High-and low-latitude forcing of the East African climate since the LGM: Inferred from the elemental composition of marine sediments off Tanzania

Quaternary Science Reviews 196 (2018) 124e136

Contents lists available at ScienceDirect

Quaternary Science Reviews journal homepage: www.elsevier.com/locate/quascirev

High-and low-latitude forcing of the East African climate since the LGM: Inferred from the elemental composition of marine sediments off Tanzania Xiting Liu a, b, *, Rebecca Rendle-Bühring c, Rüdiger Henrich c a b c

Key Laboratory of Submarine Geosciences and Prospecting Technology, College of Marine Geosciences, Ocean University of China, Qingdao, 266100, China Laboratory for Marine Geology, Qingdao National Laboratory for Marine Science and Technology, Qingdao, 266061, China MARUM e 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

a b s t r a c t

Article history: Received 5 March 2018 Received in revised form 2 August 2018 Accepted 2 August 2018

We present X-Ray Fluorescence (XRF) Scanner measurements from a 6 m long sediment core (GeoB12624-1) on the upper slope of Tanzania to reconstruct the climatic evolution in East Africa since the Last Glacial Maximum (LGM). Log-ratios of Fe/Ca and Ti/Ca are indicative for sediment discharge of the Rufiji River, which is controlled by climatic conditions in the Rufiji catchment area. The most significant changes in major elemental composition occurred at 15.1 and 7.4 ka highlighted by the regime shift index values. The data set records distinct precipitation peaks during the early Holocene. This corresponds a maximum in the Northern Hemisphere (NH) summer insolation and results in a transition from the arid LGM to the humid early Holocene. Our geochemical record also indicates that this climatic transition was interrupted by two severe droughts that occurred during NH cold intervals: the Heinrich stadial 1 (HS1) and the Younger Dryas (YD). Through a comparison with other nearby paleoclimatic records, we suggest that arid climatic conditions only occurred in East Africa north of 8e10
S, whereas in southern East Africa around 15e20
S increased humidity during the HS1 and YD prevailed. We thus conclude that these two drought events were caused by a southward migration of the Intertropical Convergence Zone (ITCZ) which was fostered by the NH cooling during the HS1 and YD. Hence, our new geochemical record clearly documents that the East African climate not only responded to low-latitude insolation forcing on sub-orbital time scales, but also, was strongly influenced by high-latitude cooling during the HS1 and YD periods. © 2018 Elsevier Ltd. All rights reserved.

Keywords: Paleoclimatology Younger Dryas Heinrich stadial 1 African Humid Period X-ray fluorescence scanner

1. Introduction Since the Last Glacial Maximum (LGM), climatic conditions in tropical Africa have been widely related to precessional summer low-latitude insolation (Kutzbach and Street-Perrott, 1985; Gasse et al., 2008; Shanahan et al., 2015); or seasonal contrasts between Northern Hemisphere (NH) and Southern Hemisphere (SH) insolation (Verschuren et al., 2009; Berke et al., 2012; Collins et al., 2017). The humid conditions in North Africa during the early and mid-Holocene, termed by deMenocal et al. (2000) as the African Humid Period (AHP), are generally considered to have been

* Corresponding author. Key Laboratory of Submarine Geosciences and Prospecting Technology, College of Marine Geosciences, Ocean University of China, Qingdao, 266100, China. E-mail address: [email protected] (X. Liu). https://doi.org/10.1016/j.quascirev.2018.08.004 0277-3791/© 2018 Elsevier Ltd. All rights reserved.

triggered by an increase in low-latitude NH summer insolation which strengthened the African monsoon. In East Africa, at the same time, humid conditions are suggested to have been caused by increased rainfall occurred during the NH summer season, which reduced rainfall variety between different seasons (Barker et al., 2011; Tierney et al., 2011a). However, when the NH was relatively cold, e.g., the Heinrich stadial 1 (HS1; ~18e14.6 ka) and the Younger Dryas (YD; ~12.8e11.5 ka; Barker et al., 2009), forcing from highlatitudes might have controlled the humid and arid phases in tropical Africa, which modulates the seasonal movement of the Intertropical Convergence Zone (ITCZ) (Tierney and deMenocal, ~ eda et al., 2016; Bastian 2013; Otto-Bliesner et al., 2014; Castan et al., 2017). According to marine records, the ITCZ shifted southward during these cold periods, which caused humid conditions in southern East Africa (Schefuß et al., 2011; Mohtadi et al., 2014; van der Lubbe et al., 2014), and drought throughout equatorial and

X. Liu et al. / Quaternary Science Reviews 196 (2018) 124e136

~ eda northern East Africa (Tierney and deMenocal, 2013; Castan et al., 2016). Such southward migration of the ITCZ have been related to slowdown of the Atlantic Meridional Overturning Circulation (AMOC) during HS1 and the YD stadial (Ritz et al., 2013; Mohtadi et al., 2016). These changes are also manifested in various lake records (Johnson et al., 2002; Brown et al., 2007; Garcin et al., 2007, 2009; Chevalier and Chase, 2015). High lake levels during the HS1 from Lakes Malawi and Chilwa (Filippi and Talbot, 2005; Thomas et al., 2009) indicate that during the HS1 the ITCZ moved southward and caused wet conditions in southern East Africa. Increased northerly winds over the Lake Malawi lake during the YD (Johnson ~ eda et al., 2007, 2009; Talbot et al., 2007); et al., 2002; Castan however, Lake Masoko, located close to Lake Malawi, experienced wet conditions at the same time (Garcin et al., 2006a, 2007). Such spatial and temporal heterogeneity in lake records has limited the comprehensive understanding of the East African evolving climate since the LGM. In spite of controlling factors such as the migration of the ITCZ, other factors including distance to the sea, local wind regimes, and land topography could also be involved, creating different lake climate records. In contrast, marine sediments deposited off the African continent have provided complete and well-dated paleoclimatic records (deMenocal, 2014). High-resolution of geochemical scanner profiles have improved time resolution of sampling (Bard, 2013), which allowed us to obtain continuous sediment sequences with high temporal resolution marine records. To date, unlike the well study on the western north tropical Africa (Shanahan et al., 2015), only few studies on marine sediments off East Africa, demonstrating inland climatic conditions since the LGM, have been carried out. These include sedimentological and geochemical archives representing Nile hydrology from the eastern Mediterranean Sea (Box et al., 2011; Revel ~ eda et al., 2016), biomarker based precipitation et al., 2014; Castan variability in marine sediments off northern and southern East Africa (Schefuß et al., 2011; Tierney and deMenocal, 2013), and geochemical (Foraminiferal Ba/Ca) and sedimentological (grain size) proxy data off the Zambezi River mouth, in the Mozambique Channel (Just et al., 2014; van der Lubbe et al., 2014; Weldeab et al., 2014a). Our key objective is therefore to present a continuous, highresolution climatic record from a marine sediment gravity core off the Rufiji River mouth, Tanzania, to establish the timing and pace of climatic change in southern East Africa and its driving force since the LGM. Through comparisons with other marine and terrestrial paleoclimatic records, and the latest SST data from the tropical Indian Ocean, we will illustrate what the predominant forcing mechanisms on the regional climatic change are in different intervals such as the HS1, the YD, and the AHP. These new data will help model and foresee extreme variations in the water cycle with either episodes of drought or periods of extreme flooding in Africa in the future (Defrance et al., 2017).

