1
Supplementary Information
R
ecurring summer and winter droughts
from
4.2
-
3.97
thousand years ago
in
north India
Alena Giesche
a*
, David A. Hodell
a
, Cameron A. Petrie
b
, Gerald H. Haug
c
, Jess F. Adkins
d
, Birgit Plessen
e
,
Norbert Marwan
f
, Harold J. Bradbury
a
g
, Adam Hartland
h
, Amanda D. French
h
, Sebastian F. M.
Breitenbach
i
a
Godwin Laboratory for Palaeoclimate
Research, Department of Earth Sciences, University of
Cambridge, Cambridge CB2 3EQ, United Kingdom
b
Department of Archaeology, University of Cambridge, Cambridge CB2 3DZ, United Kingdom
c
Climate Geochemistry Department, Max Planck Institute for Chemistr
y, Mainz 55020, Germany
d
Division of Geological and Planetary Sciences, California Institute of Technology, Pasadena, CA 91125,
USA
e
Section Climate Dynamics and Landscape Evolution, Helmholtz Centre Potsdam, German Research
Centre for Geosciences GFZ, P
otsdam 14473, Germany
f
Potsdam Institute for Climate Impact Research (PIK), Potsdam 14473, Germany
g
Department of Earth, Ocean, and Atmospheric Sciences, University of British Columbia, Vancouver, BC,
Canada
h
Environmental Research Institute, School of
Science, Faculty of Science and Engineering, University of
Waikato, Hamilton 3216, New Zealand
i
Department of Geography and Environmental Sciences, Northumbria University, Newcastle upon Tyne
NE1 8ST, UK
* To whom correspondence should be addressed.
Corr
esponding Author
: Alena Giesche, Department of Earth Sciences, Cambridge CB2 3EQ, UK
Email:
alena.giesche@gmail.com
This file includes:
Supplementa
ry
Discussion
Dharamjali Cave study site
Climate and tectonics affecting stalagmite growth in Dharamjali Cave
Carbon isotopes and cave ventilation dynamics
Transition metals as a drip rate proxy
Supplementary
R
eferences
for Discussion
Figures S1
-
S
11
Table S1
-
S2
Supplementary
References for
Figures & Tables
1
Supplementa
ry
Discussion
Dharamjali Cave study site
Dharamjali Cave (29°31’28.9” N, 80°12’28.0” E) is located at 2162 m a.s.l
. near Deodar village in the lesser Himalayas
in Uttarakhand state, northern India. The cave is developed within the Middle Proterozoic (1300
-
1200 Ma) Sor
Formation, which includes underlying Sor Slate (interbedded slates and shales with argillaceous and d
olomitic
limestone) and the overlying Thalkedar Limestone (finely laminated dolomitic limestone), occasionally intruded by basic
igneous rocks (Nautiyal, 1990). The area above the cave is dominated by C
3
-
type vegetation (with a lower δ
13
C
signature compare
d to C
4
-
type), composed of dense forest including mainly oaks (
Quercus incana
),
Rhododendron
,
and
Pinus roxburghii
along with smaller shrubs and herbs on top of a relatively thin (20
-
30 cm) soil cover (Sanwal et
al., 2013).
The cave is 30 m long, made up
of two main chambers, with one narrow entrance at its highest point (
Supplementary
Fig. S
10
) (Sanwal et al., 2013). In winter, the cave’s geometry promotes replacement of warmer air inside the cave with
denser cold outside air, which enhances ventilation
and CO
2
exchange with the atmosphere. Conversely, in summer,
the narrow entrance and downward passage orientation supports formation of a cold
-
air lake (cold, dense air
accumulates in the cave) that suppresses ventilation, with ventilation then limited to
cold summer nights. Tectonics are
active in this region, and Dharamjali Cave has been previously investigated as a recorder of seismic events, because
it contains tilted stalagmites with regrowth on different axes, as well as evidence for multiple ceiling
collapses
(Rajendran et al., 2016).
