Pliocene expansion of C4 vegetation in the core monsoon zone on the Indian Peninsula

The expansion of C4 vegetation during the Neogene was one of the largest reorganizations of Earth’s terrestrial biome. Once thought to be globally synchronous in the late Miocene, site-specific studies have revealed differences in the 10 timing of the expansion and suggest that local conditions play a substantial role. Here, we examine the expansion of C4 vegetation on the Indian Peninsula since the late Miocene by constructing a ~6 million year paleorecord with marine sediment from the Bay of Bengal at Site U1445 drilled during International Ocean Discovery Program Expedition 353. Analyses of element concentrations indicate the marine sediment originates from the Mahanadi River in the Core Monsoon Zone (CMZ) of the Indian Peninsula. Hydrogen isotopes of the fatty acids of leaf waxes reveal an overall decrease in the CMZ precipitation 15 since the late Miocene. Carbon isotopes of the leaf wax fatty acids suggest C4 vegetation on the Indian Peninsula existed before the end of the Miocene, but expanded to even higher abundances during the mid-Pliocene to mid-Pleistocene (~3.5 to 1.5 million years ago). Similar to the CMZ on the Indian Peninsula, a Pliocene expansion or re-expansion has previously been observed in northwest Australia and in East Africa, suggesting that these tropical ecosystems surrounding the Indian Ocean remained highly sensitive to changes in hydroclimate after the initial spread of C4 plants in late Miocene. 20

regions. Resolving the precise timing and factors leading to major changes in vegetation demands site-specific studies 30 (Strömberg et al., 2011;Zhou et al., 2014).
Our study provides a novel piece of the puzzle in unraveling the complexities of C4 expansion by constructing a 6 million year (Myr) record of C4 vegetation and aridity on the Indian Peninsula. The marine sediment record is from International Ocean Discovery Program (IODP) Site U1445 (17°44.72'N, 84°47.25'E; 2,503m water depth; Fig. 1A) drilled in the Bay of Bengal (BoB) close to the mouths of the Mahanadi River. Lithologies at Site U1445 include calcareous fossils, 35 biosilica, silt, and clays (including glauconite), and are overall described as hemipelagic sediment (Clemens et al., 2016). The Indian Monsoon dictates climate patterns in the Mahanadi River drainage basin: rainy summers, dry winters, and an annual reversal of wind direction (Gadgil, 2003;Sarkar et al., 2015). Highly sensitive to the seasonal changes, more than 80% of runoff from the Mahanadi River into the BoB occurs during the summer (Chakrapani and Subramanian, 1990).
Although agriculture dominates present-day vegetation, models of seasonal climate predict the natural flora of the Mahanadi basin would be closed-canopy, moist deciduous forests and moist-to-dry woodlands with rare open spaces (Fig. 1C, Zorzi et al., 2015 and references therein). Today the region encompasses a range of C3 and C4 vegetation, but proxies and models suggest that the plant communities are highly sensitive to glacial-interglacial changes with nearly all flora utilizing a 50 C4 pathway during the last glacial maximum (Galy et al., 2008;Phillips et al., 2014;Zorzi et al., 2015, and references therein).
The behavior of vegetation in the CMZ over million-year timescales is unknown.
Here, we use inorganic bulk geochemical analyses to fingerprint the origin of sediment at Site U1445 to be from the Mahanadi River. Then we use bulk organic and compound-specific biomarkers at the same site, including carbon and hydrogen isotope measurements of leaf wax fatty acids, to reconstruct the changes in C4 vegetation and rainfall in the CMZ of the Indian 55 Peninsula over the last ~6 Myr (Fig. 2).
We fit a locally weighted spline to biostratigraphic and magnetostratigraphic age constraints from Hole U1445A (Clemens et al., 2016) using CLAM software in R (Blaauw, 2010) to estimate the age of our samples (Fig. S1). Our samples and the age observed thin turbiditic sequences (~2-20cm thick) throughout Site U1445 and the expansion and dissociation of gas hydrates upon recovery that may muddle a higher-resolution record (Clemens et al., 2016). However, Site U1445 has fewer and smaller turbidite deposits relative to other sites drilled in this region and the records spanning million-year timescales are likely relatively undisturbed. 65 To determine sediment provenance, we measured major, trace, and rare earth element concentrations on 30 bulk sediment samples spanning 0 to 6 Myr including light and dark layers of sediment (Sec. 2.1.). To reconstruct hydrological and vegetation changes, we analyzed bulk organics (Sec. 2.2.) as well as compound-specific biomarkers (Sec. 2.3.) from 57 samples. We constructed the sampling plan to characterize the differences in organic content and isotope composition between the lighter and darker layers of sediment. As such, samples for organic and biomarker analysis were collected from Site U1445 70 in pairs, visually targeting relatively light and dark layers at similar depths to capture the variability range on shorter timescales while characterizing longer trends.

