Temporal characteristics of disaster data: how to avoid misconceptions, misinterpretation, or misuse

Temporal trends in disaster data can lead to contradictory conclusions, with some studies suggesting increasing impacts while others report decreases or stabilization. This study explores how these misconceptions can arise in disaster trend analyses and presents key factors to consider in preventing misinterpretations or misuse that could impact policy, insurance, and public perceptions towards scientific rigor. Factors that influence trend claims are identified, which include the way of dealing with extreme values, data reliability, normalization for population and GDP changes, trend modeling choices, and expert judgment. Next, we present a practice-oriented summary consisting of a 10-point checklist. Its use might safeguard users and designers of disaster data against overly optimistic, overly pessimistic, or flawed claims on trend patterns. It has a simple yes–no structure and serves as list of “warning signs.” In conclusion, we advocate for the adoption of “best modeling” practices to enhance the credibility of disaster data use, as it directly informs decision-making in policy and resource allocation.