How do the chemical characteristics of organic matter explain differences among its determinations in calcareous soils?

Abstract

Nowadays, the continuous organic carbon (OC) assessment in soil and its various particle-size fractions is needed to correctly estimate the soil organic matter (OM) contents and dynamics. However, the existence of several widely used analytical methods for OC and OM determination hinders the comparison of OC and OM data taken by different laboratories, in different times, soil classes, horizons and particle-size fractions. Although these methods are usually related by means of empirical soil-dependent factors, how these coefficients vary among soils is seldom addressed. In the present work 67 samples from the A horizons (0 to 20–40 cm depth) of 58 forest and agricultural calcareous soils from the Valencia province (Eastern Spain) were taken, the silt-plus-clay separated, and both the fine earth and the silt-plus-clay analysed through the wet dichromate self-heated and externally heated oxidations, the dry combustion with CO2 measurement and the loss-on-ignition. As a consequence, the readily oxidizable carbon (RXC), the total oxidizable carbon (TXC), OC and OM were obtained, respectively. Furthermore, the coefficients to convert among these properties, namely, the Walkley-Black factor (fWB), the oxidation recovery factor (fXR), the carbon valence in the organic matter (vC,OM), and the van Bemmelen (fVB) factor were assessed. The RXC in both fine earth and silt-plus-clay was 75% of the OC thus supporting a common fWB of 1.33. However, the OM in the fine earth presented less fXR than the silt-plus-clay. This apparent inconsistency between fWB and fXR was caused by the different vC,OM in the fine earth (3.45) and the silt-plus-clay (4.56) and hence, the different oxidation state of the OC in each fraction. This was revealed by how fWB depends on fXR and vC,OM through fWB = 4 fXR/vC,OM. Therefore, the different vC,OM exactly compensated for the different fXR in each fraction to give the same fWB for both. Besides, the different vC,OM in each fraction was consistent with the fact that carbon accounted for 61% and 71% of the OM mass in, respectively, the fine earth and the silt-plus-clay, thus supporting the use of a van Bemmelen factor below the standard of 1.72 and different, i.e., 1.64 and 1.41, for each particle-size fraction. This fact can be understood taking into account that vC,OM linearly depends on fVB through vC,OM = 4 – α – β fVB, where α and β are two empirical coefficients. Therefore, it has been shown how the conversion coefficients among the RXC, TXC, OC and OM depend on two independent chemical characteristics, one OM-stoichiometry-related, the carbon mass fraction, and another OM-reactivity-related, the oxidation recovery. As a consequence, it will be better understood that the determinations of OC and OM in soils, remarkably when the Walkley-Black method is used, vary not only because of changes in OC and OM magnitude, but also because of changes in OM stoichiometry and reactivity. The differences in the soil OC and OM data obtained by using different laboratory methods, and in different times, particle-size fractions, soil depths, classes, etc., will be thus better understood.