Two phases of the Holocene East African Humid Period: Inferred from a high-resolution geochemical record off Tanzania

Earth and Planetary Science Letters 460 (2017) 123–134

Contents lists available at ScienceDirect

Earth and Planetary Science Letters www.elsevier.com/locate/epsl

Two phases of the Holocene East African Humid Period: Inferred from a high-resolution geochemical record off Tanzania Xiting Liu a,b,∗ , Rebecca Rendle-Bühring b , Holger Kuhlmann b , Anchun Li a a b

Key Laboratory of Marine Geology and Environment, Institute of Oceanology, Chinese Academy of Sciences, Qingdao 266071, China MARUM – Center for Marine Environmental Sciences and Faculty of Geosciences, University of Bremen, D-28359 Bremen, Germany

a r t i c l e

i n f o

Article history: Received 10 August 2016 Received in revised form 2 December 2016 Accepted 11 December 2016 Available online xxxx Editor: M. Frank Keywords: African Humid Period Intertropical Convergence Zone Congo Air Boundary Holocene X-ray fluorescence scanner

a b s t r a c t During the Holocene, the most notably climatic change across the African continent is the African Humid Period (AHP), however the pace and primary forcing for this pluvial condition is still ambiguous, particularly in East Africa. We present a high-resolution marine sediment record off Tanzania to provide insights into the climatic conditions of inland East Africa during the Holocene. Major element ratios (i.e., log-ratios of Fe/Ca and Ti/Ca), derived from X-Ray Fluorescence scanning, have been employed to document variations in humidity in East Africa. Our results show that the AHP is represented by two humid phases: an intense humid period from the beginning of the Holocene to 8 ka (AHP I); and a moderate humid period spanning from 8 to 5.5 ka (AHP II). On the basis of our geochemical record and regime detection, the termination of the AHP initiated at 5.5 ka and ceased around 3.5 ka. Combined with other paleoclimatic records around East Africa, we suggest that the humid conditions in this region responded to Northern Hemisphere (NH) summer insolation. The AHP I and II might have been related to an eastward shift of the Congo Air Boundary and warmer conditions in the western Indian Ocean, which resulted in additional moisture being delivered from the Atlantic and Indian Oceans during the NH summer and autumn, respectively. We further note a drought event throughout East Africa north of 10◦ S around 8.2 ka, which may have been related to the southward migration of the Intertropical Convergence Zone in response to the NH cooling event. © 2016 Elsevier B.V. All rights reserved.

1. Introduction Various well documented records have demonstrated that northern Africa experienced a more pluvial early–mid Holocene (12–5.5 ka; e.g., Gasse, 2000; Tierney et al., 2011; Junginger and Trauth, 2013; Blanchet et al., 2015; Shanahan et al., 2015). This relative humid period, termed the African Humid Period (AHP; deMenocal et al., 2000), extends from North Africa to as far south as 10◦ S in East Africa (Gasse, 2000; Castañeda et al., 2007; Burrough and Thomas, 2013). Existing hydrological reconstructions however exhibit significant heterogeneity in both the timing and abruptness of the AHP termination, and in the pace of humid–arid transitional interval, which is still the subject of intense debate (Chevalier and Chase, 2015; Castañeda et al., 2016). Records from the northwestern African margin sediments show a synchronous abrupt termination of the AHP around 5.5 ka

*

Corresponding author at: Key Laboratory of Marine Geology and Environment, Institute of Oceanology, Chinese Academy of Sciences, Qingdao 266071, China. Fax: +86 532 82898521. E-mail address: [email protected] (X. Liu). http://dx.doi.org/10.1016/j.epsl.2016.12.016 0012-821X/© 2016 Elsevier B.V. All rights reserved.

(deMenocal et al., 2000; Kuhlmann et al., 2004; McGee et al., 2013). In contrast, in the eastern Sahara, pollen and sedimentological records from Lakes Yoa and Chad document a gradual end in the humid condition (Kröpelin et al., 2008; Amaral et al., 2013; Claussen et al., 2013; Francus et al., 2013). In Eastern Mediterranean sediments, the end of the AHP is expressed by a progressive hydrological shift in the Nile River based on geochemical and sedimentological proxy data (Hamann et al., 2008; Revel et al., 2010; Blanchet et al., 2013; Ehrmann et al., 2013; Flaux et al., 2013; Weldeab et al., 2014b). Furthermore, isotope hydrological indicators from marine sediments off the central western Africa coast suggest a gradual termination of the AHP (Schefuß et al., 2005; Weldeab et al., 2007). It seems that the abrupt transition is limited in the Western Sahara, where its precipitation is controlled by the Western African monsoon with vegetation feedback (Hély et al., 2009). Recent research, however, shows that the abrupt termination of the APH extends to East Africa (Garcin et al., 2012; Tierney and deMenocal, 2013; Forman et al., 2014; Morrissey and Scholz, 2014; Castañeda et al., 2016). Based on the δ Dwax data from Lakes Challa and Tanganyika and the Gulf of Aden, Tierney and deMenocal (2013) hypothesize that the rapid termination of the AHP in East Africa reflects convection feedbacks associ-

124

X. Liu et al. / Earth and Planetary Science Letters 460 (2017) 123–134

ated with Indian Ocean sea-surface temperature (SST). In contrast, Berke et al. (2012) found that δ Dwax derived from Lake Victoria show a gradual termination of the AHP, indicating a predominantly insolation forcing. These different findings could be caused by changes in the source of moisture rather the amount effect associated with local precipitation as proposed by Leduc et al. (2013). Indeed, recent studies have demonstrated that delepted δ Dwax during the early Holocene may indicate an increase in moisture derived from the Atlantic Ocean, which is more isotopically depleted than Indian Ocean sourced moisture (Costa et al., 2014; Castañeda et al., 2016). Under the influence of the Atlantic and Indian Oceans, the climate system in East Africa is complex (Nicholson and Kim, 1997). The available data concerning the humid–arid transition in East Africa mainly depend on the lacustrine sediments and vary significantly from site to site. For instance, shoreline reconstruction documents a rapid drop in lake level for Lake Turkana (Garcin et al., 2012; Forman et al., 2014; Morrissey and Scholz, 2014), whereas a gradual humid–arid transition is recorded in the Lake Chew Bahir and Lake Suguta, located to the north and south of the Lake Turkana (Foerster et al., 2012; Junginger and Trauth, 2013). In contrast to these terrestrial archives, marine sediments accumulating along the African continent margins provide complete and well-dated climate sequence (deMenocal, 2014). However, to date, few sediment cores are available off East Africa comparing with regions off West Africa, which has led a limited understanding of the Holocene climatic changes in East Africa. Therefore, high-resolution marine climatic profiles (e.g., geochemical scanning data) are imperative to allow us to study the AHP (Bard, 2013). Here, we present a high-resolution marine sediment core, obtained off Tanzania, western Indian Ocean (WIO), to reveal the pace and magnitude of environmental changes during the AHP in East Africa. This work will make contributions to critical debates, such as, the role of insolation in forcing East African climatic change, the relative importance of the Atlantic and Indian Ocean SSTs in controlling the rainfall in East Africa, and the linkage between high and low latitude climate systems during the AHP.

Fig. 1. Present climatic system of Africa. Locations of the Inter-tropical Convergence Zone (ITCZ; solid line) and the Congo Air Boundary (CAB; dashed line) for Northern Hemisphere summer (JJA: June, July, and August) and winter (DJF: December, January, and February) season, modified from Willis et al. (2013). Locations of the research sites and paleoclimatic records mentioned in the text are presented. Solid orange circles represent terrestrial records, while blue ones represent marine record. Core GeoB12605-3 is indicated by pink star. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

warmer (colder)-than-usual SSTs in the WIO (Hastenrath, 2007; Brown et al., 2009; Bahaga et al., 2015). Such shifts in zonal circulation can be caused by anomalously warm SSTs in the western Indian Ocean (Ummenhofer et al., 2009), positive IOD events (Saji et al., 1999), and the ENSO (Nicholson and Kim, 1997).

2. Regional setting

2.2. Oceanography

2.1. Present climate system

The current system in the target region is mainly influenced by the north-flowing East African Coastal Current (EACC), which is controlled by a strong seasonal monsoon (Beal et al., 2013; Manyilizu et al., 2016). The EACC and the southward flowing Mozambique Current (MC) originate from the South Equatorial Current (SEC), which passes the northern part of Madagascar and continues westward to the coast of Africa near 11◦ S (Fig. 2). During the southeast monsoon (Fig. 2a), cold, nutrient-rich waters are upwelled to the surface (McClanahan, 1988), however, upwelling is limited to northern region of the WIO. In contrast, surface waters south of 4◦ S are stratified all year-round, characterized by relatively low surface and benthic productivity, associated with low a nutrient content (Birch et al., 2013). During the northeast monsoon the EACC moves away from the coast near 3◦ S and combines with the southward flowing Somali Current (SC) to form the Equatorial Countercurrent (ECC, Fig. 2b, Kohn and Zonneveld, 2010). The East African coast is drained by several major rivers (e.g., Pangani, Rufiji, and Ruvuma Rivers in Fig. 2a) and numerous minor rivers (Shaghude, 2007). Specific to our research locations, the major sediment discharge comes from the Pangani River (Liu et al., 2016). Headwaters of the Pangani River are located in the volcanic region of Mt. Kilimanjaro and Mt. Meru, which is covered with olivine and alkaline basalts, phonolites, trachytes, nephelinites and pyroclastics (Schluter, 2008).

