When examining historical disasters, we find they can be instrumental in shaping our preparedness strategies and increasing awareness of potential threats. Investment firms often advise against relying on historical performance to predict future outcomes, and this logic can apply to low-probability events like natural disasters and pandemics, where data is scarce. Nonetheless, modern policymakers often heed the wisdom of Shakespeare: "What’s past is prologue." With a growing database on rare but impactful events and an increase in media scrutiny, the accuracy of predictive models is improving, enhancing their practical application.

Take earthquakes, for instance. Researchers like Tom Jordan, a geologist at the University of Southern California and director of the Southern California Earthquake Center, analyze the frequency of past seismic activity to anticipate future quakes. A significant earthquake, defined as one with a magnitude over 7.0, is relatively rare, but historical records combined with geological data and probability models can clarify the risks involved.

Jordan explains, “If events are random over time, you can determine their frequency.” Historical evidence suggests that Southern California experiences major earthquakes approximately every 100 to 150 years. By employing a time-independent Poisson probability distribution model, we can estimate the likelihood of such events.

Once equipped with mathematical models, researchers seek concrete data regarding factors like bedrock composition, soil stability, and the movement of tectonic plates along fault lines, notably the San Andreas Fault, which shifts at a rate of 33 to 37 millimeters annually. This comprehensive approach yields earthquake hazard curves, illustrating the probability of varying magnitudes of quakes over time in specific regions. In 2007, the Southern California Earthquake Center estimated a 99.7% chance of a magnitude 6.7 earthquake or greater within the next 30 years for California, with the highest risk identified along the Southern California segment of the San Andreas Fault.

Utilizing past data is equally crucial for creating effective earthquake preparedness strategies. John Ebel, a geology professor at Boston College and director of the Weston Observatory, contributed to the development of Hazus, a widely used model that assesses the ground shaking resulting from earthquakes of varying magnitudes based on specific epicenters. This model considers factors such as fault characteristics and soil types, predicting how different areas will respond to seismic activity.

Hazus is particularly valuable for policymakers as it can estimate the financial impact of earthquakes using "fragility curves," which analyze how various building types withstand shaking. For instance, wooden structures tend to be more flexible than masonry ones, and shorter buildings generally perform better than taller ones. By integrating data on local infrastructure and historical earthquake costs, Hazus can provide comprehensive estimates of potential damages, including economic losses and recovery timelines.

Moreover, Hazus can also project injuries and fatalities. The model can be adapted for various disasters by substituting inputs related to ground shaking with data relevant to other hazards, such as wind speed or flood zones.

Past disasters inform not only predictions of future events but also enhance public awareness of risks. For example, the threat of an asteroid impact became more recognized after the Tunguska event in 1908, which involved a massive explosion in the atmosphere, underscoring the potential dangers. Don Yeomans, who leads NASA’s near-Earth-object program, notes that interest in this field has surged, especially after significant events like the collision of comet Shoemaker-Levy 9 with Jupiter.

The cultural impact of films such as Armageddon and Deep Impact, along with increased media coverage of near-Earth objects, has led to a dramatic rise in funding for NASA's asteroid detection initiatives. The United Nations has also established a committee to coordinate global efforts in identifying and addressing potential threats from asteroids.

This phenomenon exemplifies what cognitive psychologists term the Availability Heuristic, where people prioritize risks that are easily recalled. For instance, people often mistakenly believe that the letter "r" appears more frequently as the first letter in words rather than the third because it is more readily accessible in their memory. Emotions, particularly fear, can intensify this effect, making past traumatic experiences more salient and influential on public perception.

The awareness of the risk posed by asteroids has grown significantly in recent years. Yeomans asserts that once the public recognizes a threat, they become proactive in urging policymakers to take action. Events like the Chelyabinsk meteor explosion have only heightened this awareness.

The increasing frequency of natural disasters, partly due to climate change and rising population densities, has made policymakers more attentive to disaster preparedness. Ebel observes that advancements in technology allow for more vivid and immediate coverage of disasters, fueling public interest and engagement.

Howard Kunreuther, a risk management expert, notes that while disasters may lead to increased attention, it often takes a significant event for people to prioritize these issues. He found that corporate leaders now acknowledge the importance of disaster readiness following recent events.

Ultimately, regardless of how meticulously policymakers plan and predict, human psychology complicates decision-making, especially under uncertain conditions. Jordan candidly remarks that when faced with ambiguous numbers, individuals often resort to intuition rather than analytical reasoning. Many people struggle with basic probability, and despite improvements in disaster preparedness—evident in initiatives like New York's $20 billion flood protection plan—our understanding is still evolving, largely thanks to the lessons drawn from past disasters.

David Ropeik is a risk perception and communication consultant who teaches environmental management at Harvard Extension School. He authored *How Risky Is It Really? Why Our Fears Don’t Always Match the Facts, and his blog, Risk: Reason and Reality, can be found at BigThink.com.*

Originally published at Nautilus on August 8, 2013.