Geoscience Frontiers 10 (2019) 2251e2263

H O S T E D BY Contents lists available at ScienceDirect

China University of Geosciences (Beijing)

Geoscience Frontiers
journal homepage: www.elsevier.com/locate/gsf

Research Paper

Solar insolation driven periodicities in southwest monsoon and its impact

on NE Arabian Sea paleoceanography
Syed Azharuddin a, Pawan Govil a, *, Arun Deo Singh b, Ravi Mishra c, Mayank Shekhar a
a
Birbal Sahni Institute of Palaeosciences, Lucknow, U.P., 226007, India
b
Department of Geology, Banaras Hindu University, Varanasi, 221 005, India
c
National Centre for Antarctic and Ocean Research, Vasco-da-Gama, Goa, 403 804, India

a r t i c l e i n f o a b s t r a c t

Article history: In the northeastern (NE) Arabian Sea, the fluctuation in terrestrial and freshwater runoff directly depends
Received 20 August 2018 on southwest monsoon (SWM) precipitation as well as the meltwater flux provided by Indus River.
Received in revised form Therefore, analysis of multi-proxy spectral signatures was carried out to trace the high-resolution SWM
16 January 2019
periodicities and their influence on the productivity, regional sea level fluctuations and depositional
Accepted 9 March 2019
Available online 10 April 2019
processes in the NE Arabian Sea. The time series data of stable isotopes of oxygen (d18OG.ruber), carbon
Handling Editor: E. Shaji (d13CG.ruber and d13Corg) and nitrogen (d15N), Total Organic Carbon (TOC), planktic-benthic foraminiferal
ratio (P/B ratio) and >63 mm coarse fraction (CF) were used from two coastal sedimentary cores located
Keywords:
offshore Saurashtra, NE Arabian Sea (Core SK-240/485 having 88 m water depth; Core GC/SK-240/496
Southwest monsoon having 174 m water depth). The REDFIT based spectral analysis recorded significant periodicities
Periodicities (>90% significance) in d18O time series centered at w1609, w667, w525, w296, w290 and w256 years.
Arabian sea Further, the significant periodicities recorded in carbon isotopes time series (d13CG.ruber and d13Corg)
Spectral analysis centered at w681, w512, w471, w452, w438, w360, w292, w275, w269, w245 and w209 years. The
Foraminifera significant periodicities in TOC include w471 and w322 years whereas d15N time series recorded sig-
nificant periodicity centered at w360 years. The significant periodicities in P/B ratio time series centered
at w512, w388, w304, w250, w235, w217, w152, w139 and w135 years while CF recorded w268,
w216, w209, w198, w188, w173 and w140 years significant periodicities. The observed periodicities in
the multi-proxy record consist of similar cycles (within the radiocarbon dating error) which also match
with previously reported solar insolation influenced SWM and other global and regional cycles. Further,
the stationarity of the data has been verified using wavelet analysis and shows similar periodicities as
observed in REDFIT analysis. Thereafter, the depositional behaviour was studied using correlation
analysis of the common periods of d18O time series of both the cores. The result suggests that the
depositional behaviour was different for both the core sites during the early Holocene and became
similar during the middle Holocene. The correlation analysis of Total Solar Index (TSI) with d18O time
series reveals a significant correlation with the core SK-240/485 whereas an insignificant correlation with
the core GC/SK-240/496. These observations suggest that the solar insolation has been a leading factor
responsible for the SWM trends during the Holocene which may have further influenced the produc-
tivity, regional sea level fluctuations and depositional conditions in the NE Arabian Sea. However, these
trends are better preserved in shallow marine sediments as compared to the deeper marine sediments.
Ó 2019, China University of Geosciences (Beijing) and Peking University. Production and hosting by
Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/
licenses/by-nc-nd/4.0/).

1. Introduction are predictable, whereas quasi-periodic and non-periodic varia-

tions are semi-predictable or may be unpredictable. Several natural
Climate variability may be periodic, quasi-periodic or non- events documented in the instrumental and paleoclimatic records
periodic (Ferronsky, 1992). The periodic variations in the climate indicate that a large part of the Northern Hemisphere has experi-
enced occurrences of both gradual and rapid climate change in the
* Corresponding author.
past (Bond et al., 1997; Alley et al., 2003). These climate fluctuations
E-mail addresses: [email protected], [email protected] (P. Govil). are forced by both internal and external mechanisms which drive
Peer-review under responsibility of China University of Geosciences (Beijing). on different frequencies over different time periods. On a millennial

https://doi.org/10.1016/j.gsf.2019.03.007
1674-9871/Ó 2019, China University of Geosciences (Beijing) and Peking University. Production and hosting by Elsevier B.V. This is an open access article under the CC BY-NC-
ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
2252 S. Azharuddin et al. / Geoscience Frontiers 10 (2019) 2251e2263

timescale, the Holocene climate of the Northern Hemisphere was archives regarding the periodicities of multiple orders from the
strongly influenced by seasonal variation in solar insolation northern Indian Ocean (Nigam and Nair, 1989; Tiwari et al., 2005;
(Budyko, 1969). This seasonal redistribution of solar energy was Rana and Nigam, 2009). Hence, a better understanding of these
responsible for the southern shift of Inter Tropical Convergence aspects could be a key to understand the present and future rainfall
Zone (ITCZ), leading to weak summer monsoon circulation in the patterns in the Indian subcontinent, which could further help in
Northern Hemisphere (Braconnot et al., 2007). However, the Ho- future economic planning. For this purpose, the spectral analysis
locene climate also fluctuated between warm to cold and humid to could play a vital role in providing climate variability information in
arid states on multi-centennial to multi-decadal timescales (Jung the paleoclimate data (Ghil et al., 2002). The spectral analysis may
et al., 2002). At higher frequency timescales, these climate oscilla- be inferred by using several methods viz. maximum entropy
tions appear to be associated with the El Niño-Southern Oscillation technique (Thomson, 1982), wavelet analysis (Torrence and Compo,
(ENSO) and North Atlantic Oscillation (NAO) variations in the in- 1998), PAST (Hammer et al., 2001) and REDFIT program (Schulz and
tensity of summer and winter ITCZ and fluctuations in monsoon Mudelsee, 2002). The spectral analysis helps in better under-
variability (e.g., Sirocko et al., 1996; Marshall et al., 2001; Wanner standing of the physical process, which is responsible for variability
et al., 2011; Marchitto et al., 2010). Even though significant in the time series (Schulz and Mudelsee, 2002). Therefore, in the
climate changes also occur at longer time scales as evident from the present study, we used REDFIT 4.1 program to analyze the spectra
palaeoclimatic records, the forcing mechanisms are still poorly of d18OG. ruber, d13CG. ruber, d13Corg, d15N, Total Organic Carbon (TOC),
understood, as well as the nature of primary or secondary cycle planktonic-benthic foraminifera ratio (P/B ratio) and >63 mm
variations is still debatable (Obrochta et al., 2012). The fundamental coarse fraction (CF) time series from two marine sedimentary cores
processes responsible for these climate variations include changes located offshore Saurashtra, NE Arabian Sea. The study aimed to
in solar activity and seawater flow dynamics, teleconnection of trace the Holocene periodicities in SWM and their influence on NE
Indo-Pacific sea surface temperature, ENSO and northern Hemi- Arabian Sea productivity.
sphere monsoon systems (Emile-Geay et al., 2008). Apart from the
NAO and ice-core records from Greenland, the monsoon system of 2. Material and methods
South Asia is also a crucial climate system, which regulates the sea
surface salinity and freshwater balance in the northern Indian The two marine sedimentary cores (SK-240/485 and GC/SK-240/
Ocean. It transports heat and moisture from the warmest part of the 496) were drilled from offshore Saurashtra, NE Arabian Sea during
tropical ocean across the equator and to higher latitudes. the Ocean Research Vessel (ORV) Sagar Kanya cruise in the year
Furthermore, in India, the southwest monsoon (SWM) precipi- 2007. The sediment samples of these two cores were obtained from
tation accounts for a significant portion of the total annual rainfall the sedimentary core repository at the National Centre for Antarctic
and affects the economic growth of the country (Saha et al., 1979; and Ocean Research (NCAOR), Goa, India. Cores SK-240/485
Gadgil et al., 2005). Therefore, understanding the forcing mecha- (21160 N, 68 550 59ʺE) and GC/SK-240/496 (21160 N, 68 340 03ʺE)
nism and their natural periodicities of the SWM precipitation is (Fig. 1) were recovered from 88 and 174 m water depths respec-
essential due to its broad impact on the socio-economic develop- tively (Table 1). The cores were sub-sampled at 2 cm interval to
ment in the country. There has been a considerable increase in achieve high temporal resolution. The chronology of both the cores
interest over the last few decades regarding the study of the rela- was prepared with the help of 14C Accelerator Mass Spectrometry
tionship between solar variability and global climate change (Bond (AMS) dating done at Radiocarbon Laboratory, Institute of Physics,
et al., 2001; Mauquoy et al., 2008; Gray et al., 2010). Early astro- Centre for Science and Education, Konarskiego, Gliwice, Poland and
nomical studies of the sun recorded dark spots on the photosphere, National Science Foundation (NSF) AMS facility, University of Ari-
which are called sunspots and dated back to the seventeenth cen- zona, U.S. Six AMS dates (including core-top and bottom) were
tury (Hoyt and Schatten, 1997). The study of long-term sunspots analysed using bulk sediments from the core SK-240/485, whereas,
revealed a periodic variation in solar activity with a mean cycle of four AMS dates (including core-top and bottom) were examined in
w11 years, which leads to a 0.1% change in solar constant. Also, the core GC/SK-240/496 using mixed planktic foraminifera (Fig. 2).
there is 0.24% increased solar output since Maunder Minimum The core SK-240/485 covers a time period between w12.5 and 1.5
(little or no solar activity occurred during 1675e1715) to present, kiloyear before present (or ka BP) whereas the core GC/SK-240/
which would have significantly influenced the global climate (Rind 496 covers the time period between w10 and 4 ka BP.
and Overpeck, 1993). These solar-modulated signals allow the The processing of sediment samples for the foraminiferal
reconstruction of solar variability and its influence on climate long extraction was done as described in Azharuddin et al. (2017). Clean,
before the beginning of historical sunspot observations (Beer et al., white and unbroken specimens of Globogerinoides ruber (G.ruber)
1996). Studies from a wide range of sedimentary and ice-core ar- (size fraction 250e350 mm) from both the cores were used for the
chives have investigated the role of solar forcing (solar activity) oxygen and carbon isotope (d18O and d13CG.ruber) analysis. For
through comparison of various proxy climate data (Stuiver et al., organic carbon isotopes (d13Corg), bulk sediments were treated with
1995; Steinhilber et al., 2012). Also, the study from off Karwar 1 N hydrochloric acid to remove carbonates before analysis. For
coast western continental shelf of India reveals w77 years peri- nitrogen isotopes (d15N), untreated bulk sediments were analyzed.
odicity in the monsoon which is suggested to be related to 80
10 The isotope analysis was done using Thermo Mat 253 Continuous
years Gleissberg solar cycle (Gleissberg, 1965; Nigam et al., 1995). Flow Isotope Ratio Mass Spectrometry (CFIRMS) facility at Birbal
Furthermore, a 2200 years periodicity in upwelling record from Sahni Institute of Palaeosciences (BSIP), Lucknow, India. The TOC
the western Arabian Sea since LGM is suggested to be governed by was measured by peak area obtained from the sum of the inte-
SWM and indicate that the oceanic circulation partly influences grated CO2 mass/charge (m/z) 44, 45 and 46 signal measured in the
monsoon strength at sub-Milankovitch cycles (Naidu and CFIRMS (Jensen, 1991). The P/B ratio was calculated using planktic
Malmgren, 1995). Several other studies show that the significant and benthic foraminifera abundances per gram.
monsoon fluctuations in the longer time scale are related to the The spectral analysis was done using REDFIT 4.1 computer
changes in Earth’s elliptical orbit around the Sun, i.e. eccentricity, program running on time series anomaly of isotopes (d18OG. ruber,
obliquity and precision (Croll, 1875; Milankovitch, 1941; Prell, 1984; d13CG.ruber, d13Corg and d15N), TOC, CF, and P/B ratio data to record the
Prell and Kutzbach, 1987; Clemens et al., 1991; Howard and Prell, significant periodicities. The computer program REDFIT (Schulz and
1992). These fluctuations are observed in various paleoclimatic Mudelsee, 2002) was used to perform the spectral analysis which is
 S. Azharuddin et al. / Geoscience Frontiers 10 (2019) 2251e2263 2253

