Brief Summary of Minutes of Annual Meeting

----- Arizona ----- EXTENSION & EDUCATION Presentation. Cotton late-season meetings: Spectral sensing for cotton defoliation management. Yuma-Parker AZ (7/15-16/2014). Early-season meetings: Creating prescription files from yield monitor, field sampling and images. Safford AZ 3/4/2015. Tent talk: Implementing variable-rate technology: Hardware selection. Avondale, AZ, 7/8/2015. Presentation. Delivered field demonstrations on sensor-based management and information-based application technology for delegations visiting Arizona including: Chapingo University irrigation (11/5/2014) and agricultural machinery (2/9/2015); Arizona Department of Environmental Quality (8/28/2014 and 2/20/2015). Workshop: Use of sensor technology for farm production. Innovations in Agriculture series, Arizona Geographic Alliance – University of Arizona. 4/11/2015 Maricopa County Extension Office. Workshop: Using the Case-IH AFS Pro-600 display to record cotton yield data. On-line webinar Plant Management Network. September 2014 (http://www.plantmanagementnetwork.org/edcenter/seminars/cotton/AFSPro600Monitor/) Workshop in field-based high throughput phenotyping. NSF-funded workshop that targeted graduate students and early-career plant breeders in the use of electronic equipment such as plant sensors, GPS, data-loggers, and data post-processing and analysis. Maricopa AZ 3/16-19/2015 Community outreach. Demonstration of new tractor technology w/ auto-pilot. Maricopa Farm Day 10/25/2014. Short course in machine systems for pecan production. New Mexico State University. Las Cruces, NM. 10/20-22/2014 Popular Press Article: Cary Blake. Precision agriculture opens research doors in pecan, other crops. Western Farm Press. Agricultural Technology and Irrigation. April 7, 2015 RESEARCH Sensor-based management of mid-season N fertilizer in durum wheat. Pedro Andrade-Sanchez and Michael Ottman. Research funded by Arizona Grain Research and Promotion Council. Development of economically viable variable rate P application protocols for desert vegetable production systems. 2013-2015. Pedro Andrade-Sanchez and Charles Sanchez. California Department of Food and Agriculture. Fertilizer Research and Education Program. Data harmonization and phenomics for crop diseases under heat and drought. Special Cooperative Agreement- USDA-ARS. Pedro Andrade-Sanchez and Jeff White. ----- Florida ----- Using starch accumulation on a Huanglongbing (HLB) or citrus greening infected citrus leaf, a new prototype sensor was developed based on machine vision to identify the HLB symptomatic leaves. The sensor consisted of a sensitive monochrome camera and a narrow band polarized illumination system, and was able to detect starch accumulation in HLB infected citrus leaves and differentiate it from leaves having other stresses such as nutrient deficiency. The sensor was tested in actual citrus grove conditions with healthy, HLB symptomatic, and zinc deficient leaf samples. A classifier was developed using simple statistical histogram features and detection accuracies of over 95% were obtained. This sensor system performed better than our first prototype, and increased detection accuracies with much simpler classifiers. Additionally, the effect of grinding citrus leaves was examined toward the disease detection. Four different citrus leaf classes were compared before and after being ground. The results showed that the freeze-dried ground leaves showed better separation among different classes. Due to the HLB disease, excessive premature citrus fruit drops may occur up to 25% of the entire production each year in Florida. To objectively and accurately estimate the amount of premature fruit drop at different locations, an in-field machine vision system was developed using a hardware system for automatic image acquisition, and an image processing algorithm using thresholding and K-mean clustering. A random forest classifier was used to remove background objects. The result showed that correct detection accuracies of 92% and 62% were obtained for recently dropped citrus and decayed fruit, respectively. An in-field spatial variability map of the estimated dropped fruit was created to help growers to identify potential problems of the fruit drop. An image processing algorithm was developed to forecast the number of immature green citrus and its size prior to harvesting. As a first step, a linear color space, OHTA, was used to segment green citrus from the background. Then, color, shape and texture features of the different objects (immature green citrus, leaves, soil, sky, twigs and branches) were used for identifying green citrus and background. A support vector machine (SVM) classifier was developed using gray gradient co-occurrence matrix and Tamura texture features. Shape features were used as a final step to detect citrus fruit. The method yielded an accuracy of 82% correct identification of immature green citrus fruit. Another machine vision system on a conveyor belt was developed to inspect mechanically harvested citrus fruit. Color RGB images were used using object based classification. Three ensemble learning classifiers, AdaBoost, bagging and random forest, were developed using 74 features including color histogram, textures and histogram intersection with immature and mature citrus color model. The three classifiers showed good classification accuracy for mature citrus with a minimum of 97%. Among them, the bagging trees showed the highest accuracies of 92, 89, 97, and 85% for green immature, intermediate, mature and diseased citrus, respectively. Toward the development of a citrus yield mapping, a new approach was proposed to identify individual mature citrus from clustered overlapping citrus using a marker controlled watershed method. Seed