Title: Assessment of changes in topsoil properties in LUCAS samples between 2009/2012 and 2015 surveys
Authors: FERNANDEZ UGALDE OIHANEBALLABIO CRISTIANOLUGATO EMANUELESCARPA SIMONEJONES ARWYN
Publisher: Publications Office of the European Union
Publication Year: 2020
JRC N°: JRC120138
ISBN: 978-92-76-17430-1 (online)
ISSN: 1831-9424 (online)
Other Identifiers: EUR 30147 EN
OP KJ-NA-30147-EN-N (online)
URI: https://publications.jrc.ec.europa.eu/repository/handle/JRC120138
DOI: 10.2760/5503
Type: EUR - Scientific and Technical Research Reports
Abstract: Soil delivers fundamental ecosystem services that support human well-being. These include the provision of food, feed, fuel, fibre and genetic resources, the regulation of storage, filtration and cycling of nutrients and water, cultural (aesthetic, spiritual and recreational) values and supporting the provision of all other services. Policies for sustainable land and soil management should be based on monitoring systems that are able to provide evidence of the impact of land use/land cover changes and climate change in soil condition, both in space and in time. In this context, the topsoil assessment module of the Land Use and Cover Area Frame Survey (LUCAS) is the first harmonised soil monitoring network at European Union (EU) level that uses a common sampling procedure and standard analysis methods. Eurostat has carried out the LUCAS survey every 3 years since 2006. The surveys are based on the visual assessment of environmental and structural elements of the landscape in georeferenced control points, a subsample of which is selected to be visited to collect field-based information. In 2009, a soil assessment module was added within the LUCAS survey with the scope to create a harmonised and comparable dataset of physical and chemical properties of topsoil across the EU to support policymaking. About 20,000 soil points were selected across 27 member states (except Bulgaria and Romania) based on a stratified sampling scheme with land use and terrain information as attributes. At each point, samples were collected from a depth of 20 cm using a common sampling procedure. Subsequently, the samples were analysed for several properties in a single laboratory using standard analytical methods. The same point selection procedure, sampling method and analysis methods were extended in 2012 to Bulgaria and Romania, where samples were collected for about 2,000 soil points. The LUCAS Topsoil Survey was repeated in 2015, year in which 17,613 soil points sampled in the LUCAS 2009 and 2012 surveys were revisited. Furthermore, new soil points at an altitude of 1,000 - 2,000 m were added to the survey (the altitude limit was 1,000 m in LUCAS 2009 and 2012 surveys). The soil module was also extended by the JRC to Albania, Bosnia and Herzegovina, Croatia, Montenegro, Republic of North Macedonia and Serbia. In total, 27,069 points were selected for the topsoil survey in 2015, of which 25,947 were located in the EU-28 MS. In this report, we provide a detailed evaluation of the LUCAS topsoil sampling and the laboratory analysis. We also assess changes in topsoil properties between LUCAS 2009/2012 and 2015 surveys based on data of paired samples (i.e. samples collected in revisited LUCAS soil points in 2009/2012 and in 2015). The ultimate goal of this report is to assess the efficacy of the LUCAS Topsoil Module for the early detection of changes in soil conditions, since this is a primary objective for scientific and policy organizations to improve their policies for a sustainable land use and management. The LUCAS spade sampling is an efficient and cost-effective method for topsoil monitoring at regional/continental scale, although a better control of litter removal in woodland and sampling depth in all LC classes is needed. When comparing sampling locations of revisited points, almost 97 % of the samples taken in 2015 were taken at a distance <100 m from their baseline locations in 2009/2012 as indicated in the sampling protocol. Three percent of the samples were taken at a distance between 100 and 400 m from one survey to the other. As a result, changes in soil properties were not significantly affected by the distance between sampling locations in the 2009/2012 and 2015 surveys. Regarding laboratory analysis, the data of the properties analysed showed a coherence from the soil point of view. Organic carbon and N levels showed a positive correlation, CaCO3 content was lower in samples were pH was below 7, and the sum of sand, silt and clay percentages was between 99 and 101 in the fraction <2 mm of all samples. Overall, OC and N levels were highest in woodland, followed by grassland and cropland in both the 2009/2012 and 2015 surveys. On the contrary, P and K levels were higher in cropland and grassland than in woodland in the surveys. Carbonate content was lowest in woodland from northern member states and highest in cropland from southern member states in both surveys. In agreement with these results, pH was lower in woodland than in cropland in both surveys. Soil properties showed large standard deviations within surveys and between surveys due to uncertainties arising from the sampling. Unfortunately, some LC classes were under sampled. Consideration should be given to increase the number of sampling sites in future surveys to ensure representative data. Overall, most of the soil properties showed limited changes between 2009 and 2015 (over the six-year period) in the 27 member states. Changes in Bulgaria and Romania were even less evident over the three-year period (from 2012 to 2015). Thus, the survey confirms that soil properties change very slowly over time. From a policy perspective, a time lapse longer than six years is necessary in order to observe small variations in soil conditions, unless a marked change has occurred due to erosion processes, extreme meteorological events or land use/cover changes. Despite uncertainties arising from the sampling, it has been possible to draw some conclusions when assessing changes in soil properties between 2009/2012 and 2015 surveys in mineral soils (i.e. where OC <120 g kg-1). — Taking the revisited points, a statistically significant increase in OC content of 3.74 % was observed in grassland over six years in the 27 member states. This is in line with the annual 0.4 % increase in the topsoil (30-40 cm) targeted by the ‘4 per 1000’ initiative. This would contribute to climate change mitigation. — Similarly, for the revisited points in cropland, a statistically significant decrease in OC content of 2.5 % was observed while points that changed from grassland to cropland over six years decreased by 11 %. This suggests that cropland soils are not working as carbon sinks. — In other land cover categories, the number of repeated points was insufficient to assess statistical significance. — No tangible changes were observed in Bulgaria and Romania over three years. — Nitrogen content increased in cropland, grassland, woodland points, and in points that changed from cropland to grassland over six years in the 27 member states. In Bulgaria and Romania, N content increased in cropland points and in points that changed from cropland to grassland and vice-versa over three years. In non-agricultural conditions, this may reflect airborne deposition of nitrogen. — Phosphorus content increased in cropland, grassland and woodland points over six years in the 27 member states. On the contrary, K content decreased in cropland points in the 27 member states. In Bulgaria and Romania, no tangible changes were observed over three years. — pH in CaCl2 was a more consistent measurement and was less affected by seasonal fluctuations of electrolyte concentration in soil solution. — pH in CaCl2 increased in cropland and woodland points, and in points that changed from woodland to shrubland over six years in the 27 member states. On the contrary, pH in CaCl2 decreased in grassland points. In Bulgaria and Romania, pH in CaCl2 decreased in grassland points over three years.
JRC Directorate:Sustainable Resources

Files in This Item:
File Description SizeFormat 
lucas_changes_09-15_-_final_1.pdf8.26 MBAdobe PDFView/Open


Items in repository are protected by copyright, with all rights reserved, unless otherwise indicated.