Modern East Africa experiences a semiannual rainfall cycle (Nicholson, 2000), which is driven by the migration of the Intertropical Convergence Zone (ITCZ) back and forth across the equator (Fig. 1). There are two distinct rainy seasons in March–May (MAM; the long rains), and October–December (OND; the short rains) between the transition of monsoon. The East African longrains season is characterized by heavy rainfall in April and May that is driven by the northward migration of the ITCZ, whereas the heavy rainfall experienced during the short-rains season is a result of the migration of the ITCZ to the south (Black et al., 2003). In addition, the area is influenced by the Congo Air Boundary (CAB), separating the air masses containing moisture derived from the Atlantic and Indian Ocean. Long rains over the northeastern Tanzania are associated with advection of moisture from both the Indian Ocean and Atlantic Oceans (Kijazi and Reason, 2012). As the southward migration of the ITCZ is more rapid than the northward migration, the period of heavy rainfall is shorter during the short-rains season (Clark et al., 2003). Rainfall during the short rains is related to Indian Ocean SST anomalies, attributed to the Indian Ocean Dipole (IOD) and/or El Niño/Southern Oscillation (ENSO; Nicholson, 2000). Observations and simulations indicate that wet (dry) short rains seasons tend to correspond to colder (warmer)-than-usual SSTs in the tropical eastern Indian Ocean and

X. Liu et al. / Earth and Planetary Science Letters 460 (2017) 123–134

125

Fig. 2. Current patterns (solid lines) and wind directions (dashed lines) during (a) the southeast (SE) and (b) the northeast (NE) monsoons in East Africa, modified from Kohn and Zonneveld (2010). Location of core GeoB12605-3 is denoted with solid stars in figure b; cores GeoB12615-4 and GeoB12624-1 are also shown (open stars). SC: Somali Current; EACC: East African Coastal Current; SEC: South Equatorial Current; MC: Mozambique Current; ECC: Equatorial Counter Current.

Fig. 3. (a) Age model and downcore variations in sedimentation rates of core GeoB12605-3, modified from Kuhnert et al. (2014) and Liu et al. (2016). The upper and lower errors of the sedimentation rate are indicated by the dashed lines. Error bars mark the ±1 sigma range. (b)–(c) Distributions of total organic carbon (TOC) and carbonate (CaCO3 ) of bulk sediment (wt.%) are shown along the depth. The dashed lines indicate the average values of relative weight percentage of TOC and CaCO3 .

3. Materials and methods 3.1. Core GeoB12605-3 The gravity core GeoB12605-3 (5◦ 34.39 S, 39◦ 06.49 E) has been retrieved in the WIO during the R/V METEOR Cruise M75/2 in February 2008 (Savoye et al., 2013). It was drilled in the Pemba Channel off the Pangani River mouth from 197 m water depth, with total length 500 cm. Sediment column grades from dark gray

fine-grained siliciclastic muddy sediments to olive gray mixed biogenic calcareous-siliciclastic sandy sediments from the core base to top (Fig. 3a). The chronology for core GeoB12605-3 is based on 10 accelerator mass spectrometry (AMS) radiocarbon (14 C) dates (Table 1). 14 C-AMS dates were measured on mixed planktonic foraminifera shells at the Poznan´ Radiocarbon Laboratory and the National Ocean Sciences Accelerator Mass Spectrometry Facility, Woods

126

X. Liu et al. / Earth and Planetary Science Letters 460 (2017) 123–134

Table 1 AMS radiocarbon dates and calibrated ages for sediment core GeoB12605-3, based on Kuhnert et al. (2014). Sample ID

GeoB12605-3 GeoB12605-3 GeoB12605-3 GeoB12605-3 GeoB12605-3 GeoB12605-3 GeoB12605-3 GeoB12605-3 GeoB12605-3 GeoB12605-3

3 cm 60 cm 122 cm 168 cm 180 cm 220 cm 266 cm 352 cm 434 cm 487 cm

Measured 14 C-age (yr BP)

Conventional 14 C-age (yr BP)

Probability maximum (cal yr BP)

−1 s

+1 s

1975 ± 30 2900 ± 30 3560 ± 30 4510 ± 35 5050 ± 30 5790 ± 30 7230 ± 35 8000 ± 35 8970 ± 40 9180 ± 50

1363 ± 72 2457 ± 109 3284 ± 81 4504 ± 99 5208 ± 96 6037 ± 88 7558 ± 64 8313 ± 73 9477 ± 62 9725 ± 131

1355 2455 3305 4500 5260 6040 7560 8330 9485 9710

64 107 102 95 148 91 66 90 70 116

79 110 59 102 44 84 61 56 54 146

Hole (Table 1). Radiocarbon ages were converted into calendar years with the CALIB program (Calib 6.0), using the calibration curve marine 09 ( R = 150 ± 60).  R is based on the 10 geographically closest calibration points, where the minimum and maximum values have been omitted, and the weighted average was calculated from the remaining 8 points. This excludes the very young age from Watamu Reef (Grumet et al., 2002), and one extremely old age from the Seychelles (Southon et al., 2002).

Tjallingii, 2008). Comparison of XRF scanning data and conventional EDP-XRF measurements from 243 bulk samples of sediment core GeoB12605-3 (Liu, 2014), indicates log-ratios of intensities and concentrations show good correlation (Appendix A). Therefore, log-ratios of elemental intensities represent their corresponding sedimentary concentration log-ratios in our results.

3.2. Total organic carbon (TOC) and carbonate

Geochemical datasets were analyzed for regime shifts by Sequential Regime Shift Detector version 3.4 using red-noise estimation and pre-whitening options (Rodionov, 2004; http://www. climatelogic.com). The program uses a t-tests sequential algorithm to identify if the next value is significantly different from the previous regime. If so, the point is marked as a potential change point, and subsequent observations are used to confirm or reject the regime shift assumption (see details in Rodionov, 2004). The determination of the regimes is strongly influenced by the choice of the cut-off length l, which determines the minimum length of a regime, and the significance level P of the t-test. For the analyses of our time series, we used l = 150 and P = 0.1.

For core GeoB12605-3, 81 bulk sediment samples were freezedried and then ground into a homogeneous fine powder with an agate mortar. Approximately 1 g of powder was used for Total Carbon (TC) and TOC determination, measured with a LECO-CS 200 elemental analyzer at the Faculty of Geosciences, University of Bremen. Total Inorganic Carbon (TIC) was calculated using the equation (1):

TIC (%) = TC (%) − TOC (%)

(1)

To express TIC as a percent calcium carbonate (CaCO3 ), the concentration of CaCO3 was then calculated by applying the following equation:

CaCO3 (%) = (TC − TOC) ∗ 8.33

(2)

3.3. X-ray fluorescence (XRF) core scanning XRF core scanning data were collected down-core over a 1 cm2 area using generator settings of 10 kV, a current of 0.35 mA, and a sampling time of 30 s directly at the split core surface of the archive half with XRF Core Scanner II (AVAATECH Serial No. 2) at the MARUM – University of Bremen (Röhl et al., 2007; Tjallingii et al., 2007; Westerhold et al., 2007). The split core surface was covered with a 4 μm thin SPEXCerti Prep Ultralene1 foil to avoid contamination of the XRF measurement unit and desiccation of the sediment. The reported data have been acquired by a Canberra X-PIPS Detector (Model SXP 5C-200-1500 V3) with 200 eV X-ray resolution, the Canberra Digital Spectrum Analyzer DAS 1000, and an Oxford Instruments 50W XTF5011 X-Ray tube with rhodium (Rh) target material. Raw data spectra were processed by the analysis of X-ray spectra by Iterative Least square software (WIN AXIL) package from Canberra Eurisys. High-resolution geochemical analyses over the total length of the core sediment were performed at 1 cm sampling interval on core GeoB12605-3. Scanning results could be influenced by variable water content and grain-size distributions (Tjallingii et al., 2007; Hennekam and de Lange, 2012), thus they are presented as logarithmic intensity ratios of indicator elements, log(Fe/Ca) and log(Ti/Ca), to reduce the sediment matrix effect (Weltje and