Figure 1. Location map from the cores SK-240/485 and GC/SK-240/496 from offshore Saurashtra NE Arabian Sea. Source: Ocean Data View (Schlitzer, 2007).

Table 1 been performed between d18O time series of both the cores and
Details of the presently studied core. with the Total Solar Irradiance (TSI) (Steinhilber et al., 2009). The
Station No. Latitude Longitude Water depth (m) Core length (cm) cross-correlation and auto-correlation for the respective plots
GC/SK 240/485 21 160 N 68 550 590 E 88 340 were carried out using R graphics package (v. 3.1.2; R Core Team,
GC/SK 240/496 21 160 N 68 340 030 E 174 100 2014), and the figure is given in Supplementary Section (Fig. S1).
Further, the wavelet analysis was performed to understand vari-
ability in the d18O dataset of both the cores. The method involves
freely available via anonymous ftp from ftp.rz.uni-kiel.de (file:/pub/ the decomposition of a time-series into time-frequency space,
sfb313/mschulz/redfit35.zip). The standard parameters set for the which enables the determination of both the dominant modes of
analysis are the periodogram of the time series (n50 ¼ 1; rectan- variability and their temporal variations (Rayner et al., 2003;
gular window; c.f. Schulz and Stattegger, 1997) and Nsim ¼ 1000 Shekhar et al., 2017). The analysis was done using PAST software
Monte-Carlo simulations, the runs test indicates that the AR1 (Hammer et al., 2001).
model is indeed appropriate to characterize this record (90% sig-
nificance level). The palaeoclimate time series data was converted 3. Results
into anomaly by using Eq. (1) i.e., the raw data subtracted from
mean and divided by standard deviation. The anomaly or Z-scores In the present study, the spectra of oxygen and carbon isotopes
may be positive or negative, the positive value indicating the score of G. ruber (d18OG.ruber, d13CG.ruber) and sediment (d13Corg), isotopes of
is above the mean, and a negative score is showing it is below the nitrogen in bulk sediments (d15N), total organic carbon (TOC),
mean (See Supplementary Data). planktic-benthic foraminifera ratio (P/B ratio) and >63 mm coarse
fraction (CF) time series data were analysed from two marine
Z ¼ (x e m) /s (1) sedimentary cores located offshore Saurashtra, NE Arabian Sea
(Fig. 1).
where Z ¼ anomaly; m is the mean and s is the standard deviation of The application of foraminiferal stable oxygen and carbon iso-
the data respectively. Further, to analyze the significant trend in the topes in the field of palaeoceanography depends on the fraction-
data, the Mann-Kendall trend tests were performed using package ation of oxygen and carbon isotopes during the calcification of the
"Kendall" (McLeod, 2005) in the R programming environment foraminiferal shells in the ambient seawater. The d18O time series of
version 3.5.1 (R Core Team, 2014). the core SK-240/485 record the periodicities of w1609, w296 and
In addition, correlation, cross spectrum, and wavelet analysis w256 years (Fig. 3a) whereas the d18O time series of the core GC/
were also performed to strengthen the validity of results obtained SK-240/496 record 256 years periodicity (Fig. 3b). The periodic-
from REDFIT based spectral analysis. The correlation analysis has ities recorded in carbon isotopes of foraminifera G. ruber
2254 S. Azharuddin et al. / Geoscience Frontiers 10 (2019) 2251e2263

terrestrial biospheres (Guptha et al., 2005). Hence, the organic

carbon isotope (d13Corg & VPDB) is a robust proxy for the charac-
terization of organic matter deposited in the sediment. In the
present study, the d13Corg time series provides significant period-
icities of w452, w269, w245 and w209 years from the core SK-
240/485 (Fig. 4b) whereas the core GC/SK-240/496 provides the
significant periodicities of w438 and w360 years (Fig. 4d). The TOC
time series in the core GC/SK-240/496 provide the periodicities of
w471 and w333 years (Fig. 5).
The nitrogen isotope (d15N & Air) is a proxy to study the input/
output budget of fixed nitrogen and its circulation within the
ocean which ultimately depends on the bottom-water oxygena-
tion conditions (Sigman and Casciotti, 2001). In the present study,
the spectral analysis of d15N time series from the core GC/SK-240/
496 has been performed which recorded a periodicity of w360
years (Fig. 6).
The pattern of variation in planktic/benthic foraminiferal
abundance or P/B ratio in marine sediments generally follows the
Figure 2. Age-depth model of the studied cores prepared by using Calib 7.1 program bathymetric pattern with increased values towards increasing
(Stuiver and Reimer, 1993) and applying appropriate reservoir age correction of water depth (Van der Zwaan et al., 1990). Hence, the periodicities
129
35 years for Saurashtra coast (Dutta et al., 2001). Sedimentation rates corre-
in the P/B ratio time series data might be helpful in understanding
sponding to the dated intervals are also mentioned (modified after Azharuddin, 2019).
the regional paleobathymetric trends in the area. The significant
periodicities recorded in the P/B ratio time series data include
w512, w388, w304, w250, w235, w217, w152, w139 and w135
(d13CG.ruber) time series of core SK-240/485 include w512 and w275 years (Fig. 7).
years (Fig. 4a) whereas that in the d13CG. ruber time series of the core The variation in >63 mm coarse fraction content (CF) defines the
GC/SK-240/496 include w681, w471 and w291 years periodicities contribution of coarse (sandy) sediment in the core at a particular
(Fig. 4c). interval. Hence, the periodicities associated with the CF time series
The organic carbon cycle revolves around the CO2 fixation into might be helpful in understanding the pattern of the sediment
organic biomass through photosynthesis, in both marine and deposition/depositional environment. The significant periodicities

Figure 3. Results of spectral analysis of d18O time series of the cores. (a) Core SK-240/485; (b) Core GC/SK-240/496; (c) Core SK-240/485 (8e1 ka) and (d) Core SK-240/485
(12.4e8 ka).
 S. Azharuddin et al. / Geoscience Frontiers 10 (2019) 2251e2263 2255

Figure 4. Results of spectral analysis of (a) d13CG.ruber time series of core SK-240/485; (b) d13Corg time series in core SK-240/485; (c) d13CG.ruber time series in core GC/SK-240/496; (d)
d13Corg time series in core SK-240/496.

