Late Quaternary stratigraphic analysis of the Lake Malawi Rift, East Africa: An integration of drill-core and seismic-reflection data
Introduction
The influence of the tropical continental regions on global heating and moisture transport is poorly understood. Continents disrupt ocean circulation and the position of the Inter-Tropical Convergence Zone (ITCZ), thus playing a crucial role in modulating global climate. In order to understand the continental response to orbitally-induced insolation variability, long continuous continental climate records are needed. Only recently have these records emerged, and with the development of continental lake drilling projects around the world, these new data sets are providing insights into the continental responses to global climate change (e.g. Williams et al., 1997, Fritz et al., 2004, Scholz et al., 2007, Scholz et al., 2011).
Marine tropical climate records have generally shown the dominance of precession forcing over orbital time scales (Pokras and Mix, 1987, Clemens et al., 1991, deMenocal, 1995). Eolian dust records offshore the coasts of West and East Africa demonstrate the influence of the high latitudes on African climate, but generally show that orbital precession controlled monsoonal climate throughout these intervals (deMenocal, 1995). Pokras and Mix (1987) observed the offshore transport of the freshwater diatom Melosira as an indication of African aridity, with spectral analysis of the data showing a strong 23 kyr signal. Clemens et al. (1991) documented strong coherence between Indian Ocean tropical monsoon tracers and the precession cycle. A subtropical continental climate record from the Pretoria Saltpan extends continuously to ∼ 200 ka, where the precession signal dominates the rainfall-derived record (Partridge et al., 1997). Climate modeling demonstrates that meridional and zonal heating gradients in the tropics are greatly affected by precessional forcing (Clement et al., 2004), and are amplified during periods of high eccentricity.
Prior to lake drilling, available tropical continental data sets were either short (e.g. Broecker et al., 1998, Johnson et al., 2002) or of low resolution (e.g. Hooghiemstra et al., 1993, Olsen and Kent, 1996). Prior efforts focusing on climate records from Lake Malawi, generated from sediment cores collected in the 1980s and 1990s (Finney and Johnson, 1991, Johnson et al., 2002, Filippi and Talbot, 2005), record a short but variable climate record over the past 25 ka. These studies, as well as several others, document dry East African climate during the well-known recent high-latitude cold intervals of the Last Glacial Maximum (LGM), the Younger Dryas, and the Little Ice Age (Gasse et al., 1989, Johnson et al., 1996, Johnson et al., 2002, Brown and Johnson, 2005, Filippi and Talbot, 2005, Powers et al., 2005, Castañeda et al., 2007, Felton et al., 2007, Russell and Johnson, 2007). Some East African climate records extend further back in time, including a core from Lake Tanganyika's Kavala Ridge with an estimated 100 kyr record (Scholz et al., 2003, Scholz et al., 2007, Burnett et al., 2011-this issue). In equatorial East Africa, punctuated records from the central Kenya rift extend over the last 175 ka (Trauth et al., 2001, Trauth et al., 2003). Several very long (> 1 Ma) discontinuous low-resolution records have been collected throughout Kenya, Ethiopia and Tanzania, as well (Trauth et al., 2005 and references therein). To further investigate East African climate within a context of orbital forcing, high-resolution continuous climate records are needed beyond a single precession cycle (23 ka). In order to understand these orbital forcing mechanisms, the 2005 Lake Malawi Drilling Project collected 623 m of continuous drill core, covering 2 drill sites, and collected a single drill core ∼ 390 m below the lake floor (Scholz et al., 2011-this issue).
Lake Malawi is a useful site for studying long-term (> 25 ka) climate change due to its thick sedimentary section (> 4 km) (Flannery, 1988, Specht and Rosendahl, 1989) and its great water depth (700 m maximum). This basin configuration probably prevented the lake from experiencing long periods of desiccation throughout the late Quaternary. Furthermore, it allows for the development of anoxic bottom waters beneath ∼ 200 m (Vollmer et al., 2005), which minimize bioturbation, and accordingly preserve annual varves over extended parts of the sediment record (Pilskaln and Johnson, 1991, Johnson et al., 2002).
Seismic-reflection data provide 2-D and 3-D perspectives of lake stratigraphy and allow assessments of depositional variability. Using the principles of seismic sequence stratigraphy, base-level history reconstructions can be developed (Vail et al., 1977, Scholz, 2001). Base-level change in rift basin lakes is controlled by lake-level fluctuations, tectonic subsidence and sediment supply. These changes are manifested in seismic-reflection data as unconformities such as onlap, downlap, and erosional truncation surfaces. However, the use of seismic sequence stratigraphic analysis can only document relative changes in base level, and provides no quantitative age information on sequence stratigraphic development. Whether a given sequence boundary is caused by eustatic (lake level) change, uplift/subsidence, sediment supply variability, and/or compaction (Vail et al., 1977, Reynolds et al., 1991) can be difficult to differentiate. This is especially true in lacustrine rift basins, where subsidence and lake-level fluctuations can occur more rapidly than on marine passive margins (Scholz, 2001).
