1834 (2016).

    Google Scholar 

  13. Arz, H. W., Lamy, F., Pätzold, J., Müller, P. J. & Prins, M. Mediterranean moisture source for an Early-Holocene Humid Period in the northern Red Sea. Science 300, 118–121 (2013).

    Google Scholar 

  14. Enzel, Y., Kushnir, Y. & Quade, J. The middle Holocene climatic records from Arabia: reassessing lacustrine environments, shift of ITCZ in Arabian Sea, and impacts of the southwest Indian and African monsoons. Glob. Planet. Change 129, 69–91 (2015).

    Google Scholar 

  15. Engel, M. et al. Lakes or wetlands? A comment on ‘The middle Holocene climatic records from Arabia: Reassessing lacustrine environments, shift of ITCZ in Arabian Sea, and impacts of the southwest Indian and African monsoons’ by Enzel et al. Glob. Planet. Change 148, 258–267 (2017).

    Google Scholar 

  16. Kutzbach, J. E. & Liu, Z. Response of the African Monsoon to orbital forcing and ocean feedbacks in the middle Holocene. Science 278, 440–443 (1997).

    CAS  Google Scholar 

  17. Rosenberg, T. M. et al. Middle and Late Pleistocene humid periods recorded in palaeolake deposits of the Nafud desert, Saudi Arabia. Quat. Sci. Rev. 70, 109–123 (2013).

    Google Scholar 

  18. Shanahan, T. M. et al. The time-transgressive termination of the African Humid Period. Nat. Geosci. 8, 140–144 (2015).

    CAS  Google Scholar 

  19. Kutzbach, J. E., Chen, G., Cheng, H., Edwards, R. L. & Liu, Z. Potential role of winter rainfall in explaining increased moisture in the Mediterranean and Middle East during periods of maximum orbitally-forced insolation seasonality. Clim. Dyn. 42, 1079–1095 (2014).

    Google Scholar 

  20. Rubin, S., Ziv, B. & Paldor, N. Tropical plumes over eastern North Africa as a source of rain in the Middle East. Mon. Weather Rev. 135, 4135–4148 (2007).

    Google Scholar 

  21. Tubi, A., Dayan, U. & Lensky, I. M. Moisture transport by tropical plumes over the Middle East: a 30-year climatology. Q. J. R. Meteorol. Soc. 143, 3165–3176 (2017).

    Google Scholar 

  22. Waldmann, N., Torfstein, A. & Stein, M. Northward intrusions of low- and mid-latitude storms across the Saharo-Arabian belt during past interglacials. Geology 38, 567–570 (2010).

    CAS  Google Scholar 

  23. Skinner, C. B. & Poulsen, C. J. The role of fall season tropical plumes in enhancing Saharan rainfall during the African Humid Period. Geophys. Res. Lett. 43, 349–358 (2016).

    Google Scholar 

  24. Hausleiter, A., Eichmann, R. & Al-Najem, M. (eds.). Taymā’ I (Archaeopress, Oxford, 2018).

  25. Breeze, P. S. et al. Remote sensing and GIS techniques for reconstructing Arabian palaeohydrology and identifying archaeological sites. Quat. Int. 382, 98–119 (2015).

    Google Scholar 

  26. Bar-Matthews, M., Ayalon, A., Gilmour, M., Matthews, A. & Hawkesworth, C. J. Sea–land oxygen isotopic relationships from planktonic foraminifera and speleothems in the Eastern Mediterranean region and their implication for paleorainfall during interglacial intervals. Geochim. Cosmochim. Acta 67, 3181–3199 (2003).

    CAS  Google Scholar 

  27. Grant, K. M. et al. Rapid coupling between ice volume and polar temperature over the past 150,000 years. Nature 491, 744–747 (2012).

    CAS  Google Scholar 

  28. Cheng, H. et al. The climate variability in northern Levant over the past 20,000 years. Geophys. Res. Lett. 42, 8641–8650 (2015).

    Google Scholar 

  29. Fleitmann, D. et al. Holocene ITCZ and Indian monsoon dynamics recorded in stalagmites from Oman and Yemen (Socotra). Quat. Sci. Rev. 26, 170–188 (2007).

    Google Scholar 

  30. Cheng, H. et al. Timing and structure of the 8.2 kyr B.P. event inferred from δ18O records of stalagmites from China, Oman, and Brazil. Geology 37, 1007–1010 (2009).

    CAS  Google Scholar 

  31. Schulz, E. & Whitney, J. W. Upper Pleistocene and Holocene lakes in the An Nafud, Saudi Arabia. Hydrobiologia 143, 175–190 (1986).

