To improve catchment carbon budget models and to quantify climate change impacts we need to know where dissolved organic matter (DOM) is produced and consumed within aquifers and how it is exchanged between rivers and aquifers. Measurements of DOM in the river water, groundwater and sediments at Wellington Caves, NSW, Australia provide new insights.
Water levels in the river and aquifers were continuously measured. Water samples were collected along the river, from alluvial and karst monitoring bores, and from the caves. They were analysed for their water stable isotopes (δ18O and δ2H) and chloride chemistry, and we used UV absorbance and fluorescence spectroscopy to characterise the natural organic matter. The absorbance data were processed to provide the specific ultraviolet absorbance and spectral slopes. Parallel factor analysis was used to discriminate fluorescent components and assess their dynamics in groundwater and river water.
Groundwater levels are dynamic and respond quickly to changes in the river stage, and show that the river is predominantly losing. The chloride, δ18O and δ2H data demonstrate a direct hydraulic connection between the river and groundwater. The absorbance and fluorescence properties of DOM indicate higher molecular weight, chromophoric, and hydrophobic components in groundwater compared to river water, which has hydrophilic DOM with low molecular weight. DOM concentration, absorbance and fluorescence intensity rapidly decrease from the river water to the groundwater, suggesting the alluvial aquifer acts as a sink for the riverine DOM. We show that groundwater DOM is mostly derived from sedimentary organic matter, which has different characteristics compared to river-derived DOM. We present a conceptual DOM process model for intermittent rivers connected to karst alluvial aquifers that shows continuous processing of groundwater DOM from different sources along the flowpath. This model may be applicable to other alluvial systems and is important for research into carbon biogeochemical processing.