International Journal of Oceans and Oceanography ISSN 0973-2667 Volume 14, Number 2 (2020), pp. 197-220 © Research India Publications https://dx.doi.org/

Centennial-Millennial Climate Variability in the Makassar Strait during Early Holocene until the End of the Last Deglaciation

Marfasran Hendrizan1,3*, Nining S. Ningsih2, Sri Y. Cahyarini3, Mutiara R. Putri2, Bambang Setiadi4, Iwan P. Anwar1,2, Dwi A. Utami3, Vera C. Agusta3 1 Earth Sciences Study Program, Faculty of Earth Sciences and Technology, Bandung Institute of Technology, Labtek XI, Jl. Ganeca No. 10, Bandung 40132, Indonesia. E-mail: m_hendrizan@yahoo,com (MH); [email protected] (IPA) 2 Research Group of Oceanography, Faculty of Earth Sciences and Technology, Bandung Institute of Technology, Labtek XI, Jl. Ganeca No.10, Bandung 40132,Indonesia. E-mail:[email protected] (NSN), [email protected] (MRP) 3Paleoclimate and Paleoenvironment Research Group, Research Center for Geotechnology, Indonesian Institute of Sciences, Jl. Sangkuriang Bandung 40135, Indonesia. E-mail:[email protected] (SYC), [email protected] (DAU), [email protected] (VCA) 4Geoinformatics and Computations Research Group, Research Center for Geotechnology, Indonesian Institute of Sciences, Jl. Sangkuriang Bandung 40135, Indonesia. E-mail: [email protected] *Correspondence: [email protected]

Abstract Indonesian archipelago establishes an essential region for considering relationships between climate and human activities in the region. However; large uncertainties remain in understanding Indonesian hydroclimate under varying timescales. Here we present reanalyzed data of high-resolution proxy from high sedimentation rate (120-170 cm/kyrs) marine sediment core SO217- 18517 from 9 to 14 kyrs in the offshore Mahakam Delta. Our dataset reveals high-frequency variations at centennial and millennial time scales during the 18 period between 9 and 14 kyrs based on multi-proxies data (δ Osw; Sea Surface Temperature/SST; and δ18O). Early Holocene to the end of last deglaciation period in the Makassar Strait was characterized by mixture forcing between 198 Marfasran Hendrizan et al.

deep ocean and solar radiation with ~1100 and ~1000-years periodicity in 18 18 altering millennial-scale variability on δ Osw and δ O records. On centennial timescale; the Makassar Strait climate variability was dominated by ~500-years periodicity recorded for Sea Surface Temperature (SST) at our core site location which is more influenced by deep ocean circulation. Keywords: Centennial; Millennial; Mixture Forcing; Deep Ocean, Solar Radiation; Makassar Strait

1. INTRODUCTION Climate variability in the Indonesian archipelago may have played a confidence role on the intensity of rainfall patterns and tropical convections in the highly populated region related to global warming impact in the future. Precipitation patterns in the Indonesian archipelago are strongly governed by island topography and/or ocean-atmosphere interactions. The phase five of Coupled Model Intercomparison Project (CMIP5) climate model predicts Southern Hemisphere monsoon including Indonesia will get less precipitation in the future (Lee & Wang, 2014). However, large uncertainties remain in understanding Indonesian hydroclimate under different timescales in the past. Therefore, it is crucial to study hydrologic variability in the Indonesian archipelago and dynamic process on a wide range spatial and temporal timescales. Holocene climate archived have been identified centennial to millennial-scale in tropical region (Khider et al., 2014). It is also challenging to look further those centennial to millennial-scale climate into the last deglacial period in the tropical region. Examples of millennial-scale variability such as the Heinrich Stadial 1 (HS1), Bølling-Allerød (B-A) and the Younger Dryas (YD) exist in the Indonesian region which is associated with precipitation changes in the region (Hendrizan et al., 2017; Kuhnt et al., 2015). However, centennial to millennial-scale variability of Indonesian hydroclimate since the last deglaciation until the early Holocene might be controlled by more than single forcing on tropical rain belt changes (Hendrizan et al., 2017). Southward swings of the Intertropical Convergence Zone (ITCZ) (Hendrizan et al., 2017; Kuhnt et al., 2015; Muller et al., 2008), El Niño Southern Oscillation (ENSO) changes (Levi et al., 2007; Stott et al., 2004), and sea level (Griffiths et al., 2009; Partin et al., 2007) are likely mechanisms in controlling precipitation changes at Indonesian region. The questions will be appeared, do centennial to millennial-scale climate variability occur in the Indonesian region during the last deglaciation and the Holocene? What is the main forcing on precipitation changes in the Indonesian region? we will extend the strength of the Atlantic Meridional Overturning Circulation (AMOC) proposed by previous study (Khider et al., 2014) which may influence the ocean/atmosphere system on centennial-millennial timescales in the Indonesian region. Centennial-Millennial Climate Variability in the Makassar Strait… 199

In addition, solar forcing is also identified in the Holocene period as an external forcing (Bond et al., 2001; Debret et al., 2009). However, recent study mentions Holocene centennial to millennial-scale variability might be controlled by deep water circulation in the western Pacific region (Khider et al., 2014) or ocean-ice sheet interaction (Bakker et al., 2017) as an internal forcing. Both external and internal forcing could trigger heat transfer in the Indonesian region which can be traced in the longer period until the last deglacial period in the region. Several marine sediment cores in the Indonesian region record a long temporal resolution until the last deglaciation in understanding the influence of Sea Surface Temperature (SST) with regional climate in the region (Mohtadi et al., 2011, 2017; Schroder et al., 2016; Schröder et al., 2018; Visser et al., 2003). Core SO217-18517 (Hendrizan et al., 2017) is more challenging to be reanalyzed due to the influence of river runoff and the ITF intensity in the core site. In addition, the core provides the highest resolution timescales until < 7 years during the last 18 deglacial period. δ Osw records in the Makassar Strait (Fan et al., 2013) interpreted millennial-scale variability during Holocene controlled by ENSO variability. Other revealed deep sea circulation is the main forcing of millennial-scale variability in the West Pacific Warm Pool (WPWP) during Holocene (Khider et al., 2014). Different forcing on millennial-scale variability in the WPWP especially central Indonesia is still 18 open question related to complex influences of δ Osw in the central Indonesia (Schröder et al., 2018). One of the most important is local river input which local river 18 input is known to be an important factor in controlling δ Osw – salinity variability in the WPWP (Katz et al., 2010; Morimoto et al., 2002). Therefore, Core SO217-18517 closes to the Mahakam River is suitable to understand high resolution climate signal in the Makassar Strait. Here we present high-resolution proxy data from high sedimentation rate (120-170 cm/kyr) marine sediment core SO217-18517 from 9 to 14 kyr in the offshore Mahakam Delta ((Hendrizan et al., 2017); Figure 1). The previous study (6) in Core SO217-18517 has not explained whether similar or not the Younger Dryas climate mechanisms in the Makassar Strait with the early Holocene period. Previously, uneven time series of Core SO217-18517 based on multi-proxies analysis was reported in (Hendrizan et al., 2017). In this study we have improved the evenness of the time series to 10 years/sample resolution, and concentrated on the time period 9 to 14 kyr. We present multi-proxies 18 18 (δ O, SST, δ Osw) wavelet analysis of Core SO217-18517 to characterize centennial and millennial climate variability spanning from early Holocene and the end of the last deglaciation in the Makassar Strait. We compare our multi-proxies data with other hydroclimate records in the central Indonesia particularly Kalimantan and Sulawesi region.

