Comparative Analysis of Machine Learning and Deep Learning Approaches for Hourly Soil Temperature Estimation At ‎Multiple Depths from Meteorological Data in A Semi-Arid Re‎gion of Burkina Faso

  • Authors

    https://doi.org/10.14419/wcverd98

    Received date: August 30, 2025

    Accepted date: October 1, 2025

    Published date: October 12, 2025

  • Soil Temperature; Subsurface Modeling; Deep Learning; SHAP Analysis; Semi-Arid Climate

  • Abstract

    Accurate estimation of soil temperature is essential for understanding land–atmosphere interactions and managing agricultural systems, par-‎particularly in semi-arid regions. This study evaluates and compares the performance of multiple modeling approaches: machine learning (Random Forest, XGBoost, LightGBM, SVR, GPR, Decision Tree), and deep learning (MLP-ANN, LSTM, ConvLSTM, Hybrid CNN–‎LSTM) to predict hourly soil temperature at depths of 10, 20, 30, and 40 cm in a semi-arid land from Burkina Faso. Using high-frequency ‎meteorological data collected, the models were assessed based on their accuracy, temporal generalization, and variable importance. Results ‎show that among ML models, Random Forest demonstrates strong performance up to 30 cm. Deep learning models, particularly ConvLSTM and Hybrid CNN–LSTM, outperform all others across depths, capturing the spatio-temporal variability of subsurface temperature ‎with high fidelity. SHAP analysis reveals that air temperature (Ta) and solar radiation (RAD) dominate at the surface, while dew point tem-‎temperature (DPT) and vapor pressure deficit (VPD) gain importance in deeper layers. Variables such as rainfall, wind direction, and relative ‎humidity show minimal contribution and can be excluded without sacrificing performance‎.

