Source: UNIVERSITY OF FLORIDA submitted to NRP
MAINTAINING THE COMPETITIVENESS OF TREE FRUIT PRODUCTION THROUGH PRECISION AGRICULTURE
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
COMPLETE
Funding Source
Reporting Frequency
Annual
Accession No.
0190376
Grant No.
2001-52103-11323
Cumulative Award Amt.
(N/A)
Proposal No.
2001-04813
Multistate No.
(N/A)
Project Start Date
Sep 15, 2001
Project End Date
Sep 14, 2005
Grant Year
2001
Program Code
[(N/A)]- (N/A)
Recipient Organization
UNIVERSITY OF FLORIDA
G022 MCCARTY HALL
GAINESVILLE,FL 32611
Performing Department
AGRICULTURAL & BIOLOGICAL ENGINEERING
Non Technical Summary
Precision agriculture combines information from many sources to address the spatial and temporal variability of production systems. However, precision agriculture has not been adopted as rapidly for horticultural fruits and vegetables as for agronomic crops, but potential benefits may be even greater. We propose to develop precision agriculture for Florida citrus and tree fruit production in the Pacific NW that can be adapted to other horticultural crops at other locations. Yield mapping is a beginning point. Development of state-of-the-art sensing technology for tree status such as nutrient, disease, pest, and environmental stress is a key issue addressed in this proposal.
Animal Health Component
35%
Research Effort Categories
Basic
30%
Applied
35%
Developmental
35%
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
1020999202025%
1021119202025%
4020999202025%
4021119202025%
Goals / Objectives
Use yield mapping methods developed in Florida to improve citrus production, and adapt these methods to tree fruit production in the Pacific NW. Improve tree fruit production through the use of sensors that provide information on site characteristics, tree condition, and leaf density and identify biotic and non-biotic factors contributing to stress and limiting profits. Develop and use the tools of precision agriculture to improve nitrogen management of tree fruit production in order to optimize profits, to reduce energy consumption, to insure fruit quality, and to minimize environmental impact.
Project Methods
Techniques developed for citrus yield mapping in Florida will be adapted to tree fruit production in the Pacific NW. Harvesting of most tree fruits is a manual unit operation in which fruit is placed in a container adjacent to the tree and is subsequently loaded onto a truck and transported to a packinghouse or processing plant. The location of each container is recorded by GPS and the density of containers per unit area is the basis for the yield map. Automatic triggering to record the location is required to insure accurate results. Improved sensing methods are going to be used to identify and measure the most important factors and to rapidly identify sources of stress that can limit yield. The soils used for tree fruit production vary greatly and include very deep well drained soils as well as poorly drained soils with a high water table. Sensors including soil conductivity, physical and chemical measurements of soil and leaf samples, water table recording, and aerial photography will be evaluated individually and in combination. Productivity and fruit quality are influenced by light interception and distribution in the canopy. Laser distance sensor will be used to accurately establish canopy shape and volume, and provide information on leaf density and the leaf area index. These canopy characteristics will also provide the basis for prescription pesticide applications to obtain maximum spray efficacy. Ground based, aerial, and satellite imagery will be evaluated for rapid stress detection and identification of stress factors in tree fruit production. Elevation and soil electrical conductivity will be mapped in each orchard using high resolution RTK DGPS and either EM38 and/or Veris conductivity measuring systems. Soils in each experimental unit will be sampled for soil profile nitrates to a depth of 1.2 m in 0.3 m increments while trees are still dormant to establish baseline levels of soil N availability. Within an orchard, four rates of N fertilizer will be applied to induce a range of tree N status - 0, 0.5x, 1x and 2x, where x is the standard N recommendation for the given orchard. Four replications of this these treatments will be arranged in a randomized, complete block design. Selected trees in each N fertilizer plot will be monitored for canopy vigor and leaf tissue N content using standard measurement techniques at three times during the growing season - early, mid-season and pre-harvest. Airborne multi-spectral 1-m resolution images will be obtained at these times along with hand-held leaf chlorophyll measurements using a Minolta SPAD meter and a hand-held NDVI sensor.