200 Marfasran Hendrizan et al.

Figure 1. a. Location of core SO217-18517 (1°32.198’S, 117°33.756’E; 698 m water depth), off Mahakam Delta, East Borneo (modificafion Figure from Supp. Figure 1 in (Hendrizan et al., 2017)). Red stars indicates core locations from other studies, marine sediment cores SO217-18519 and TGS-931 (Schröder et al., 2018), SO217-18515 (Schroder et al., 2016), Lake sediment from Towuti (Konecky et al., 2016), and Gunung Buda speleothem (Partin et al., 2007) in a bathymetric map from ETOPO 1 (http://maps.ngdc.gov/viewer/bathimetry); b. Marine sediment photograph of Core SO217-18517 in-depth scale.

2. REGIONAL CLIMATE AND OCEANOGRAPHY Marine sediment core SO217-18517 (1°32.198’S, 117°33.756’E; 698 m water depth) is situated in the offshore Mahakam Delta, East Borneo (Hendrizan et al., 2017) which extends for about ~90 km from the coastal line (Figure 1). The core was collected as a part of the SO217 (Makassar-Java/MAJA) cruise using R/V Sonne. The core site is connected to the southeast distributary channel of the Mahakam River. However, the core site is not only influenced by Mahakam river runoff but also controlled by lower thermocline Indonesian Throughflow (ITF) ((Hendrizan et al., 2017); Figure 1). During boreal summer, the changes in wind direction shift towards Asia due as the Intertropical Convergence Zone (ITCZ) moves northward, leading to reduced precipitation in eastern Borneo, and increased SST follows the peak sun radiation (Figure 2). Saline Centennial-Millennial Climate Variability in the Makassar Strait… 201 surface water during boreal summer indicates a reduced freshwater input related to the monsoonal wind shift ((Gordon, 2005); Figure 2). During boreal winter, wind direction moves along the Indonesian archipelago, dipping into northern Australia. Both increased precipitation in the eastern Borneo and decreased the SST also cause a reduced salinity in the Makassar Strait (Figure 2). The core site is situated in the Makassar Strait which is the inflow pathways of the ITF (Hendrizan et al., 2017). Recent study found the ITF changes during decadal time scales, the ITF strengthened when the Pacific Decadal Oscillation (PDO) positive and weakened when the PDO negative (Sprintall et al., 2019).

Figure 2. Monthly climatology in the offshore Mahakam river (a) average salinity (World Ocean Atlas 2009), (b) average SST (World Ocean Atlas 2013) (Locarnini et al., 2013), (d) average wind velocities (NCEP reanalysis data at http://www.esrl.noaa.gov/psd/), (d) CMAP Estimated precipitation from NOAA NCEP CPC Merged Analysis over the 1979-2011 period (Xie & Arkin, 1997).

202 Marfasran Hendrizan et al.

3. METHODS 3.1. Age model AMS 14C records of Core SO217-18517 used in this study are obtained by the previous study in the similar core (Hendrizan et al., 2017). Summary of previous study (Hendrizan et al., 2017), approximately 6 mg of well-preserved pteropod shells were picked from the >355 µm size fractions and mixed planktonic foraminifera were picked from the size fractions 250-315 µm for accelerator mass spectrometry (AMS) 14C dating. AMS14C analysis was performed at the Leibniz Laboratory, Christian- Albrechts-University, Kiel. The age model was constrained by 8 AMS 14C dates based on planktonic foraminifers and pteropods. An interpolated curve was fitted through the 8 AMS 14C tie points using a Stineman function (smooth function in Kaleidagraph).

3.2. Data

18 18 We used reanalysis data of Mg/Ca derived SST, δ O, and δ Osw from core SO217- 18517 that was published in (Hendrizan et al., 2017). In addition, a detailed explanation of the analytical technique for Mg/Ca and δ18O is shown in (Hendrizan et al., 2017) and summarized below. For δ18O analysis, approximately 10 tests of G. ruber were picked from the size fraction 315-250 µm at 2 cm interval from 14 to 9 ka. δ18O was measured with a Finnigan MAT 253 mass spectrometer (Carbo-Kiel Device (Type IV) for automated CO2 preparation from carbonate samples for stable isotopic analysis) at the Leibniz Laboratory, Christian-Albrechts-University, Kiel. For Mg/Ca measurement, thirty well-preserved tests of G. ruber were selected from the size fraction 315-250 µm. The samples were measured using a radial viewing simultaneous ICP-OES (Spectro Ciros SOP CCD, Spectro Analytical Instruments, Germany) at the Institute of Geosciences, Christian-Albrechts-University Kiel. Foraminifera Mg/Ca were converted into temperature equation (Anand et al., 2003). Mg/Ca=0.38 exp 0.09 T for Globigerinoides ruber 18 18 δ Osw reconstruction was calculated based on paired Mg/Ca and δ O measurements of G. ruber using equation (Bemis et al., 1998). 18 18 δ Osw = 0.27 + (T (°C) – 16.5 + 4.8 × δ O (V – PDB))/4.8 18 δ Osw Core SO217-18517 indicates paleosalinity indicator in the Makassar Strait, 18 which is assumed similar with relationship between δ Osw-salinity in the Western Pacific Warm Pool (Morimoto et al., 2002). Centennial-Millennial Climate Variability in the Makassar Strait… 203