  • References

    1. Mampitiya, L., Rozumbetov, K., Rathnayake, N., Erkudov, V., Esimbetov, A., Arachchi, S., ... & Rathnayake, U. (2024). Artificial intelligence to predict soil temperatures by development of novel model. Scientific Reports, 14(1), 9889. https://doi.org/10.1038/s41598-024-60549-x.
    2. Elsayed, S., Gupta, M., Chaudhary, G., Taneja, S., Gaur, H., Gad, M., ... & Schmidhalter, U. (2023). Interpretation the influence of hydrometeoro-logical variables on soil temperature prediction using the potential of deep learning model. Knowledge-Based Engineering and Sciences, 4(1), 55-77. https://doi.org/10.51526/kbes.2023.4.1.55-77.
    3. Sattari, M. T., Avram, A., Apaydin, H., & Matei, O. (2020). Soil Temperature Estimation with Meteorological Parameters by Using Tree-Based Hybrid Data Mining Models. Mathematics, 8(9), 1407. https://doi.org/10.3390/math8091407.
    4. Hsieh, C. I., Chiu, C. J., Huang, I. H., & Visessri, S. (2023). Estimating Ground Heat Flux from Net Radiation. Atmosphere, 14(12), 1778. https://doi.org/10.3390/atmos14121778.
    5. Edmondson, J. L., Stott, I., Davies, Z. G., Gaston, K. J., & Leake, J. R. (2016). Soil surface temperatures reveal moderation of the urban heat island effect by trees and shrubs. Scientific Reports, 6(1), 33708. https://doi.org/10.1038/srep33708.
    6. Asadzadeh, F., Emami, S., Elbeltagi, A., Akiner, M. E., Rezaverdinejad, V., Taran, F., & Salem, A. (2025). Investigating the impact of meteorolog-ical parameters on daily soil temperature changes using machine learning models. Scientific Reports, 15(1), 19988. https://doi.org/10.1038/s41598-025-04605-0.
    7. Taheri, M., Schreiner, H. K., Mohammadian, A., Shirkhani, H., Payeur, P., Imanian, H., & Cobo, J. H. (2023). A review of machine learning ap-proaches to soil temperature estimation. Sustainability, 15(9), 7677. https://doi.org/10.3390/su15097677.
    8. Šimůnek, J., Brunetti, G., Jacques, D., van Genuchten, M. T., & Šejna, M. (2024). Developments and applications of the HYDRUS computer soft-ware packages since 2016. Vadose Zone Journal, 23(4), e20310. https://doi.org/10.1002/vzj2.20310.
    9. Zhang, Y., Li, X., Šimůnek, J., Chen, N., Hu, Q., & Shi, H. (2024). Evaluating the effects of different irrigation water sources on soil temperature using HYDRUS (2D/3D) and considering the coupled movement of water and heat. Soil and Tillage Research, 244, 106259. https://doi.org/10.1016/j.still.2024.106259.
    10. Progga, J. F., Khan, M. N. H., & Amın, M. M. (2023). Meteorological Parameters–Soil Temperature Relations in a Sub-Tropical Summer Grassland: Physically-Based and Data-Driven Modeling. Atatürk Üniversitesi Ziraat Fakültesi Dergisi, 54(2), 48-56. https://doi.org/10.5152/AUAF.2023.23126.
    11. Alizamir, M., Ahmed, K. O., Kim, S., Heddam, S., Gorgij, A. D., & Chang, S. W. (2023). Development of a robust daily soil temperature estima-tion in semi-arid continental climate using meteorological predictors based on computational intelligent paradigms. PLoS One, 18(12), e0293751. https://doi.org/10.1371/journal.pone.0293751.
    12. Alizamir, M., Kisi, O., Ahmed, A. N., Mert, C., Fai, C. M., Kim, S., ... & El-Shafie, A. (2020). Advanced machine learning model for better predic-tion accuracy of soil temperature at different depths. PLoS One, 15(4), e0231055. https://doi.org/10.1371/journal.pone.0231055.
    13. Farhangmehr, V., Cobo, J. H., Mohammadian, A., Payeur, P., Shirkhani, H., & Imanian, H. (2023). A Convolutional Neural Network Model for Soil Temperature Prediction under Ordinary and Hot Weather Conditions: Comparison with a Multilayer Perceptron Model. Sustainability, 15(10), 7897. https://doi.org/10.3390/su15107897.
    14. Li-Xiaoning, X., Zhu, Y., Li, Q., Zhao, H., Zhu, J., & Zhang, C. (2023). Interpretable spatio-temporal modeling for soil temperature predic-tion. Frontiers in Forests and Global Change, 6, 1295731. https://doi.org/10.3389/ffgc.2023.1295731.
    15. Geng, Q., Wang, L., & Li, Q. (2024). Soil temperature prediction based on explainable artificial intelligence and LSTM. Frontiers in Environmental Science, 12, 1426942. https://doi.org/10.3389/fenvs.2024.1426942.
