In the context of the EU project Land Use / Cover Area Frame Survey (LUCAS) the land use and land cover of 250’000 locations spread over 23 countries are periodically assessed. At around 10 % of the sites soil samples are taken and analysed for soil properties like soil organic carbon, pH and texture. The results are mapped and published.
On the occasion of the International Year of Soil 2015, Switzerland and other non-EU countries had the opportunity to participate in the EU project and to contribute to an established and well-organised research project. A total of more than 30 countries took part. In Switzerland, 160 sites were sampled, in accordance with the EU Directive up to a maximum altitude of 1500 m above sea level.
Figure 1: Sampling sites of LUCAS soil Switzerland. In red the areas above 1500 m a.s.l. which are out of scope
A special feature for sampling in Switzerland was that, in addition to the official EU method using spades, the NABO standard sampling procedure using a gouge auger could also be carried out. All 160 sites were sampled twice. This was the first time that it was possible to directly compare the two methods and quantify their influence on the results.
The most important results are:
In arable land the LUCAS spade and gouge auger method produced very similar results. The Lin’s concordance coefficient is 0.98, 0.97 in meadow and 0.93 in pasture.
The soil carbon and total nitrogen contents in permanent meadows and under deciduous forest show good correlation and low mean errors for both methods.
The comparison of the two sampling methods for the monitoring of pastures and sites under coniferous forest for total nitrogen indicates a clear difference. The large mean error of the spade sampling indicates that the spade sampling is not suitable for the long-term monitoring of these land uses and soil properties in its current form.
Sampling with the gouge auger allows better control of the 20 cm penetration depth than the spade method. These deviations can be a source of error in the analysis of soil carbon and total nitrogen in topsoil samples and have greater effects in forest, pasture and permanent meadow sites than in arable sites due to the high depth gradients of these soil parameters. The arable sites are frequently ploughed and have a comparatively homogeneous distribution in the topsoil.
Overall, this study shows that the LUCAS spade method can be a valid sampling method for continental scale surveys due to relatively low sampling costs, provided that improvements in depth control are introduced. These recommendations will feed into the design of the next Europe-wide sampling campaign.
