Title: Real-time Sensor Systems for Fertility Management. Author
Submitted to: Cotton Incorporated Crop Management Seminar Proceedings
Publication Type: Abstract Only
Publication Acceptance Date: November 6, 2008
Publication Date: November 24, 2008
Citation: Vories, E.D. 2008. Real-time Sensor Systems for Fertility Management [abstract]. Crop Incorporated Crop Management Seminar & Workshops, Cary, NC. 2 pp. 2008 [unpaginated CD-ROM] Technical Abstract: As production costs continue to increase, farmers are always looking for additional ways to improve the efficiency of their operations. For many years we used field-wide recommendations for supplemental nutrient applications, even though we knew that some areas would receive more than the optimum amount while others got less. Higher costs for the materials along with environmental concerns for the runoff from over-fertilized areas have combined to increase the emphasis on knowing exactly how much is needed and where. Since we know that the distribution of the nutrients is variable, then variable rate application (VRA) is the best way to correct deficiencies. Grid sampling and remote sensing have sometimes been used as the basis for VRA, but currently much of the work is focused on using in-field sensors to measure nutrient status. Agricultural sensors are broadly divided into two types, sensing either the soil or the plant. Many different sensors are available or being developed for sensing soil properties. Soil pH sensor systems are currently commercially available from Veris Technologies (mention of trade names or commercial products is solely for purpose of providing specific information and does not imply recommendation or endorsement by the U.S. Department of Agriculture). Systems for measuring nutrients are available for laboratory use, but require much modification to be rugged enough to collect large amounts of data in agricultural applications. Most of the current research in plant sensor systems for nutrient management is focused on nitrogen, and the most commonly used sensors are based on spectral reflectance from the crop canopy. Since these sensors don't have to contact the plants, they are well suited for real-time management, where the measurements, the application rate determination, and the variable-rate application all occur in the same trip through the field. Sensors from NTech Industries (GreenSeeker) and Holland Scientific (Crop Circle) are probably the most common reflectance-based sensors. CROPSCAN, Inc. and Yara UK Ltd. also make reflectance-based sensors that are less commonly used, and Texas A&M University and others are developing their own sensors. These reflectance sensors have been used with wheat and corn for several years. While the corn system appears to work adequately with one sensor above the plant row, many researchers feel that additional information will be required to accurately predict the nitrogen needs of cotton and avoid the adverse growth effects of excessive nitrogen. One approach has been to place an ultrasonic sensor above the row to determine plant height. Another approach is to place an additional reflectance sensor between two rows. Testing will show whether one or both of these methods will provide the information needed or some other approach will be necessary. To summarize, a great deal of research is continuing on sensor systems for nutrient management in cotton, with much of it focusing on reflectance-based systems. Cotton Incorporated is encouraging the different research teams to work together as much as possible to develop systems that farmers can use. On-farm field-scale demonstrations are being conducted and a Conservation Innovation Grant from USDA-NRCS will allow more such demonstrations. It is reasonable to expect that an effective, reliable, real-time sensor system for cotton nitrogen management will be available soon, with systems for other nutrients to follow.