Xiaoyun Su

Accurate high-resolution temperature predictions at the subseasonal-to-seasonal (S2S) timescale are vital for applications in agriculture, energy management, and disaster mitigation. However, current S2S forecasting capacity is hindered by declining predictive skill with increasing lead time and limited capability in capturing extreme events and fine-scale spatial variability, particularly over regions with complex terrain. To bridge this gap, this study develops a deep learning (DL)-based framework, termed Terrain-ResNet-U-Net (TRU-Net), for the downscaling (from 0.5 ∘ to 0.1 ∘ resolution) and bias correction of 2 m air temperature forecasts over North China. Unlike existing approaches that treat topographic information as auxiliary input, TRU-Net integrates terrain information through a dedicated feature extraction module and incorporates it into a U-Net architecture with a ResNet34 backbone and a spatial attention mechanism, enabling enhanced nonlinear feature learning and improved representation of terrain-induced spatial heterogeneity. Comparative experiments show that TRU-Net consistently outperforms the original forecasts for daily 2 m temperature prediction at the S2S timescale. Averaged across the six lead times, TRU-Net reduces the root mean squared error by 22.97% and increases the Pearson correlation coefficient and Accuracy by 25.16% and 24.48%, respectively, relative to the original forecasts. It also outperforms representative baseline methods, including Linear Regression, U-Net, and several DL baselines. The largest overall improvements are observed at lead times of 3–5 weeks, while the ablation results confirm the contributions of terrain information and the proposed multi-scale terrain feature extraction and fusion strategy. Similar advantages are maintained for representative high-impact events, where the proposed framework also provides positive correction and downscaling effects under these challenging conditions. These results indicate that an effective terrain-aware framework can substantially improve S2S temperature bias correction and downscaling over North China.

The apparent temperature (APT), or human-perceived temperature, is commonly defined as a function of the surface air temperature (SAT), vapour pressure (or humidity) and wind speed. This paper demonstrates that the APT over China, as revealed by daily station- observed data, has generally increased faster than the SAT during summertime in the past 50 years (1968-2017). The rate of increase in APT was significantly faster than that of SAT in 60.1% of stations, and the difference between the average China-wide APT and SAT was 0.11°C decade−1. This phenomenon is occurring nationwide, but it is more intense over western, north-eastern and eastern coastal China. The more rapid increasing trend in APT indicates that human beings actually experience surplus heat stress under a certain change in SAT, and the increased SAT explains 67.0% of the average APT warming for the country, contributing to the change in the base APT. Apart from the increasing SAT, a decrease in surface wind speed and an increase in surface vapour pressure have also been observed, contributing to 21.6% of the increase in APT and explaining the remaining 11.4%, respectively.