Expanded glaciers during a dry and cold Last Glacial Maximum in
equatorial East Africa
Meredith A. Kelly1, James M. Russell2, Margaret B. Baber1, Jennifer A. Howley1, Shannon E. Loomis2,
Susan Zimmerman3, Bob Nakileza4, and Joshua Lukaye5
1
Department of Earth Sciences, Dartmouth College, Hanover, New Hampshire 03750, USA
Department of Geological Sciences, Brown University, Providence, Rhode Island 02912, USA
3
Center for Accelerator Mass Spectrometry, Lawrence Livermore National Laboratory, Livermore, California 94550, USA
4
Mountain Resource Centre, Makerere University, Kampala, Uganda
5
Petroleum Exploration and Production Department, Ministry of Energy and Mineral Development, Entebbe, Uganda
2
ABSTRACT
Glaciers on the world’s highest tropical mountains are among the most sensitive components
of the cryosphere, yet the climatic controls that influence their fluctuations are not fully understood. Here we present the first 10Be ages of glacial moraines in Africa and use these to assess
the climatic conditions that influenced past tropical glacial extents. We applied 10Be surface
exposure dating to determine the ages of quartz-rich boulders atop moraines in the Rwenzori
Mountains (~1°N, 30°E), located on the border of Uganda and the Democratic Republic of
Congo. The 10Be ages document expanded glaciers ca. 23.4 and 20.1 ka, indicating that glaciers
in equatorial East Africa advanced during the global Last Glacial Maximum (ca. 26–19.5 ka).
A comparison of these moraine ages with regional paleoclimate records indicates that Rwenzori glaciers expanded contemporaneously with dry and cold conditions. Recession from the
moraines occurred after ca. 20.1 ka, similar in timing to a rise in air temperature documented
in East African lake records. Our results suggest that, on millennial time scales, past fluctuations of Rwenzori glaciers were strongly influenced by air temperature.
INTRODUCTION
While most middle- and high-latitude glacial
extents are influenced by summer air temperatures and, to a lesser extent, winter precipitation
(e.g., Oerlemans, 1994; Anderson and Mackintosh, 2006), the mass balance of tropical glaciers
is sensitive to changes in air temperature and to
a variety of climatic parameters including precipitation, humidity, and cloudiness (Kaser et
al., 2004; Mölg et al., 2006; Taylor et al., 2006a,
2006b; Thompson et al., 2009; Hastenrath,
2010). Understanding the climatic controls that
influence tropical glaciers is critical to inferring
climatic conditions from past glacial extents
(e.g., Mark et al., 2005) and to predicting tropical glacial response to future global warming.
The Rwenzori Mountains, located on the border of Uganda and the Democratic Republic of
Congo, host the most extensive glacial and moraine systems in Africa (Kaser and Osmaston,
2002) and provide a unique opportunity to apply
surface exposure dating using the cosmogenic
nuclide 10Be because they are predominantly
composed of Precambrian quartz-rich gneisses.
At least four prior glacial extents are marked
by moraine systems that are in multiple valleys
in the Rwenzori Mountains (Osmaston, 1989).
We targeted the Lake Mahoma Stage moraines
that H. Osmaston mapped and suggested were
deposited during the last glacial period (Livingstone, 1962; Osmaston, 1989). At the type
locality near Lake Mahoma, a kettle lake in the
Mubuku Valley on the eastern side of the mountains, a series of moraines marks former glacial
extents down to ~2070 m above sea level (asl)
(Fig. 1). At present, Rwenzori glaciers are restricted to elevations above ~4800 m asl, and
therefore the Lake Mahoma Stage moraines
represent a dramatic glacial response to past climatic conditions.
STUDY SITE AND METHODS
In the Mubuku Valley, the Lake Mahoma
Stage moraines include three moraines, which
we informally call, from oldest to youngest, the
Mahoma-3, Mahoma-2, and Mahoma-1 (Fig. 1).
The Mahoma-3 and Mahoma-2 moraines are
arcuate terminal moraines at ~3000 m asl that
were deposited by a glacier with a catchment
area likely located on the eastern side of Mount
Baker and Mount Luigi di Savoia that flowed
down the Kuruguta and Mubuku Valleys. The
Mahoma-2 moraine forms the southeastern margin of Lake Mahoma. The Mahoma-1 moraines
are a pair of lateral moraines that extend much
farther down the Mubuku Valley to an elevation
of ~2070 m asl. Glaciers from Kuruguta, Mubuku, and Bujuku Valleys merged to form the
Mahoma-1 extent. The right-lateral Mahoma-1
moraine crosscuts the Mahoma-2 moraine and
forms the northwestern margin of Lake Mahoma.
