Multitemporal Analysis of Sugarcane Crops Using Satellite Indices and Agronomic Variables

Traditional monitoring of extensive sugarcane plantations faces challenges associated with limited field visibility and fragmented information, which can delay the detection of anomalies such as water stress, pests, and crop stand reduction. To address this issue, the Geomatics and Agricultural Technologies Department at Ingenio San Antonio developed a geospatial monitoring methodology that integrates remote sensing and data analytics.

The system combines PlanetScope (3 m) and Sentinel-2 (10 m) imagery with agronomic and operational variables from ERP systems, including variety, irrigation, fertilization, agroclimatic variables, and historical yield. Using cloud infrastructure integrating AWS S3, Google Earth Engine, and Google Colab, the system automates the analysis of spectral indices such as NDVI, MSAVI, NDWI, and Contrast-Enhanced NDVI at 60, 90, and 120 days.

The methodology enables comparison of up to four historical crop seasons per field, identifying trends in crop vigor and moisture. Results are presented through field-level reports and dashboards, supporting decision-making. Its implementation has enabled timely detection of water stress and crop stand reduction, strengthening precision agriculture and operational traceability from satellite monitoring to field action.