Research on Estimating Rice Canopy Height and LAI Based on LiDAR Data

Area of Science:

• Agricultural Science
• Remote Sensing Technology
• Crop Physiology

Background:

• Accurate estimation of crop phenotypic traits like canopy height and density is vital for biomass prediction in rice.
• Non-destructive methods are needed for rapid and precise measurement of these parameters.

Purpose of the Study:

• To establish a platform using LiDAR point cloud data for non-destructive detection and estimation of rice phenotypic parameters.
• To develop accurate models for predicting rice canopy height and Leaf Area Index (LAI).

Main Methods:

• A LiDAR-based platform was used to collect canopy layer data across multiple rice plots.
• Canopy-top point clouds were identified using a highest percentile method (optimal at 0.975).
• Canopy height was calculated as the difference between ground elevation and the percentile value; LAI was assessed via gap fractions and ground returns.

Main Results:

• The prediction model based on LiDAR-detected canopy height showed a strong correlation with actual height (R² = 0.941, RMSE = 0.019).
• Models predicting LAI based on ground return counts and intensity had moderate correlations (R² = 0.24, R² = 0.28).
• A prediction model using LiDAR-detected canopy height for LAI demonstrated higher accuracy (R² = 0.77, RMSE = 0.03) compared to other LAI models.

Conclusions:

• LiDAR technology provides a highly accurate and non-destructive method for estimating rice canopy height.
• Canopy height derived from LiDAR data is a reliable predictor of rice Leaf Area Index (LAI).
• This approach offers a significant advancement for precision agriculture and crop monitoring.