Related Experiment Video
Updated: Sep 17, 2026

Determination of Total Lipid and Lipid Classes in Marine Samples
Published on: December 11, 2021
Region-specific LOI-TOC conversion models for marine sediment blue carbon assessment in the Western Irish Sea
Anthony Grey1, Ciara Ní Mhaoldomhnaigh2, Mark Chatting3
1School of Chemical Science, Dublin City University, Dublin, Ireland. greyanthony@gmail.com.
Abstract:
Continental-shelf sediments (< 200 m water depth; ~ 9% of the global seafloor) store large amounts of organic carbon (OC) and accumulate OC annually, but management and accounting applications require regional evidence and transparent uncertainty. Loss-on-ignition (LOI) is a rapid, low-cost proxy for organic matter and OC, but LOI-to- total organic carbon (TOC) conversions vary with sediment type and analytical protocol. This study used multiagency Western Irish Sea data to test whether region- and sediment-specific LOI-to-TOC regressions improve OC estimation relative to a pooled conversion. Paired LOI and TOC measurements were strongly correlated (R2 = 0.948). Structural-break and analysis of covariance (ANCOVA) tests identified changes in the LOI-TOC relationship associated with mud content, iron (Fe) concentration and water depth. Splitting samples into high-mud and low-mud groups produced group-specific models (R2 = 0.97 and 0.72, respectively) and improved fit for high-mud sediments, while the low-mud model retained greater uncertainty. The results support the use of separate sediment-class-specific LOI-to-TOC calibration models for Western Irish Sea sediments and comparable settings, provided that paired LOI-TOC calibration, the range of sediments to which each model applies, and propagated uncertainty are reported. By integrating archived samples, routine geotechnical measurements, grain-size data, LOI results and targeted TOC analyses from multiple survey and research sources, the study also demonstrates how existing marine monitoring and seabed-mapping activities can be leveraged to strengthen national sediment carbon datasets.