    Google Scholar 

  32. Parton, A. et al. Middle-late Quaternary palaeoclimate variability from lake and wetland deposits in the Nefud Desert, Northern Arabia. Quat. Sci. Rev. 202, 78–97 (2018).

    Google Scholar 

  33. Zielhofer, C. et al. Climate forcing and shifts in water management on the Northwest Arabian Peninsula (mid-Holocene Rasif wetlands, Saudi Arabia). Quat. Int. 473, 120–140 (2018).

    Google Scholar 

  34. Engel, M. et al. The early Holocene humid period in NW Saudi Arabia—sediments, microfossils and palaeo-hydrological modelling. Quat. Int. 266, 131–141 (2012).

    Google Scholar 

  35. Neugebauer, I. et al. Implications of S1 tephra findings in Dead Sea and Tayma palaeolake sediments for marine reservoir age estimation and palaeoclimate synchronisation. Quat. Sci. Rev. 170, 269–275 (2017).

    Google Scholar 

  36. Pint, A. et al. How to discriminate athalassic and marginal marine microfaunas: Foraminifera and other fossils from an early Holocene continental lake in northern Saudi Arabia. J. Foram. Res. 47, 175–187 (2017).

    Google Scholar 

  37. Parker, A. et al. A record of Holocene climate change from lake geochemical analyses in southeastern Arabia. Quat. Res. 66, 465–476 (2006).

    Google Scholar 

  38. Sachse, D. et al. Molecular paleohydrology: interpreting the hydrogen-isotopic composition of lipid biomarkers from photosynthesizing organisms. Annu. Rev. Earth Planet. Sci. 40, 221–249 (2012).

    CAS  Google Scholar 

  39. Collins, J. A. et al. Estimating the hydrogen isotopic composition of past precipitation using leaf-waxes from western Africa. Quat. Sci. Rev. 65, 88–101 (2013).

    Google Scholar 

  40. Ginau, A., Engel, M. & Brückner, H. Holocene chemical precipitates in the continental sabkha of Tayma (NW Saudi Arabia). J. Arid Environ. 84, 26–37 (2012).

    Google Scholar 

  41. Garber, R., Levy, Y. & Friedman, G. The sedimentology of the Dead Sea. Carbonates Evaporites 2, 43–57 (1987).

    Google Scholar 

  42. Heim, C., Nowaczyk, N., Negendank, J., Leroy, S. & Ben-Avraham, Z. Near East desertification: evidence from the Dead Sea. Naturwissenschaften 84, 398–401 (1997).

    CAS  Google Scholar 

  43. Leng, M. J. & Marshall, J. D. Palaeoclimate interpretation of stable isotope data from lake sediment archives. Quat. Sci. Rev. 23, 811–831 (2004).

    Google Scholar 

  44. Ben Dor, Y. et al. The paleohydrological implications of aragonite precipitation under contrasting climates in the endorheic Dead Sea and its precursors revealed by experimental investigations. Chem. Geol. 576, 120261 (2021).

    Google Scholar 

  45. Torfstein, A., Gavrieli, I., Katz, A., Kolodny, Y. & Stein, M. Gypsum as a monitor of the paleo-limnological–hydrological conditions in Lake Lisan and the Dead Sea. Geochim. Cosmochim. Acta 72, 2491–2509 (2008).

    CAS  Google Scholar 

  46. Arz, H. W., Lamy, F. & Pätzold, J. A pronounced dry event recorded around 4.2 ka in brine sediments from the northern Red Sea. Quat. Res. 66, 432–441 (2006).

    Google Scholar 

  47. IAEA & WMO. Global Network of Isotopes in Precipitation. The GNIP Database. (2018).

  48. Rohling, E. J. & Palike, H. Centennial-scale climate cooling with a sudden cold event around 8,200 years ago. Nature 434, 975–979 (2005).

    CAS  Google Scholar 

  49. Gasse, F. Hydrological changes in the African tropics since the Last Glacial Maximum. Quat. Sci. Rev. 19, 189–211 (2000).

    Google Scholar 

  50. Blanchet, C. L. et al. High- and low-latitude forcing of the Nile River regime during the Holocene inferred from laminated sediments of the Nile deep-sea fan. Earth Planet. Sci. Lett. 364, 98–110 (2013).

    CAS  Google Scholar 

  51. Rohling, E. J., Marino, G. & Grant, K. M. Mediterranean climate and oceanography, and the periodic development of anoxic events (sapropels). Earth Sci. Rev. 143, 62–97 (2015).

    CAS  Google Scholar 

  52. Tesi, T. et al. Large-scale response of the Eastern Mediterranean thermohaline circulation to African monsoon intensification during sapropel S1 formation. Quat. Sci. Rev. 159, 139–154 (2017).

