東華大學圖書館 |

Understanding Hurricane-Induced Water Quantity and Quality Dynamics Using Machine Learning and Environmental Data Analytics Approaches.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Understanding Hurricane-Induced Water Quantity and Quality Dynamics Using Machine Learning and Environmental Data Analytics Approaches./
作者:	Fidan, Emine Nur.
面頁冊數:	1 online resource (157 pages)
附註:	Source: Dissertations Abstracts International, Volume: 84-10, Section: A.
Contained By:	Dissertations Abstracts International84-10A.
標題:	Water quality. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=30364015click for full text (PQDT)
ISBN:	9798377685050

Understanding Hurricane-Induced Water Quantity and Quality Dynamics Using Machine Learning and Environmental Data Analytics Approaches.
Fidan, Emine Nur.

Understanding Hurricane-Induced Water Quantity and Quality Dynamics Using Machine Learning and Environmental Data Analytics Approaches. - 1 online resource (157 pages)

Source: Dissertations Abstracts International, Volume: 84-10, Section: A.

Thesis (Ph.D.)--North Carolina State University, 2023.

Includes bibliographical references

Hurricanes are disruptive and alter water quantity and quality dynamics in surface waters. Process based models are commonly used to assess flood extent and water quality changes. However, these models typically require sophisticated boundary conditions, an understanding of complex drainage systems, abundance of validation data, expert domain knowledge, and high computational power. This dissertation focuses on advancing the use of machine learning, statistical modeling, and data analytics to evaluate flood dynamics and water quality trends induced by extreme events.The specific objectives of this research were to (1) evaluate the application of the Random Forest, a machine learning algorithm, for daily flood extent modeling in lowlying and flat terrains using rainfall and physiographic characteristics as predictors, (2) explain floodwater nutrient concentrations as a function of watershed characteristics using Bayesian statistical learning methods, and (3) quantify hurricane impacts on dissolved oxygen trends within different estuaries using data analytics.Objective 1 was accomplished by processing satellite imagery and delineating landscape characteristics to produce a timeline of pluvial flooding from Hurricanes Matthew (2016) and Florence (2018). A Random Forest model was trained within the North Carolina (NC) Coastal Plain and results indicated that the model predicts pluvial flooding well, with overall accuracy of 0.96 and F1 score of 0.83. An evaluation of model variables indicated that physically-based predictor variables, distance to the nearest stream, distance to the nearest road, and height above nearest drainage, were the most important in flood extent prediction. Objective 2 identified the nutrient water quality signatures in Hurricane Florence (2018) floodwaters and characterized the potential drivers of nutrient pollution as a function of different environmental variables using geostatistical Bayesian modeling techniques. Results showed that rainfall and pollution point sources were important variables in explaining nutrient responses, with concentrated animal feeding operations (CAFOs) and wastewater treatment plant (WWTP) variables being important factors influencing floodwater nutrient concentrations. This study suggests that better management of CAFOs and WWTPs for extreme storm events may be necessary to reduce nutrient pollution in floodwaters. Study 3 quantified how hurricanes affect dissolved oxygen trends within different estuaries along the U.S. Atlantic coast using data science tools. Our findings revealed no change in dissolved oxygen concentrations after hurricane impact within the tidal estuaries in our study region. We believe that the short residence time within the estuaries in this study likely induced freshwater flushing and dilution of nutrients and organic matter entering the coastal basins; this effect may have feasibly muted the DO response.These three studies above explore hurricane impacts on surface water quantity and quality, as well as the potential drivers of surface water dynamics. These studies reveal how extreme events sometimes impact our rivers and estuaries, but not always, and the variables that potentially drive the observed impacts.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2023

Mode of access: World Wide Web

ISBN: 9798377685050Subjects--Topical Terms:

556913
Water quality.
Index Terms--Genre/Form:

542853
Electronic books.

Understanding Hurricane-Induced Water Quantity and Quality Dynamics Using Machine Learning and Environmental Data Analytics Approaches.
LDR:04662nmm a2200373K 4500 001 2361068
005 20231024102923.5
006 m o d
007 cr mn ---uuuuu
008 241011s2023 xx obm 000 0 eng d
020 $a 9798377685050
035 $a (MiAaPQ)AAI30364015
035 $a (MiAaPQ)NCState_Univ18402040245
035 $a AAI30364015
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a Fidan, Emine Nur. $3 3701719
245 1 0 $a Understanding Hurricane-Induced Water Quantity and Quality Dynamics Using Machine Learning and Environmental Data Analytics Approaches.
264 0 $c 2023
300 $a 1 online resource (157 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: 84-10, Section: A.
500 $a Advisor: Doll, Barbara;Emanuel, Ryan;Harris, Angela;Sagues, Natalie Nelson.
502 $a Thesis (Ph.D.)--North Carolina State University, 2023.
504 $a Includes bibliographical references
520 $a Hurricanes are disruptive and alter water quantity and quality dynamics in surface waters. Process based models are commonly used to assess flood extent and water quality changes. However, these models typically require sophisticated boundary conditions, an understanding of complex drainage systems, abundance of validation data, expert domain knowledge, and high computational power. This dissertation focuses on advancing the use of machine learning, statistical modeling, and data analytics to evaluate flood dynamics and water quality trends induced by extreme events.The specific objectives of this research were to (1) evaluate the application of the Random Forest, a machine learning algorithm, for daily flood extent modeling in lowlying and flat terrains using rainfall and physiographic characteristics as predictors, (2) explain floodwater nutrient concentrations as a function of watershed characteristics using Bayesian statistical learning methods, and (3) quantify hurricane impacts on dissolved oxygen trends within different estuaries using data analytics.Objective 1 was accomplished by processing satellite imagery and delineating landscape characteristics to produce a timeline of pluvial flooding from Hurricanes Matthew (2016) and Florence (2018). A Random Forest model was trained within the North Carolina (NC) Coastal Plain and results indicated that the model predicts pluvial flooding well, with overall accuracy of 0.96 and F1 score of 0.83. An evaluation of model variables indicated that physically-based predictor variables, distance to the nearest stream, distance to the nearest road, and height above nearest drainage, were the most important in flood extent prediction. Objective 2 identified the nutrient water quality signatures in Hurricane Florence (2018) floodwaters and characterized the potential drivers of nutrient pollution as a function of different environmental variables using geostatistical Bayesian modeling techniques. Results showed that rainfall and pollution point sources were important variables in explaining nutrient responses, with concentrated animal feeding operations (CAFOs) and wastewater treatment plant (WWTP) variables being important factors influencing floodwater nutrient concentrations. This study suggests that better management of CAFOs and WWTPs for extreme storm events may be necessary to reduce nutrient pollution in floodwaters. Study 3 quantified how hurricanes affect dissolved oxygen trends within different estuaries along the U.S. Atlantic coast using data science tools. Our findings revealed no change in dissolved oxygen concentrations after hurricane impact within the tidal estuaries in our study region. We believe that the short residence time within the estuaries in this study likely induced freshwater flushing and dilution of nutrients and organic matter entering the coastal basins; this effect may have feasibly muted the DO response.These three studies above explore hurricane impacts on surface water quantity and quality, as well as the potential drivers of surface water dynamics. These studies reveal how extreme events sometimes impact our rivers and estuaries, but not always, and the variables that potentially drive the observed impacts.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2023
538 $a Mode of access: World Wide Web
650 4 $a Water quality. $3 556913
650 4 $a Hurricanes. $3 551051
650 4 $a Floods. $3 549458
650 4 $a Stormwater management. $3 3685325
650 4 $a Watersheds. $3 544815
650 4 $a Civil engineering. $3 860360
650 4 $a Engineering. $3 586835
650 4 $a Meteorology. $3 542822
650 4 $a Water resources management. $3 794747
655 7 $a Electronic books. $2 lcsh $3 542853
690 $a 0543
690 $a 0537
690 $a 0454
690 $a 0557
690 $a 0595
710 2 $a ProQuest Information and Learning Co. $3 783688
710 2 $a North Carolina State University. $3 1018772
773 0 $t Dissertations Abstracts International $g 84-10A.
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=30364015 $z click for full text (PQDT)