東華大學圖書館 |

Application of Geophysical and Machine Learning Methods for Interpreting Marine Gas Hydrates /

Record Type:	Electronic resources : Monograph/item
Title/Author:	Application of Geophysical and Machine Learning Methods for Interpreting Marine Gas Hydrates // Fawz Naim.
Author:	Naim, Fawz,
Description:	1 electronic resource (197 pages)
Notes:	Source: Dissertations Abstracts International, Volume: 86-12, Section: B.
Contained By:	Dissertations Abstracts International86-12B.
Subject:	Energy. -
Online resource:	https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=32142659
ISBN:	9798315779612

Application of Geophysical and Machine Learning Methods for Interpreting Marine Gas Hydrates /
Naim, Fawz,

Application of Geophysical and Machine Learning Methods for Interpreting Marine Gas Hydrates /Fawz Naim. - 1 electronic resource (197 pages)

Source: Dissertations Abstracts International, Volume: 86-12, Section: B.

The formation and accumulation of natural gas hydrate systems across Earth's offshore basins is still unknown. In this thesis, I present new knowledge on gas hydrate systems on continental slopes around the world by studying the distribution, accumulation and the source of hydrate. Also, there is often lack of sufficient data to interpret gas hydrate in the sediments on continental slope. To address this issue of missing data, I develop a machine learning model to predict downhole logs that will be important for characterizing hydrate systems in marine sediments.In Chapter 2, I develop a machine learning model to predict compressional velocity and bulk density logs in marine gas hydrate systems. Compressional velocity and bulk density are two most important downhole logging measurements for interpreting gas hydrate. However, these logs are often not acquired or sometimes poor quality in the hydrate stability zone due to borehole washouts or tool issues. I use six different supervised machine learning algorithms to estimate compressional velocity and bulk density logs in discrete intervals including water saturated sediments, hydrate in the primary pore space and hydrate in near-vertical fractures. I use input data from logging-while drilling tools to train the machine learning model. Out of the six machine learning algorithms (multilinear regression, polynomial regression, polynomial regression with ridge regularization, K nearest neighbors, random forest and multilayer perceptron), I find that random forest and K-nearest neighbors are the best algorithms to predict the two logs in marine gas hydrate.In Chapter 3, I conduct the first basin wide study in the North Carnarvon Basin, offshore Western Australia to interpret the hydrate system using well and seismic data from the petroleum industry. I find hydrate in 43% wells (52 out of 120) distributed throughout the hydrate stability zone mostly in low concentrations. Moreover, there is no connection between the presence of hydrate in wells and deep thermogenic reservoir below. On seismic data, I observe a bottom simulating reflection (BSR) in a small area; this shows that BSRs are not common in the area. All these observations present the evidence for a microbial source for the hydrate system in the North Carnarvon Basin, offshore Western Australia.In Chapter 4, I identify and interpret the hydrate system for the first time in the Faroe-Shetland Basin, offshore UK, using well and seismic data from the petroleum industry. I find that the hydrate system in the Faroe-Shetland Basin is predominantly microbial in origin, which contradicts with a consensus opinion that the hydrate system across Europe are thermogenic in origin. In addition, I argue that most hydrate systems across Europe are likely microbial in origin. In the Faroe-Shetland Basin, I identify hydrate in almost all the wells (19 out of 20 wells) that is mostly accumulated in low concentrations in discrete intervals throughout the hydrate stability zone, signifying a microbial source of hydrate. On the seismic data, I identify the first BSR offshore UK, which covers an area of 66 km2 on the continental slope. The BSR is associated with faults that are not deep seated and is intersected by two low concentration hydrate wells. Moreover, the root mean square amplitude values near the two hydrate wells and in the area above the BSR is consistently low. These observations suggest that there is an insignificant accumulation of hydrate in the vicinity of the BSR, which means a microbial source could likely supply the source gas needed for this hydrate system.In Chapter 5, I assess hydrate at International Ocean Discovery Program (IODP) Expedition 400 drill sites in the Baffin Bay, offshore NW Greenland using downhole logging and core data acquired from Sites U1603, U1604, U1607 and U1608. Moreover, I use core data acquired at Site U1606 for assessing hydrate as no downhole logging data was available at this site. At Site U1606, there are intervals within the hydrate stability zone where porewater salinity drops however, a further interpretation for hydrate assessment was not possible due to the unavailability of the downhole logging data. Due to the limitations of the acquired data, evidence for the presence of hydrate could not be found at any drill site. However, hydrate might still be present in the region.