3.4. Sequential regime shift detection

4. Results 4.1. Age model, TOC and CaCO3 Ages are linearly interpolated between the 10 calibrated ages. Smoothing functions have not been applied to the age-depth relationship, because these might conceal true abrupt sedimentation rate changes (Kuhnert et al., 2014). According to the age model, core GeoB12605-3 covers continues sedimentation from 9.7 ka until 1.3 ka, with an average sedimentation rate of almost 60 cm/ka (Kuhnert et al., 2014). The sedimentation rates are high due to its location close to the coast (Fig. 2b), and a decreasing trend from core base to top could be observed (Fig. 3a). The TOC content of core GeoB12605-3 is generally low, ranging from 0.3 to 0.8 wt.% with an average value at 0.5 wt.% (Fig. 3c). We observe a decrease between 500 and 250 cm (10–7 ka) and an increase in TOC between 250 and 150 cm (7–4 ka). After that, there is a distinct decrease in the content of TOC, which becomes stable with an average of 0.4 wt.%. CaCO3 contents have been calculated for core GeoB12605-3, which range from 20 to 40 wt.% with an average of 28 wt.% (Fig. 3b). The carbonate record shows an increase between 500 and 320 cm (10–8 ka; the early Holocene), followed by relative stable until 190 cm (5.5 ka). Then CaCO3 concentrations exhibit a further increase between 190 and 130 cm (the upper mid-Holocene), which reaches a maximum around 130 cm (3.5 ka), matching the start of the late Holocene. Since that time, there is a steady decrease in the amount of CaCO3 in the sediment.

X. Liu et al. / Earth and Planetary Science Letters 460 (2017) 123–134

127

5. Discussion 5.1. Indicator for climatic changes

Fig. 4. Temporal geochemical profiles and regime shifts in the XRF scanning data for GeoB12605-3. (a)–(c) Log-ratios of Fe/Ca, Ti/Ca, and Fe/Ti, obtained from XRF scanner on core GeoB12605-3, are plotted against age. Regime shifts in the log-ratios of Fe/Ca are indicated with a pink line. (d) Solid black lines on bottom indicate Regime Shift Index (RSI) values. A significant decrease in log-ratios of Fe/Ca and Ti/Ca around 8.2 ka is marked. Solid triangles represented radiocarbon ages referred from Kuhnert et al. (2014). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

4.2. XRF scanning results Log-ratios of Fe/Ca, Ti/Ca, and Fe/Ti against time, derived off equatorial East Africa, are presented in Fig. 4. We observe a similar pattern through time between log-ratios of Fe/Ca and Ti/Ca (Fig. 4a–b), thus we only describe the variation in log(Fe/Ca), representing variations in log(Fe/Ca) and log(Ti/Ca). During the early Holocene, the upper log-ratios of Fe/Ca between 9.7 and 8 ka in core GeoB12605-3 are stable and high (Fig. 4a). At the end of early Holocene, a sharp decrease in log(Fe/Ca) is recorded around 8.2 ka. For the period of the midHolocene (8–3.5 ka), the log(Fe/Ca) obtained from XRF scanner is represented by a moderate decrease between 8 and 7.5 ka, followed by a relative stable interval between 7.5 and 5.5 ka. The log(Fe/Ca) decreasing trend in the mid-Holocene resumes after 5.5 ka and reaches a minimum at ∼3.5 ka. The late Holocene (3.5–1.3 ka) part of core GeoB12605-3 is characterized by a slight increasing trend in log-ratios of Fe/Ca. In addition, log-ratios of Fe/Ti show relative constant patterns throughout the Holocene (Fig. 4c). 4.3. Regime shifts On the basis of the sequential t-test method (cut-off length = 150, P = 0.1) applied to log-ratios of Fe/Ca data set, five distinct regimes: 9.7–8.3 ka, 8.3–7.7 ka, 7.7–5.5 ka, 5.5–3.5 ka, and 3.5–1.3 ka, are determined (the red line in Fig. 4a). The strongest regime shift (displayed by the Regime Shift Index; RSI) in XRF scanning data is found at 3.5 ka (Fig. 4d), marking the beginning of the late Holocene. Another strong regime shift occurred around 5.5 ka, corresponding a transition period between 5.5 and 3.5 ka. Further a comparatively weak shift is detected around 8.2 ka (Fig. 4d).

5.1.1. Controlling factors for log-ratios of Fe/Ca and Ti/Ca In marine sediments, Fe and Ti are derived from the terrigenous component, whereas Ca commonly represents the biogenic carbonate component (Arz et al., 1998; Stuut et al., 2014). Therefore, log-ratios of Fe/Ca and/or Ti/Ca have been employed to trace changes in supply of river-suspended materials around East African continental margin (Ziegler et al., 2013; Revel et al., 2015). However, carbonate concentration is controlled not only by dilution by terrigenous materials but also influenced by changes in carbonate production and dissolution (Govin et al., 2012). In order to evaluate the contribution of marine productivity to the concentration of Ca, we plot the intensity of Ca as function of Ba/Ti intensity ratios, a productivity proxy (Paytan and Griffith, 2007), which shows very weak correlation (Fig. 5a). Thus, changes in carbonate production probably cannot significantly impact the carbonate concentration in the sediment due to low surface and benthic productivity in such oligotrophic environments (Birch et al., 2013). Aerobic organic matter degradation could cause carbonate dissolution in sediments above the chemical lysocline (e.g., Schulte and Bard, 2003; De Villiers, 2005). Weak correlation between TOC and carbonate (R 2 = 0.0015, Fig. 5b), and wellpreserved coccoliths in the surface sediment samples observed by Stolz et al. (2015), suggest that a significant aerobic TOC degradation can be ruled out. In addition, Fe in marine sediments is sensitive to redox changes in the sediment (Peterson et al., 2000), however it shows good correlation with Ti (R 2 ≈ 0.9, Fig. 5c), which is less sensible to diagenesis, indicating that sediments in this study have not been significantly affected by diagenesis. In contrast, the intensity of Ca and Fe (Ti) represents a significantly negative statistical correlation (Fig. 5d), which suggests that terrigenous dilution plays a primary role in their downcore distributions. Consequently, we suggest that the carbonate content of core sediments is mainly controlled by terrigenous dilution, and log(Fe/Ca) and (Ti/Ca) can be used to indicate the changes in the amount of terrigenous matter discharged into the WIO by the Pangani River. 5.1.2. Potential effects of sea-level change Since core GeoB12605 is located close to the coast, sediment discharge is potentially influenced by changes in the positions of the shoreline and the Pangani River mouth. Therefore, it is important to discuss the role of relative sea-level changes on these geochemical proxies in the research area. During the early Holocene, the sea level was ∼25 m below the present average level, and the shoreline would have migrated seawards (red line in Fig. 6a). Following a rapid sea-level rise between 9.6 and 8 ka (Zinke et al., 2003; Camoin et al., 2004; Punwong et al., 2013; Woodroffe et al., 2015), the distance between river mouth and the research site progressively increased (Fig. 6b–c). A decreasing trend in sediment discharge would be expected during such transgression period but our geochemical record (Fig. 6g), combined with sedimentation rate (Fig. 3a), indicates constant and high fluvial discharge of the Pangani River during the early Holocene. Meanwhile, high sediment inputs occurred from the Nile and Rufiji Rivers (Fig. 6e, h–i), which has been related to the AHP (Liu, 2014; Revel et al., 2015; Romahn et al., 2015). Thus high fluvial discharge of the Pangani, Rufiji and Nile Rivers, in spite of a rapid sea-level rise, reveals that sea-level changes could not have been the dominant factor controlling the fluvial discharge during the early Holocene. Sediments deposited in the studied region have become coarser and dominated by autochthonous quartz and feldspars since 8 ka (Fig. 6d; Liu et al., 2016), which indicates that the shoreline has

128

X. Liu et al. / Earth and Planetary Science Letters 460 (2017) 123–134

Fig. 5. (a) Correlation between Ca and Ba/Ti. (b) Correlation between contents of carbonate and TOC. (c) Elemental intensity of Fe plotted as a function of Ti intensity. (d) Elemental intensity of Ca plotted as a function of Fe intensity.

