SUGARCANE YIELD MONITORING: A PROTOCOL FOR YIELD MAP INTERPOLATION AND KEY CONSIDERATIONS IN THE COLLECTION OF YIELD DATA

YIELD MONITORING IS a key element of Precision Agriculture (PA) given the importance of yield maps to the process of ‘management zone’ delineation and decisions around the targeting of inputs such as fertilisers. This paper presents a protocol for the interpolation of yield maps from yield monitor data and considers some critical issues in the data acquisition process relating to the calibration of the yield monitor to tonnages recorded at the mill. These strongly suggest that successful adoption of yield monitoring and mapping in the sugar industry, and thus of PA more broadly, will depend on significant improvements being made to the consignment system; electronic consignment presents as one possible solution to yield monitor calibration errors. In addition, yield mapping requires: geolocation with an accuracy of dGPS or better; yield monitor calibration on a per-harvest event basis, logging of yield data at 3 second intervals, or where more frequent logging is used, data thinning to an equivalent of 3 second logging; removal of errors and aberrant values from the yield monitor data based on harvester speeds of < 0.75 m/s and zero yields followed by cleaning of values that are more than three standard deviations from the mean yield; and map interpolation using local block kriging onto a 2 m grid, using exponential variograms, blocks of 10 m and a data cloud of 100 points.