Figure 5. Results of spectral analysis of TOC time series in the core GC/SK-240/496. Figure 6. Results of spectral analysis of d15N time series in the core GC/SK-240/496.

The wavelet analysis of d18O of core SK-240/485 marks high

recorded in the CF time series include w268, w216, w209, w198, energy bands in and around 300 and 580 years extending between
w188, w173 and w140 years (Fig. 8). 1.4 and 3 ka (Fig. 9). A significant high energy band around 1250
The correlation analysis between d18O time series of common years is also visible in the early Holocene extending between 8 and
time periods from both the cores identified three different phases, 11 ka. Another significant high energy band in and around 520
i.e., phase 1 having a significant positive correlation (r ¼ 0.48572, years also appears to extend during early-Holocene between 10.5
n ¼ 13) whereas phases 2 and 3 show insignificant negative cor- and 11.7 ka. On the other hand, the d18O of core GC/SK-240/496 does
relation (r ¼ 0.18821, n ¼ 7; and r ¼ 0.08285, n ¼ 13 not record any significant high energy band in the wavelet analysis
respectively). (Fig. 10). Similarly, the wavelet-based periodicities of ~530, ~260
2256 S. Azharuddin et al. / Geoscience Frontiers 10 (2019) 2251e2263

inhibit the top w25 m of the sea surface of tropical to subtropical

waters (Be, 1977; Fairbanks et al., 1980; Hilbrecht, 1996). This spe-
cies prefers oligotrophic condition of the mixed layer thus abun-
dantly occurs when surface water is more stratified due to more
freshwater discharge in the sea. Thus, oxygen isotope signal of
G. ruber is likely to be related to the freshening of sea surface water
attributed to SWM related overhead precipitation and freshwater
discharge from the land. In the NE Arabian Sea, the freshwater
discharge is primarily related to Indus River, the main source of
which are the SWM rainfall and meltwater flux from Himalayan
glaciers (Milliman et al., 1984). Therefore, the d18O signatures
around the present shallow cores are directly influenced by the
freshwater input from the Indus River (Azharuddin et al., 2017).
Hence, it can be assumed that the variation in d18O signals of the
presently studied cores directly attribute to the fluctuation in SWM
strength. We used reconstructed total solar irradiance (TSI)
anomaly (Z-Score) for past 9300 obtained from the cosmogenic
nuclide 10B based ice-cores measurement (Steinhilber et al., 2009).
The TSI anomaly time series plotted with d18O time series anomaly
Figure 7. Results of spectral analysis of P/B foraminiferal ratio time series in the core of cores SK-240/485 and GC/SK 240/496 from NE Arabian Sea is
SK-240/485.
shown in Fig. 11. The statistical significance regarding trends and
magnitudes of all the three-time series anomaly were compared
using the non-parametric Mann-Kendall trend test (Mann, 1945;
Kendall, 1975). The results suggest significant decreasing trend in
TSI [s ¼ 0.17; p-value (two-tailed): <0.0001] whereas the d18O of
both the cores [SK-240/485: s ¼ 0.688; p-value (two-tailed):
<0.0001], [GC/SK-240/496: s ¼ 0.443; p-value (two-tailed):
<0.0001] show significant increasing trend (positive excursion)
during the Younger Dryas (YD) and Holocene (Fig. 11). These ob-
servations suggest the influence of solar insolation on the intensi-
fication of SWM. Previous studies from the Arabian Sea (Nigam and
Nair, 1989; Naidu and Malmgren, 1995; Nigam et al., 1995;
Agnihotri et al., 2002; Gupta and Mélice, 2003; Gupta et al., 2005,
2013; Tiwari et al., 2005; Thamban et al., 2007) and terrestrial re-
cords of India (Dutt et al., 2015; Sandeep et al., 2015; Warrier et al.,
2017) also suggested that the solar forcing is one of the vital factor
responsible for the fluctuations of SWM. Further, the spectral sig-
natures of oxygen isotopes (Fig. 3), carbon isotopes (Fig. 4), TOC
(Fig. 5), nitrogen isotopes (Fig. 6), P/B ratio (Fig. 7) and CF (Fig. 8)
time series have been performed to understand the impact of solar
insolation driven SWM periodicities on the Arabian Sea paleo-
Figure 8. Results of spectral analysis of the >63 mm coarse fraction (CF) time series in
ceanography regarding productivity, regional sea level fluctuation,
the core SK-240/485.
and depositional settings.

4.1. Periodicities in SWM

and ~380 years in other multi-proxy datasets appear to be common
with the periodicities based on REDFIT analysis of the same dataset.
Spectral analysis is a useful technique which helps in delineating
the factors forcing the monsoon and its cyclicity (Tiwari et al.,
4. Discussions 2005). Spectral analysis has been done on the oxygen isotope
(d18OG.ruber) records of both the studied cores (SK-240/485 and GC/
The periodicities recorded in the multi-proxy time series data do SK-240/496) using REDFIT 4.1 program. The oxygen isotope
not point to the same solar cycle number but show a close (d18OG.ruber) time series record of the core SK-240/485 exhibits three
resemblance with each other and with previously reported cycles. significant periodicities (significance >90%) of w256, w296 and
This uncertainty may arise due to the time lag in the response of w1609 years (Fig. 3a). However, the d18OG. ruber record of the core
different geochemical (isotopes) and biological (foraminifera GC/SK-240/496 exhibits a significant (significance >95%) period-
abundance ratio) proxies and the uncertainties in the radiocarbon icity of w256 years (Fig. 3b). The w256 years periodicity observed
dating. The radiocarbon error reported in the AMS dating of the from d18OG.ruber time series data of both the cores point towards the
studied sedimentary cores (SK-240/485 and GC/SK-240/496) sug- fact that SWM follows a quasi-periodicity of w210
50 years,
gest an average of w50 years of uncertainty. Tiwari et al. (2005) also which is similar to that of DeVries/Suess cycle (Usoskin and
proposed an uncertainty of w50 years in the periodicity data based Mursula, 2003; Raspopov et al., 2008). The spectral analysis of
on simulation experiments which could be a result of discrete Globogerinoides bulloides time series in the Hole 723A from the
sampling and radiocarbon uncertainties. western Arabian Sea (Gupta et al., 2013) reported statistically
Spectral analysis was performed on d18OG.ruber data of cores SK- stronger DeVries/Suess solar cycle of 208 years. The spectral anal-
240/485 and GC-SK 240/496 (Fig. 3). G. ruber is the surface- ysis of d18O data of G.ruber and G.menardii from SK145-9 (south-
dwelling planktic foraminifera species which predominantly eastern Arabian Sea) (Tiwari et al., 2005) reported periodicities of
 S. Azharuddin et al. / Geoscience Frontiers 10 (2019) 2251e2263 2257

Figure 9. Results of the wavelet power spectrum in the core SK-240/485. (a) d18O; (b) d13CG.ruber; (c) d13Corg; (d) > 63 mm coarse fraction (CF); (e) P/B foraminiferal ratio. The cross-
hatched region is the cone of influence. Black contour is the 95% significance level, using a red-noise (autoregressive lag 1).