markers were obtained by morphological filtering and intensity based region growing. Compared to the traditional watershed algorithm, marker controlled watershed did not yield over-segmentation. Then Hough circle detection was used to identify each separated fruit and count them. This proposed method yielded a correct detection accuracy of 86% for a validation set of images. A novel laser weeding system was designed and built towards elimination of in-row weeds. The system consisted of machine vision/image processing, peripheral operations and central processing subsystems. A webcam was used to acquire images at regular intervals of travel distance. A binary image was created using OpenCV identifying only the vegetation in the picture. The system was able to differentiate plants (crops, weeds) from soil background. If the vegetation occupied considerable area in small grids in an image, a decision was made to fire laser on weed. A prototype was built and tested for its functionality as well as its real time performance. ----- Kansas ----- Use of sUAVS in agricultural applications: Research projects are being conducted in Manhattan (KS) and across the state. Faculty involved with UAVS research projects are related to the Agronomy Department, Agricultural Biological Engineering (ABE), Veterinary, and Agricultural Economics at the Kansas State University. A PrecisionAg team was created in the last year, the main focus of the group is to facilitate the development and utilization of new technologies on all farming operations. Extension Education for Ag Professionals: A total of 20 meetings were provided on the uses of sUAVS in agriculture since the last August 2014 (more than a year). This topic continues to be in high-demand for our clientele, participation (presentations) and demonstrations on field days and other Extension meetings is requested every month from our group in order to provide some insights on how sUAVS can assist farmers and key-stakeholders in developing support tools. Participation: Providing training to farmers, extension Ag agents, crop consultants in the use and agronomic applications of sUAVS technology. Research: Measurements in corn, soybean involved estimation of plant height, biomass, canopy cover, stand counts, leaf area index, and number of green leaves. For corn, determination of early stand count is also pursued comparing “ground truthing” (stand counts at multiple growth stages) data with the imagery collected from the sUAVS. Similar approach has been taken for measuring plant height and biomass (using site-specific positioning). Uniformity of corn canopy is also estimated via imagery collection, in calibration with biomass samples across the entire cornfield (approx. 3 acres). Measurements related to thermal imagery projects are performed around the state in farming settings, but also in greenhouse and growth chambers (for calibration purposes). The latter technique was developed by Dr. Sharda (ABE Department) and it is currently calibrated under field conditions (different hybrids) on several on-farm fields. Overall, the past year was primarily dedicated to evaluate examples on the use of sUAVS together with the beginning of preparation of support decision tools for implementing those in the farming decision-making process. Still, there is a lot to be done on our side to determine the “true” potential of this new technology within the precision agriculture discipline. Theoretical to pragmatic projects are currently implemented with the final goal of developing sUAVS technologies that can assisting producers, crop advisors, and other agri-business professionals for facilitating the decision-making process. ----- Missouri ----- EXTENSION & EDUCATION The University of Missouri has developed a Precision Agriculture Certificate program that is offered to full time MU students and through extension programming to producers and consultants. The certificate requires successful completion of four courses: 1) Precision Agriculture Science and Technology; 2) Machinery Management Using Precision Agriculture Technology; 3) Data Management and Analysis Using Precision Agriculture Technology; and 4) Profit Strategies Using Precision Agriculture Technologies. The courses are taught using classroom, online, and lab experience methods. In addition, course #1 listed above is a core class for a precision agriculture emphasis for several MU degree programs, with average annual enrollment of about 35. RESEARCH - A completed journal manuscript describes how soil and landscape properties were used to develop higher-resolution crop management zones than those provided by traditional soil survey soil maps. Yield maps were used to validate these management zones. - Agricultural Policy/Environmental eXtender (APEX) simulation of annual crop yields at different landscape positions worked well for average yields, but the temporal variability of simulated yields in response to water availability (dry and wet years) was lower than measured. The reason for this discrepancy is being investigated. - A collaborative project with researchers at 8 Land Grant universities is in its second year. This project examines the performance of in-season corn nitrogen management tools, including canopy sensing methods over a wide range of soil and weather scenarios (16 sites per year). Data analysis and interpretation to evaluate these tools relative to yield, profitability, fertilizer use efficiency, and off-field nitrogen loss has begun, with first year’s results to be reported at the 2015 ASA-CSSA-SSSA annual meeting in November. - Yield and profitability assessment data from a precision agriculture system (PAS) research field located near Centralia, MO continues. The PAS plan takes advantage of targeted management that addresses