Explore the Knowledge Hub

Data from the FP7 iSOIL research projectdealing with new technologies for assessing soil properties along depth accurately and with high resolution by combining existing sensors as well as exploring emerging technologies (mainly geophysical); data available for the test sites of iSoil, located in Germany, Austria, Czech Republic and Bulgaria.

Soil structure is an essential indicator of soil health. The data set consists of DICOM data from 36 soil cores, assessed by medical X-ray CT, from 36 soil cores (height and width 12 cm) sampled in 13 European countries sampled 2017/18 within the framework of the COST Action ES1406 (KEYSOM). Undisturbed soil samples from forest and grassland sites were taken manually with a special device, shipped to Germany and scanned centrally with a medical CT scanner at the Leibniz-Institute for Zoo- and Wildlife Research in Berlin.

From Scopus we extracted the research papers containing the word ‘soil’ in the title that on July 13th 2018 counted more than 500 citations. From this collection of 310 papers we excluded the unobtainable ones (in orange), those that do not contain original data or report aggregated data not attributable to any single soil (in green), and those clearly not dealing with soil in spite of the title (in yellow). So, we selected 174 suitable papers, which were checked for i) the presence of the soil name (according to any classification system) in the main text, the Supplementary Materials, external databases linked to the paper, or publications quoted in the main paper, and ii) the following info: geographic coordinates of the sampled soil, particle size distribution/texture class, pH, organic carbon content, cation exchange capacity, and any other ancillary soil data.

This dataset consists of a number of data layers (raster GRID maps) that are associated to the peer-reviewed publication "Assessment of soil organic carbon stocks under future climate and land cover changes in Europe" . Layers cover the current Soil Organic Carbon Stocks (2016) and the projected Soil Organic Carbon Stocks by 2050, for various Climate Scenarios (CCSM4, HadGEM2-AO , IPSL-CM5A-LR MRI-CGCM3) and Representative Concentration Pathways (RCPs).

We provide information on the follwing: Article ID: A unique study ID First Author: The name of the first author Year (soil collected): Year of collection (whenever available) of the soil. This can differ from the publication year lat: latitude long: longitude: C mg/g: soil organic carbon N mg/g: soil total nitrogen δ13C: carbon stable isotope of soil δ15N: nitrogen stable isotope of soil Forest type: type of forest n: replication number pH: soil pH       Data collection On September 20, 2024, we carried out a systematic literature search in the Web of Science Core Collection using the search string "(delta or 15N or 13C) and soil and forest". The initial search yielded 9917 non-review articles. We carried out the initial filtering based on titles and abstracts, which filtered out a total of 9499 articles. We subsequently checked the main text of the remaining articles, which resulted in a set of 236 papers that met our inclusion criteria (Fig. S1). Our inclusion criteria were that (a) at least one isotopic value (δ¹³C or δ¹⁵N); (b) the observations were taken from forest mineral soils. Over the literature search phase, we preserved entries that only contained one of the two isotopes but for subsequent analyses we worked on the subset of entries for which we had both soil δ¹³C and δ¹⁵N values. For each selected paper, we extracted the following variables: the mean values and standard deviation of δ¹³C, δ¹⁵N, soil sample replicates (n), total nitrogen (TN), total carbon (TC), total phosphorus (TP), soil pH, sampling depth, geographic coordinates (latitude and longitude) of sampling locations, year of soil collection, forest succession stage (classified as primary forest, secondary forest, or plantation), forest type (categorized as evergreen broadleaf, deciduous broadleaf, coniferous forest or a mix of the above). Data were obtained directly from published tables or, when necessary, extracted from figures using PlotDigitizer (https://plotdigitizer.com). The forest type classification was based on either explicit descriptions or the dominant species of the forest. The forest succession stage was determined through textual information regarding: the presence of anthropogenic or natural disturbances, natural growth history, and evidence of management. We further extracted information, based on coordinates on the mean annual temperature and mean annual precipitation (WorldClim(Hijmans et al., 2005)), total, dry and wet deposition(Ackerman et al., 2018), aridity index and evaporotranspiration(Zomer et al., 2022) and Köppen climate(Beck et al., 2018).

A soil sample is a sample of the soil that is taken for further analysis in the field or in a laboratory. A sample is always taken at a certain depth (from/to). Optionally, a certain technique can be described (e.g. disturbed or undisturbed), as well as the conditions of sampling (e.g. atmospheric conditions, etc.). The results of analyses carried out on the sample are stored as observations linked to the sample. A sample can possibly be linked to one or more commands and attachments can also be linked to a sample (for example, analysis results or reports). A soil sample is either a single sample or a composite sample (this is the type). A single sample is always linked to one soil location or one depth interval. A soil location or depth interval may have 0 or more single samples. A composite sample can be linked to one soil site, one soil location or one soil depth interval. These may have 0 or more composite samples.