Fig. 6. Comparison geochemical data with climatic and sea-level changes. (a) Bathymetric map covering the area between the shoreline and the core location. The shoreline at 10 ka is indicated by the red line. (b) A distance–water depth profile between the shoreline and core site. Orange arrow represents the location of the shoreline at 10 ka, and the blue one indicates its present location. (c) Composite sea-level curve in the WIO based on the date from Mayotte (Zinke et al., 2003; Camoin et al., 2004). (d) Sand fraction in bulk sediments of core GeoB12605-3 (Liu et al., 2016). (e) Log(Fe/Ca) data from Core MS27PT (Site 2–1), indicating the amount of terrigenous materials supplied to the marine core site by the Nile River (Revel et al., 2015). (f) Lake Challa (Site 19) diatom-based oxygen isotope values based on Barker et al. (2011), representing ratios between precipitation and evaporation. (g) Log(Fe/Ca) derived from XRF scanner on core GeoB12605-3, this study, indicating the Pangani River discharge. (h) Ratios of Fe/Ca from core GeoB12615-4 (Site 21), representing the fluvial discharge of the Rufiji River (Romahn et al., 2015). (i) Log-ratios of Fe/Ca, indicating terrestrial versus marine component to sediments of core GeoB12624-1; modified from Liu (2014). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

X. Liu et al. / Earth and Planetary Science Letters 460 (2017) 123–134

129

Fig. 7. Comparison of paleoclimatic records around East Africa. Original radiocarbon dates have been converted into calendar years. The locations of each site have been shown in Fig. 1. The black bar indicates humid conditions, gray represents intermediate conditions, and blank marks arid conditions. Dashed black lines cover the drought event around 8.2 ka. For detail, see the supplementary material (Appendix A).

been stabilized at its present-day position (Woodroffe et al., 2015). After 8 ka, fluvial discharge of the Pangani and Rufiji Rivers decreased gradually (Fig. 6g–i), though sea-level has been stabilized (Fig. 6c). In comparison, log-ratios of Fe/Ca from core GeoB12605-3 correspond to fluctuations of ratios between precipitation and evaporation in Lake Challa (Fig. 6f; Barker et al., 2011). In addition, the Holocene log(Fe/Ca) data of core GeoB12605-3 follows the pace of geochemical records from the Mediterranean Sea (e.g., core MS27PT; Fig. 6e), representing the fluvial discharge of the Nile River in response climate conditions (Blanchet et al., 2013; Revel et al., 2014). Thus, we suggest that log-ratios of Fe/Ca and Ti/Ca from GeoB12605-3 are also mainly controlled by rainfall in the hinterland of East Africa. In fact, log-ratios of Fe/Ca and Ti/Ca have already been used for paleoclimatic reconstructions around East African continental margins, such as offshore northern East Africa (Revel et al., 2010, 2014, 2015; Blanchet et al., 2013) and southern East Africa (Schefuß et al., 2011; Ziegler et al., 2013; Romahn et al., 2015). According to the XRF scanning data and regime shift detection, climate in Tanzania can be distinguished into several phases during the Holocene (Fig. 6): 10–8 ka (AHP I): a intense humid period in East Africa; 8–5.5 ka (AHP II): a moderate humid period; 5.5–3.5 ka

(Transition): a gradual transitional period from humid to arid condition; 3.5–1.3 ka (Arid): an arid late Holocene. 5.2. Holocene African Humid Period in East Africa The expression of AHP in East Africa appears highly variable in space and time (Fig. 7). Nile River discharge (Sites 1–3), based on sedimentological and geochemical signals from the Eastern Mediterranean Sea (Hamann et al., 2008; Blanchet et al., 2013; Flaux et al., 2013; Revel et al., 2014; Weldeab et al., 2014b), as well as, palaeohydrological records from Lakes Victoria, Edward and Naivasha (Sites 14–17) (Maitima, 1991; Russell et al., 2003; Stager et al., 2003; Berke et al., 2012; Cockerton et al., 2015), demonstrate that the transition between the AHP and the arid late Holocene is gradual. Similarly, geochemical records from Lake Tana (Site 5) and marine core sediments (Site 8) reveal that the termination of the AHP was completed in two steps: one step at 8.6–7.8 and 8.5 ka, a second one between 6.8–4.2 ka and 6–3.8 ka (Jung et al., 2004; Marshall et al., 2011). In contrast, the isotopic signals (δ Dleaf wax ), obtained from the Gulf of Aden (Site 6), Lake Challa (Site 19), and Lake Tanganyika (Site 20) demonstrate a rapid humid–arid transition around 5 ka (Tierney et al., 2008, 2011; Tierney and deMenocal, 2013). Meanwhile, the lake-level reconstruction of Lakes Abhé (Site 7), Ziway-Shalla (Site 9) and

130

X. Liu et al. / Earth and Planetary Science Letters 460 (2017) 123–134

Turkana (Site 12) experienced a rapid drop in lake level around 5 ka (Gillespie et al., 1983; Garcin et al., 2012). In addition, there is a hinge zone near the Lake Malawi (Site 25), where sites located in the north of it experienced a humid early Holocene, while southern sites (Sites 25–27) were relative arid (Gasse, 2000; Castañeda et al., 2007; Schefuß et al., 2011; Wang et al., 2013). These paleoclimatic records indicate an abrupt or gradual termination of the AHP, raising the debate on the pace of the AHP in East Africa. Different proxies may show distinct response to the same climatic events even in the same archive, thus the controversy of the abruptness of the AHP might be partially attributed to the type of proxy applied (Castañeda et al., 2016). Though these climatic records seem heterogeneous, due to regional differences, various type of proxies, and different temporal resolution; there is still general pattern for Holocene climatic changes: East Africa north of 10◦ S experienced a intense humid period from the beginning of Holocene till 8 ka (APH I), and a second moderate humid phase between 8–5.5 ka during the midHolocene (AHP II). The second humid condition ended with an abrupt or gradual termination between ∼5.5 and 3.5 ka, followed by an arid late Holocene. Furthermore, a dramatic fluctuation in the palaeo-record has been documented as a significant dry spell at ∼8.2 ka during the AHP (Fig. 7). 5.3. Paleoclimatic implications In the NH monsoon region, the early Holocene African humid period has been related to a northward shift of the ITCZ, associated with intensified monsoon in response to NH summer insolation (deMenocal et al., 2000; Haug et al., 2001; Dykoski et al., 2005; Marriner et al., 2012; Marzin et al., 2013). Rainfall pattern inferred from our geochemical records also follows the NH summer insolation (Berger and Loutre, 1991), which is consistent with Asian summer monsoon (Fleitmann et al., 2003; Dykoski et al., 2005), progressively declined during the mid-late Holocene transition (Fig. 8a–d). During the early Holocene, the ITCZ reached its northernmost position, sites located in the north of 10◦ S experienced was humid, while southern sites was relative arid (Chevalier and Chase, 2015; Castañeda et al., 2016; van der Lubbe et al., 2016). However, the humid–arid transition during the mid- to late Holocene in East Africa cannot be explained only by the shift of ITCZ (Tierney et al., 2008). Alternatively, increased rainfall in East Africa has been linked to ocean–atmosphere system in the Indian and Atlantic Oceans, such as, the shift of the CAB and the fluctuation of SSTs in the Indian Ocean. Junginger et al. (2014) hypothesized that the eastwards shift in the position of the CAB during the NH summer season led to the superior rainfall during the AHP, via delivering additional moisture from the Atlantic Ocean. This seasonal migration of the CAB is due to enhanced atmospheric pressure gradient between India and East Africa, and abnormally strong westerly winds, attributing to enhanced strength of the Asian monsoon (Camberlin, 1997). We observe that the AHP I, inferred from log(Fe/Ca) in this study, is coherent with the period when the CAB migrated eastwards during the early Holocene, corresponding to the period of strong Asian summer monsoon (Fig. 8b–d). This zonal migration of the CAB has been documented by D-depleted to D-enriched waxes in Lake Tana around 8.5 ka (Costa et al., 2014). Therefore, the AHP I in East Africa would have been benefited to additional moisture conveyed from the Atlantic Ocean in the present dry summer season due to the eastward shift of the CAB, in response to the strong Asian and African monsoon. This explanation also helps us to understand the abruptness of δ Dwax data from lacustrine sediments, indicating that the CAB migration contributed to the abrupt climatic record in East Africa.

Fig. 8. Time series of East African hydrologic record combined with Asian summer monsoon and SST in the Indian Ocean. (a) Insolation is summer at 20◦ N (Berger and Loutre, 1991). (b) Oxygen isotope (δ 18 O) profiles of Holocene stalagmites from Dongge Cave, southern China (Dykoski et al., 2005). (c) Stalagmite δ 18 O from Qunf Cave, southern Oman (Fleitmann et al., 2003), b–c representing the strength of Asian summer monsoon. (d) Log(Fe/Ca) based on XRF scanner data in core GeoB12605-3, this study, indicator for precipitation in East Africa. (e) 50-year averages of the Holocene δ 18 O from Kilimanjaro ice cores, indicating local temperature or precipitation (Thompson et al., 2002). (f) SST anomalies (series mean subtracted) in the WIO from core GeoB12605-3 based on Mg/Ca of G. ruber (Kuhnert et al., 2014).

