A Simple and Robust Spectral Index for Identifying Lodged Maize Using Gaofen1 Satellite Data

Area of Science:

• Agricultural Science
• Remote Sensing Technology
• Geospatial Analysis

Background:

• Crop lodging significantly impacts agricultural productivity and necessitates efficient assessment methods.
• Satellite remote sensing offers large-scale data but faces limitations in detecting lodging due to data resolution and event complexity.
• The Gaofen1 satellite's characteristics present an opportunity for improved lodging identification.

Purpose of the Study:

• To develop and validate a cost-efficient method for assessing crop lodging using satellite data.
• To explore the potential of Gaofen1 satellite data for identifying lodged maize.
• To propose a novel spectral index for distinguishing lodged from non-lodged maize.

Main Methods:

• Analysis of spectral features of lodged and non-lodged maize.
• Development of a spectral sum index for lodging detection.
• Validation using ground sample points, Planet data comparison, and farmer surveys.
• Application of the method to Gaofen1 satellite data in two study areas.

Main Results:

• Lodged maize exhibited significantly increased reflectance across all spectral bands compared to non-lodged maize.
• The proposed spectral sum index demonstrated effectiveness in distinguishing lodged maize.
• Validation achieved high overall accuracies (92.86% and 88.24%) and showed strong agreement with finer-resolution data and farmer surveys.

Conclusions:

• The spectral sum index is a promising tool for maize lodging identification.
• Gaofen1 satellite data holds significant potential for rapid and large-scale crop lodging monitoring.
• The developed method offers a cost-efficient approach to reduce lodging-related agricultural losses.