On Digital Twin for a High Yielding Soybean Variety Towards Optimal Field Recommendations

Soybean in India is pivotal for its nutritional value and as livestock feed. Its cultivation faces challenges like pests, diseases, and adverse weather, impacting yields and farmer incomes. Sustainable soybean production can enhance farmer livelihoods and ensure long-term agricultural viability. Creating a digital twin of the crop which models how it matures under specific conditions and operations can be an effective way to understand the yield outputs and environmental impact. For a high-yielding soybean variety KDS 992 (Phule Durva), we present a digital twin based model for monitoring sustainability practices and crop health, validated with a field study over a Kharif (monsoon) season in India. Results showed that measured yield for the selected variety aligned well with the model's predicted yield. With organic practices, soil health improved with a net increase of 1.50 <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{tCO}_{2}\mathbf{e}/\mathbf{ha}$</tex> in soil organic carbon (SOC). As a contrast, the model projected an SOC loss of 5.88 <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{tCO}_{2}\mathbf{e}/\mathbf{ha}$</tex> with conventional practices. AI-driven models integrated with the framework detected and contained soybean stress timely, boosting crop health. We believe the integrated digital insights would assist in giving specific recommendations to farmers planting soybean in general and the variety in particular.