A length of “uneven dataset” (~5000 years), original dataset obtained from geochemical measurements with different time spacing (δt), between 9 and 14 kyrs is prepared to analyze further for wavelet analysis. Before wavelet analysis applied, unevenly dataset at Core SO217-18517 from previous study (Hendrizan et al., 2017) should be interpolated to acquire equal time spacing (δt) in the entire timeseries between 9 and 14 kyrs for reanalysis data in this paper. Interpolated data with equal time spacing (δt) of 10 years in this study will be identified as “evenly dataset”. We applied linear interpolation in this study by (Lepot et al., 2017; Mathworks, 2019). Linear interpolation of our dataset will obtain a straight line passed through the end point χa 18 18 and χb of original dataset including SST, δ O, and δ Osw (Hendrizan et al., 2017). Equivalent equation for the method shown below

휒 − 휒 푋 = 푎 푏 (푖 − 푏) + 휒 푖 푎 − 푏 푏

Interpolated data are bound between xA and xB, and true values are, in average, underestimated: this affirmation is strongly dependent on the distribution of data and should be verified for each data set (Lepot et al., 2017). A Number of dataset before interpolated or original dataset are 208 data of SST, 232 data of δ18O, and 229 data of 18 δ Osw with various time spacing (δt) between 10 and 150 years. After interpolation applied, our interpolated data increase to be 484 data of SST, 481 data of δ18O, and 482 18 data of δ Osw with 10 years’ time spacing (δt) data.

3.3. Wavelet Analysis Before wavelet analysis performed, interpolated data in this study have removed their trends using equation of linear detrending. Removing a trend from our dataset help us to focus on the fluctuations of the data. We applied linear detrending using (Mathworks, 2019). The mean in linear detrending is given by the linear regression line Xt=St + I over the period T (=Ni∆푡), the regression slope S and intercept I shown as below (Rannik & Vesala, 1999)

푁 ∑ 푡푋 − ∑ 푡 ∑ 푋 푆 = 푖 푡 푡 2 2 푁푖 ∑ 푡 − (∑ 푡)

∑ 푋 − 푆 ∑ 푡 퐼 = 푡 푁푖

Where t=1∆t and the summation is made of i=1,……., Ni. Detrending data has a mean = 0 and standard deviation = 1.

204 Marfasran Hendrizan et al.

Wavelet analysis is useful for analyzing localized variations of power within time series 18 18 records. We analyze our multiproxy records (Mg/Ca, δ O, and δ Osw) using a modified script by (Torrence & Compo, 1997). Evenly datasets (~5000 years) between 18 18 9 and 14 kyrs in our multiproxy (Mg/Ca derived SST, δ O, and δ Osw) are prepared to obtain several millennial and centennial signal in the records with equal time spacing (δt) with 10 years periodicity. Wavelet transform is a band-pass filter which consists of convoluting the signal with scaled and translated forms of a highly time-localized wave function (the filter), the so-called “mother wavelet”. Morlet transform consists of a Gaussian plane wave as shown below:

2 −1/4 푖휔0휂 −휂 /2 ѱ0(휂) = 휋 푒 푒

Where ѱ0(휂) is wavelet function that depends on a nondimensional “time” parameter

휂 and 휔0 is the nondimensional frequency. Normalization is applied to ensure the wavelet transforms at each scale are directly comparable to each other and to the transforms of additional time series. Confidence levels refer to the range of confidence value in a given time series, we use the 95% confidence level to measure the mean power spectrum at Core SO17-18517. Global wavelet spectrum is the time-averaged wavelet spectrum of a vertical slice over a local wavelet spectrum plot as shown at the formula below,

2 1 푛2 2 푊푛 (s)= ∑푛=푛1|푊푛(푠)| 푛푎

Explanation of the formula indicates n is the midpoint of n1 and n2, and na=n2-n1+1 is the average number of points.

4. RESULTS 4.1. Distribution of Datasets 18 18 The box plot of our δ O, SST, and δ Osw which can be seen in Figure 3 indicates a number summary of critical information in Core SO217-18517 datasets. We see from the box plot (Figure 3), several SST distribution exceeds the upper whisker compare to 18 18 other δ O, and δ Osw in a core location. Exceeding the upper whisker of SST could be as ‘outlier’ observations, we see the SST with the value of maximum box plot (~29.5°C) could be ignored. We make a limitation of the upper whisker in the SST based on maximum calcification temperature based on foraminiferal Mg/Ca is more less 29.5°C (Anand et al., 2003; Regenberg et al., 2009). Therefore, outliers are lied more than upper whisker as a maximum calcification temperature should be ignored. The Centennial-Millennial Climate Variability in the Makassar Strait… 205 summary of our datasets in Core SO217-18517 is shown in table 1. Statistical summary in Core SO217-18517 shows 25th and 75th percentile range (blue box), median (red lines), and outliers (red cross) in Figure 3; Table 1 to approximately 99.5% the data range (blue ‘whiskers’) which are normally distributed (Suppl. Figure 1).

Table 1. Critical information of statistical summary in Core SO217-18517 datasets.

Datasets Mean Median Upper Lower Maximum Minimum Quartile Quartile

δ18O -0.1086 -0.1060 0.0594 -0.3 0.5764 -0.6729

SST 27.8621 27.8191 28.3 27.75 29.5 26.3858

18 δ Osw -2.2764 -2.3335 -1.8 -2.7 -1.13 -3.275

Figure 3. multi-proxies box plot of Core SO217-18517. a. δ18O, b. Sea Surface Temperature, c. δ18Osw. By reducing the outlier’s effect in SST, our multi-proxies reconstruction in Core SO217-18517 is more considerable.

4.2. Normalized Datasets Uneven records of core SO217-18517 are interpolated into evenly datasets with time resolution of 0.01 kyrs (Figure 4). It shows that linear interpolation between original and interpolated data intersects with each other. Coefficient correlation (r value) 18 between original and interpolated data about 41.2% of δ Osw, 29.7% of SST, and 73% 18 18 of δ O (Suppl. Figure 2). Depleted δ Osw records at our core site occur in the periods 18 of the 11-10 kyrs and 12-13 kyrs and. The amplitude of δ Osw shows higher salinity value at the Younger Dryas (12-13 kyrs) compare to other salinity records at 10-11 kyrs

206 Marfasran Hendrizan et al.

Figure 4. Composite multi-proxies Core SO217-18517 between original and 18 interpolated data with time resolution of 0.01 kyrs. a. Composite δ Osw, b. Composite SST, c. Composite δ18O. These records present a significant high-resolution paired Mg/Ca-δ18O with average sampling interval ~10 years during the last deglacial until early Holocene in maritime continent providing hydrological changes during the past 14,000 years.

