Discovering geochemical patterns using self-organizing neural networks: A new perspective for sedimentary provenance analysis

Juan Pablo Lacassie, Barry Roser, Javier Ruiz Del Solar, Francisco Hervé

Resultado de la investigación: Article

27 Citas (Scopus)

Resumen

An unsupervised neural network technique, Growing Cell Structures (GCS), was used to visualize geochemical differences between sandstones of four different sedimentary provenance groups: P1 (mafic), P2 (intermediate), P3 (felsic), and P4 (recycled). Multidimensional data of four sandstone data sets comprising major elements, log-normalized major elements, trace elements, and high field strength elements (HFSE) were projected into colored two-dimensional maps that can be visually and quantitatively interpreted. The cluster structure and variable distributions produced show that each sedimentary provenance group can be distinguished in the neural maps according to a unique combination of major or trace element concentrations. In these terms, the distinguishing features of each provenance group are: P1-high Fe2O3t, TiO2, MgO, MnO, CaO, P2O5, Sc, V, Cr, and Cu; P2-intermediate Fe2O3t, TiO2, MgO, MnO, CaO, Sc, V, and Cu; P3-intermediate to high K2O, intermediate SiO2 and Al2O3, low Fe2O3t and TiO2, and intermediate to low Nb, Rb, and Th; P4-high SiO2, Y, Nb, Rb, Th, Ba, and Zr, coupled with low Al2O3, CaO, Na2O, Fe2O3t, MgO, MnO, and TiO2. The elemental associations in P1, P2, and P3 reflect petrogenetic evolution of first-cycle sources, whereas the associations in P4 are compatible with the combined effects of recycling, weathering, and heavy mineral concentration.

Idioma originalEnglish
Páginas (desde-hasta)175-191
Número de páginas17
PublicaciónSedimentary Geology
Volumen165
N.º1-2
DOI
EstadoPublished - 1 mar 2004

Huella dactilar

provenance
sandstone
trace element
heavy mineral
recycling
weathering
analysis
effect
distribution

ASJC Scopus subject areas

  • Geology
  • Stratigraphy

Citar esto

Lacassie, Juan Pablo ; Roser, Barry ; Ruiz Del Solar, Javier ; Hervé, Francisco. / Discovering geochemical patterns using self-organizing neural networks : A new perspective for sedimentary provenance analysis. En: Sedimentary Geology. 2004 ; Vol. 165, N.º 1-2. pp. 175-191.
@article{5509849ec1be4bf5a11f6a7f361cba31,
title = "Discovering geochemical patterns using self-organizing neural networks: A new perspective for sedimentary provenance analysis",
abstract = "An unsupervised neural network technique, Growing Cell Structures (GCS), was used to visualize geochemical differences between sandstones of four different sedimentary provenance groups: P1 (mafic), P2 (intermediate), P3 (felsic), and P4 (recycled). Multidimensional data of four sandstone data sets comprising major elements, log-normalized major elements, trace elements, and high field strength elements (HFSE) were projected into colored two-dimensional maps that can be visually and quantitatively interpreted. The cluster structure and variable distributions produced show that each sedimentary provenance group can be distinguished in the neural maps according to a unique combination of major or trace element concentrations. In these terms, the distinguishing features of each provenance group are: P1-high Fe2O3t, TiO2, MgO, MnO, CaO, P2O5, Sc, V, Cr, and Cu; P2-intermediate Fe2O3t, TiO2, MgO, MnO, CaO, Sc, V, and Cu; P3-intermediate to high K2O, intermediate SiO2 and Al2O3, low Fe2O3t and TiO2, and intermediate to low Nb, Rb, and Th; P4-high SiO2, Y, Nb, Rb, Th, Ba, and Zr, coupled with low Al2O3, CaO, Na2O, Fe2O3t, MgO, MnO, and TiO2. The elemental associations in P1, P2, and P3 reflect petrogenetic evolution of first-cycle sources, whereas the associations in P4 are compatible with the combined effects of recycling, weathering, and heavy mineral concentration.",
keywords = "Geochemistry, Neural networks, Sandstone, Sedimentary provenance, Unsupervised learning, Visual analysis",
author = "Lacassie, {Juan Pablo} and Barry Roser and {Ruiz Del Solar}, Javier and Francisco Herv{\'e}",
year = "2004",
month = "3",
day = "1",
doi = "10.1016/j.sedgeo.2003.12.001",
language = "English",
volume = "165",
pages = "175--191",
journal = "Sedimentary Geology",
issn = "0037-0738",
publisher = "Elsevier",
number = "1-2",

}

Discovering geochemical patterns using self-organizing neural networks : A new perspective for sedimentary provenance analysis. / Lacassie, Juan Pablo; Roser, Barry; Ruiz Del Solar, Javier; Hervé, Francisco.