Inorganic analyses of bulk major, trace, rare earth element concentrations
The samples we analyzed for major, trace, and rare earth element concentrations were originally collected for moisture and density (MAD) measurements onboard the JOIDES Resolution during IODP Expedition 353. Each sample was 75 collected with a 2 cm diameter plastic syringe that fits into the top of a 10 cm 3 glass vial, allowing for the vial to be completely filled with sediment (Clemens et al., 2016). The samples were dried in a convective oven at 105°C ± 5°C for 24 hours (Clemens et al., 2016). The remaining sample preparation, digestions, and analyses were conducted at Boston University and a detailed description of the analytical geochemical procedures are presented in . In summary here, sediment samples were hand-powdered with an agate mortar and pestle. For major elements, sample powders were digested by flux fusion 80 (Murray et al., 2000) and analyzed by inductively coupled plasma-emission spectrometry (ICP-ES). For analysis of trace and rare earth elements, sample powders were dissolved in a heated acid cocktail (HNO3, HCl, and HF, with later additions of HNO3 and H2O2 after samples were dried down) under clean-lab conditions and analyzed by inductively couple plasma-mass spectrometry (ICP-MS). Three separate digestions of a matrix-matched in-house sediment standard were analyzed with each batch and determined precision [(standard deviation)/(average) x 100] was ~2% of the measured value for each element. The 85 international Standard Reference Material BHVO-2 was analyzed as an unknown with each batch and results were consistently found to be accurate within precision for each element.

Analyses of carbon and nitrogen content and isotopes
Analyses of the abundance of total carbon (TC), total inorganic carbon (TIC), total organic carbon (TOC), nitrogen (N), and the δ 13 C of the TOC component were performed at Woods Hole Oceanographic Institution and methods are described 90 in Whiteside et al. (2011). In brief here, samples for TOC were weighed into tared silver boats and then acidified to remove carbonates in a closed desiccator for 3 days at 60-65°C over concentrated hydrochloric acid. All samples were flash combusted in a Costech 4010 Elemental Analyzer coupled via a Finnigan-MAT Conflo-II interface to a Thermo DeltaVPlus isotope ratio mass spectrometer. Data were recorded and integrated using the Isodat software package. Post-run calculations were performed for blank corrections, quantifications, and final calibrations. 95