4.3. Wavelet Power Spectrum 18 18 Three multi-proxies of core SO217-18517 including δ Osw, SST, and δ O show high- frequency variations at centennial and millennial timescale. These multi-proxies have been interpolated and normalized to evenly space data with time resolution of 0.01 kyrs (Figures 5a, 6a, 7a). These multi-proxies will be explained in each section below:

18 4.3.1. δ Osw reconstruction 18 Wavelet power spectrum of δ Osw reveals a number of centennial from 9 to 14 kyrs. 18 Result of wavelet analysis shows that ~ 1.1-kyrs or 1100-years oscillation of δ Osw core SO217-18517 is the most stable and dominant among periodicities between 9 and Centennial-Millennial Climate Variability in the Makassar Strait… 207

14 kyrs (Figure 5b). Although, low-power global wavelet spectrum on centennial and millennial-climate variability (Figure 5c) but those climate variability can be traced on wavelet power spectrum in a range of 95% confidence level (Figure 5b). An increase 18 18 in (heavy) δ Osw during 10-11 kyrs and 12-13 kyrs associated with an increase in δ Osw average variance between 0.01-0.04 ‰ using bandpass filter output of 0.1-1 kyrs 18 (Figure 5d). In contrast, light δ Osw have slightly constant average variance during 9- 10 kyrs, 11-12 kyrs, and 13-14 kyrs. Occurrences of several centennial-scale variability including 300 years, 150 years, and 63 years is getting stronger on periods of heavy 18 δ Osw at 10-11 kyrs and 12-13 kyrs (Figures 5b-c).

18 Figure 5. a. δ Osw reconstruction of core SO217-18517 after interpolation to evenly 18 spaced data with time resolution of 0.01 kyrs; b. Wavelet power spectrum δ Osw core SO217-18517 (High power is indicated by red whereas low power is indicated by blue), areas cycles by black solid lines represent the confidence level greater than 95 %; c. 18 Global wavelet spectrum using mean power spectrum δ Osw signal from 9 to 14 kyrs 1, 0.3, and 0.15 kyrs periodicities; orange dotted lines indicate confidence level greater than 95 %; d. Average variance with band pass filter skala ±0.05 -0.03 kyrs. It shows 18 increasing δ Osw variance during 10-11 kyrs and 12-13 kyrs. Significant power spectrum shown at millennial timescale of 1.1 kyrs or 1100 years-periodicity.

4.3.2. Sea surface temperature (SST) Wavelet power spectrum of SST reveals a number of centennial-timescale in periodicity of 125, 230 and 500-years (Figure 6b). A dominant climate signal of centennial-scale on 500-years periodicity is the most stable and dominant during 9-14 kyrs

208 Marfasran Hendrizan et al.

(Figures 6b-c). Bandpass filter output (after the 0.1-1 kyrs bandpass filter) of SST reconstructions shows SST variance between 0.03-0.17 °C (Figure 6D). 230 years periodicity is the second dominant signal during 11-14 kyrs and followed by 125 years periodicity during 9-11 kyrs. A trend of 500-years periodicity in our SST records strengthened from 10-14 kyrs and reduced after 10 kyrs (Figure 6b).

Figure 6. a. SST reconstruction of core SO217-18517 after interpolation to evenly spaced data with time resolution of 0.01 kyrs; b. Wavelet power spectrum SST core SO217-18517 (High power is indicated by red whereas low power is indicated by blue), areas cycles by black solid lines represent the confidence level greater than 95 %; c. Global wavelet spectrum using mean power spectrum SST signal from 9 to 14 kyrs with dominant signal 125, 230, and 500-years periodicities; orange dotted lines indicate confidence level greater than 95 %; d. Average variance with band pass filter skala ±0.1 -1 kyrs. Significant power spectrum occurs at centennial timescale of 0.5 kyrs.

4.3.3. δ18O reconstruction Wavelet power spectrum of δ18O core SO217-18517 reveals little variance signals (Figures 7b, d). The result shows a statistically confidence above 95% level centennial periodicity centered at 375 and 1000-years periodicity (Figures 7b, c). The result of wavelet analysis shows that ~ 1000-years periodicity of δ18O at core SO217-18517 are the most stable and dominant among periodicities of 9 to 14 kyrs (Figure 7b). Bandpass filter output (after the 0.05-1 kyrs bandpass filter) of δ18O reconstructions shows variance between 0.01-0.13‰ Centennial-Millennial Climate Variability in the Makassar Strait… 209

5. DISCUSSION The ~1100-years or 1.1-kyrs and 1000-years periodicities are detected in planktonic 18 18 foraminifera δ Osw and δ O records of SO217-18517 between 9 and 14 kyrs timeseries (Figures 5b & 7b). These millennial-scale variabilities of Core SO217-18517 confirm similar millennial-scale in Holocene planktonic records in the North Atlantic and the North Pacific (Khider et al., 2014; Lüning & Vahrenholt, 2016; Oppo et al., 2009; Wanner et al., 2008) which extended to the last deglaciation in the Makassar Strait. However, SST records SO217-18517 represent centennial-scale variabilities (Figure 6b) which is contrast to millennial-scale Holocene SST records in the North Pacific (Khider et al., 2014). In the next sections, we will explain this centennial to millennial- scale climate variabilities in the Makassar Strait. We will also compare our records with other records in the central Indonesian region to see how the salinity changes response to precipitation particularly in the period of 10-11 kyrs and 12-13 kyrs.

Figure 7. a. δ18O reconstruction of core SO217-18517 after interpolation to evenly spaced data with time resolution of 0.01 kyrs; b. Wavelet power spectrum SST core SO217-18517 (High power is indicated by orange whereas low power is indicated by blue), areas cycles by black solid lines represent the confidence level greater than 95 %; orange dotted lines indicate confidence level greater than 95 %; c. Global wavelet spectrum using mean power spectrum SST signal from 9 to 14 kyrs with dominant signal 375 and 1000-years periodicities; d. Average variance with band pass filter skala ±0.05 -1 kyrs. Significant power spectrum occurs at millennial timescale of 1000 years.

