Brief Summary of Minutes of Annual Meeting

NCERA-180 Summary of Accomplishments 2008-2009 Research NCERA-180 activities and the relationships formed through these activities have facilitated many research and extension accomplishments over the past year. The following is a brief summary of some of these activities as reported by participants from Arkansas, California, Florida, Kentucky, Michigan, Nebraska, North Carolina, and North Dakota. Various Projects in Arkansas: Researchers in Arkansas have been working on problems related to soil compaction, in-season site-specific determination of plant nutrient needs, and statistical site-specific evaluation techniques for grain and cotton crops. They are also evaluating the utility of variable-rate lime application for validating lime recommendation rates. Their research also extends beyond traditional grain and cotton production to the nursery industry where they are evaluating methods of inventorying field-grown shade trees. Planting-site-specific application in orchards. The goal of this research was to use recent advances in the global positioning system and computer technology to apply just the right amount of fumigant where it is most needed (i.e., in the neighborhood of each tree planting site or tree-planting-site-specific application) to decrease the incidence of replant disease, and achieve the environmental and economical benefits of reducing the application of these toxic chemicals. In this study we retrofitted a shank type chemical applicator with a high-performance global positioning system receiver (accuracy in the range of 10 to 20 cm), an embedded controller to read GPS data, a Pulse Width Module (PWM) and solenoid actuated nozzles to provide precision rate on demand. Extensive testing indicated that the new system had a RMS error of less than 15 cm. The system was field tested in three almond orchards in California during the Fall of 2007. The system performed well in all three locations. Precision Weed Control System for Processing Tomatoes. The goal of this research was to develop an automated system for control of weeds in the seedline of direct seeded processing tomato plants. The system is based upon two main components, a hyperspectral imaging system for weed species identification, and a precision thermal weed treatment system for targeted application of a lethal high temperature organic oil application. In 2008 significant progress was made toward the development of both the weed species identification system and the thermal weed control system. We successfully conducted an outdoor test of the system where weeds in the seedline were automatically detected, identified, and treated by the system. Test results show that 95.8% of black nightshade and 92.7% of pigweed plants automatically treated by the heated oil spray were dead 15 days post application. Only 2.4% of the tomato seedlings accidentally sprayed by the system were killed. Nearly all (97.6%) of the tomato plants in the seedline were alive and had a dry mass not significantly (a=0.05) less than that of the untreated control tomato plants at 15 days post treatment. Pistachio Yield Monitor: A yield monitoring system has been developed for studying the yield behavior of pistachio trees to improve orchard productivity, field management and to investigate the mechanisms that cause alternate bearing. Two pistachio yield monitoring system sets developed and fabricated by UCD were enhanced for the 2008 harvesting season. These sets were retrofitted on two pistachio catch-frame harvester sets that were capable of maintaining nearly commercial harvesting field capacity. An automated, DGPS based row identification system, has been designed and developed to map tree locations inside mature orchards. An algorithm was developed to use real-time sensed odometric information to localize the trees inside the orchard canopy. In 2008, approximately 20,000 tree yields obtained from 64 ha of a young and a mature orchard were recorded and have been analyzed separately by UCD and Paramount Farming Co, CA. Seven-year yield data of very carefully monitored 4288 trees have been analyzed for alternate bearing studies. The accuracy of the tree yield data independently measured by the grower and this analysis supports the successful development of the current design. Olive Yield Monitor: In Oct 2008 experimental data were obtained using a harvester provided by Dave Smith Engineering (DSE), a modified Korvan prototype, while working on Manzanillo cultivars at the Rocky Hill Ranch, CA. Field trials were conducted to evaluate our design specifications to improve machine performance on a 3-acre area of mature trees planted on a 11 x 26 ft spacing. To monitor machine efficiency and field capacity the DSE picking head olive harvester prototype was retro-fitted with a UCD Bio-automation Lab (BAL) custom built yield monitoring system (OYMO8) and an AgLeader Insight monitor unit. The OYMO8 logged the signal from a custom built bin scale mounted on the harvester. The Insight yield monitor was able to measure picking head speeds (rpm) and record GPS position data. The OYM08 successfully recorded selected variables: date, UTC time, latitude, longitude, ground speed (GPS) and yield. Results of 2008 olive harvest trials are being evaluated. Various Projects in Florida: Researchers in Florida contributed work related to NCERA-180 objectives. Under the leadership of Dr. Wonsuk Daniel Lee, research continues on yield mapping of unripe green citrus. A new NIR camera from the Electrophysics was tested for the feasibility of detecting green citrus. However, in field tests it did not reliably distinguish green citrus from green canopies. Daniel Lee, working with Reza Ehsani and others, developed a new machine vision system to detect the amount of trash (leaves and twigs) resulting from mechanical harvesting with a continuous canopy shake and catch harvester. The system consisted of two digital color cameras, four halogen lamps, a laptop, and a DGPS receiver. Due to the catastrophic citrus greening disease (Huanglongbing or HLB), citrus growers want to know how much trash is generated from mechanical