We obtained samples for 10Be dating from the
Mahoma-2 and Mahoma-1 moraines because of
their distinct stratigraphic relationship. Both moraines are prominent (~50–150 m high), sharpcrested ridges that host large (as much as 5 m3)
gneissic boulders (Fig. 2). The 10Be samples
were processed in the laboratory at Dartmouth
College (New Hampshire, USA) and measured
at the Center for Accelerator Mass Spectrom-
GEOLOGY, June 2014; v. 42; no. 6; p. 1–4; Data Repository item 2014184
|
doi:10.1130/G35421.1
|
etry at Lawrence Livermore National Laboratory (California, USA). We determined surface
exposure ages from 10Be concentrations using
the CRONUS-Earth online calculator (Balco et
al., 2008) and a production rate calibrated in a
low-latitude, high-altitude location (i.e., PQuel
[Quelccaya], 3.78 ± 0.09 atoms g–1 yr –1, with
time-invariant scaling after Lal, 1991; Stone,
2000; i.e., “St”; Kelly et al., 2013) (Fig. 1; Table
DR1 in the GSA Data Repository1). We do not
apply any corrections for the influence of snow
or vegetation cover or for boulder surface erosion. 10Be ages are shown with 1σ measurement
uncertainties. We interpret the 10Be ages of boulders on the moraines to represent the timing of
final moraine construction at the glacial margin.
We report 10Be ages in years ago (yr ago; i.e., prior to the date of sample collection in A.D. 2012)
and thousands of years ago (ka). In contrast, previously published radiocarbon ages described
herein and interpretations based on radiocarbon
dating are reported as thousands of calibrated yr
before present (i.e., prior to A.D. 1950; kyr B.P.).
RESULTS
We measured eight high-precision 10Be surface exposure ages of boulders on the Lake
Mahoma Stage moraines (Fig. 1; also see Table DR1). Four 10Be ages from the Mahoma-2
moraine range from 22,930 ± 440 yr ago to
24,040 ± 540 yr ago (10Be ages ± 1σ measurement uncertainties) with a mean age of 23,370
± 470 yr ago (arithmetic mean ± 1σ). Four 10Be
ages from the right-lateral Mahoma-1 moraine
range from 19,240 ± 370 yr ago to 20,520
± 700 yr ago (10Be ages ± 1σ measurement uncertainties) with a mean age of 20,140 ± 610 yr
ago (arithmetic mean ± 1σ).
In general, uncertainties in 10Be dating arise
from 10Be production rate and measurement
uncertainties as well as geological uncertainties such as boulder surface erosion, cover by
1
GSA Data Repository item 2014184, Table DR1
(10Be sample data and calculated 10Be surface exposure ages), is available online at www.geosociety
.org/pubs/ft2014.htm, or on request from editing@
geosociety.org or Documents Secretary, GSA, P.O.
Box 9140, Boulder, CO 80301, USA.
Published online XX Month 2014
GEOLOGY
©
2014 Geological
2014 | ofwww.gsapubs.org
America. For permission to copy, contact Copyright Permissions, GSA, or [email protected].
| JuneSociety
1
A
29°50'E
29°55'E
30°0'E
B
Bu
ju
ku
Mubuku Valley
Va
lle
y
UGANDA
0°25'N
0°25'N
DEMOCRATIC
REPUBLIC OF
CONGO
Mt. Speke
y
lle
Va
ku
u
b
Mu
Bigo Bogs
Bujuku
Valley
Lake Mahoma
Insets B, C
Mt. Stanley
1 km
Mt. Baker
N
Mubuku Valley
Mt. Luigi di Savoia
Lake Mahoma
0°20'N
3000 m
0°20'N
C
Bu
ju
Kuruguta Valley
ku
Mubuku Valley
Va
l
le
y
RZ-1
RZ-9
RZ-7,
RZ-8
RZ-2, RZ-3
ey
all
uV
RZ-4, RZ-5
k
u
3000 m
b
Mu
0°15'N
Lake Mahoma
0°15'N
29°50'E
Contour interval 200 meters
29°55'E
30°0'E
km
Elevation (meters)
2500
5000
0
2
4
6
8
Legend
10
Be age
Mahoma-1 moraine
Mahoma-2 moraine
Mahoma-3 moraine
Undifferentiated moraine
1 km
N
Mahoma-1
moraine
20,410±390 (RZ-1)
20,530±700 (RZ-7)
20,410±420 (RZ-8)
19,240±370 (RZ-9)
Mahoma-2
moraine
22,930±440 (RZ-2)
23,230±440 (RZ-3)
24,040±540 (RZ-4)
23,280±450 (RZ-5)
Figure 1. Location map of Rwenzori Mountains and 10Be ages of Lake Mahoma Stage moraines in the Mubuku and Bujuku Valleys.