Progress 09/15/01 to 09/14/05

Outputs
Existing and potential yield mapping technologies were researched. A bin location and count method was developed and demonstrated. Mapping unit size was studied. Tree fruit yield mapping technology was transferred between Florida, Washington, and Oregon and to clientele. An investigation of a machine vision approach for classifying citrus diseases was conducted on four classes of leaves. Various statistical approaches were tested and resulted accuracies over 86%. Analysis demonstrated that such methods can be used for classifying citrus diseased leaves under controlled laboratory lighting conditions. A multi-site, multi-sensor, multi-temporal study was performed to explore the capability of remote sensing to identify stress levels and factors in Florida citrus. Measures of tree physiological condition and yield were studied. Initial GIS based analyses indicated that the technology transfer from hyperspectral research mode to selected-narrowband multispectral operational mode may be feasible. A nitrogen sensing system for citrus was developed. Significant wavelengths were determined by statistical methods. A reflectance housing, a light source, detectors, filters and a data acquisition board were used. The developed calibration model had 75% unknown samples which had predicted N concentrations with less than 2 g/kg error. A study was conducted to identify a reliable system to characterize tree canopy structure for variable rate applications. The feasibility of using a light-measuring instrument to quantify leaf area index of tree canopies was explored. An ultrasonic and a laser scanning systems were developed that could quantify structure in field applications. The system was found to be capable of generating useful data for variable rate applications. A study was conducted to develop a new precision agriculture application, using ground conductivity measured with an electromagnetic soil profiler for the mapping of shallow water table depths in citrus groves in Florida. Calibrations were developed and successfully tested in five months and two sites, where water table depth accuracy ranged from 4 to 15 cm. Significant bias was however evident when comparing calibrations developed on different sites, so water table mapping should be done with local calibrations only. In Washington, apple bin mapping techniques were developed to assess the variability within field for fruit production, and relationships between yield and remote sensing estimates of canopy vigor were evaluated. No relationships were found between canopy vigor and fruit quantity/quality. Significant relationships were found between canopy vigor and fruit maturation. A nitrogen trial was performed for three years. Approximately 40% of the variability in the second harvest could be explained by the relationship between yield and N rate. Analysis of measurements from three handheld chlorophyll meters suggest that multiple measurements in an area could differentiate an area with insufficient or excessive N. The data suggest that use of the chlorophyll meters likely is better for within field differential management than for uniform N fertilizer adjustment.

Impacts
Tree fruit yield mapping was transferred to Washington and Oregon, and will contribute to precise tree management for better yield and profit. Identification of diseased leaves by machine vision technique will help develop an in-field disease detection system for better citrus management. The results from the GIS based remote sensing analysis will help develop a narrowband multispectral system for detecting tree stress levels. The nitrogen sensing system will make in-field nitrogen assessment more convenient and help determine nitrogen application rates at different locations. The ultrasonic and a laser scanning systems will contribute to fast and accurate characterization of tree structures for variable rate applications. The water table assessment using ground conductivity will help grower determine optimal placement of tile drains and develop best management practices for irrigation. The developed apple bin mapping techniques will provide an estimate of fruit yield and profit and help develop site-specific management of orchards. The nitrogen trial results will help develop precision nitrogen management techniques for pome fruit.