harvesting, so that the growers could prepare how to remove the disease infected trash. Work on citrus interpretation from satellite or aerial imagery continued under the leadership of L. Gene Abrigo. Evaluation of NDVI changes showed the decline of citrus blocks affected with Sudden Death in Brazil, and then their recovery as trees were replanted over a 7 year monitoring period. NDVI values were higher at the end of the period than before the disease indicating the better condition due to a higher density and installation of irrigation in the replanted blocks. In another comparison using multispectral data from aerial imaging, 62% of the variation in yield was correlated with percentage of ground area covered by citrus trees and the NDVI of those trees for a group of 53 Valencia orange blocks. Arnold Schumann and Kevin Hostler redesigned a popular optical sensor-based controller used for applying fertilizer and pesticides in Florida citrus orchards. By adding substantial GPS speed-regulated, microcontroller-based look-ahead capacity, the canopy sensors could be mounted to the front of the tractor instead of to the applicator, thus allowing multiple applicator devices to share the same control system and reduce costs. A PocketPC-based GUI software interface with Bluetooth wireless connects to the controller for advanced setup and monitoring. The new control system is now being commercialized for the benefit of all interested tree crop growers. Reza Ehsani is beginning a new major USDA-funded project working on autonomous tractors with Carnegie Mellon University, Cornell University and John Deere entitled "Integrated Automation for Sustainable Specialty Crop Farming". The goal of this 3-year project is to deploy, operate, and evaluate a networked fleet of autonomous tractors performing variable rate fertilizer and spraying applications, yield estimation, mowing, plant status monitoring and scouting in a large south Florida commercial citrus groves. The project brings the value of autonomous tractors (24/7 operation and reduced labor) and precision agriculture (maximizing profit) together. It also involves an extension effort to transfer the technology to other fruit and vegetable crops. John K. Schueller is concentrating on promoting the international sharing of precision agriculture technologies, including chairing Section III: Equipment Engineering for Plant Production of the International Commission of Agricultural Engineering. Advanced Machine Control Technologies. Researchers in Kentucky have focused heavily on advanced machine control technologies to increase productivity. Autosteer and lightbar technologies have been shown to greatly increase field and input use efficiency and to facilitate new management strategies such as strip tillage and precise input placement. Automatic section control has been shown to reduce input usage as much as 15% in irregularly shaped fields. Standardized Testing of Precision Agriculture Technologies. Researchers at Kentucky and Nebraska continue to refine procedures and facilities for testing GNSS receivers, autosteer systems, and yield monitors. As a result of these activities, one ISO standard has been initiated and nearly completed, another ISO standard will begin the evaluation process this year, and one ASABE standard has been completed. Site-specific cover crop management. Cover crops generate important benefits to agro-ecosystems and their importance will further increase if above ground biomass is removed for biofuel production. Better understanding of how variations in topography and soil properties affect cover crops will contribute to increased cover crop adoption, allowing producers to tailor cover crop management to site-specific features of their fields. We are studying a range of issues that need to be addressed prior to site-specific cover crop implementation. The issues include tools available for quick non-destructive assessment of cover crop growth patterns and biomass production; analysis of soil and topographical factors that affect cover crops on an agricultural field scale; and assessment of spatial variability in the amounts of biomass and nutrient inputs contributed by cover crops to the subsequent main crop and how it affects variability in the main crop performance. We are working with 8 agricultural fields that are part of Scale-up portion of the Kellogg Biological Station Long Term Ecological Research site experiment located in southwest Michigan. The fields are in organically managed corn-soybean-wheat rotation with red clover cover crop used in wheat years. We collect cover crop biomass data several times during the season and prior to cover crop incorporation, monitor performance and yield of the following main crops, examine topographical and soil characteristics of the studied fields. A range of statistical and geostatistical techniques, including but not limited to multiple regression, random coefficient regressions, hierarchical mixed models, regression kriging and cokriging are used for data analyses. Preliminary results were reported at the 9th International Conference on Precision Agriculture (Denver, CO) and will be reported at the Joint International Agricultural Conference 2009 (Wageningen, Netherlands). Spatial Variability in Energy Crop Quality Components. A better understanding of how soil and landscape features affect energy crop quality component yield is necessary to maximize efficiency in bioenergy cropping systems. Bioenergy crop quality components including ethanol production in corn grain and methyl ester profile and oil content in biodiesel crops were evaluated. State-wide data (2006) indicate a trend for increasing corn grain ethanol quality component (g ethanol produced per g of grain) with increasing latitude. On a within field basis, corn grain quality component yield was found to vary as much as 13% across a 50 ha field. Corn grain ethanol quality component yield was positively correlated with corn grain yield, rotational soybean yield, and soil carbon. Conversely, corn grain ethanol quality component yield was inversely correlated with elevation. Soybean biodiesel methyl ester profile was