Kuhnert et al. (2014) recently estimated SSTs in the central WIO based on the shell Mg/Ca ratio of G. ruber s.s from core GeoB12605-3, indicating that a warm phase between 7.8 and 5.6 ka is followed by a cool interval during 5.6–4.2 ka (Fig. 8f). We note that the AHP II is consistent with warmer SSTs in the WIO (Fig. 8d, f), so variation in precipitation of East Africa might have been influenced by SST changes in the WIO as observed in the present and last deglacial intervals (Ummenhofer et al., 2009; Weldeab et al., 2014a). Moreover, the termination of AHP in East Africa, as indicated by the decrease in log-ratios of Fe/Ca in this study, is consist with a decrease in SSTs of short rains in the WIO after 5 ka (Zinke et al., 2014). Thus, we conclude that the humid–arid climate transition in East Africa is related to the cooling local SSTs, resulting in less moisture from the Indian Ocean during the short rains. In addition, we observe a good correlation between log(Fe/Ca) and oxygen isotope of Kilimanjaro ice cores during the mid Holocene transitional period and arid late Holocene (5.5–1.3 ka; Fig. 8d–e), which may indicate that the Kilimanjaro ice-core δ 18 O signal represents the rainfall amount rather than air temperature. However, SSTs in the Indian Ocean show various trend during the Holocene (Abram et al., 2007, 2009; Saher et al., 2007; Mohtadi et al., 2014; Romahn et al., 2014; Weldeab et al., 2014a), so the Holocene SST gradient (i.e., the mean state of the IOD) in the Indian Ocean during this period is still controversy (Griffiths et al., 2010; de Boer et al., 2014;

X. Liu et al. / Earth and Planetary Science Letters 460 (2017) 123–134

131

linked to 8.2 ka cooling event in the NH (Gasse and Van Campo, 1994), associated with the weakening of Asian summer monsoon (Fig. 9a–c). Our geochemical data, coupled with other archives, suggest that East African climatic conditions are sensitive to the NH cooling event (Fig. 9d–e). Such cooling event resulted in a serve drought in sites north of 10◦ S around 8.2 ka (Fig. 7), indicating a southward migration of the ITCZ. 6. Conclusions Our high-resolution geochemical results reveal that the AHP was represented by two phases in East Africa, an intense humid phase from the beginning of the Holocene to ∼8 ka (AHP I), and a moderate one ∼8–5.5 ka (AHP II). According the sequential t-test, the AHP II ended gradually within 5.5–3.5 ka. Comparing our XRF scanning data with other paleoclimatic records, we note that the timing and abruptness of the termination of the AHP may differ depending on the proxy applied. Wet conditions prevailed in East Africa during the early to mid-Holocene correspond to the period of maximum NH insolation. We thus propose that the AHP in East Africa (north of 10◦ S) responded to NH summer insolation during the Holocene, which influences coupled ocean–atmosphere dynamics in the Atlantic and Indian Oceans. We suggest that the AHP I was mainly linked to elevated Atlantic moisture because of an eastward shift of the CAB, while the AHP II was related to additional moisture from the Indian Ocean due to warmer SSTs in the WIO. When both mechanisms failed in the late Holocene, the AHP ended, leading to arid conditions throughout East Africa. In spite of the humid conditions during the AHP I, this study further observes a short-term drought event in East Africa around 8.2 ka. Combined with other records, opposing precipitation trends between sites north of 10◦ S (arid) and sites at 15–20◦ S (humid) may have been related to the southward migration of the ITCZ in response to the NH cooling event. Fig. 9. (a) Oxygen isotope record for the Greenland ice cap (Thomas et al., 2007). (b) Oxygen isotope record from Dongge Cave, southern China (Wang et al., 2005). (c) Oxygen isotope record from Qunf Cave, southern Oman (Fleitmann et al., 2003). (d)–(e) Log-ratios of Fe/Ca and Ti/Ca of core GeoB12605-3, this study.

Niedermeyer et al., 2014; Kwiatkowski et al., 2015). Further modeling study thus is needed to estimate the influence of SSTs in the Indian Ocean on the climatic conditions of East Africa. 5.4. The 8.2 ka event Apart from humid period during the early and mid-Holocene, a drought event is indicated by a dramatic drop in log-ratios of Fe/Ca and Ti/Ca around 8.2 ka (Fig. 9). This drought event, recognized by the sequential t-test method (Fig. 4d), has also been documented by various archives throughout East Africa (Fig. 7). These paleoclimatic records include dust event in the Kilimanjaro ice record around 8.4 ka (Thompson et al., 2002), and changes in diatom assemblages in Lakes Victoria (Stager et al., 2003), Abiyata (Chalié and Gasse, 2002), and Massoko (Barker et al., 2003). It has been recorded in lake levels, such as, short-term low lake level culminated at 8.7–8.1 ka in Lakes Ziway-Shalla and Abhé (Gasse, 2000) and in Lakes Turkana and Suguta (Garcin et al., 2009, 2012; Forman et al., 2014; Junginger et al., 2014). As well as, geochemical records indicate a decrease of moisture at ∼8.4 ka at Lake Tana (Marshall et al., 2011) and between 8 and 7.8 ka at Chew Bahir basin (Foerster et al., 2012). Likewise, decline of Ba/Ca and Fe/Ca have been reported at 8.6–8.2 ka, indicating a decrease in the Nile River runoff (Weldeab et al., 2014b; Revel et al., 2015). This short period of aridity has been described as abrupt phases of maximum aridity during the AHP in East Africa, which has been

Acknowledgements We acknowledge two anonymous reviewers for their thoughtful and constructive comments that improved the manuscript, and Associate Editor Martin Frank for his valuable comments. Financial support for this study was provided by the DFG Research Center/Cluster of Excellence “The Ocean in the Earth System”. We thank the crew and scientists of RV Meteor Cruise M75/2 for collecting the gravity cores used in this study. We acknowledge to Ms. Brit Kockisch for the TOC measurements, as well as to Dr. Jürgen Pätzold for XRF core scanner measurements. This research used data acquired at the XRF Core Scanner Lab at the MARUM – Center for Marine Environmental Sciences, University of Bremen, Germany. XTL was supported by the National Natural Science Foundation of China (Grant Nos. 41606062; 41430965), China Postdoctoral Science Foundation (Grant No. 2016M592257), and Open Fund of the Key Laboratory of Marine Geology and Environment, Chinese Academy of Sciences (Grant No. MGE2016KG05). Appendix A. Supplementary material Supplementary material related to this article can be found online at http://dx.doi.org/10.1016/j.epsl.2016.12.016. References Abram, N.J., Gagan, M.K., Liu, Z.Y., Hantoro, W.S., McCulloch, M.T., Suwargadi, B.W., 2007. Seasonal characteristics of the Indian Ocean Dipole during the Holocene epoch. Nature 445, 299–302. Abram, N.J., McGregor, H.V., Gagan, M.K., Hantoro, W.S., Suwargadi, B.W., 2009. Oscillations in the southern extent of the Indo-Pacific Warm Pool during the midHolocene. Quat. Sci. Rev. 28, 2794–2803.