Publications

  • Schueller, J. K., S. M. Baker, W. S. Lee, C. L. Montague, E. J. Phlips, T. F. Burks, J. Jordan, J. W. Mishoe, M. Salyani, and A. W. Schumann. 2003. Development of precision agriculture sensing technologies for clam and citrus production. Werner, A., and A. Jarfe. Programe book of the joint conference of ECPA -- ECPLF. 4th European Conference on Precision Agriculture. Berlin. June 15-19. pp. 565-566.
  • Schueller, J. K. 2004. What is precision agriculture? Precision Agriculture for Florida Citrus. Lake Alfred, Florida. January 21. CD-ROM.
  • Schueller, J. K. 2005. Tree fruit hand harvest yield mapping: Information and opinions from Florida. Unpublished report to Washington State University and Oregon State University and their clientele.
  • Schumann, A. W., and R. Munoz-Carpena. 2002. A simple, self-contained canal stage recorder. Applied Eng. in Agric. 18(6): 691-696.
  • Schumann, A. W. Zaman, Q. U., and Hostler, K. H. 2004. Soil organic matter affects productivity of Florida citrus soils. American Society of Agronomy Annual Meeting Abstracts, November 2004. 1 p.
  • Schumann, A. W. and Zaman, Q. U. 2003. Mapping water table depth by electromagnetic induction. Applied Engineering in Agriculture 19 (6): 675-688.
  • Wei, J. and M. Salyani. 2004. Development of a laser scanner for measuring tree canopy characteristics, Phase 1: Prototype development. Transactions of the ASAE 47 (6): 2101-2107.
  • Wei, J. and M. Salyani. 2005. Development of a laser scanner for measuring tree canopy characteristics, Phase 2: Foliage density measurement. Transactions of the ASAE 48 (4): 1595-1601.
  • Zaman, Q. U. and M. Salyani. 2004. Effects of foliage density and ground speed on ultrasonic measurement of citrus tree volume. Applied Engineering in Agriculture 20(2): 173-178.
  • Pierce, F. J. 2004. Interviewed on October 15, 2004 for a documentary to be aired in December, 2004, by KIMA TV in Yakima, WA on The future of the apple industry. The reporter on this project is George Finch.
  • Pierce, F. J. 2004. Monitoring and Control for Frost Protection. Presented at the 7th International Conference on Precision Agriculture, July 27, 2004, Minneapolis, MN.
  • Pierce, F. J. 2004. Invited presentation The Role of Site-Specific Sensor Networks for Agriculture", at the Second AMS Users Conference, part of the Annual Meeting of the American Meteorological Society, Seattle WA, January 13, 2004.
  • Pydipati, R., T. F. Burks, and W. S. Lee. 2005. Statistical and neural network classifiers for citrus disease detection using machine vision. Trans. ASAE 48(5): 2007-2014.
  • Salyani, M. and J. Wei. 2005. Effect of travel speed on characterizing citrus canopy structure with a laser scanner. In: Precision Agriculture 2005, J. V. Stafford, Ed., pp 185-192.
  • Schueller, J. K., J. D. Whitney, T. A. Wheaton, and L. A. Balastreire. 2002. Accuracy and data characterization analysis in precision agriculture using MATLAB. ASAE Paper No. 0211044. ASAE Annual International Meeting. Chicago. July 28-31.
  • Corsi, W. and Pierce, F. J. 2004. Working session with fruit growers on the use of sensor networks in orchards, a presentation at the NeATA conference on Telemetry Networks for Agriculture Yakima, WA, July 13, 2004.
  • Davenport, J. R., E. M. Perry, N. S. Lang, and R. G. Stevens. 2003. Leaf spectral reflectance for non-destructive measurement of plant nutrient status. HortScience 38:743. (Published Abstract)
  • Davenport, J. R., E. M. Perry, N. S. Lang, and R. G. Stevens. Leaf spectral reflectance for non-destructive measurement of plant nutrient status. Invited workshop presentation at the 2003 American Society for Horticulture Science Annual Meeting, Providence, RI, 4 October 2003.
  • Jordan, J. D., and J. J. Schultz, 2003. Hyperspectral Remote Sensing in Citrus Precision Agriculture. Seven Hills Regional User Group (SHRUG) GIS Workshop, Tallahassee, FL, 19-21 Nov. 2003.
  • Jordan, J. D., and J. J. Schultz, 2003. Overview and Status of IFAFS Project, Task 3: Remote Sensing of Citrus Tree Stress Levels and Stress Factors. Precision Agriculture and Remote Sensing (PARS) seminar series, University of Florida, Gainesville, 26 Sept. 2003.
