Filtern
Dokumenttyp
Schlagworte
- Soil (3)
- LIBS (2)
- RFA (2)
- Agricultural (1)
- Agriculture (1)
- Boden (1)
- Chemometrics (1)
- Laser-induced breakdown spectroscopy (LIBS) (1)
- Multivariate (1)
- Multivariate data analysis (1)
Organisationseinheit der BAM
The quantification of the elemental content in soils with laser-induced breakdown spectroscopy (LIBS) is challenging because of matrix effects strongly influencing the plasma formation and LIBS signal. Furthermore, soil heterogeneity at the micrometre scale can affect the accuracy of analytical results. In this paper, the impact of univariate and multivariate data evaluation approaches on the quantification of nutrients in soil is discussed. Exemplarily, results for calcium are shown, which reflect trends also observed for other elements like magnesium, silicon and iron. For the calibration models, 16 certified reference soils were used. With univariate and multivariate approaches, the calcium mass fractions in 60 soils from different testing grounds in Germany were calculated. The latter approach consisted of a principal component analysis (PCA) of adequately pre-treated data for classification and identification of outliers, followed by partial least squares regression (PLSR) for quantification. For validation, the soils were also characterised with inductively coupled plasma optical emission spectroscopy (ICP OES) and X-ray fluorescence (XRF) analysis. Deviations between the LIBS quantification results and the reference analytical results are discussed.
Currently, there is almost no comprehensive mapping of agricultural cropland because of the lack of fast and affordable mapping methods for important soil properties. Arable land of some hectares in size show a broad range of different nutrient compositions. Because of this heterogeneity wrong fertilization can occur and can cause environmental pollution or lead to smaller harvests and this is simply a waste of resources. The goal of I4S (intelligence for soil) is to develop an integrated system for site-specific soil fertility management. The I4S consortium consists of 10 different institutions, which are testing various sensors for their suitability for the requested applications. Besides the preparation of reference materials, the main task of the Federal Institute for Materials Research and Testing (BAM) in this project is the method development for online-XRF (x-ray fluorescence spectroscopy) and for online-LIBS (laser-induced breakdown spectroscopy) sensor systems. Both methods have the advantage that there is no or only little sample preparation necessary. Each method provides a fast and simultaneous multi-element analysis. Both measurement setups can be utilized for a mobile application which is fundamental for reaching the goals of I4S. Having finally a set of some reference materials, further work will focus on the online-XRF and online-LIBS methods in order to obtain useable calibration models. The calibration models will then be tested using stationary and moving samples.
Per Online-RFA und -LIBS Elementgehalte in Böden bestimmen.
Die ortsspezifische Steuerung der Bodenfruchtbarkeit durch angepasste Düngung und andere Maßnahmen hilft die Bodenfunktionen zu verbessern und Umweltbelastungen zu vermindern. Dabei zeigt das Beispiel die hohe Relevanz schneller, robuster Vor-Ort-Analysen für viele umweltrelevante Fragestellungen.
Within the framework of precision agriculture, the determination of various soil properties is moving into focus, especially the demand for sensors suitable for in-situ measurements. Energy-dispersive X-ray fluorescence (EDXRF) can be a powerful tool for this purpose. In this study a huge diverse soil set (n = 598) from 12 different study sites in Germany was analysed with EDXRF. First, a principal component analysis (PCA) was performed to identify possible similarities among the sample set.
Clustering was observed within the four texture classes clay, loam, silt and sand, as clay samples contain high and sandy soils low iron mass fractions. Furthermore, the potential of uni- and multivariate data evaluation with partial least squares regression (PLSR) was assessed for accurate Determination of nutrients in German agricultural samples using two calibration sample sets. Potassium and iron were chosen for testing the performance of both models. Prediction of these nutrients in 598 German soil samples with EDXRF was more accurate using PLSR which is confirmed by a better overall averaged deviation and PLSR should therefore be preferred.