both crop production and environmental issues. The plan includes no-till, cover crops, growing wheat instead of corn for field areas where depth to the argillic horizon is shallow, site-specific N for wheat and corn using canopy reflectance sensing, variable-rate P, K, and lime using intensively grid sampled data, and targeting of herbicides based on weed pressure. Yield has slightly improved for corn (5%) and soybean (9%) with PAS over pre-PAS management. Risk as measured by grid-cell year-to-year yield coefficient of variation decreased 57% when comparing where wheat replaced corn with PAS, but has remained unchanged for soybeans. This field and the PAS system are now part of the USDA-ARS Central Mississippi River Basin Long Term Agroecosystem Research site. New soil, air, and weather instrumentation has been installed and is being evaluated. - The combination of near-infrared reflectance spectroscopy, soil apparent electrical conductivity sensing, and soil strength sensing was used to estimate parameters important in soil quality. Initial results were inconclusive, and a larger, more diverse dataset is being assembled. ----- Nebraska ----- EDUCATION Site-Specific Crop Management: AGRO/MSYM/AGEN 431 This senior-level course combines agronomic and engineering/technology aspects of site-specific crop management. The course is 3 credit hours, offered fall semesters. Enrollment has steadily increased, from 28 students in 2012 to 67 students in 2015. Curriculum is adjusted on a yearly basis to include state-of-the-art technologies and analysis methods that reflect real-world applications to give students experience with challenges they will face as they enter the job marketplace. Curricula Development The University of Nebraska – Lincoln is part of a multi-state effort – “Precision Farming Workforce Development: Standards, Working Groups and Experimental Learning Curricula” working towards updating curriculum for using in precision agriculture courses. At this stage of the project Nebraska has been marginally involved; efforts have focused on evaluating educational needs from industry, and on assembling syllabi from precision ag courses nationally. In 2016 we will become more involved with development of draft curricular materials, and evaluation of those materials in AGRO/MSYM/AGEN 431, Site-Specific Crop Management. The project is coordinated by South Dakota State University. EXTENSION Precision Agriculture Data Management Workshops A recent survey of Nebraska producers indicated the need for more educational outreach regarding the collection, management, analysis, and usage of precision agriculture data. In response to this feedback, hands-on workshops were created where attendees utilize farm management software with actual precision agriculture datasets to learn concepts and skills for addressing these needs. In 2015, ten one-day workshops were held in various counties across the state. Nearly 140 producers, consultants, retailers, and other agricultural professionals attended the workshops. In addition, 36 other attendees have taken part in this extension activity in North Dakota and Kentucky in 2015. Based on post-workshop surveys received to date, 60% to 70% of attendees thought they would start utilizing practices learned or expand their current practices based on the workshops. A majority of survey respondents have indicated moderate to significant improvements in knowledge regarding agricultural data management. The program has been innovative in the methods of presenting complex information into useful practices that can be readily applied by producers in their own operations. While many field operations are overwhelmed with too much information, this program is unique in providing hands-on tools for using data to make better decisions and in this way, the program serves as a model for educational programming in the applied data science area. Impact will be large as it is leading toward an increase in producer adoption of precision agriculture based technologies and for them to apply this information directly for management decisions (rather than merely collecting data and images for archival purposes). Project SENSE In response to continuing expansion of areas in Nebraska with elevated groundwater nitrate levels, and observations that long-term trends for increasing fertilizer nitrogen use efficiency (NUE) may be plateauing, an educational/on-farm research project was initiated in 2015 to encourage crop producers to increase their use of in-season nitrogen fertilization. Project SENSE (Sensors for Efficient Nitrogen Use and Stewardship of the Environment) is a joint effort of the University of Nebraska-Lincoln, the Nebraska Corn Board, five Natural Resources Districts (NRDs), USDA-NIFA, and producers participating in the Nebraska On-Farm Research Network. In 2015 a total of 17 sites were used on cooperating producers fields in which active crop canopy sensor-based in-season N fertilization occurred. Field days were held in each NRD in July and August. Yield data will be collected this fall and information shared during grower meetings this winter on the efficacy of sensor-based fertilization to increase NUE and profitability. A research component of the project with graduate student support will begin in 2016. The project will continue in 2016 and 2017. Nebraska Agricultural Technologies Association (NEATA) This organization is composed of crop producers, their advisors, input providers, and researchers related to site-specific crop management and other emerging agricultural technologies. The association holds an annual conference, in recent years with a pre-conference symposium on variable rate technologies. The last conference and symposium was held in Grand Island, NE, February 4-5, 2015. More information