  • Jordan, J. D., and J. J. Schultz, 2003. Remote Sensing for Citrus Condition and Productivity-Hickory Branch Site. (Poster highlighting preliminary RS/GIS work at that site for benefit of grove manager)
  • Jordan, J. D., J. J. Schultz, and K. L. Huang, 2003. (Miscellaneous fieldwork photos used in UF Agricultural and Biological Engineering dept. brochures)
  • MacArthur, D. K., J. K. Schueller, and C. D. Crane. 2005. Remotely-piloted mini-helicopter imaging of citrus. ASAE Paper No. 051055. ASAE Annual International Meeting. Tampa. July 17-20.
  • Perry, E. M., D. Faubion, C. Seavert, U. Schulthess, J. R. Davenport, R. G. Stevens, J. Leal and Pierce, F. J.. 2003. Crop Traceability to Assess Spatial Variability of Crop Quality. ASA/CSSA/SSSA Abstracts.
  • Perry, E. M., R. Rupp, J. Davenport, J. Leal, Pierce, F. J., U. Schulthess, 2003. Crop traceability and remote sensing in tree fruit, in Proceedings of SPIE Vol. 5153 Ecosystems Dynamics, Agricultural Remote Sensing and Modeling, and Site-Specific Agriculture, edited by Wei Gao, David R. Shaw (SPIE, Bellingham, WA, 2003) 103-108.
  • Pierce, F. J. Presentation to the students in graduate course on Field Plant Pathology on precision agriculture on July 21, 2004.
  • Pierce, F. J. and W. Corsi. 2004. Sensor Network: AgWeatherNet and AgFrostNet. Presented at the Cascade Ag Services Recertification Meeting, December 17, 2004, Chelan, WA.
  • Pierce, F. J., E. M. Perry, T. V. Elliot, and G. R. Matthews. 2004. Telemetry Networks for Monitoring and Control in Agricultural Systems. Presented at the Symposium, Precision Monitoring for Agricultural Systems, held on September 22, 2004 at the International Conference on Monitoring Science and Technology, Denver, Colorado
  • Pierce, F. J., E. M. Perry, W. Corsi, T. V. Elliott, G. Matthews, L. Hartz. 2004. AgWeatherNet: A Network of Networks. Presented at the 100th Annual Meeting of the Washington State Horticultural Association, December 7, 2004, Yakima, WA.
  • Pierce, F. J. 2004. Real-time Monitoring and Control in Agriculture: Regional and On-Farm Networks. Presented at the 18th Annual Pacific Northwest Vegetable Association Conference and Trade Show, December 14, 2004.
  • Pierce, F. J. 2004. Networks for Washington. WSU Extension Agent Training, Dec 13, 2004, Irrigated Agriculture Research and Extension Center, Prosser, WA.
  • Pierce, F. J. 2004. Real-time Monitoring and Control in Agriculture: Regional and On-Farm Networks. Presented at the 18th Annual Pacific Northwest Vegetable Association Conference and Trade Show, December 14, 2004.
  • Pierce, F. J. 2004. Networks for Washington. WSU Extension Agent Training, Dec 13, 2004, Irrigated Agriculture Research and Extension Center, Prosser, WA.
  • Pierce, F. J. 2004. Field day presentation on sensor networks to the Washington State Commission on Pesticide Registration Annual Tour, July 20, 2004.
  • Pierce, F. J. 2004. Regional and On-Farm Telemetry Networks for Agriculture. Presented at the University of Georgia, On-Farm Wireless Networking, An Information Summit For Researchers and Industry, Tifton, GA, June 16 & 17, 2004
  • Min, M. and W. S. Lee. 2005. Determination of significant wavelengths and prediction of nitrogen content for citrus. Trans. ASAE 48(2): 455-461.
  • Min, M. and W. S. Lee. 2005. Hyperspectral nitrogen sensing system for citrus. ASAE Paper No. 051068. St. Joseph, Mich.: ASAE.
  • Min, M. and W. S. Lee. 2003. Spectral based nitrogen sensing for citrus. ASAE Paper No. 031137. St. Joseph, Mich.: ASAE.
  • Perry, E. M., J. Davenport, A. Antunez, and F. Pierce. 2005. Reflectance-based measures of leaf chlorophyll in a perennial crop, presented at the American Society of Agronomy Annual Meetings, Nov 6-10 2005, Salt Lake City, UT.
  • Perry, E. M., U. Schulthess, C. Seavert, D. Faubion, Pierce, F. J. 2004. Measuring Differences in Apple Yield, Quality and Vigor Within Orchard Blocks. Poster presented at the WA Hort Show December 7-8 2004 Yakima WA.