A: Topographic map of the Rwenzori Mountains on the border of Uganda and the Democratic Republic of Congo showing locations
discussed in text, including the Kuruguta, Mubuku, and Bujuku Valleys, Lake Mahoma, and the Bigo Bogs. White rectangle shows
locations of B and C. B: Worldview2 (www.satimagingcorp.com/gallery-worldview-2.html) 0.5-m-resolution satellite image taken in
January 2012 showing the Mubuku and Bujuku Valleys located in the eastern part of Rwenzori Mountains. C: Same satellite image
showing Lake Mahoma Stage moraines, 10Be sample locations, and 10Be ages corresponding to sample locations. We have mapped
undifferentiated moraines based only on satellite imagery and currently have no chronological data to constrain their ages. 10Be
ages shown are in years ago (i.e., prior to date of sample collection in A.D. 2012) with 1σ measurement uncertainties.
sediment, snow or vegetation, and postdepositional boulder movement. The 10Be ages presented here are calculated using a 10Be production rate determined at a similarly low-latitude,
high-altitude location (Kelly et al., 2013) with
an estimated uncertainty of ~6% (cosmognosis.
wordpress.com). Measurement uncertainties are
<3.5% (Table DR1). The 10Be ages show excellent internal consistency on individual moraines
and suggest that the geological uncertainties are
small. For example, two boulders from <10 m
apart on the Mahoma-2 moraine (RZ-2 and
RZ-3) yield nearly identical ages of ~22,930
± 440 yr ago and 23,230 ± 440 yr ago (Figs. 1
and 2). Moreover, the 10Be ages are consistent
with the crosscutting relationship of the moraines showing that the Mahoma-1 moraine is
younger than the Mahoma-2 moraine. Our 10Be
moraine chronology is also consistent with a radiocarbon age of organic material in a sediment
core from Lake Mahoma (uncalibrated age is
14,750 ± 290 14C yr B.P.; Livingstone, 1962).
2
This material (comprising ~20 cm of organicrich sediment located 60 cm above gravelly till
at the base of the core) provides a minimumlimiting age on glacial retreat of 17.2–18.6 kyr
B.P. (2σ calibrated age range based on IntCal13;
Reimer et al., 2013).
DISCUSSION
Based on our 10Be chronology, we suggest
that glaciers in the Rwenzori Mountains advanced to near their maximum extent during the
last glacial period by ca. 23.4 ka. The next significant moraine set outside of the Lake Mahoma Stage moraines is ~1 km downvalley and is
estimated to have been deposited during a prior
glacial period, possibly during Marine Isotope
Stage 6 (Osmaston, 1989). While glaciers were
likely present in the Rwenzori throughout the
last glacial period, climatic conditions between
ca. 23.4 and 20.1 ka caused significant glacial
growth, with glaciers in the Kuruguta, Mubuku,
and Bujuku Valleys merging and advancing
downvalley to 2070 m asl, an elevation currently occupied by tropical montane forest. Subsequent retreat occurred shortly after ca. 20.1 ka
and may have been rapid. The next significant
moraine set located upvalley of the Lake Mahoma Stage moraines is near the Bigo Bogs,
~7 km up the Bujuku Valley at ~3400 m asl, and
is estimated to be Holocene in age (Osmaston,
1989; Fig. 1).