Figure 10. Results of the wavelet power spectrum in the core GC/SK-240/496. (a) d18O; (b) d13CG.ruber; (c) d13Corg; (d) d15N; (e) Total Organic Carbon (TOC). The cross-hatched region
is the cone of influence. Black contour is the 95% significance level, using a red-noise (autoregressive lag 1).

w230, w215 and w210 years which also correspond to the Suess south India also reported similar periodicities of w220, w209 and
solar cycle. The magnetic susceptibility records from Arabian Sea w232 years respectively. These all cycles are also reported to
(Thamban et al., 2007), Pookot (PK) lake (Sandeep et al., 2015) and correspond the w205 years sunspot cycle (Solanki et al., 2004).
Thimmannanayakanakere (TK) lake (Warrier et al., 2017) from Similar cycles are also reported from the D14C record which is
2258 S. Azharuddin et al. / Geoscience Frontiers 10 (2019) 2251e2263

Figure 11. Holocene variation in (a) Total Solar Irradiance (TSI) anomaly time series (Steinhilber et al., 2009); (b) d18O anomaly time series in the cores SK 240/485 and (c) d18O
anomaly time series in the cores GC/SK-240/496.

ascribed to be originated from the solar influence (Stuiver and 240/485 also reveals two distinct periodicities of w667 and w290
Braziunas, 1993). Another periodicity of w296 years (>90% signif- years. The w667 years periodicity mentioned above matches with
icance) has been recorded in the d18OG. ruber time series of the core w640 years periodicity in magnetic susceptibility record of the
SK-240/485 (Fig. 3a). This w296-year periodicity may correspond Arabian Sea (Thamban et al., 2007), which is considered to be
to the w282 years cycle in the North Atlantic cold event recorded related to subtle variation in solar activity. The w290 years peri-
from the Chesapeake Bay, eastern North America which is attrib- odicity in the middle-late Holocene d18OG.ruber time-series of SK
uted to the variation in solar cycle (Willard et al., 2005). It is to be 240/485 (Fig. 3c) appears to be similar as w296 years periodicity of
noted that the w296 years periodicity was not observed in GC/SK- the d18OG.ruber time series of the entire Holocene (Fig. 3a, as dis-
240/496 record. It may be due to bathymetric condition at the core cussed above). The spectral analysis was also performed on Dongge
site around SK-240/485 being shallower as compared to other cave d18O data (Dykoski et al., 2005) to test the existence of these
deeper cores. Additionally, a cyclicity of w1609 years (>90% sig- SWM periodicities in the Himalayan region. The d18O time series of
nificance) is noted in the d18OG. ruber time-series of core SK-240/485 Dongge cave displays periodicities of w304 and w243 years
(Fig. 3a), which may correspond to the 1500
500 years Holocene (Fig. 12), which match to those recorded in the d18OG.ruber time se-
cycle as described by Bond et al. (1997, 2001). ries of the present study. These recorded periodicities also show
The spectral analysis of the d18OG.ruber time series of SK-240/485
representing the Younger Dryas (YD) and early Holocene (12.4e8
ka) (Fig. 3d) and middle-late Holocene (8e1 ka) (Fig. 3c) intervals
was performed separately in the core SK-240/485. The YDeearly
Holocene d18OG.ruber time-series spectral analysis reveals two
prominent periodicities of w945 and w525 years, which are not
found in Holocene spectral record (Fig. 3d). The w945 years peri-
odicity during YDeearly Holocene d18OG.ruber time series is similar
to the w950 and w906 years spectral signatures recorded in the
magnetic susceptibility data of the Arabian Sea (Thamban et al.,
2007) and TK lake (Warrier et al., 2017) respectively. These peri-
odicities have also been suggested to be of solar origin. Similar
periodicities have also been reported from the early Holocene
stalagmite records of Oman (Neff et al., 2001), northern Hemi-
sphere tree-ring record (Lean, 2002) and lake sediments record of
Alaska (Hu et al., 2003). Another w525 years periodicity in
YDeearly Holocene d18OG.ruber time-series match with solar influ-
enced w490 years periodicity observed in varved sediment record
from off Pakistan (Berger and Von Rad, 2002), w500 years peri-
odicity based on tree ring D14C (Lean, 2002) and w470 years
periodicity in magnetic susceptibility records from the eastern
Arabian Sea (Thamban et al., 2007) and TK lake (Sandeep et al., Figure 12. Results of spectral analysis of d18O time series data from Dongge cave, South
2015). The middleelate Holocene d18OG.ruber time-series of SK China (Dykoski et al., 2005).
 S. Azharuddin et al. / Geoscience Frontiers 10 (2019) 2251e2263 2259

Table 2
Proxy-wise periodicities recorded in the studied cores with their possible origins and references of similar periodicities recorded in global paleoclimatic
records.

Proxy (core) Recorded periodicities Possible origins Similar periodicities in other paleoclimatic records
(year)
d18O (both cores) 1609, 667, 525, Solar influences SWM cycle, Nigam and Nair, 1989; Stuiver and Braziunas, 1993;
296, 290 and 256 210 years DeVries/Suess Naidu and Malmgren, 1995; Nigam et al., 1995;
cycle of the Sun Bond et al., 1997, 2001; Neff et al., 2001;
Agnihotri et al., 2002; Berger and Von Rad, 2002;
Lean, 2002; Gupta and Mélice, 2003;
Usoskin and Mursula, 2003; Solanki et al., 2004;
Gupta et al., 2005, 2013; Tiwari et al., 2005;
Willard et al., 2005; Thamban et al., 2007;
Raspopov et al., 2008; Sandeep et al., 2015;
Warrier et al., 2017
d13CG.ruber and d13Corg 681, 512, 471, 452, Solar influences SWM cycle, Usoskin and Mursula, 2003; Baker et al., 2005;
(both cores) 438, 360, 292, 275, 210 years DeVries/Suess Gupta et al., 2005, 2013; Thamban et al., 2007;
269, 245 and 209 cycle of the Sun Raspopov et al., 2008; Rana and Nigam, 2009;
Warrier et al., 2017
d15N (GC/SK-240/496) 360 Solar influences SWM cycle Thamban et al., 2007
TOC (GC/SK-240/496) 471 and 322 Solar influences SWM cycle Willard et al., 2005
CF (SK-240/485) 268, 216, 209, 198, Solar influences SWM cycle, Agnihotri et al., 2002; Solanki et al., 2004; Gupta et al., 2005;
188, 173 and 140 210 years DeVries/Suess Thamban et al., 2007; Rana and Nigam, 2009;
cycle of the Sun Sandeep et al., 2015; Warrier et al., 2017
P/B ratio (SK-240/485) 512, 388, 304, 250, Solar influences SWM cycle, Agnihotri et al., 2002; Solanki et al., 2004;
235, 217, 152, 139 210 years DeVries/Suess Baker et al., 2005; Gupta et al., 2005;
and 135 cycle of the Sun Willard et al., 2005; Rana and Nigam, 2007;
Thamban et al., 2007; Sandeep et al., 2015;
Warrier et al., 2017

similarities with other terrestrial and marine records of the Arabian The d13CG.ruber is a proxy for surface water productivity, which is
Sea and Indian lakes (Nigam and Nair, 1989; Naidu and Malmgren, related to the degree of carbon sequestration occurred during
1995; Nigam et al., 1995; Agnihotri et al., 2002; Gupta and Mélice, photosynthesis. The d13CG.ruber record of the cores SK-240/485 de-
2003; Gupta et al., 2005, 2013; Tiwari et al., 2005; Thamban et al., picts the periodicities of w512 and w275 years (Fig. 4a). The w512
2007; Dutt et al., 2015; Sandeep et al., 2015; Warrier et al., 2017) years periodicity matches with w525 years periodicity in d18OG.ruber
within the range of proposed chronological errors (Table 2). Hence, time series of the core SK-240/485. The d13CG.ruber time series from
the recorded periodicities in the monsoonal proxy (d18O of planktic the core GC/SK-240/496 reflects the periodicities of w681, w471
foraminifera) of the present study suggest a linkage of SWM pre- and w292 years (Fig. 4c). Out of which, the w292 years periodicity
cipitation with the activity of solar cycle even though the exact matches with the w296 years cycle observed in d18OG.ruber time
linking mechanism is yet to be explained. The periodicities in SWM series of the core SK-240/485.
variability of the YDeearly Holocene period are different from The w681 years periodicity recorded in the d13CG.ruber time se-
those observed in the middle and late Holocene. It may be due to ries of the core GC/SK-240/496 matches with the w667 years
the difference in magnitude of solar irradiance during the different periodicity marked in d18OG.ruber time series of the core SK 240/485,
time intervals within the Holocene. which might be similar to w640 years magnetic susceptibility cycle
related to the solar variation (Thamban et al., 2007). The observed
4.2. Periodicities in carbon and nitrogen cycles and associated periodicities in d13Corg time series data from the SK-240/485
paleoproductivity include w452, w269, w245 and w209 years (Fig. 4b). The w452
years periodicity is similar to w471 years periodicity recorded in
Sediment trap studies reveal that the surface water productivity d13CG. ruber time series. These cycles may correspond to 434e454
in the NE Arabian Sea is a function of the seasonal southwest years bond like cycles recorded from South America (Baker et al.,
(summer) monsoon upwelling and northeast (winter) monsoon- 2005). The w245 and w209 years periodicities recorded in
induced convective mixing (Nair et al., 1989; Madhupratap et al., d13Corg time series of core SK-240/485 match with the w256 years
1996). The primary productivity is also responsible for the regula- periodicity marked in d18OG.ruber time series. These periodicities are
tion of the carbon cycle by the exchange of organic and inorganic reported to be similar with deVries/Suess solar cycle which points
carbon between the atmosphere and the deeper ocean (Archer, towards the pronounced effect of solar activity on the global
2003). Since the studied cores come from very shallow waters, climate processes (Usoskin and Mursula, 2003; Raspopov et al.,
therefore, the recorded productivity signatures are mainly influ- 2008). The d13Corg time series of the core GC/SK-240/496 shows a
enced by SWM runoff from Indus River while internal cycling of periodicity of w438 years which is similar to the w452 years
nutrients within water column through upwelling/convecting periodicity recorded in the d13Corg time series of the core SK-240/
mixing have negligible effects (Azharuddin et al., 2017). It is also 485 (Fig. 4d). The w360 years periodicity recorded in d13Corg time
supported by the ratio of planktic and benthic foraminifera per- series of the core GC/SK-240/496 is similar with w360 years peri-
centage in the core SK-240/485 (Azharuddin et al., 2017) which odicity marked in d15N time series of the core GC/SK-240/496
demonstrates a reduction in productivity in the periods corre- (Fig. 6). These cycles may correspond to the solar influenced
sponding to the weakened SWM conditions. A significant decline in w320 years SWM cycle reported in the magnetic susceptibility
productivity due to weakened monsoon circulation during cold record from the Arabian Sea (Thamban et al., 2007). Moreover, the
periods is also reported from the eastern Arabian Sea (Singh et al., TOC time series data of GC/SK-240/496 recorded the periodicities of
2011). We, therefore, attempted the spectral analysis of paleo- w471 and w322 years (Fig. 5). The w471 and w322 years peri-
productivity proxies d13CG.ruber, TOC, d13Corg and d15N (Figs. 4e6). odicities recorded in the TOC data match with w525 and w292
2260 S. Azharuddin et al. / Geoscience Frontiers 10 (2019) 2251e2263

