Paleoclimate - Marine Sediments
Stable Isotope
| d18O | δ13C | |
|---|---|---|
| Definition | δ18O is the per-mil deviation in the ratio of the 18O to 16O in a sample of calcium-carbonate relative to an international standard (VPDB). | δ13C is the per-mil deviation in the ratio of the 13C to 12C in a sample of calcium-carbonate relative to an international standard (VPDB). |
| Basic controls | …the controlled by global hydrological cycle, primarily driven by variation in ice volumes. ① H216O has higher vapour pressure than H218O, evaporates more readily and condenses less readily. ② During glacial periods, the buildup of ice sheet requires large volumes of water to be evaporated from the ocean and transported to high altitude, where it is precipitated as snow. During this process, the water becomes increasingly depleted in 18O (Rayleigh distillation). ③ Therefore, the ice sheets lock away more 16O, while the remaining ocean water becomes relatively enriched in 18O (δ18O 1.1‰ higher in LGM than today, Adkins et al. 2002). ④ During interglacial periods, the melting of glacier ice returns 16O-rich freshwater to oceans, lower the δ18O of sea water. ⑤ A rough estimate for the change in the average δ18O of the global ocean due to glaciation is 0.09‰/-10 m of sea level (Hillaire-Marce et al., 2007). | …controlled by biological productivity and oceanic circulation. Biological productivity is the primary driver of δ13C fractionation. 1. Atmospheric CO2 exchanges with surface ocean waters, establishing an equilibrium δ13C signature within the surface dissolved inorganic carbon (DIC) pool. 2. ① In the euphotic zone, phytoplankton takes up more 12C for photosynthesis, since it is lighter and easier to diffuse and respond. ② This isotopic fractionation leaves the residual surface DIC relatively enriched in 13C. ③ Planktonic foraminifera, living in these surface waters, calcify their calcium carbonate shells using this ambient DIC, thus recording the 13C-enriched isotopic signal. 3. ① The 12C-enriched organic matter produced by primary productivity sinks into the deep ocean as part of the biological pump. ② As it descends, microbial remineralization decomposes this organic matter, releasing 12C-rich CO2 back into the DIC pool at depth, progressively lowering the δ13C of deep water DIC. ③ Benthic foraminifera living near the seafloor precipitate their shells from this DIC reservoir, incorporating the isotopic signature of the local deep water. |
| Other factors | ① Also controlled by the balance between evaporation and freshwater input. This is reflected in the relationship between contemporary sea water salinity: roughly 0.5‰/1.0 psu (Hillaire-Marce et al., 2007). Saltier regions with intense evaporation (e.g., subtropical gyres) therefore more enriched in δ18O, whereas large river plumes are fresher and deplete δ18O. ② However, this δ18O-salinity relationship also sees exceptions, especially in Antarctica where sea-ice formation significantly raises salinity via brine rejection without proportionally increasing δ18O. The deep Southern Ocean waters end up being relatively salty but with low δ18O (~34.7 psu and ~−0.3‰). This forms clear contrast to the deep water formed in the North Atlantic (eg NADW), where the high salinity is resulted not from brine rejection but evaporation-enhanced cooling, resulting in both high salinity and high δ18O (~35.0 psu and ~+0.3‰). | Ocean circulation modulates the spatial and temporal distribution of these δ13C signals. Spatial ①: Holocene NADW featured by 1.4‰ δ13C, in comparison to AABW 0.4‰ δ13C (Howe et al, 2016). This is because NADW originates from the cooling and sinking of warm, saline surface waters in the North Atlantic, which is relatively young and has been recently in contact with the atmosphere - better “ventilation” (exchange between surface water and deep waters). The low residence time also means that NADW has little time to accumulate 12C-rich respired carbon from organic matter sink and remineralization. Temporal ②: Oppo et al. (2015) revealed that during Heinrich Event, δ13C value decreases in most sediment cores from the upper Atlantic and some from the deeper N Atlantic, likely attributed to (1) a reduced fraction of high-δ13C North Atlantic Deep Water (NADW) in comparison to AABW and/or (2) to a decrease in the NADW δ13C source value. |