125

Webster et al., 1999; Bahaga et al., 2015). The present climatic system of tropical East Africa is complicated because moisture transfer from the Atlantic, as well as, the Indian Ocean plays a role (Tierney et al., 2011a). The convergence of these different moisture sources marks the Congo Air Boundary (CAB). Previous studies ~ o-Southern Oscillation (ENSO) also impact suggest that the El Nin the East African rainfall, which could cause abundant rainfall during the short-rains season during warm ENSO events (Nicholson and Kim, 1997). The WIO is mainly influenced by the East African Coastal Current (EACC), a branch of the South Equatorial Current (SEC; Fig. 1b). The SEC splits into the north-flowing EACC and the south-flowing Mozambique Current (MC) at the coast of East Africa near 11
S after passing the northern part of Madagascar (Beal et al., 2013). During the southeast (SE) monsoon, the EACC can has a higher velocities of 2 m/s; which reduce to lower than 0.2 m/s during the northeast (NE) monsoon (Newell, 1957). Combined with the southflowing Somali Current (SC), the EACC leaves eastward off the mainland around ~3
S during the NE monsoon, which forms the east-flowing Equatorial Countercurrent (ECC, Fig. 1b) (Kohn and Zonneveld, 2010). This seasonal reversal in the wind patterns causes upwelling (Fig. 1b), which occurs off the Kenyan and Somalian coasts (McClanahan, 1988; Birch et al., 2013). In contrast, to south of 4
S, surface waters (such as our core location) are represented by low nutrient contents and low surface and benthic productivity (Birch et al., 2013). 2.2. The Rufiji River The Rufiji River, the largest river in Tanzania draining the rift mountains, drains a basin of 177 000 km2 (Shaghude, 2005). Most of the plateau and mountain sections of the basin are covered by gneissic and schistose metamorphic rocks, whereas southeastern section of the basin are underlain by arenaceous stratified rocks mainly of Karroo age (Temple and Sundborg, 1972, Fig. 2a). Because of the narrow continental shelf and channel system off Tanzania (Fig. 2bec), the Rufiji River is the main terrigenous sediment source to our research location (Liu et al., 2016a), with a sediment yield of ~95 t/km2/yr (Milliman and Syvitski, 1992). Due to deposition of sediment discharged by the Rufiji River, the Rufiji delta protrudes 15 km into the Mafia channel, covering an area of ~1200 km2 (Shaghude, 2005). The Rufiji delta, composed of by fluvial sand, silt and clay (Shaghude, 2005; Punwong et al., 2013), contains the largest estuarine mangrove forest in East Africa, with a total area of ~500 km2 (Duvail and Hamerlynck, 2007). Together with other rivers (e.g., the Zambezi and Limpopo Rivers; Fig. 1a), in this region, large amounts of terrigenous sediments are discharged into the WIO (Shaghude, 2007), which would impede the growth of carbonate-producing biota (e.g., corals; Arthurton, 2003). Mangroves beds, however could filter freshwater derived from theses inland rivers and promote the growth of coral reefs offshore (Shaghude, 2005; Romahn et al., 2015).

2. Regional setting 3. Materials and methods 2.1. Climatic and oceanographic setting 3.1. Marine sediment cores East Africa receives seasonal rainfall as a result of the passage of the ITCZ twice a year (Fig. 1a): a short one in the NH autumn (short rains; September to November) and a longer one in the NH spring (long rains; March to May; Nicholson, 2000). In addition, rainfall during the short rains is also modulated by SSTs of the Western Indian Ocean (WIO), which brings heavier rainfall when the SST of the seawater adjacent to the East African mainland is warmer (Ummenhofer et al., 2009) or higher SST gradients occur between the WIO and eastern tropical Indian Ocean (Saji et al., 1999;

Marine sediment core GeoB12624-1 (8
14.050 S; 39
45.160 E) was retrieved from the upper continental slope off Tanzania (Fig. 2bec) at the water depth of 655 m, during a R/V METEOR cruise (M75/2) in February 2008 (Savoye et al., 2013). The 600 cm long core sediments are composed by dark olive-gray fine-grained muddy deposits. They derive from the discharged sediments of the Rufiji River and flow into the WIO at the southwest of the Mafia Island (Liu et al., 2016a). The age model of core GeoB12624-1 was based on

126

X. Liu et al. / Quaternary Science Reviews 196 (2018) 124e136

Fig. 1. Climatic and oceanographic systems of East Africa. (a) Seasonal movement of the ITCZ and the CAB between June and August (JJA; solid lines) and between December and February (DJF; dashed lines), based on Romahn et al. (2015); Mean monthly rainfall of lower Rufiji catchment according to Shaghude (2005). (b) Current system during the southeast and northeast monsoons, represented by the solid and dashed lines, respectively, modified from Kohn and Zonneveld (2010). Core site of GeoB12624-1is indicated by the red dot, while other 2 cores discussed in the text are also illustrated by blue solid dots: GeoB12615-4 (Romahn et al., 2014) and GeoB9307-3 (Schefuß et al., 2011). GeoB12623-25 represents core GeoB12623-1, core GeoB12624-3 and core GeoB12625-1. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

7 accelerator mass spectrometry (AMS) radiocarbon age values (Table 1), which were converted to calendar ages with CALIB 6.11 software, using a constant reservoir correction of 140 ± 25 yr (Bouimetarhan et al., 2015). In addition, 3 surface samples (GeoB12623-1, GeoB12624-3 and GeoB12625-1; Fig. 1a), near the core location of core GeoB12624-1, were obtained by multicorer during the expedition. 3.2. X-ray diffraction (XRD) analysis XRD analyses were carried out on the 3 bulk samples from cores GeoB12623-25 in the laboratories of Crystallography and Applied Material Sciences (Faculty of Geosciences, University of Bremen). Freeze-dried bulk sediments were ground to a fine powder (<20 mm) and prepared with the Philips back-loading system. The XRD measurements were performed with a Philips X'Pert Pro multipurpose diffractometer equipped with a Cu-tube (ka 1.541, 45 kV, 40 mA), a fixed divergence slit of ¼
, a 16 sample changer, a secondary Ni filter and the X'Celerator detector system. The measurements were made with a continuous scan from 3
to 85
2q, with a calculated step size of 0.016
2q and a calculated time per step of 100 s. Semi-quantitative interpretations of the mineral content were carried out with a Quantitative Phase-Analysis with X-ray Powder Diffraction (QUAX). 3.3. X-ray fluorescence (XRF) core scanning XRF core scanning was carried out directly on the split core surface of the archive half at MARUM (University of Bremen) using the XRF Core Scanner II (AVAATECH Serial No. 2). XRF data were collected with a resolution of 2 cm, with a sampling time of 30 s, using generator settings of 10 kV and a current of 0.35 mA. In order to prevent the XRF scanning unit from contamination, and to avoid sediment desiccation, a 4 mm thin SPEXCerti Prep Ultralene1 foil was used to cover the sediment surface before measurement. The data was acquired and processed following the outlines described

€hl et al., 2007; Tjallingii et al., 2007; in previous studies (Ro Westerhold et al., 2007). The principal component analysis (PCA) has performed on the measured XRF data (Al, Si, K, Ca, Ti, Fe, and Ba) using the XLSTAT software package. XRF data were presented by log-ratios of elemental intensities (i.e., Al/Si, Fe/K, Fe/Ca, and Ti/Ca), which have been proven to represent the relative concentrations of respective elements, via minimizing the effects of sample geometry and physical properties of the core sediments (Weltje and Tjallingii, 2008). A regime shift index (RSI) algorithm (Rodionov, 2004) was then applied (P > 0.1, cut-off ¼ 100) to identify regime shifts in geochemical composition. Regime shift was determined by a quick shift in the data from one interval characterized by a specific mean and variance to an interval with a statistically significant difference in mean and variance.

4. Results 4.1. Core sediments and sedimentation rates According to our previous studies Liu et al. (2016a), the sediments of core GeoB12624-1 are mainly composed of the silt fraction (55%), and secondly the sand fraction (30%; quartz, feldspar, and mica), with the clay fraction representing the minor grain-size component (15%; Fig. 3). The sediments reveal a coarsening upward trend indicated by a continuous increase in the sand fraction from the core bottom to top. In addition, a total of 12 turbidite layers were observed (Fig. 3). Sedimentation rates (SR; cm/ka) were calculated according to the age model attained by AMS 14C dating (excluding the thicknesss of turbidite beds), with a highest SR (average of 68.4 cm/ka) at the central part of the core (Fig. 3). The lowermost (19.3e15.7 ka) and uppermost (9.4e2.6 ka) sections of the core were characterized by lower SR (13.8 and 17.1 cm/ka, respectively). Core GeoB12624-1 is divided into the early (EH) and mid-to late Holocene (MLH) sediment section which is recognized by a clear shift in grain size patterns at 9.4 ka.

X. Liu et al. / Quaternary Science Reviews 196 (2018) 124e136

127

Fig. 2. Geological map of Eastern Tanzania and the location of marine sediment core GeoB12624-1. The simplified geological map is modified from Semkiwa and Hester (2005).

Table 1 Radiometric ages of core GeoB12624-1 based on DR ¼ 140 ± 25 years (Bouimetarhan et al., 2015). Depth (cm)

Lab code

Conventional14C age (yr BP)

1s calendar age ranges (yr BP)

Calibrated age (cal. yr BP)

2 124 210 300 398 512 596

Poz-30420 Poz-47931 OS-79104 Poz-47932 Poz-47933 Poz-47934 Poz-30421

2810 ± 35 8680 ± 50 9540 ± 65 10 410 ± 60 11 240 ± 60 13 200 ± 70 16 630 ± 80

2308e2419 9091e9265 10172-10 332 11184-11 312 12564-12 664 14781-15 116 19244-19 417

2340 (þ79/-32) 9178 (þ87/-87) 10223 (þ109/-51) 11212 (þ100/-28) 12610 (þ54/-46) 15 040 (þ126/-259) 19380 (þ37/-136)

4.2. Mineral assemblage of surface sediments

4.3. XRF scanning data

The sediments from core GeoB12623-25 generally consists of a mixture of calcite, feldspar (plagioclase and K-feldspar), quartz, clay minerals (e.g., illite, chlorite), and of other trace minerals (<5 wt.%; include pyroxene, apatite, rhodochrosite, and epidote, and pyrite; Table 2). The mean calcite content is 35.0 wt.% and forms the most abundant mineral in the bulk sediment. The feldspar represents the second most abundant mineral groups and constitute 14.2 wt.% of bulk sediment. The clay minerals are dominated by illite and mica, with a mean of 12.0 wt.%, and quartz concentration ranges from 9.6 to 11.2 wt.% (mean 10.4 wt.%).