210 Marfasran Hendrizan et al.

5.1. Centennial-Millennial-scale variability since the last deglaciation 18 18 The δ Osw and δ O records on G. ruber of Core SO217-18517 are characterized by millennial-scale variability with 1100-years and 1000-years periodicity between 9 and 14 kyrs (Figures 5 and 7). Wavelet power spectrum indicates a period of millennial- 18 18 scale variability is the most stable event in the entire δ Osw and δ O records between 9 and 14 kyrs (Figures 5b & 7b), even though this millennial power occurs below 95 % 18 confidence level (Figure 5c & 5d). We see δ Osw trends are strong on millennial-scale variability in the Makassar Strait shown by the wavelet power spectrum (Figure 5b). This millennial-scale variability is robust based on age from the previous study (Hendrizan et al., 2017), analytical, and calibration uncertainties (Figures 3 and 4). In 18 addition, a dominant signal of 1100-years periodicity from δ Osw Core SO217-18517 can be found as well at Mindanao marine sediment core MD81 (Khider et al., 2014). Signal of 1100 years periodicity at Mindanao during Holocene probably extend to last deglacial period in the Makassar Strait as shown in Figure 5b. We propose mixture forcing between the deep ocean and solar radiation could be an amplifier for forcing and altering millennial-scale variability from 9 to 14 kyrs at our core location. 18 The most prominent δ Osw maximum occurs at 10-11 kyrs and 12-13 kyrs, coincides with the Early Holocene and Younger Dryas interval (Figure 4). We interpret the peak 18 of δ Osw at 10-11 kyrs and 12-13 kyrs as a response to a decreased in precipitation and a reduced in Mahakam River input into the Mahakam Strait. In those period at 10-11 kyrs and 12-13 kyrs, several centennial-scale variability including 300 years, 150 years, and 63 years are stronger than other periods at 9-10 kyrs, 11-12 kyrs, and 13-14 kyrs, which indicate strengthening centennial-scale periodicity (Figure 5b). Those centennial-scale variability can be related to strengthening of solar radiative forcing (Liu et al., 2009). It would affect intensified precipitation in tropical region (Novello et al., 2016) particularly regions influenced by Australia-Indonesia Monsoon (Mohtadi et 18 al., 2016). This mechanism probably could explain an increase in δ Osw during 10-11 kyrs and 12-13 kyrs with existing several centennial-scale variability in our core SO217-18517 (Figure 5b).

5.2. Mixture forcing of climate variability Wavelet analysis of our SST time series (Fig. 6) reveals a confidence period of ~500- years periodicity. A common periodicity of ~500 years is found in global ocean from North Atlantic to the East Asia region (H. Cheng et al., 2015; Zielhofer et al., 2019). A reduced in North Atlantic Deepwater (NADW), cooling of the ocean surface and high- latitude continent around North Atlantic and North Pacific, weakening of the Asian monsoon (H. Cheng et al., 2015; Hai Cheng et al., 2012) could be mechanisms in controlling 500-years periodicity at SST records Core SO217-18517 (Figure 6b). To explain deep water circulation on our SST records, we see on log (Zr/Rb) as proxy of Centennial-Millennial Climate Variability in the Makassar Strait… 211 bottom current intensity in the ITF region from Core SO217-18517 (Hendrizan et al., 2017). Intriguing finding show the ITF intensity based on log (Zr/Rb) (Figure 8g; left panel) associated with replacement of the ITCZ from 10 to 14 kyrs (Figure 8). This evidence indicates likely the ITF transfer heat to our Core SO217-18517 site during 10- 14 kyrs which is reflected by 500-years periodicity in our SST records (Figure 5b). We also looked furher on solar radiation forcing from Jun-August local insolation records near Core SO217-18517 (Figure 8g: right panel). When the the ITCZ shifted northward 18 during 11-12 kyrs and 13-14 kyrs, light δ Osw (Figure 8a) associated with both increase in solar radiation forcing and stronger ITF intensity (Figure 8g). In contrast, southward 18 movement of the ITCZ reflects heavy δ Osw associated with opposite pattern between solar radiation forcing and the ITF intensity during 10-11 kyrs and 12-13 kyrs (Figure 18 8). Here we suppose the deep ocean is not the main forcing in driving SST and δ Osw in this core-site, other external forcing from solar radiation is the second forcing to change climate proxy of Core SO217-18517.

5.3 Comparison hydroclimate proxy records Today, precipitation patterns over Kalimantan and Sulawesi are associated with the ITCZ movement along these region (Hendrizan et al., 2017; Konecky et al., 2016). Precipitation pattern in Kalimantan shows the ITCZ position shifts twice which cause rain falls along the year with two precipitation peaks on May and October (Dubois et al., 2014). Precipitation pattern in Sulawesi indicates the ITCZ shifts to the south during December-January with maximum precipitation and minimum precipitation when reverse movement of the ITCZ to the north during June-August (Dubois et al., 2014). 18 Multi-factors in controlling δ Osw reconstruction include sea level, current circulation, and river discharge (Schröder et al., 2018) need to be considered when interpreting 18 fluctuation of precipitation changes in the past. δ Osw from marine sediment cores of the northwest Banda Sea (TGS-931) and Makassar Strait (SO217-18517, 18519, and 18515) reflects seasonality of rainfall in Kalimantan and Sulawesi (Schröder et al., 2018). Those records will be compared with δ18O Gunung Buda speleothem records

(Cobb et al., 2007) and δDprecip (Konecky et al., 2016) as proxy of precipitation to more comprehend interpretation of Kalimantan and Sulawesi precipitation pattern during the last deglacial until early Holocene period. Temporal evolution of proxy records occurred in the central Indonesia (Fig. 8), 18 indicating a decrease in δ Osw particularly during 10-11 ka and 12-13 ka in the Makassar Strait. It is also associated with reduced precipitation in Kalimantan (Fig. 8e) 18 but the changes is uncertain in Sulawesi (Figs. 8b-d). An increase δ Osw is represented

