Methodological evolution of potato yield prediction: a comprehensive review.

Area of Science:

• Agricultural Science
• Remote Sensing
• Data Science

Background:

• Potato (Solanum tuberosum L.) is a global staple crop, necessitating improved yield prediction for food security and economic stability.
• Accurate crop yield forecasting supports agricultural planning, insurance, and trade.
• Methodological advancements are crucial for enhancing potato production efficiency.

Purpose of the Study:

• To review the evolution of potato yield prediction methodologies.
• To highlight the roles of remote sensing (RS), crop growth models (CGM), and yield limiting factor (LF) analysis.
• To provide a theoretical basis for future potato yield prediction research.

Main Methods:

• Comprehensive literature survey of potato yield prediction studies.
• Analysis of methodologies including remote sensing (satellite and UAV-based), crop growth models, and yield limiting factor analysis.
• Evaluation of the integration of artificial intelligence (AI) with remote sensing techniques.

Main Results:

• Remote sensing, particularly satellite-based, is key for large-scale potato yield prediction and decision support.
• Crop growth models are effective for optimizing management and addressing climate change impacts.
• Unmanned aerial vehicle (UAV) remote sensing combined with AI offers significant potential for precision agriculture in potato farming.
• Current research is limited by the number of potato varieties and field sample sizes studied.

Conclusions:

• Future potato yield prediction requires time-series data from multiple sources, encompassing a wider range of varieties and larger field sample sizes.
• Integrating advanced techniques like UAV-based RS and AI is crucial for precision potato management.
• Continued research is essential to refine prediction accuracy and support sustainable potato production.