CHARACTERIZATION OF NATURAL ORGANIC MATTER WITHIN CLASTIC CAVE SEDIMENTS USING PARAFAC MODELLING

Natural sediments contain a heterogeneous mixture organic compounds at varying concentrations. Natural organic matter (NOM) plays an essential role in most biogeochemical cycles and contaminant fate and transport. However, dynamics of transport, storage, and transformation of NOM remains unclear. In this project we characterized NOM extracted from clastic sediment cores from a cave in the northern karst region of Puerto Rico. The clastic sediment samples (n=29) were treated with an alkaline solution and emission-excitation matrices were collected on the extracted organic matter using fluorescence spectroscopy. Using the EEM data, a parallel factor (PARAFAC) analysis was performed. This analysis decomposes a set of EEMs into modelled components that describe the entire dataset by mathematically separating spectra of overlapping fluorescent peaks. A PARAFAC model with three components was chosen.

The EEM data collected from the extracted organic matter had peaks consistent with humic-like substances. Fluorescence index (FI) and biological index (BIX) are used to differentiate between terrestrial-and microbial-derived organic carbon. FI values of 1.4 or less and BIX values of 0.8 or less indicate terrestrial NOM. Sample FI and BIX values ranged from 0.89 to 1.54 and 0.68 to 1.87, respectively, indicating that both microbially and terrestrially derived NOM was present. Samples with the lowest BIX and FI values were found to contain the highest NOM content. All three modelled PARAFAC components plotted within the humic-like region of the EEM space. The proportion of each component was normalized to the total contribution of each sample. It was found that the proportion of the modelled components differed in samples collected from above and below the water level in the cave.

The data collected through fluorescence spectroscopy reveal that the type and structure of the NOM is variable and could be linked to the surrounding environmental conditions (i.e. sediment saturation). The data presented suggest that PARAFAC modelling applied to EEM data collected from cave sediment is a valuable tool to better understand NOM dynamics occurring within the subsurface.