Analyses of compound specific biomarkers abundances and isotopes
The analyses of compound-specific biomarkers were performed at Brown University (e.g., Daniels et al., 2017).
Samples were freeze-dried and lipids were extracted from 3.5 to 4.5 g of sediment using a Dionex 350 Accelerated Solvent Extractor (ASE) with dichloromethane:methanol (9:1 v/v). The fatty acids in the total lipid extract were separated from the neutral lipids using aminopropyl silica gel chromatography, eluting with a dichloromethane:isopropanol solution followed by 100 ether with 5% acetic acid.
The fatty acids were methalyated to form fatty acid methyl ester (FAME) by dissolving dried down acid fraction in in ~0.3 mL of toluene and ~1mL of 5:95 acetyl chloride:methanol with known isotopic composition. Nitrogen replaced the headspace in the vial before they were capped tightly and heated at 60°C for 12 hours. Once the reaction was complete, the FAMEs were separated from the water by-products formed during the methylation reaction. Sample received ~1mL of 105 synthetic saline solution (50g NaCl/L of double-distilled water) and ~1mL of hexane, were vigorously shaken, and then allowed to rest until the hexane separated from the water. The hexane fraction was pipetted into a new vial, avoiding the water.
Another ~1mL of hexane was added to the sample, shaken, and pipetted into the new vial. To clean the solution and isolate the fatty acids, samples were run through a second silica gel column, eluting with hexane to remove unwanted acids and then DCM to acquire the clean FAME fraction. 110 The FAME fraction was analyzed on an Agilent 6890 gas chromatograph with a flame ionization detector (GC-FID).
Sample blanks were analyzed with every batch. The isotope ratios of the FAME fraction (δDn-acid and δ 13 Cn-acid) were measured on a Thermo Finnigan Delta + XL isotope ratio mass spectrometer with a HP 6890 gas chromatograph and a high-temperature pyrolysis reactor for sample introduction. For δD, three injections of each sample were analyzed and two injections of each sample were analyzed for δ 13 C. Between every six injections, an in-house lab standard mixture containing known amounts of 115 various n-acids was analyzed to monitor instrument accuracy and precision. The δD and δ 13 C were corrected for the methyl groups added during derivatization using the following equations δDcorrected=[(2n+2)*δDmeasured+123.7*3]/(2n-1) and δ 13 Ccorrected =[(n+1)*δ 13 Cmeasured + 36.52]/n, where n is the carbon chain length of the compound and the methanol added has δD=-123.6‰ and δ 13 C=-36.52‰. Analytical uncertainty was calculated by [standard deviation/average] of the injections and is typically less than 3% for δD and less than 1% for δ 13 C. The standard deviations are reported in Table S3. For every 120 instrument run, samples were analyzed in random order. 2012). Then we approximate that C3 plants have a δD that is 30 ‰ lighter than C4 plants (Smith and Freeman, 2006;Chikaraishi and Naraoka, 2007; and use the fraction of C3 versus C4 to correct the δD of each sample for the differences in plant physiology (Fig. S5). The corrected data shows the same overall trend as the uncorrected data, except the values are shifted to be more negative and the overall change in δD is steeper (Fig. S5).

Results 130
To determine the provenance of the aluminosilicate fraction, we examined the proportions of Al, Ti, Sc, Nb, La, and Th concentrations, because other elements (e.g., Fe, K, Mg, Si, Zr, Hf) may be affected by continental weathering, sorting during transport, and post-depositional authigenic processes. The results from 30 samples have almost constant element proportions of the selected elements, indicating that the aluminosilicate fraction of sediment did not significantly vary over the past 6 Myr. The composition of the 30 samples, even amongst the light and dark layers, matches the composition of 135 lithologies that comprise the Mahanadi basin such as Precambrian granite and gneisses of the Indian craton and associated sedimentary deposits (Sharma, 2009;Fig. S2; Table S1). Marine sediment deposits in other parts of the Bay of Bengal closer to the Krishna and Godavari Rivers or Ganges-Brahmaputra Rivers have a more mafic or highly variable composition that is not observed at Site U1445 (e.g., Tripathy et al., 2014;Fig. S2). As such, we interpret our results as recording terrestrial changes in the CMZ, specifically the Mahanadi drainage basin. 140 The pairs of samples used for organic analyses are spaced ~28 m apart (~260 kyr intervals) and the adjacent light and dark layers within each pair were 0.2 m to 4.3 m apart in the sediment core (2 kyr to 46 kyr; Fig. S1). The color difference can be related to the total organic carbon content (wt%; TOC) and total nitrogen content with darker layers having 1.0 to 2.8 times more than adjacent lighter layers ( Fig. 2A; Fig. S3; Table S2).
Long-chain normal fatty acids of leaf waxes are derived from land plants and are well preserved during transport and 145 burial in marine sediment (Eglinton and Eglinton, 2008). We focus on the C30 chain length to avoid possible contaminations from non-terrestrial sources that contribute shorter chain length fatty acids ( Fig. S4; Table S3). The results of the δ 13 C of C30 fatty acid of leaf waxes (δ 13 CFA) show a 5‰ increase from mid-Pliocene to mid-Pleistocene (~3.5 to ~1.5 Ma), after which δ 13 CFA decreases and becomes more variable from ~1.5 Ma to the present (Fig. 2B). The hydrogen isotope compositions of the leaf wax fatty acids (δDFA) increase gradually over the past 6 Myr, but have a wide range amongst light and dark layers and 150 shorter time intervals ( Fig. 2C; Table S3). Comparing the earlier and later time intervals when δ 13 CFA changes, before the mid-Pliocene (3.5 Ma) δDFA ranges from -179‰ to -147‰ and after the mid-Pleistocene (1.5 Ma) the δDFA increases to between -166‰ to -126‰ (Fig. 2C).