The only other locations in Africa where moraines have been dated directly are in Kenya
and Tanzania on Mounts Kenya and Kilimanjaro, respectively. Here, the basaltic bedrock
types preclude the application of 10Be dating,
and 36Cl surface exposure dating was applied
to determine moraine ages. On Mount Kenya,
36
Cl ages of boulders on presumed Last Glacial
Maximum (LGM; i.e., Liki II; Mahaney, 1990)
moraines in the Gorges Valley range from 24
± 1 ka to 32 ± 2 ka, and boulders on Liki II deposits in the Teleki Valley yielded 36Cl ages of
ca. 19.2–135 ka (Shanahan and Zreda, 2000).
www.gsapubs.org
|
June 2014
|
GEOLOGY
B
Mahoma-1 moraines
20,410±390
Sample
(panel B)
C
Mahoma-2 moraine
D
22,930±440
23,230±440
Figure 2. Photos of Lake Mahoma Stage moraines and boulders that were sampled and 10Be
dated. A: Right- and left-lateral Lake Mahoma Stage moraines in Mubuku Valley. B: Very large
boulder (sample RZ-1) on the right-lateral Lake Mahoma Stage moraine in Mubuku Valley. 10Be
age of the boulder is shown in years ago with 1σ measurement uncertainties. White circle
marks standing person for scale. C: Lake Mahoma Stage (Mahoma-2) moraine that bounds
Lake Mahoma. D: Two smaller boulders (white arrows; samples RZ-2 and RZ-3) on the Mahoma-2 moraine yield very similar 10Be ages. See Figure 1 for moraine and sample locations.
On Kilimanjaro, 36Cl ages of boulders on the
Fourth or Main Glaciation moraines range from
16.4 ± 0.7 ka to 103 ± 5 ka on Mawenzi Peak,
from 13.6 ± 0.6 ka to 23 ± 1 ka on the Saddle,
and from 12.0 ± 0.4 ka to 28 ± 1 ka on Kibo
Peak (Shanahan and Zreda, 2000). Shanahan
and Zreda (2000) generally attributed the high
variability of the ages to either the influence of
nonuniform erosional processes or the presence
of 36Cl inherited from a prior period of exposure.
Although rejection of some ages that are clearly
outliers yields moraine ages similar to the global
LGM (i.e., 20 ± 1 ka for the Main Glaciation
moraines on Mawenzi Peak), the scatter in these
36
Cl ages makes comparison with other paleoclimate proxy data difficult. In contrast, the much
more tightly defined ranges of our 10Be ages
from the Rwenzori Mountains readily facilitate
comparison with other paleoclimate proxy data
to assess the climatic conditions that influenced
past glacial extents.
To infer the climatic conditions that led to
the Rwenzori glacial fluctuations, we compare
the Lake Mahoma Stage moraine ages with regional paleoclimate records. Virtually all of the
African Great Lakes were nearly or completely
desiccated during the LGM (Johnson et al.,
1996; Beuning et al., 1997; Gasse, 2000; Russell et al., 2003). Lake Albert, at the northern
end of the Rwenzori, was reduced to a shallow,
GEOLOGY
|
June 2014
|
www.gsapubs.org
swampy lake surrounded by an arid grassland
between ca. 35 and 21.5 kyr B.P., with complete desiccation between 21 and 15 kyr B.P.
(Beuning et al., 1997). To the south, lowstand
deltas in Lake Edward indicate a 65% reduction in lake volume during the LGM relative to
present (McGlue et al., 2006). Lakes Albert and
Edward both receive substantial runoff from the
Rwenzori Mountains and therefore respond at
least in part to hydrological changes within the
mountains. To the east, aridity is marked by desiccation surfaces in Lake Victoria that suggest
a reduction in precipitation of as much as 30%
during the LGM relative to present (Johnson et
al., 1996). To the west, geochemical data in the
Congo Basin suggest an arid LGM (Weijers et
al., 2007). While there is limited evidence for
relatively moist conditions elsewhere in tropical Africa during the LGM, the records from
near the Rwenzori Mountains clearly indicate
that the glacial advances to the Mahoma-2 and
Mahoma-1 moraines, just prior to ca. 23.4 ka
and ca. 20.1 ka, respectively, occurred during
regionally dry climatic conditions.
Paleotemperature estimates have been sparse
in tropical Africa due to the confounding effects
of precipitation on many temperature proxies.
