Zhang Fuyao, Li Xiubin, Wang Xue, Tan Minghong, Xin Liangjie
Key Laboratory of Land Surface Pattern and Simulation, Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing, 100101, China.
College of Resources and Environment, University of Chinese Academy of Sciences, Beijing, 100049, China.
Sci Data. 2025 Apr 30;12(1):720. doi: 10.1038/s41597-025-05071-z.
Global warming has intensified extreme weather events, posing challenges to regional climate and hydro-ecological systems. To address the low-resolution limitations of current multi-climate variables and potential evapotranspiration (PET), this study develops a super-resolution fusion framework based on deep residual attention mechanisms, establishing China's 10-km resolution multi-model-multi-scenario high-resolution climate and PET dataset (SRCPCN10). The Residual Channel Attention Network (RCAN) demonstrates exceptional downscaling performance for temperature, radiation, and pressure (R/KGE > 0.99), while precipitation exhibits significantly lower accuracy (R = 0.897) due to spatial discontinuity. The findings reveal distinct emission-gradient responses in China's future climate variables under SSP scenarios, with temperature, radiation, and precipitation increases escalating alongside radiative forcing intensification. The comparison of annual mean differences between original CMIP6 and downscaled data showed excellent agreement, with most climate indices differing by less than 1%. This work overcomes traditional limitations, providing kilometer-scale multivariate data for watershed hydrological modeling, agricultural climate risk assessment, and carbon-neutral pathway optimization, enhancing the precision of regional adaptation strategies.
全球变暖加剧了极端天气事件，给区域气候和水文生态系统带来了挑战。为了解决当前多气候变量和潜在蒸散量（PET）的低分辨率限制问题，本研究基于深度残差注意力机制开发了一种超分辨率融合框架，建立了中国10公里分辨率的多模型多情景高分辨率气候和PET数据集（SRCPCN10）。残差通道注意力网络（RCAN）在温度、辐射和气压的降尺度方面表现出卓越的性能（R/KGE > 0.99），而由于空间不连续性，降水的精度显著较低（R = 0.897）。研究结果揭示了在共享社会经济路径（SSP）情景下，中国未来气候变量存在明显的排放梯度响应，温度、辐射和降水的增加随着辐射强迫的增强而加剧。原始CMIP6数据和降尺度数据之间的年平均差异比较显示出高度一致性，大多数气候指数的差异小于1%。这项工作克服了传统局限性，为流域水文建模、农业气候风险评估和碳中和路径优化提供了公里尺度的多变量数据，提高了区域适应策略的精度。