English

ISBN: 9798315779612Subjects--Topical Terms:

876794
Energy.
Subjects--Index Terms:

Well logs

Application of Geophysical and Machine Learning Methods for Interpreting Marine Gas Hydrates /
LDR:05951nmm a22004933i 4500 001 2437443
005 20251119142816.5
006 m o d
007 cr|nu||||||||
008 261011s2024 miu||||||m |||||||eng d
020 $a 9798315779612
035 $a (MiAaPQD)AAI32142659
035 $a (MiAaPQD)OhioLINKosu1735049308721074
035 $a AAI32142659
040 $a MiAaPQD $b eng $c MiAaPQD $e rda
100 1 $a Naim, Fawz, $e author. $3 3828418
245 1 0 $a Application of Geophysical and Machine Learning Methods for Interpreting Marine Gas Hydrates / $c Fawz Naim.
264 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2024
300 $a 1 electronic resource (197 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertations Abstracts International, Volume: 86-12, Section: B.
500 $a Advisors: Cook, Ann Committee members: Cole, David; Sawyer, Derek; Moortgat, Joachim.
502 $b Ph.D. $c The Ohio State University $d 2024.
520 $a The formation and accumulation of natural gas hydrate systems across Earth's offshore basins is still unknown. In this thesis, I present new knowledge on gas hydrate systems on continental slopes around the world by studying the distribution, accumulation and the source of hydrate. Also, there is often lack of sufficient data to interpret gas hydrate in the sediments on continental slope. To address this issue of missing data, I develop a machine learning model to predict downhole logs that will be important for characterizing hydrate systems in marine sediments.In Chapter 2, I develop a machine learning model to predict compressional velocity and bulk density logs in marine gas hydrate systems. Compressional velocity and bulk density are two most important downhole logging measurements for interpreting gas hydrate. However, these logs are often not acquired or sometimes poor quality in the hydrate stability zone due to borehole washouts or tool issues. I use six different supervised machine learning algorithms to estimate compressional velocity and bulk density logs in discrete intervals including water saturated sediments, hydrate in the primary pore space and hydrate in near-vertical fractures. I use input data from logging-while drilling tools to train the machine learning model. Out of the six machine learning algorithms (multilinear regression, polynomial regression, polynomial regression with ridge regularization, K nearest neighbors, random forest and multilayer perceptron), I find that random forest and K-nearest neighbors are the best algorithms to predict the two logs in marine gas hydrate.In Chapter 3, I conduct the first basin wide study in the North Carnarvon Basin, offshore Western Australia to interpret the hydrate system using well and seismic data from the petroleum industry. I find hydrate in 43% wells (52 out of 120) distributed throughout the hydrate stability zone mostly in low concentrations. Moreover, there is no connection between the presence of hydrate in wells and deep thermogenic reservoir below. On seismic data, I observe a bottom simulating reflection (BSR) in a small area; this shows that BSRs are not common in the area. All these observations present the evidence for a microbial source for the hydrate system in the North Carnarvon Basin, offshore Western Australia.In Chapter 4, I identify and interpret the hydrate system for the first time in the Faroe-Shetland Basin, offshore UK, using well and seismic data from the petroleum industry. I find that the hydrate system in the Faroe-Shetland Basin is predominantly microbial in origin, which contradicts with a consensus opinion that the hydrate system across Europe are thermogenic in origin. In addition, I argue that most hydrate systems across Europe are likely microbial in origin. In the Faroe-Shetland Basin, I identify hydrate in almost all the wells (19 out of 20 wells) that is mostly accumulated in low concentrations in discrete intervals throughout the hydrate stability zone, signifying a microbial source of hydrate. On the seismic data, I identify the first BSR offshore UK, which covers an area of 66 km2 on the continental slope. The BSR is associated with faults that are not deep seated and is intersected by two low concentration hydrate wells. Moreover, the root mean square amplitude values near the two hydrate wells and in the area above the BSR is consistently low. These observations suggest that there is an insignificant accumulation of hydrate in the vicinity of the BSR, which means a microbial source could likely supply the source gas needed for this hydrate system.In Chapter 5, I assess hydrate at International Ocean Discovery Program (IODP) Expedition 400 drill sites in the Baffin Bay, offshore NW Greenland using downhole logging and core data acquired from Sites U1603, U1604, U1607 and U1608. Moreover, I use core data acquired at Site U1606 for assessing hydrate as no downhole logging data was available at this site. At Site U1606, there are intervals within the hydrate stability zone where porewater salinity drops however, a further interpretation for hydrate assessment was not possible due to the unavailability of the downhole logging data. Due to the limitations of the acquired data, evidence for the presence of hydrate could not be found at any drill site. However, hydrate might still be present in the region.
546 $a English
590 $a School code: 0168
650 4 $a Energy. $3 876794
650 4 $a Geophysics. $3 535228
650 4 $a Geology. $3 516570
650 4 $a Acoustics. $3 879105
650 4 $a Marine geology. $3 3173821
650 4 $a Petroleum geology. $3 3173827
653 $a Well logs
653 $a Machine learning
653 $a Gas hydrate
653 $a Gas source
653 $a Petrophysics
690 $a 0583
690 $a 0791
690 $a 0556
690 $a 0373
690 $a 0372
690 $a 0986
690 $a 0467
710 2 $a The Ohio State University. $b Earth Sciences. $e degree granting institution. $3 3826345
720 1 $a Cook, Ann $e degree supervisor.
773 0 $t Dissertations Abstracts International $g 86-12B.
790 $a 0168
791 $a Ph.D.
792 $a 2024
856 4 0 $u https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=32142659