Wide & deep learning for predicting relative mineral compositions of sediment cores solely based on XRF scans, a case study from Pleistocene Paleolake Olduvai, Tanzania

This study develops a method to use deep learning models to predict the mineral assemblages and their relative abundances in paleolake cores using high-resolution XRF core scan elemental data and X-ray diffraction (XRD) mineralogical results from the same core taken at coarser resolution. It uses th...

Full description

Saved in:
Bibliographic Details
Main Authors: Gayantha R.L. Kodikara, Lindsay J. McHenry, Ian G. Stanistreet, Harald Stollhofen, Jackson K. Njau, Nicholas Toth, Kathy Schick
Format: Article
Language:English
Published: KeAi Communications Co. Ltd. 2024-12-01
Series:Artificial Intelligence in Geosciences
Online Access:http://www.sciencedirect.com/science/article/pii/S2666544124000297
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850180246413770752
author Gayantha R.L. Kodikara
Lindsay J. McHenry
Ian G. Stanistreet
Harald Stollhofen
Jackson K. Njau
Nicholas Toth
Kathy Schick
author_facet Gayantha R.L. Kodikara
Lindsay J. McHenry
Ian G. Stanistreet
Harald Stollhofen
Jackson K. Njau
Nicholas Toth
Kathy Schick
author_sort Gayantha R.L. Kodikara
collection DOAJ
description This study develops a method to use deep learning models to predict the mineral assemblages and their relative abundances in paleolake cores using high-resolution XRF core scan elemental data and X-ray diffraction (XRD) mineralogical results from the same core taken at coarser resolution. It uses the XRF core scan data along with published mineralogical information from the Olduvai Gorge Coring Project (OGCP) 2014 sediment cores 1A, 2A, and 3A from Paleolake Olduvai, Tanzania. Both regression and classification models were developed using a Keras deep learning framework to assess the predictability of mineral assemblages with their relative abundances (in regression models) or at least the mineral assemblages (in classification models) using XRF core scan data. Models were created using the Sequential class and Functional API with different model architectures. The correlation matrix of element ratios calculated from XRF element intensity records from the cores and XRD-derived mineralogical information was used to select the most useful features to train the models. 1057 training data records were used for the models. Lithological classes were also used for some models using Wide & Deep neural networks since those combine the benefits of memorization and generalization for mineral prediction. The results were validated using 265 validation data records unseen by the model and discuss the accuracy of models using six test records. The optimized Deep Neural Network (DNN) classification model achieved over 86% binary accuracy while the regression models were also able to predict the relative mineral abundances of samples with high accuracies. Overall, the study shows the efficacy of a carefully crafted Deep Learning (DL) model for predicting mineral assemblages and abundances using high-resolution XRF core scan data.
format Article
id doaj-art-296d774553e749c4a8cbcaa3a49c6a0c
institution OA Journals
issn 2666-5441
language English
publishDate 2024-12-01
publisher KeAi Communications Co. Ltd.
record_format Article
series Artificial Intelligence in Geosciences
spelling doaj-art-296d774553e749c4a8cbcaa3a49c6a0c2025-08-20T02:18:15ZengKeAi Communications Co. Ltd.Artificial Intelligence in Geosciences2666-54412024-12-01510008810.1016/j.aiig.2024.100088Wide & deep learning for predicting relative mineral compositions of sediment cores solely based on XRF scans, a case study from Pleistocene Paleolake Olduvai, TanzaniaGayantha R.L. Kodikara0Lindsay J. McHenry1Ian G. Stanistreet2Harald Stollhofen3Jackson K. Njau4Nicholas Toth5Kathy Schick6Department of Geosciences, University of Wisconsin-Milwaukee, 3209 N, Maryland Ave, Milwaukee, WI, 53211, USA; Corresponding author.Department of Geosciences, University of Wisconsin-Milwaukee, 3209 N, Maryland Ave, Milwaukee, WI, 53211, USADepartment of Earth, Ocean and Ecological Sciences, University of Liverpool, Brownlow Street, Liverpool, L69 3GP, UK; The Stone Age Institute, Bloomington, IN, 47407-5097, USAGeoZentrum Nordbayern, Friedrich-Alexander-University (FAU) Erlangen-Nümberg, Schloβgarten 5, 91054, Erlangen, GermanyThe Stone Age Institute, Bloomington, IN, 47407-5097, USA; Department of Earth and Atmospheric Sciences, Indiana University, 1001 East 10th Street, Bloomington, IN, 47405-1405, USAThe Stone Age Institute, Bloomington, IN, 47407-5097, USAThe