Sustainable system design for gridded, spatio-temporal, agroecosystem forecasting models

Abstract

Decision support systems able to capitalize on publicly available high resolution datasets have become increasingly valuable to agroecosystem, hydrologic and urban system stakeholders. In this paper we address the common agroecosystem modeling problem of weather-based risk forecasting. We compare storage system designs for an expandable crop disease forecasting system that relies on multiple gridded weather forecast inputs to artificial neural network disease risk models. A traditional relational database management system (PostgreSQL), a NoSQL database system (MongoDB) and a scientific file format version (netCDF) of a single crop disease risk modeling system in one region of the country, for potato late blight in the US Great Lakes region, were designed and compared for speed. To test expandability, another crop disease risk modeling system, for modeling risk of economically significant deoxynivalenol (eDON) accumulation due to Fusarium head blight of barley in the northern US Great Plains, was also created in the three formats. Speeds for the three types of systems were fairly similar. Expandability, which is becoming highly desirable in agroecosystem model design, differed based on designer priorities.