A nutrient effect on the TEX₈₆ paleotemperature proxy

Published

R. Rattanasriampaipong, J. E. Tierney, J. T. Abell, L. D. Gilmore

Geophysical Research Letters 52 · 2025

In brief

Lab cultures show that when marine archaea are short of nutrients they build more rings into their membrane lipids, which raises TEX₈₆ as if the water were warmer. This study shows that the same “nutrient effect” is recorded in seafloor sediments, and that it can be separated from the temperature signal.

Key findings

A global signal. In an expanded coretop database, TEX₈₆ residuals (what is left after removing the thermal effect) correlate negatively with nitrate: ρ = −0.31 (P < 0.001). Low-nutrient sediments read too warm.
Stronger where nitrate gradients are steep. In regions with sharp nutrient gradients the correlation reaches ρ = −0.73 to −0.91.
A simpler explanation for the Red Sea. Its anomalously high TEX₈₆ fits extreme nutrient stress in an oligotrophic sea, without needing endemic archaeal clades.
Visible in the past. In the Arabian Sea, TEX₈₆ warms by up to 7 °C during the Younger Dryas and Heinrich Stadial 1, when alkenone (U^K′₃₇) records cool and δ¹⁵N points to weaker upwelling and fewer nutrients. In the Tasman Sea, interglacial TEX₈₆ temperatures overshoot U^K′₃₇ as the nutrient-poor subtropical gyre expands.

Key figure

Panel a: coretop TEX86 against thermocline-integrated temperature. Panel b: TEX86 residuals against nitrate on a log scale. Panel c: world map of coretop TEX86 residuals over modern nitrate. — **Figure 1.** Positive TEX₈₆ residuals in coretop sediments match low-nutrient regions. (a) TEX₈₆ against thermocline-integrated temperature. (b) Residuals from that fit fall as nitrate rises, most clearly below 2.7 µmol/L (blue). (c) Map of residuals over modern nitrate. Rattanasriampaipong et al. (2025), *GRL*.

From finding to predictor → TEXAS

Nitrate became one of the two nonthermal predictors in TEXAS, our proxy system model for TEX₈₆ paleothermometry. Below 1 µmol/L it shifts the TEXAS S-curve to account for nutrient stress, and the Tasman Sea records from this paper return as a TEXAS case study.

Cite

(Rattanasriampaipong et al., 2025). Data: 10.5281/zenodo.14806961.

The TEX₈₆ paleotemperature proxy is widely used to reconstruct ocean surface temperatures over the past 100 million years. However, archaeal culture experiments show that nutrient stress elevates TEX₈₆ values by increasing glycerol dialkyl glycerol tetraether (GDGT) cyclization. Here, we demonstrate that this “nutrient effect" is also recorded in sedimentary GDGTs. Using an expanded core-top database, we find a significant negative correlation between TEX₈₆ and nitrate concentrations (ρ= -0.31; P<.001) once the thermal effect is removed. There are stronger correlations (ρ-0.73 to -0.91; P<.001) in regions with steep nitrate gradients. Comparisons between TEX₈₆ and U^K’₃₇-based reconstructions from the Arabian Sea and Tasman Sea suggest that nutrient stress influenced GDGT distributions during glacial-interglacial cycles. Our findings underscore the need to account for nutrient effects when applying TEX₈₆ paleothermometry.

In brief

Key findings

Key figure

From finding to predictor → TEXAS

Cite

References

2025