Dissolved organic matter: artefacts, definitions, and functions

In this concept paper, some basic problems in the obtaining of the dissolved organic matter (DOM) are described. Some of them are fairly obvious such as adsorption and desorption on the filters used to isolate DOM. However, some are subtler such as alteration in the tertiary physicochemical structure of DOM because of cavitation (filtration), concentration (ultrafiltration, reverse osmosis), or decomposition (electrophoresis). These structure changes may not have an influence on a variety of analytical techniques, but they may alter the ability of DOM to interact with certain ecosystem components such as the biota and pollutants. Therefore, the interpretation of the analyses done with such DOM cannot necessarily be extrapolated to in situ conditions. Fluorescence has been found to be a sensitive tool to follow such tertiary structure changes. In structured matrices with pore spaces such as soils and sediments, DOM may have different fates and functions depending on the pore spaces in which it is located. This again needs to be considered in sampling strategies. For example, DOM in macropores may play important roles in groundwater quality but presumably does not play a major role in the in situ production of the greenhouse gas N2O via denitrification, since this pore space tends to be well aerated. A protocol for the separate sampling of the different pore spaces is presented. Another difficulty, which is shown here, is that minor changes in extraction methodology (such as the extractant-soil ratio) can have a serious effect on the apparent amount of DOM. DOM samples from a field can have considerable temporal and spatial variability, which is complicated by the fact that DOM from unsaturated systems can be strongly influenced by ambient water content. This again needs to be considered, especially when fields from different regions are sampled and compared. Another important point is that the results obtained from one DOM source cannot necessarily be extrapolated to the DOM from another source. For example, soil pore water DOM differs both from aquatic DOM and from the material, which can be extracted as the humic substance fraction. Finally, with the increasing awareness of the ecological functions of DOM, which are briefly presented here, there is also an increase in DOM research being done in different disciplines. This is laudable, but presentations of results need to define DOM more precisely to prevent misunderstandings. (C) 2002 Elsevier Science B.V All rights reserved.