C4 Expansion on the Indian Peninsula 155
The δ 13 CFA of terrestrial plants is primarily a function of the photosynthetic pathway and isotopic composition of atmospheric CO2 (e.g., Farquhar et al., 1989). In this study, the 5‰ increase in δ 13 CFA is greater than the reconstructed δ 13 C of atmospheric CO2 (≤1‰; Tipple et al., 2010), suggesting that a correction for δ 13 CCO2 would only slightly adjust our results.
Thus we interpret δ 13 CFA as reflecting the amount of C4 relative to C3 vegetation produced in the CMZ.
From ~6 Ma until ~3.5 Ma, approximately 51% to 81% (avg. 69% ± 9% s.d.) of the vegetation in the CMZ utilized 160 a C4 photosynthetic pathway (Fig. S5). Thus, the environmental threshold for C4 photosynthetic pathway had already been crossed before the end of the late Miocene. Later in the mid-Pliocene, the reconstruction shows another distinct expansion reaching 64% to 92% (avg. 81% ± 7% s.d.) of C4 vegetation in the early Pleistocene (Fig. 2B). The change in vegetation from ~3.5 to ~1.5 Ma suggests multiple steps of C4 expansion in the CMZ, rather than a singular late-Miocene expansion. From ~1.5 Ma to the present, the average proportion of C4 vegetation decreased and became more variable (58% to 92%, avg. 76% 165 ± 10% s.d.; Fig. 2B), which may reflect the sensitivity of the region to glacial-interglacial variations observed in shorter records from this region (e.g., Zorzi et al., 2015, and references therein).

Aridification of the Indian Peninsula
After correcting for the effects of plant physiology (Sec. 2.4.), the amount of precipitation and mixing of different air masses can each vary the hydrogen isotopic composition of leaf wax fatty acids (δDFA; e.g., Eglinton and Eglinton, 2008). The 170 mixing of two air masses with unique δD values was recently observed to drive δD of rainfall in New Delhi, India, but, similar to the amount of precipitation, the relatively depleted δD corresponded with wetter conditions (Hein et al., 2017). Thus, we interpreted the δDFA as a qualitative proxy for aridity or the relative amount of precipitation. The δDFA results suggest an overall drying of the CMZ on the Indian Peninsula over the past 6 Myr. The shorter-term scatter in the δDFA record may reflect higher frequency variations in aridity or rainfall. 175

