Climate change poses great threats to the planet and human societies, warranting urgent mitigation and adaptation strategies. To motivate individual-level behavioral changes and build social consensus on collective actions, it is essential to promote social awareness of the urgency of climate change1,2. In light of this, researchers have made efforts to measure climate change risk perception, mostly focusing on developed countries in the Global North. In the US, for example, studies have investigated county-level spatial variations in a range of climate change beliefs3, temporal shifts over decades4, and misalignments between climate-related risks and risk perceptions5.
For many highly-populated developing countries, knowledge about public perception of climate risks remains limited. China is experiencing an increased intensity of climate change shocks and has taken ambitious climate actions in recent years6,7. Efforts made to measure public perception of climate change in China, however, are currently fragmented. Yu et al. conducted one of the earliest studies on Chinese perception of climate change, surveying around 400 college students in Beijing8. Most of the subsequent studies have focused on limited samples, such as farmers, teenagers9, or residents in a few cities10. Yet, the impacts of climate change are geographically heterogeneous, meaning that certain areas and population groups are much more vulnerable to climate change than others. To design more targeted communication strategies, it is necessary to understand the spatial pattern of climate change perceptions and how they have changed over time.
In this study, we quantified the shifts in public perceptions of climate change in China at both provincial and prefecture-city levels, based on two national surveys conducted in 2010 and 2023 (N = 11783 and N = 4050, respectively). Two questions were used to measure climate change issue priority and risk perception: “Is climate change the most urgent environmental problem in China?” (Perceived Priority) and “Will climate change have the greatest impact on you and your families compared to other environmental issues?” (Perceived Impact). Our analysis measured the changes in perceived priority and perceived impact over the 14-year period, revealed disparities between and within provinces, and identified the factors influencing public perception in China.
Growing climate change perception in China. We deployed a multilevel regression and poststratification model (hereinafter MRP) to generate reliable estimates of climate change perceptions at both provincial and prefecture-city levels based on national surveys 11,12. Using the location-scale model 13, we then compared the shifts in national mean climate change perception and its provincial variability. We find that the perceived priority of climate change in China has increased in all provinces (Fig. 1A). The national mean value rose from 5.91% in 2010 to 23.4% in 2023 (t = 19.84, P-value < 0.001; disaggregation methods). In 2010, the percentage of residents viewing climate change as the top priority ranged from 4.4% in Beijing (BJ), China’s capital, to 11.5% in Yunnan (YN), one of the least developed provinces in southern China. In 2023, this range increased to 16.9% in Guangdong (GD), China’s most populous province, to 40.1% in Liaoning (LN), the most populous province in coastal Northeast China.
In comparison, the perceived impact of climate change (Fig. 1B) in China is lower, but still exhibits significant increases over the last decade. Nationally, the perceived impact of climate change increased from 4.04% in 2010 to 15.6% in 2023 (t = 25.45, P-value < 0.001). In 2010, people who believed that climate change would have the greatest impacts ranged from 2.4% in Beijing (BJ) to 8.4% in Yunnan (YN). By 2023, these percentages increased to a range of 12.9% in Shandong (SD), a coastal province in East China, to 22.9% in Heilongjiang (HL), China’s northernmost and easternmost province. These increases can be attributed to factors such as broader media coverage14, more frequent and widespread personal experience2, and intensified domestic and international policy efforts.
We then calculated the logarithmic standard deviation of provincial perception scores, a measurement of provincial variability, to investigate the disparities in climate change perceptions across the country. We find that despite strong increases in perceived priority and impact of climate change in China, the variations in climate change perceptions across provinces were amplified from 2010 to 2023. The logarithmic standard deviation values have increased by 0.999 for perceived priority (t = 5.437, p < 0.001) and by 0.977 for perceived impact (t = 5.563, p < 0.001). It suggests that climate change perceptions have not grown in the same fashion across the different regions of China.
