Climate-denying rumor propagation in a coupled socio-climate model: Impact on average global temperature

equal products of the parameters (d * k). The constant product value (d * k), helps in analyzing how the rejector population and temperature anomaly vary, with changes in the parameter values. The size of the rejector population determines the temperature anomaly, for instance, a higher rejector population leads to a lower temperature anomaly (Fig 5(a) and 5(b)). Higher variations in the size of the rejector population (and temperature anomaly) are observed when the average number of messages transmitted is low. This implies that both rejector population and temperature anomaly are more sensitive to the variations in the average number of mes- sages transmitted per group than the network degree. In other words, the messaging rate in groups is more important than the number of individual connections in terms of determining the size of the rejector population (and thus in determining how much climate change is miti- gated). As long as the frequency of messages about climate change in groups remains high, reduction in individual connections does not influence the temperature anomaly. The interactions between rumor propagation parameters (group believing and rejecting probabilities) and the forcing due to doubling of CO2 (Wm-2) shows that temperature anomaly is more sensitive to changes in forcing than the changes in the believing or rejecting probabil- ity (S17 Fig). The forcing due to doubling of CO2 (F2x), causes an increase in the temperature anomaly even at lower believing probability (higher rejecting probability). Being a an environ-

mental variable, forcing has its own upper hand in influencing climate change compared to human actions but human actions are the reason for these changes is also an underlying fact. The tornado diagram is a sensitivity analysis tool that helps examine the model parameters’ relative priority. The sensitivity of the temperature anomaly (in 2200) demonstrates the rela- tive sensitivity of each parameter. S18 Fig is the tornado plot showing the temperature anomaly sensitivity when the model parameters vary individually at 10%. The analysis confirms the results from S18 Fig that temperature change is most sensitive to the emission limiting con- stant C (S18 Fig). The temperature is more sensitive to the group parameters than the individ- ual parameters. The hypothesis that temperature change is more sensitive to forgetting probability in rejectors is also true in the case of 10% change in the parameters even though there are no visible changes because of the extremely low baseline value considered. The tor- nado plot confirms the results of the previous sensitivity analysis and helps to understand the relative influence for each parameter. 4 Discussion A rumor propagation model was linked with an Earth System model to understand how the spread of climate rumors responds to climate change, and how resulting behavioral change can feedback on climate. Differing from the existing literature, this model focuses on how rumors about climate change can influence the climate system itself, using methods from com-

partmental modeling and Earth System Modeling [59–65]. The study has shown that the miti- gation strategies by rumor rejectors to limit emissions are vital in reducing emission levels and temperature anomalies. Acknowledging that climate change is a major issue is not enough but working efficiently towards mitigation and adopting efficient mitigation strategies are vital in controlling the effects of climate change. The parameter planes and tornado plots suggest that temperature anomaly is most sensitive to these mitigation strategies by the rejectors. The miti- gation plan is the crucial link that connects human behavior and climate change in our model. Our results support that human mitigation strategies help in keeping climate change under control by avoiding undesirable and uninhabitable situations in the future, similar to the sug- gestions from various studies [29–32, 66]. The extent of climate change we will experience in the future depends on the strategies we adopt today and how soon we start acting according to PLOS ONE Climate-denying rumor propagation in a coupled socio-climate model: Impact on average global temperature PLOS ONE | https://doi.org/10.1371/journal.pone.0317338 January 16, 2025 12 / 18