212 Marfasran Hendrizan et al. as reduced in Mahakam River runoff in Core SO217-18517 and 18519 (Figure 8). 18 Similar changes of δ Osw especially during 10-11 ka corresponding to our core sites also found in marine core MD98-2178 at Celebes Sea (Fan et al., 2013). However, records in Sulawesi show more complex mechanisms during 10-11 ka and 12-13 ka (Figure 8). During 12-13 ka, marine sediment cores TGS-931 and SO217-18515 from 18 Sulawesi has same pattern of an increase in δ Osw with marine sediment cores off Mahakam Delta. Those records also associated with reduced precipitation based on 18 δDprecip from Lake Towuti (Figure 8). In contrast, light δ Osw reflected more fresh water supplied to the marine sediment cores from Sulawesi region which are consistent with increase in precipitation at Lake Towuti during 10-11 ka. We interpret difference between Kalimantan and Sulawesi during 10-11 kyrs due to movement of the ITCZ to the south but Sulawesi has received more rainfall and Kalimantan is still drying. A proposed mechanism in driving precipitation pattern at Kalimantan and Sulawesi could be related to the ITCZ position changes during the end of the last deglacial period until the early Holocene. During the period of 12-13 ka which was coincided with the YD, the ITCZ shifted further to the south (Figure 8). Therefore, it will reduce the Australian-Indonesian Summer Monsoon in almost entire region of Indonesia, including Kalimantan and Sulawesi. Enhanced precipation in the Flores and Australia during the YD (Ayliffe et al., 2013; Griffiths et al., 2009; Kuhnt et al., 2015). The influence of ocean circulation via AMOC during the last deglaciation (Henry et al., 2016) to atmospheric variability in the central Indonesia could be terminated at 10 kyrs 18 as both the ITF and solar radiation weakened but light δ Osw which indicates the ITCZ shifted to the north (Figure 8). Centennial-scale variability of SST with 500-years periodicity show coherency of stronger centennial-scale from 10 to 14 kyrs (Figure 6). However, this interpretation still need to be examined in the Holocene records in the future. We suppose centennial and millennial-scale variability of AMOC intensity associated with the ITF contribute to climate variability in the central Indonesia until 10 kyrs ago. Centennial-Millennial Climate Variability in the Makassar Strait… 213

Figure 8. Comparison of hydroclimate proxies between Kalimantan and Sulawesi. a) 18 Interpolated δ Osw data Core SO217-18517 offshore Mahakam Delta with smoothing 18 18 trend in red line, b,f) δ Osw Core TGS-931 in the northwest Banda Sea and δ Osw Core 18 SO217-18519 in the offshore Mahakam Delta (Schröder et al., 2018) c) δ Osw Core

SO217-18515 in the Mandar Bay, Offshore Sulawesi (Schroder et al., 2016), d) δDprecip records from Lake Towuti (Konecky et al., 2016), e) δ18O stalagmite records from

214 Marfasran Hendrizan et al.

Gunung Buda, Kalimantan (Partin et al., 2007), g) Left panel: Log (Zr/Rb) based on X- ray fluorescence (XRF) measurements at 1 cm intervals from Core SO217-18517 (Hendrizan et al., 2017), right panel: mean summer (21 June to 21 September) 18 insolation at 0°S (Laskar et al., 2004). Green shaded colors indicate heavy δ Osw on 10-11 kyrs and 12-13 kyrs as reduced precipitation in Kalimantan during those periods.

6. CONCLUSION Marine Sediment core SO217-18517 records the Makassar Strait climate variability on a wide range between early Holocene and the end of the last deglaciation. During the 18 18 period between 9 and 14 kyrs, multi-proxies data (δ Osw, SST, and δ O) show high- 18 18 frequency variations at centennial and millennial timescale. δ Osw and δ O records show the confidence power spectrum occurs at millennial time scales of ~1100-years and 1000-years periodicity. However, the confidence power spectrum with ~500 years periodicity (centennial) of SST is more stable than millennial time scales. Our results 18 find peaks of δ Osw at 10-11 kyrs and 12-13 kyrs as a response to a decrease in precipitation and a reduced in Mahakam River input into the Mahakam Strait. In those period at 10-11 kyrs and 12-13 kyrs, several centennial-scale variability including 300 years, 150 years, and 63 years are stronger than other periods at 9-10 kyrs, 11-12 kyrs, and 13-14 kyrs, which indicate strengthening centennial-scale periodicity. A proposed mechanism in driving precipitation pattern at Kalimantan and Sulawesi could be related to the ITCZ position changes during the end of the last deglacial period until the early Holocene. The influence of ocean circulation via AMOC to atmospheric variability in the central Indonesia could be terminated at 10 kyrs as both the ITF and solar radiation 18 weakened but light δ Osw which indicates the ITCZ shifted to the north.

Author Contributions: All the authors are the main contributors to this manuscript. MH wrote the manuscript and analyzed the data; NSN, SYC, and MRP wrote and reviewed the manuscript; BS and IPA analyzed the data using wavelet analysis, DAU and VCA drew the map, analyzed regional oceanography the Makassar Strait.

Acknowledgments: We gratefully acknowledge the Deutsche Forschungsgemeinchaft (grant KU649/29-1 to Wolfgang Kuhnt) for funding this geochemical analysis in this research and the Germany Ministry for Education and Science (BMBF, grant SO-217, and MAJA, 03G0217A to Wolfgang Kuhnt) for funding the Sonne-217 Cruise. We thank Beasiswa Saintek Kemenristek/BRIN for funding the publication fee.

