信息时代，互联网和社交媒体的出现大大降低了信息传播的成本。尽管这有助于扩散包括灾害预警信息和救灾信息在内的一些有益于灾害应对的灾害信息，但同样也令一些给社会带来负面影响的灾害信息更容易产生和传播。因此，了解灾害信息的传播规律十分必要，有利于加强对灾害信息传播的管理，以更好地从信息层面应对灾害。然而，前期主要聚焦在仅涉及一种灾害的单一灾害场景中的灾害信息传播规律，同时涉及不同灾害的多灾场景中的灾害信息传播情形却并未得到足够的关注，多灾场景中的灾害信息传播规律尚不明朗。因此，需要增加对多灾场景中灾害信息的关注，理解并厘清多灾场景中灾害信息的传播规律和传播机制，从而制定适用于多灾场景的灾害信息治理策略，以减少灾害信息给社会带来的负面影响。本文聚焦于一种典型的多灾场景——自然灾害—流行病双灾场景，开展了其中的灾害信息传播规律研究，旨在探索多灾场景中的灾害信息传播规律是否出现变化，以更好地治理多灾场景中的灾害信息。本文通过理论建模的方法厘清了该场景中公众对灾害的关注随时间的演化规律，以及公众对自然灾害的负面情绪随时间的演化规律，并通过真实数据对理论模型进行了验证。发现相比于单一灾害场景，这两者均发生了变化。基于该现象，本文发现了多灾信息耦合效应，即不同灾害在信息层面会相互作用，从而影响灾害信息的传播规律。本文还验证了多灾信息耦合效应呈现形式的多样性，为多灾场景中的灾害信息治理提供了科学依据。基于“多灾信息耦合效应”，本文对自然灾害—流行病双灾场景中的两种特殊灾害信息——双灾复合灾害谣言及其辟谣信息的传播进行了探究，并厘清了两者的传播规律。提出了能够描述两种灾害信息传播规律的信息传播模型，并通过真实数据对理论模型进行了验证，发现两者均存在传播放大现象。结果揭示了“多灾信息耦合效应”既能促进正面信息传播，也能促进负面信息传播。该结果为多灾场景中的灾害谣言治理提供了指导和帮助。针对双灾复合灾害谣言的传播放大效应出现原因进行了进一步研究，探索并明确了该场景下社交媒体意见领袖的信息发布策略。理论建模和实证分析的结果表明，“发布双灾信息”是社交媒体意见领袖在该场景下获得公众影响力的核心策略，为多灾场景中的社交媒体治理提供了支持。

In the information age, the emergence of the Internet and social media has significantly reduced the cost of information propagation. While this facilitates the spread of disaster-related information, including disaster warning information and disaster relief information beneficial for disaster response, it also leads the generation of disaster-related information that can have adverse effects on society and make it easier propagate. Therefore, understanding the propagation patterns of disaster-related information is necessary and it facilitates enhancing the management of its propagation, thereby enabling a more effective response to disasters from the information level. However, previous studies have primarily focused on the propagation patterns of disaster-related information within the single-hazard scenario that only involve one disaster, and the propagation patterns within multi-hazard scenarios involving various disasters has not received sufficient attention. Consequently, the propagation patterns of disaster-related information within multi-hazard scenarios remain unclear. Hence, there is a need to emphasize the study of disaster-related information within multi-hazard scenarios, comprehend and elucidate the propagation patterns and mechanisms therein, and devise governance strategies that are appropriate for multi-disaster scenarios to mitigate the adverse impacts of disaster-related information on society.This paper focuses on a typical multi-hazard scenario—namely, the double-hazard scenario of natural disasters and pandemics. It investigates the propagation patterns of disaster-related information within this context, aiming to explore whether the patterns of disaster-related information propagation undergo changes in multi-hazard scenarios. The research goal is to enhance the management of disaster-related information in such multi-hazard scenarios. This paper employs theoretical modeling to elucidate the temporal evolution of public attention to disasters and the evolution of negative public emotions towards natural disasters within this scenario. The theoretical model is validated by real data, revealing that both aspects undergo changes compared to single-hazard scenarios. Based on this observation, this paper identifies the multi-hazard information coupling effect, indicating that different disasters interact at the informational level, thereby influencing the propagation patterns of disaster-related information. Also, this paper verifies the diversity in the manifestation of the multi-hazard information coupling effect, providing a scientific basis for disaster-related information governance within multi-hazard scenarios.Based on the "multi-hazard information coupling effect", this paper investigates the propagation of two types of specific disaster-related information within the double-hazard scenario—double-hazard compound rumors and their debunking information—and clarifies their propagation patterns. It proposes information propagation models capable of describing the propagation patterns of both types of disaster-related information and validates the theoretical models by using real data, revealing that both types of information exhibit propagation amplification phenomena. The results reveal that the "multi-hazard information coupling effect" can both facilitate the propagation of positive information and exacerbate the propagation of negative information, demonstrating its ambivalence. These findings provide guidance and assistance for managing disaster-related rumors within multi-hazard scenarios. Further research is conducted to investigate the causes of the propagation amplification effect of double-hazard compound rumors and to explore and clarify the information propagation strategies of social media opinion leaders within this scenario. The results of theoretical modeling and empirical analysis indicate that the behavior of "publishing double-hazard information" is a core strategy for social media opinion leaders to gain influence over the public within this scenario, thereby providing support for the governance of social media within multi-disaster scenarios.