Authors: Simone Priori and Leonardo Pace (Dep. of Agriculture and Forest Sciences, University of Tuscia, Viterbo, Italy)

eLTER TA site: TERENO Harz - Central German lowland - Hohes Holz experimental site, Magdeburg (Germany)

Research stay dates: 19-22 August 2024

Duration: 3 days

It is of great importance to monitor the turnover of organic matter (SOM) in forests, as forest management has a significant impact on surface radiation and SOM mineralisation, which in turn leads to an increase in CO₂ emissions. It is imperative to identify cost-effective and straightforward techniques for monitoring soil carbon stock and organic matter mineralisation in forests. This is essential for evaluating the impact of management strategies in the short and medium term. Proximal soil sensing techniques, such as near infrared diffuse reflectance (NIR-DRIFT) spectroscopy, represent a promising approach for the rapid and inexpensive collection of data, which could facilitate the provision of the necessary information. In recent years, low-cost spectrometers based on micro-electromechanical systems (MEMS) have demonstrated the capacity to predict a range of soil characteristics, particularly soil organic carbon (SOC), total nitrogen (TN) and texture (Thomas et al., 2020; Priori et al., 2022).  Nevertheless, these studies have been conducted on dried samples, sieved through a 2 mm mesh, within a laboratory setting. Meanwhile, there are several field-handheld sensors that have been developed and which we would like to test for their applicability in a forest environment.

This TA-RA proposal, entitled "Mapping Soil Organic Carbon and Nitrogen in Forest Soils with Portable NIR Spectroscopy (SONIR)", was accepted in February 2024. The objective of the SONIR project was to test the replicability of the same NIR spectrometer (Neospectra Scanner, Si-Ware) using two different devices and another model of spectrometer produced by the same company, namely the Neospectra PUCK, in forest soils. Furthermore, the most suitable method of spectra acquisition in the field was also tested.

The Italian-German team at work in the TERENO HARZ site in Hohes Holz. 

An eLTER PLUS site, such as TERENO Harz (Central German lowland, Hohes Holz), provides long-term data on soil and ecosystems, offering the potential to utilise the site as a baseline for monitoring carbon changes in the soil. The TERENO Harz site was subdivided into five areas, selected on the basis of differing forest conditions. These were: (i) scattered beech trees; (ii) scattered mixed forest (oak, beech, hornbeam); (iii) dense mixed forest; (iv) dense beech trees; (v) forest clearing with ferns. In each area, three plots were selected as replicates, situated a few metres apart. This resulted in a total of 15 plots. The humus form of each plot was described, resulting in the identification of both Oligomull and, in a limited number of cases, Dysmull. In general, the litter thickness was found to be approximately 2-3 cm, with a sparse layer of fragmented litter (OF) measuring between 0.5-1 cm in thickness. It was observed that the organic OH horizon was consistently absent.

Figure 2: Once the study plots had been selected, the litter was carefully removed in order to facilitate soil scanning with the Neospectra Scanner.

The procedures employed for the acquisition of NIR spectra were as follows:

1. The litter was removed with minimal disturbance.

2. The NIR spectra were collected using two Neospectra scanners positioned on the surface (DIRECT mode) and the surface was compacted to remove any irregularities (COMPACTION mode). Three spectra were collected per plot, with a distance of a few centimetres between each replicate.

3. The topsoil (A horizon) was collected from the 0-5 cm depth interval and transferred to three Petri dishes. The two Neospectra Scanners and the Neospectra PUCK were employed for the collection of soil spectra on Petri dishes (DISTURBED mode). Additionally, the soil samples were collected into Petri dishes for subsequent laboratory-based soil spectroscopy acquisition and conventional laboratory analysis.

4. The soil moisture content of each plot was determined in the field using a time-domain reflectometer (TDR), and in the laboratory through the thermo-gravimetric method.

Figure 3: One of the objective of the work was to verify the replicability of the spectral acquisition by testing two Neospectra scanners in the same points.


Figure 4: Spectral acquisition in disturbed samples inside Petri dishes.

Following the conventional laboratory analysis of organic carbon and total nitrogen, it would be feasible to calibrate a predictive model based on near-infrared (NIR) spectra and to assess the precision of the diverse spectrometers and acquisition techniques.

The opportunity to receive an “eLTER PLUS TA-RA” grant for this transnational collaboration has enabled us to engage in this field activity and to disseminate knowledge on the best practices for using NIR spectroscopy to monitor forest soils. We would like to thank our local hosts from the Helmholtz Centre for Environmental Research (UFZ) in Leipzig, in particular Ulrike Werban for excellent coordination and accommodation facilities; Felix Thomas, Linda Au, Simon Paasch for their invaluable support in field activity; as well as Hannes Mollenhauer, Hannah Boedeker and Maximilian Nerlich (Smart Farming Lab, Uni Leipzig) for interesting discussions about soil and viticulture.

About the authors:

Simone Priori is an Associate Professor of Pedology at the Department of Agriculture and Forest Sciences, University of Tuscia (Viterbo, Italy). His principal research interests are soil survey, mapping and monitoring, including the use of proximal soil sensors such as electromagnetic induction, gamma-ray spectroscopy and NIR diffuse reflectance spectroscopy. His research focuses on soil studies, particularly in the context of viticulture and other perennial crops, such as hazelnuts and olive trees, but also in arable lands and forest soils.

Leonardo Pace is a PhD student at the University of Tuscia (PhD Course in Plant and Animal Production Sciences), who obtained a master’s degree in forestry in 2023, with a thesis on the use of remote sensing for forest monitoring. In November 2023, he was awarded a PhD scholarship, focused on a project on the application of proximal sensing methods for soil mapping and monitoring in tree crops.