Tiny Ocean Algae Reveal How Monsoon Rains Weakened the Indian Ocean’s Carbon Pump During the Last Ice Age

Deep in the sediments of the Bay of Bengal, a team of micropaleontologists has recovered a 53,000-year-old archive written in microscopic calcium carbonate plates, and it tells a surprising story about one of the planet’s most important carbon conveyor belts. The marine biological pump, the suite of processes that carries organic carbon from the sunlit surface ocean down to the deep sea, is a cornerstone of Earth’s carbon cycle and a natural regulator of atmospheric carbon dioxide. Yet how this pump behaved in the northeastern Indian Ocean across the last glacial cycle has remained stubbornly unclear, largely because the region sits under one of the most complex hydrological regimes on Earth. A new high-resolution study published in the Journal of Micropalaeontology by Nilufar Yasmin Liza of the South China Sea Institute of Oceanology and colleagues now fills that gap, showing that the biological pump in the Bay of Bengal was at its weakest during the depths of the last ice age and strengthened dramatically as the world warmed into the Holocene.

The research hinges on coccolithophores, single-celled calcifying phytoplankton that build minute plates called coccoliths around their cells. These organisms occupy a privileged position in marine carbon cycling because they participate in both the organic pump, fixing carbon dioxide through photosynthesis, and the carbonate pump, producing calcium carbonate that sinks toward the seafloor. When coccolithophores die, their plates rain down onto the ocean floor, where they are exceptionally well preserved. Because different species thrive at different depths and nutrient levels, the composition of fossil assemblages encodes a detailed record of past upper-ocean conditions. In tropical waters, the deep-dwelling species Florisphaera profunda is particularly valuable: it lives below the euphotic zone near the nutricline, the depth at which nutrient concentrations rise sharply, and its relative abundance in sediments tracks how deep that nutrient reservoir sits.

To exploit this proxy, the team analyzed piston core YDY05, a 2.39-meter sediment core retrieved from the southern Bay of Bengal at a water depth of 3,300 meters during a research cruise of the R/V Shiyan 3. The core was subsampled at 2-centimeter intervals, yielding 120 samples that were processed with a standard drop technique and examined under a polarizing-light microscope at 1,000-fold magnification. At least 300 coccoliths were counted per sample, and the chronological framework was anchored by six accelerator mass spectrometry radiocarbon dates on mixed planktonic foraminifera, calibrated with the Marine20 dataset and refined through Bayesian age-depth modeling. The resulting record extends back roughly 52,680 years, spanning Marine Isotope Stages 3, 2, and 1, with sedimentation rates ranging from about 3.7 to 8.6 centimeters per thousand years.

The assemblages turned out to be abundant and beautifully preserved throughout, dominated by three ecologically telling species: Florisphaera profunda, Gephyrocapsa oceanica, and Emiliania huxleyi, which together account for 85 to 97 percent of the community. From the relative abundance of F. profunda, the researchers computed primary productivity using an updated calibration for the tropical Indian Ocean derived from a global core-top dataset and satellite-based productivity estimates, a relationship with a correlation coefficient squared of 0.74 and a residual standard error of 1.5 grams of carbon per square meter per year. The reconstructed productivity ranged from roughly 108 to 412 grams of carbon per square meter per year, and the mean value of about 192 grams closely matches modern satellite estimates for the southern Bay of Bengal, a reassuring sign that the proxy is capturing real oceanographic signal rather than artifact.

The temporal pattern is striking. Between about 53,000 and 27,000 years ago, productivity averaged around 166 grams of carbon per square meter per year, then dipped to a minimum of roughly 108 grams during the Last Glacial Maximum. As the deglaciation unfolded, productivity climbed steadily, surging from about 146 grams at 18,000 years ago to a remarkable peak of 412 grams at around 11,000 years ago in the Early Holocene, before easing back to about 239 grams during the mid-Holocene. Species-level trends mirror this arc: F. profunda dominated before the glacial maximum, at one point spiking to about 80 percent of the assemblage between 27,000 and 25,000 years ago, while the upper-euphotic zone taxa E. huxleyi and G. oceanica surged during the deglaciation, reaching pronounced maxima near 11,000 years ago. Total coccolith fluxes followed suit, bottoming out before 27,000 years ago and peaking between 18,000 and 11,000 years ago.