is available on the association website: http://neata.org/ RESEARCH Improving Irrigation Water and Energy Use Efficiency through Accurate Spatial and Temporal Management The adoption of variable-rate irrigation (VRI) technology has been slow across the state of Nebraska, primarily due to lack of information regarding cost-to-benefit ratios and knowledge for successful management of these systems throughout the growing season. This project sought to begin quantifying the potential benefits from adopting VRI with respect to utilizing stored root zone water holding capacity (RZWHC). This project has resulted in three significant accomplishments related to irrigation management using advanced technologies. Geospatial data layers including EC, OM, historical yield, etc. were combined with terrain analysis data (from field elevation) and used to relate these readily available data to RZWHC measurements from field monitoring sites. These data were then extrapolated to other field areas to determine spatial variability at the field scale. Georeferenced VRI pivot control scenarios were created in GIS software to quantify how commercially available sector or zone control systems could be used to address field AWHC variability and potentially improve irrigation water use efficiency. Using these data layers, dynamic irrigation control maps could potentially be generated based on any data (weather, soil moisture, crop growth) being collected. This project’s use of public data to estimate VRI irrigation savings for the majority of center pivots in Nebraska sets it apart from other VRI research was also seen as a potential breakthrough given that potential benefits of VRI have not been quantified beyond a small number of intensely studied fields and potential variability within the state of Nebraska can be assessed. Use of Sensing Systems to Detect Crop Stress Several projects related to nitrogen management use a variety of sensors to either evaluate treatment effects – such as fertilizer N rate or use of urease or nitrification inhibitors – on canopy N status of irrigated corn. The Holland Scientific RapidScan CS-45 is the most commonly used ground-based sensor for such projects, along with the SPAD 502 chlorophyll meter. Aerial sensing systems are used on unmanned aerial vehicles (UAVs) for evaluation in a number of projects as well, most commonly using a Tetracam MCA-6 with an integrated incoming radiometer for spectral correction. Several other projects have made use of various aerial sensors, either multispectral or RGB, to assess crop status. These include; use of RGB sensing on a bi-weekly basis in the spring and fall to evaluate buffalograss cultivar variation in green color duration; multispectral phenotyping of maize breeding lines for drought tolerance. Macro-scale Spatial and Temporal Distribution of Nutrient Pulses in Relation to Grazing Strategies This interdisciplinary project involves entomology, range science, and soil science perspectives in evaluating carbon and nitrogen cycling in grazed rangeland systems. Multispectral UAV imagery is being used to assess distribution patterns of dung pats as influenced by grazing management system. ----- Ohio ----- EDUCATION ASM 4580 – INTRODUCTION TO PRECISION AGRICULTURE was taught the fall of 2014. This senior-level, 3-credit hour course provides overview on precision ag technologies and practices allow students to 1) Identify the major terminology associated with each topic and be able to use those terms correctly when discussing material from the course, 2) demonstrate a familiarity with equipment and software used in the course, and 3) demonstrate knowledge of basic procedures discussed in lecture AND be able to apply that knowledge through hands-on laboratory skill activities. The Fall 2014 enrollment was 45 students. EXTENSION Precision Ag Online Certification Course – a series of 8 educational modules were developed and recorded covering various precision ag technologies. Professional students watch each module then complete an online test. The course is management through eXtension and support by the Alabama Cooperative Extension System. An Ohio State Precision Ag web presence was created to provide timely information and other sources of precision ag material. The team is developing a website (http://fabe.osu.edu/precisionag) along with a social media presence to disseminate general precision ag information and research findings. The Ohio State Precision Ag made over eighty five presentations focused on “Big Data”, precision ag services and new technologies. The team continues to speak and educate data topics. Events that team members help facilitate and speak at on data included, the Big Data Workshop: Managing Your Most Exclusive Farm Assets, Ames, IA (25AUG2014) and Big Data: Understanding and Leveraging The Most Elusive Farm Asset, Columbus, Ohio (16FEB2015). We also hosted the Top Farmers of Ohio group for a day-long overview of Precision Ag research at Ohio State (5AUG2014). RESEARCH Downforce By Seeding Depth: Row-unit Downforce (or Margin commonly termed) can significantly influence final seeding depth. Soil texture significantly affected final corn seeding depth with deepest occurring in Sandy Loam versus Silty Clay Loam and Silt Loam. Emergence timing and final live corn plant populations were influenced by final seeding depth and downforce. In general and across all sites, heavy downforce reduced the final live population and delayed emergence. Precision Seeding Meters Evaluation: Significant differences existed between the John Standard meter setup and the JD ProMax40 and Precision Planting eSet. The John Deere Standard corn seed plates and meter can be inconsistent at times under field conditions and can be sensitive to changes in seed size. No