also found to vary spatially within a 50 ha field. The cumulative levels of fully saturated methyl esters (palmitic 16:0 and stearic 18:0) was significantly correlated with soil conductivity measurements. Levels of other soybean biodiesel quality components including oleic methyl esters (18:1) and the cumulative unsaturated methyl esters (linoleic 18:2 and linolenic 18:3), varied across the field landscape but were not significantly correlated with field or crop yield parameters. Various Projects in North Dakota. Use of site-specific technologies, tools, and management continue to increase in North Dakota. The increased cost of inputs and crop prices this past growing season resulted in a surge of demand for site-specific services from providers. Site-specific based businesses in the region have hired additional people to help service the demand. A number of these businesses have recently banded together to form the Alliance of Site-Specific Providers. The group will serve to enhance the ability of the mostly entrepreneurial members to work together through improved interaction. Two major site-specific projects were completed in 2008; The NRCS terrain modeling project, and the summary of 40 years of grid sampling in two Illinois fields. The summaries of both projects are available on Dave Franzens web page. An attempt to secure funding for cooperative work between NDSU and several site-specific businesses in the state and region through a state Center of Excellence proposal for about $1,000,000 in total funding was not successful. Three site-specific circulars- What is Site-Specific Agriculture?; Zone Sampling; and Yield Mapping; were published in 2008 and are available on the web. These are major updates of their 1998 publication. Crop and Soil Management Systems for Water Quality Protection and Agricultural Sustainability. Development of the Crop Circle active sensor system and protocols for sensor use was finalized through a cooperative research development agreement (CRADA) with Holland Scientific. Holland Scientific has in turn signed an agreement with Ag Leader Technology (Ames, IA), a leading manufacturer of precision farming technologies and Ag Leader announced May 1, 2008 that it has become the exclusive distributor in 2009 for Holland Scientifics second generation active light reflectance sensor products for use in production agriculture. Our research results showed that sensor readings acquired during vegetative growth and expressed as chlorophyll index (CI), (CI= sensor NIR / sensor visible  1) would have the greatest potential for assessing canopy N content and directing spatially variable in-season N applications. An algorithm for converting sensor readings to variable N applications was developed and was evaluated in on-farm field studies. Preliminary results from field studies show a potential for savings in N application from 20 to 50% compared to traditional N management practices. Integrated Soil Sensing for Site-Specific Crop Management. The process of targeted (also called guided or smart) sampling proposed to enhance the use of on-the-go soil sensing data has been analyzed. An objective function that accounts for representing the entire range of sensor data, spreading across the field and local homogeneity was developed. Constrained categorical separation and Latin hypercube sampling were used to simultaneously address all established criteria for multiple data layers while prescribing a random set of targeted sampling locations. A MatLab program was written to implement the method developed. In the spatial clustering research, we were able to combine univariate observations into groups based on values of the multivariate normal likelihood. The spatial location of these observations is taken into account in the variance-covariance matrix in the likelihood itself. With involvement of partner external scientists and industry representatives, the concept for integrated sensor platform has been established. Substantial resources have been pulled together to invest in a new Veris Mobile Sensor Platform and Trimble RTK-level GPS/GLONASS equipment. Both pieces integrated with the instrumented tractor and other sensors developed at UNL will serve as a platform for continued research in this area. Substantial funding has been secured to implement sensor fusion approach to manage soil acidity in Eastern Nebraska. Several large datasets were obtained using optical crop canopy sensors and recently added ultrasonic distance sensor during summer months. The analysis of sensor density effect has been accepted for publication in a refereed journal. The analysis of potential for in-season N management that is based on soil maps obtained using on-the-go soil sensor and aerial imaging platforms (in addition to the optical crop canopy sensing) is in progress. Also, the relationship between measured corn height and detected chlorophyll level status is studied with respect to the potential improvement of in-season N-management algorithm developed. Once implemented, the method presented will allow users unbiased prescription of optimized targeted soil sampling schemes necessary to build data calibration models. Further, successful implementation of the new clustering approach will allow users delineation of field areas that require differentiated treatments and are based on several calibrated self-generated data sources. Extension NCERA-180 participants are working toward the development of an eXtension web site on site-specific management. Several meetings have already taken place and the project seems well on its way to fruition. This is an extensive effort involving many different researchers and industry personnel. In addition, participants have conducted extension workshops on GIS, utilizing digital imagery, profitability, precision agriculture, ecosystem management, automatic section control, precision guidance, variable rate application equipment, greenseeker technology, and the use of software tools for nutrient management and yield data manipulation. Several states have also developed grower groups to help facilitate organized on-farm research and educational activities.