Regional disparities in growth of Chinese climate change perceptions. We used Analysis of Variance (ANOVA) tests to compare the 14-year changes in perceived priority and impact between two pairs of province groups (i.e., Northern and Southern provinces, Eastern and Western provinces) to investigate how the estimated growth in climate change perception differed across regions (Fig. 1C and Fig. 1D). Regions were divided according to conventions commonly adopted in previous studies 15,16. Our results show that people living in Eastern China experienced more pronounced increases in perceived priority (F = 17.090, p < 0.001) and impact (F = 8.797, p < 0.01) compared to those in the West. Views of people in the Southern China have grown faster for perceived impacts (F = 8.341, P < 0.01), but not for perceived priority (F = 0.693, P = 0.412). Slow progress in raising climate change perceptions of residents in the North and the West may induce spatial misalignment between climate change risks and perceptions, as these large regions are at increasing risk of more intense water scarcity and extreme heat events in the future 17.
We also used the MRP model to estimate fine-scale prefecture-city level perceived priority and impact in China for the year 2023 (Fig. 2A; Fig. 2C). The results show significant heterogeneity between cities. The percentage of the population who identify climate change as a priority range from a low of 18% to a high of 28% at the city level (a 10% gap). The percentage of residents who perceive climate change impact show even greater city-level variations, ranging from a low of 7% to a high of 29% (a 22% gap). We applied spatial autocorrelation analysis to discover hot spots and cold spots of climate change perception. A global univariate Moran’s I test suggests that both city-level perceived priority and perceived impact are spatially clustered (global Moran’s I = 0.54 and 0.46, respectively). A Local Moran’s I test analysis further identifies significant locations as High-High and Low-Low spatial clusters (Fig. 2B; Fig. 2D). In this context, 'High-High' clusters refer to areas where both a particular location and its neighboring locations have high values of perceived priority and impact of climate change. These 'High-High' clusters, or hotspots of elevated climate change perception, are mainly found along the southern and eastern coasts of China. Conversely, cities located in the North and Central China Plain and a large swath in Western Xinjiang Province form ‘Low-Low’ clusters, implying lower levels of perceived priority and impact from climate change.
Influences on climate change issue salience and risk perception in China. After conducting our MRP estimation, we explored various city-level factors that could potentially influence the average perceived priority and impact of climate change within each city. This analysis, distinct from our MRP estimation, focused on natural, demographic, and socio-economic features at the city level rather than individual-level predictors used in MRP model. We chose influencing factors based on their relevance in affecting people's experience and understanding of climate change reported in previous literature 1–3,18–20. Here, we present those factors which were found to have a significant impact.
Firstly, cities with more urban residents were associated with higher perceived priority towards climate change compared to those with a high proportion of rural residents (r = 0.28, P < 0.001). The correlations between the urban population's age structure and climate change perceptions were relatively weak. However, cities with a higher proportion of individuals aged 46–55 exhibited significantly lower perceived priority and impact towards climate change (r = -0.17, P < 0.005; r = -0.14, P < 0.01, respectively), compared to cities with a different age composition. It shows that urbanization process and changing demographics reshape climate change perception in China. Elder population groups could be more vulnerable to physical risks due to lower awareness of climate change. Secondly, environmental factors including urban rainfall and proximity to green space had significant impacts on climate change perception. Cities with more precipitation events predict higher public perceived priority (r = 0.41, P < 0.001) and perceived impact (r = 0.47, P < 0.001) of climate change. In addition, urban green space coverage was also positively correlated with both perceived priority (r = 0.16, P < 0.005) and perceived impact (r = 0.17, P < 0.001). It suggests that personal experience with extreme weather impacts helps enhance public climate change awareness. Lastly, we collected the annual average frequency of 'climate change' searches on Baidu, the largest search engine in China. We find that the internet search behavior, as a measurement of public attention, was positively correlated with increased perceived priority (r = 0.15, P < 0.01). The collective influence of these diverse elements, spanning societal context to environmental factors, suggests a complex array of drivers underlying the spatial disparities in climate change perception across China 20.
In this study, we provide the first high-resolution estimate of climate change perception in China and report a substantial overall increase in public awareness. The findings also underscore the spatial heterogeneity in public perception of climate change, which could result in diverse adaptation and mitigation behaviors across various regions. This disparity may exacerbate the unequal distribution of climate change impacts, as individuals in regions with higher perceived risk might be more inclined to adopt adaptation and mitigation strategies, while those in areas with lower perceived risk may be less likely to take actions (Fig. A3). Moreover, significant differences between subregions within the same province highlight the necessity for targeted policies and communication strategies to ensure a coordinated and comprehensive approach to address climate change at both regional and local levels. Further investigation into the determinants of climate change perception will be crucial for guiding these processes both in China and beyond.