Patterns of C4 Expansion surrounding the Indian Ocean
In this section, we compare our record of C4 expansion with other compound-specific biomarker records of C4 expansion at sites in the Indian Ocean or adjacent land and seas. Multiple proxy records document a late-Miocene C4 expansion in the Ganges or Brahmaputra River basins such as the Siwalik Group in northern Pakistan or BoB sites receiving outflow sediment (Fig. 3A France -Lanord and Derry, 1994;Quade and Cerling, 1995;Cerling et al., 1997;Freeman and Colarusso, 180 2001;Sanyal et al., 2004;Behrensmeyer et al., 2007;. Collectively, the reported timing of C4 expansions in the Himalaya region ranges from 9 to 5 Myr, most commonly 8 to 6 Myr (Behrensmeyer et al., 2007). Rather than a uniform timing, detailed sampling of various deposits around the Siwalik regions shows that C4 vegetation expansion was staggered amongst nearby sub-environments with different local conditions , and references therein). Another biomarker record documents a late-Miocene expansion in a wide continental region north and west of the Arabian Sea (Site 185 722; Fig. 1A, Fig. 3B; Huang et al., 2007). Once C4 vegetation expanded at each of these sites, the records suggest there is overall little or no systematic change in the amount of C4 vegetation after the late Miocene.
In contrast, the CMZ of the Indian Peninsula and a few other records around the Indian Ocean document an expansion of C4 vegetation during the Pliocene (Fig. 3). Marine deposits in the Gulf of Aden originate from northeast Africa and record a late-Miocene C4 expansion, followed by a relapse to predominantly C3 vegetation ~4.3 Ma and a re-expansion of C4 plants 190 in the Pliocene (Site 231; Fig. 1A, Fig. 3C; Feakins et al., 2005Feakins et al., , 2013Liddy et al., 2016). The Pliocene re-expansion is consistent with other records from tropical East Africa (e.g., Levin et al., 2004;Cerling et al., 2011). A C4 expansion in the Pliocene is also observed in northwest Australia (Site 763A; Fig. 3E; Andrae et al., 2018). There is little evidence of significant C4 vegetation prior to the Pliocene, suggesting a relatively late onset of C4 vegetation expansion in northwest Australia (Fig.   3). Additionally, there is evidence that East, South, and Central Asia experienced multiple steps of C4 expansion through the 195 Pliocene (e.g., Passey et al., 2009;Zhou et al., 2014;Miao et al., 2017;Koutsodendris et al., 2019).
Collectively, a significant regional expansion in the Pliocene, distinctly after the first late-Miocene expansion, is common at least amongst tropical East Africa, Northwest Australia, Asia, and the Indian Peninsula.

