How to take volume-based peat samples down to mineral soil?

Abstract

Determination of soil organic carbon stocks in peat soils is of major importance for prioritization and evaluation of mitigation measures. This requires the accurate assessment of bulk density, which is commonly undertaken by measuring the oven-dry weight of a volume-based sample. Sample rings (steel cylinders) are widely used and have become the method of choice. If sample rings cannot be taken (e.g. in deep peat layer), sampling needs to be performed with other sampling devices. The aim of this study is to evaluate the accuracy and precision of different sampling devices in determining bulk density and total peat masses of the entire peat profiles down to mineral soil. Four sampling devices (driving hammer device with a sheath probe, gouge auger, Russian corer and Wardenaar corer) were compared with sample rings at one bog peat site and one fen peat site. The gouge auger and Russian corer – the only sampling devices in this study applicable for depths below approximately −1m – were also compared with one another.The results varied depending on peat type and horizon characteristics. Sample rings and the driving hammer were the only sampling devices that could be used to sample the amorphous or aggregated peat of the upper, unsaturated part of the profiles. However, samples taken with the driving hammer significantly underestimated bulk densities and thus caused a high systematic error of −0.068 g cm−3. In the sampling depths with slightly to moderately decomposed peat, bulk density values determined with the driving hammer, gouge auger and Wardenaar corer were not significantly different from the data acquired using sample rings. At these depths, all the sampling devices had low systematic errors, with −0.002 g cm−3 for the gouge auger, 0.005 g cm−3 for the driving hammer, −0.006 g cm−3 for the Wardenaar corer and 0.012 g cm−3 for the Russian corer. The Russian corer caused an overestimation of bulk density in the unsaturated sampling depths, whereas in the saturated sampling depths, the values were similar to those determined with the gouge auger. Total peat masses determined using the tested sampling devices differed only slightly. As those devices which can acquire samples from the amorphous or aggregated horizons are not suited for deep peat profiles, a combination of different devices will be necessary for determining bulk density and thus soil organic carbon stocks at many peatland sites. We could show that this is a reliable approach when considering site-specific conditions.