    16. Farhangmehr, V., Imanian, H., Mohammadian, A., Cobo, J. H., Shirkhani, H., & Payeur, P. (2025). A spatiotemporal CNN-LSTM deep learning model for predicting soil temperature in diverse large-scale regional climates. Science of The Total Environment, 968, 178901. https://doi.org/10.1016/j.scitotenv.2025.178901.
    17. Mohammed, S., Arshad, S., Ocwa, A., Al-Dalahmeh, M., ALDabbas, A., Alzoubi, M. M., ... & Harsányi, E. (2025). Advanced Attention-Driven Deep Learning Architectures for Multi-Depth Soil Temperature Prediction. Results in Engineering, 106508. https://doi.org/10.1016/j.rineng.2025.106508.
    18. Wang, Y., Shi, L., Hu, Y., Hu, X., Song, W., & Wang, L. (2023). A comprehensive study of deep learning for soil moisture prediction. Hydrology and Earth System Sciences Discussions, 2023, 1-38. https://doi.org/10.5194/hess-28-917-2024.
    19. Yu, F., Hao, H., & Li, Q. (2021). An Ensemble 3D Convolutional Neural Network for Spatiotemporal Soil Temperature Forecast-ing. Sustainability, 13(16), 9174. https://doi.org/10.3390/su13169174.
    20. Huang, F., Zhang, Y., Zhang, Y., Shangguan, W., Li, Q., Li, L., & Jiang, S. (2023). Interpreting Conv-LSTM for spatio-temporal soil moisture pre-diction in China. Agriculture, 13(5), 971. https://doi.org/10.3390/agriculture13050971.
    21. Ouoba Nebie, B. A., Lawane, A., Hema, C., & Siroux, M. (2025). Energy efficiency in the building sector in Burkina Faso: Literature review, SWOT analysis, and recommendations. Energies, 18(11), 2689. https://doi.org/10.3390/en18112689.
    22. Yıldırım, H. (2024). The multicollinearity effect on the performance of machine learning algorithms: Case examples in healthcare modelling. Asia-Pacific Journal of Educational, Social and Scientific Research, 12(3), 68–80. https://doi.org/10.21541/apjess.1371070.
    23. Chan, J. Y.-L., Leow, S. M. H., Bea, K. T., Cheng, W. K., Phoong, S. W., Hong, Z.-W., & Chen, Y.-L. (2022). Mitigating the Multicollinearity Problem and Its Machine Learning Approach: A Review. Mathematics, 10(8), 1283. https://doi.org/10.3390/math10081283.
    24. Mienye, I. D., & Jere, N. (2024). A survey of decision trees: Concepts, algorithms, and applications. IEEE access, 12, 86716-86727. https://doi.org/10.1109/ACCESS.2024.3416838.
    25. Ke, G., Meng, Q., Finley, T., Wang, T., Chen, W., Ma, W., ... & Liu, T. Y. (2017). LightGBM: A highly efficient gradient boosting decision tree. Advances in neural information processing systems, 30.
    26. Fan, J., Ma, X., Wu, L., Zhang, F., Yu, X., & Zeng, W. (2019). Light Gradient Boosting Machine: An efficient soft computing model for estimat-ing daily reference evapotranspiration with local and external meteorological data. Agricultural water management, 225, 105758. https://doi.org/10.1016/j.agwat.2019.105758.
    27. Airlangga, G.; Bata, J.; Adi Nugroho, O.I.; Lim, B.H.P. (2025). Hybrid CNN-LSTM Model with Custom Activation and Loss Functions for Pre-dicting Fan Actuator States in Smart Greenhouses. AgriEngineering , 7, 118. https://doi.org/10.3390/agriengineering7040118.
    28. Qian, L., Wu, L., Zhang, Z., Fan, J., Yu, X., Liu, X., ... & Cui, Y. (2024). A gap filling method for daily evapotranspiration of global flux data sets based on deep learning. Journal of Hydrology, 641, 131787. https://doi.org/10.1016/j.jhydrol.2024.131787.Li, A.; Zhang, W.; Zhang, X.; Chen, G.; Liu, X.; Jiang, A.; Zhou, F.; Peng, H. (2024). A Deep U-Net-ConvLSTM Framework with Hydrodynamic Model for Basin-Scale Hydrodynamic Prediction. Water, 16, 625. https://doi.org/10.3390/w16050625.
    29. Lundberg, S. M., & Lee, S.-I. (2017). A unified approach to interpreting model predictions. arXiv. https://arxiv.org/abs/1705.07874.
    30. Douanla, A. Y., Dembélé, A., Ossénatou, M., & Lenouo, A. (2022, February). Prediction of daily direct solar energy based on XGBoost in Came-roon and key parameter impacts analysis. In 2022 IEEE Multi-conference on Natural and Engineering Sciences for Sahel's Sustainable Development (MNE3SD) (pp. 1–7). IEEE. https://doi.org/10.1109/MNE3SD53781.2022.9723309.

  • Downloads

  • How to Cite

    Dabilgou, F., Kébré, M. B., Gandema, S., & Koalaga, Z. (2025). Comparative Analysis of Machine Learning and Deep Learning Approaches for Hourly Soil Temperature Estimation At ‎Multiple Depths from Meteorological Data in A Semi-Arid Re‎gion of Burkina Faso. International Journal of Basic and Applied Sciences, 14(6), 242-255. https://doi.org/10.14419/wcverd98