years periodicities recorded in the d18OG.ruber time series of the core periodicities match with the results observed in the time series of
SK-240/485. Hence, it can be summarised that the solar forcing is other proxies which suggest the strong interdependence of various
the primary factor driving the SWM and that regulates the paleo- climatic processes occurring in the coastal area which influenced
productivity as well as carbon and nitrogen cycles of the Arabian the depositional conditions of the area.
Sea during the Holocene.
4.4. Validity and stationarity of the spectral analysis results
4.3. Periodicities in sea level fluctuation
Correlation analysis has been performed between the d18O time
The depositional environment in coastal areas remains very series of both the cores and with the TSI. Firstly, the phase analysis
dynamic due to intermixing of terrestrial and marine sediments was performed between common time periods of both the cores
and waters as well as their interaction with atmospheric winds. The (i.e., w10e4 ka). Total, three different phases were identified within
deposited sediments include the terrigenous material supplied by the studied time period i.e. phase 1 consists of the time period
the fluvial source (rivers) and the biogenic inputs resulted due to between 3.8 ka and 6.2 ka having a significant positive correlation
biological productivity in the basin (Azharuddin et al., 2017). The (r ¼ 0.48572; n ¼ 13) whereas phases 2 and 3 consist of insignifi-
sediment characteristics and processes involved are mainly gov- cant negative correlation (Phase 2ebetween 8 ka and 7 ka:
erned by the distance from the coast as well as fluctuation in r ¼ 0.18821; n ¼ 7; Phase 3ebetween 9.8 ka and 8 ka:
regional sea level in combination with oceanographic conditions r ¼ 0.08285; n ¼ 13). These results indicate that both the cores
around the core site. The planktic/benthic foraminifera ratio (P/B although belongs to a nearby location, but the behavior of proxies is
ratio) in the sediment of continental margin is a well-developed not similar due to the difference in bathymetric conditions. Hence,
proxy to study the regional sea level fluctuation (Grimsdale and the periodicities recorded in the proxy time series of one core may
Van Morkhoven, 1955; Nigam and Henrique, 1992). Increased dis- or may not be observed in the same proxy time series of the other
tance from the coast due to deepening of core site and increased core.
regional sea level is marked by the increased P/B ratio (Van der Correlation analysis of d18O time series of both the cores was
Zwaan et al., 1990; Azharuddin et al., 2017) and decreased CF % done with the TSI time series for the overlapping time periods. The
(Azharuddin et al., 2017). The spectral analysis of P/B ratio time phase estimate of d18O time series of the core SK-240/485 (having
series in the core SK-240/485 reveals significant periodicities (>90% 88 m water depth) with TSI indicates significant negative correla-
significance) centered at w512, w388, w304, w250, w235, w217, tion (r ¼ 0.34136; P < 0.001) whereas the d18O time series of the
w152, w139 and w135 years (Fig. 7). core GC/SK-240/496 (having 174 m water depth) shows an insig-
The w512 years periodicity recorded in the P/B ratio time series nificant relationship with the TSI time series (r ¼ 0.09065, P ¼ not
shows a similarity with the w525 years monsoonal periodicity significant). Hence, the core SK-240/485 has a strong influence of
observed in d18OG.ruber time series. This similarity may be due to the Indus River discharge and is capable of preserving maximum SWM
result of SWM influence on the regional sea level fluctuations. The signals as compared to the core GC/SK-240/496. The main reason
w388 years periodicity in the P/B ratio time series can be correlated for these observations could be possibly explained by the deposi-
with the w438 years cycle recorded in d13Corg time series in the tional conditions at these core locations. The average sedimenta-
core SK-240/496 (Fig. 4d). These two cycles are related to 434e454 tion rate around the core SK-240/485 is 32.9 cm/ka whereas that of
years cycles recorded in Lake Titicaca, Altiplano plateau, South core SK-240/496 is 21.1 cm/ka. Therefore, the high rate of sedi-
America, which are associated with Bond events of North Atlantic mentation enabled the core SK-240/485 to have a better temporal
(Baker et al., 2005). Further, the w250, w235 and w217 years resolution (40e110 years per sample) than the core GC/SK-240/496
(Fig. 7) periodicities in P/B ratio time series match with w256 years (100e200 years per sample). These observations are also supported
periodicities recorded in d18OG.ruber time series which are reported by the records of d18O, d13C and foraminifera percentage from these
to be influenced with SWM (as discussed in the previous sections). two cores (Azharuddin et al., 2017; Azharuddin, 2019). The high-
Hence it may be concluded that the SWM has a substantial influ- resolution record of core SK-240/485 marked several abrupt
ence on the regional sea level fluctuation in the coastal areas. Other events (e.g., Younger Dryas (YD), 8.2 ka, 4 ka cold events, etc.)
periodicities in the P/B ratio time series centering at w152, w139 corresponding to the weakened SWM in response to the North-
and w135 years also suggest the similarity with already reported Atlantic cold oscillations.
solar influence SWM periodicities from the Arabian Sea. These cy- On the other hand, in the core GC/SK-240/496, the same proxies
cles include w152 years (Gupta et al., 2005), w156 and w126 years (d18O, d13C and foraminifera percentage) do not mark those
(Thamban et al., 2007) periodicities, which are similar and reported important abrupt events significantly due to relatively lower sam-
to be of solar origin and also match with w154 years sunspot cycle ple resolution. These observations suggest that the solar influenced
(Solanki et al., 2004). A study from the western Bay of Bengal based signals are most suitable to be studied using the near-shore coastal
on benthic foraminiferal abundance also suggests ~200
50 years marine sedimentary cores due to the high rate of sediment depo-
periodicity in the southwest monsoon precipitation indicating the sition which provides the high-resolution dataset. In contrast, the
effect of solar variability on the monsoonal changes (Rana and depositional rates as well as temporal resolution towards the
Nigam, 2009). Similar w151 and w147 years solar influenced deeper waters decreases and hence low-resolution signals are more
SWM cycle have also been recorded in magnetic susceptibility re- prone to get diluted by the deeper oceanic currents and/or the
cords of Arabian Sea (Agnihotri et al., 2002) and TK lake, South India localised oxic conditions in the water column. Further, the cross-
(Warrier et al., 2017). On a global level, these are similar with the spectrum analysis shows high significance between both the
w148 years periodicity recorded in the Chesapeake Bay, eastern cores, i.e. there is no lag in the power spectra (e.g., significant
North Atlantic (Willard et al., 2005) which is attributed to the cold correlation at 0 lag) (Fig. S1).
events in the North Atlantic. Wavelet analysis was performed using PAST data analysis
The spectral analysis of >63 mm coarse fraction (CF) was per- software version 3.21 (Hammer et al., 2001) in support of the
formed to look at the possible periodicities present in the deposi- results obtained from REDFIT based spectral analysis. The analysis
tional processes. The significant periodicities (>90% significance) was done to understand the stationarity of the data. The analytical
recorded in CF time series in the core SK-240/485 include w268, parameters were set to detect the periodicities of >95% signifi-
w216, w209, w198, w188, w173 and w140 years (Fig. 8). All these cance. The d18O of core SK-240/485 marks high energy bands in
 S. Azharuddin et al. / Geoscience Frontiers 10 (2019) 2251e2263 2261