| From sw to shell | The shell δ18O reflect sea water δ18O, but there would also be slight deviation, controlled by factors such as the temperature during calcification. (When the shell calcite is in oxygen isotopic equilibrium with seawater), and the isotopic separation factor between calcite and seawater (ε~δ18Oshell-δ18Osw) is inversely related to calcification temperature, with the δ18O of calcite decreases by ~0.21–0.23‰ for 1°C increase in temperature for a given δ18OH2O. The carbonate ion effect (see right) results in 1.4 * 10-3 ‰ to -4.5 * 10-3 ‰ per μmol kg-1 of [CO32−] for planktic foraminifera (Spero et al., 1999). Slightly weaker than d13C. | There is a deviation of the d13C value in foraminifera shells from the surrounding environment, referred to as vital effects. One key factor determining the fractionation of d13C during calcite formation is the concentration of carbonate ion (Carbonate Ion Excursion, CIE): at higher [CO32−], calcification proceeds faster, leading to kinetic fractionation which favors the lighter isotope (12C), resulting in lower δ13C in the shell. Early lab experiments (eg Spero et al., 1999) found the effect to be species-specific, ranging from -4.7 * 10-3 ‰ to -13.0 * 10-3 ‰ per μmol kg-1 of [CO32−] for planktic foraminifera. However, recent study by Köhler and Mulitza (2024) using stack compilations and BICYCLE-SE simulation concluded that this effect is negligible for interpreting glacial–interglacial δ13C changes. |
| Case study | Hodell et al. (2023) used planktic and benthic δ18O records from IODP Site U1385 (Iberian Margin) to reconstruct millennial-scale climate variability (MCV) over the past 1.5 million years, demonstrating that: 1. MCV was most active during intermediate glacial conditions (benthic δ18O ≈ 3.8–4.8‰), while suppressed during full interglacials (δ18O < 3.8‰) and some peak glacials (δ18O > 4.8‰), reflecting a threshold response. 2. Prior to ~0.9 Ma, the amplitude of MCV was strongly modulated by obliquity (41 kyr), whereas after the Middle Pleistocene Transition (MPT), MCV became increasingly influenced by precession and eccentricity-paced variability, consistent with the emergence of 100 kyr cycles. | Blaser et al. (2025) found that the d13C data from cores from subpolar N Atlantic cluster between 0.7 and 1.0‰ during Heinrich Stadial 1, indicating that the water filling the basin was still young and well ventilated, rather than being replaced by old south water. |
Paleo-thermometry
| Mg/Ca | UK37 | |
|---|---|---|
| Definition | The Mg/Ca thermometer is based on the principle that certain calcareous marine micro-organisms, especially foraminifera (unicellular zooplankton that secrete CaCO3 shells), incorporate magnesium (Mg2+) and calcium (Ca2+) into their shells in ratios that depend on the temperature of the seawater in which they grow. | The UK′37 thermometer is based on the unsaturation ratio of long-chain alkenone lipid biosynthesized by haptophyte algae. Unsaturated lipids (lipids with double bonds) serve to enhance fluidity in cell membranes by preventing tight molecular packing. |
| Principle – qualitative description | The substitution of Mg2+ for Ca2+ in the calcite lattice is endothermic, i.e. it requires energy (Barker et al., 2005). Therefore, as temperature increases, so does the favorability of this substitution, resulting in higher Mg/Ca ratio. | At low temperatures, algae increase the production of more unsaturated alkenones (C37:3, which has 3 double bonds) to maintain membrane flexibility, while at high temperatures when molecular motion already promotes membrane fluidity, algae produce less unsaturated forms (C37:2, two double bonds) to prevent excessive fluidity. Thus, the UK′37 index (the ratio of C37:2 to C37:3) increase with temperature. |
| Principle – quantitative relationship | The relationship is exponential (6-10% increase in Mg/Ca for every 1°C warming, Rosenthal et al., 2022), approximated by: Mg/Ca = B * exp(A*T), Where T is temperature (°C), and A and B are empirically-derived constants based on core-top or culturing studies. The precise response varies by species and habitat depth. | According to empirical studies, the relationship is linear, with a form of: SST = (UK′37 − 0.044)/0.033 In other words, as UK′37 increases by about 0.033, the temperature increases by 1°C. |
