TY - JOUR
T1 - Elemental analysis of Cerrado agricultural soils via portable X-ray fluorescence spectrometry
T2 - Inferences for soil fertility assessment
AU - Lima, Tatiane Melo de
AU - Weindorf, David C.
AU - Curi, Nilton
AU - Guilherme, Luiz R.G.
AU - Lana, Regina M.Q.
AU - Ribeiro, Bruno T.
N1 - Publisher Copyright:
© 2019 Elsevier B.V.
PY - 2019/11/1
Y1 - 2019/11/1
N2 - Conventional soil chemical analyses are time-consuming and laboratory-based. Spectroscopic techniques have been introduced to make soil measurements more rapid and cost-effective. While portable X-ray fluorescence spectrometry (pXRF) has been increasingly adopted for advanced soil analysis in temperate regions, comparatively few applications have been developed for tropical soils. Using PXRF, soil total elemental composition can be obtained in seconds, with properties inferred from the collected data. In the present study, pXRF was used to screen samples (n = 252) from tropical Cerrado agricultural areas under different land uses in Brazil. The objective was to determine the total elemental composition of soil samples and predict soil fertility properties using simple or multiple linear regression with pXRF data as a proxy. Air-dried and sieved soil samples (<2 mm) under laboratory conditions were screened using a pXRF, then compared to soil chemical characterization by conventional methods. pXRF data properly corroborated the known soil mineralogy and geochemical background of Cerrado soils. Based exclusively on pXRF data, clay content was successfully predicted. Also, soil organic matter and plant-available forms of Ca, Mn and Cu were reasonably predicted from pXRF results.
AB - Conventional soil chemical analyses are time-consuming and laboratory-based. Spectroscopic techniques have been introduced to make soil measurements more rapid and cost-effective. While portable X-ray fluorescence spectrometry (pXRF) has been increasingly adopted for advanced soil analysis in temperate regions, comparatively few applications have been developed for tropical soils. Using PXRF, soil total elemental composition can be obtained in seconds, with properties inferred from the collected data. In the present study, pXRF was used to screen samples (n = 252) from tropical Cerrado agricultural areas under different land uses in Brazil. The objective was to determine the total elemental composition of soil samples and predict soil fertility properties using simple or multiple linear regression with pXRF data as a proxy. Air-dried and sieved soil samples (<2 mm) under laboratory conditions were screened using a pXRF, then compared to soil chemical characterization by conventional methods. pXRF data properly corroborated the known soil mineralogy and geochemical background of Cerrado soils. Based exclusively on pXRF data, clay content was successfully predicted. Also, soil organic matter and plant-available forms of Ca, Mn and Cu were reasonably predicted from pXRF results.
KW - Proximal sensors
KW - Soil analysis
KW - Spectroscopic methods
UR - http://www.scopus.com/inward/record.url?scp=85068615618&partnerID=8YFLogxK
U2 - 10.1016/j.geoderma.2019.06.045
DO - 10.1016/j.geoderma.2019.06.045
M3 - Article
AN - SCOPUS:85068615618
SN - 0016-7061
VL - 353
SP - 264
EP - 272
JO - Geoderma
JF - Geoderma
ER -