New organic geochemical proxies based upon
the relative abundances of isoprenoidal and
branched glycerol dialkyl glycerol tetraethers
[i.e., TEX86 (tetraether index of tetraethers with
86 carbon atoms) and MBT-CBT (methylation index of branched tetraethers–cyclization
index of branched tetraethers) proxies] have
substantially improved our understanding of
late Quaternary temperature changes in Africa. Reconstructions from Lake Malawi (Powers
et al., 2005), Lake Tanganyika (Tierney et al.,
2008), Sacred Lake on Mount Kenya (Loomis
et al., 2012), and the Congo Basin (Weijers et
al., 2007) indicate that LGM temperatures were
3–5 °C cooler than at present (Fig. 3), compatible with paleotemperature reconstructions
from fossil pollen in the Burundi highlands
(Bonnefille et al., 1990) that indicate 4 ± 2 °C
cooling. In addition, the organic geochemical records from East Africa (i.e., Tanganyika,
Malawi, and Sacred Lakes) show the onset of
deglacial warming between 21 and 20 kyr B.P.,
contemporaneous with glacial recession from
the Mahoma-1 moraines (Fig. 3). The combined
Rwenzori moraine chronology and paleoclimate
records thus suggest that the LGM in equatorial
Africa was cold enough to drive significant glacial advances despite a decrease in precipitation.
Tropical paleotemperatures are also commonly inferred from past glacial extents using
former glacial equilibrium line altitudes (ELAs)
and an assumed adiabatic lapse rate. ELAs inferred from Lake Mahoma Stage glaciers in
Mubuku Valley are ~3800–4000 m asl and are
lower than recent (A.D. 1965) ELAs by ~800–
1000 m (Osmaston, 1989; Kaser and Osmaston,
2002; Mark et al., 2005). Present-day ELAs are
likely much higher than those determined in
A.D. 1965. Assuming an adiabatic lapse rate of
~5.5 °C/km in the Rwenzori Mountains (Eggermont et al., 2010), Lake Mahoma Stage gla-
Mahoma
Glaciation
30 Malawi
Temperature (oC)
A
25
Tanganyika
20
Sacred Lake
15
10 Mahoma Moraine Ages
0
5
10
15
20
25
30
k.y. BP
Figure 3. Quantitative paleotemperature reconstructions based on TEX86 (see text) from
Lake Malawi (dashed line; Powers et al.,
2005) and Lake Tanganyika (Tierney et al.,
2008) and on branched glycerol dialkyl glycerol tetraethers from Sacred Lake, Kenya
(Loomis et al., 2012). Gray bar shows time
of deposition of Mahoma-2 and Mahoma-1
moraines, indicating that these were deposited during cold conditions in equatorial
Africa. The y-axis value of the moraine ages
plotted is arbitrary.
3
ciers were influenced by temperatures at least
4–6 °C lower than at present. Mark et al. (2005)
inferred cooling of ~3–8 °C from ELAs associated with Lake Mahoma Stage glaciers. Due to
a drier atmosphere during the LGM, the lapse
rate and thus the magnitude of cooling inferred
from LGM ELAs may have been larger; however, there are no data independent of the glaciers
to quantify past changes in the lapse rate. The
highest elevation organic geochemical proxy
record from Africa (i.e., from Sacred Lake at
~2400 m asl on Mount Kenya) indicates temperatures ~5 °C lower than at present at the end
of the last glacial period (Loomis et al., 2012).
Thus, the amplitude of LGM cooling inferred
from glacial ELAs is compatible with organic
geochemical proxy records.
CONCLUSIONS
10
Be surface exposure dating of quartz-rich
boulders on glacial moraines in the Rwenzori
Mountains yields ages that show excellent internal consistency and are in agreement with a
previously published radiocarbon age. Eight
10
Be ages of Lake Mahoma Stage moraines indicate deposition during the LGM. These data,
combined with dry and cold LGM climatic conditions inferred from regional paleoclimate proxies, strongly suggest a direct correlation between
Rwenzori Mountains glacial extents and temperature on millennial time scales. Future research
should focus on applying 10Be dating to the
extensive glacial moraine system in the Rwenzori Mountains to examine questions about the
relationship of the LGM to older glaciations, the
structure of the LGM and subsequent deglaciation, as well as Holocene glacial extents.
ACKNOWLEDGMENTS
This research was funded by grants from the National Geographic Society (grant 8886-11) and the
National Science Foundation (grant EAR-122705).
We thank the guides and porters at Rwenzori Mountain Services and Rwenzori Trekking Services in
Uganda for essential field logistics. The Ugandan
Wildlife Authority and the Ugandan Council on Science and Technology provided permits to access the
region and obtain samples. We thank G. Balco for help
in calculating 10Be ages and J. Chipman for assistance
with satellite imagery and Figure 1. This is LLNLJRNL-651669.
