Relationship between loss on ignition (LOI) and total organic carbon (TOC) in sediments from two coastal settings in Japan

Loss on ignition (LOI) is widely used as a simple and cost-effective proxy for estimating organic matter content in sediments. However, the relationship between LOI and total organic carbon (TOC) varies with sediment characteristics and depositional setting, and its applicability to coastal terrestrial sediments remains insufficiently constrained. In this study, we investigate the relationship between LOI at 550 °C (LOI 550 ) and TOC using muddy and sandy sediments collected from coastal lowland and coastal forest environments in Japan. A total of 147 samples, including muddy sediments formed under normal depositional conditions and sandy event deposits, were analyzed. Strong linear relationships were observed between LOI 550 and TOC for both sediment types. The regression equations were TOC = 0.61 × LOI 550 − 0.78 ( R 2 = 0.96) for mud and TOC = 0.86 × LOI 550 − 0.76 ( R 2 = 0.80) for sand. ANCOVA showed no significant difference in either slope or intercept between the Sendai coastal lowland and Hasunuma coastal forest for mud samples, whereas the regression slopes differed significantly between the two settings for sand samples ( p = 0.003). Comparison of linear, quadratic, and piecewise linear models provided no strong evidence that the observed relationships were explained by systematic nonlinearity across the LOI 550 range. Comparison with previous studies indicates broad similarities in regression trends, although differences in intercepts, sediment characteristics, and depositional environments limit direct comparison and preclude interpretation as independent validation. Increased variability was also observed at low LOI 550 values, where non-organic contributions to mass loss become relatively important. These results provide a regional calibration of the LOI 550 –TOC relationship for Japanese coastal terrestrial sediments and demonstrate its potential as a practical tool for rapid TOC estimation in coastal event-deposit studies. However, the significant site-dependent difference observed for sandy sediments and the limited LOI 550 range of the sand dataset highlight the need for caution when applying the regression equations beyond the sediment types, LOI 550 ranges, and depositional settings examined here. Broader validation across additional coastal environments is required before general applicability can be established.