Deep Learning Meets Process-Based Models: A Hybrid Approach to Agricultural Challenges

TL;DR

This paper surveys process-based and deep learning approaches for agricultural modelling and advocates hybrid PBM-DL frameworks that blend mechanistic knowledge with data-driven learning. It classifies hybrids into DL-informed PBMs and PBM-informed DL, and provides a case study on crop dry biomass showing that hybrids outperform standalone PBMs and DL under noisy data, limited data, and cross-location generalization. The authors discuss differentiable modeling, surrogate and physics-informed strategies, and a range of agricultural applications from crop monitoring to risk control. They conclude with actionable recommendations to improve transparency, data standards, uncertainty quantification, and computational efficiency while outlining open challenges in interpretability, scalability, and data integration for scalable, robust agro-environmental decision-making.

Abstract

Process-based models (PBMs) and deep learning (DL) are two key approaches in agricultural modelling, each offering distinct advantages and limitations. PBMs provide mechanistic insights based on physical and biological principles, ensuring interpretability and scientific rigour. However, they often struggle with scalability, parameterisation, and adaptation to heterogeneous environments. In contrast, DL models excel at capturing complex, nonlinear patterns from large datasets but may suffer from limited interpretability, high computational demands, and overfitting in data-scarce scenarios. This study presents a systematic review of PBMs, DL models, and hybrid PBM-DL frameworks, highlighting their applications in agricultural and environmental modelling. We classify hybrid PBM-DL approaches into DL-informed PBMs, where neural networks refine process-based models, and PBM-informed DL, where physical constraints guide deep learning predictions. Additionally, we conduct a case study on crop dry biomass prediction, comparing hybrid models against standalone PBMs and DL models under varying data quality, sample sizes, and spatial conditions. The results demonstrate that hybrid models consistently outperform traditional PBMs and DL models, offering greater robustness to noisy data and improved generalisation across unseen locations. Finally, we discuss key challenges, including model interpretability, scalability, and data requirements, alongside actionable recommendations for advancing hybrid modelling in agriculture. By integrating domain knowledge with AI-driven approaches, this study contributes to the development of scalable, interpretable, and reproducible agricultural models that support data-driven decision-making for sustainable agriculture.

Figures (22)

• Figure 1: Number of publications related to the agriculture modeling of hybrid PBM-DL models, comparing purely data-driven DL models and process-based models in the domain of agricultural analysis.
• Figure 2: Simplified overview of a sub-model solution within the SIMPLACE framework enders2023simplace.
• Figure 3: Schematic overview of the major processes implemented in WOFOST de201925.
• Figure 4: Schematic overview of an APSIM simulation holzworth2014apsim.
• Figure 5: Schematic overview of an DSSAT simulation jones2003dssat.