Land Surface Temperature (LST), as a key indicator for characterizing surface energy exchange and urban thermal environments, plays a crucial role in climate change research and sustainable urban development. To address the challenges of strong spatial heterogeneity, pronounced temporal periodicity, and difficulty in modeling long-term dependencies in monthly LST sequences, this study proposes an MSG-ConvLSTM model that integrates multi-scale spatial feature extraction with a periodic awareness mechanism. The model enhances spatial representation through parallel multi-scale convolutions and incorporates a sinusoidal positional encoding–based periodic gating mechanism to explicitly capture seasonal variations, thereby improving its ability to model long-term dependencies. Experimental results based on the Pearl River Delta dataset from 2003 to 2023 demonstrate that the proposed model outperforms several mainstream methods in multi-step forecasting tasks. Compared with ConvLSTM, the proposed model reduces MSE by approximately 10%–15%; compared with CNN-BiLSTM, by 12%–18%; compared with Swin-Transformer, by 8%–12%; and compared with PredRNN, by 9%–14%. Similar improvements are also observed in terms of MAE. Notably, in medium- and long-term forecasting, the model exhibits slower error accumulation and stronger stability. Ablation studies further verify the effectiveness of the multi-scale feature extraction and periodic modeling mechanisms in improving prediction accuracy and mitigating error propagation. This study provides an effective and robust approach for monthly LST forecasting using only historical LST data.

Yajun Xiao. A Multi-Scale ConvLSTM with Seasonal Encoding for Monthly Land Surface Temperature Prediction. Academic Journal of Computing & Information Science (2026), Vol. 9, Issue 5: 1-8. https://doi.org/10.25236/AJCIS.2026.090501.

[1] Gaur A, Deb C. Machine learning methods and approaches for Urban Heat Island (UHI) assessment: A comprehensive review[J]. Renewable and Sustainable Energy Reviews, 2026, 234: 116903.

[2] Hu L, Sun Y, Collins G, et al. Corrigendum to “Improved estimates of monthly land surface temperature from MODIS using a diurnal temperature cycle (DTC) model”[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2021, 171: 118.

[3] Kesavan R, Muthian M, Sudalaimuthu K, et al. ARIMA modeling for forecasting land surface temperature and determination of urban heat island using remote sensing techniques for Chennai city, India[J]. Arabian Journal of Geosciences, 2021, 14(11): 1016.

[4] Nurwanda A, Honjo T. The prediction of city expansion and land surface temperature in Bogor City, Indonesia[J]. Sustainable Cities and Society, 2020, 52: 101772.

[5] Zhao H, Cui Y, Wang J, et al. A multiscale and seasonal model for urban surface temperature prediction based on landscape, land use and spectral indices[J]. Sustainable Cities and Society, 2025, 131: 106783.

[6] Suthar G, Singh S, Kaul N, et al. Prediction of land surface temperature using spectral indices, air pollutants, and urbanization parameters for Hyderabad city of India using six machine learning approaches[J]. Remote Sensing Applications: Society and Environment, 2024, 35: 101265.

[7] Zhang J, Xiao C, Liang X, et al. Machine learning based on a swarm intelligence algorithm and explainable AI for the prediction of reservoir temperature[J]. Energy, 2025, 341: 139412.

[8] Deo R C, Şahin M. Forecasting long-term global solar radiation with an ANN algorithm coupled with satellite-derived (MODIS) land surface temperature (LST) for regional locations in Queensland[J]. Renewable and Sustainable Energy Reviews, 2017, 72: 828-848.

[9] Li Q, Zheng H. Prediction of summer daytime land surface temperature in urban environments based on machine learning[J]. Sustainable Cities and Society, 2023, 97: 104732.

[10] Xu S, Dai D, Cui X, et al. A deep learning approach to predict sea surface temperature based on multiple modes[J]. Ocean Modelling, 2023, 181: 102158.

[11] Xin N, Su J, Hasan M M. MMformer with adaptive attention: Advancing multivariate time series forecasting for environmental applications[J]. Applied Soft Computing, 2026, 186(Part B): 114090.

[12] Shi X, Chen Z, Wang H, et al. Convolutional LSTM network: a machine learning approach for precipitation nowcasting[M]. MIT Press, 2015.

[13] Rajesh J, Pande C B. Estimation of land surface temperature for Rahuri Taluka, Ahmednagar District (MS, India) using remote sensing data and algorithm[M]. Climate Change Impacts on Natural Resources, Ecosystems and Agricultural Systems. Cham: Springer, 2023: 565-577.

[14] Pande C B. Land use/land cover and change detection mapping in Rahuri watershed area (MS), India using the Google Earth Engine and machine learning approach[J]. Geocarto International, 2022, 37(26): 13860-13880.

[15] Zhu J, Huang J, Li H, et al. Multi-scenario simulation of land use change and carbon stock assessment in the Pearl River Delta urban agglomeration[J]. Journal of South China Normal University (Natural Science Edition), 2025, 57(03): 62-73.

[16] Kim S H, Lee S-H, Chung C C. Phase shift calibration method in optical sinusoidal encoder signals applied to servo track writer[J]. IFAC-PapersOnLine, 2016, 49(21): 1-6.

[17] Karadag Y M, Talaz I, Dino I G, et al. ms-mamba: Multi-scale mamba for time-series forecasting[J]. Neurocomputing, 2026, 680: 133226.