Stone Age Institute, Bloomington, IN, 47407-5097, USAThis study develops a method to use deep learning models to predict the mineral assemblages and their relative abundances in paleolake cores using high-resolution XRF core scan elemental data and X-ray diffraction (XRD) mineralogical results from the same core taken at coarser resolution. It uses the XRF core scan data along with published mineralogical information from the Olduvai Gorge Coring Project (OGCP) 2014 sediment cores 1A, 2A, and 3A from Paleolake Olduvai, Tanzania. Both regression and classification models were developed using a Keras deep learning framework to assess the predictability of mineral assemblages with their relative abundances (in regression models) or at least the mineral assemblages (in classification models) using XRF core scan data. Models were created using the Sequential class and Functional API with different model architectures. The correlation matrix of element ratios calculated from XRF element intensity records from the cores and XRD-derived mineralogical information was used to select the most useful features to train the models. 1057 training data records were used for the models. Lithological classes were also used for some models using Wide & Deep neural networks since those combine the benefits of memorization and generalization for mineral prediction. The results were validated using 265 validation data records unseen by the model and discuss the accuracy of models using six test records. The optimized Deep Neural Network (DNN) classification model achieved over 86% binary accuracy while the regression models were also able to predict the relative mineral abundances of samples with high accuracies. Overall, the study shows the efficacy of a carefully crafted Deep Learning (DL) model for predicting mineral assemblages and abundances using high-resolution XRF core scan data.http://www.sciencedirect.com/science/article/pii/S2666544124000297
spellingShingle Gayantha R.L. Kodikara
Lindsay J. McHenry
Ian G. Stanistreet
Harald Stollhofen
Jackson K. Njau
Nicholas Toth
Kathy Schick
Wide & deep learning for predicting relative mineral compositions of sediment cores solely based on XRF scans, a case study from Pleistocene Paleolake Olduvai, Tanzania
Artificial Intelligence in Geosciences
title Wide & deep learning for predicting relative mineral compositions of sediment cores solely based on XRF scans, a case study from Pleistocene Paleolake Olduvai, Tanzania
title_full Wide & deep learning for predicting relative mineral compositions of sediment cores solely based on XRF scans, a case study from Pleistocene Paleolake Olduvai, Tanzania
title_fullStr Wide & deep learning for predicting relative mineral compositions of sediment cores solely based on XRF scans, a case study from Pleistocene Paleolake Olduvai, Tanzania
title_full_unstemmed Wide & deep learning for predicting relative mineral compositions of sediment cores solely based on XRF scans, a case study from Pleistocene Paleolake Olduvai, Tanzania
title_short Wide & deep learning for predicting relative mineral compositions of sediment cores solely based on XRF scans, a case study from Pleistocene Paleolake Olduvai, Tanzania
title_sort wide amp deep learning for predicting relative mineral compositions of sediment cores solely based on xrf scans a case study from pleistocene paleolake olduvai tanzania
url http://www.sciencedirect.com/science/article/pii/S2666544124000297
work_keys_str_mv AT gayantharlkodikara wideampdeeplearningforpredictingrelativemineralcompositionsofsedimentcoressolelybasedonxrfscansacasestudyfrompleistocenepaleolakeolduvaitanzania
AT lindsayjmchenry wideampdeeplearningforpredictingrelativemineralcompositionsofsedimentcoressolelybasedonxrfscansacasestudyfrompleistocenepaleolakeolduvaitanzania
AT iangstanistreet wideampdeeplearningforpredictingrelativemineralcompositionsofsedimentcoressolelybasedonxrfscansacasestudyfrompleistocenepaleolakeolduvaitanzania
AT haraldstollhofen wideampdeeplearningforpredictingrelativemineralcompositionsofsedimentcoressolelybasedonxrfscansacasestudyfrompleistocenepaleolakeolduvaitanzania
AT jacksonknjau wideampdeeplearningforpredictingrelativemineralcompositionsofsedimentcoressolelybasedonxrfscansacasestudyfrompleistocenepaleolakeolduvaitanzania
AT nicholastoth wideampdeeplearningforpredictingrelativemineralcompositionsofsedimentcoressolelybasedonxrfscansacasestudyfrompleistocenepaleolakeolduvaitanzania
AT kathyschick wideampdeeplearningforpredictingrelativemineralcompositionsofsedimentcoressolelybasedonxrfscansacasestudyfrompleistocenepaleolakeolduvaitanzania