Scientific uncertainties in climate change projections are generally addressed using an ensemble method, in which multiple models are used to generate climate projections. In the interest of transparent and honesty, such uncertainty should be communicated to the general public. Thus, it is important to investigate how such uncertainty should be communicated to the general public. This study explored three uncertainty representation formats—average, range, and multi-value—to investigate how each format affected the general public’s trust, perceived accuracy, perceived likelihood, and concern after acknowledging the presence of uncertainty in climate projections (i.e., the use of multi-model climate projections). We conducted a web survey of 2400 participants in Japan, in which we randomly assigned each participant to one of three formats by which climate projection uncertainty was presented. We then asked participants to rate trust, perceived accuracy, perceived likelihood, and concern regarding the climate projections. The multi-value format enhanced trust and perceived accuracy and partially increased perceived likelihood and concern regarding the climate projections compared to the average and range formats, regardless of participants’ numeracy and education level. This study suggests that the multi-value format might be effective for communicating multi-model projections and promoting public trust and support for climate polices.

气候变化预测中的科学不确定性通常使用集合方法来解决，其中使用多个模型来生成气候预测。为了透明和诚实，这种不确定性应该向公众传达。因此，调查如何向公众传达这种不确定性非常重要。本研究探索了三种不确定性表示格式——平均值、范围和多值——以调查在承认气候预测中存在不确定性（即使用多模式气候预测）。我们对日本的 2400 名参与者进行了网络调查，我们将每个参与者随机分配到呈现气候预测不确定性的三种格式之一。然后，我们要求参与者对气候预测的信任度、感知准确性、感知可能性和担忧进行评分。与平均和范围格式相比，多值格式增强了信任和感知准确性，并部分增加了对气候预测的感知可能性和关注，无论参与者的计算能力和教育水平如何。这项研究表明，多值格式可能有效地传达多模型预测并促进公众对气候政策的信任和支持。与平均和范围格式相比，多值格式增强了信任和感知准确性，并部分增加了对气候预测的感知可能性和关注，无论参与者的计算能力和教育水平如何。这项研究表明，多值格式可能有效地传达多模型预测并促进公众对气候政策的信任和支持。与平均和范围格式相比，多值格式增强了信任和感知准确性，并部分增加了对气候预测的感知可能性和关注，无论参与者的计算能力和教育水平如何。这项研究表明，多值格式可能有效地传达多模型预测并促进公众对气候政策的信任和支持。