significant differences existed between John Deere ProMax 40 and Precision Planting eSet meter setups for corn. Testing indicated for best field results to run the vacuum gauge on the high side for the recommendations provided by the manufacturer: Row-Crop Planter Requirements to Support Variable-Rate Seeding of Maize: Results from this investigation indicated that final seeding depth of maize was impacted by both the planter depth setting and downforce applied on the gauge wheels. Final seeding depth did not equal the target depth for both Fields 1 and 2. Maize emergence was affected by both target planting depth and downforce in Fields 1 and 2. More variability in planting depth was measured at the 2.5-cm treatment compared to the 5.1-cm depth treatment. Final yield for both fields was most influenced by soil type which was expected since these fields had different yield potential for maize. In Field 3 where VRS was implemented, the time to make a rate change (e.g. response time) was less than 1.0 sec regardless of the magnitude in the rate change and ground speed. No trends existed for the time to make a rate change as ground speed and seeding rate varied indicating quick and consistent performance of the VRT used on this planter. However, a delay or lag was observed when a rate change occurred when crossing a management zone but varied depending up travel direction. The delay was 7.7 m when traveling East-to-West versus 3.8 m for West-to-East. Therefore, the correct planter and display setups must be used including defining the GPS location relative to the seed meter and entering the right look-ahead time within the display. Improper setup can impact final maize population and rate changes can initiate before or after the preferred MZ boundary. Significant differences were found between the two different metering technologies evaluated. The eSet meter setup provided a more consistent and better quality of seed metering in terms of singulation and seed spacing. Overall, the quality of seed metering degraded regardless of meter type at higher meter speeds (> 38 rpm) with this aspect not clearly indicated at times in the as-planted maps. The as-planted maps from the two commercial systems provided general representation of the planter population across the field but did not reflect the correct location of rate changes or did they take into consideration the actual planter performance when comparing to the final, emerged seed spacing. This study recommended that operators need to ensure the correct planter and display setups in order to achieve needed seed placement performance to support variable-rate seeding. In conclusion, implementing VRS in maize needs to consider the setup of the VR planter and technology to maintain desired seeding depth and final emergence while as-planted data must be improved and possibly include other parameters such as downforce and seeding depth. ----- Washington ----- Team at Washington State University (Manoj Karkee, Lav Khot and Qin Zhang) conducted several research projects and achieved substantial accomplishments in developing systems and technologies for precision and automated agriculture. Major projects and corresponding accomplishments are listed below. Unmanned Aerial Systems (UASs) for Mitigating Bird Damage in Blueberry Crops Every year, significant fruit yield loss is attributed to bird damage in WA and other parts of the country. The issue is particularly prevalent to Washington and Oregon vineyards but is also a critical issue for cherries and other fruits including blueberries and raspberries. Washington State grape, blueberry, cherry, and Honeycrisp apple farmers lose $80 million annually to bird damage. Netting, auditory scare devices, visual scare devices, chemical applications, and active methods such as trapping, falconry, and lethal shooting are the most common ways that bird control is practiced. However, netting is the only method viewed by most farmers as effective, which also is costly and lethal to a host of wildlife. In this work, we plan to investigate the efficacy of using fixed wing, quadrotor and/or other Unmanned Aerial Systems (UASs) to deter birds from vineyards. After showing that human-guided UASs can effectively deter birds, our longer term goal is to apply machine learning techniques to autonomously deter birds out of an area. In-field Sensing and Decision Support System to Prevent Cherry Fruit Cracking due to Rainwater Fruit cracking due to early summer rain remains the key concern for fresh market sweet cherry growers worldwide. Existing mechanical rainwater removal techniques (e.g. orchard sprayers or fans, aerial helicopters) are used by growers but there has been little systematic research on when and how much water needs to be removed from cherry canopies and the effectiveness of water removal. Through this research efforts, we have developed an in-field sensing to monitor real-time rainwater level of orchard canopies to assist grower decision making. Sensing system constitutes array of wetness sensors placed in canopies transmitting logged data in real-time to base station over wireless network. In year 2015, we plan to extensively test the robustness of the sensing system and develop decision rules through field studies. Field studies towards evaluating efficiency of orchard sprayer airblast, manned and unmanned helicopter downwash in rainwater removal will be conducted. 