Based on data collected in 2009, relative humidity in Dharamjali Cave ranges between 50 and 70%, and temperature
ranges from 8 to 15°C (Sanwal et al., 2013), supporting the notion of active ventilation. Rainfall from G
PCC data (Meyer
-
Christoffer et al., 2018) shows that rainfall at this location averages 225 mm during the winter half of the year (Nov
-
Apr), compared to 1475 mm in the summer half (May
-
Oct) (
Supplementary
Fig. S
3
). The cave is actively dripping
throughout the year, though maximum infiltration occurs during the summer season with drip rates of 20
-
30 drops/min
compared to 3
-
8 drops/min in the winter (Sanwal et al., 2013). The cave therefore is highly sensitive and responsive to
seasonal changes in
precipitation amounts. Today’s average winter monsoon precipitation at the cave location, c. 10%
of the annual total, barely exceeds potential evapotranspiration during the winter months.
Based on the Online Isotopes in Precipitation Calculator (OIPC) v3
.2 model database (Bowen and Revenaugh, 2003),
the average summer (Jun
-
Sep) rainfall, weighted by monthly precipitation amount, has a δ
18
O value of
-
9.1‰, a δD
value of
-
60.2‰, and d
-
excess of 12.3‰, distinct from the rest of the year (Oct
-
May), also weigh
ted by amount, with
δ
18
O =
-
7.5‰, δD =
-
48.4‰, and d
-
excess = 11.6‰ (
Supplementary
Fig. S
2
). Published modern Dharamjali Cave
dripwater δ
18
O ranges between
-
6.5‰ and
-
5.5‰ (Sanwal et al., 2013), which may reflect sampling during spring
months or evaporatio
n along the flow path. However, dripwater data collected by our team in 2006 ranges between
-
9.5‰ in July and
-
8.6‰ in October, which matches unevaporated precipitation. Because the amount of rainfall in just
the four months of June
-
September (c. 1300 mm)
vastly exceeds the rest of the year (<400 mm), and the rainfall from
these four months has significantly lower δ
18
O and δD values, the average annual water isotope composition would be
skewed towards more negative ISM values. A relatively stronger or longe
r summer monsoon would consequently also
result in more negative δ
18
O / δD, and vice versa. Finally, evaporation should be relatively weaker during the ISM
season, and cave temperatures are warmer, both resulting in more negative rain
-
and dripwater δ
18
O a
nd δD values.
This mixed annual rainfall composition and changes in cave conditions would be incorporated by any speleothem
growing in this location. Since the residence time of the water in the epikarst is quite short, growth layers in a speleothem
might
also reflect distinct seasonal changes, including summer and winter rainfall δ
18
O. However, modern potential
evapotranspiration is higher than precipitation for all winter months except January and February, resulting in minimal
effective contribution of w
inter rainfall to the dripwater signal under today’s conditions.
Climate and tectonics affecting stalagmite growth in Dharamjali Cave
Previously studied Dharamjali
Cave speleothems include DH
-
1, DH
-
3 and DH
-
4 (Rajendran et al., 2016; Kotlia et al.,
2018). In this study, DHAR
-
1 shows a change in calcite to aragonite mineralogy just before 4.2 ka BP that corresponds
to a shift in growth axis around 4.2 ka BP in sample
DH
-
4 (Rajendran et al., 2016). Possibly, the onset of
dr
ier
conditions
during the mid
-
late Holocene transition rerouted water in Dharamjali Cave and led to the growth of DHAR
-
1. During the
1100
-
year period after 4.2 ka BP, no growth axis shifts occur in D
HAR
-
1 (or in any of the other previously studied
stalagmites from Dharamjali Cave), and trace elements and carbon and oxygen stable isotopes likely serve as
tectonically
-
unperturbed climate proxies. After 3.1 ka BP, tectonic disturbances may have occurred:
in DHAR
-
1, two
growth axis shifts occur at 3.1 ka BP and 2.78 ka BP, and a distinct break is visible just after 2.55 ka BP. In the DH
-
3
sample, a growth axis shift at 2.78 ka BP and a break at 2.5 ka BP were also attributed to tectonic events (Rajendran
e
t al., 2016).
2
Carbon isotopes
and cave ventilation dynamics
Speleothem
δ
13
C can be influenced by multiple factors, including vegetation composition and activity, soil microbial
activity, organic carbon content in the soil and epikarst, soil temperature, host carbonate rock
δ
13
C, prior
carbonate/aragonite precipitation (and thus
local effective moisture regime), and CO
2
degassing in epikarst and cave
(e.g., Fohlmeister et al., 2020). The latter is further influenced by cave air pCO
2
level and cave ventilation, with lower
pCO
2
levels and stronger ventilation fostering CO2 degassing
and kinetic fractionation. All the factors act with varying
relative importance at a given cave, and force the
δ
13
C signal in the same direction (higher
δ
13
C values under drier
conditions, except soil respiration).