Triggers of C4 Expansion in the Pliocene
The adaptations of the C4 photosynthetic pathway provides a competitive advantage over C3 vegetation in 200 environmental conditions with low pCO2, high temperature, high aridity or extreme seasonality that can lead to frequent floods, droughts, and fires (e.g., Edwards et al., 2010, and references therein). Many studies hypothesize that the global expansion of C4 vegetation in the late Miocene was triggered by pCO2 decreasing below a temperature-dependent threshold (Herbert et al., 2016 and references therein). During the Pliocene, decreasing pCO2 may have also contributed to the expansion of C4 vegetation ( Fig. 3a; Pagani et al., 2009;Tripati et al., 2009;Seki et al., 2010;Bartoli et al., 2011). However, the heterogenous 205 regional response of the Pliocene C4 expansion suggests pCO2 cannot be the only driver. Given that the Pliocene C4 expansion is observed in multiple regions surrounding the Indian Ocean that have highly seasonal rainfall, regional hydrodynamics likely played a role in the Pliocene C4 expansion.
In the modern era, a complex interplay of multiple modes of variability dictate the unique seasonal and multi-decadal precipitation patterns in the regions surrounding the Indian Ocean. For example, rainfall on the Indian Peninsula follows 210 quintessential monsoon behavior, but is also tied to the Inter-Tropical Convergence Zone (ITCZ), Walker Circulation, and the Indian Ocean Dipole (IOD; Gadgil, 2003;Wang et al., 2017). The biannual rains of the (semi)arid tropical East Africa are related to the ITCZ, monsoon winds, sea surface temperature (SST), and Walker Circulation (e.g., Williams and Funk, 2011;Tierney et al., 2015;Yang et al., 2015). Monsoon rains annually quench northern Australia, but the El Niño Southern Oscillation (ENSO), Walker Circulation, and the amount of Indonesian Throughflow (ITF) better explain the precipitation in 215 other parts of Australia (Ummenhofer et al., 2009(Ummenhofer et al., , 2011a(Ummenhofer et al., , 2011b. Previous studies reason that changes in monsoon precipitation or frequency of fires likely triggered local expansion of C4 vegetation Passey et al., 2009;Zhou et al., 2014;Miao et al., 2017;Andrae et al., 2018). However, the sampling resolution of our million-year study cannot resolve these seasonal, decadal, or centennial rhythms in precipitation, so we examine the underlying processes driving variations in each of these modes of variability. 220 The physical mechanism of each of these modes of variability can be related back to atmospheric pressure gradients.
Monsoons are classically defined as a seasonal wind reversal induced by the pressure gradient caused by differential heating of land and sea (Wang et al., 2017). Broadly, the ITCZ is formed by the convergence and uplift of air masses near the equator due to differential heating between low and high latitudes and the formation of Hadley Cells. The IOD and ENSO are cyclic changes in the zonal gradient across the Indian and Pacific Oceans, respectively, and affect Walker circulation. Variable 225 Walker circulation affects ocean upwelling and thermocline depth, which in turn will influence sea surface temperature. Sea surface temperatures can interact with the atmosphere and reinforce or dampen the atmospheric pressure gradients affiliated with the monsoon, ITCZ, IOD, or ENSO dynamics. Changes in any combination of these atmospheric gradients may have altered the seasonal precipitation in the regions surrounding the Indian Ocean during the Pliocene (Wang et al., 2017).
What could have changed atmospheric pressure gradients in the Pliocene? Indonesian Throughflow was being 230 restricted from 5 to 3 Myr and there is evidence for strengthening SST gradients during the Pliocene (Cane and Molnar, 2001; Wara et al., 2005;Karas et al., 2009;Ford et al., 2012;Zhang et al., 2014;Burls and Fedorov, 2017;Christensen et al., 2017;White and Ravelo, 2020). Changes in the zonal and meridional SST gradients would have affected atmospheric pressure gradients in and surrounding the Indian Ocean and modified seasonal precipitation patterns. At the same time, the onset of northern hemisphere glaciation likely also helped reorganize the atmospheric pressure gradients surrounding the Indian Ocean 235 (e.g., Koutsodendris et al., 2019). Before the intensification of glaciation at 2.7 Ma, there is evidence of four early local glaciation events at 4.9-4.8 Myr, 4.0 Myr, 3.6 Myr and 3.3 Myr (De Schepper et al., 2014). The latter two events are within age model uncertainty of 3.5 Myr when C4 vegetation expands in various regions around the Indian Ocean. Thus, it is possible that the early local growth of ice, not the intensification of glaciation, altered the Siberian high or other atmospheric pressure gradients near the Indian Ocean changing cyclic precipitation patterns that lead to a regional expansion in C4 vegetation. 240 On the Indian Peninsula, C4 vegetation expands gradually from 3.5 Ma until the Mid-Pleistocene when the range of variability in C4 vegetation increases. The timing of the change coincides within uncertainty with the Mid-Pleistocene transition (~1.3 Ma) where the glacial periods transitioned from 41 kyr cycles to 100 kyr cycles. Although the trigger for the switch in cyclicity is unknown, the processes responsible for the change or the ice sheets themselves may have affected atmospheric gradients and the modes of climate variability that altered the dynamics of C4 expansion on the Indian Peninsula. 245 The sampling resolution from 1.5-0 Ma in our study is too low to resolve the change in cyclicity. However, previous studies documenting variations in C4 vegetation relative to C3 since the last glacial maximum demonstrate the sensitivity of plant communities to recent glacial-interglacial cycles (Galy et al., 2008;Phillips et al., 2014;Zorzi et al., 2015, and references therein).