3D Machine Vision for Improved Apple Crop Load Estimation Accurate estimation of apple crop-load is essential for efficient orchard management. We continued field evaluation of an over the row platform to capture images from two side of apple canopies, that helped minimize the occlusions and improve the accuracy of crop-load estimation. A color camera, and a 3D camera were mounted in the sensor platform and were moved along rows of apple trees in two different commercial orchards of Allan Bros. Inc., Prosser, WA. Images were capture in both day and night times. A apple identification algorithm and a 3D mapping algorithm was used to count and estimate size of apples while avoiding duplicate counting of apples that were visible from both sides of tree canopies. Apple counting accuracy improved approximately by 20% when imaged from two sides compared to that with single-side imaging. Apple sizing accuracy was 89%. Human machine collaboration for automated harvesting of tree fruit The long-term goal of this work is to reduce dependency on human labor through mechanization and human-machine collaboration while increasing yields of premium quality fruit. The overall objective is to develop a framework for knowledge transfer and collaboration between human and machine. The team focused on understanding the dynamics of the hand picking of fruit, development of an effective end-effector based on the knowledge of hand picking, and a framework of hardware and software for optimal collaboration between human and machine for fruit identification. The machine vision system developed in this project achieved a fruit detection accuracy of 98%. The preliminary results of the first prototype of the integrated robotic harvester achieved a cycle time of ~7 sec per fruit. Shake and Catch Apple Harvesting Formally training fruiting wall architecture of modern apple orchards provides an opportunity to shake tree branches locally and capture fruit very close to where they are, potentially leading to a shake and catch harvesting system with acceptable fruit damage rate. In this project, a trans-disciplinary team of researcher is working on developing various types of shaking and catching mechanisms and understanding basics of energy transmission and fruit removal efficiency around tree canopies. Development and Optimization of Solid-Set Canopy Delivery Systems for Resource-Efficient, Ecologically Sustainable Apple and Cherry Production This was a multidisciplinary research and extension project to develop, evaluate, and deliver resource-efficient, innovative management technologies and tactics for apple and cherry production systems. The aim was to establish innovative delivery technologies for canopy and orchard floor inputs (including high efficiency irrigation systems, precision-activated micro-emitters, and reduced risk pesticides). The study showed that a solid set delivery system with ~1” diameter hose could be installed to ~400 ft rows with acceptable pressure loss. In a tall spindle apple orchard in Prosser, WA, the system achieved a similar or higher level of chemical coverage on upper-side of leaves but lower coverage on under-side of leaves compared to the coverage achieved by a conventional airblast sprayer. Mechanizing Red Raspberry Pruning and Tying System Cane management in red raspberry production is highly labor intensive. Labor availability is uncertain at best and labor cost is increasing. Currently, Washington growers estimate the pruning and tying cost in red-raspberry production to be from $500 to $800 per acre. In addition, labor is at risk for chronic and acute injury. Mechanization has the potential to substantially reduce labor use from cane management. In this project, we designed a few types of cane bundling and tying end effectors and developed a prototype. Field evaluation will be carried out in a red raspberry plot established in Prosser, WA though this project. Precision Canopy and Water Management of Specialty Crops through Sensor-Based Decision Making This project is a subcontract to a SCRI project with UC Davis as the leading institution. WSU team is contributing to nine different objectives and is playing critical roles in a few objectives. We have been refining the sensor system and perform the canopy PAR/shape assessment in tree fruit orchards; and we have been developing a research-grade sensing and mapping system to gather the data for each plant using multiple sensors to predict plant water status. WSU investigators have been leading the development of a visualized decision support system to meet the decision support needs of growers, university researchers; and have been involved in the development and implementation of site-specific application of water and fertilizer using autonomous units. Collaborating with external partners, we have also been conducting studies on assessing social impacts of developed innovative technologies through collecting, analyzing and summarizing collected data.

Chung, S., Sudduth, K.A., Drummond, S.T., Kitchen, N.R. 2014. Spatial variability of soil properties using nested variograms at multiple scales. Journal of Biosystems Engineering. 39(4):377-388. Sudduth, K.A., Kim, H.J., Motavalli, P.P. 2014. Soil. In: Moretto, L., Kalcher, K. editors. Environmental Analysis by Electrochemical Sensors and Biosensors, Vol. 1: Fundamentals. New York, NY: Springer. p. 23-61. Yang, W., Wikle, C.K., Holan, S.H., Myers, D.B., Sudduth, K.A. 2015. Bayesian analysis of spatially-dependent functional responses with spatially-dependent multi-dimensional functional predictors. Statistica Sinica. 25:205-223. Veum, K.S., Sudduth, K.A., Kremer, R.J., Kitchen, N.R. 2015. Estimating a soil quality index with VNIR reflectance spectroscopy. Soil Science Society of America Journal. 79:637-649. Wikle, C.K., Holan, S.H., Sudduth, K.A., Myers, D. 2014. Soil property estimation and design for agroecosystem management using hierarchical geospatial functional data models. Journal of the Indian Society of Agricultural Statistics. 68(2):203-216. Vories, E.D., Jones, A., Sudduth, K.A., Drummond, S.T., Benson, R. 2014. Sensing nitrogen requirements for irrigated and rainfed cotton. Applied Engineering in Agriculture. 