The
δ
13
C
proxy is complex and its main driver possibly changes throughout the
record. Overall, we favor the option that PAP
drives
δ
13
C
from 4.2 to 3.97 ka BP
when drought is the most severe
, and
that a combination of
shifts in
soil carbon
via
vegetation, ventilat
ion, and
CO
2
degassing (all primarily related to winter
precipitation and/or temperature) drive the
proxy thereafter.
Below, we expand on
how
cave
ventilation
might affect
the
δ
13
C values
. One
notable result
was that
δ
13
C did not
perfectly match the
δ
44
Ca proxy, which is thought to exclusively reflect PAP (e.g., Owen et al., 2016; Magiera et al.,
2019; de Wet et al., 2021). So, PAP does not seem to be the primary driver for
δ
13
C, at least after 3.97 ka BP.
The
δ
13
C
proxy
could still reflect a
long
-
term
soil carbon shift if we assume that
the PAP driver passes a threshold and
causes an abrupt change in the baseline
δ
13
C
values after 3.97 ka BP. Alternatively,
an increase in cave ventilation
due to a change in season length
could explain
the divergence in
δ
13
C from
δ
18
O and
δ
44
Ca. If the dry winter/cold
season is longer and ventilation is favored in winter, then proportionally more aragonite (with corresponding
ly higher
δ
13
C) could form during the colder months. Winter dripwater would either stem from the epikarst that is still saturated
from summer monsoon rainfall late into the fall (i.e., a scenario where the summer monsoon is both stronger but also
more seas
onally compressed into a shorter span of months), or the dripwater represents directly infiltrating winter
precipitation (which has a more positive
δ
18
O
signature
than summer precipitation at Dharamjali Cave,
Supplementary
Fig. S2a). In both cases,
active CO
2
degassing in a highly ventilated cave environment would promote both higher
δ
13
C
(
and
δ
18
O
)
. However, between 3.97 and 3.4 ka BP
there is
a trend towards decreasing
δ
18
O while
δ
13
C increases. This
suggests that the
δ
18
O signal may in fact be dam
pened or muted
-
with summer monsoon being comparatively so
strong (lower
δ
18
O) that even increased winter ventilation is not able to compensate (without a more intense summer
monsoon,
DHAR
-
1 would show
an increase in
δ
18
O). One caveat for the ventilation interpretation is that a longer dry
winter season is not the only option; in fact, simply colder winter temperatures or longer winters (even with more
wintertime precipitation) could also lead to an increase in
δ
13
C. The
independent drip rate proxies (transition metals
Ni/Ca, Zn/Ca, and Cu/Ca)
, however,
may shed some light on distinguishing these possibilities
.
I
ndeed
,
the
transition
metals
indicate a slower drip rate around 3.6 ka BP (Fig. 2A) that would support the inter
pretation of
δ
13
C related to
winter aridity rather than simply cooler temperatures.
Since the drip rate proxies could reflect either summer or winter
precipitation, the
δ
18
O
(ISM proxy) becomes a useful tool for distinguishing when drip rate changes reflect winter v
ersus
summer rainfall.
At Dharamjali Cave, the ventilation situation is understood as outlined here: the site experiences a pronounced wet
ISM season between Jun
e and September, and a dry season during the rest of the year (
225 mm
winter precipitation
is observed
from Nov
-
Apr
) (Kotlia et al. 2018, Sanwal et al. 2013) (
Supplementary
Fig. S2). Effective infiltration
(precipitation minus runoff and evapotranspiration
) is confined largely to the ISM season (
Supplementary
Fig. S2a). At
the same time, surface air temperature varies significantly on a seasonal scale (
Supplementary
Fig. S2b), with highest
temperatures recorded in the summer.
The cave itself opens on the s
houlder of a narrow valley and is a single
downward
-
oriented passage (ca. 28 degree slope) (
Supplementary
Fig. S11) that forms a cul de sac after ca. 50 m or
so. This geometry has important bearings on the ventilation regime of the cave.
Observations
(th
is study, and
Sanwal et al.
,
2013) show that drips are active year
-
round, with ca. 20
-
30 drips/minute
during the ISM season, and 3
-
8 drips/minute during the dry season. This drip behavior suggests a rapid response to
infiltration events, with a short (thou
gh not quantified) lag period. The ventilation regime of the cave has not been
monitored in detail, but given its geometry and local temperature conditions four scenarios can be deduced
(
Supplementary
Fig. S11).
