On the use of loss‐on‐ignition techniques to quantify fluvial particulate organic carbon

Abstract

The fluvial flux of carbon (C) from terrestrial to marine environments represents an important component of the global C-cycle, which can transfer C from the atmosphere to sedimentary storage. Fluvial fluxes of C are also an essential resource for freshwater ecosystems, critical for habitat heterogeneity and function. As such it is crucial that we are able to quantify this flux accurately. However, at present there are a number of different techniques used to quantify concentrations of fluvial C, and these techniques vary in their accuracy. In this article, we compare particulate organic carbon (POC) measurements derived from two commonly-used techniques; a simple combustion and loss-on-ignition (LOI) technique, and an oxidative–combustion and carbon dioxide (CO2) detection technique. The techniques were applied to water samples collected from 10 contrasting reference-condition, temperate river ecosystems. The POC measurements derived from the LOI technique were up to 16 times higher (average four times higher), than those derived from the oxidative–combustion and CO2 detection technique. This difference was highly variable both across the different river ecosystems and within each river ecosystem over time, suggesting that there is no simple way of converting the mass measured by LOI to estimates of fluvial POC. It is suggested that the difference in POC measured by these two techniques is a consequence of: (1) the loss of inorganic carbon at LOI combustion temperatures of > 425 °C, (2) the potential during the LOI combustion stage to lose hygroscopic and intercrystalline water, not completely driven off by the drying stage at temperatures of < 150 °C, and (3) the variable C content of fluvial organic matter, meaning that the simple application of a fixed correction factor to values obtained from the LOI technique may not be appropriate. These findings suggest that oxidative–combustion and CO2detection techniques are preferential for quantifying fluvial POC

Export of fine particulate organic carbon from redwood‐dominated catchments

“…Recently, Grove and Bilotta () questioned the loss‐on‐ignition method to determine POC in United Kingdom catchments because of the possible loss of CO 2 from certain minerals or of water from clay minerals. Those authors were most concerned with samples collected in catchments with limestone‐dominated geology, which is not the case in the present study.…”

“…This figure is available in colour online at wileyonlinelibrary.com/journal/espl 1535 EXPORT OF FINE PARTICULATE ORGANIC CARBON discharge or turbidity were not complete for each year, and annual organic and inorganic sediment flux could only be calculated for three years at PRW and LLM and for two years at PRU. Finally, a conversion factor of 0 · 5 was applied to convert organic matter to carbon content, a typical value used in previous studies using loss-on-ignition (Sutherland, 1998;Grove and Bilotta, 2014), although these authors report a range of 0 · 33 to 0 · 58 as conversion factors used in various other studies of POC.…”

A Bayesian network model for estimating stoichiometric ratios of lake seston components

“…Estimates of carbon content introduce another source of uncertainty to the current estimates of seston stoichiometry. We estimated carbon content as a fixed fraction of volatile suspended sediment, but this fraction can vary with conditions (Grove and Bilotta 2014). Others have employed different approaches for measuring organic carbon, including the use of a relationship between phytoplankton cell volume and carbon content (Hessen et al 2003), and direct measurements of elemental carbon (Urbansky 2001).…”

Predicting suspended sediment concentration from nephelometric turbidity in organic‐rich waters

“…These filters were oven‐dried at 105°C for 24 hr and weighed and redried three times. The glass fibre filters were dried in a muffle furnace at 500°C for 30 min and reweighed to determine the loss on ignition, which is an index of POM (see Grove and Bilotta, ). Samples below the method detection limit of 0.3 mg/L were excluded from the data set.…”