Development and Validation of a Phenotyping Computational Workflow to Predict the Biomass Yield of a Large Perennial Ryegrass Breeding Field Trial

Alem Gebremedhin,Pieter Badenhorst,Fan Shi,Khageswor Giri,Ed Breen,Junping Wang,German C Spangenberg,Kevin Smith

doi:10.3389/fpls.2020.00689

Abstract

Increasing dry matter yield (DMY) is the most important objective in perennial ryegrass breeding programs. Current yield assessment methods like cutting are time-consuming and destructive, non-destructive measures such as scoring yield on single plants by visual inspection may be subjective. These assessments involve multiple measurements and selection procedures across seasons and years to evaluate biomass yield repeatedly. This contributes to the slow process of new cultivar development and commercialisation. This study developed and validated a computational phenotyping workflow for image acquisition, processing and analysis of spaced planted ryegrass and investigated sensor-based DMY yield estimation of individual plants through normalized difference vegetative index (NDVI) and ultrasonic plant height data extraction. The DMY of 48,000 individual plants representing 50 advanced breeding lines and commercial cultivars was accurately estimated at multiple harvests across the growing season. NDVI, plant height and predicted DMY obtained from aerial and ground-based sensors illustrated the variation within and between cultivars across different seasons. Combining NDVI and plant height of individual plants was a robust method to enable high-throughput phenotyping of biomass yield in ryegrass breeding. Similarly, the plot-level model indicated good to high-coefficients of determination (R2) between the predicted and measured DMY across three seasons with R2 between 0.19 and 0.81 and root mean square errors (RMSE) values ranging from 0.09 to 0.21 kg/plot. The model was further validated using a combined regression of the three seasons harvests. This study further sets a foundation for the application of sensor technologies combined with genomic studies that lead to greater rates of genetic gain in perennial ryegrass biomass yield.

Highlights

Increasing biomass yield is the most important trait to improve during the breeding of perennial ryegrass (Smith et al, 2001; McDonagh et al, 2016; Herridge et al, 2018; Ghamkhar et al, 2019)
We have recently shown that normalized difference vegetative index (NDVI) and plant height correlated with perennial ryegrass biomass yield in four seasons (Gebremedhin et al, 2019a)
The results show a strong coefficient of determination between predicted dry matter yield (DMY) from automated and manual polygons with R2 values ranging 0.82-0.93 for the winter2018 and late-spring2018_1 season and root mean square errors (RMSE) values ranging from 1.81 to 3.83 g/plant

Summary

Introduction

Increasing biomass yield is the most important trait to improve during the breeding of perennial ryegrass (Smith et al, 2001; McDonagh et al, 2016; Herridge et al, 2018; Ghamkhar et al, 2019). The early stages of perennial ryegrass breeding programs depend on the assessment of populations based on large numbers of genotypes planted as spaced plants or small sward plots in the field (Lootens et al, 2016; Ghamkhar et al, 2019). These assessments involve multiple measurement and selection procedures across seasons and years to repeatedly evaluate biomass yield (Leddin et al, 2018). The application of sensorbased high-throughput phenotyping (HTP) technologies on aerial and ground-based mobile platforms have the potential to offer a non-destructive, rapid and efficient method to assess biomass yield in the field (Inostroza et al, 2016; Gebremedhin et al, 2019b; Ghamkhar et al, 2019)

Objectives

Methods

Results

Conclusion