A.
Summertime day ventilation (
Supplementary
Fig. S11a)
During summer (and ISM season) the daytime surface air temperature is higher than inside the cave. Together
with the cave’s geometry this results in the formation of a cold air lake that keeps cool air in the cave. This
situation h
inders active ventilation of the cave, and allows accumulation of CO
2
(which is also brought in with
active drips). Thus, high pCO
2
levels are expected during the wet summer season during the day. Under these
conditions (high pCO
2
levels and reduced cave/s
urface air exchange) it is likely that both overall CO
2
degassing and kinetic fractionation of carbon isotopes (i.e., preferential degassing of
12
C from dripwater into
cave air) is suppressed.
δ
13
C in speleothem carbonate growing in this scenario would be
relatively negative
(little enrichment).
3
B.
Summertime night ventilation (
Supplementary
Fig. S11b)
During summer (and ISM season) the nighttime surface air temperature is likely still higher than inside the
cave, although the difference between surface and ca
ve air temperature (
D
T) would be lower. This would
result in the retention of the cold air lake in the cave and/or partial ventilation by replacement of cool cave air
with slightly colder surface air (if nighttime air cools sufficiently). This
situation allows some limited ventilation
of the cave and possibly some removal of CO
2
(brought in with active drips). Thus, pCO
2
levels are expected
to be either stagnant or lower slightly during ISM season nights.
Under these conditions (constant or lowe
ring
p CO
2
levels and limited cave/surface air exchange) it is likely that both overall CO
2
degassing and kinetic
fractionation of carbon isotopes are limited, but at times active.
δ
13
C in speleothem carbonate growing in this
scenario would likely remain r
elatively negative, or possibly slightly enriched.
C.
Wintertime day ventilation (
Supplementary
Fig. S11c)
During the dry winter season the situation changes significantly at the cave site. Daytime outside temperatures
are colder and
D
T would be low, such tha
t sometimes the cave would be warmer, and sometimes colder than
the surface air. This would allow for some exhaling of the cave (if cave air is warmer) and removal of cave air
CO
2
. The drip rates are lower compared to the summer and less CO
2
is transported
with dripwater into the
cave. Combining the lower CO
2
supply and somewhat limited ventilation would allow for only limited CO
2
degassing and isotope fractionation. This would result in speleothem
δ
13
C values being mid
-
ranging (not too
kinetically enriched
).
D.
Wintertime night ventilation (
Supplementary
Fig. S11d)
During the dry winter nights cave air would be significantly warmer than the surface air, which would result in
rigorous exhaling of the warm (rising) cave air and inflow of cold surface air (i.e.,
strong ventilation). Since the
drip rates are low (Sanwal et al. 2013) during the dry season, under these conditions prolonged CO
2
degassing
from incoming drips (which hang longer on the soda straws/stalactites) combines with fast removal of CO
2
from the c
ave and therefore active kinetic fractionation. This means that
δ
13
C in speleothems would be
enriched.
Considering these four scenarios
,
it is unlikely that active ventilation would frequently fall together with times of high
drip rates. Importantly (considering the sampling resolution achievable in the stalagmite) we argue that
δ
13
C is lower
during the ISM season and higher during
the win
ter
season (we cannot resolve night/daytime differences in the
stalagmites).
Whether more carbonate is deposited during the dry or the wet season remains speculative,
due to a
lack
of
monitoring.
However,
there are no longterm shifts in growth rate of th
e speleothem between 4.2 and 3.1 ka BP,
and
faster
accumulation occurs
only
after 3.1 ka BP (
Supplementary
Fig. S4). With stronger ventilation during the dry
season,
one
could expect to observe faster saturation of the incoming dripwater and thus faster gr
owth. Alternatively, the growth
during the ISM season might outpace the dry season deposition due to higher drip rates and loading of dripwater with
dissolved inorganic carbon.
This
can only be resolved with seasonal monitoring of the cave.
Transition metals as a drip rate proxy
Transition metals such as zinc, copper, cobalt, and nickel have received increasing attention because of their tendency
to bind to organic ligands (Hartland et al., 2014; Hartland and Zitoun, 2018). These elements ha
ve predicted and
empirically derived distribution coefficients >1 in calcite (Lindeman et al.,
2022
), and this prediction also holds for
aragonite although experimental research is still largely lacking (Wang and Xu, 2001; Fairchild and Baker, 2012).
