How natural disasters and environmental fears shape American climate attitudes across political orientation

Climate change poses a significant threat, yet its global scale and gradual impact make subjective assessment difficult. This complicates public communication and motivates mitigation efforts. Political orientation is a key predictor of climate concern in the US, with Republicans and conservatives showing more skepticism. This polarization hinders mitigation efforts, demanding interventions to improve support across the political spectrum. Existing models suggest threat exposure leads to ideological shifts toward conservatism or intensifies existing preferences, potentially hindering interventions. However, these models often neglect the political and affective context of threats, overrepresenting threats that appeal to conservatism. This study examines the influence of threat perception and fear on climate concern across the political spectrum, offering both practical and theoretical insights.

Extensive research demonstrates that US political conservatives perceive lower climate risk, express more skepticism, and distrust climate science compared to liberals and moderates. This polarization is increasingly mirrored internationally. One explanation is the difference in threat perception: conservatives focus on local, intentional threats, while liberals are concerned with global threats from inaction. These differences may reflect underlying values—conservatism prioritizing stability and liberalism favoring change. Ideological differences in threat prioritization may explain the higher climate concern among liberals, as climate change is a global threat from indirect anthropogenic processes. Fear related to climate change influences risk perceptions and policy support, with worry linked to greater risk perceptions, policy support, and adaptive behavior. However, climate change's abstract nature makes galvanizing concern difficult without highlighting tangible climate experiences.

This study uses data from the SPEER survey, a comprehensive online survey administered by Qualtrics to 2188 US adults. Quota-based recruitment aimed for a representative sample. The survey measured climate concern through four linear dependent variables: belief in anthropogenic climate change, perception of risk, personal importance, and worry. Independent variables included self-reported fear across six domains (natural disasters, environmental disasters, strangers, crime, human-made events, and personal tragedy), and political orientation (liberal, moderate, conservative). Regression analyses examined the relationships between fear, political orientation, and climate concern, including interaction terms. Post-hoc tests examined marginal effects at various fear levels to assess moderation.

Regression analyses showed that compared to liberals, moderates and conservatives had lower levels of climate concern (supporting H1), consistent with prior literature. Fear of environmental disasters was positively associated with all four climate concern domains, more so than other fear variables. Fear of natural disasters positively correlated with climate risk, importance, and worry but not belief alone (largely supporting H2). Interestingly, fear of strangers and crime negatively predicted climate risk and importance. Interaction analyses showed that fear of natural disasters increased climate concern among conservatives relative to liberals (supporting H3). Graphical depictions and post-hoc tests revealed that at higher levels of natural disaster fear, the difference in climate concern across political orientations diminishes. For environmental disasters, the interaction effect was less consistent. At higher levels of environmental disaster fear, conservatives showed more pronounced increases in climate belief and risk than liberals, though the impact on worry and importance was less pronounced. Marginal effects analyses further illustrated this convergence across political orientations at higher fear levels for both natural and environmental disasters.

The findings strongly support the hypotheses that conservatives show lower climate concern and that fear of natural and anthropogenic environmental disasters increases climate concern across all political orientations. The more pronounced effect of natural disaster fear on conservatives suggests that focusing on tangible, local threats may be more effective in increasing their climate concern. The mixed support for Hypothesis 3 regarding environmental disasters might be explained by the role of anger, which may increase threat perceptions and motivate responses independently of fear. This suggests that threats evoking strong negative emotions may be effective at motivating attitudinal shifts. Theoretically, these findings challenge existing threat perception and political orientation paradigms. They support the ideology-affordance model, suggesting that threatened individuals align with the ideology perceived as capable of resolution. Practically, the results guide climate communicators and policymakers to increase climate concern among conservatives by connecting local threats to global warming and offering accessible behavioral responses.

This study provides evidence that fear of natural and anthropogenic environmental disasters significantly impacts climate concern, especially among conservatives. High levels of fear lead to convergence in climate concern across political orientations, highlighting potential for effective messaging strategies. Future research should explore the role of other emotions like anger and design interventions that connect local threats to global warming while offering accessible climate-friendly behaviors.

The study's online sample may skew toward liberals and undercount conservatives. The lack of behavioral measures limits the assessment of the relationship between climate concern and actual behavior. While interaction effects were statistically significant, they didn't greatly increase model fit, suggesting that direct effects of independent variables are more impactful.