Progress 10/01/03 to 09/30/04

Outputs
The tasks related to develop yield mapping system for tree crops were completed. Throughout the project there was a sharing of information by phone calls and at meetings. Yield mapping technologies were discussed during visits to Florida, Washington, and Oregon. A short report was created and shared among the states. Progress has been made in remote sensing of citrus tree levels and stress factors. Multispectral and hyperspectral images have been acquired, processed, and analyzed. Another set of images were acquired in May 2004. A feasibility study was conducted to determine whether classification algorithms could be used to identify diseased and normal citrus leaves. Algorithms for feature extraction using color co-occurrence method and classification by image processing were designed. The analysis achieved classification accuracies of over 95%. Investigation has been continued to find important nitrogen absorption bands for citrus leaves. Some important wavelengths were identified for nitrogen prediction of citrus leaves. A nitrogen sensing system is being designed and fabricated. The development of a laser system for tree canopy measurement was continued. The system parameters were determined. A calculation algorithm for tree volume and foliage density was formulated to process tree images and generate continuous graphic output of tree height, tree volume, and foliage density on the go. The objectives of in-field sensing of shallow watertables with an electromagnetic induction (EMI) profiler have been achieved. Measurements by the EMI devices were found to be a rapid and accurate solution for the mapping of water tables in citrus groves. The results indicated that EMI instruments could be useful to delineate different soil series within a grove. An ultrasonic system was successfully implemented for measuring canopy volume and tree size in citrus groves. A fruit tracking system has been designed to support traceability. A radio modem unit was developed to monitor flow rate, pressure, and solenoid valves in a cherry orchard. New remote control functionality was developed for a 900 MHz frequency radio to improve line-of-sight coverage and to allow growers to control wind machines and irrigation systems for frost protection. Automated insect traps are being developed targeting some insects based on attract and kill principles. An irrigation system was designed to provide water via drip and sprinklers in support of deficit irrigation experiment. Acquisition of imagery was continued. Relationships between different N rates and apple quality factors were studied. Apple plant tissue samples were collected to assess plant N status. Measurements were also made on the leaves with hand held green-ness monitoring devices. Various correlations were found among C, N, and readings of the devices. A 2-ac block of danjou pears was harvested and were analyzed for pounds of total yield, pounds of culls, and distribution of fruit size. A regression analysis was conducted to find the relationships between yield and tree cross sectional area per tree. The results suggest growers to track pear yields during harvest and assess low yielding areas of the orchard.

Impacts
The outcome of this tree fruit project will enhance the competitiveness of tree fruit production through application of precision technologies. Different sensing technologies have been developed for tree fruit crop production to reduce management costs and increase yield and profit.

Publications

  • Jordan, J. D., and J. J. Schultz, 2004. Hyperspectral to Narrowband with Indexing and Fusion: A Remote Sensing Example in Florida Citrus. Presentation at Florida Section American Society of Agricultural Engineers conference, June 4, Stuart, FL.
  • Min, M., Lee, W.S., and Bogrekci, I. 2004. The effect of water and variety on nitrogen sensing of citrus leaf. ASAE Paper No. 041080. St. Joseph, Mich.: ASAE.
  • Schumann, A.W. and Zaman, Q.U. 2003. Using electromagnetic induction methods to map groundwater in Florida citrus soils. American Society of Agronomy Annual Meeting Abstracts, Denver, Co.
  • Schumann, A.W. 2003. New Technologies for Soil Profiling. 2nd Workshop of Precision Agriculture for Florida Citrus, UF/IFAS CREC, Lake Alfred, FL.
  • Zaman, Q.U. and Schumann, A.W. 2003. Spatial Variability of Soil Properties and Citrus Tree performance. American Society of Agronomy Annual Meeting Abstracts, Denver, Co.


