Deciphering public attention to geoengineering and climate issues using machine learning and dynamic analysis

TL;DR

This study investigates how public attention to geoengineering evolves in response to climate news using a data driven approach that combines Google Trends (2018–2022) with a BBC NYTimes article corpus (n=30,773). The authors apply BERTopic for topic modeling, BERT sentiment analysis, and time series models to quantify links between media coverage and geoengineering interest, testing three hypotheses. Key findings show that energy and disaster related climate news, particularly with positive sentiment, predict higher geoengineering attention over time, while religion topics show negative associations and topic overlap varies across geoengineering techniques. The results illuminate how media framing could influence public engagement with solar radiation management and greenhouse gas removal, offering guidance for science communication and policy deliberation while noting limitations of proxy measures and dataset representativeness. Overall, the work demonstrates a data driven pathway to map public perception dynamics for emerging climate technologies and informs targeted public engagement strategies.

Abstract

As the conversation around using geoengineering to combat climate change intensifies, it is imperative to engage the public and deeply understand their perspectives on geoengineering research, development, and potential deployment. Through a comprehensive data-driven investigation, this paper explores the types of news that captivate public interest in geoengineering. We delved into 30,773 English-language news articles from the BBC and the New York Times, combined with Google Trends data spanning 2018 to 2022, to explore how public interest in geoengineering fluctuates in response to news coverage of broader climate issues. Using BERT-based topic modeling, sentiment analysis, and time-series regression models, we found that positive sentiment in energy-related news serves as a good predictor of heightened public interest in geoengineering, a trend that persists over time. Our findings suggest that public engagement with geoengineering and climate action is not uniform, with some topics being more potent in shaping interest over time, such as climate news related to energy, disasters, and politics. Understanding these patterns is crucial for scientists, policymakers, and educators aiming to craft effective strategies for engaging with the public and fostering dialogue around emerging climate technologies.

Figures (6)

• Figure 1: The BBC and the NYTimes news temporal trends for 2018–2022. This figure shows the temporal trends of news articles produced per week containing terms associated with geoengineering and climate change, as per Table 1. The total number of relevant news articles for our analysis is 30,773.
• Figure 2: Relationship between attention to environmental issues and geoengineering for both the public and media. (a) The correlation between public interest in various climate topics (represented on the y-axis) and public interest in geoengineering is illustrated using data sourced from Google Trends. The x-axis values represent weighted Pearson correlation coefficients (R$^{2}$), with the dots and lines illustrating the mean and 95% confidence level. (b) Correlations between media coverage on different environmental issues in the BBC and NYTimes (y-axis) and on geoengineering are calculated as the Word2Vec distance between words. We report the Eucledian distance at 95% confidence levels.
• Figure 3: Relationship between sentiment scores of environment and climate-related news articles and public attention towards geoengineering. The x-axis shows the Google Trends Index scores, representing public interests in geoengineering, and the y-axis represents the sentiment scores across the five topic categories from the BBC and the NYTimes corpus. The trend line shows the Pearson correlations, with shaded regions showing uncertainties at 95% levels. The correlation coefficient value for "Nature" is 0.10, -0.08 for "Politics", -0.18 for "Disaster", -0.24 for "Religion", and 0.18 for "Energy".
• Figure 4: Sentiment scores of climate-related topics in the news articles and public attention towards the various geoengineering approaches. The x-axis shows the Google Trend Index, representing interests in types of geoengineering, and the y-axis represents sentiment scores across the 5 categories of climate-related topics in the news data. The trend lines show the Pearson coefficient, with shaded regions showing uncertainties at 95% levels.
• Figure 5: Estimates of public attention to geoengineering across the five topic categories: disaster, energy, nature, politics, and religion. Model 1 represents the associations between public interests across the topics (disaster and energy at 99% CI and politics at 95% CI). Model 2 shows the relationship between embedded sentiments across the topics (energy at 99% CI and disaster at 95% CI). Error bars are shown at 95% CI, model summary statistics presented in SI Table 2.