and around w300 and w580 years extending between 1.4 ka and salinity effect on the isotopic fractionation which may depend on
3 ka (Fig. 9). A significant high energy band around 1250 years is several other parameters hence diluting the solar influenced SWM
also visible in the early Holocene extending between 8 ka and 11 signatures in the isotopic proxies. Therefore, a strong connection
ka (Fig. 9). Another significant high energy band in and around exists between the high solar activities causing strengthened
w520 years also appears to extend during early-Holocene be- monsoon which ultimately affects the depositional system of the
tween 10.5 ka and 11.7 ka. On the other hand, the d18O of core GC/ NE Arabian Sea.
SK-240/496 does not record any significant high energy band in
the wavelet analysis (Fig. 10). 5. Conclusions
The w300 years periodicity during the late Holocene (1.4e3 ka)
seems to be similar as the w290 years periodicity observed in the The spectral analysis in the two cores from offshore Saurashtra,
time-sliced spectral analysis of middle-late Holocene (1e8 ka) NE Arabian reveals that similar significant periodicities exist in
(Fig. 3c). In addition, the w520 years cyclicity in the wavelet data SWM, productivity, regional sea level fluctuation, and deposi-
appears similar to the w525 years periodicity observed in REDFIT tional processes of the NE Arabian Sea during the Holocene. The
based spectral analysis of YD and early Holocene time slice (12.4e8 recorded significant (>90% significance) SWM periodicities
ka) (Fig. 3d). Similarly, the wavelet based periodicities of w530, include w1609, w945, w667, w525, w290, w296 and w256
w260, w380 years in other multi-proxy datasets appear to be years. The periodicities recorded in productivity include w681,
common with the periodicities based on REDFIT analysis of the w512, w471, w452, w438, w360, w322, w292, w275, w269,
same dataset (Fig. 9). However, there are few periodicities which w245 and w209 years. Moreover, the periodicities recorded in the
are not common for both the analyses (Wavelet and REDFIT). This regional sea-level fluctuations and depositional system include
difference could be due to the different confidence limits and/or w512, w388, w304, w268, w250, w235, w217, w216, w209,
analysis parameter settings. w152, w139, w198, w188, w173, w140 and w135 years. Further,
the stationarity and periodicities have also been verified using
4.5. Possible amplifying mechanism of the climate periodicities due Wavelet analysis which infers the similar periodicities as observed
to solar activity in REDFIT analysis. The correlation analysis for common time
periods suggests the different depositional behaviour of both the
The change in energy associated with solar activity on the Ho- core sites during early Holocene and thereafter both the cores
locene time-scale is w1.3% or 4 Wme2 (Renssen et al., 2006). It show similar behaviour during the middle Holocene. Moreover,
arises the question that with changes in solar activity, how the the TSI reveals a significant relationship with d18O time series of
climate system reacts? The amplitude of climate response due to the core SK-240/485 whereas that of core GC/SK-240/496 shows
variation in solar activity can be studied by two possible feedback the insignificant relationship. The similarity in SWM periodicities
mechanisms (Crosta et al., 2007; Govil et al., 2018). These mecha- with productivity, sea-level and depositional processes (within
nisms have been working through atmospheric processes either the radiocarbon and sampling errors) during the Holocene in-
alone or in conjugation with each other, and also superimpose with dicates a strong coherency among these parameters which is ul-
different internal climate variability and external forcing to intro- timately governed by solar forcings. The study also indicates that
duce additional climate variability at decadal and centennial time sediment flux may be a more reliable indicator of SWM cycles
scale (Rind, 2002). The first feedback mechanism is related to the than the biogeochemical proxies. However, these trends are better
UV radiations and production of ozone. During the high solar ac- preserved in shallow water marine sediment core (i.e., core SK-
tivity, the stratospheric ozone absorbs more UV radiations which 240/485) as compared to the deeper marine sediment core (GC/
increase the temperature in the stratosphere (Shindell et al., 1999). SK-240/496). More similar multiproxy spectral analysis studies
The strong winds thus produced in the lower stratosphere descend are required to get a better understanding of the relation between
to the troposphere which may influence changes in the tropical monsoon, external forcing factors and the mechanism responsible
Hadley circulation and primarily affects the SWM induced evapo- for these periodicities.
ration and precipitation (Dogar, 2018). The second feedback
mechanism consists of the variation in cosmic rays and cloud cover
due to the change in solar activity which causes amplification in Acknowledgment
climate forcing. Low and optically thick cloud cover causes cooling
around the earth surface due to low radiative effect whereas high The authors are thankful to the Director, BSIP, Lucknow for
and optically thin clouds cover causes warming (Hartmann, 1993). providing the necessary facilities to carry out this work and
A strong relationship exists between solar activity and the cloud permitting us to publish the same. Extended thanks to the ship-
cover causing a certain degree of correlation (Laut, 2003). It is board team of Sagar Kanya cruise for sampling two sediment cores.
believed that during high solar activity, less cosmic particles reach SA is also indebted to SERB-DST Project No. SR/FTP/ES-53/2013 and
to the atmosphere which reduces the number of nuclei available for Council of Scientific and Industrial Research (CSIR) (Grant No. 09/
the condensation of cloud resulting in reduced cloud cover which 528/0022/2018/EMR-1) for the financial assistance in the form of
causes warming (Marsh and Svensmark, 2000) and hence fellowships. We are also thankful to Dr. P. Morethekai, BSIP, Luck-
strengthening the SWM. Conversely, more low clouds are produced now, India for his suggestions during statistical analyses. This work
during low solar activity which causes cooling on the earth surface was supported by fast-track grant of SERB-DST under project No.
(Crosta et al., 2007). SR/FTP/ES-53/2013. It is the BSIP contribution no. BSIP/RDCC/Pub-
Furthermore, the present study documents that multi-proxy lication No. 77/2017-2018. We are also thankful to the Associate
study based on spectral analysis reveals that the proxies related Editor, Prof. E. Shaji and two anonymous reviewers for their
to sediment flux (i.e., P/B ratio and CF) recorded more significant constructive comments, which improved MS substantially.
periodicities (ranging up to 99% significance value) as compared
to the periodicities observed in the biogeochemical (isotopic) Appendix A. Supplementary data
proxies (ranging of 90%e95% significance). It indicates that sedi-
ment flux may be a more reliable indicator of SWM cycles than the Supplementary data to this article can be found online at
biogeochemical proxies. It may be due to the temperature and https://doi.org/10.1016/j.gsf.2019.03.007.
2262 S. Azharuddin et al. / Geoscience Frontiers 10 (2019) 2251e2263

