Ecologically-based intensification is a key strategy for sustainably increasing agricultural production

Intercropping could potentially improve productivity via improving crop plant defenses. However, identifying plants that will grow well together and trigger chemical upregulation is not a trivial task. Plants, their phytochemistries, and their symbionts form a complex system whose dynamics are difficult to predict via trial-and-error studies alone. Using data science and greenhouse trials I propose to (a) determine the evolutionary, chemical, and/or ecological factors that predict chemical defense synergies between crops, companion plants, and arbuscular mycorrhizal fungi (AMF), and (B) develop an analytical program that can be used to identify candidate species for novel intercropping systems that maximize crop productivity and defense.using a metadataset containing phylogenetic, phytochemical, and ecological descriptors of over 600 plant species and their associated AMF, I will: 1) test its utility by sampling volatiles from plants that are not currently present in the dataset and match the compound identifications with those in the database. 2) use a combination of phylogenetic relatedness and phytochemical similarity to select 8 candidate intercrops to be grown with 3 companion crops, and empirically test our algorithm's ability to predict successful plant pairings. 3) sample headspace volatiles from plants grown in polyculture to test if plants can alter the chemical profile of their neighbors. 4) use information gained in objectives 1-3 to develop an electronic application that packages this groundbreaking approach into a convenient tool that allows users to select plant species to intercrop within their own fields.