    Google Scholar 

  53. Pross, J. et al. Massive perturbation in terrestrial ecosystems of the Eastern Mediterranean region associated with the 8.2 kyr BP climatic event. Geology 37, 887–890 (2009).

    Google Scholar 

  54. Dinies, M., Neef, R., Plessen, B. & Kürschner, H. In The Archaeology of North Arabia: Oases and Landscapes (ed. Luciani, M.) 57–78 (Austrian Academy of Sciences Press, 2016).

  55. Weltje, G. J. & Tjallingii, R. Calibration of XRF core scanners for quantitative geochemical logging of sediment cores: theory and application. Earth Planet. Sci. Lett. 274, 423–438 (2008).

    CAS  Google Scholar 

  56. Weltje, G. J. et al. In Micro-XRF Studies of Sediment Cores (eds Croudace, I. W. & Rothwell R. G.) 507–534 (Springer, Dordrecht, 2015).

  57. Aitchison, J. The statistical analysis of compositional data. J. R. Stat. Soc. Ser. B 44, 139–160 (1982).

    Google Scholar 

  58. Brown, T. A., Nelson, D. E., Mathewes, R. W., Vogel, J. S. & Southon, J. R. Radiocarbon dating of pollen by accelerator mass spectrometry. Quat. Res. 32, 205–212 (1989).

    CAS  Google Scholar 

  59. Vandergoes, M. J. & Prior, C. A. AMS Dating of pollen concentrates—a methodological study of late Quaternary sediments from South Westland, New Zealand. Radiocarbon 45, 479–491 (2003).

    Google Scholar 

  60. Regnéll, J. & Everitt, E. Preparative centrifugation—a new method for preparing pollen concentrates suitable for radiocarbon dating by AMS. Veg. Hist. Archaeobot. 5, 201–205 (1996).

    Google Scholar 

  61. Nakagawa, T. et al. Dense-media separation as a more efficient pollen extraction method for use with organic sediment/deposit samples: comparison with the conventional method. Boreas 27, 15–24 (1998).

    Google Scholar 

  62. Brauer, A. & Casanova, J. Chronology and depositional processes of the laminated sediment record from Lac d’Annecy, French Alps. J. Paleolimnol. 25, 163–177 (2001).

    Google Scholar 

  63. Blaauw, M. & Christen, J. A. Flexible paleoclimate age-depth models using an autoregressive gamma process. Bayesian Anal. 6, 457–474 (2011).

    Google Scholar 

  64. Reimer, P. J. et al. IntCal13 and Marine13 radiocarbon age calibration curves 0–50,000 Years cal BP. Radiocarbon 55, 1869–1887 (2013).

    CAS  Google Scholar 

  65. Spötl, C. & Vennemann, T. W. Continuous-flow isotope ratio mass spectrometric analysis of carbonate minerals. Rapid Commun. Mass Spectrom. 17, 1004–1006 (2003).

    Google Scholar 

  66. Bowen, G. J. Isoscapes: spatial pattern in isotopic biogeochemistry. Annu. Rev. Earth Planet. Sci. 38, 161–187 (2010).

    CAS  Google Scholar 

  67. Kim, S.-T. & O’Neil, J. R. Equilibrium and nonequilibrium oxygen isotope effects in synthetic carbonates. Geochim. Cosmochim. Acta 61, 3461–3475 (1997).

    CAS  Google Scholar 

  68. Sachse, D., Radke, J. & Gleixner, G. δD values of individual n-alkanes from terrestrial plants along a climatic gradient—implications for the sedimentary biomarker record. Org. Geochem. 37, 469–483 (2006).

    CAS  Google Scholar 

  69. Aichner, B. et al. Hydroclimate in the Pamirs was driven by changes in precipitation‐evaporation seasonality since the last glacial period. Geophys. Res. Lett. 46, 13972–13983 (2019).

    Google Scholar 

  70. Horita, J. & Wesolowski, D. J. Liquid-vapor fractionation of oxygen and hydrogen isotopes of water from the freezing to the critical temperature. Geochim. Cosmochim. Acta 58, 3425–3437 (1994).

    CAS  Google Scholar 

  71. Hausleiter, A. In Cultural heritages of water—the cultural heritages of water in the Middle East and Maghreb (ed. ICOMOS) 313–343 (UNESCO World Heritage Convention, International Council on Monuments and Sites, 2017).