Becau
se the elements Zn
2+
, Cu
2+
, Co
2+
and Ni
2+
all have Dx >1, increased prior calcite and aragonite precipitation
during arid conditions would effectively remove a higher proportion of these components from dripwater and result in
relatively lower X/Ca ratios
in the speleothem. However, because of the ligand
-
binding characteristic of transition
metals, the X/Ca ratio in the speleothem is also controlled by the length of time that the water film resides on the
speleothem (Hartland and Zitoun, 2018). A lower drip
rate with longer residence time on the speleothem allows for
increased dissociation of the metals from the ligands. This means that higher X/Ca ratios can be found in the
speleothem under more arid conditions, despite the occurrence of PCP or PAP and a tr
ansition metal D
x
>1. The relative
importance of this process is still being investigated (H
ö
pker et al., 2021), and robust D
x
values are still lacking for
aragonite speleothems. Nevertheless, the PCA of trace elements in DHAR
-
1 (Fig. 3) suggests that some transition
metals (e.g., Cu
2+
, Ni
2+
, and Zn
2+
) may be a useful proxy of drip rate in aragonite stalagmites.
4
Supplementa
ry
References
for Discussion
1.
Nautiyal,
A. C.
Microfacies microfossils (organic
-
walled) in Middle Proterozoic Tejam Group of Kumaun Lesser
Himalaya and paleoenvironmental significance.
J. Palaeontol. Soc. India
,
35
, 177
-
187 (1990).
2.
Sanwal
,
J.
et al. Climatic variability in Central Indian Himalaya during the last
∼
1800 years: Evidence from a
high resolution speleothem record.
Quat. Int.
,
304
, 183
-
192 (2013).
3.
Rajendran
,
C. P.
et al. Stalagmite growth perturbations from the Kumaun Himalaya as potential earthquake
recorders.
J. Seismol.
,
20
, 579
-
594 (2016).
4.
Meyer
-
Christoffer,
A.
,
Becker,
A.
,
Finger,
P.
,
Schneider,
U.
,
Ziese,
M.
Data from “GPCC Climatology Version
2018 at 0.25°:
Monthly Land
-
Surface Precipitation Climatology for Every Month and the Total Year from Rain
-
Gauges built on GTS
-
based and Historical Data.” GPCC at DWD. Available at https://doi.org/
10.5676/DWD_GPCC/CLIM_M_V2018_025, Deposited 2018.
5.
Bowen,
G. J.
,
Revenaug
h,
J.
Interpolating the isotopic composition of modern meteoric precipitation.
Water
Resour. Res.
,
39
, (2003).
6.
Kotlia,
B. S.,
Singh,
A. K.,
Joshi,
L. M.,
Bisht,
K.
Precipitation variability over Northwest Himalaya from
∼
4.0
to 1.9 ka BP with likely impact on civilization in the foreland areas.
J. Asian Earth Sci.
,
162
, 148
-
159 (2018).
7.
Fohlmeister, J. Main controls on the stable carbon isotope composition of speleothe
ms.
Geochim.
Cosmochim. Acta
,
279
, 67
-
87 (2020).
8.
Magiera
, M. et al. Local and regional Indian Summer Monsoon precipitation dynamics during Termination II
and the Last Interglacial.
Geophys. Res. Lett.
,
46
, 12454
-
12463 (2019).
9.
Owen, R. A. et al. Calcium isotopes in caves as a proxy for aridity: Modern calibratio
n and application to the
8.2 kyr event.
Earth Planet. Sci. Lett.
,
443
, 129
-
138 (2016).
10.
Rawal, R. S., Pangtey, Y. P. S. Distribution and phenology of climbers of Kumaun in Central Himalaya, India.
Vegetatio
,
97
, 77
-
87 (1991).
11.
de Wet, C. B. et al. Semiquant
itative Estimates of Rainfall Variability During the 8.2 kyr Event in California
Using Speleothem Calcium Isotope Ratios.
Geophys. Res. Lett.,
48
, e2020GL089154, (2021).
12.
Hartland,
A.
,
Fairchild,
I. J.
,
Müller,
W.
,
Dominguez
-
Villar,
D.
Preservation of NOM
–
m
etal complexes in a
modern hyperalkaline stalagmite: Implications for speleothem trace element geochemistry.