References Grimsdale, T.F., Van Morkhoven, F.P.C.M., 1955 January. 4. The ratio between pelagic
and benthonic foraminifera as a means of estimating depth of deposition of. In:
4th World Petroleum Congress. World Petroleum Congress.
Agnihotri, R., Dutta, K., Bhushan, R., Somayajulu, B.L.K., 2002. Evidence for solar
Gupta, A.K., Mélice, J.L., 2003. Orbital forcing of the Plio-Pleistocene Indian mon-
forcing on the Indian monsoon during the last millennium. Earth and Planetary
soons: benthic foraminiferal proxies from ODP Site 758. Current Science
Science Letters 198 (3), 521e527.
179e184.
Alley, R.B., Marotzke, J., Nordhaus, W.D., Overpeck, J.T., Peteet, D.M., Pielke, R.A.,
Gupta, A.K., Das, M., Anderson, D.M., 2005. Solar influence on the Indian summer
Pierrehumbert, R.T., Rhines, P.B., Stocker, T.F., Talley, L.D., Wallace, J.M., 2003.
monsoon during the Holocene. Geophysical Research Letters 32 (17).
Abrupt climate change. Science 299 (5615), 2005e2010.
Gupta, A.K., Mohan, K., Das, M., Singh, R.K., 2013. Solar forcing of the Indian summer
Archer, D., 2003. Biological fluxes in the ocean and atmospheric pCO2. Treatise on
monsoon variability during the Ållerød period. Scientific Reports 3, 2753.
Geochemistry 6, 625.
Guptha, M.V.S., Naidu, P.D., Haake, B.G., Schiebel, R., 2005. Carbonate and carbon
Azharuddin, S., 2019. Late Quaternary Oceanographic and Climatic Reconstructions
fluctuations in the Eastern Arabian Sea over 140 ka: implications on produc-
Based on Foraminifera and Sediment Geochemical Signatures from the North-
tivity changes? Deep Sea Research Part II: Topical Studies in Oceanography 52
eastern Arabian Sea. Banaras Hindu University, Varanasi, India. Ph. D. Thesis.
(14e15), 1981e1993.
Azharuddin, S., Govil, P., Singh, A.D., Mishra, R., Agrawal, S., Tiwari, A.K., Kumar, K.,
Hammer, Ø., Harper, D.A.T., Ryan, P.D., 2001. PAST-Palaeontological statistics, p. 2009.
2017. Monsoon-influenced variations in productivity and lithogenic flux along
www.uv.es/wpardomv/pe/2001_1/past/pastprog/past.pdf. acessado em, 25(07).
offshore Saurashtra, NE Arabian Sea during the Holocene and Younger Dryas: a
Hartmann, D.L., 1993. Radiative effects of clouds on Earth’s climate. International
multi-proxy approach. Palaeogeography, Palaeoclimatology, Palaeoecology 483,
Geophysics Series 54, 151.
136e146.
Hilbrecht, H., 1996. Extant Planktic Foraminifera and the Physical Environment
Baker, P.A., Fritz, S.C., Garland, J., Ekdahl, E., 2005. Holocene hydrologic variation at
in the Atlantic and Indian Oceans: an Atlas Based on Climap and Levitus
Lake Titicaca, Bolivia/Peru, and its relationship to North Atlantic climate vari-
(1982) Data.
ation. Journal of Quaternary Science: Published for the Quaternary Research
Howard, W.R., Prell, W.L., 1992. Late Quaternary surface circulation of the southern
Association 20 (7e8), 655e662.
Indian Ocean and its relationship to orbital variations. Paleoceanography 7 (1),
Be, A.W., 1977. An ecological, zoogeographic and taxonomic review of recent
79e117.
planktonic foraminifera. Oceanic Micropaleontology 1, 1e100.
Hoyt, D.V., Schatten, K.H., 1997. The Role of the Sun in Climate Change. Oxford
Beer, J., Mende, W., Stellmacher, R., White, O.R., 1996. Intercomparisons of proxies
University Press.
for past solar variability. In: Climatic Variations and Forcing Mechanisms of the
Hu, F.S., Kaufman, D., Yoneji, S., Nelson, D., Shemesh, A., Huang, Y., Tian, J., Bond, G.,
Last 2000 Years. Springer, Berlin, Heidelberg, pp. 501e517.
Clegg, B., Brown, T., 2003. Cyclic variation and solar forcing of Holocene climate
Berger, W.H., Von Rad, U., 2002. Decadal to millennial cyclicity in varves and tur-
in the Alaskan subarctic. Science 301 (5641), 1890e1893.
bidites from the Arabian Sea: hypothesis of tidal origin. Global and Planetary
Jensen, E.S., 1991. Evaluation of automated analysis of 15N and total N in plant
Change 34 (3), 313e325.
material and soil. Plant and Soil 133, 83e92.
Bond, G., Kromer, B., Beer, J., Muscheler, R., Evans, M.N., Showers, W., Hoffmann, S.,
Jung, S.J.A., Davies, G.R., Ganssen, G., Kroon, D., 2002. Decadal-centennial scale
Lotti-Bond, R., Hajdas, I., Bonani, G., 2001. Persistent solar influence on North
monsoon variations in the Arabian Sea during the early Holocene. Geochem-
Atlantic climate during the Holocene. Science 294 (5549), 2130e2136.
istry, Geophysics, Geosystems 3 (10), 1e10.
Bond, G., Showers, W., Cheseby, M., Lotti, R., Almasi, P., Priore, P., Cullen, H.,
Kendall, M.G., 1975. Rank Correlation Methods. Griffin, London.
Hajdas, I., Bonani, G., 1997. A pervasive millennial-scale cycle in North Atlantic
Laut, P., 2003. Solar activity and terrestrial climate: an analysis of some purported
Holocene and glacial climates. Science 278 (5341), 1257e1266.
correlations. Journal of Atmospheric and Solar-Terrestrial Physics 65 (7), 801e812.
Braconnot, P., Otto-Bliesner, B., Harrison, S., Joussaume, S., Peterchmitt, J.Y., Abe-
Lean, J., 2002. Solar forcing of climate change in recent millennia. In: Climate Devel-
Ouchi, A., Crucifix, M., Driesschaert, E., Fichefet, T., Hewitt, C.D., Kageyama, M.,
opment and History of the North Atlantic Realm. Springer, Berlin, pp. 75e88.
2007. Results of PMIP2 coupled simulations of the Mid-Holocene and Last
Madhupratap, M., Prasanna Kumar, S., Bhattathiri, P.M.A., Kumar, M.D.,
Glacial MaximumePart 1: experiments and large-scale features. Climate of the
Raghukumar, S., Nair, K.K.C., Ramaiah, N., 1996. Mechanism of the biological
Past 3 (2), 261e277.
response to winter cooling in the Northeastern Arabian Sea. Nature 384,
Budyko, M.I., 1969. The effect of solar radiation variations on the climate of the
549e552.
earth. Tellus 21 (5), 611e619.
Mann, H.B., 1945. Nonparametric tests against trend. Econometrica 13, 245e259.
Clemens, S., Prell, W., Murray, D., Shimmield, G., Weedon, G., 1991. Forcing mech-
Marchitto, T.M., Muscheler, R., Ortiz, J.D., Carriquiry, J.D., van Geen, A., 2010.
anisms of the Indian Ocean monsoon. Nature 353 (6346), 720.
Dynamical response of the tropical Pacific Ocean to solar forcing during the
Croll, J., 1875. Climate and time. Nature 12 (304), 329.
early Holocene. Science 330 (6009), 1378e1381.
Crosta, X., Debret, M., Denis, D., Courty, M.A., Ther, O., 2007. Holocene long- and
Marsh, N.D., Svensmark, H., 2000. Low cloud properties influenced by cosmic rays.
short-term climate changes off Adélie Land, East Antarctica. Geochemistry,
Physical Review Letters 85 (23), 5004.
Geophysics, Geosystems 8 (11), Q11009. https://doi.org/10.1029/2007GC001718.
Marshall, J., Kushnir, Y., Battisti, D., Chang, P., Czaja, A., Dickson, R., Hurrell, J.,
Dogar, M.M., 2018. Impact of tropical volcanic eruptions on Hadley circulation using
McCartney, M., Saravanan, R., Visbeck, M., 2001. North Atlantic climate vari-
a high-resolution AGCM. Current Science. https://doi.org/10.18520/cs/v114/i06/
ability: phenomena, impacts and mechanisms. International Journal of Clima-
1284-1294.
tology 21 (15), 1863e1898.
Dutt, S., Gupta, A.K., Clemens, S.C., Cheng, H., Singh, R.K., Kathayat, G., Edwards, R.L.,
Mauquoy, D., Yeloff, D., Van Geel, B., Charman, D.J., Blundell, A., 2008. Two decadally
2015. Abrupt changes in Indian summer monsoon strength during 33,800 to
resolved records from north-west European peat bogs show rapid climate
5500 years BP. Geophysical Research Letters 42 (13), 5526e5532.
changes associated with solar variability during the midelate Holocene. Journal
Dutta, K., Bhushan, R., Somayajulu, B.L.K., 2001. DR correction values for the
of Quaternary Science 23 (8), 745e763.
Northern Indian Ocean. Radiocarbon 43, 483e485.
McLeod, A.I., 2005. Kendall rank correlation and Mann-Kendall trend test. R Package
Dykoski, C.A., Edwards, R.L., Cheng, H., Yuan, D., Cai, Y., Zhang, M., Lin, Y., Qing, J.,
Kendall.
An, Z., Revenaugh, J., 2005. A high-resolution, absolute dated Holocene and
Milankovitch, M.M., 1941. Canon of Insolation and the Iceage Problem. Koniglich
Deglacial Asian monsoon record from Dongge Cave, China. Earth and Planetary
Serbische Akademice Beograd Special Publication, p. 132.
Science Letters 233, 71e86.
Milliman, J.D., Quraishee, G.S., Beg, M.A.A., 1984. Sediment discharge from the Indus
Emile-Geay, J., Seager, R., Cane, M.A., Cook, E.R., Haug, G.H., 2008. Volcanoes and
River to the ocean: past, present and future. Marine geology and oceanography
ENSO over the past millennium. Journal of Climate 21 (13), 3134e3148.
of Arabian Sea and Coastal Pakistan 65e70.
Fairbanks, R.G., WIEBE, P.H., Bé, A.W., 1980. Vertical distribution and isotopic
Naidu, P.D., Malmgren, B.A., 1995. A 2,200 years periodicity in the Asian monsoon
composition of living planktonic foraminifera in the western North Atlantic.
system. Geophysical Research Letters 22 (17), 2361e2364.
Science 207 (4426), 61e63.
Nair, R.R., Ittekkot, V., Manganini, S.J., Ramaswamy, V., Haake, B., Degens, E.T.,
Ferronsky, V.I., 1992. The nature of short-periodic climate variation and its relation
Desai, B.T., Honjo, S., 1989. Increased particle flux to the deep ocean related to
with water exchange. GeoJournal 27 (3), 237e242.
monsoons. Nature 338 (6218), 749e751.
Gadgil, S., Rajeevan, M., Nanjundiah, R., 2005. Monsoon predictiondwhy yet
Neff, U., Burns, S.J., Mangini, A., Mudelsee, M., Fleitmann, D., Matter, A., 2001. Strong
another failure. Current Science 88 (9), 1389e1400.
coherence between solar variability and the monsoon in Oman between 9 and
Ghil, M., Allen, M.R., Dettinger, M.D., Ide, K., Kondrashov, D., Mann, M.E.,
6 kyr ago. Nature 411 (6835), 290e293.
Robertson, A.W., Saunders, A., Tian, Y., Varadi, F., Yiou, P., 2002. Advanced
Nigam, R., Henrique, P.J., 1992. Planktonic percentage of foraminiferal fauna in
spectral methods for climatic time series. Reviews of Geophysics 40 (1).
surface sediments of the Arabian Sea (Indian Ocean) and a regional model for
Gleissberg, W., 1965. The eighty-year cycle in auroral frequency numbers. The
paleodepth determination. Palaeogeography, Palaeoclimatology, Palaeoecology
Journal of the British Astronomical Association 75, 227.
91 (1), 89e98.
Govil, P., Mazumder, A., Ram, R., Singh, D.S., Azharuddin, S., 2018. Melt water flux
Nigam, R., Nair, R.R., 1989. Cyclicity in monsoon. Bulletin of Science (India) 5, 42e44.
and climate change record of last 18.5 ka from Schirmacher Oasis, East
Nigam, R., Khare, N., Nair, R.R., 1995. Foraminiferal evidences for 77-year cycles of
Antarctica. Polar Science 18, 135e141.
droughts in India and its possible modulation by the Gleissberg solar cycle.
Gray, L.J., Beer, J., Geller, M., Haigh, J.D., Lockwood, M., Matthes, K., Cubasch, U.,
Journal of Coastal Research 1099e1107.
Fleitmann, D., Harrison, G., Hood, L., Luterbacher, J., 2010. Solar influences on
Obrochta, S.P., Miyahara, H., Yokoyama, Y., Crowley, T.J., 2012. A re-examination of
climate. Reviews of Geophysics 48 (4), RG4001. https://doi.org/10.1029/
evidence for the North Atlantic “1500-year cycle” at Site 609. Quaternary Sci-
2009RG000282.
ence Reviews 55, 23e33.
 S. Azharuddin et al. / Geoscience Frontiers 10 (2019) 2251e2263 2263