30(5):707-716. Vories, E.D., Stevens, G., Sudduth, K.A., Drummond, S.T., Benson, R. 2015. Impact of soil variability on irrigated and rainfed cotton. Journal of Cotton Science. 19(1):1-14. Khedher Agha, M. K., W. S. Lee, R. A. Bucklin, A. A. Teixeira, and A. R. Blount. 2014. Sorption isotherms for triticale seed. Trans. ASABE 57(3): 901-904. Kurtulmus, F., W. S. Lee, and A. Vardar. 2014. Immature peach detection in colour images acquired in natural illumination conditions using statistical classifiers and neural network. Precision Agriculture 15: 57-79. http://dx.doi.org/10.1007/s11119-013-9323-8. Lee, W. S., and R. Ehsani. 2014. Sensing systems for precision agriculture in Florida. Computers and Electronics in Agriculture. doi:10.1016/j.compag.2014.11.005. Li, H., W. S. Lee, K. Wang, R. Ehsani, and C. Yang. 2014. ‘Extended spectral angle mapping (ESAM)’ for citrus greening disease detection using airborne hyperspectral imaging. Precision Agriculture 15: 162-183. http://dx.doi.org/10.1007/s11119-013-9325-6. Li, H., W. S. Lee, and K. Wang. 2014. Identifying blueberry fruit of different growth stages using natural outdoor color images. Computers and Electronics in Agriculture 106: 91-101. Ma, H., H. Ji, and W. S. Lee. 2014. Detection of citrus greening based on VIS-NIR spectroscopy and spectral feature analysis. Spectroscopy and spectral analysis 34(10): 2713-2718. Pourreza, A. W. S. Lee, E. Raveh, R. Ehsani, and E. Etxeberria. 2014. Citrus Huanglongbing detection using narrow band imaging and polarized illumination. Trans. ASABE 57(1): 259-272. Sengupta, S., and W. S. Lee. 2014. Identification and determination of the number of immature green citrus fruit under different ambient light conditions. Biosystems Engineering 117: 51-61. http://dx.doi.org/10.1016/j.biosystemseng.2013.07.007. Yang, C., W. S. Lee, and P. Gader. 2014. Hyperspectral band selection for detecting different blueberry fruit maturity stages. Computers and Electronics in Agriculture 109: 23-31. Pourreza, A., W. S. Lee, R. Ehsani, J. K. Schueller, and E. Raveh. 2015. An optimum method for real-time in-field detection of Huanglongbing disease using a vision sensor. Computers and Electronics in Agriculture 110: 221-232. Pourreza, A., W. S. Lee, E. Etxeberria, and A. Banerjee. 2015. An evaluation of a vision based sensor performance in Huanglongbing disease identification. Biosystems Engineering 130: 13-22. Kweon, G., E. D. Lund, C. Maxton, W. S. Lee, and D. B. Mengel. 2015. Comparison of soil phosphorus measurements. Trans ASABE 58(2): 405-414 Li, X., W. S. Lee, M. Li, R. Ehsani, A. R. Mishra, C. Yang, and R. L. Mangan. 2015. Feasibility study on Huanglongbing (citrus greening) detection based on WorldView-2 satellite imagery. Biosystems Engineering 132: 28-38. Choi, D., W. S. Lee, C. Yang, R. Ehsani, and F. Roka. 2014. Post-harvest quality evaluation system on conveyor belt for mechanically harvested citrus. 12th International Conference on Precision Agriculture. July 20-23, 2014, Hyatt Regency, Sacramento, California. Pourreza, A., and W. S. Lee. 2014. Effect of starch accumulation in Huanglongbing symptomatic leaves on reflecting polarized light. 12th International Conference on Precision Agriculture. July 20-23, 2014, Hyatt Regency, Sacramento, California. Chen, Y., and W. S. Lee. 2014. Identification and Determination of the Number of Overlapping Citrus Fruit in Natural Outdoor Conditions. ASABE Paper No. 1903766. St. Joseph, Mich.: ASABE. Choi, D., W. S. Lee, R. Ehsani, and F. Roka. 2014. Estimation of Prematurely Dropped Citrus Count on the Ground before Harvesting. ASABE Paper No. 1892107. St. Joseph, Mich.: ASABE. Ni, Z., T. F. Burks, and W. S. Lee. 2014. 3D Reconstruction of Plant/Tree Canopy from Multiple Views Using Machine Vision. ASABE Paper No. 1893190. St. Joseph, Mich.: ASABE. Pourreza, A., W. S. Lee, and R. Ehsani. 2014. A vision based sensor for Huanglongbing disease detection under a simulated field condition. ASABE Paper No. 1900251. St. Joseph, Mich.: ASABE. Pourreza, A., and W. S. Lee, and E. Etxeberria. 2014. Rapid in-field diagnosis of Huanglongbing disease using computer vision. Paper presented at the 127th Florida State Horticultural Society Annual Meeting, Clearwater, Florida, USA, June 1-3, 2014. Vyasaraja, S. and W. S. Lee. 2014. Laser weeding system for elimination of in-row weeds: new hardware system. The 5th International Workshop: Applications of Computer Image Analysis and Spectroscopy in Agriculture to be held on July 12-13, 2014 in Montreal, Quebec, Canada. Zhao, C., W. S. Lee, and D. He. 2014. Immature green citrus detection by multiple features using machine vision. ASABE Paper No. 1911329. St. Joseph, Mich.: ASABE. Marx, S.E.1, J.D. Luck, R.M. Hoy, S.K. Pitla, M.J. Darr, and E. Blankenship. Validation of machine CAN Bus J1939 fuel rate accuracy using Nebraska Tractor Test Laboratory fuel rate data. Computers Electronics Agric. (accepted, in press). Luck, J.D., J.P. Fulton, and J. Rees. Hands-on precision agriculture data management workshops for producers and industry professionals: Development and Assessment. J. Extension. 53(4). DOI: http://www.joe.org/joe/2015august/tt10.php Sama, M.P., J.D. Luck, and T.S. Stombaugh. 2015. Scalable control architecture for variable-rate turn compensation. App. Eng. Agric. 31(3): 425-435. Pitla, S.K., L. Nin, J.D. Luck, and S.A. Shearer. 2015. Use of controller area network (CAN) data to determine field efficiencies of agricultural machinery. App. Eng. Agric. 30(6): 829-839. Ferguson, Richard B. 2015. Groundwater quality and nitrogen use efficiency in Nebraska’s Central Platte River Valley. Journal of Environmental Quality doi:10.2134/jeq2014.02.0085. Krienke, B., R. Ferguson, B. Maharjan. 