132

X. Liu et al. / Earth and Planetary Science Letters 460 (2017) 123–134

Amaral, P.G.C., Vincens, A., Guiot, J., Buchet, G., Deschamps, P., Doumnang, J.C., Sylvestre, F., 2013. Palynological evidence for gradual vegetation and climate changes during the African Humid Period termination at 13◦ N from a MegaLake Chad sedimentary sequence. Clim. Past 9, 223–241. Arz, H.W., Patzold, J., Wefer, G., 1998. Correlated millennial-scale changes in surface hydrography and terrigenous sediment yield inferred from last-glacial marine deposits off northeastern Brazil. Quat. Res. 50, 157–166. Bahaga, T.K., Mengistu Tsidu, G., Kucharski, F., Diro, G.T., 2015. Potential predictability of the sea-surface temperature forced equatorial East African short rains interannual variability in the 20th century. Q. J. R. Meteorol. Soc. 141, 16–26. Bard, E., 2013. Out of the African Humid Period. Science 342, 808–809. Barker, P., Williamson, D., Gasse, F., Gibert, E., 2003. Climatic and volcanic forcing revealed in a 50,000-year diatom record from Lake Massoko, Tanzania. Quat. Res. 60, 368–376. Barker, P.A., Hurrell, E.R., Leng, M.J., Wolff, C., Cocquyt, C., Sloane, H.J., Verschuren, D., 2011. Seasonality in equatorial climate over the past 25 k.y. revealed by oxygen isotope records from Mount Kilimanjaro. Geology 39, 1111–1114. Beal, L.M., Hormann, V., Lumpkin, R., Foltz, G.R., 2013. The response of the surface circulation of the Arabian Sea to Monsoonal forcing. J. Phys. Oceanogr. 43, 2008–2022. Berger, A., Loutre, M.F., 1991. Insolation values for the climate of the last 10 million years. Quat. Sci. Rev. 10, 297–317. Berke, M.A., Johnson, T.C., Werne, J.P., Grice, K., Schouten, S., Damste, J.S.S., 2012. Molecular records of climate variability and vegetation response since the Late Pleistocene in the Lake Victoria basin, East Africa. Quat. Sci. Rev. 55, 59–74. Birch, H., Coxall, H.K., Pearson, P.N., Kroon, D., O’Regan, M., 2013. Planktonic foraminifera stable isotopes and water column structure: disentangling ecological signals. Mar. Micropaleontol. 101, 127–145. Black, E., Slingo, J., Sperber, K.R., 2003. An observational study of the relationship between excessively strong short rains in coastal East Africa and Indian Ocean SST. Mon. Weather Rev. 131, 74–94. Blanchet, C.L., Contoux, C., Leduc, G., 2015. Runoff and precipitation dynamics in the Blue and White Nile catchments during the mid-Holocene: a data–model comparison. Quat. Sci. Rev. 130, 222–230. Blanchet, C.L., Tjallingii, R., Frank, M., Lorenzen, J., Reitz, A., Brown, K., Feseker, T., Brückmann, W., 2013. 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. Brown, J., Lynch, A.H., Marshall, A.G., 2009. Variability of the Indian Ocean Dipole in coupled model paleoclimate simulations. J. Geophys. Res., Atmos. 114, D11105. Burrough, S.L., Thomas, D.S.G., 2013. Central southern Africa at the time of the African Humid Period: a new analysis of Holocene palaeoenvironmental and palaeoclimate data. Quat. Sci. Rev. 80, 29–46. Camberlin, P., 1997. Rainfall anomalies in the source region of the Nile and their connection with the Indian summer monsoon. J. Climate 10, 1380–1392. Camoin, G.F., Montaggioni, L.F., Braithwaite, C.J.R., 2004. Late glacial to post glacial sea levels in the Western Indian Ocean. Mar. Geol. 206, 119–146. Castañeda, I.S., Schouten, S., Pätzold, J., Lucassen, F., Kasemann, S., Kuhlmann, H., Schefuß, E., 2016. Hydroclimate variability in the Nile River Basin during the past 28,000 years. Earth Planet. Sci. Lett. 438, 47–56. Castañeda, I.S., Werne, J.P., Johnson, T.C., 2007. Wet and arid phases in the southeast African tropics since the Last Glacial Maximum. Geology 35, 823–826. Chalié, F., Gasse, F., 2002. Late Glacial–Holocene diatom record of water chemistry and lake level change from the tropical East African Rift Lake Abiyata (Ethiopia). Palaeogeogr. Palaeoclimatol. Palaeoecol. 187, 259–283. Chevalier, M., Chase, B.M., 2015. Southeast African records reveal a coherent shift from high- to low-latitude forcing mechanisms along the east African margin across last glacial–interglacial transition. Quat. Sci. Rev. 125, 117–130. Clark, C.O., Webster, P.J., Cole, J.E., 2003. Interdecadal variability of the relationship between the Indian Ocean zonal mode and East African coastal rainfall anomalies. J. Climate 16, 548–554. Claussen, M., Bathiany, S., Brovkin, V., Kleinen, T., 2013. Simulated climatevegetation interaction in semi-arid regions affected by plant diversity. Nat. Geosci. 6, 954–958. Cockerton, H.E., Street-Perrott, F.A., Barker, P.A., Leng, M.J., Sloane, H.J., Ficken, K.J., 2015. Orbital forcing of glacial/interglacial variations in chemical weathering and silicon cycling within the upper White Nile basin, East Africa: stable-isotope and biomarker evidence from Lakes Victoria and Edward. Quat. Sci. Rev. 130, 57–71. Costa, K., Russell, J., Konecky, B., Lamb, H., 2014. Isotopic reconstruction of the African Humid Period and Congo Air Boundary migration at Lake Tana, Ethiopia. Quat. Sci. Rev. 83, 58–67. de Boer, E.J., Tjallingii, R., Vélez, M.I., Rijsdijk, K.F., Vlug, A., Reichart, G.-J., Prendergast, A.L., de Louw, P.G., Florens, F., Baider, C., 2014. Climate variability in the SW Indian Ocean from an 8000-yr long multi-proxy record in the Mauritian lowlands shows a middle to late Holocene shift from negative IOD-state to ENSO-state. Quat. Sci. Rev. 86, 175–189. De Villiers, S., 2005. Foraminiferal shell-weight evidence for sedimentary calcite dissolution above the lysocline. Deep-Sea Res., Part I, Oceanogr. Res. Pap. 52, 671–680.

deMenocal, P., Ortiz, J., Guilderson, T., Adkins, J., Sarnthein, M., Baker, L., Yarusinsky, M., 2000. Abrupt onset and termination of the African Humid Period: rapid climate responses to gradual insolation forcing. Quat. Sci. Rev. 19, 347–361. deMenocal, P.B., 2014. Marine sediment records of African climate change: progress and puzzles. In: Holland, H.D., Turekian, K.K. (Eds.), Treatise on Geochemistry, 2nd ed. Elsevier, Oxford, pp. 99–108. Dykoski, C.A., Edwards, R.L., Cheng, H., Yuan, D.X., Cai, Y.J., Zhang, M.L., Lin, Y.S., Qing, J.M., An, Z.S., Revenaugh, J., 2005. A high-resolution, absolute-dated Holocene and deglacial Asian monsoon record from Dongge Cave, China. Earth Planet. Sci. Lett. 233, 71–86. Ehrmann, W., Seidel, M., Schmiedl, G., 2013. Dynamics of Late Quaternary North African humid periods documented in the clay mineral record of central Aegean Sea sediments. Glob. Planet. Change 107, 186–195. Flaux, C., Claude, C., Marriner, N., Morhange, C., 2013. A 7500-year strontium isotope record from the northwestern Nile delta (Maryut lagoon, Egypt). Quat. Sci. Rev. 78, 22–33. Fleitmann, D., Burns, S.J., Mudelsee, M., Neff, U., Kramers, J., Mangini, A., Matter, A., 2003. Holocene forcing of the Indian monsoon recorded in a stalagmite from southern Oman. Science 300, 1737–1739. Foerster, V., Junginger, A., Langkamp, O., Gebru, T., Asrat, A., Umer, M., Lamb, H.F., Wennrich, V., Rethemeyer, J., Nowaczyk, N., Trauth, M.H., Schaebitz, F., 2012. Climatic change recorded in the sediments of the Chew Bahir basin, southern Ethiopia, during the last 45,000 years. Quat. Int. 274, 25–37. Forman, S.L., Wright, D.K., Bloszies, C., 2014. Variations in water level for Lake Turkana in the past 8500 years near Mt. Porr, Kenya and the transition from the African Humid Period to Holocene aridity. Quat. Sci. Rev. 97, 84–101. Francus, P., Von Suchodoletz, H., Dietze, M., Donner, R.V., Bouchard, F., Roy, A.J., Fagot, M., Verschuren, D., Kropelin, S., 2013. Varved sediments of Lake Yoa (Ounianga Kebir, Chad) reveal progressive drying of the Sahara during the last 6100 years. Sedimentology 60, 911–934. Garcin, Y., Junginger, A., Melnick, D., Olago, D.O., Strecker, M.R., Trauth, M.H., 2009. Late Pleistocene–Holocene rise and collapse of Lake Suguta, northern Kenya Rift. Quat. Sci. Rev. 28, 911–925. Garcin, Y., Melnick, D., Strecker, M.R., Olago, D., Tiercelin, J.-J., 2012. East African mid-Holocene wet–dry transition recorded in palaeo-shorelines of Lake Turkana, northern Kenya Rift. Earth Planet. Sci. Lett. 331–332, 322–334. Gasse, F., 2000. Hydrological changes in the African tropics since the Last Glacial Maximum. Quat. Sci. Rev. 19, 189–211. Gasse, F., Van Campo, E., 1994. Abrupt post-glacial climate events in West Asia and North Africa monsoon domains. Earth Planet. Sci. Lett. 126, 435–456. Gillespie, R., Streetperrott, F.A., Switsur, R., 1983. Post-glacial arid episodes in Ethiopia have implications for climate prediction. Nature 306, 680–683. Govin, A., Holzwarth, U., Heslop, D., Keeling, L.F., Zabel, M., Mulitza, S., Collins, J.A., Chiessi, C.M., 2012. Distribution of major elements in Atlantic surface sediments (36◦ N–49◦ S): imprint of terrigenous input and continental weathering. Geochem. Geophys. Geosyst. 13, Q01013. Griffiths, M.L., Drysdale, R.N., Gagan, M.K., Frisia, S., Zhao, J.-x., Ayliffe, L.K., Hantoro, W.S., Hellstrom, J.C., Fischer, M.J., Feng, Y.-X., 2010. Evidence for Holocene changes in Australian–Indonesian monsoon rainfall from stalagmite trace element and stable isotope ratios. Earth Planet. Sci. Lett. 292, 27–38. Grumet, N.S., Guilderson, T.P., Dunbar, R.B., 2002. Meridional transport in the Indian Ocean traced by coral radiocarbon. J. Mar. Res. 60, 725–742. Hamann, Y., Ehrmann, W., Schmiedl, G., Krüger, S., Stuut, J.B., Kuhnt, T., 2008. Sedimentation processes in the Eastern Mediterranean Sea during the Late Glacial and Holocene revealed by end-member modelling of the terrigenous fraction in marine sediments. Mar. Geol. 248, 97–114. Hastenrath, S., 2007. Circulation mechanisms of climate anomalies in East Africa and the equatorial Indian Ocean. Dyn. Atmos. Ocean. 43, 25–35. Haug, G.H., Hughen, K.A., Sigman, D.M., Peterson, L.C., Rohl, U., 2001. Southward migration of the Intertropical Convergence Zone through the Holocene. Science 293, 1304–1308. Hély, C., Braconnot, P., Watrin, J., Zheng, W., 2009. Climate and vegetation: simulating the African humid period. C. R. Géosci. 341, 671–688. Hennekam, R., de Lange, G., 2012. X-ray fluorescence core scanning of wet marine sediments: methods to improve quality and reproducibility of high-resolution paleoenvironmental records. Limnol. Oceanogr., Methods 10, 991–1003. Jung, S.J.A., Davies, G.R., Ganssen, G.M., Kroon, D., 2004. Stepwise Holocene aridification in NE Africa deduced from dust-borne radiogenic isotope records. Earth Planet. Sci. Lett. 221, 27–37. Junginger, A., Roller, S., Olaka, L.A., Trauth, M.H., 2014. The effects of solar irradiation changes on the migration of the Congo Air Boundary and water levels of paleoLake Suguta, Northern Kenya Rift, during the African Humid Period (15–5 ka BP). Palaeogeogr. Palaeoclimatol. Palaeoecol. 396, 1–16. Junginger, A., Trauth, M.H., 2013. Hydrological constraints of paleo-Lake Suguta in the Northern Kenya Rift during the African Humid Period (15–5 ka BP). Glob. Planet. Change 111, 174–188. Kijazi, A.L., Reason, C.J.C., 2012. Intra-seasonal variability over the northeastern highlands of Tanzania. Int. J. Climatol. 32, 874–887. Kohn, M., Zonneveld, K.A.F., 2010. Calcification depth and spatial distribution of Thoracosphaera heimii cysts: implications for palaeoceanographic reconstructions. Deep-Sea Res, Part 1, Oceanogr Res. Pap. 57, 1543–1560.

