Submitted to: Transactions of the ASAE
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 8/19/1999
Publication Date: N/A
Interpretive Summary: With the recent availability of yield sensors and positioning technologies, producers, scientists and agri-businesses are currently investigating and adopting the practice of yield mapping of agricultural crops. The creation and interpretation of yield maps is a complex process, and the measurement of the actual harvested area per unit time is an important component in the ecreation of accurate, useful yield maps. For row crops, such as corn, the swath width of crop entering the combine is discrete, constant and easily measured. However, in drilled or broadcast crops, a more accurate and automated method of determining the actual combine harvest area is required. In this study, a vector method is developed to determine the actual combine harvest area at each time step of the harvest process from a global positioning system (GPS) trajectory. The method involves the creation of polygons representing the area covered by the header during each time step of the crop harvest. Each polygon is then processed in harvest order by subtracting, in the boolean sense, all previously processed polygons from the current one. The resulting polygon(s) represents the actual harvested area during that time step. The algorithm was coded in a geographic information system (GIS) and compared to a previously reported raster method. The proposed method produced excellent results, and had a number of distinct advantages over other methods of harvest area determination. This technique will benefit scientists, producers and agri-business by minimizing a major error source affecting the creation of accurate yield maps, and by providing additional mapping techniques which can aid in yield map interpretation.
Technical Abstract: A vector method for determining actual combine harvest area using combine position data was developed and implemented in a geographic information system (GIS). The method involves the creation of polygons representing the area covered by the header during each time step of the crop harvest. Polygons are processed in harvest order by subtracting, in the boolean sense, all previously processed polygons from the current one. The resulting polygons represent the actual harvested areas during each time interval of harvest. The procedure was tested on a 10 ha drilled soybean crop in the fall of 1997, with data collected on 1 s intervals. Over the entire field, individual yield measurements would be in error by an average of 11% compared to the conventional assumption of a constant full swath width. Within different regions of the test field containing varying harvest pattern conditions, average error results ranged from 8 to 12%. The vector method was compared to a previously reported raster method for harvest area determination, and provided potential for improved accuracy at the cost of some increased computational complexity. Improvements to the algorithm were implemented to significantly reduce the complexity and processing time required. A useful byproduct of the proposed vector method was the ability to create classed polygon maps. These maps prove extremely useful in the visual interpretation of yield maps, since harvest patterns and actual harvested areas are apparent.