Beyond high‐throughput: leveraging plant phenotyping to improve understanding and prediction of plant growth through process‐based models

The last decade has marked a period of rapid innovation in high-throughput phenotyping (HTP) of plants. This includes the establishment of robotic phenotyping infrastructure, development of new sensors, and improvements in computation for downstream analysis. While HTP approaches have revolutionized data collection, meaningful insights into plant function require a yet deeper connection between resultant HTP-based information and biological responses. We suggest that dynamic process-based plant models, which simulate growth and physiology in a time-explicit manner, can serve as a functional link between high-throughput methods and whole-plant mechanisms of growth. Using this framework, we review recent research that has leveraged HTP approaches for estimation of plant traits that are commonly used as process-based model (PBM) variables. Through this analysis, we review successes and identify emerging directions for future research. Finally, we highlight the varied ways that HTP can be used in conjunction with PBMs as a tool to advance discovery and improve prediction of plant growth.