  72. Wellbrock, K. et al. In Taymāʾ I (eds. Hausleiter, A., Eichmann, R. & al-Najem, M.) 145–198 (Archaeopress, Oxford, 2018)

  73. Al-Sagaby, A. & Moallim, A. Isotopes based assessment of groundwater renewal and related anthropogenic effects in water scarce areas: sand dunes study in Qasim area. Saudi Arabia. IAEA-TECDOC 1246, 221–229 (2001).

    Google Scholar 

  74. Alyamani, M. S. Isotopic composition of rainfall and ground-water recharge in the western province of Saudi Arabia. J. Arid Environ. 49, 751–760 (2001).

    Google Scholar 

  75. Michelsen, N. et al. Isotopic and chemical composition of precipitation in Riyadh, Saudi Arabia. Chem. Geol. 413, 51–62 (2015).

    CAS  Google Scholar 

  76. Neugebauer, I. et al. Geochemical and sedimentological data of the Tayma palaeolake record. GFZ Data Services.

  77. Fick, S. E. & Hijmans, R. J. WorldClim 2: new 1-km spatial resolution climate surfaces for global land areas. Int. J. Climatol. 37, 4302–4315 (2017).

    Google Scholar 

  78. Akçar, N. & Schlüchter, C. Paleoglaciations in Anatolia: a schematic review and first results. E&G Quat. Sci. J. 55, 102–121 (2005).

    Google Scholar 

  79. Wellbrock, K., Strauss, M., Külls, C. & Grottker, M. In Des refuges aux oasis: Vivre en milieu aride de la Préhistoire à aujourd’hui. XXXVIIIe rencontres internationales d’archéologie et d’histoire d’Antibes (eds. Purdue, L., Charbonnier, J. & Khalidi, L.) 231–249 (Éditions APDCA, 2018).

  80. Danielson, J. J. & Gesch, D. B. Global multi-resolution terrain elevation data 2010 (GMTED2010). USGS Open-File Rep. 2011–1073 (2011).

  81. Laskar, J. et al. A long-term numerical solution for the insolation quantities of the Earth. Astron. Astrophys. 428, 261–285 (2004).

    Google Scholar 

  82. Thomas, E. R. et al. The 8.2 ka event from Greenland ice cores. Quat. Sci. Rev. 26, 70–81 (2007).

    Google Scholar 

Download references


This study is a contribution to the interdisciplinary research project “CLEAR—Holocene CLimatic Events of Northern ARabia” (, funded by Deutsche Forschungsgemeinschaft (DFG grants EN977/2-1, PL535/2-1, FR1489/5-1). The CLEAR project was kindly invited and supported by the Saudi-German Joint Archaeological Project at Tayma, led by Ricardo Eichmann, Arnulf Hausleiter (German Archaeological Institute) and Mohammed H. al-Najem (Museum of Archaeology and Ethnography, Tayma). We are grateful for the research permit issued by the Saudi Commission for Tourism and National Heritage, represented by HRH Prince Sultan bin Salman bin Abdulaziz al-Sa’ud. Permissions to export core samples were kindly granted under the umbrella of the Joint Saudi-German Archaeological Project at Tayma. Further support by Jenny Friedrich (sample preparation), Sylvia Pinkerneil (GFZ) (hydrogen isotope measurements on water), Doreen Noack, Andrea Vieth-Hillebrand (GFZ) (hydrogen isotope measurements on n-alkanes), Dieter Berger (GFZ) (preparation of thin sections), Christin Kramer (documentation) and Kirstin Jacobson (language editing) is greatly appreciated. Detailed comments of three anonymous reviewers greatly improved the paper. For the publication fee we acknowledge financial support by Deutsche Forschungsgemeinschaft within the funding programme “Open Access Publikationskosten” as well as by Heidelberg University and GFZ Potsdam.

Author information

Authors and Affiliations



B.P., I.N., M.E. and P.F. designed the study. M.E., H.B., M.D. and A.P. collected the sediment cores. M.D. and I.N. constructed the age model. I.N., B.P., R.T., P.H. and A.B. contributed the sedimentological and microfacies data. B.P and I.N. contributed stable-isotope data on water and carbonates. N.D., V.F.S. and G.G. contributed the leaf-wax n-alkane data. A.P. and P.F. contributed foraminiferal and ostracod analyses. A.S. and K.J.K. contributed the diatom analysis. I.N., M.E. and B.P. wrote the paper. All authors discussed and commented on the paper.

Corresponding author

Correspondence to Max Engel.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Communications Earth & Environment thanks Jonathan Dean and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Primary Handling Editor: Joe Aslin. Peer reviewer reports are available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Neugebauer, I., Dinies, M., Plessen, B. et al. The unexpectedly short Holocene Humid Period in Northern Arabia. Commun Earth Environ 3, 47 (2022).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI:


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing