A cross-channel temporal correlation transformer for diurnal near-surface air temperature prediction from multivariate observations
- 1School of Mathematics, Hefei University of Technology
Abstract
Time series forecasting is fundamental to numerous fields, yet long-range dependencies, multivariate coupling, and nonlinear trend capture remain persistent challenges, especially in fine-grained 24 h surface temperature prediction. To address these issues, this paper presents a novel Cross-Channel Temporal Correlation Transformer (CCT). Its core innovations are threefold: (i) A multi-branch adaptive fusion architecture that, for the first time in time series forecasting, dynamically integrates temporally enhanced, multi-scale trend, and original embedding features through a learnable gated attention mechanism, avoiding the information loss inherent in single-branch designs. (ii) A cross-temporal multi-head attention block augmented with relative positional bias and a parallel local convolutional enhancement branch, which jointly capture global patch relationships and fine-grained local dynamics. (iii) A trend-aware attention module that employs a learnable group of multi-scale convolutions and cross-variable attention, enabling adaptive trend decomposition and interaction beyond fixed-scale moving averages. These components enable the decoupled modeling and adaptive fusion of daily-scale weather evolution and hourly-scale details. Evaluated on a meteorological dataset covering six representative Chinese cities across coastal and inland climates, CCT outperforms 12 mainstream models, achieving an average MSE reduction of 6.2% and MAE reduction of 7.5%. In extreme heat conditions such as Chongqing, the MSE reduction exceeds 9%.
Related
- Quantifying turbulent kinetic energy in the middle atmosphere from wind and temperature observations: a method validated against a global model
- Are surface temperature and chlorophyll in a large deep lake related? An analysis based on satellite observations in synergy with hydrodynamic modelling and in-situ data
- Emergent constraints improve near-term temperature projections over China: evaluations against observations based on the preindustrial baseline
- Steering Diffusion Priors with Sparse Observations for High-Resolution Temperature Downscaling
- Machine-learning prediction of freezing temperature in Pb-contaminated soils remediated with biochar
BibTeX
@article{ding2026crosschannel,
title = {{A cross-channel temporal correlation transformer for diurnal near-surface air temperature prediction from multivariate observations}},
author = {Qianji Ding and Xing Huo},
journal = {Discover Environment},
year = {2026},
month = oct,
doi = {10.1007/s44274-026-01044-0},
url = {https://doi.org/10.1007/s44274-026-01044-0},
note = {CC BY},
}
RIS
TY - JOUR
TI - A cross-channel temporal correlation transformer for diurnal near-surface air temperature prediction from multivariate observations
AU - Qianji Ding
AU - Xing Huo
PY - 2026
DA - 2026/10/25
JO - Discover Environment
DO - 10.1007/s44274-026-01044-0
UR - https://doi.org/10.1007/s44274-026-01044-0
AB - Time series forecasting is fundamental to numerous fields, yet long-range dependencies, multivariate coupling, and nonlinear trend capture remain persistent challenges, especially in fine-grained 24 h surface temperature prediction. To address these issues, this paper presents a novel Cross-Channel Temporal Correlation Transformer (CCT). Its core innovations are threefold: (i) A multi-branch adaptive fusion architecture that, for the first time in time series forecasting, dynamically integrates temporally enhanced, multi-scale trend, and original embedding features through a learnable gated attention mechanism, avoiding the information loss inherent in single-branch designs. (ii) A cross-temporal multi-head attention block augmented with relative positional bias and a parallel local convolutional enhancement branch, which jointly capture global patch relationships and fine-grained local dynamics. (iii) A trend-aware attention module that employs a learnable group of multi-scale convolutions and cross-variable attention, enabling adaptive trend decomposition and interaction beyond fixed-scale moving averages. These components enable the decoupled modeling and adaptive fusion of daily-scale weather evolution and hourly-scale details. Evaluated on a meteorological dataset covering six representative Chinese cities across coastal and inland climates, CCT outperforms 12 mainstream models, achieving an average MSE reduction of 6.2% and MAE reduction of 7.5%. In extreme heat conditions such as Chongqing, the MSE reduction exceeds 9%.
SN - 2731-9431
ER -