Pasar al contenido principal

Resultados relacionados

Resultados totales (Incluyendo duplicados): 3
Encontrada(s) 1 página(s)

DIGITAL.CSIC. Repositorio Institucional del CSIC
Publicación Artículo científico (JournalArticle)

Confidence of a k-Nearest Neighbors Python Algorithm for the 3D Visualization of Sedimentary Porous Media

  • Bullejos, Manuel
  • Cabezas, David
  • Martín-Martín, Manuel
  • Alcalá, Francisco J.
In a previous paper, the authors implemented a machine learning k-nearest neighbors (KNN) algorithm and Python libraries to create two 3D interactive models of the stratigraphic architecture of the Quaternary onshore Llobregat River Delta (NE Spain) for groundwater exploration purposes. The main limitation of this previous paper was its lack of routines for evaluating the confidence of the 3D models. Building from the previous paper, this paper refines the programming code and introduces an additional algorithm to evaluate the confidence of the KNN predictions. A variant of the Similarity Ratio method was used to quantify the KNN prediction confidence. This variant used weights that were inversely proportional to the distance between each grain-size class and the inferred point to work out a value that played the role of similarity. While the KNN algorithm and Python libraries demonstrated their efficacy for obtaining 3D models of the stratigraphic arrangement of sedimentary porous media, the KNN prediction confidence verified the certainty of the 3D models. In the Llobregat River Delta, the KNN prediction confidence at each prospecting depth was a function of the available data density at that depth. As expected, the KNN prediction confidence decreased according to the decreasing data density at lower depths. The obtained average-weighted confidence was in the 0.44−0.53 range for gravel bodies at prospecting depths in the 12.7−72.4 m b.s.l. range and was in the 0.42−0.55 range for coarse sand bodies at prospecting depths in the 4.6−83.9 m b.s.l. range. In a couple of cases, spurious average-weighted confidences of 0.29 in one gravel body and 0.30 in one coarse sand body were obtained. These figures were interpreted as the result of the quite different weights of neighbors from different grain-size classes at short distances. The KNN algorithm confidence has proven its suitability for identifying these anomalous results in the supposedly well-depurated grain-size database used in this study. The introduced KNN algorithm confidence quantifies the reliability of the 3D interactive models, which is a necessary stage to make decisions in economic and environmental geology. In the Llobregat River Delta, this quantification clearly improves groundwater exploration predictability., Research Project PID2020-114381GB-100 of the Spanish Ministry of Science and Innovation, Research Project 101086497 of the Horizon Europe Framework Programme HORIZON-CL6-2022-GOVERNANCE-01-07, Research Groups and Projects of the Generalitat Valenciana from the University of Alicante (CTMA-IGA), and Research Groups FQM-343 and RNM-188 of the Junta de Andalucía., Peer reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publicación Artículo científico (JournalArticle)

A K-Nearest Neighbors Algorithm in Python for Visualizing the 3D Stratigraphic Architecture of the Llobregat River Delta in NE Spain

  • Bullejos, Manuel
  • Cabezas, David
  • Martín-Martín, Manuel
  • Alcalá, Francisco J.
The k-nearest neighbors (KNN) algorithm is a non-parametric supervised machine learning classifier; which uses proximity and similarity to make classifications or predictions about the grouping of an individual data point. This ability makes the KNN algorithm ideal for classifying datasets of geological variables and parameters prior to 3D visualization. This paper introduces a machine learning KNN algorithm and Python libraries for visualizing the 3D stratigraphic architecture of sedimentary porous media in the Quaternary onshore Llobregat River Delta (LRD) in northeastern Spain. A first HTML model showed a consecutive 5 m-equispaced set of horizontal sections of the granulometry classes created with the KNN algorithm from 0 to 120 m below sea level in the onshore LRD. A second HTML model showed the 3D mapping of the main Quaternary gravel and coarse sand sedimentary bodies (lithosomes) and the basement (Pliocene and older rocks) top surface created with Python libraries. These results reproduce well the complex sedimentary structure of the LRD reported in recent scientific publications and proves the suitability of the KNN algorithm and Python libraries for visualizing the 3D stratigraphic structure of sedimentary porous media, which is a crucial stage in making decisions in different environmental and economic geology disciplines., Research Project PID2020-114381GB-100 of the Spanish Ministry of Science and Innovation, Research Groups and Projects of the Generalitat Valenciana from the University of Alicante (CTMA-IGA), and Research Groups FQM-343 and RNM-188 of the Junta de Andalucía., Peer reviewed
Dipòsit Digital de Documents de la UAB
Publicación Artículo científico (JournalArticle)

Sharp-based, mixed carbonate-siliciclastic shallow-marine deposits (upper Miocene, Betic Cordillera, Spain), the record of ancient transgressive shelf ridges?

  • Poyatos-Moré, Miquel|||0000-0001-7813-8868
  • García-García, F.
  • Rodríguez-Tovar, F. J.
  • Soria, J.
  • Viseras Iborra, César
  • Pérez-Valera, Fernando|||0000-0002-8623-9148
  • Midtkandal, I.
Isolated sharp-based sedimentary bodies in shelf settings can develop via the reworking of regressive deposits during transgressions. An example of these are shelf ridges, formed under a wide range of processes, and widely studied due to their high reservoir potential. However, there is still a lack of examples in mixed (carbonate-siliciclastic) successions. This study presents an outcrop example from the Upper Miocene of the Betic Cordillera (Spain), with the aim to propose a model for the development of transgressive sharp-based mixed carbonate-siliciclastic deposits, and to provide criteria to differentiate these from their regressive counterparts. The studied succession is ca. 300 m-thick, and shows a cyclic alternation of coarse and fine-grained mixed deposits. Depositional cycles start with siliciclastic-dominated offshore to offshore transition deposits, progressively replaced by lower shoreface deposits. These are abruptly truncated by sharp erosive contacts bioturbated by passively-infilled large burrows; their ichnological features allow assignation to the Glossifungites ichnofacies. These contacts are interpreted as ravinement surfaces. They are overlain by mixed carbonate-siliciclastic barforms, rich in skeletal fragments and extraclasts, and displaying large-scale cross bedding. These form several m-thick and hundreds of m-long depositional elements interpreted as mixed shelf ridges. These ridges formed in a fine-grained, shallow-water shelf, which occasionally received coarse siliciclastic sediment supply via gravity flows, but had a coeval offshore carbonate factory, which provided the skeletal fragments. The sharp-based, coarser-grained nature and lithological break at the base of these mixed carbonate-clastic deposits could lead to their misinterpretation as forced-regressive wedges. However, the nature of their lower contact, combined with the reworked offshore skeletal fragments, and their stacking pattern are consistent with these mixed units forming during transgression. Other studies in relatively time-equivalent deposits have demonstrated the existence of coeval regressive, coarser siliciclastic-dominated shoreline systems in relatively close localities. These evidence a complex basin configuration in the area during the upper Miocene, with the development of local depocentres and relatively narrow corridors or seaways in the Mediterranean-Atlantic connection, which could have favoured shelf reworking processes, but also promoted the development of diverse stacking patterns, reflecting the differential interaction between active tectonics and sedimentation across the region.