REFERENCES CITED
Anderson, B., and Mackintosh, A., 2006, Temperature
change is the major driver of late-glacial and Holocene glacier fluctuations in New Zealand: Geology, v. 34, p. 121–124, doi:10.1130/G22151.1.
Balco, G., Stone, J.O., Lifton, N.A., and Dunai, T.J.,
2008, A complete and easily accessible means of
calculating surface exposure ages or erosion rates
from 10Be and 26Al measurements: Quaternary
Geochronology, v. 3, p. 174–195, doi:10.1016/j
.quageo.2007.12.001.
Beuning, K.R.M., Talbot, M.R., and Kelts, K.R., 1997,
A revised 30,000-year paleoclimatic and paleo-
4
hydrologic history of Lake Albert, East Africa:
Palaeogeography, Palaeoclimatology: Palaeoecology, v. 136, p. 259–279, doi:10.1016/S0031
-0182(97)00034-5.
Bonnefille, R., Roeland, J.C., and Guiot, J., 1990,
Temperature and rainfall estimates for the
past 40,000 years in equatorial Africa: Nature,
v. 346, p. 347–349, doi:10.1038/346347a0.
Eggermont, H.E., Heiri, O., Russell, J.M., Vuille, M.,
Audenaert, L., and Verschuren, D., 2010, Paleotemperature reconstruction in tropical Africa
using fossil Chironomidae (Insecta: Diptera):
Journal of Paleolimnology, v. 43, p. 413–435,
doi:10.1007/s10933-009-9339-2.
Gasse, F., 2000, Hydrological changes in the African
tropics since the Last Glacial Maximum: Quaternary Science Reviews, v. 19, p. 189–211,
doi:10.1016/S0277-3791(99)00061-X.
Hastenrath, S., 2010, Climatic forcing of glacier thinning on the mountains of equatorial East Africa:
International Journal of Climatology, v. 30,
p. 146–152.
Johnson, T.C., Scholz, C., Talbot, M.R., Kelts, K.R.,
Ricketts, R.D., Ngobi, G., Beuning, K.R.M.,
Ssemmanda, I., and McGill, J.W., 1996, Late
Pleistocene desiccation of Lake Victoria and
rapid evolution of cichlid fishes: Science, v. 273,
no. 5278, p. 1091–1093, doi:10.1126/science
.273.5278.1091.
Kaser, G., and Osmaston, H., 2002, Tropical glaciers: Cambridge, UK, Cambridge University
Press, 207 p.
Kaser, G., Hardy, D., Mölg, T., Bradley, R.S., and
Hyera, T.M., 2004, Modern glacier retreat on
Kilimanjaro as evidence of climate change:
Observations and facts: International Journal
of Climatology, v. 24, p. 329–339, doi:10.1002
/joc.1008.
Kelly, M.A., Lowell, T.V., Applegate, P.J., Phillips,
F.M., Schaefer, J.M., Smith, C.A., Kim, H., and
Leonard, K.C., 2013, A locally calibrated, late
glacial 10Be production rate from a low-latitude,
high-altitude site in the Peruvian Andes: Quaternary Geochronology, doi:10.1016/j.quageo
.2013.10.007.
Lal, D., 1991, Cosmic-ray labeling of erosion surfaces: In situ nuclide production rates and erosion models: Earth and Planetary Science Letters, v. 104, p. 424–439, doi:10.1016/0012-821X
(91)90220-C.
Livingstone, D.A., 1962, Age of deglaciation in the Ruwenzori Range, Uganda: Nature, v. 194, p. 859–
860, doi:10.1038/194859b0.
Loomis, S.E., Russell, J.M., Ladd, B., Street-Perrott,
F.A., and Sinnighe-Damsté, J.S., 2012, Calibration and application of the branched GDGT temperature proxy on East African lake sediments:
Earth and Planetary Science Letters, v. 357–358,
p. 277–288, doi:10.1016/j.epsl.2012.09.031.
Mahaney, W.C., 1990, Ice on the equator: Quaternary
geology of Mount Kenya: Sister Bay, Wisconsin, WM Caxon Ltd., 386 p.
Mark, B.G., Harrison, S.P., Spessa, A., New, M., Evans, D.J.A., and Helmens, K.F., 2005, Tropical
snowline changes at the Last Glacial Maximum: A global assessment: Quaternary International, v. 138–139, p. 168–201, doi:10.1016/j
.quaint.2005.02.012.