En: Sedimentary Geology, Vol. 165, N.º 1-2, 01.03.2004, p. 175-191.

Resultado de la investigación: Article

TY - JOUR

T1 - Discovering geochemical patterns using self-organizing neural networks

T2 - A new perspective for sedimentary provenance analysis

AU - Lacassie, Juan Pablo

AU - Roser, Barry

AU - Ruiz Del Solar, Javier

AU - Hervé, Francisco

PY - 2004/3/1

Y1 - 2004/3/1

N2 - An unsupervised neural network technique, Growing Cell Structures (GCS), was used to visualize geochemical differences between sandstones of four different sedimentary provenance groups: P1 (mafic), P2 (intermediate), P3 (felsic), and P4 (recycled). Multidimensional data of four sandstone data sets comprising major elements, log-normalized major elements, trace elements, and high field strength elements (HFSE) were projected into colored two-dimensional maps that can be visually and quantitatively interpreted. The cluster structure and variable distributions produced show that each sedimentary provenance group can be distinguished in the neural maps according to a unique combination of major or trace element concentrations. In these terms, the distinguishing features of each provenance group are: P1-high Fe2O3t, TiO2, MgO, MnO, CaO, P2O5, Sc, V, Cr, and Cu; P2-intermediate Fe2O3t, TiO2, MgO, MnO, CaO, Sc, V, and Cu; P3-intermediate to high K2O, intermediate SiO2 and Al2O3, low Fe2O3t and TiO2, and intermediate to low Nb, Rb, and Th; P4-high SiO2, Y, Nb, Rb, Th, Ba, and Zr, coupled with low Al2O3, CaO, Na2O, Fe2O3t, MgO, MnO, and TiO2. The elemental associations in P1, P2, and P3 reflect petrogenetic evolution of first-cycle sources, whereas the associations in P4 are compatible with the combined effects of recycling, weathering, and heavy mineral concentration.

AB - An unsupervised neural network technique, Growing Cell Structures (GCS), was used to visualize geochemical differences between sandstones of four different sedimentary provenance groups: P1 (mafic), P2 (intermediate), P3 (felsic), and P4 (recycled). Multidimensional data of four sandstone data sets comprising major elements, log-normalized major elements, trace elements, and high field strength elements (HFSE) were projected into colored two-dimensional maps that can be visually and quantitatively interpreted. The cluster structure and variable distributions produced show that each sedimentary provenance group can be distinguished in the neural maps according to a unique combination of major or trace element concentrations. In these terms, the distinguishing features of each provenance group are: P1-high Fe2O3t, TiO2, MgO, MnO, CaO, P2O5, Sc, V, Cr, and Cu; P2-intermediate Fe2O3t, TiO2, MgO, MnO, CaO, Sc, V, and Cu; P3-intermediate to high K2O, intermediate SiO2 and Al2O3, low Fe2O3t and TiO2, and intermediate to low Nb, Rb, and Th; P4-high SiO2, Y, Nb, Rb, Th, Ba, and Zr, coupled with low Al2O3, CaO, Na2O, Fe2O3t, MgO, MnO, and TiO2. The elemental associations in P1, P2, and P3 reflect petrogenetic evolution of first-cycle sources, whereas the associations in P4 are compatible with the combined effects of recycling, weathering, and heavy mineral concentration.

KW - Geochemistry

KW - Neural networks

KW - Sandstone

KW - Sedimentary provenance

KW - Unsupervised learning

KW - Visual analysis

UR - http://www.scopus.com/inward/record.url?scp=1542319212&partnerID=8YFLogxK

U2 - 10.1016/j.sedgeo.2003.12.001

DO - 10.1016/j.sedgeo.2003.12.001

M3 - Article

AN - SCOPUS:1542319212

VL - 165

SP - 175

EP - 191

JO - Sedimentary Geology

JF - Sedimentary Geology

SN - 0037-0738

IS - 1-2

ER -