Geochim.
Cosmochim. Acta
,
128
, 29
-
43 (2014).
13.
Hartland
A.
,
Zitoun,
R.
Transition metal availability to speleothems controlled by organic binding ligands.
Geochem. Perspect. Lett.
8
, 22
-
25 (2018).
14.
Lindeman,
I.
,
Hansen,
M.
,
Scholz,
D.
,
Breitenbach,
S.F.M.
,
Hartland,
A.
Effects of organic matter complexation
on partitioning of
transition metals into calcite: cave
-
analogue crystal growth experiments.
Geochim.
Cosmochim. Acta
,
317
, 118
-
137
(
2022
).
15.
Wang
,
Y.
,
Xu,
H.
Prediction of trace metal partitioning between minerals and aqueous solutions: A linear free
energy correlation approach.
Geochim. Cosmochim. Acta
65
, 1529
-
1543 (2001).
16.
Fairchild
,
I. J.
,
Baker,
A.
Speleothem Science: From Process to Past Environments. Wiley, Hoboken, UK
(2012).
17.
H
ö
pker
, S.
et al.
Partitioning of trace metals in cave (and cave
-
analogue) carbonate precipitates
–
towards a
quantitative hydrological proxy in stalagmites.
EGU General Ass
embly Conference Abstracts
,
EGU21
-
13057
(2021).
5
Supplementary Figures
Fig. S1.
Global map
of the proportion of winter:summer precipitation.
Global overview, including a box
encompassing the close
-
up of the Indus region, and
reconstructions for a dry (orange) and wet (blue) climate at 4.2 ka
BP (
numbered
records
are
listed
in Table S1).
Contour lines represent the ratio of IWM (Nov
–
Apr) to ISM (May
–
Oct)
precipitation based on the 0.25° GPCC v2018 dataset from 1951
–
2000 rain ga
uge data (Meyer
-
Christoffer et al., 2018).
6
Fig.
S2
.
Dharamjali Cave
climate
with
modern monthly precipitation
(blue bars)
(
a
)
from the 0.25° GPCC v2018
dataset from 1951
-
2000 rain gauge data (Meyer
-
Christoffer
et al., 2018), shown with OIPC model
-
based monthly δ
18
O
values (purple circles) (Bowen and Revenaugh, 2003) where circle size denotes the weighting based on monthly
precipitation amount. The average weighted δ
18
O of summer months between June and Septembe
r (dark purple
dashed line) is compared to the average weighted δ
18
O during the rest of the year between October and May (light
purple dashed line). The yellow shading indicates monthly potential evapotranspiration, calculated using the
Thornthwaite method
(Palmer and Havens, 1958).
In
(b)
,
monthly rainfall, minimum and maximum temperatures from
Pithoragarh (1981
-
1985) (dark red), modified after Rawal and Pangtey (1991), and cave air temperature (min & max)
(light red) observed by Sanwal et al. (2013). The
cave air is colder than surface air between roughly March and
November, and warmer between November and
the
end of February. This suggests that for most of the year, the cave
acts as
a
cold air trap.
Fig. S
3
.
Map of δ
18
O of precipitation over India
and surrounding regions, with rainfall contours.
Shading denotes
δ
18
O
precip
values for winter (February) and summer (July), respectively, from the OIPC model (Bowen and Revenaugh,
2003). Contours show winter (Nov
-
Apr)
(
a
)
and summer (May
-
Oct)
(
b
)
precipita
tion from the 0.25° GPCC v2018 dataset
from 1951
-
2000 rain gauge data (Meyer
-
Christoffer et al., 2018). The location of Dharamjali Cave (this study, red
triangle) is shown with other nearby records (orange triangles) including marine core 63KA (Giesche et
al., 2019),
Sahiya Cave stalagmite SAH
-
2 (Kathayat et al., 2017), Kotla Dahar Lake KOT
-
D (Dixit et al., 2014), and Mawmluh
Cave stalagmite KM
-
A (Berkelhammer et al., 2012).
7
Fig. S
4
.
DHAR
-
1 mean depth
-
age model (blue) with 95% confidence limits (red)
based on COPRA (Breitenbach et
al., 2012). Individual ages are plotted as circles with ±2σ error bars. Stalagmite mean growth rate is 90 μm/year
where age is >3.1 ka BP and 200 μm/year where age is <3.1 ka BP.