Steven D. Meyers, Marcus W. Beck, Edward T. Sherwood, Mark E. Luther
{"title":"开发一种遗传算法,用于选择优化旧坦帕湾水动力冲洗的基础设施改造方案","authors":"Steven D. Meyers, Marcus W. Beck, Edward T. Sherwood, Mark E. Luther","doi":"10.1007/s12237-024-01387-y","DOIUrl":null,"url":null,"abstract":"<p>Impermeable infrastructure such as traffic causeways can reduce the natural hydrodynamic flushing of an estuary, resulting in reduced water quality and increased incidence of harmful algal blooms (HABs). A series of cuts through the three causeways spanning Old Tampa Bay, FL, (OTB) are being considered to help restore the natural circulation of the region, but the number of possible location combinations is computationally challenging to fully assess. A prototype genetic algorithm (GA) was developed to identify the optimal configuration of these cuts through one of the bridge sections that maximizes flushing as represented in a numerical ocean circulation model of OTB. Flushing was measured by integrating the trajectories of over 21,000 passive Lagrangian “particles” using the model velocity fields. The rate of loss of particles initialized near Feather Sound (a region subject to frequent HABs) was used to quantify the “fitness” over which the configurations were optimized. The highest-scoring solution produced a 42% increase in net flushing compared to a no-change baseline. Six independently initialized applications of the GA were conducted. All converged to the same solution within no more than 7 generations. The small population size of the prototype allowed testing of the complete solution space, and verification the found solution was optimal. Elitism (preservation of the highest-ranking solution) was required for convergence. The GA also identified configurations that had similar, but slightly slower, flushing rates. These results will help area managers prioritize or rank combinations of causeway modifications to improve overall water quality conditions in Tampa Bay.</p>","PeriodicalId":11921,"journal":{"name":"Estuaries and Coasts","volume":"47 1","pages":""},"PeriodicalIF":2.3000,"publicationDate":"2024-07-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Developing a Genetic Algorithm for Selecting Infrastructure Modifications that Optimize Hydrodynamic Flushing in Old Tampa Bay\",\"authors\":\"Steven D. Meyers, Marcus W. Beck, Edward T. Sherwood, Mark E. Luther\",\"doi\":\"10.1007/s12237-024-01387-y\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p>Impermeable infrastructure such as traffic causeways can reduce the natural hydrodynamic flushing of an estuary, resulting in reduced water quality and increased incidence of harmful algal blooms (HABs). A series of cuts through the three causeways spanning Old Tampa Bay, FL, (OTB) are being considered to help restore the natural circulation of the region, but the number of possible location combinations is computationally challenging to fully assess. A prototype genetic algorithm (GA) was developed to identify the optimal configuration of these cuts through one of the bridge sections that maximizes flushing as represented in a numerical ocean circulation model of OTB. Flushing was measured by integrating the trajectories of over 21,000 passive Lagrangian “particles” using the model velocity fields. The rate of loss of particles initialized near Feather Sound (a region subject to frequent HABs) was used to quantify the “fitness” over which the configurations were optimized. The highest-scoring solution produced a 42% increase in net flushing compared to a no-change baseline. Six independently initialized applications of the GA were conducted. All converged to the same solution within no more than 7 generations. The small population size of the prototype allowed testing of the complete solution space, and verification the found solution was optimal. Elitism (preservation of the highest-ranking solution) was required for convergence. The GA also identified configurations that had similar, but slightly slower, flushing rates. These results will help area managers prioritize or rank combinations of causeway modifications to improve overall water quality conditions in Tampa Bay.</p>\",\"PeriodicalId\":11921,\"journal\":{\"name\":\"Estuaries and Coasts\",\"volume\":\"47 1\",\"pages\":\"\"},\"PeriodicalIF\":2.3000,\"publicationDate\":\"2024-07-20\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Estuaries and Coasts\",\"FirstCategoryId\":\"93\",\"ListUrlMain\":\"https://doi.org/10.1007/s12237-024-01387-y\",\"RegionNum\":3,\"RegionCategory\":\"环境科学与生态学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q3\",\"JCRName\":\"ENVIRONMENTAL SCIENCES\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Estuaries and Coasts","FirstCategoryId":"93","ListUrlMain":"https://doi.org/10.1007/s12237-024-01387-y","RegionNum":3,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"ENVIRONMENTAL SCIENCES","Score":null,"Total":0}
Developing a Genetic Algorithm for Selecting Infrastructure Modifications that Optimize Hydrodynamic Flushing in Old Tampa Bay
Impermeable infrastructure such as traffic causeways can reduce the natural hydrodynamic flushing of an estuary, resulting in reduced water quality and increased incidence of harmful algal blooms (HABs). A series of cuts through the three causeways spanning Old Tampa Bay, FL, (OTB) are being considered to help restore the natural circulation of the region, but the number of possible location combinations is computationally challenging to fully assess. A prototype genetic algorithm (GA) was developed to identify the optimal configuration of these cuts through one of the bridge sections that maximizes flushing as represented in a numerical ocean circulation model of OTB. Flushing was measured by integrating the trajectories of over 21,000 passive Lagrangian “particles” using the model velocity fields. The rate of loss of particles initialized near Feather Sound (a region subject to frequent HABs) was used to quantify the “fitness” over which the configurations were optimized. The highest-scoring solution produced a 42% increase in net flushing compared to a no-change baseline. Six independently initialized applications of the GA were conducted. All converged to the same solution within no more than 7 generations. The small population size of the prototype allowed testing of the complete solution space, and verification the found solution was optimal. Elitism (preservation of the highest-ranking solution) was required for convergence. The GA also identified configurations that had similar, but slightly slower, flushing rates. These results will help area managers prioritize or rank combinations of causeway modifications to improve overall water quality conditions in Tampa Bay.
期刊介绍:
Estuaries and Coasts is the journal of the Coastal and Estuarine Research Federation (CERF). Begun in 1977 as Chesapeake Science, the journal has gradually expanded its scope and circulation. Today, the journal publishes scholarly manuscripts on estuarine and near coastal ecosystems at the interface between the land and the sea where there are tidal fluctuations or sea water is diluted by fresh water. The interface is broadly defined to include estuaries and nearshore coastal waters including lagoons, wetlands, tidal fresh water, shores and beaches, but not the continental shelf. The journal covers research on physical, chemical, geological or biological processes, as well as applications to management of estuaries and coasts. The journal publishes original research findings, reviews and perspectives, techniques, comments, and management applications. Estuaries and Coasts will consider properly carried out studies that present inconclusive findings or document a failed replication of previously published work. Submissions that are primarily descriptive, strongly place-based, or only report on development of models or new methods without detailing their applications fall outside the scope of the journal.