4.3.1. Principal component analysis (PCA) Three principal components (PC1, PC2 and PC3) were recognized, which explain 94.26% of the total variance (Fig. 4). According the loadings of the elements on PC1, two basic clusters for elements can are identified: a terrigenous cluster (Al, Si, K, Ti and Fe) and a marine cluster (Ca and Ba), which show strong negative correlation on PC1 (Fig. 4a). Two sub-clusters of the terrigenous elements are observed: Al-Si cluster verse Ti-Fe cluster, which are anti-correlated on PC2 and PC3; whereas K cannot be classified into one sub-cluster groups due to its opposite loading on PC2 and PC3. On PC3, Ca is

128

X. Liu et al. / Quaternary Science Reviews 196 (2018) 124e136

Fig. 3. Description and the grain-size distributions of terrigenous component for core GeoB12624-1. Sedimentation rates are calculated after subtracting the thickness of the turbidites. The distribution of turbidite beds indicated by triangles, modified from Liu et al. (2016a).

Table 2 Bulk mineralogy composition of surface sediments off the Rufiji River mouth. Core sites

Depth (m)

Longitude

Latitude

Quartz

Feldspar

Calcite

Illite & Mica

GeoB12623-1 GeoB12624-3 GeoB12625-1 Average

643 648 542

39
45.700 39
45.190 39
44.510

8
12.600 8
13.780 8
12.030

10.5 9.6 11.2 10.4

10.2 15.2 17.2 14.2

39.0 33.3 32.8 35.0

11.6 11.5 12.8 12.0

Feldspars ¼ Plagioclase þ K-feldspar.

anti-correlated to Ba (Fig. 4b), which indicates that these two elements might have different behaviors.

4.3.2. Elemental records since the LGM In marine sediments, specific elemental ratios have been used widely for paleoclimatic reconstructions. Here we present logratios of selected elements (i.e., Al/Si, Fe/K, Fe/Ca, and Ti/Ca) in Fig. 5. The log-ratios of Al/Si and Fe/K share a very similar changing

trend (Fig. 5aeb), which are high and consistent prior to the Holocene. A rapid decrease in log (Al/Si) and (Fe/K) occurs at the beginning of the Holocene, reaching a lower state during the EH with significant fluctuations. At the end of EH, log-ratios of Al/Si and Fe/K decrease sharply but recover to a moderate around 7.4 ka, followed by a slight decrease. Both the log-ratios of Fe/Ca and Ti/Ca are low with gently increasing during the LGM and HS1 (Fig. 5ced), followed by a rapid increase at the end of HS1, reaching the first

X. Liu et al. / Quaternary Science Reviews 196 (2018) 124e136

129

Fig. 4. Correlations between elements and the first three principal components derived from the PCA carried out on the XRF-core scanning data. (a) Loadings of each element along the first two principal components (PC1 and PC2); (b) Loadings of each element along the first and third principal component (PC1 and PC3). PC1, PC2 and PC3 represent respectively 57.52%, 23.65% and 13.08% of the total variance.

maximum corresponding to the BA interval. At the onset of the YD, a rapid and substantial decrease in log (Fe/Ca) and log (Ti/Ca) is found, followed by a rapid increase to the second maximum at the onset of the EH. During the Holocene, these ratios are marked by a decreasing trend till ~8.7 ka, and then back to a moderate state, followed by another decrease. In general, the log-ratios of Fe/Ca and Ti/Ca increased during the deglacial period; however one reversed trend was found, which perfectly phases with the YD chronozone confined by two 14C ages (arrows in Fig. 5ced). According to the regime-shift algorithm applied to the time series of log-ratios of Fe/Ca (Fig. 5e), four regimes were identified: 19.3e15.1 ka (mean ¼ 0.27), 15.1e12.1 ka (mean ¼ 0.60), 12.1e10.4 ka (mean ¼ 0.83), 10.4e7.4 ka (mean ¼ 0.44), and 7.4e2.5 ka (mean ¼ 0.13). On the basis of RSI values, the largest magnitude of significant shifts occurred at 7.4 ka and a second largest one at 15.1 ka (Fig. 5e).

5. Discussion 5.1. Element ratios as proxies for paleoclimatic change The variation in concentrations of major elements in marine sediments has been related to climatic change; however the interpretation of elements differs from site to site with partly controversial interpretations (Govin et al., 2012). Among the logratios of elements considered here, to date, Al/Si/and Fe/K have been used as indicators for chemical weathering (Hu et al., 2012; Lupker et al., 2013; Huang et al., 2016). Theses elemental ratios however, are also most likely to be related to grain-size sorting because, Si and K are connected to both fine-grained clay minerals such as illite and chlorite, as well as coarse-grained minerals such as quartz and feldspar (Clift et al., 2014; Govin et al., 2014; Liu et al., 2017a). In core GeoB12624-1, the log-ratios between the elements from the terrigenous cluster (Al/Si and Fe/K) show negative correlation to grain-size median values (Fig. 6aeb). These simultaneous changes in elemental ratios and grain-size properties are caused by hydrodynamic sorting (Bloemsma et al., 2012), because elements, such as Al and Fe, are associated with the clay minerals, whereas Si and K are linked to quartz and feldspar. Hydrodynamic sorting effects thus play a more important role than chemical weathering in

controlling elemental ratios of Al/Si and Fe/K in core GeoB12624-1. During the EH, such elemental ratios show large amplitude fluctuations (Fig. 5aeb), corresponding to the development of turbidite deposits (dashed line in Fig. 5a). Higher river discharge to the continental margin during humid periods might result in increased deposition of sediments on the adjacent continental slope and initiate turbidity currents (Covault and Graham, 2010). The logratios Al/Si and Fe/K therefore indicated an interval with intense turbidity current activity on the upper slope off East Africa during the EH, which has been related to humid inland climatic conditions (Fig. 5a; Liu et al., 2016a). Consequently, to trace the intensity of the chemical weathering, it would be better to perform destructive sediment analysis of elemental ratios of Al/Si and Fe/K strictly on the clay fraction of the bulk sediment (Bastian et al., 2017). On the basis of PCA results (Fig. 4), Fe/Ca and Ti/Ca ratios account for two end-members of the elemental composition of marine sediments, because Fe and Ti are commonly indicative of terrigenous sediments, while the Ca is mostly derived from the marine biogenic carbonate (Steinke et al., 2014; Stuut et al., 2014). In our research area, marine carbonate production and preservation is mainly controlled by terrigenous input from fluvial discharge according to study on the coccolith species in surface samples from (Stolz et al., 2015). Ba/Al ratio, as an indicator of marine productivity (Dean et al., 1997), shows negative correlation with Fe intensity (Fig. 6c), suggesting that high terrigenous input would suppress the carbonate production. Carbonate dissolution caused by degradation of organic carbon has been proved to be minor due to the low content of organic carbon (Liu et al., 2017b). In addition, Fe, a sensitive element to redox conditions in marine sediments, show good correlation with Ti (Fig. 6d), indicating core sediments in this study have not been significant influenced by early diagenesis. Therefore, calcium intensity in our core sediments is possibly related to carbonate production and terrigenous dilution or the interplay between them. The log-ratios of Fe/Ca and Ti/Ca in the Tanzanian slope sediments are therefore possibly controlled by terrigenous discharge by the Rufiji River, which have been widely used to indicate the fluvial discharge (Blanchet et al., 2013; Revel et al., 2014). This interpretation is further supported by the benthic foraminiferal assemblage (dominated by Saidovina karreriana: a fauna indicating the Rufiji River runoff), as well as ratios

130

X. Liu et al. / Quaternary Science Reviews 196 (2018) 124e136

Fig. 5. Down-core profiles of various elemental intensity ratios of GeoB12624-1. (aed) Logarithms of elemental ratios: (a) Al/Si; (b) Fe/K; (c) Fe/Ca; (d) Ti/Ca. (e) Regime Shift Index (RSI) values for log-ratios of Fe/Ca. The turbidite frequency and sedimentation rate of core GeoB12624-1 are indicated by dashed lines in Figure (a) and Figure (d), according to Liu et al. (2016a). Five regimes are determined in the log-ratios of Fe/Ca, indicated with a dashed line in Figure (c). Vertical gray bars denote different time intervals indicated at the top raw of the figure. Solid triangles at the bottom indicate the stratigraphic locations of radiocarbon dates referred from Bouimetarhan et al. (2015).

