Precision Agriculture (PA) as a conceptual framework for farming operations responds to the need to manage inter-field and intra-field variability on farms, within watersheds, regionally and internationally. How PA is used, the objectives involved, and the technologies that support it have changed substantially since the inception of modern PA in the 1980s when the U.S. Global Positioning System (GPS) became available for public use. Coupled with geographical information system (GIS) computer technologies that were first developed for satellite imagery, PA became a mainstream tool for farmers to plan site-specific agricultural operations, early on including fertilizer application, followed by seeding rate, seed variety, pesticide spraying and now site-specific irrigation. Equipment with GPS steering and position-aware supervisory control systems allowed pre-determined site-specific prescription maps to be downloaded into equipment and used, for example, to turn off a spraying system as it passed over a waterway. GPS-enabled harvesting equipment produced yield maps that were some of the first data to be used for site-specific management, often with confusing results due to a lack of co-varying field data and adequate decision support systems (DSS)based on how soil spatiotemporal properties influence plant development. This kind of passive and indirect PA has evolved, however, to provide more capable solutions that, for example, provide for variable rate application of fertilizers based on georeferenced soil sampling that leads to prescription maps of fertilizer need. Or for another example, spatially variable irrigation management based on 30-m resolution maps of crop water use based on multi-satellite sensor fusion. Many of the more successfulPA technologies involve on-board sensor systems that feed data to embedded computing platforms that make on-the-fly adjustments to equipment. Such active and direct PA systems use modern technology that provides the ability, for instance, to turn spray equipment on in the presence of weeds and off otherwise, or to turn on variable rate irrigation nozzles where abiotic stress sensors indicate crop water stress. Such supervisory control and data acquisition (SCADA) systems rely on algorithms based on sophisticated understanding of biophysics and biological systems. Today the confluence of computing power, data acquisition and management infrastructure, new modeling paradigms, and spatial decision support systems ushers in new possibilities for PA. Providers of PA services now include government institutions from national to local levels, private providers (often using publically available data from government ground, aerial and satellite sensing systems), university extension systems and farmer cooperatives. Sources of data range from public domain to private data held by farmers or third parties. Questions around data standards, data sharing, data ownership, and public and private rights add further complexity to modern PA, but are actively being addressed by both public and private institutions.
展开▼
