Soil sampling – the basis for soil protection

Along with water and air, soils are one of the natural foundations of life and fulfil numerous functions for plants, animals and humans. However, a variety of substances are introduced into soils either intentionally (e.g. pesticides, fertilisers) or unintentionally (via the air, slurry, sewage sludge, etc.), which can impair these functions. The aim of soil protection is to preserve these functions. To this end, soil monitoring programmes are carried out, which regularly take samples at specified locations. In addition, there are individual samples taken to investigate specific issues, such as the determination of background values. The quality of the research results depends to a large extent on standardised soil sampling that is tailored to the respective research question.

Sampling for the German Environmental Specimen Bank

Since autumn 2002, Fraunhofer IME has been responsible for soil sampling for the German Environmental Specimen Bank (ESB). The aim of this sampling is the long-term archiving of soil samples under constant conditions. These are stored in a pure nitrogen atmosphere at approximately -150 °C.

Samples are taken at four year intervals at eleven sites throughout the country and have different uses e.g. forest, arable land and parks. The areas are permanently marked and protected against changes by contractual agreements. The samples are taken in September and October.

Composite samples are taken from each sampling area at three soil horizons: humus layer or root mat, topsoil and subsoil. For this purpose, 16 systematically distributed individual samples are taken. The sampling area is first divided into a 4 x 4 grid, with each grid unit in turn being subdivided into a 4 x 4 sub-grid (Fig. 1). Approximately 5 kg of fresh matter (cover, root mat) or 15 kg (topsoil and subsoil) is taken per horizon.

The samples are taken based on volume: the vegetation horizon is sampled with a frame up to the upper limit of the mineral soil (Fig. 2), while the topsoil and subsoil are taken with a divisible split-tube sampler from the previously prepared area (Fig. 3). The samples are then sieved in the field (5 mm for the surface layer/root mat, 2 mm for mineral soil) and frozen with liquid nitrogen during the sieving process (Fig. 4). They are then stored in stainless steel containers over liquid nitrogen.

The material is then transferred layer by layer into stainless steel containers and stored in a storage tank over liquid nitrogen. This procedure ensures that the samples are transferred to storage in liquid nitrogen within about an hour of sampling, largely excluding the possibility of chemical or biological alteration of the samples. All details of the sampling procedure are documented in a sampling protocol.

The samples stored over liquid nitrogen are prepared in the laboratory while still maintaining the required cold conditions. The material is homogenised in a liquid nitrogen-cooled mixer. The sample material is then divided into individual samples (approx. 100 g fresh weight for mineral soil and approx. 50 g for the layer and root mat) and stored in 100 ml Duran glass bottles in the environmental specimen bank's archive. Basic soil properties are determined on four sub-samples: grain size composition, pH value and carbon content, as well as water content. The samples stored under controlled conditions over liquid nitrogen also enable retrospective investigations for various contaminants

Sampling for determining background values and for the contamination of agricultural soils

Between 2022 and 2024, samples were taken for two projects of the German Environment Agency:

1. Determining background levels of PFAS and microplastics.
2. Investigating the contamination of agricultural soils by pesticides, biocides and veterinary medicinal products.

Before sampling, sampling concepts were created for each project that defined criteria such as representativeness, area acquisition, sample preparation and storage. Land use as arable land or grassland was selected as the representativeness criterion for the background values, with orchards and viticulture being added to this for the contamination of agricultural soils. The locations were determined according to their distribution in the individual federal states. Furthermore, the humus content of the soils was taken into account as a representativeness criterion as far as possible.

The soil samples were taken using the satellite method and corresponded to the procedure used, for example, in the soil status monitoring for the Soil survey forest. For this purpose, sub-samples were taken in at least eight segments in a circular pattern at 10 m intervals around a central point in the arable or grassland area to be sampled. The drilling pattern corresponded to a compass rose (Fig. 5).

When taking samples for the background values, care had to be taken to ensure that the samples were, as far as possible, free of any contaminants. This meant that plastic products (e.g. plastic buckets and bags) had to be avoided when taking samples. As far as clothing was concerned, care was also taken to use only natural fibres as far as possible, since ‘modern’ functional clothing always contains plastic components and often PFAS as well.

The sampling depths varied:

• Background values: 0-10 cm (grassland), plough horizon (arable land: approx. 0-20 cm to 0-30 cm).
• Contamination of agricultural soils: 0-5 cm and 5-20 cm (regardless of use).

For each sampling site and depth, approximately 5 kg of freshly sampled soil was collected and transported chilled to Fraunhofer IME. Sampling was carried out as for the ESB, using a split-tube sampler. All sampling details were also documented in a sampling protocol for these projects. In addition, a pedological drill core analysis was carried out at the sampling sites.

In the laboratory, the samples were sieved to 2 mm or 5 mm to determine the PFAS and microplastics, a sub-sample was air-dried at max. 40 °C and sent to the analytical laboratories. Another sub-sample was frozen at -20 °C with its field moisture content as a retained sample for further investigations. The samples for the contamination of agricultural soils were also sieved to 2 mm, a reference sample was frozen and sub-samples were taken for analysis both in the field-moist and air-dried state.