of organic carbon to nitrogen of the bulk sediment from core GeoB12615-4 (Romahn et al., 2015). Recent studies suggested that sea-level fluctuations could influence the sedimentation pattern on the southern East African margin during the last deglacial and the Holocene, and thus impacted the sedimentary characteristics of the marine sediments in the WIO (Just et al., 2014; van der Lubbe et al., 2014; Liu et al., 2016b). Thus, when we use log-ratios of Fe/Ca and Ti/Ca to trace the terrigenous input into the WIO, the effect of the sea-level changes should be calculated carefully. Paleo-sea-level rise in the research area was firstly recorded around 18e17 ka from coral data from Mayotte in the WIO (Camoin et al., 2004). At that time, the sea level was ~110e115 m lower than present (Fig. 7a). Due to such a significant sea-level rise during the last deglaciation (arrow in Fig. 7a), the mouth of the Rufiji River would have progressively retreated from its more proximal glacial location to the research site to become farther away. As a

consequence, the volume of terrigenous sediments discharged into the WIO would be expected to decrease. However, our elemental log-ratios of Fe/Ca from core of GeoB12624-1, as well as Fe/Ca ratios from the nearby site of core GeoB12615-4 (Romahn et al., 2015), share a similar trend which indicate an increase in sediment discharge of the Rufiji River during the last deglaciation (Fig. 7bec). This suggests that seal-level rise during the deglacial time interval played a minor role in controlling the amount and properties of the terrigenous sediments discharged from the Rufiji River into the WIO. In addition, the change in log-ratios of Fe/Ca of core GeoB12624-1 is synchronous with the relative amount of the clay fraction of bulk sediment and sedimentation rate (Fig. 7cee). These clays are expected to be transported from land to sea with minimal transport time because of their small size and density (Bastian et al., 2017). Previous studies off East Africa showed that the elemental concentrations and sedimentation rates could response to climatic conditions without detectable time lag (Romahn et al., 2015; Liu

X. Liu et al. / Quaternary Science Reviews 196 (2018) 124e136

131

Fig. 6. (aeb) Elemental intensity ratios of Al/Si (a) and Fe/K (b) are plotted against the median grain size of the terrigenous sediments. Grain-size data are cited from Liu et al. (2016a). (c) Ba/Al ratios plotted as a function of Fe intensity. (d) Elemental intensity of Fe plotted as a function of Ti intensity.

et al., 2016a). In conclusion, the log-ratios of Fe/Ca and Ti/Ca of core GeoB12624-1 could serve as proxies for terrigenous sediment discharged into this region of the WIO, and suggest that the sediment discharge is primarily controlled by climatic conditions in the Rufiji River catchment area and not sea-level change. 5.2. East African climatic evolution since the LGM Climatic conditions in East Africa show spatial and temporal complexity since the LGM, due to a regional response to various components of climatic forcing. In order to illustrate the climatic evolution, we make a synthesis of East African paleoclimatic records (Fig. 8), including well-dated climate records from northern East Africa to southern East Africa: The Nile deep sea fan (Revel et al., 2014), Chew Bahir Basin (Foerster et al., 2012), Lake Challa (Verschuren et al., 2009), Lake Tanganyika (Tierney et al., 2008), Lake Massoko (Garcin et al., 2006b), Lake Malawi (Johnson et al., 2002) and Zambezi catchment (Schefuß et al., 2011). The logratios of Fe/Ca, indicative of fluvial discharge for the Rufiji River (Fig. 8e), correspond to hydrologic records from sites located north of the research site (Fig. 8aed), but anti-correlate with a precipitation record of plant waxes off the Zambezi River mouth (Schefuß et al., 2011), southern East Africa (Fig. 8h).

In general, arid climatic conditions prevailed in East Africa (Fig. 8aee) during the LGM, then transited to the AHP during the Holocene by two distinctive dry-wet transitions, which occurred around 15e14.5 ka and 11.5e11 ka, respectively (Gasse, 2000; Gasse et al., 2008). During millennial-scale NH cooling intervals, such as the HS1 and the YD, the sites located north of 8e10
S experienced a severe drought, e.g., the decrease in fluvial discharge reflected in the log-rations of Fe/Ca of core GeoB12624-1 (Fig. 8e), while the south sites (e.g., the Zambezi catchment; Fig. 8f) were humid. Our elemental records, as well a pollen record for GeoB12624-1 (Bouimetarhan et al., 2015), reveal that a decrease in precipitation occurred in Rufiji catchment during the HS1 (Fig. 8e), which is in agreement with previous marine records (Tierney and deMenocal, ~ eda et al., 2016). Increased aridity 2013; Revel et al., 2014; Castan also coincides with terrestrial studies (Fig. 8bed), such as geochemical composition from the Chew Bahir Basin (Foerster et al., 2012), river runoff of Lake Challa indicated by BIT index (Verschuren et al., 2009), and the isotopic composition of leaf wax of Lakes Tanganyika (Tierney et al., 2008, 2010) and Malawi ~ eda et al., 2007, 2009). The HS1 period is termed as one of (Castan the most the most severe aridity over the Afro-Asian monsoon region during the past 50 ka (Stager et al., 2011). Thomas et al. (2012), however, examined the paleoclimatic records across East

132

X. Liu et al. / Quaternary Science Reviews 196 (2018) 124e136

Fig. 7. Various geochemical and grain-size parameters compared to the sea-level fluctuations in the western Indian Ocean (WIO) during the past 20 ka. (a) Reconstructed sea-level curve in the WIO by Camoin et al. (2004). (b) Log-ratios of Fe/Ca of core GeoB12624-1, this study. (c) Fe/Ca ratios of core GeoB12615-4, indicating the rations between terrigenous and marine components (Romahn et al., 2015). (dee) Relative volume contents of the clay fraction (<4 mm) in the terrigenous sediment and the sedimentation rate of core GeoB12624-1 (Liu et al., 2016a).

Africa and found that records from equatorial and northern basins indicated arid conditions, whereas records from southern East Africa (e.g., Zambezi catchment; Fig. 8h) didn't demonstrate a severe aridity. We determined a shift towards more humid conditions ~15 ka from RSI values, at the end of HS1 (Fig. 5e). Such a shift has also been demonstrated by the vegetation records from the same sediment core (Bouimetarhan et al., 2015), as well as numerous records of terrestrial and marine deposition from northern to southern East Africa (Fig. 8aee). Therefore, we infer that arid conditions throughout East Africa during the HS1, while the most south sites (e.g., the Zambezi catchment) experienced humid conditions at this time (Schefuß et al., 2011; Chevalier and Chase, 2015). Although chronological differences cannot be ruled out, the termination of HS1 around 15 ka and hence the initiation of humid conditions, as reflected by our elemental ratios and many other paleoclimatic documents, is in good agreement with modeling study (Mohtadi et al., 2014). As indicated by log-ratios of Fe/Ca and Ti/Ca, there was a rapid decrease in sediment discharge of the Rufiji River during the YD period (Fig. 8e). However, the reconstructed vegetation history from the same core documents an alternation between humid (e.g., Rhizophora, Typha, fern spores) and dry (e.g., Astercaeae and Boscia) taxa without a significant climatic trend (Bouimetarhan et al., 2015). A possible cause for these discrepancies between the data sets is that the short duration of the YD was not captured by some climatic archives. The YD interval in southern East Africa is characterized by increased northerly winds (Johnson et al., 2002), which is inferred from high primary productivity in the northern basin of Lake Malawi documented in a silica record (Fig. 8g). This time with strong northerly winds, as well as, increased aridity in Lake Malawi, indicates a southward shift of the ITCZ during the YD ~ eda et al., 2007, 2009). It should be noted that cold interval (Castan Lake Masoko (Garcin et al., 2006b), located to the north of Lake Malawi, shows evidence for a humid YD interval (Fig. 8f). Lake

Fig. 8. Comparison of paleoclimatic records from East African. (a) The log Fe/Ca of the core MS27PT, Nile deep sea fan (Revel et al., 2014); (b) XRF scanning data from Chew Bahir (Foerster et al., 2012); (c) Lake Challa BIT index (Verschuren et al., 2009); (d) dD of leaf waxes (inverted scale) from Lake Tanganyika (Tierney et al., 2008); (e) Log-ratios of Fe/Ca, indication the Rufiji river discharge, this study; (f) Magnetic susceptibility from Lake Massoko (Garcin et al., 2006b); (g) Biogenic silica mass accumulation rate (BSi MAR) of Lake Malawi (Johnson et al., 2002); (h) Zambezi River freshwater discharge inferred from dD of leaf waxes (inverted scale) of GeoB9307-3 (Schefuß et al., 2011).

