Monitoring yellow rust progression during spring critical wheat growth periods using multi-temporal Sentinel-2...

Area of Science:

• Agricultural remote sensing
• Plant pathology
• Geospatial analysis

Background:

• Yellow rust (Puccinia striiformis f. sp. tritici) severely impacts wheat yield and food security in China.
• Monitoring disease progression during critical spring growth stages is vital for prediction and management.

Purpose of the Study:

• To develop and validate period-specific remote sensing models for mapping wheat yellow rust progression.
• To track the spatiotemporal changes of yellow rust during wheat jointing, heading, and filling stages.

Main Methods:

• Acquired Sentinel-2 imagery for three key wheat growth periods.
• Extracted spectral, texture, and color features to create period-specific datasets.
• Developed and validated three targeted monitoring models for disease mapping and spatiotemporal analysis.

Main Results:

• Models accurately mapped wheat yellow rust progression with high accuracy (87.4% classification, R² of 0.77-0.76 for regression).
• Spatiotemporal change detection revealed high conformity (98.2% and 84.4%) with expected disease patterns.
• Validated models demonstrated realistic and credible representation of disease spread.

Conclusions:

• The multi-period targeted monitoring model group overcomes single-phase limitations.
• Effectively reveals wheat yellow rust spatiotemporal epidemics in spring.
• Provides timely disease trend updates for optimized management and predictive model calibration.