LAUSR

Abstract Long-term paleovegetational records from continental settings help in comprehending regional and global forcing on the abundance of C3-C4 plants in the past. The carbon isotopic composition of soil carbonates (δ13CSC), soil organic matter (δ13CSOM), organic matter occluded in soil carbonate nodules (δ13CNOM) and biomarkers in paleosol organic matter (long-chain fatty acid; δ13CFAME) are often used to estimate changes in the past-vegetational composition. However, it has been observed that depending on the type of proxy, the estimated abundance of C3-C4 plants varies, which can lead to uncertainty in paleovegetational records. Hence, the present study aims to comprehend the factors affecting the δ13CSC, δ13CSOM, δ13CNOM and δ13CFAME values within a paleosol. In this context, available δ13CSC, δ13CSOM, δ13CFAME and newly measured δ13CNOM values from the late Quaternary sequences of the Ganga Plain, India has been used. The abundance of C4 plants calculated from the δ13CSC and δ13CFAME values is ~2% to 89% higher compared to the δ13CSOM and δ13CNOM values-based estimates. Even with a common source of organic matter, the δ13CSOM values indicate a higher abundance of C4 plants (~2% to 50%) compared to the estimates from δ13CNOM values. We suggest that the disparity is due to the variation in the response of proxies to perturbations in the paleovegetational regime, growing season condition, and isotopic fractionation during decomposition and incorporation of organic matter into the soil. For example, the organic matter is incorporated into the soil throughout the year and represents average annual biomass, whereas SC precipitates under warmer and often drier condition when the ratio of C4 to C3 plant respiration is higher. Additionally, preferential degradation of 13C enriched labile compounds and C4 plants derived organic matter may lower the δ13CNOM values resulting in an under estimation of C4 plants in a mixed C3-C4 environment. The higher abundance of C4 plants estimated from the δ13CFAME values is due to the isotopic fractionation (13C enrichment of ~2‰ to 7‰) during incorporation of plant-derived long-chain fatty acids into the soil. The disparity in the abundance of C4 plants estimated from δ13CSOM and δ13CNOM values is due to the difference in the preservation of SOM and NOM. Contrary to the NOM (which is in a closed system), SOM in open system within the soil matrix is subjected to 13C enrichment due to the long-term humification of organic matter. Various factors such as grain size and pedogenesis that are inherent to the depositional environment also control the δ13C values of paleosol components. Considering the uncertainties associated with the δ13C values of paleosol components, reporting the absolute abundance of C4 plants would be uncertain. Therefore, we recommend presenting the relative change in abundance of C3-C4 plants during paleovegetational reconstruction.