Previous seismic-reflection studies of Lake Malawi provide evidence for high-magnitude base-level changes that impact the stratigraphic record. Deep basin-scale multichannel reflection profiles were collected during Project PROBE and were used to interpret basin-scale rifting processes and stratigraphy (Rosendahl, 1987, Versfelt and Rosendahl, 1989, Specht and Rosendahl, 1989, Flannery and Rosendahl, 1990, Scholz et al., 1990). Basin-wide stratigraphic unconformities observed in Lake Malawi and neighboring Lake Tanganyika are thought to have been induced by severe low lake stages (Scholz and Rosendahl, 1988, Finney et al., 1996, Scholz et al., 2003). High-resolution imaging of some of the lake's many delta systems, including abandoned delta deposits offshore of the Songwe (Buoniconti, 2000) and Dwangwa Rivers (Martin, 1997) were identified in water depths as much as 500 m below present lake level (BPLL), providing evidence for the lake's high sensitivity to tectonic and climatic change (Scholz, 1995c). Furthermore, evidence for cyclic stacking of sublacustrine fans offshore of the South Rukuru River is thought to have been controlled in part by lake level (Scholz, 1995c, Soreghan et al., 1999). Prior to scientific drilling in 2005, no coring programs had recovered samples deep enough for dating these ancient features observed in seismic profiles.
The objectives of this study are to further constrain the framework developed through interpretation of seismic-reflection data alone, and to gain an understanding of how orbital-scale climate variability in the tropics affects the temporal and spatial variability of seismic- and lithofacies within the Lake Malawi Rift Basin. We present geochemical and geophysical analyses to constrain the timing of sedimentological response to changes in lake level in Lake Malawi over the past ∼ 150 ka, and show how variations in orbital insolation force lake-level change within the basin. Initial reports of the upper ∼ 150 ka of the sedimentary section are found in Scholz et al. (2007), Cohen et al. (2007), and Brown et al. (2007). This paper presents a seismic stratigraphic synthesis of new and existing seismic-reflection data, seismic–drill-core correlations, and a lithostratigraphic summary of each drill site.
Section snippets
Tectonics and structure
The East African Rift System (EARS) is a north–south alignment of rift basins on the east side of Africa (Fig. 1), that defines the incipient Africa–Somalia plate boundary (Chu and Gordon, 1999). A continuum of extensional processes define the EARS, from rift-related seismicity and minor subsidence in southern Botswana (Wright et al., 2003) to fully developed sea floor spreading in the Afar region to the north (e.g. Yirgu et al., 2006). The EARS is divided into two structural branches (Fig. 1),
Seismic-reflection surveys
To develop a sequence stratigraphic framework for the entire Lake Malawi Rift Basin, we used over 10,000 km of seismic-reflection profiles collected over the past ∼ 20 yrs (Fig. 3). These data vary in scale from a low-resolution, basin-scale (6-second record length) multichannel data set collected in the 1980s by Project PROBE (Rosendahl, 1987, Scholz and Rosendahl, 1988, Flannery and Rosendahl, 1990), to high-resolution single-channel (2–3 s record length) data sets collected in the mid-90s and
Seismic sequence stratigraphy
Each of the three main structural basins in Lake Malawi exhibits unique stratal architecture and seismic facies characteristics due to contrasting structural geometry. There is evidence for considerable amounts of erosional truncation and stratigraphic thinning of sequences in each basin due to both rapid lake-level change and localized rift-related subsidence. Whereas shallow reflections and sequence boundaries can be identified across the basins, the deeper stratigraphy in the North and
Sedimentation patterns and depositional environments
The river systems in the Lake Malawi catchment enter the lake at a variety of structural settings (Fig. 2). Therefore, stratigraphic evidence for lake-level variability is manifested slightly differently in each basin. However, we develop a conceptual model based on the several patterns that we observe in all basins that can be related to changes in seismic amplitude on seismic-reflection profiles and lithologic variability of the drill core. During highstands, coarse-grained sedimentation is
Conclusions
- (1)
Seismic-reflection data from Lake Malawi reveal evidence for extreme low lake stages, manifested as coarse-grained seismic facies, such as lowstand deltas, fans, turbidites, and sublacustrine channels in as much as 500 m modern water depth. Erosional truncation surfaces in the Central Basin are in as much as 550 m modern water depth.
- (2)
We identify three sequences (I–III) in the North Basin and interpret them throughout the lake. Each sequence represents a complete lake-level cycle, where a basal
Acknowledgements
General contracting and barge modifications during drilling operations were carried out by the University of Rhode Island and Lengeek Vessel Engineering, Inc. We thank ADPS Ltd. and Seacore Ltd. for marine operations and drilling support, respectively, on the Viphya drilling vessel. DOSECC, Inc., the Geological Survey of Malawi, Malawi Department of Surveys, and Malawi Lake Services provided technical and logistical support. Drill-core processing and analysis were performed at the National Lake
References (92)
- et al.
Quaternary sedimentary history of Adana, Cilicia and Iskenderun Basins, northeast Mediterranean Sea
Marine Geology
(1992) - et al.
New evidence for a reduced water balance in East Africa during the Last Glacial Maximum: implication for model-data comparison
Quaternary Science Reviews
(2003) - et al.
Calibration of marine sediment core loggers for quantitative acoustic impedance studies
Marine Geology
(1999) - et al.
Vegetation response to glacial–interglacial climate variability near Lake Malawi in the southern African tropics
Palaeogeography, Palaeoclimatology, Palaeoecology
(2011) - et al.
Modeling of Pliocene–Pleistocene abyssal mudwaves using synthetic seismograms
Marine Geology
(1998) - et al.
Antiphasing between rainfall in Africa's Rift Valley and North America's Great Basin
Quaternary Research
(1998) Lake Malawi's response to “megadrought” terminations: Sedimentary records of flooding, weathering and erosion
Palaeogeography, Palaeoclimatology, Palaeoecology
(2011)- et al.
Tropical East African climate change and its relation to global climate: A record from Lake Tanganyika, Tropical East Africa, over the past 90+ kyr
Palaeogeography, Palaeoclimatology, Palaeoecology
(2011) The East African Rift System
Journal of African Earth Sciences
(2005)- et al.
Three-dimensional quantitative modeling of clinoform development
Marine Geology
(1999)