2015. Using an unmanned aerial vehicle to evaluate nitrogen availability and distance effect with an active crop canopy sensor. In J. Stafford (ed.): Precision Agriculture ’15, Proceedings of the 10th European Conference on Precision Agriculture, Tel Aviv, Israel, 12-16 July 2015. Pan, L., Adamchuk, V.I., Ferguson, R.B., Dutilleul, P.R.L. and Prasher, S.O. 2014. Analysis of Water Stress Prediction Quality as Influenced by the Number and Placement of Temporal Soil-Water Monitoring Sites. Journal of Water Resource and Protection, 6:961-971. Luck, J.D. and J.P. Fulton. 2014. Best Management Practices for Collecting Accurate Yield Data and Avoiding Errors during Harvest. Institute of Agriculture and Natural Resources, University of Nebraska-Lincoln, EC-2004. Luck, J.D., J.P. Fulton, and N.M. Mueller. 2015. Improving Yield Map Quality by Reducing Errors through Yield Data File Post-Processing. Institute of Agriculture and Natural Resources, University of Nebraska-Lincoln, EC-2005. Luck, J.D. 2014. Understanding the Benefits and Limitations of Current Pesticide Application Field Equipment. In Proceedings of the 2014 University of Nebraska-Lincoln Crop Production Clinics. Amatya, S., M. Karkee, A. K. Alva, P. A. Larbi, and B. Adhikari. 2014. Hyperspectral Imaging for Detecting Water Stress in Potatoes. GSTF Journal on Agricultural Engineering, 1(1): 52-61. Ampatzidis, Y.G., S.G. Vougioukas, M.D. Whiting, and Q. Zhang. 2014. Applying the machine repair model to improve efficiency of harvesting fruit. Biosystems Engineering, 120: 25-33. Karkee, M., B. Adhikari, S. Amatya, and Q. Zhang. 2014. Identification of Pruning Branches in Tall Spindle Apple Trees for Automated Pruning. Computers and Electronics in Agriculture, 103:127-135. Larbi, P. A., and M. Karkee. 2014. Effects of Orchard Characteristics and Operator Performance on Harvesting Rate of a Mechanical Sweet Cherry Harvester. GSTF Journal on Agricultural Engineering, 1(1): 1-11. Sharda, A., M. Karkee, Q. Zhang, J. Brunner, I. Ewlanow, and U. Adameit. 2014. Effect of emitter type and mounting configuration on spray coverage for Solid Set Canopy Delivery Systems. Computers and Electronics in Agriculture, 112: 184-192. Silwal, A., A. Gongal, and M. Karkee. 2014. Apple identification in field environment with over the row machine vision system. Agricultural Engineering International: CIGR Journal, 16(4): 66-75. Zhou, J., L. He, Q. Zhang, and M. Karkee. 2014. Effect of Excitation Position of a Handheld Shaker on Fruit Removal Efficiency and Damage in Mechanical Harvesting of Sweet Cherry. Biosystems Engineering, 125:36-44. Gongal, A., S. Amatya, M. Karkee*, Q. Zhang, and K. Lewis. 2014. Identification of Repetitive Apples for Improved Crop-Load Estimation with Dual-Side Imaging. Proceedings of The 19th World Congress of the International Federation of Automatic Control; 24-29 August 2014; Cape Town, South Africa. Ma, S., P. Scharf, M. Karkee, Q. Zhang, J. Tong, and L. Yu. 2014. Effects of Off-Track Errors of a Sugarcane Harvester On Stubble Height and Weight. Proceedings of the 6th Automation Technology for Off-road Equipment Conference (ATOE),15-19 September 2014, Beijing, China. Silwal, A., A. Gongal, and M. Karkee. 2014. Apple Identification In Field Environment With Over The Row Machine Vision System. Proceedings of the 6th Automation Technology for Off-road Equipment Conference (ATOE), 15-19 September 2014, Beijing, China. Sharda, A., D. Mangus, M. Karkee, and Q. Zhang. 2014. Effect Of Time Of Application On Spray Coverage Using Solid Set Canopy Delivery System. Proceedings of 12th International Conference on Precision Agriculture; July 20-23, Sacramento, CA. De Kleine, M.E., M. Karkee, K. Lewis, and Q. Zhang. 2014. A Dual Motor Actuator used to Detach Fruit by Shaking Limbs of Fruit Trees. Proceedings of 12th International Conference on Precision Agriculture; July 20-23, Sacramento, CA. Gongal, A., S. Amatya, and M. Karkee. 2014. Identification of Repetitive Apples for Improved Crop-Load Estimation with Dual-Side Imaging. ASABE Paper No.141888882. St. Joseph, Mich.: ASABE. Tong, J., Q. Zhang, M. Karkee, H. Jiang, and J. Zhou. 2014. Understanding the Dynamics of Hand Picking Patterns of Fresh Market Apples. ASABE Paper No.141898024. St. Joseph, Mich.: ASABE. De Kleine, M. E., M. Karkee, K. Lewis, and Q. Zhang. 2014. An End Effector Concept for Removing Fresh-Market Apples from a Tree Limb. ASABE Paper No.141906284. St. Joseph, Mich.: ASABE. Larbi, P. A., M. Karkee, S. Amatya, Q. Zhang, and M. D. Whiting. 2014. Field Evaluation of a Modified Mechanical Sweet Cherry Harvester. ASABE Paper No.141896871. St. Joseph, Mich.: ASABE. Sharda, A., J.P. Fulton, and T.P. McDonald. 2015. Impact of control system response characteristics on nozzle flow stabilization during simulated field scenarios. Computers and Electronics in Agriculture. 112(2015): 139-148. (doi: http://dx.doi.org/10.1016/j.compag.2014.11.001). Torino, M.S., B.V. Ortiz, J.P. Fulton, K.S. Balkcom, and C.W. Wood. 2014. Evaluation of vegetation indices for early assessment of corn status and yield potential in Southeastern U.S. Agronomy Journal. 106(4): 1389-1401. Virk, S.S., A. Poncet, J.P. Fulton, K.S. Balkcom, B. Ortiz, T.P. McDonald, and G.L. Pate. 2014. Row-Crop Planter Requirements to Support Variable-rate Seeding of Maize. In Proceedings of the 12th International Conference on Precision Agriculture, Sacramento, CA, July 20-23. Virk, S.S., J.P. Fulton, T.P. McDonald, K.S. Balkcom, A. Poncet, and A.B. Brooke. 2014. Impact of Ground Speed and Varying Seeding Rates on Meter Performance. ASABE Paper No. 141897985. St. Joseph, Mich.: ASABE. Virk, S.S., J.P. Fulton, T.P. McDonald, K.S. Balkcom, A. Poncet, and A.B. Brooke. 2014. Current Performance of Planter Technology to Support Variable-rate Seeding in the Southern US. ASABE Paper No. 141898019. St. Joseph, Mich.: ASABE.