Progress 10/01/02 to 10/01/03

Outputs
To develop yield mapping system for tree crops, qualitative comparisons made for different mapping techniques. Information was studied about traceability technologies and issues relevant to tree fruit crops. For DGPS field data acquisition system, a software interface for the machine vision system, and a GPS driver interface has been under development. To develop remote sensing techniques for citrus tree levels and stress factors, leaf sample reflectance analysis has been conducted and several vegetation indices have been developed. Yield maps and other information were obtained for all four sites. Algorithm for conversion of hyperspectral imagery raw digital numbers to spectral reflectance has been prepared. Leaves were collected from grapefruit trees infected with common diseases to determine if machine vision techniques can be used to identify leaf-based citrus diseases. The analysis of the Color Co-occurrence Method showed that the method was able to classify with an accuracy of 96%. Accuracies of 100% were achieved on leaves with intensity features. To find out the relationship between reflectance of citrus leaves and actual N content, citrus leaves were collected. The validation of the calibration models showed very promising results. Several important wavelength regions were identified. The evaluation of an ultrasonic system for tree volume measurement found that there was no significant difference between manual and ultrasonic measurements of the canopy volume. For a laser sensor measurement, the system hardware and software were completed, and evaluated the repeatability of the measurements of tree volume and canopy leaf density. Water table depths and Electromagnetic Induction (EMI) were measured. Correlations between the vertical dipole mode soil conductivity and measured water table depths were significant. The objective of mapping groundwater with EMI techniques was achieved and can be used for improving citrus grove management and profitability. In Washington, bin locations were mapped and yield maps were produced. NDVI values from the corresponding imagery were compared to yield. Work was begun on mapping apple quality by bin and traceability. Measurements were made on the trees to assess trunk diameter and visual bloom. New shoot growth differed by year and by N rate. Leaf samples were collected to assess plant N status. Measurements were made on leaves with three hand held green-ness monitoring devices. As N rate increased, there was an increase in fruit in color, indicating that fruit red color decreased with increasing N rate. In Oregon, it was found that as the price structure for the various sizes and grades for dAnjou pears come together in the marketplace, the importance of achieving packouts for individual pear trees diminish. One measure of assessing the vigor of a tree was its trunk cross sectional area (TCSA). Since canopy sizes were much more variable than N concentration, the amount of leaves on a tree had more impact on the amount of total N in the canopy than the N content in the leaves. As TCSA increased, so did the dollar returns on an individual tree basis.

Impacts
This tree fruit project will enhance the competitiveness of tree fruit production through application of precision technologies. While comprising two geographically separated industries with their own production specifics, both citrus and apple tree fruit crops have similar problems constraining their global competitiveness. This joint project on implementing precision agriculture into these two industries will provide the utilities and economies of shared expertise and provide outcomes useful nationally in the tree fruit industries.

Publications

  • Davenport, J. R., E. M. Perry, N. S. Lang, and R. G. Stevens. 2003. Leaf spectral reflectance for non-destructive measurement of plant nutrient status. Invited workshop presentation at the 2003 American Society for Horticulture Science Annual Meeting, Providence, RI, October 4, 2003.
  • Davenport, J. R., E. M. Perry, N. S. Lang, and R. G. Stevens. 2003. Leaf spectral reflectance for non-destructive measurement of plant nutrient status. HortScience 38:743.
  • Jordan, J. D., and J. J. Schultz. 2003. Hyperspectral remote sensing in citrus precision agriculture. Seven Hills Regional User Group (SHRUG) GIS Workshop, 19-21 Nov. 2003, Tallahassee, FL.
  • Jordan, J. D., and J. J. Schultz. 2003. Remote sensing for citrus condition and productivity-Hickory Branch Site. Special poster for the grove manager, to summarize preliminary remote sensing research at that site, Lake Placid, FL.
  • Min, M., and W. S. Lee. 2003. Spectral-based nitrogen sensing for citrus. ASAE Paper No. 031137. St. Joseph, Mich.: ASAE.
  • Lee, W. S., T. F. Burks, J. D. Jordan, J. K. Schueller, J. D. Whitney, M. Salyani, A. W. Schumann, T. A. Wheaton, W. M. Miller, F. J. Pierce, E. Perry, J. R. Davenport, R. G. Stevens, C. Seavert, T. Rigetti, and S. Castagnoli. 2002. Maintaining the Competitiveness of Tree Fruit Production through Precision Agriculture. Ag2020/IFAFS Annual Meeting, NASA Stennis Space Center, Mississippi, Dec. 4, 2002.
  • Perry, E. M., D. Faubion, C. Seavert, U. Schulthess, J. R. Davenport, R. G. Stevens, J. Leal and F. J. Pierce. 2003. Crop traceability to assess spatial variability of crop quality. ASA/CSSA/SSSA Abstracts.
  • Cugati, S., W. M. Miller, and J. K. Schueller. 2003. Automation concepts for the variable rate fertilizer applicator for tree farming. Presented at the 4th European Conference on Precision Agriculture, June 15-19, 2003, Berlin, Germany.
  • Perry, E. M., R. Rupp, J. Davenport, J. Leal, F. J. Pierce, and U. Schulthess. 2003. Crop traceability and remote sensing in tree fruit. In Proceedings of SPIE Vol. 5153 Ecosystems' Dynamics, Agricultural Remote Sensing and Modeling, and Site-Specific.
  • Schueller, J. K., S. M. Baker, W. S. Lee, C. L. Montague, E. J. Phlips, T. F. Burks, J. D. Jordan, J. W. Mishoe, M. Salyani, and A. W. Schumann. 2003. Development of precision agriculture sensing technologies for clam and citrus production. pp. 565-566. A. Werner and A. Jarfe, eds., Program Book of the Joint Conference of the 4th European Conference on Precision Agriculture and the 1st European Conference on Precision Livestock Farming (ECPA-ECPLF), June 15-19, 2003, Berlin, Germany, Wageningen Academic Publishers, The Netherlands.
  • Schumann, A. W. and Q. Zaman. 2003. Mapping water table depth by electromagnetic induction. Applied Engineering in Agriculture 19(6): 675-688.
  • Seavert, C. F. 2002. Assessing the economic benefits of technologies. Washington State Horticultural Association Annual Meeting, 98th Annual Meeting, pp. 47-51.
  • Seavert, C. F. 2002. An evaluation of the economic benefits of adopting site-specific management technologies in the tree fruit industry. Acta Horticulturae No. 596. Proceedings of the Eighth International Symposium on Pear, Volume 1. 2002. pp. 123-126.
  • Zaman, Q. U. and M. Salyani. 2003. Effects of foliage density and ground speed on ultrasonic measurement of citrus tree volume. ASAE Paper No. FL03-129. St. Joseph, Mich.: ASAE.


