No uncertain terms

Three words—forecast, projection, and prediction—in scientific terms are similar, but they have distinct implications in specific contexts. I find it concerning when read research papers, announcements by government agencies modelers, or popular science coverage that use the incorrectly. Such misuse may cause misinterpretations. And worst-case scenario, correct meanings be dismissed incorrect enforced. It is responsibility of scientists to correctly appropriately interpret communicate them with caution.“Forecasts” COVID-19 pandemic been offered agencies, institutions, teams around world. As this issue Physics Today goes press, Forecasting Mathematical Modeling webpage, via US Centers for Disease Control Prevention (CDC), each week provides four-week “forecasts” hospitalizations deaths.11. For forecasts new total deaths US, see https://www.cdc.gov/coronavirus/2019-ncov/science/forecasting/mathematical-modeling.html. (Case not posted since December 2021.) The results presented include both ensemble independent ones numbers based on.Obviously, huge uncertainties associated those forecasts. CDC’s hospitalization death forecast pages state “models make various assumptions about levels social distancing other interventions, which reflect recent changes behavior.” Thus possible scenarios necessarily probable because unpredictable factors such as national policies human behaviors.22. J. Cepelewicz, “The hard lessons modeling coronavirus pandemic,” Quanta Magazine, 28 January 2021. CDC case page states, “While among most reliable performance over time, even reliably predicted rapid trends reported cases, hospitalizations, deaths. They should relied upon making decisions possibility timing trends.”To calculate some future events conditions, usually a result study analysis available pertinent data. A forecasted event occurrence. term frequently used reference weather—which on basis correlated meteorological observations. If weather shows going snow tomorrow, means rather condition next day. whether will tomorrow does least depend how humans behave politicians debate.The way many groups “forecast” confusion public, policymakers, decision makers, leaving wrong impression comparable forecast. In fact, “projection” more appropriate projection offers only conditional response depends validity assumed scenarios.To help explain distinction between forecast, prediction, consider three sample sentences:1. Bureau Meteorology tomorrow. (The occurrence.)2. team’s world population rise 11 billion 2100. tells possibilities.)3. Astronomers can accurate predictions an eclipse occur. prediction inference certainty.)I seen modelers. good example work from October 2021 reporting projections contacts during end-of-year holiday Mexico City would potentially affect outcomes, including infections, deaths, hospitalizations.33. F. Alarid-Escudero et al., MDM Policy Pract. 6, 238146832110492 (2021), doi: https://doi.org/10.1177/23814683211049249. Some argue “prediction” modeling. But implies certainty convey possibilities implied projection.Confusion two no rare working climate science, revealed survey.44. D. Bray, H. von Storch, Sci. Commun. 30, 534 (2009). https://doi.org/10.1177/1075547009333698 Unfortunately, paper relationship coastal carbon sequestration change, authors “go beyond soil C stock estimates reveal global tidal wetland accumulation predict under relative sea level rise, temperature precipitation.”55. Wang Natl. Rev. 8, nwaa296 (2021). https://doi.org/10.1093/nsr/nwaa296 there large regarding underlying scenarios, unforeseen socioeconomic technological conditions uncertain growth coming decades. So “project” here talking scenarios.I strongly recommend careful particularly regard change COVID-19. policymakers gradually get projections.ReferencesSection:ChooseTop pageReferences <<1. Google Scholar2. Scholar3. ScholarCrossref4. https://doi.org/10.1177/1075547009333698, ScholarCrossref5. https://doi.org/10.1093/nsr/nwaa296, ScholarCrossref© 2023 American Institute Physics.