Abstract
Automated field phenotyping provides continuous and precise measures of adaptation and performance traits that are key to today’s breeding and agricultural practices. Besides monitoring morphological changes of crop growth and development, highresolution and high-frequency of phenotypic measures can empower an accurate delineation of the genotype to phenotype pathway enabling the assessment of genes controlling yield potential and environmental adaptation. Here, we present CropQuant, a cost-effective Internet of Things (IoT) powered phenotyping platform, designed to be easily used and widely deployed in any environment. To manage and process data generated by the platform, we developed an automatic in-field control system, highthroughput trait analysis algorithms, and machine-learning based modelling to explore the dynamics between genotypes, phenotypes and environment. We used the platform in a 95-day field experiment to generate dynamic developmental profiles of five wheat genotypes within the single genetic background of Paragon (a UK spring wheat variety) and demonstrated a successful example of how this technology could be applied to breeding, crop research and digital agriculture.