A Digital Twin of an Agrocenosis for Modeling Plant Growth Based on an Automated Mapping System

This paper presents the concept and architecture of a digital twin of an agrocenosis developed for crop growth modeling based on an automated mapping system. The digital twin is structured as a multi-layer system integrating physical-geographical, technological, biophysical, observational, and modeling components. The physical-geographical layer includes a digital elevation model, soil properties, and climatic and hydrological factors that determine spatial variability of growing conditions. The technological layer formalizes agricultural operations in space and time, forming a digital field management map. The biophysical layer describes crop state variables such as phenological stage, biomass, and leaf area index (LAI). The observation layer combines satellite imagery, UAV data, and ground-based sensors to ensure continuous synchronization between the real field and its digital representation. A modular mathematical framework integrating agrophysical balance models and data-driven approaches is proposed. The digital twin enables scenario analysis, yield forecasting, and optimization of agricultural technologies, supporting precision agriculture and risk management under changing climatic conditions.