Digital Transformation of Crop Production Systems: Data-Driven and Model-Informed Approaches for High-Yield and Sustainable Agriculture

Global crop production systems face unprecedented challenges driven by increasing climate variability, extreme weather events, freshwater limitations, and the imperative to reduce chemical and resource inputs. Meeting global food and economic demands under these constraints requires moving beyond conventional empirical management toward predictive, scalable, and resource-efficient agricultural practices. Advances in digital technologies, sensor networks, and data science offer unprecedented capabilities to monitor field conditions and crop performance. However, purely data-driven methods may lack physiological interpretability and can face challenges when predicting crop responses under unprecedented climatic extremes. Conversely, mechanistic, process-based crop models capture biophysical and physiological principles with high explanatory power, but their application can be constrained by calibration requirements and heterogeneous field conditions.Bridging these paradigms through model-informed and hybrid digital approaches represents an important frontier in digital crop science. By integrating mechanistic understanding with artificial intelligence, machine learning, multi-source sensing, and advanced data analytics, researchers can improve the robustness, interpretability, and timeliness of predictions of crop development, resource use, stress responses, and yield. This Research Topic aims to consolidate recent advances in the digital transformation of crop production systems, with particular emphasis on how data-driven and process-based approaches can jointly support high-yield, climate-resilient, and sustainable crop management across major food and economic crops.We invite Original Research, Reviews, Mini Reviews, Methods, Perspectives, and Technology and Code articles. Studies that connect data-driven tools with dynamic crop growth, physiological processes, process-based models, or predictive modeling frameworks are particularly welcome. We also encourage contributions using remote sensing, proximal sensing, high-throughput phenotyping, artificial intelligence, machine learning, and digital decision-support technologies, provided that they demonstrate clear relevance to crop growth, stress assessment, productivity, resource-use efficiency, or sustainable crop management.Topics of interest include, but are not limited to:• Hybrid crop modeling frameworks: Integrating mechanistic process-based models (such as DSSAT, APSIM, or functional-structural plant models) with machine learning and deep learning for enhanced prediction of crop growth, phenology, and yield.• Dynamic data assimilation and sensor fusion: Integrating proximal sensing, UAV imaging, satellite observations, and Internet of Things environmental data for crop monitoring, state estimation, stress detection, model updating, and predictive assessment across spatial and temporal scales.• Data- and model-informed climate risk and stress forecasting: Using process-based models, remote sensing, machine learning, or hybrid approaches to assess and predict crop vulnerability, canopy dynamics, phenological responses, and yield stability under drought, heat waves, and compound environmental stresses.• Digital twins and decision-support systems: Developing and evaluating data-driven, model-informed, or hybrid digital twins and decision-support tools for precision irrigation, nitrogen and nutrient management, resource-use efficiency, and adaptive agronomic interventions.• Scaling from high-throughput phenotyping to field performance: Linking multi-scale phenotypic measurements from controlled environments, proximal sensing, UAVs, and satellites with crop traits, canopy dynamics, model parameterization, and field-scale performance prediction.• Uncertainty quantification and model generalizability: Assessing prediction reliability, model uncertainty, parameter optimization, domain adaptation, and spatial-temporal transferability across diverse soils, cultivars, environments, and agroecological zones. Topic editors: Chenyao Yang, Andrej Ceglar, Helder Fraga, Bing Liu, Liujun Xiao Source: https://www.frontiersin.org/research-topics/86460/digital-transformation-of-crop-production-systems-data-driven-and-model-informed-approaches-for-high-yield-and-sustainable-agriculture