X. Liu et al. / Earth and Planetary Science Letters 460 (2017) 123–134

Kröpelin, S., Verschuren, D., Lézine, A.-M., Eggermont, H., Cocquyt, C., Francus, P., Cazet, J.-P., Fagot, M., Rumes, B., Russell, J.M., Darius, F., Conley, D.J., Schuster, M., von Suchodoletz, H., Engstrom, D.R., 2008. Climate-driven ecosystem succession in the Sahara: the past 6000 years. Science 320, 765–768. Kuhlmann, H., Meggers, H., Freudenthal, T., Wefer, G., 2004. The transition of the monsoonal and the N Atlantic climate system off NW Africa during the Holocene. Geophys. Res. Lett. 31, L22204. Kuhnert, H., Kuhlmann, H., Mohtadi, M., Meggers, H., Baumann, K.H., Patzold, J., 2014. Holocene tropical western Indian Ocean sea surface temperatures in covariation with climatic changes in the Indonesian region. Paleoceanography 29, 423–437. Kwiatkowski, C., Prange, M., Varma, V., Steinke, S., Hebbeln, D., Mohtadi, M., 2015. Holocene variations of thermocline conditions in the eastern tropical Indian Ocean. Quat. Sci. Rev. 114, 33–42. Leduc, G., Sachs, J.P., Kawka, O.E., Schneider, R.R., 2013. Holocene changes in eastern equatorial Atlantic salinity as estimated by water isotopologues. Earth Planet. Sci. Lett. 362, 151–162. Liu, X.T., 2014. Sediment dynamics off the East African continental margin during the last deglaciation and the Holocene: constrained by changes in climate and sea level. Universität Bremen, Bremen, Germany. Liu, X.T., Rendle-Bühring, R., Meyer, I., Henrich, R., 2016. Holocene shelf sedimentation patterns off equatorial East Africa constrained by climatic and sea-level changes. Sediment. Geol. 331, 1–11. Maitima, J.M., 1991. Vegetation response to climatic change in central Rift Valley, Kenya. Quat. Res. 35, 234–245. Manyilizu, M., Penven, P., Reason, C.J.C., 2016. Annual cycle of the upper-ocean circulation and properties in the tropical western Indian Ocean. Afr. J. Mar. Sci. 38, 81–99. Marriner, N., Flaux, C., Kaniewski, D., Morhange, C., Leduc, G., Moron, V., Chen, Z., Gasse, F., Empereur, J.Y., Stanley, J.D., 2012. ITCZ and ENSO-like pacing of Nile delta hydro-geomorphology during the Holocene. Quat. Sci. Rev. 45, 73–84. Marshall, M.H., Lamb, H.F., Huws, D., Davies, S.J., Bates, R., Bloemendal, J., Boyle, J., Leng, M.J., Umer, M., Bryant, C., 2011. Late Pleistocene and Holocene drought events at Lake Tana, the source of the Blue Nile. Glob. Planet. Change 78, 147–161. Marzin, C., Braconnot, P., Kageyama, M., 2013. Relative impacts of insolation changes, meltwater fluxes and ice sheets on African and Asian monsoons during the Holocene. Clim. Dyn. 41, 2267–2286. McClanahan, T.R., 1988. Seasonality in East Africa’s coastal waters. Mar. Ecol. Prog. Ser. 44, 191–199. McGee, D., deMenocal, P.B., Winckler, G., Stuut, J.B.W., Bradtmiller, L.I., 2013. The magnitude, timing and abruptness of changes in North African dust deposition over the last 20,000 yr. Earth Planet. Sci. Lett. 371, 163–176. Mohtadi, M., Prange, M., Oppo, D.W., De Pol-Holz, R., Merkel, U., Zhang, X., Steinke, S., Luckge, A., 2014. North Atlantic forcing of tropical Indian Ocean climate. Nature 509, 76–80. Morrissey, A., Scholz, C.A., 2014. Paleohydrology of Lake Turkana and its influence on the Nile River system. Palaeogeogr. Palaeoclimatol. Palaeoecol. 403, 88–100. Nicholson, S.E., 2000. The nature of rainfall variability over Africa on time scales of decades to millenia. Glob. Planet. Change 26, 137–158. Nicholson, S.E., Kim, E., 1997. The relationship of the El Nino Southern oscillation to African rainfall. Int. J. Climatol. 17, 117–135. Niedermeyer, E.M., Sessions, A.L., Feakins, S.J., Mohtadi, M., 2014. Hydroclimate of the western Indo-Pacific Warm Pool during the past 24,000 years. Proc. Natl. Acad. Sci. USA 111, 9402–9406. Paytan, A., Griffith, E.M., 2007. Marine barite: recorder of variations in ocean export productivity. Deep-Sea Res., Part 2, Top. Stud. Oceanogr. 54, 687–705. Peterson, L.C., Haug, G.H., Hughen, K.A., Rohl, U., 2000. Rapid changes in the hydrologic cycle of the tropical Atlantic during the last glacial. Science 290, 1947–1951. Punwong, P., Marchant, R., Selby, K., 2013. Holocene mangrove dynamics and environmental change in the Rufiji Delta, Tanzania. Veg. Hist. Archaeobot. 22, 381–396. Revel, M., Colin, C., Bernasconi, S., Combourieu-Nebout, N., Ducassou, E., Grousset, F.E., Rolland, Y., Migeon, S., Bosch, D., Brunet, P., Zhao, Y.L., Mascle, J., 2014. 21,000 years of Ethiopian African monsoon variability recorded in sediments of the western Nile deep-sea fan. Reg. Environ. Change 14, 1685–1696. Revel, M., Ducassou, E., Grousset, F.E., Bernasconi, S.M., Migeon, S., Revillon, S., Mascle, J., Murat, A., Zaragosi, S., Bosch, D., 2010. 100,000 years of African monsoon variability recorded in sediments of the Nile margin. Quat. Sci. Rev. 29, 1342–1362. Revel, M., Ducassou, E., Skonieczny, C., Colin, C., Bastian, L., Bosch, D., Migeon, S., Mascle, J., 2015. 20,000 years of Nile River dynamics and environmental changes in the Nile catchment area as inferred from Nile upper continental slope sediments. Quat. Sci. Rev. 130, 200–221. Rodionov, S.N., 2004. A sequential algorithm for testing climate regime shifts. Geophys. Res. Lett. 31. Röhl, U., Westerhold, T., Bralower, T.J., Zachos, J.C., 2007. On the duration of the Paleocene–Eocene thermal maximum (PETM). Geochem. Geophys. Geosyst. 8.