Masoko, however, might be influenced by local orographic effects due to its small size, indicating a local rather than a regional cli~ eda et al., 2009). An increase in aridity during matic signal (Castan the YD has been demonstrate by various indicators of both marine and terrestrial records from northern to southern East Africa (Fig. 8aee). These include the YD arid interval recorded in the Nile catchment (Revel et al., 2015), decrease in precipitation in equatorial Lakes Challa (Verschuren et al., 2009) and Albert (Berke et al., 2014), arid conditions indicated by isotopic signals of leaf wax from Lake Tanganyika (Tierney et al., 2008), and decreased fluvial discharge of the Rufiji River recorded by our elemental ratios. To the south, relative wet conditions occurred in Zambezi catchment (Fig. 8h), documented by multi-proxies derived marine sediments off the Zambezi River mouth (Schefuß et al., 2011; Wang et al., 2013; van der Lubbe et al., 2016), which is in anti-phase with the paleoclimatic records from the northern sites. At the end of YD, elemental ratios of Fe/Ca record a resumption of fluvial discharge of

X. Liu et al. / Quaternary Science Reviews 196 (2018) 124e136

the Rufiji River (Fig. 8e), indicating increased precipitation in southern East Africa. This regime shift of climatic conditions is coincident with a change in surface winds recorded in sediment offshore northern East Africa (Garcin et al., 2007; Talbot et al., 2007). Through a comparison with a range of other regional paleoclimatic data, this regime shift is accompanied by a fast recovery to a humid climate (the AHP) in the sites north of 8e10
S and the onset of arid conditions south of it (Fig. 8). In summary, the periods during the HS1 and YD stadials are characterized by arid conditions along the East African margin and increased humidity in the most southern East Africa (e.g., the Zambezi catchment), which were terminated with a regime shift in climatic conditions. The AHP in East Africa during the early and mid-Holocene has been recorded by our XRF core scanning data: (1) an increase in fluvial discharge of the Rufiji River reflected in log-ratios Fe/Ca and Ti/Ca; and (2) an increase in turbidity current deposition, indicated by log-ratios of Al/Si and Fe/K. Such an increase in fluvial discharge, caused by humid conditions in the hinterland of the Rufiji catchment, is also evident in multi-proxy data of nearby marine records from core GeoB12615-4 (Romahn et al., 2015). Humid conditions during the AHP enhanced the fluvial discharge of the river from East Africa, such as the Nile River (Blanchet et al., 2013; Revel et al., ~ eda et al., 2016) and the Pangani River (Liu et al., 2017a, 2015; Castan 2017b). The AHP has also been confirmed by lake-level re and Gasse, 2002; Garcin et al., 2012; constructions (Chalie Junginger and Trauth, 2013; Morrissey and Scholz, 2014; van der Lubbe et al., 2017) and biogeochemical indicators derived from lake sediments (Tierney et al., 2008; Tierney et al., 2011b; Tierney ~ eda et al., 2016). A and deMenocal, 2013; Costa et al., 2014; Castan “hinge zone” however, was observed around 8e10
S (Liu et al., 2017b): the northern sites were humid (e.g., the Rufiji catchment) ~ eda while the southern sites (e.g., Lake Malawi) were arid (Castan et al., 2007, 2009). Furthermore, the AHP terminated abruptly around 5 ka inferred from lake-level reconstructions (Garcin et al., 2009, 2012) and hydrogen isotopes in leaf wax (Tierney et al., 2008, 2011a), our geochemical record and many other archives however, indicate a more gradual or stepwise pace (Weldeab et al., 2014b; Revel et al., 2015; Liu et al., 2017b). These abrupt responses to the AHP termination might be caused by a difference in the sensitivity ~ eda of the diverse proxies to the same climatic variability (Castan et al., 2016), or by a no-linear response to gradual climatic forcing (Collins et al., 2017; van der Lubbe et al., 2017). 5.3. Climate change forcing on different time scales A comparison of the log-ratios with other paleoclimatic records around East Africa shows that low-latitude insolation might have dominated the occurrence of humid phases in this region (Verschuren et al., 2009; Berke et al., 2012; Cockerton et al., 2015). The increase in precipitation at north of 8e10
S since the LGM, is primarily related to summer insolation which increased in amplitude in two steps that coincide with the end of cold intervals at high northern latitudes. The synchronicity between the noticeable paleoclimatic changes in East Africa and the cold intervals (HS1 and YD) at high latitudes (Fig. 9a) indicates that extra-tropical driving forcing might have impressed the tropical climatic system during the last deglaciation. On the basis of integrated paleoclimatic proxies and modeling results, Mohtadi et al. (2014) proposed that drastic climatic changes in tropical East Africa might respond to the AMOC (Fig. 9b), which have been slowdown or shutdown completely during HS1 and the YD stadial (Johnson et al., 2011; Ritz et al., 2013). During such cold intervals, the ITCZ moved southwards (Mohtadi et al., 2016), most probably resulting from NH cooling (Schefuß et al., 2011; Deplazes et al., 2013). Under this scenario, the

133

Fig. 9. Climatic changes and its driving forcing in East Africa. (a) Temperature in central Greenland, referred from Alley (2000). (b) Sedimentary 231Pa/230Th against 238U-based age (solid) and 232U -based age values (dashed line), indicating the strength of the Atlantic Meridional Overturning Circulation (AMOC), based on McManus et al. (2004). (c) Solid line: Log-ratios of Fe/Ca, indicating the Rufiji river discharge, this study; Dashed line: Insolation in July at 20
N (Berger and Loutre, 1991). (d) Hydrogen isotope compositions of the n-C31 alkane off the Zambezi River mouth (Schefuß et al., 2011). (e) SST gradient between the western and eastern Indian Ocean, based Mg/Ca SST derived from the foraminiferal tests from core GeoB12615-4 in the western Indian Ocean (Romahn et al., 2014), and core 39 KL in eastern Indian Ocean (Mohtadi et al., 2014).

strong southward migration of the ITCZ during the HS1 and the YD, might have resulted in a decrease in precipitation in the equatorial and northern tropical East Africa (Otto-Bliesner et al., 2014) and an increase in precipitation in most southern East Africa (Schefuß et al., 2011; Bouimetarhan et al., 2015). The broad geographic range of drought during the HS1 and YD can not only be attributed to a southward movement of the ITCZ, but is also related to changes in the SSTs in the Indian Ocean (Stager et al., 2011; Tierney et al., 2011b; Weldeab et al., 2014a; Pausata and Tierney, 2015). Cooler SSTs in the WIO could reduce moisture that transported from the Indian Ocean, causing aridity in East Africa (Tierney et al., 2011b). If SSTs in the WIO were not decreased, the rainfall in East Africa would not be largely affected, in spite that the cooling in the NH made a southward migration of the ITCZ (Tierney et al., 2015). However, recent SST reconstructions based on foraminiferal tests from core sediments off East Africa, show a slight warming during the HS1 (Romahn et al., 2014; Rippert et al., 2015), suggesting that at least in our research area, arid climatic conditions was not a result of local SST cooling. The SST cooling in the WIO can therefore not explain the tropical East African climate during the HS1 and the YD. The AHP, in East Africa, have been related to insolation forcing due to intense monsoonal precipitation (Verschuren et al., 2009;