| Other controls / limitation | The first factor is pH - lower pH shifts the carbonate system equilibrium toward more CO2 and less CO32−, which affects calcite saturation and potentially inhibits the substitution of Mg2+ for Ca2+ in the crystal lattice, artificially lower the Mg/Ca ratio (~4% per psu, lab culture experiment by Honisch et al., 2013). Secondly, there are possible post-depositional diagenesis, particularly the dissolution of calcite shells in low calcite saturation environment, often the deep water, which reduces Mg/Ca ratios. (Regenberg et al., 2014) | Notably, this empirical calibration works best for SSTs between roughly 8°C and 29°C. At very high temperatures (> 29°C), the UK′37 index tends to saturate; at very low temperature, alkenone formation is suppressed. |
| Advantages | When combined with δ18O of the same shell, it enables separation of temperature and ice volume/salinity effects, a major advantage over δ18O alone. | Purely organic and resistant to dissolution |
Paleo-circulation
| Sortable silt | 231Pa/230Th | |
|---|---|---|
| Principle | • Mean size of the 10–63 μm terrigenous silt in current-sorted marine sediments is good proxy for near-bottom current strength (Tegzes, Jansen and Telfod, 2015). • It controlled by selective deposition: strong currents winnow away finer grains and deposit coarser silt, resulting in a larger mean size; weaker currents allow finer particles to settle, leading to a smaller mean size. • Grains > 63 μm are rarely moved by ocean currents, and those < 10 μm behave cohesively, acting as aggregates rather than individual grains (McCave and Hall, 2006; McCave et al., 1995). Thereby, silt with size of 10–63 μm are used as proxies for near-bottom flow speed, with a relationship of 1.36 ± 0.19 cm s−1/μm as calibrated by McCave (2017). • Notably, this proxy is most effective when sediment supply is relatively constant and inputs from biological or ice-rafting are minimal. | • 231Pa (half-life 32.5 kyr) and 230Th (half-life 75.2 kyr) are naturally occurring in seawater and produced through the α-decay of soluble U at a near-constant activity ratio of ~0.093. • Both radio-nuclides are particle reactive, i.e. they are scavenged from the water column by attaching to settling particles, but at different rates. • 230Th is highly particle reactive, therefore gets removed from water column rapidly and deposits near its production site. On the other hand, 231Pa is less particle reactive, thus remaining longer in the water and can be advected by deep ocean circulation (e.g. NADW) before settling. • Thus, low 231Pa/230Th ratio indicates more Pa is exported by water mass rather than scavenge locally, and therefore representing stronger ocean current. • This method is most applicable in N Atlantic where the current moves fast (resulting in a Holocene value of 0.055, McManus, 2004) and therefore the 231Pa/230Th ratio is sensitive to the shift in current strength. |
| Case study | Nagai et al. (2025) reconstructed a 45-kyr Intermediate Western Boundary Current (IWBC) speed from SS records from Brazil’s Santos Basin. Converting mean size to flow velocity shows the IWBC peaked near 22 cm s−1 before 40 ka (~MIS 3), weakened into the LGM, but sped up during H 1, 3 and 4 and YD, implying IWBC intensification when the northern AMOC slowed, likely associated with a compensating export of Antarctic Intermediate Water toward the tropics. | McManus et al (2004) used the sediment core from Bermuda Rise to show that during the LGM overturning rate is reduced by 30-40% compared to the Holocene (0.068 for the former and 0.055 for the later). At the onset of Heinrich Stadial 1a (17.5 ka), the ratio climbed steeply towards the production value (0.093) and stayed elevated for ~2000 years, indicating the collapse of the AMOC. The AMOC recovered with the ratio falling back to 0.065 during the Bolling-Allerod warming, and then spiked at 0.08 during the Younger Dryas (12.7 ka). |
| Limitations | Only reflects near-bottom flow – does not track full water-column overturning | Only reflect strength not sign |
Enjoy Reading This Article?
Here are some more articles you might like to read next: