Natural hazards are associated with tens of thousands of deaths per year, cause widespread damage, and inhibit development. Over the past century, the global death rate from natural hazard-related disasters has decreased by an order of magnitude, saving tens of millions of lives (EM-DAT, 2024). Sustaining and accelerating this decline in death rate requires, among other factors, continued advances in our understanding and management of natural hazards. Natural hazards were defined as “natural processes or phenomena that may cause loss of life, injury or other health impacts, property damage, loss of livelihoods and services, social and economic disruption, or environmental damage”1. A 2016 report redefines hazard as "A process, phenomenon or human activity that may cause loss of life, injury or other health impacts, property damage, social and economic disruption or environmental degradation.", further noting that "Natural hazards are predominantly associated with natural processes and phenomena", although it does not provide a full redefinition for natural hazards2. In today’s world, humans have extensively modified the atmosphere and Earth surface3,4, altering the characteristics, frequency, and magnitude of many hazards and challenging the concept of a “natural process or phenomenon”. Here, I explore whether the concept of “natural hazards” remains relevant today and if existing definitions are fit for purpose.

The role of human activity in disasters has long been recognized5. However, distinguishing ‘natural’ from ‘anthropogenically modified’ processes is increasingly challenging in today’s world, and the separation of disasters into a “natural” trigger and a “human” context often oversimplifies complex realities. Examples of anthropogenically modified hazards include newly formed unstable glacial lakes6, intensified hurricanes or typhoons (Emanuel, 2013), frequent landsliding around newly built roads or active mines7, earthquakes triggered by subsurface fluid injection8, and even volcanic eruption rates modulated by deglaciation9. This growing influence of human activities is clearly observed in various hydrometeorological and geological hazards, as explored in the following paragraphs.

The fingerprints of human activities are particularly apparent for many hydrometeorological hazards, including storms and flooding, making it increasingly challenging to classify these hazards as purely natural. Anthropogenic climate change has intensified hurricanes and typhoons; warmer sea surface temperatures due to greenhouse gas emissions have increased storm energy, leading to more destructive events10,11. Data reveals a rise in Category 4 and 5 hurricanes since the 1970s12. Additionally, flood risk has been exacerbated by both climatic and other human factors. One example is the city of Jakarta, where multiple additive factors contribute to severe flooding: sea-level rise driven by global warming, increased precipitation due to altered weather patterns, and subsidence caused by excessive groundwater extraction and infrastructure development13. This subsidence, combined with rising sea levels, has led to frequent and severe flooding, making Jakarta one of the most vulnerable cities in the world to climate change14. These compound effects underscore how human actions not only amplify individual hazards but can transform them into complex multihazard networks that challenge traditional approaches to hazard management.

Anthropogenic activities can also exacerbate landslide hazard by modifying rainfall patterns and transforming land surface cover or topography. Climate-driven changes in rainfall duration and intensity can increase the probability of landslides, particularly in regions where vegetation has been removed. In the Himalaya, studies have recorded a substantial rise in landslide frequency following vegetation removal15. Other changes in land surface cover, particularly through infrastructure developments such as road building, quarrying, dam construction, and mining, can further affect slope stability7. For example, the 2023 Glacial Lake Outburst Flood (GLOF) at South Lhonak Lake in Sikkim, India, was triggered by a landslide failure into a proglacial lake, itself formed in response to climate warming-driven glacier recession16. This case exemplifies how anthropogenic environmental changes can contribute to disasters, even where their role in triggering the initial hazard is less evident.

Viewing hazards through a multihazard lens is essential in a world where human activity increasingly impacts natural processes. As seen in the earlier example of South Lhonak Lake, the hazard cascade was preconditioned by glacial retreat, triggered by a landslide, and exacerbated by the failure of a downstream dam, illustrating how hazards may be amplified in multiple ways by human activity. Similarly, earthquakes in deforested areas may trigger additional landslides, dam rivers, and increase flood risk, while hurricanes exacerbate coastal flooding and landslides across a range of urban and rural environments. The Sendai Framework for Disaster Risk Reduction (UN, 2015) emphasizes multihazard approaches, integrating risk assessments, governance, and early warning systems (EWS) to address interconnected risks. Supported by initiatives such as Early Warnings for All, EWS enhance preparedness and response by integrating hazard warnings across sectors, addressing the complex interactions in a human-modified environment. Adopting a multihazard perspective is essential for understanding these interconnected risks and developing mitigation strategies that consider both natural processes and human influences (e.g., Figure 1).

Fig. 1
Fig. 1
Full size image

Example of potential anthropogenic influences and some of the hazards these may apply to.

Given the increasing role of human activity in shaping “natural hazards”, it is appropriate to question whether this term remains valid or useful. Alternative terms, such as “environmental hazards" or simply “hazards" with further classifications (e.g., technological, geological, or meteorological), have been proposed to capture better the broad spectrum of processes (e.g. UNDRR hazard profiles, 2016 UNDRR report). These alternatives offer the advantage of encompassing terminology for events with either natural and anthropogenic origins, recognizing the significant role that human activities play in hazard generation and modification. However, these terms also have limitations. “Environmental hazards" can be overly broad, potentially diluting the focus on the natural processes central to many hazards. The term “hazards," while inclusive, might obscure the specific natural elements that differentiate geological or meteorological hazards from purely technological or human ones, yet still not clearly recognize the crossing between these definition boundaries. By noting that "Natural hazards are predominantly associated with natural processes and phenomena", the 2016 UNDRR hazard definition raises the possibility that anthropogenic activity may influence natural hazards, but does not explicitly address this.

The persistence of traditional terminology presents both challenges and opportunities for our conception of hazard and risk. While the disaster risk reduction community has moved away from the term “natural disasters" for its problematic implications of inevitability and reduced agency, “natural hazards" remains embedded in scientific, policy, and public discourse. This terminological entrenchment, demonstrated by the continued use of “natural disasters" despite decades of critique and the evolution of medical terminology during COVID-1917, suggests that redefining established terms can be more effective than replacement. I propose reconceptualising “natural hazards" to explicitly acknowledge interplay between human and environmental processes while maintaining the term’s utility for cross-sectoral communication and policy implementation. This approach aligns with major policy frameworks, including the Sendai Framework for Disaster Risk Reduction and IPCC assessments, which increasingly emphasize the cascading and compound nature of hazards in today’s world. I propose the following working definition: Natural hazards are interconnected geological and/or meteorological processes, often influenced by or interacting with anthropogenic activity, that can cause loss of life, injury or other health impacts, property damage, social and economic disruption, or environmental degradation.. No definition is perfect and none solve problems on their own, but a useful definition can help improve communication and guide solutions. This definition captures how multiple processes interact and amplify in a multihazard framework and acknowledges the possible role of human influences in shaping otherwise natural events. By explicitly addressing these interactions, it facilitates clearer communication across disciplines and stakeholders and removes the need to artificially separate natural and anthropogenic processes, reducing potential confusion. By emphasizing interconnected processes, this definition supports systematic analysis of cascading and compound hazards, enabling more effective risk assessment, management strategies, and early warning systems in a world increasingly–but not exclusively–shaped by human activity.