Enhancing management strategies for crop yield improvement via advanced data-informed decisions

Abstract

Generating multi-scale information to understand crop variability and optimize crop management is crucial for increasing yields and improving input efficiency to meet global food demand. This dissertation uses available technological tools to generate actionable insights and improve our understanding of crop behavior at different scales. This dissertation is divided into five chapters. The first chapter serves as an introduction to crop yield improvement strategies and describes technologies that facilitate the generation of knowledge. The remaining chapters present case studies focusing on yield improvement, testing different crop management practices and using data approaches. Chapter 2 uses crop modelling to explore different combinations of soybean planting dates and maturity groups at the regional level in Kansas, USA, under both current and future weather scenarios. Chapter 3 uses the Canopeo tool, crop modelling and Bayesian analysis to assess the effects of equidistant and non-equidistant soybean planting patterns at different site densities in three US states. Chapter 4 uses multiple years of geo-referenced yield monitor data from GPS enabled combines and spatio-temporal analysis to establish a data pipeline that quantifies within-field yield variation in response to precipitation. Finally, Chapter 5 summarizes the key findings and outlines directions for future research to further improve crop management information and yield outcomes, a critical step in meeting the world’s growing food demand.