134

X. Liu et al. / Quaternary Science Reviews 196 (2018) 124e136

Berke et al., 2012). However, other authors suggested that an increase in precipitation in East Africa was due to warmer SSTs in the WIO or a positive west-east temperature gradient in the Indian Ocean (Tierney et al., 2011a). Here, we applied Mg/Ca SST derived from the foraminiferal tests from core GeoB12615-4 in the WIO (Romahn et al., 2014) and core 39 KL in eastern Indian Ocean (Mohtadi et al., 2014) to estimate the SST gradient in the tropical Indian Ocean (Fig. 9e). The SST gradient in the tropical Indian Ocean doesn't show similar trend with the fluvial discharge of the Rufiji River during the Holocene, thus cannot account for the AHP in East Africa. In fact, the fluvial discharge of the Rufiji River, as reflected in elemental ratios, is roughly synchronous with NH summer insolation at 20
N (Fig. 9c), indicating the strong influence of summer insolation on climate in East Africa. In addition, an eastward migration of the CAB might have been another factor that could have enhanced the rainfall in East Africa, which could transport additional moisture from the Atlantic Ocean (Junginger and Trauth, 2013; Costa et al., 2014; Liu et al., 2017b; van der Lubbe et al., 2017). Therefore, we suggest that the climatic transition from the arid LGM to the humid EH in East Africa is primary controlled by lowlatitude insolation, which modulates the migration path of the ITCZ and CAB to influence the precipitation in the research area. Recently, the termination of AHP has been related to a decrease in artic air temperatures, which caused a decrease on the Tropical Easterly Jet and consequently aridification in tropical Africa (Collins et al., 2017). However, a recent modeling experiment suggests that the high-cooling over the Arctic regions was enhanced by the African precipitation decrease at the end of the AHP (Muschitiello et al., 2015). This connection may have contributed to a tipping point behavior between low and high latitudes, and thus the relationship between the Arctic and African climate at the end of the AHP needs further research. The other studies suggested an relationship between the extreme rainfall anomalies in equatorial East Africa and the ENSO (Wolff et al., 2011). However, it has recently been suggested that the influence of the tropical Pacific on rainfall in East Africa is minor on multidecadal and perhaps longer timescales (Tierney et al., 2013). The statistical relationship between East African rainfall and ENSO described in earlier studies is due to the indirect effects from the Pacific via the Indian Ocean (Clark et al., 2003). In South Africa (south of 20
S), precipitation during the Holocene has been linked to the latitudinal position of the SH westerlies in response to changes in sea-ice extent around Antarctica (Chevalier and Chase, 2015). Recently, the Talos Dome record suggests that, in the Western Ross Sea region, a change in the transport dynamics of wind dust occurred in two steps: a first one (12e8 ka) corresponded to the complete suppression of dust transport from remote extra-Antarctic sources; the second phase characterized the middle and late Holocene (after 7 ka) likely related to internal climatic dynamics affecting the entire Ross Sea region (Albani et al., 2012; Delmonte et al., 2010; Mezgec et al., 2017). The influence of such changes in atmospheric dynamics could explain (or be a consequence) the change in climatic conditions in southern East Africa, which needs to be carefully considered in further study. 6. Conclusions Our geochemical composition results show an interplay between the influence of high-latitude and low-latitude climate forcing on East African climate during the last deglaciation and Holocene: (1) Elemental log-ratios of Fe/Ca and Ti/Ca of core GeoB12624-1, indicative for the Rufiji River discharge, reveal arid condition in East Africa north of 8e10
S during the LGM and a humid

phase (AHP) during the early Holocene. This transition corresponds to summer insolation at 20
N, implying a predominant low-latitude forcing. (2) Two drastic droughts however, occurred during the last deglaciation, are consistent with the cold intervals in the NH (i.e., HS1 and YD): During these periods, the ITCZ moved to close to most southern part of East Africa (to around 15e20
S. This led to an increasing precipitation and humid phases in southern East Africa but, a decrease in precipitation and arid phases. in equatorial and northern East Africa (north of 8e10
S). After the YD, summer insolation reached its maximum, and the ITCZ moved northward and reached close to the equator, thus the AHP was observed in East Africa north of 8e10
S. These results highlight the high-latitude forcing during these cold intervals. (3) In addition, we estimate the impact of the Indian Ocean's SSTs on precipitation in East Africa, and find that SST is not the primary controlling forcing for climatic changes in this case study. Therefore, we conclude that arid phases occurred in equatorial and northern East Africa when the ITCZ was forced southwards during the NH cooling at high latitudes, such as the HS1 and the YD, or when low-latitude summer insolation declined after the termination of the AHP during the late Holocene. Acknowledgements We thank the crew and scientists of RV Meteor Cruise M75/2 for collecting the gravity core. Dr. Holger Kuhlmann and Dr. Jürgen P€ atzold are acknowledged for XRF scanning. XTL was funded through the National Natural Science Foundation of China (41606062) and supported by State Key Laboratory of Marine Geology, Tongji University (MGK1824). This project was funded by the Deutsche Forschungsgemeinschaft as part of the DFG-Research Center/Cluster of Excellence MARUM “The Ocean in the Earth System”. Appendix A. Supplementary data Supplementary data related to this article can be found at https://doi.org/10.1016/j.quascirev.2018.08.004. References Albani, S., Delmonte, B., Maggi, V., Baroni, C., Petit, J.R., Stenni, B., Mazzola, C., Frezzotti, M., 2012. Interpreting last glacial to Holocene dust changes at Talos Dome (East Antarctica): implications for atmospheric variations from regional to hemispheric scales. Clim. Past 8, 741e750. Arthurton, R., 2003. The fringing reef coasts of eastern Africadpresent processes in their long-term context. West. Indian Ocean J. Mar. Sci. 2, 1e13. Alley, R.B., 2000. The Younger Dryas cold interval as viewed from central Greenland. Quat. Sci. Rev. 19, 213e226. 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. Roy. Meteorol. Soc. 141, 16e26. Bard, E., 2013. Out of the African humid period. Science 342, 808e809. 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, 1111e1114. Barker, S., Diz, P., Vautravers, M.J., Pike, J., Knorr, G., Hall, I.R., Broecker, W.S., 2009. Interhemispheric Atlantic seesaw response during the last deglaciation. Nature 457, 1097e1102. Bastian, L., Revel, M., Bayon, G., Dufour, A., Vigier, N., 2017. Abrupt response of chemical weathering to Late Quaternary hydroclimate changes in northeast Africa. Sci. Rep. 7, 44231. 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, 2008e2022. Berger, A., Loutre, M.F., 1991. Insolation values for the climate of the last 10 million

X. Liu et al. / Quaternary Science Reviews 196 (2018) 124e136 years. Quat. Sci. Rev. 10, 297e317. 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, 59e74. Berke, M.A., Johnson, T.C., Werne, J.P., Livingstone, D.A., Grice, K., Schouten, S., Damste, J.S.S., 2014. Characterization of the last deglacial transition in Tropical East Africa: insights from Lake Albert. Palaeogeogr. Palaeoclimatol. Palaeoecol. 409, 1e8. 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, 127e145. 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, 98e110. Bloemsma, M.R., Zabel, M., Stuut, J.B.W., Tjallingii, R., Collins, J.A., Weltje, G., 2012. Modelling the joint variability of grain size and chemical composition in sediments. Sediment. Geol. 280, 135e148. Bouimetarhan, I., Dupont, L., Kuhlmann, H., Patzold, J., Prange, M., Schefuss, E., Zonneveld, K., 2015. Northern Hemisphere control of deglacial vegetation changes in the Rufiji uplands (Tanzania). Clim. Past 11, 751e764. Box, M., Krom, M., Cliff, R., Bar-Matthews, M., Almogi-Labin, A., Ayalon, A., Paterne, M., 2011. Response of the Nile and its catchment to millennial-scale climatic change since the LGM from Sr isotopes and major elements of East Mediterranean sediments. Quat. Sci. Rev. 30, 431e442. Brown, E.T., Johnson, T.C., Scholz, C.A., Cohen, A.S., King, J.W., 2007. Abrupt change in tropical African climate linked to the bipolar seesaw over the past 55,000 years. Geophys. Res. Lett. 34, L20702. 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, 119e146. ~ eda, I.S., Werne, J.P., Johnson, T.C., 2007. Wet and arid phases in the southeast Castan African tropics since the last glacial maximum. Geology 35, 823e826. ~ eda, I.S., Werne, J.P., Johnson, T.C., Filley, T.R., 2009. Late Quaternary vegetaCastan tion history of southeast Africa: the molecular isotopic record from Lake Malawi. Palaeogeogr. Palaeoclimatol. Palaeoecol. 275, 100e112. ~ eda, I.S., Schouten, S., Pa €tzold, J., Lucassen, F., Kasemann, S., Kuhlmann, H., Castan Schefuß, E., 2016. Hydroclimate variability in the Nile River basin during the past 28,000 years. Earth Planet Sci. Lett. 438, 47e56. , F., Gasse, F., 2002. Late GlacialeHolocene diatom record of water chemistry Chalie and lake level change from the tropical East African Rift Lake Abiyata (Ethiopia). Palaeogeogr. Palaeoclimatol. Palaeoecol. 187, 259e283. 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, 117e130. 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. Clim. 16, 548e554. Clift, P.D., Wan, S., Blusztajn, J., 2014. Reconstructing chemical weathering, physical erosion and monsoon intensity since 25Ma in the northern South China Sea: a review of competing proxies. Earth Sci. Rev. 130, 86e102. 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: stableisotope and biomarker evidence from Lakes Victoria and Edward. Quat. Sci. Rev. 130, 57e71. Collins, J.A., Prange, M., Caley, T., Gimeno, L., Beckmann, B., Mulitza, S., Skonieczny, C., Roche, D., Schefuß, E., 2017. Rapid termination of the African Humid Period triggered by northern high-latitude cooling. Nat. Commun. 8, 1372. 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, 58e67. Covault, J.A., Graham, S.A., 2010. Submarine fans at all sea-level stands: tectonomorphologic and climatic controls on terrigenous sediment delivery to the deep sea. Geology 38, 939e942. Dean, W.E., Gardner, J.V., Piper, D.Z., 1997. Inorganic geochemical indicators of glacial-interglacial changes in productivity and anoxia on the California continental margin. Geochem. Cosmochim. Acta 61, 4507e4518. Defrance, D., Ramstein, G., Charbit, S., Vrac, M., Famien, A.M., Sultan, B., Swingedouw, D., Dumas, C., Gemenne, F., Alvarez-Solas, J., Vanderlinden, J.P., 2017. Consequences of rapid ice sheet melting on the Sahelian population vulnerability. Proc. Natl. Acad. Sci. U. S. A. 114, 6533e6538. Delmonte, B., Baroni, C., Andersson, P.S., Schoberg, H., Hansson, M., Aciego, S., Petit, J.-R., Albani, S., Mazzola, C., Maggi, V., Frezzotti, M., 2010. Aeolian dust in the Talos Dome ice core (East Antarctica, Pacific/Ross Sea sector): Victoria Land versus remote sources over the last two climate cycles. J. Quat. Sci. 25, 1327e1337. 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, 347e361. 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, second ed. Elsevier, Oxford, pp. 99e108. Deplazes, G., Luckge, A., Peterson, L.C., Timmermann, A., Hamann, Y., Hughen, K.A.,