Centennial-Millennial Climate Variability in the Makassar Strait… 215

Conflicts of Interest: The authors declare no conflict of interest

REFERENCES Anand, P., Elderfield, H., & Conte, M. H. (2003). Calibration of Mg/Ca thermometry in planktonic foraminifera from a sediment trap time series. Paleoceanography, 18(2), 1–15. https://doi.org/10.1029/2002PA000846 Ayliffe, L. K., Gagan, M. K., Zhao, J., Drysdale, R. N., Hellstrom, J. C., Hantoro, W. S., et al. (2013). Rapid interhemispheric climate links via the Australasian monsoon during the last deglaciation. Nature Communications, 4(2908), 1–6. https://doi.org/10.1038/ncomms3908 Bakker, P., Clark, P. U., Golledge, N. R., Schmittner, A., & Weber, M. E. (2017). Centennial-scale Holocene climate variations amplified by Antarctic Ice Sheet discharge. Nature Publishing Group. https://doi.org/10.1038/nature20582 Bemis, B. E., Spero, H. J., Bijma, J., & Lea, D. W. (1998). Reevaluation of the oxygen isotopic composition of planktonic foraminifera: Experimental results and revised paleotemperature equations. Paleoceanography, 13(2), 150–160. https://doi.org/10.1029/98PA00070 Bond, G., Kromer, B., Beer, J., Muscheler, R., Evans, M. N., Showers, W., et al. (2001). Persistent solar influence on north atlantic climate during the Holocene. Science, 294(5549), 2130–2136. https://doi.org/10.1126/science.1065680 Cheng, H., Sinha, A., Verheyden, S., Nader, F. H., Li, X. L., Zhang, P. Z., et al. (2015). The climate variability in northern Levant over the past 20,000 years. Geophysical Research Letters, 42(20), 8641–8650. https://doi.org/10.1002/2015GL065397 Cheng, Hai, Sinha, A., Wang, X., Cruz, F. W., & Edwards, R. L. (2012). The Global Paleomonsoon as seen through speleothem records from Asia and the Americas. Climate Dynamics, 39(5), 1045–1062. https://doi.org/10.1007/s00382-012-1363-7 Cobb, K. M., Adkins, J. F., Partin, J. W., & Clark, B. (2007). Regional-scale climate influences on temporal variations of rainwater and cave dripwater oxygen isotopes in northern Borneo. Earth and Planetary Science Letters, 263, 207–220. https://doi.org/10.1016/j.epsl.2007.08.024 Debret, M., Sebag, D., Crosta, X., Massei, N., Petit, J., Chapron, E., & Bout- Roumazeilles, V. (2009). Evidence from wavelet analysis for a mid-Holocene transition in global climate forcing. Quaternary Science Reviews, 28(25–26), 2675–2688. https://doi.org/10.1016/j.quascirev.2009.06.005

216 Marfasran Hendrizan et al.

Dubois, N., Oppo, D. W., Galy, V. V., Mohtadi, M., Van Der Kaars, S., Tierney, J. E., et al. (2014). Indonesian vegetation response to changes in rainfall seasonality over the past 25,000 years. Nature Geoscience, 7(7), 513–517. https://doi.org/10.1038/ngeo2182 Fan, W., Jian, Z., Bassinot, F., & Chu, Z. (2013). Holocene centennial-scale changes of the Indonesian and South China Sea throughflows: Evidences from the Makassar Strait. Global and Planetary Change, 111, 111–117. https://doi.org/10.1016/j.gloplacha.2013.08.017 Gordon, A. L. (2005). Oceanography of the Indonesian Seas. Oceanography, 18(4), 14–27. https://doi.org/10.5670/oceanog.2005.18 Griffiths, M. L., Drysdale, R. N., Gagan, M. K., Zhao, J., Ayliffe, L. K., Hellstrom, J. C., et al. (2009). Increasing Australian – Indonesian monsoon rainfall linked to early Holocene sea-level rise. Nature Geoscience, 2, 636–639. https://doi.org/10.1038/ngeo605 Hendrizan, M., Kuhnt, W., & Holbourn, A. (2017). Variability of Indonesian Throughflow and Borneo Runoff During the Last 14 kyr. Paleoceanography, 32, 1054–1069. https://doi.org/10.1002/2016PA003030 Henry, L. G., Mcmanus, J. F., Curry, W. B., Roberts, N. L., Piotrowski, A. M., & Keigwin, L. D. (2016). North Atlantic ocean circulation and abrupt climate change during the last glaciation. Science, 353(6298), 470–474. https://doi.org/10.1126/science.aaf5529 Katz, M. E., Cramer, B. S., Franzese, A., Hönisch, B., Miller, K. G., Rosenthal, Y., & Wright, J. D. (2010). Traditional and Emerging Geochemical Proxies in Foraminifera. The Journal of Foraminiferal Research, 40(2), 165–192. https://doi.org/10.2113/gsjfr.40.2.165 Khider, D., Jackson, C. S., & Stott, L. D. (2014). Assessing millennial-scale variability during the Holocene: A perspective from the western tropical Pacific. Paleoceanography, 29, 143–159. https://doi.org/10.1002/2013PA002534 Konecky, B., Russell, J., & Bijaksana, S. (2016). Glacial aridity in central Indonesia coeval with intensified monsoon circulation. Earth and Planetary Science Letters, 437, 15–24. https://doi.org/10.1016/j.epsl.2015.12.037 Kuhnt, W., Holbourn, A., Xu, J., Opdyke, B., Deckker, P. De, Röhl, U., & Mudelsee, M. (2015). Southern Hemisphere control on Australian monsoon variability during the late deglaciation and Holocene. Nature Communications, 6(5916), 1–7. https://doi.org/10.1038/ncomms6916 Centennial-Millennial Climate Variability in the Makassar Strait… 217

Laskar, J., Robutel, P., Joutel, F., Gastineau, M., Correia, A. C. M., & Levrard, B. (2004). A long-term numerical solution for the insolation quantities of the Earth. Astronomy and Astrophysics, 428(1), 261–285. https://doi.org/10.1051/0004- 6361:20041335 Lee, J. Y., & Wang, B. (2014). Future change of global monsoon in the CMIP5. Climate Dynamics, 42, 101–119. https://doi.org/10.1007/s00382-012-1564-0 Lepot, M., Aubin, J., & Clemens, F. H. L. R. (2017). Interpolation in Time Series : An Introductive Overview of Existing Methods , Their Performance Criteria and Uncertainty Assessment. Water, 9(796), 1–20. https://doi.org/10.3390/w9100796 Levi, C., Bassinot, F., Guichard, F., Cortijo, E., Waelbroeck, C., Caillon, N., et al. (2007). Low-latitude hydrological cycle and rapid climate changes during the last deglaciation. Geochemistry Geophysics Geosystems, 8, 1–11. https://doi.org/10.1029/2006GC001514 Liu, J., Bin, W., Ding, Q., Kuang, X., Soon, W., & Zorita, E. (2009). Centennial Variations of the Global Monsoon Precipitation in the Last Millennium : Results from ECHO-G Model. Journal of Climate, 22, 2356–2371. https://doi.org/10.1175/2008JCLI2353.1 Locarnini, R. A., Mishonov, A. V., Antonov, J. I., Boyer, T. P., Garcia, H. E., Baranova, O. K., et al. (2013). NOAA Atlas NESDIS 73 WORLD OCEAN ATLAS 2013 Volume 1 : Temperature, 1(September). Lüning, S., & Vahrenholt, F. (2016). The Sun’s role in climate. In Evidence-Based Climate Science: Data Opposing CO2 Emissions as the Primary Source of Global Warming: Second Edition (pp. 283–305). https://doi.org/10.1016/B978-0-12- 804588-6.00016-1 Mathworks. (2019). Data Analysis R 2019 b. The MathWorks, Inc. 1 Apple Hill Drive Natick, MA 01760-2098. Mohtadi, M., Oppo, D. W., Steinke, S., Stuut, J. W., Pol-holz, R. De, Hebbeln, D., & Lückge, A. (2011). Glacial to Holocene swings of the Australian – Indonesian monsoon. Nature Geoscience, 4, 540–544. https://doi.org/10.1038/ngeo1209 Mohtadi, M., Prange, M., & Steinke, S. (2016). Review Palaeoclimatic insights into forcing and response of monsoon rainfall. Nature, 533, 191–199. https://doi.org/10.1038/nature17450 Mohtadi, M., Prange, M., Schefuß, E., & Jennerjahn, T. C. (2017). Late Holocene slowdown of the Indian Ocean Walker circulation. Nature Communications, 8(1), 1–8. https://doi.org/10.1038/s41467-017-00855-3