Conclusion 250
Our study provides a piece of the puzzle in unraveling the complexities of C4 vegetation expansion. Although C4 vegetation was established in the CMZ on the Indian Peninsula before the end of the Miocene, the results of this study show another significant expansion in the Pliocene (~3.5 to 1.5 Myr). The latter expansion is not observed in many records from the orographically-wet Himalaya emphasizing the spatial heterogeneities in C4 vegetation response -even within the same monsoon system. However, other regions adjacent to the Indian Ocean, such as tropical East Africa, Asia, and Northwest 255 Australia, corroborate the observed expansion in the CMZ of the Indian Peninsula and show C4 vegetation patterns sensitive to the changes in hydroclimate during the Pliocene. The heterogeneous response suggests that pCO2 cannot be the exclusive driver of the expansion of the C4 vegetation in the Pliocene and changes in regional hydrodynamics likely contributed.
Restriction of Indonesian Throughflow and the onset on Northern Hemisphere glaciation may have altered the atmospheric pressure gradients and the modes of variability that determine seasonal precipitation patterns in the continents surrounding the 260 Indian Ocean, which may have caused the regional expansion of C4 vegetation in the Pliocene.

Data Availability
Data is included in the supplementary tables and will be made publically available in the NSF funded EarthChem database (https://www.earthchem.org/) maintained by the Interdisciplinary Earth Data Alliance (IEDA) at the Lamont-Doherty Earth Observatory of Columbia University. 265

Author Contribution
A.G.D. participated in the IODP Expedition 353 sampling party, performed the geochemical analyses, and lead writing and revisions of the manuscript. L.G. participated on IODP Expedition 353, was involved in the project's conceptualization, sample acquisition, and provided supervision. Y.H. also participated on IODP Expedition 353, was involved in the project's conceptualization, provided resources and funding acquisition, advised on the methodology, supervision, and 270 aided in interpretation of data.

Competing Interests
The authors declare that they have no conflict of interest.    -32 -30 -28 -26 -24 -22 -190 -170 -150 -130 Age ( Figure S1. Age-depth model for Site U1445. To determine the ages of our samples, we fit the biostratigraphic and magentostratigraphic age constraints (Clemens et al., 2016) with an age-depth model using CLAM software in R (Blaauw, 2010). We ran iterations of the model with different types of fit and levels of smoothing, and identified a locally weighted spline with 0.4 490 smoothing to best represent the trends observed in the age constraints. The differences between the age models iterations are not significant and would not change the interpretations of this study. . Carbonate and bulk organic analyses at Site U1445. Analysis of 57 samples at IODP Site U1445 for (a) bulk calcium carbonate content (weight %) calculated as (total inorganic carbon x (8.33313 CaCO3 wt/C wt)), (b) total organic carbon concentration (weight %), (c) total acidified nitrogen content (weight %), (d) Ratio of total organic carbon to total nitrogen 505 (TOC/TN, wt.%/wt.%) and (e) carbon isotopes of the total organic carbon (per mil). Black dots represent visually darker layers relative to a lighter layer (white dot) at a similar depth. TOC/TN shows a distinct increase in the mid-Pliocene, but remains within the range of TOC/TN expected for marine organic material.  Table S1. Concentrations of major, trace, and rare earth elements. Inorganic analyses of major, trace, and rare earth element concentrations for 30 bulk sediment samples at Site U1445 and additional samples from other sites in the Bay of Bengal for reference.

525
For methods see Appendix A or details in . Table S2. Carbonate and bulk organic analyses data at Site U1445. Analysis of 57 samples at Site U1445 for bulk calcium carbonate, total organic carbon, total carbon, total acidified nitrogen, carbon isotopes of the total organic carbon, and the designation of visually lighter versus darker samples at similar depths. Table S3. Carbon and hydrogen isotope analysis of leaf wax fatty acids at Site U1445. Hydrogen isotopes and carbon isotope analyses 530 of leaf wax fatty acids extracted from 57 samples at Site U1445. Measurements from fatty acid chainlengths C26, C28, and C30 are reported with their standard deviation. The correction for C3-C4 physiological differences in the hydrogen isotopes of C30 fatty acids is reported, estimating C3 vegetation as having a δ 13 C of -35.4 ‰ and C4 vegetation as -21.4‰ (Fig. S5).