135

Rohl, U., Laj, C., Cane, M.A., Sigman, D.M., Haug, G.H., 2013. Links between tropical rainfall and North Atlantic climate during the last glacial period. Nat. Geosci. 6, 213e217. Duvail, S., Hamerlynck, O., 2007. The Rufiji River flood: plague or blessing? Int. J. Biometeorol. 52, 33e42. Filippi, M., Talbot, M., 2005. The palaeolimnology of northern Lake Malawi over the last 25ka based upon the elemental and stable isotopic composition of sedimentary organic matter. Quat. Sci. Rev. 24, 1303e1328. 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, 25e37. Garcin, Y., Vincens, A., Williamson, D., Guiot, J., Buchet, G., 2006a. Wet phases in tropical southern Africa during the last glacial period. Geophys. Res. Lett. 33. Garcin, Y., Williamson, D., Taieb, M., Vincens, A., Mathe, P.E., Majule, A., 2006b. Centennial to millennial changes in maar-lake deposition during the last 45,000 years in tropical Southern Africa (Lake Masoko, Tanzania). Palaeogeogr. Palaeoclimatol. Palaeoecol. 239, 334e354. Garcin, Y., Vincens, A., Williamson, D., Buchet, G., Guiot, J., 2007. Abrupt resumption of the African monsoon at the younger dryas-Holocene climatic transition. Quat. Sci. Rev. 26, 690e704. Garcin, Y., Junginger, A., Melnick, D., Olago, D.O., Strecker, M.R., Trauth, M.H., 2009. Late pleistoceneeHolocene rise and collapse of Lake Suguta, northern Kenya rift. Quat. Sci. Rev. 28, 911e925. Garcin, Y., Melnick, D., Strecker, M.R., Olago, D., Tiercelin, J.-J., 2012. East African mid-Holocene wetedry transition recorded in palaeo-shorelines of Lake Turkana, northern Kenya Rift. Earth Planet Sci. Lett. 331e332, 322e334. Gasse, F., 2000. Hydrological changes in the African tropics since the last glacial maximum. Quat. Sci. Rev. 19, 189e211. Gasse, F., Chalie, F., Vincens, A., Williams, M.A.J., Williamson, D., 2008. Climatic patterns in equatorial and southern Africa from 30,000 to 10,000 years ago reconstructed from terrestrial and near-shore proxy data. Quat. Sci. Rev. 27, 2316e2340. Govin, A., Chiessi, C.M., Zabel, M., Sawakuchi, A.O., Heslop, D., Horner, T., Zhang, Y., Mulitza, S., 2014. Terrigenous input off northern South America driven by changes in Amazonian climate and the North Brazil Current retroflection during the last 250 ka. Clim. Past 10, 843e862. 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
Ne49
S): imprint of terrigenous input and continental weathering. Gcubed 13, Q01013. Hu, D.K., Boning, P., Kohler, C.M., Hillier, S., Pressling, N., Wan, S.M., Brumsack, H.J., Clift, P.D., 2012. Deep sea records of the continental weathering and erosion response to East Asian monsoon intensification since 14 ka in the South China Sea. Chem. Geol. 326, 1e18. Huang, J., Wan, S.M., Xiong, Z.F., Zhao, D.B., Liu, X.T., Li, A.C., Li, T.G., 2016. Geochemical records of Taiwan-sourced sediments in the South China Sea linked to Holocene climate changes. Palaeogeogr. Palaeoclimatol. Palaeoecol. 441, 871e881. Johnson, T.C., Brown, E.T., McManus, J., Barry, S., Barker, P., Gasse, F., 2002. A highresolution paleoclimate record spanning the past 25,000 years in southern East Africa. Science 296, 113e132. Johnson, T.C., Brown, E.T., Shi, J.M., 2011. Biogenic silica deposition in Lake Malawi, East Africa over the past 150,000 years. Palaeogeogr. Palaeoclimatol. Palaeoecol. 303, 103e109. 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). Global Planet. Change 111, 174e188. Just, J., Schefuss, E., Kuhlmann, H., Stuut, J.B.W., Patzold, J., 2014. Climate induced sub-basin source-area shifts of Zambezi River sediments over the past 17 ka. Palaeogeogr. Palaeoclimatol. Palaeoecol. 410, 190e199. Kohn, M., Zonneveld, K.A.F., 2010. Calcification depth and spatial distribution of Thoracosphaera heimii cysts: implications for palaeoceanographic reconstructions. Deep-Sea Res. Part I Oceanogr. Res. Pap. 57, 1543e1560. Kutzbach, J.E., Street-Perrott, F.A., 1985. Milankovitch forcing of fluctuations in the level of tropical lakes from 18 to 0 kyr BP. Nature 317, 130e134. Liu, X.T., Rendle-Bühring, R., Henrich, R., 2016a. Climate and sea-level controls on turbidity current activity on the Tanzanian upper slope during the last deglaciation and the Holocene. Quat. Sci. Rev. 133, 15e27. Liu, X.T., Rendle-Bühring, R., Meyer, I., Henrich, R., 2016b. Holocene shelf sedimentation patterns off equatorial East Africa constrained by climatic and sealevel changes. Sediment. Geol. 331, 1e11. Liu, X.T., Rendle-Bühring, R., Henrich, R., 2017a. Geochemical composition of Tanzanian shelf sediments indicates Holocene climatic and sea-level changes. Quat. Res. 87, 442e454. Liu, X.T., Rendle-Bühring, R., Kuhlmann, H., Li, A., 2017b. Two phases of the Holocene East African humid period: inferred from a high-resolution geochemical record off Tanzania. Earth Planet Sci. Lett. 460, 123e134. Lupker, M., France-Lanord, C., Galy, V., Lave, J., Kudrass, H., 2013. Increasing chemical weathering in the Himalayan system since the last glacial maximum. Earth Planet Sci. Lett. 365, 243e252. McClanahan, T.R., 1988. Seasonality in East Africa's coastal waters. Mar. Ecol. Prog. Ser. 44, 191e199. McManus, J.F., Francois, R., Gherardi, J.M., Keigwin, L.D., Brown-Leger, S., 2004. Collapse and rapid resumption of Atlantic meridional circulation linked to