218 Marfasran Hendrizan et al.

Morimoto, M., Abe, O., Kayanne, H., Kurita, N., Matsumoto, E., & Yoshida, N. (2002). Salinity records for the 1997 – 98 El Nino from Western Pacific corals. Geophysical Research Letters, 29(11), 1–4. https://doi.org/10.1029/2001GL013521 Muller, J., Kylander, M., Wüst, R. A. J., Weiss, D., Martinez-cortizas, A., Legrande, A. N., et al. (2008). Possible evidence for wet Heinrich phases in tropical NE Australia : the Lynch’s Crater deposit. Quaternary Science Reviews, 27, 468– 475. https://doi.org/10.1016/j.quascirev.2007.11.006 Novello, V. F., Vuille, M., Cruz, F. W., Stríkis, N. M., Paula, M. S. De, Edwards, R. L., et al. (2016). Centennial-scale solar forcing of the South American Monsoon System recorded in stalagmites. Scientific Reports, 6(24762), 1–8. https://doi.org/10.1038/srep24762 Oppo, D. W., Rosenthal, Y., & Linsley, B. K. (2009). 2,000-year-long temperature and hydrology reconstructions from the Indo-Pacific warm pool. Nature, 460, 1113– 1116. https://doi.org/10.1038/nature08233 Partin, J. W., Cobb, K. M., Adkins, J. F., Clark, B., & Fernandez, D. P. (2007). Millennial-scale trends in west Pacific warm pool hydrology since the Last Glacial Maximum. Nature, 449, 452–456. https://doi.org/10.1038/nature06164 Rannik, U., & Vesala, T. (1999). Autoregressive filtering versus linear detrending in estimation of fluxes by the eddy covariance method. Boundary-Layer Meteorology, 91, 259–280. Regenberg, M., Steph, S., Nürnberg, D., Tiedemann, R., & Garbe-schönberg, D. (2009). Calibrating Mg / Ca ratios of multiple planktonic foraminiferal species with δ18O-calcification temperatures : Paleothermometry for the upper water column. Earth and Planetary Science Letters, 278, 324–336. https://doi.org/10.1016/j.epsl.2008.12.019 Schroder, J. F., Holbourn, A., Kuhnt, W., & Küssner, K. (2016). Variations in sea surface hydrology in the southern Makassar Strait over the past 26 kyr. Quaternary Science Reviews, 154, 143–156. Schröder, J. F., Kuhnt, W., Holbourn, A., Beil, S., & Zhang, P. (2018). Deglacial Warming and Hydroclimate Variability in the Central Indonesian Archipelago. Paleoceanography and Paleoclimatology, 974–993. https://doi.org/10.1029/2018PA003323 Sprintall, J., Gordon, A. L., Wijffels, S. E., Feng, M., Hu, S., Koch-Larrouy, A., et al. (2019). Detecting Change in the Indonesian Seas. Frontiers in Marine Science, 6, Centennial-Millennial Climate Variability in the Makassar Strait… 219

1–24. https://doi.org/10.3389/fmars.2019.00257 Stott, L., Cannariato, K., Thunell, R., Haug, G. H., Koutavas, A., & Lund, S. (2004). Decline of surface temperature and salinity in the western tropical Pacific Ocean in the Holocene epoch. Nature, 431, 2–5. https://doi.org/10.1038/nature02809.1.2.3.4.5.6.7.8.9.10.11.12.13.14.15. Torrence, C., & Compo, G. P. (1997). A Practical Guide to Wavelet Analysis. Bulletin of the American Meteorological Society, 79(1), 61–78. https://doi.org/10.1016/j.biopha.2017.10.142 Visser, K., Thunell, R., & Stott, L. (2003). Magnitude and timing of temperature change in the Indo-Pacific warm pool during deglaciation. Nature, 421, 3667–3670. https://doi.org/10.1038/nature01331.1. Wanner, H., Beer, J., Bütikofer, J., Crowley, T. J., Cubasch, U., Flückiger, J., et al. (2008). Mid- to Late Holocene climate change: an overview. Quaternary Science Reviews, 27, 1791–1828. https://doi.org/10.1016/j.quascirev.2008.06.013 Xie, P., & Arkin, P. A. (1997). Global Precipitation: A 17-Year Monthly Analysis Based on Gauge Observations, Satellite Estimates, and Numerical Model Outputs. Bulletin of the American Meteorological Society, 78(11), 2539–2558. https://doi.org/10.1175/1520-0477(1997)078<2539:GPAYMA>2.0.CO;2 Zielhofer, C., Köhler, A., Mischke, S., Benkaddour, A., Mikdad, A., & Fletcher, W. J. (2019). Western Mediterranean hydro-climatic consequences of Holocene ice- rafted debris (Bond) events. Climate of the Past, 15(2), 463–475. https://doi.org/10.5194/cp-15-463-2019

220 Marfasran Hendrizan et al.

Supplementary Figure 1. Histogram of interpolated SO217-18517 dataset. a. Normal distribution of sea surface temperature (SST), b. Normal distribution of 18 18 δ Osw, c. Normal distribution of δ O

Supplementary Figure 2. Coefficient correlation between original and interpolated 18 18 SO217-18517 dataset. a. Sea surface temperature (SST), b. δ Osw, c. δ O