Development and implementation of quantitative models for the study and management of agricultural epiphytotics
Grant
Overview
Affiliation
View All
Overview
abstract
Plant diseases relevant to agricultural production are variously transmitted, affected by environmental conditions, distributed across geographic and landscape scales, behaviorally and technologically managed, exhibitive of evolutionary and ecological dynamics, and subject to change. Agricultural plant disease systems (hereafter "pathosystems") are composed of many components including a given pathogen with its characteristics, as well as all the additional aspects of agricultural production and disease management that affect the pathogen and its impacts. Models allow multiple characteristics and components of pathosystems to be studied and understood in mutual context, while also supporting the application of this understanding to the management, description, and prediction of disease (Jeger 2004, Jones et al. 2010). Models meeting criteria for beneficial deployment as predictive tools are absent or lacking in many pathosystems, especially those of recent origin or introduction, or those characterized by relative complexity and variability, such as those involving arthropod-vectored pathogens. In systems where models exist and are used, developments in computational technology, means of agricultural data acquisition (Honkravaara et al. 2012, Rosenheim & Gratton 2017), or empirical modeling methods may support improvement of existing models and implementations.A primary problem in disease management is uncertainty surrounding pathosystem-level mechanisms and dynamics. Even after disease etiological bases and molecular genetic mechanisms are understood, variation in causal factors ranging from environment to management practices results in uncertainty concerning outcomes unless there is accounting for roles of these factors. Pathosystem complexity results in there being multiple dimensions necessarily attended by any empirical approach, limiting feasibility of controlled or factorially- manipulated experimentation at this level. Observational and quasi-experimental methods in concert with model-based inference make up a readily feasible approach to answering questions about pathosystem-level relationships and dynamics. These methods offer broad complementarity with the reductionist experimental methods that are used to characterize disease-causing agents and their interactions with agricultural hosts. In addition to advancing understanding of pathosystems to the benefit of management, informative models also support experimental research by identifying likely factors of interest and factors that must be controlled or attended during experimentation.........