McGlue, M., Scholz, C., Karp, T., Ongodia, B., and
Lezzar, K.E., 2006, Facies architecture of flexural
margin lowstand delta deposits in Lake Edward,
East African rift: Constraints from seismic reflection imaging: Journal of Sedimentary Research,
v. 76, p. 942–958, doi:10.2110/jsr.2006.068.
Mölg, T., Rott, H., Kaser, G., Fischer, A., and Cullen,
N.J., 2006, Recent glacial recession in the Rwenzori Mountains of East Africa due to rising air temperature: Comment: Geophysical Research Letters, v. 33, L20404, doi:10.1029/2006GL027254.
Oerlemans, J., 1994, Quantifying global warming
from the retreat of glaciers: Science, v. 264,
p. 243–245, doi:10.1126/science.264.5156.243.
Osmaston, H., 1989, Glaciers, glaciations, and equilibrium line altitudes on the Rwenzori, in Mahaney, W.C., ed., Quaternary and environmental
research on East African Mountains: Rotterdam,
Balkema, p. 31–104.
Powers, L.A., Johnson, T.C., Werne, J.P., Castaneda,
I.S., Hopmans, E.C., Sinninghe-Damste, J.S., and
Schouten, S., 2005, Large temperature variability in the southern African tropics since the last
glacial maximum: Geophysical Research Letters,
v. 32, L08706, doi:10.1029/2004GL022014.
Reimer, P.J., and 29 others, 2013, IntCal13 and Marine13 radiocarbon age calibration curves
0–50,000 years cal BP: Radiocarbon, v. 55,
p. 1869–1887, doi:10.2458/azu_js_rc.55.16947.
Russell, J.M., Johnson, T.C., Kelts, K.R., Lærdal, T., and
Talbot, M.R., 2003, An 11,000-year lithostratigraphic and paleohydrologic record from equatorial Africa: Lake Edward, Uganda-Congo:
Palaeogeography, Palaeoclimatology, Palaeoecology,v.193,p.25–49,doi:10.1016/S0031-0182
(02)00709-5.
Shanahan, T., and Zreda, M., 2000, Chronology of
Quaternary glaciations in East Africa: Earth
and Planetary Science Letters, v. 177, p. 23–42,
doi:10.1016/S0012-821X(00)00029-7.
Stone, J.O., 2000, Air pressure and cosmogenic isotope production: Journal of Geophysical Research, v. 105, p. 23753–23759, doi:10.1029
/2000JB900181.
Taylor, R.G., Mileham, L., Tindimugaya, C., Majugu,
A., Muwanga, A., and Nakileza, B., 2006a, Recent glacial recession in the Ruwenzori Mountains of East Africa due to rising air temperature:
Geophysical Research Letters, v. 33, L10402,
doi:10.1029/2006GL025962.
Taylor, R.G., Mileham, L., Tindimugaya, C., Majugu,
A., Muwanga, A., and Nakileza, B., 2006b, Recent glacial recession in the Ruwenzori Mountains of East Africa due to rising air temperature:
Reply: Geophysical Research Letters, v. 33,
L20405, doi:10.1029/2006GL027606.
Tierney, J.E., Russell, J.M., Huang, Y., SinningheDamste, J.S., Hopmans, E.C., and Cohen, A.S.,
2008, Northern Hemisphere controls on tropical
southeast African climate during the last 60,000
years: Science, v. 322, p. 252–255, doi:10.1126
/science.1160485.
Thompson, L.G., Brecher, H.H., Mosley-Thompson,
E., Hardy, D.R., and Mark, B.G., 2009, Glacier
loss on Kilimanjaro continues unabated: National Academy of Sciences Proceedings, v. 106,
p. 19770–19775, doi:10.1073/pnas.0906029106.
Weijers, J.W.H., Schefuß, E., Schouten, S., and Sinninghe Damsté, J.S., 2007, Coupled thermal and
hydrological evolution of tropical Africa over
the last deglaciation: Science, v. 315, p. 1701–
1704, doi:10.1126/science.1138131.
Manuscript received 29 December 2013
Revised manuscript received 14 March 2014
Manuscript accepted 19 March 2014
Printed in USA
www.gsapubs.org
|
June 2014
|
GEOLOGY