133

Romahn, S., Mackensen, A., Groeneveld, J., Pätzold, J., 2014. Deglacial intermediate water reorganization: new evidence from the Indian Ocean. Clim. Past. 10, 293–303. Romahn, S., Mackensen, A., Kuhlmann, H., Patzold, J., 2015. Benthic foraminiferal response to Late Glacial and Holocene sea level rise and rainfall variability off East Africa. Mar. Micropaleontol. 119, 34–48. Russell, J.M., Johnson, T.C., Kelts, K.R., Laerdal, T., Talbot, M.R., 2003. An 11,000-year lithostratigraphic and paleohydrologic record from Equatorial Africa: Lake Edward, Uganda–Congo. Palaeogeogr. Palaeoclimatol. Palaeoecol. 193, 25–49. Saher, M.H., Jung, S.J.A., Elderfield, H., Greaves, M.J., Kroon, D., 2007. Sea surface temperatures of the western Arabian Sea during the last deglaciation. Paleoceanography 22, PA2208. Saji, N.H., Goswami, B.N., Vinayachandran, P.N., Yamagata, T., 1999. A dipole mode in the tropical Indian Ocean. Nature 401, 360–363. Savoye, B., Ridderinkhof, H., Pätzold, J., Schneider, R., 2013. Western Indian Ocean Climate and Sedimentation - Cruise No. M75 - December 29, 2007 April 08, 2008 - Port Louis (Mauritius) – Cape Town (South Africa). DFGSenatskommission für Ozeanographie, Bremen. Schefuß, E., Kuhlmann, H., Mollenhauer, G., Prange, M., Pätzold, J., 2011. Forcing of wet phases in southeast Africa over the past 17,000 years. Nature 480, 509–512. Schefuß, E., Schouten, S., Schneider, R.R., 2005. Climatic controls on central African hydrology during the past 20,000 years. Nature 437, 1003–1006. Schluter, T., 2008. Geological atlas of Africa: with Notes on Stratigraphy, Tectonics, Economic Geology, Geohazards, Geosites and Geoscientific Education of Each Country, 2nd ed. Springer-Verlag, Berlin, Heidelberg. Schulte, S., Bard, E., 2003. Past changes in biologically mediated dissolution of calcite above the chemical lysocline recorded in Indian Ocean sediments. Quat. Sci. Rev. 22, 1757–1770. Shaghude, Y.W., 2007. Shore morphology and sediment characteristics south of Pangani River, coastal Tanzania. West. Indian Ocean J. Mar. Sci. 3, 93–104. Shanahan, T.M., McKay, N.P., Hughen, K.A., Overpeck, J.T., Otto-Bliesner, B., Heil, C.W., King, J., Scholz, C.A., Peck, J., 2015. The time-transgressive termination of the African Humid Period. Nat. Geosci. 8, 140–144. Southon, J., Kashgarian, M., Fontugne, M., Metivier, B., Yim, W.W.S., 2002. Marine reservoir corrections for the Indian Ocean and southeast Asia. Radiocarbon 44, 167–180. Stager, J.C., Cumming, B.F., Meeker, L.D., 2003. A 10,000-year high-resolution diatom record from Pilkington Bay, Lake Victoria, East Africa. Quat. Res. 59, 172–181. Stolz, K., Baumann, K., Mersmeyer, H., 2015. Extant coccolithophores from the western equatorial Indian Ocean off Tanzania and coccolith distribution in surface sediments. Micropaleontology 61, 473–488. Stuut, J.-B.W., Temmesfeld, F., De Deckker, P., 2014. A 550 ka record of aeolian activity near North West Cape, Australia: inferences from grain-size distributions and bulk chemistry of SE Indian Ocean deep-sea sediments. Quat. Sci. Rev. 83, 83–94. Thomas, E.R., Wolff, E.W., Mulvaney, R., Steffensen, J.P., Johnsen, S.J., Arrowsmith, C., White, J.W., Vaughn, B., Popp, T., 2007. The 8.2 ka event from Greenland ice cores. Quat. Sci. Rev. 26, 70–81. Thompson, L.G., Mosley-Thompson, E., Davis, M.E., Henderson, K.A., Brecher, H.H., Zagorodnov, V.S., Mashiotta, T.A., Lin, P.N., Mikhalenko, V.N., Hardy, D.R., Beer, J., 2002. Kilimanjaro ice core records: evidence of Holocene climate change in tropical Africa. Science 298, 589–593. Tierney, J.E., deMenocal, P.B., 2013. Abrupt shifts in Horn of Africa hydroclimate since the Last Glacial Maximum. Science 342, 843–846. Tierney, J.E., Lewis, S.C., Cook, B.I., LeGrande, A.N., Schmidt, G.A., 2011. Model, proxy and isotopic perspectives on the East African Humid Period. Earth Planet. Sci. Lett. 307, 103–112. Tierney, J.E., Russell, J.M., Huang, Y., Damste, J.S., Hopmans, E.C., Cohen, A.S., 2008. Northern hemisphere controls on tropical southeast African climate during the past 60,000 years. Science 322, 252–255. Tjallingii, R., Rohl, U., Kolling, M., Bickert, T., 2007. Influence of the water content on X-ray fluorescence core-scanning measurements in soft marine sediments. Geochem. Geophys. Geosyst. 8, Q02004. Ummenhofer, C.C., Sen Gupta, A., England, M.H., Reason, C.J.C., 2009. Contributions of Indian Ocean Sea surface temperatures to enhanced East African rainfall. J. Climate 22, 993–1013. van der Lubbe, H.J.L., Frank, M., Tjallingii, R., Schneider, R.R., 2016. Neodymium isotope constraints on provenance, dispersal, and climate-driven supply of Zambezi sediments along the Mozambique Margin during the past ∼45,000 years. Geochem. Geophys. Geosyst. 17, 181–198. Wang, Y., Cheng, H., Edwards, R.L., He, Y., Kong, X., An, Z., Wu, J., Kelly, M.J., Dykoski, C.A., Li, X., 2005. The Holocene Asian Monsoon: links to solar changes and North Atlantic climate. Science 308, 854–857. Wang, Y.V., Larsen, T., Leduc, G., Andersen, N., Blanz, T., Schneider, R.R., 2013. What does leaf wax δ D from a mixed C3 /C4 vegetation region tell us? Geochim. Cosmochim. Acta 111, 128–139. Weldeab, S., Lea, D.W., Oberhänsli, H., Schneider, R.R., 2014a. Links between southwestern tropical Indian Ocean SST and precipitation over southeastern Africa over the last 17 kyr. Palaeogeogr. Palaeoclimatol. Palaeoecol. 410, 200–212.

134

X. Liu et al. / Earth and Planetary Science Letters 460 (2017) 123–134

Weldeab, S., Lea, D.W., Schneider, R.R., Andersen, N., 2007. 155,000 years of West African monsoon and ocean thermal evolution. Science 316, 1303–1307. Weldeab, S., Menke, V., Schmiedl, G., 2014b. The pace of East African monsoon evolution during the Holocene. Geophys. Res. Lett. 41, 1724–1731. Weltje, G.J., Tjallingii, R., 2008. Calibration of XRF core scanners for quantitative geochemical logging of sediment cores: theory and application. Earth Planet. Sci. Lett. 274, 423–438. Westerhold, T., Röhl, U., Laskar, J., Raffi, I., Bowles, J., Lourens, L.J., Zachos, J.C., 2007. On the duration of magnetochrons C24r and C25n and the timing of early Eocene global warming events: implications from the Ocean Drilling Program Leg 208 Walvis Ridge depth transect. Paleoceanography 22, PA2201. Willis, K.J., Bennett, K.D., Burrough, S.L., Macias-Fauria, M., Tovar, C., 2013. Determining the response of African biota to climate change: using the past to model the future. Philos. Trans. R. Soc. Lond. B, Biol. Sci. 368, 20120491.

Woodroffe, S.A., Long, A.J., Punwong, P., Selby, K., Bryant, C.L., Marchant, R., 2015. Radiocarbon dating of mangrove sediments to constrain Holocene relative sealevel change on Zanzibar in the southwest Indian Ocean. Holocene 25, 820–831. Ziegler, M., Simon, M.H., Hall, I.R., Barker, S., Stringer, C., Zahn, R., 2013. Development of Middle Stone Age innovation linked to rapid climate change. Nat. Commun. 4, 1905. Zinke, J., Pfeiffer, M., Park, W., Schneider, B., Reuning, L., Dullo, W.-C., Camoin, G., Mangini, A., Schroeder-Ritzrau, A., Garbe-Schönberg, D., 2014. Seychelles coral record of changes in sea surface temperature bimodality in the western Indian Ocean from the Mid-Holocene to the present. Clim. Dyn. 43, 689–708. Zinke, J., Reijmer, J.J.G., Thomassin, B.A., Dullo, W.C., Grootes, P.M., Erlenkeuser, H., 2003. Postglacial flooding history of Mayotte lagoon (Comoro Archipelago, southwest Indian Ocean). Mar. Geol. 194, 181–196.