Exploring innovative techniques for identifying geochemical elements as fingerprints of sediment sources in an agricultural catchment of Argentina affected by soil erosion

TORRES ASTORGA, ROMINA; DOMÍNGUEZ-QUINTERO, OLGIOLY; DIAWARA, YACOUBA; VELASCO, HUGO; MEIGIKOS DOS ANJOS, ROBERTO; MABIT, LIONEL; DE LOS SANTOS VILLALOBOS, SERGIO; PEREIRA CARDOSO, RENAN; DERCON, GERD

ENVIRONMENTAL SCIENCE AND POLLUTION RESEARCH

SPRINGER HEIDELBERG

Lugar: HEIDELBERG; Año: 2018 vol. 25 p. 20868 - 20879

0944-1344

Identification of hot spots of land degradation is strongly related with the selection of soil tracers for sediment pathways. Thisresearch proposes the complementary and integrated application of two analytical techniques to select the most suitable finger-print tracers for identifying the main sources of sediments in an agricultural catchment located in Central Argentina with erosiveloess soils. Diffuse reflectance Fourier transformed in the mid-infrared range (DRIFT-MIR) spectroscopy and energy-dispersiveX-ray fluorescence (EDXRF) were used for a suitable fingerprint selection. For using DRIFT-MIR spectroscopy as fingerprintingtechnique, calibration through quantitative parameters is needed to link and correlate DRIFT-MIR spectra with soil tracers.EDXRF was used in this context for determining the concentrations of geochemical elements in soil samples. The selectedtracers were confirmed using two artificial mixtures composed of known proportions of soil collected in different sites withdistinctive soil uses. These fingerprint elements were used as parameters to build a predictive model with the whole set of DRIFT-MIR spectra. Fingerprint elements such as phosphorus, iron, calcium, barium, and titanium were identified for obtaining asuitable reconstruction of the source proportions in the artificial mixtures. Mid-infrared spectra produced successful predictionmodels (R 2 = 0.91) for Fe content and moderate useful prediction (R 2 = 0.72) for Ti content. For Ca, P, and Ba, the R 2 were 0.44,0.58, and 0.59 respectively.