关键词: 极端气候指数, 格网尺度, 碳排放, 皮尔逊相关系数, 时空分析

Abstract:

The greenhouse effect, driven by the continuous rise in carbon emissions, has increased the likelihood of extreme climate events. In turn, these events have further amplified carbon emission intensity and accelerated emission rates. While the academic community widely acknowledges the correlation between the two, most studies have focused solely on a single variable—either extreme climate or carbon emissions—with limited literature analyzing them jointly to explore their interrelationships. Additionally, most research has been conducted at the scale of traditional administrative divisions, which tends to obscure intra-regional differences in climate and carbon emissions. This makes it challenging to accurately identify local high-carbon emission hotspots and extreme climate zones. By contrast, analysis at the grid scale can overcome the constraints of administrative boundaries; the relationship between extreme climate and carbon emissions at a high-resolution grid scale thus merits further investigation. This study employs 20 km×20 km grids covering China as research units, integrating annual carbon emission data from the EDGAR database (2002-2022) and daily maximum temperature, minimum temperature, and precipitation data from 397 meteorological stations. After quality control and data cleaning, 18 extreme climate indices were calculated using RClimDex1.0 software, and categorized into three dimensions: frequency, intensity, and duration. The CRITIC method was adopted to standardize these indices and assign objective weights, thereby constructing a comprehensive extreme climate index. Finally, ArcGIS 10.8 was used to complete spatial interpolation and grid-scale matching, and Pearson correlation analysis was applied to quantify the correlation coefficients between grid-based carbon emissions and extreme climate indices. The results yield three key findings: 1) From 2002 to 2022, China’s carbon emissions increased by 6.6×10⁸ t, with an average annual growth rate of 4.99%; however, the growth rate slowed to 1.56% after 2018. Spatially, emissions show a pattern of gradual decline radiating outward from central cities. The grid-based distribution of carbon emissions is distinctly delineated by the Hu Line: most grids east of the line fall within the range of [1, 10)×10⁴ t, while those west of the line are mostly in the (0, 1)×10⁴ t range. 2) Extreme climate indices exhibit significant regional differences: the duration of extreme temperature and precipitation events has shortened, a trend particularly notable in North China. In contrast, the frequency and intensity of extreme temperature and precipitation have increased significantly, especially in multiple central provinces of South China. Overall, the frequency and intensity of extreme climate events have risen markedly in many regions while their duration has shortened, indicating that extreme climate events have gradually become normalized in China. 3) The extreme climate frequency index shows the most prominent positive correlation with carbon emissions nationwide—among the 34 provinces, autonomous regions, and municipalities directly under the Central Government, only Hong Kong shows a negative correlation. Conversely, the extreme precipitation duration index has the strongest negative correlation with carbon emissions, with only Xinjiang and Hong Kong showing a positive correlation. The correlations between extreme precipitation frequency/intensity indices and carbon emissions vary significantly by region: strong negative correlations are observed in Hunan and Jiangxi, while strong positive correlations are found in provinces such as Xinjiang and Inner Mongolia. Building on these findings, we observe that during the research period, the spatial distribution characteristics of China’s carbon emissions and extreme climate indices have become increasingly distinct. In 2002, carbon emissions were highest in eastern China, followed by the central region, with the lowest in the western region. In recent years, however, emissions in the western region have surpassed those in the central region, ranking second only to the eastern region. Additionally, the inter-provincial spatial correlation of carbon emissions in China presents a stable network structure. China’s extreme climate has entered a normalized phase characterized by “high frequency, strong intensity, and short duration.” The frequency and intensity of extreme precipitation exhibit a north-south gradient of “increasing in the north and decreasing in the south,” which together constitute the core characteristics of extreme climate evolution. Significant regional differences exist in the correlations between carbon emissions and different extreme climate indices, and the grid scale reveals the refined characteristics of this heterogeneity—highlighting the complex coupling relationship between human activities and extreme climate. From a mechanism perspective, the correlation between carbon emissions and extreme climate is jointly shaped by human activities, land-surface processes, and natural climatic drivers. Anthropogenic factors such as fossil-fuel combustion in energy bases and industrial clusters intensify greenhouse warming and inject considerable waste heat into the atmosphere, which further enhances local temperature increases. At the same time, rapid urbanization and the expansion of impervious surfaces reduce vegetation cover and weaken evaporative cooling, allowing extreme temperature signals associated with emissions to become more pronounced. In some economically dense southern cities, the preservation of green areas and wetlands partially offsets the warming effect. In coastal regions, atmospheric circulation and oceanic influences, including sea-land thermal contrasts, sea breezes, and ocean temperature anomalies, dilute the impact of emissions on extreme temperature and precipitation, revealing stronger external forcing and more complex drivers. These mechanisms imply that regional strategies should be differentiated. High-emission and high-risk regions, such as northwestern energy bases and eastern manufacturing areas, need to coordinate carbon mitigation with heat-risk reduction. High-emission regions with relatively weaker climatic sensitivity may leverage ecological assets to buffer warming, while coastal areas should strengthen interactions between urban and ocean systems and enhance coastal ecological barriers to improve climate resilience. The grid perspective therefore provides a clearer basis for developing region-specific mitigation and adaptation pathways. Overall, the spatial patterns of carbon emissions and extreme climate in China have become increasingly pronounced, and their linkages reflect the feedbacks between human activities and the climate system. The grid-based perspective uncovers substantial local heterogeneity that remains hidden under administrative statistics and helps identify climate-sensitive hotspots and high-emission clusters, even within the same province or city. These refined spatial details strengthen the basis for targeted climate governance and suggest that mitigation and adaptation should be planned with differentiated regional priorities. Looking ahead, integrating land-use, energy systems and socioeconomic processes into dynamic coupled models would help clarify the bidirectional interactions between extreme climate and carbon emissions, and provide more robust support for carbon mitigation, climate risk identification and adaptive management under increasingly normalized extreme climate conditions.

Key words: ETCCDI climate extremes indices, grid scale, carbon emissions, Pearson correlation analysis, spatiotemporal evolution