Water narratives in local newspapers within the United States

Frontiers in Environmental Science

Environmental Science

Jonathan M. Gilligan¹  Thushara Gunda²  Matthew D. Sweitzer² 

[1] Department of Earth and Environmental Sciences, Vanderbilt University, Nashville, TN, United States;Sandia National Laboratories, Albuquerque, NM, United States;

关键词: water;    newspapers;    big data;    natural language processing;    structural topic modeling;    named entity recognition;    sentiment analyis;   

DOI  :  10.3389/fenvs.2023.1038904

 received in 2022-09-07, accepted in 2023-02-17,  发布年份 2023

来源: Frontiers

【 摘 要 】

Sustainable use of water resources continues to be a challenge across the globe. This is in part due to the complex set of physical and social behaviors that interact to influence water management from local to global scales. Analyses of water resources have been conducted using a variety of techniques, including qualitative evaluations of media narratives. This study aims to augment these methods by leveraging computational and quantitative techniques from the social sciences focused on text analyses. Specifically, we use natural language processing methods to investigate a large corpus (approx. 1.8M) of newspaper articles spanning approximately 35 years (1982–2017) for insights into human-nature interactions with water. Focusing on local and regional United States publications, our analysis demonstrates important dynamics in water-related dialogue about drinking water and pollution to other critical infrastructures, such as energy, across different parts of the country. Our assessment, which looks at water as a system, also highlights key actors and sentiments surrounding water. Extending these analytical methods could help us further improve our understanding of the complex roles of water in current society that should be considered in emerging activities to mitigate and respond to resource conflicts and climate change.

【 授权许可】

Unknown   
Copyright © 2023 Sweitzer, Gunda and Gilligan.