Interpreting these numbers requires understanding why the Bay of Bengal behaves so differently from its neighbor across the Indian subcontinent, the Arabian Sea. Both basins feel the full force of the Indian Summer Monsoon, yet the Bay of Bengal is consistently less productive. The reason lies in freshwater. Summer monsoon precipitation there can reach about 20 millimeters per day, compounded by enormous discharge from the Ganges, Brahmaputra, Meghna, and Irrawaddy river systems. This deluge lowers surface salinity, creating a low-density lid over saltier subsurface water. The resulting halocline and barrier layer suppress vertical mixing, so nutrient-rich waters below the nutricline never reach the sunlit euphotic zone. During the spring intermonsoon, modern measurements show primary productivity between roughly 13 and 174 millimoles of carbon per square meter per day, but particulate organic carbon export stays low, between zero and 7.7 millimoles, a signature of chronic nutrient limitation imposed by stratification.

The glacial story that emerges from core YDY05 is one in which this freshwater control was amplified. During the Last Glacial Maximum, global sea level stood about 120 meters lower than today, restricting the Indonesian Throughflow and reorganizing Indo-Pacific circulation. But in the southern Bay of Bengal, the decisive factor was the regional freshwater balance: reduced precipitation, diminished river discharge, and disrupted shelf-basin connectivity kept the upper ocean strongly stratified, deepened the nutricline, and starved surface waters of nutrients. The persistently high abundance of the deep-dwelling F. profunda during glacial intervals, including its elevated absolute abundance during the glacial maximum itself, therefore reflects enhanced upper-ocean stability rather than enhanced productivity, a subtle distinction that only an assemblage-based proxy can resolve. As the planet deglaciated, rising seas restored shelf connectivity, and processes such as cyclonic eddies, island-induced upwelling near the Maldives, and equatorial circulation associated with the Wyrtki Jet periodically shoaled the thermocline, delivering nutrients upward and driving the Early Holocene productivity peak.

Perhaps the most consequential finding comes from the basin-wide comparison. The team compiled productivity proxy records from six sediment cores spanning the Indian Ocean, including coccolith-derived productivity, biogenic barium, total organic carbon accumulation, and coccolith strontium-to-calcium ratios, interpolated them to a common 1,700-year grid, and ran a principal component analysis. The first two components explain 80 percent of the variance. The dominant mode, PC1, tracks the global benthic oxygen isotope stack almost inversely, confirming that glacial-interglacial climate change is the primary driver of basin-scale productivity. Yet the spatial pattern of loadings reveals something deeper: northeastern Indian Ocean sites load positively while western and eastern tropical Indian Ocean sites load negatively. In the upwelling-dominated western and eastern basins, wind-driven systems produced the opposite productivity history, with glacial intervals often more productive than interglacial ones. The same planetary-scale climate forcing thus produced regionally opposite biological responses, depending on whether wind or freshwater governed nutrient supply.

The secondary mode, PC2, aligns with the Indian Monsoon Stack and separates the monsoon’s two faces: its atmospheric expression, seen in wind-driven upwelling in the Arabian Sea, and its hydrological expression, seen in precipitation, runoff, and salinity proxies from the Bay of Bengal. Site YDY05 loads only weakly on this wind-driven component and clusters instead with the hydrological records, demonstrating that in the southern Bay of Bengal, monsoon rains and river discharge, not monsoon winds, held the upper hand. Even during the intensified summer monsoons of the mid-Holocene, increased freshwater input reinforced stratification and offset the benefits of stronger wind-driven nutrient supply, which is why productivity declined after the Early Holocene peak. For scientists projecting how the ocean’s carbon pump will respond to ongoing warming and changing monsoon rainfall, the message from this 53,000-year record is sobering and clear: in stratified, river-influenced basins, the fate of carbon sequestration may rest less on how hard the winds blow and more on how much fresh water falls from the sky.

Subject of Research: Reconstruction of biological pump and primary productivity variability in the northeastern Indian Ocean since the last glaciation using coccolithophore assemblages

Article Title: Biological pump variations in the northeastern Indian Ocean since the last glaciation: evidence from coccolithophores

Article References: Biological pump variations in the northeastern Indian Ocean since the last glaciation: evidence from coccolithophores. (n.d.). https://doi.org/10.5194/jm-45-699-2026

Image Credits: AI Generated

DOI: 10.5194/jm-45-699-2026

Keywords: biological pump, coccolithophores, Bay of Bengal, Indian Ocean, primary productivity, monsoon, last glacial maximum, paleoceanography, Florisphaera profunda, carbon cycle, ocean stratification, sediment core