136

X. Liu et al. / Quaternary Science Reviews 196 (2018) 124e136

deglacial climate changes. Nature 428, 834e837. Mezgec, K., Stenni, B., Crosta, X., Masson-Delmotte, V., Baroni, C., Braida, M., Ciardini, V., Colizza, E., Melis, R., Salvatore, M.C., Severi, M., Scarchilli, C., Traversi, R., Udisti, R., Frezzotti, M., 2017. Holocene sea ice variability driven by wind and polynya efficiency in the Ross Sea. Nat. Commun. 8, 1334. Milliman, J.D., Syvitski, J.P.M., 1992. Geomorphic tectonic control of sediment discharge to the ocean - the importance of small mountainous rivers. J. Geol. 100, 525e544. 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, 76e80. Mohtadi, M., Prange, M., Steinke, S., 2016. Palaeoclimatic insights into forcing and response of monsoon rainfall. Nature 533, 191e199. Morrissey, A., Scholz, C.A., 2014. Paleohydrology of Lake Turkana and its influence on the Nile River system. Palaeogeogr. Palaeoclimatol. Palaeoecol. 403, 88e100. Muschitiello, F., Zhang, Q., Sundqvist, H.S., Davies, F.J., Renssen, H., 2015. Arctic climate response to the termination of the African humid period. Quat. Sci. Rev. 125, 91e97. Newell, B.S., 1957. A Preliminary Survey of the Hydrography of the British East African Coastal Waters, 9. Colonial Office Fishery Publication, pp. 1e20. Nicholson, S.E., Kim, E., 1997. The relationship of the El Nino southern oscillation to African rainfall. Int. J. Climatol. 17, 117e135. Nicholson, S.E., 2000. The nature of rainfall variability over Africa on time scales of decades to millenia. Global Planet. Change 26, 137e158. Otto-Bliesner, B.L., Russell, J.M., Clark, P.U., Liu, Z., Overpeck, J.T., Konecky, B., deMenocal, P., Nicholson, S.E., He, F., Lu, Z., 2014. Coherent changes of southeastern equatorial and northern African rainfall during the last deglaciation. Science 346, 1223e1227. Pausata, F.S., Tierney, J.E., 2015. Deglacial Indian monsoon failure and North Atlantic stadials linked by Indian Ocean surface cooling. Nat. Geosci. 9, 46. Punwong, P., Marchant, R., Selby, K., 2013. Holocene mangrove dynamics and environmental change in the Rufiji Delta, Tanzania. Veg. Hist. Archaeobotany 22, 381e396. 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, 1685e1696. 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, 200e221. Rippert, N., Baumann, K.H., Patzold, J., 2015. Thermocline fluctuations in the western tropical Indian Ocean during the past 35 ka. J. Quat. Sci. 30, 201e210. Ritz, S.P., Stocker, T.F., Grimalt, J.O., Menviel, L., Timmermann, A., 2013. Estimated strength of the Atlantic overturning circulation during the last deglaciation. Nat. Geosci. 6, 208e212. Rodionov, S.N., 2004. A sequential algorithm for testing climate regime shifts. Geophys. Res. Lett. 31. € hl, U., Westerhold, T., Bralower, T.J., Zachos, J.C., 2007. On the duration of the Ro Paleocene-Eocene thermal maximum (PETM). G-cubed 8. €tzold, J., 2014. Deglacial intermediate Romahn, S., Mackensen, A., Groeneveld, J., Pa water reorganization: new evidence from the Indian Ocean. Clim. Past 10, 293e303. 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, 34e48. Saji, N.H., Goswami, B.N., Vinayachandran, P.N., Yamagata, T., 1999. A dipole mode in the tropical Indian Ocean. Nature 401, 360e363. €tzold, J., Schneider, R., 2013. Western Indian Ocean Savoye, B., Ridderinkhof, H., Pa Climate and Sedimentation - Cruise No. M75 - December 29, 2007 - April 08, 2008 - Port Louis (Mauritius) - Cape Town (South Africa). DFG-Senatskommission für Ozeanographie, Bremen. €tzold, J., 2011. Forcing of Schefuß, E., Kuhlmann, H., Mollenhauer, G., Prange, M., Pa wet phases in southeast Africa over the past 17,000 years. Nature 480, 509e512. Semkiwa, P.M., Hester, B., 2005. Opportunities for Mineral Resource Development: Tanzania, fourth ed. Ministry of Energy and Minerals, Republic of Tanzania, Dar es Salaam. Shaghude, Y., 2005. Coastal Impacts of Water Abstraction and Impoundment in Africa: the Case of Rufiji (River). Shaghude, Y.W., 2007. Shore morphology and sediment characteristics south of Pangani River, coastal Tanzania. West. Indian Ocean J. Mar. Sci. 3, 93e104. 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, 140e144. Stager, J.C., Ryves, D.B., Chase, B.M., Pausata, F.S.R., 2011. Catastrophic drought in the Afro-Asian monsoon region during Heinrich event 1. Science 331, 1299e1302. Steinke, S., Mohtadi, M., Prange, M., Varma, V., Pittauerova, D., Fischer, H.W., 2014. Mid- to Late-Holocene AustralianeIndonesian summer monsoon variability. Quat. Sci. Rev. 93, 142e154.

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, 473e488. 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, 83e94. Talbot, M.R., Filippi, M.L., Jensen, N.B., Tiercelin, J.J., 2007. An abrupt change in the African monsoon at the end of the Younger Dryas. G-cubed 8. Temple, P.H., Sundborg, Å., 1972. The Rufiji River, Tanzania hydrology and sediment transport. Geogr. Ann. Phys. Geogr. 54, 345e368. Thomas, D.S.G., Bailey, R., Shaw, P.A., Durcan, J.A., Singarayer, J.S., 2009. Late Quaternary highstands at Lake Chilwa, Malawi: frequency, timing and possible forcing mechanisms in the last 44 ka. Quat. Sci. Rev. 28, 526e539. Thomas, D.S.G., Burrough, S.L., Parker, A.G., 2012. Extreme events as drivers of early human behaviour in Africa? The case for variability, not catastrophic drought. J. Quat. Sci. 27, 7e12. 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, 252e255. Tierney, J.E., Russell, J.M., Huang, Y.S., 2010. A molecular perspective on Late Quaternary climate and vegetation change in the Lake Tanganyika basin, East Africa. Quat. Sci. Rev. 29, 787e800. Tierney, J.E., Lewis, S.C., Cook, B.I., LeGrande, A.N., Schmidt, G.A., 2011a. Model, proxy and isotopic perspectives on the east African humid period. Earth Planet Sci. Lett. 307, 103e112. Tierney, J.E., Russell, J.M., Damste, J.S.S., Huang, Y.S., Verschuren, D., 2011b. Late quaternary behavior of the east African monsoon and the importance of the Congo air boundary. Quat. Sci. Rev. 30, 798e807. Tierney, J.E., deMenocal, P.B., 2013. Abrupt shifts in Horn of Africa hydroclimate since the last glacial maximum. Science 342, 843e846. Tierney, J.E., Smerdon, J.E., Anchukaitis, K.J., Seager, R., 2013. Multidecadal variability in east African hydroclimate controlled by the Indian Ocean. Nature 493, 389e392. Tierney, J.E., Pausata, F.S.R., deMenocal, P., 2015. Deglacial Indian monsoon failure and North Atlantic stadials linked by Indian Ocean surface cooling. Nat. Geosci. 9, 46e50. 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. Gcubed 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. Clim. 22, 993e1013. van der Lubbe, J.J.L., Tjallingii, R., Prins, M.A., Brummer, G.-J.A., Jung, S.J.A., Kroon, D., Schneider, R.R., 2014. Sedimentation patterns off the Zambezi River over the last 20,000 years. Mar. Geol. 355, 189e201. 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. G-cubed 17, 181e198. van der Lubbe, H.J.L., Krause-Nehring, J., Junginger, A., Garcin, Y., Joordens, J.C.A., Davies, G.R., Beck, C., Feibel, C.S., Johnson, T.C., Vonhof, H.B., 2017. Gradual or abrupt? Changes in water source of Lake Turkana (Kenya) during the African Humid Period inferred from Sr isotope ratios. Quat. Sci. Rev. 174, 1e12. Verschuren, D., Sinninghe Damste, J.S., Moernaut, J., Kristen, I., Blaauw, M., Fagot, M., Haug, G.H., CHALLACEA project members, 2009. Half-precessional dynamics of monsoon rainfall near the East African Equator. Nature 462, 637e641. Wang, Y.V., Larsen, T., Leduc, G., Andersen, N., Blanz, T., Schneider, R.R., 2013. What does leaf wax dD from a mixed C3/C4 vegetation region tell us? Geochem. Cosmochim. Acta 111, 128e139. Webster, P.J., Moore, A.M., Loschnigg, J.P., Leben, R.R., 1999. Coupled oceanatmosphere dynamics in the Indian Ocean during 1997-98. Nature 401, 356e360. €nsli, H., Schneider, R.R., 2014a. Links between Weldeab, S., Lea, D.W., Oberha southwestern tropical Indian Ocean SST and precipitation over southeastern Africa over the last 17 kyr. Palaeogeogr. Palaeoclimatol. Palaeoecol. 410, 200e212. Weldeab, S., Menke, V., Schmiedl, G., 2014b. The pace of East African monsoon evolution during the Holocene. Geophys. Res. Lett. 41, 1724e1731. 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, 423e438. €hl, U., Laskar, J., Raffi, I., Bowles, J., Lourens, L.J., Zachos, J.C., 2007. Westerhold, T., Ro 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. Wolff, C., Haug, G.H., Timmermann, A., Damste, J.S.S., Brauer, A., Sigman, D.M., Cane, M.A., Verschuren, D., 2011. Reduced interannual rainfall variability in East Africa during the last ice age. Science 333, 743e747.