Prell, L.W., 1984. Monsoonal climate of the Arabian Sea during the late Quaternary: Singh, A.D., Jung, S.J., Darling, K., Ganeshram, R., Ivanochko, T., Kroon, D., 2011.
a response to changing solar radiation. In: Milankovitch and Climate: Under- Productivity collapses in the Arabian Sea during glacial cold phases. Paleo-
standing the Response to Astronomical Forcing, p. 349. ceanography 26 (3). https://doi.org/10.1029/2009PA001923.
Prell, W.L., Kutzbach, J.E., 1987. Monsoon variability over the past 150,000 years. Sirocko, F., Garbe-Schönberg, D., McIntyre, A., Molfino, B., 1996. Teleconnections
Journal of Geophysical Research: Atmosphere 92 (D7), 8411e8425. between the subtropical monsoons and high-latitude climates during the last
R Core Team, 2014. R: A Language and Environment for Statistical Computing. R deglaciation. Science 272 (5261), 526e529.
Foundation for Statistical Computing, Vienna, Austria. Solanki, S.K., Usoskin, I.G., Kromer, B., Schüssler, M., Beer, J., 2004. Unusual activity
Rana, S.S., Nigam, R., 2009. Cyclicity in the late Holocene monsoonal changes from of the Sun during recent decades compared to the previous 11,000 years. Nature
the western Bay of bengal: foraminiferal approach. Journal of the Palae- 431 (7012), 1084e1087.
ontological Society of India 54 (2), 165e170. Steinhilber, F., Abreu, J.A., Beer, J., Brunner, I., Christl, M., Fischer, H., Heikkilä, U.,
Raspopov, O.M., Dergachev, V.A., Esper, J., Kozyreva, O.V., Frank, D., Ogurtsov, M., Kubik, P.W., Mann, M., McCracken, K.G., Miller, H., 2012. 9,400 years of cosmic
Kolström, T., Shao, X., 2008. The influence of the de Vries (w 200-year) solar radiation and solar activity from ice cores and tree rings. Proceedings of the
cycle on climate variations: results from the Central Asian Mountains and their National Academy of Sciences 109 (16), 5967e5971.
global link. Palaeogeography, Palaeoclimatology, Palaeoecology 259 (1), 6e16. Steinhilber, F., Beer, J., Fröhlich, C., 2009. Total solar irradiance during the Holocene.
Rayner, N.A., Parker, D.E., Horton, E.B., Folland, C.K., Alexander, L.V., Rowell, D.P., Geophysical Research Letters 36 (19), L19704. https://doi.org/10.1029/
Kent, E.C., Kaplan, A., 2003. Global analyses of sea surface temperature, sea ice, 2009GL040142.
and night marine air temperature since the late nineteenth century. Journal of Stuiver, M., Braziunas, T.F., 1993. Modeling atmospheric 14 C influences and 14 C
Geophysical Research 108 (D14), 4407. https://doi.org/10.1029/2002JD002670. ages of marine samples to 10,000 BC. Radiocarbon 35 (1), 137e189.
Renssen, H., Goosse, H., Muscheler, R., 2006. Coupled climate model simulation of Stuiver, M., Grootes, P.M., Braziunas, T.F., 1995. The GISP2 d18O climate record of the
Holocene cooling events: oceanic feedback amplifies solar forcing. Climate of past 16,500 years and the role of the sun, ocean, and volcanoes. Quaternary
the Past 2 (2), 79e90. Research 44 (3), 341e354.
Rind, D., Overpeck, J., 1993. Hypothesized causes of decade-to-century-scale climate Stuiver, M., Reimer, P.J., 1993. Extended 14C database and revised CALIB radiocarbon
variability: climate model results. Quaternary Science Reviews 12 (6), 357e374. calibration program. Radiocarbon 35, 215e230.
Rind, D., 2002. The Sun’s role in climate variations. Science 296 (5568), 673e677. Thamban, M., Kawahata, H., Rao, V.P., 2007. Indian summer monsoon variability
Saha, K.R., Mooley, D.A., Saha, S., 1979. The Indian monsoon and its economic during the Holocene as recorded in sediments of the Arabian Sea: timing and
impact. GeoJournal 3 (2), 171e178. implications. Journal of Oceanography 63 (6), 1009e1020.
Sandeep, K., Shankar, R., Warrier, A.K., Weijian, Z., Xuefeng, L., 2015. The environ- Thomson, D.J., 1982. Spectrum estimation and harmonic analysis. Proceedings of
mental magnetic record of palaeoenvironmental variations during the past the IEEE 70 (9), 1055e1096.
3100 years: a possible solar influence? Journal of Applied Geophysics 118, Tiwari, M., Ramesh, R., Somayajulu, B.L.K., Jull, A.J.T., Burr, G.S., 2005. Solar control of
24e36. southwest monsoon on centennial timescales. Current Science 89 (9), 1583.
Schlitzer, R., 2007. Ocean data view. http://odv.awi.de. Torrence, C., Compo, G.P., 1998. A practical guide to wavelet analysis. Bulletin of the
Schulz, M., Mudelsee, M., 2002. REDFIT: estimating red-noise spectra directly from American Meteorological Society 79 (1), 61e78.
unevenly spaced paleoclimatic time series. Computers & Geosciences 28 (3), Usoskin, I.G., Mursula, K., 2003. Long-term solar cycle evolution: review of recent
421e426. developments. Solar Physics 218 (1e2), 319e343.
Schulz, M., Stattegger, K., 1997. SPECTRUM: spectral analysis of unevenly spaced Van der Zwaan, G.J., Jorissen, F.J., De Stigter, H.C., 1990. The depth dependency of
paleoclimatic time series. Computers & Geosciences 23 (9), 929e945. planktonic/benthic foraminiferal ratios: constraints and applications. Marine
Shekhar, M., Bhardwaj, A., Singh, S., Ranhotra, P.S., Bhattacharyya, A., Pal, A.K., Geology 95 (1), 1e16.
Roy, I., Martín-Torres, F.J., Zorzano, M.P., 2017. Himalayan glaciers experienced Wanner, H., Solomina, O., Grosjean, M., Ritz, S.P., Jetel, M., 2011. Structure and origin
significant mass loss during later phases of little ice age. Scientific Reports 7. of Holocene cold events. Quaternary Science Reviews 30 (21e22), 3109e3123.
Article number: 10305. https://doi.org/10.1038/s41598-017-09212-2. Warrier, A.K., Sandeep, K., Shankar, R., 2017. Climatic periodicities recorded in lake
Shindell, D., Rind, D., Balachandran, N., Lean, J., Lonergan, P., 1999. Solar cycle sediment magnetic susceptibility data: further evidence for solar forcing on
variability, ozone, and climate. Science 284 (5412), 305e308. Indian summer monsoon. Geoscience Frontiers 8 (6), 1349e1355.
Sigman, D.M., Casciotti, K.L., 2001. Nitrogen isotopes in the ocean. In: Steele, J.H., Willard, D.A., Bernhardt, C.E., Korejwo, D.A., Meyers, S.R., 2005. Impact of
Turekian, K.K., Thorpe, S.A. (Eds.), Encycopedia of Ocean Sciences. Academic, millennial-scale Holocene climate variability on eastern North American
London. terrestrial ecosystems: pollen-based climatic reconstruction. Global and Plan-
etary Change 47 (1), 17e35.