Progress 10/01/01 to 10/01/02

Outputs
In Florida, work related to fruit yield mapping was concentrated on the development of techniques to visually represent the yield. Example citrus yield data gathered with volumetric and weight measurements was manipulated and mapped. The development of a DGPS based field data acquisition system has been initiated. The project has begun with the selection and purchase of a DGPS system for collecting geo-spatial position of the monitoring system. Four citrus study sites were selected in different Florida County to develop remote sensing techniques for citrus tree levels and stress factors. Image files (IKONOS, RDACS, ATLAS, and hyperspectral) from May 2002 satellite/airborne data collection were delivered from NASA. In order to determine if machine vision techniques can be used to identify leaf-based citrus diseases, leaves were collected from grapefruit trees infected with common diseases. The Color Co-occurrence Method was used to discriminate diseases of citrus leaves and the image texture analysis was able to classify between the disease classes and healthy leaves with an accuracy of over 96%. One thousand leaf samples were obtained from citrus groves to develop nitrogen sensing techniques for citrus production. Their spectral reflectance was measured in 400-2498 nm with 2 nm increment. The relationship between spectral measurements and actual concentration is being studied to identify important absorption bands of nitrogen. A light measuring instrument was used to explore the utility of the photosynthetically active radiation to characterize tree canopy structure. The results showed that leaf area index was a function of solar radiation. An ultrasonic system was used to quantify tree canopy volume. Results showed a good repeatability in tree volume measurements. Two citrus groves were chosen to develop in-field sensing techniques of shallow watertables and hydromorphic soils with an electromagnetic induction profiler. Ground conductivity data were regressed against the watertable depth and the coefficient of determination appeared to be very good (0.845). In Washington, Fall 2002 was the first opportunity under this grant to yield map multiple-picked fruit. The location of the full bins was tagged with a unique ID number, linking the bin with the location of the bin. At multiple times during the picking period (the orchard was picked three times), an ArcGIS shapefile was created with the bin locations and ID numbers. A nitrogen rate study was established in an apple orchard for precise N management. In mid June, plots were evaluated for new growth and there was a significant relationship between N rate and new shoot length. In Oregon, yield, fruit quality, and economic returns have been determined for different locations within an orchard. It seemed that the large differences in tree size, N use efficiency, leaf efficiency and partitioning efficiency within the orchard would complicate the interpretation and utilization of leaf biomass and vigor estimates. Research is in progress to find ways to utilize remote sensing to help develop management plans in situations where soil limitations exist but obvious differences are not apparent.

Impacts
This tree fruit project will enhance the competitiveness of tree fruit production through application of precision technologies. While comprising two geographically separated industries with their own production specifics, both citrus and apple tree fruit crops have similar problems constraining their global competitiveness. This joint project on implementing precision agriculture into these two industries will provide the utilities and economies of shared expertise and provide outcomes useful nationally in the tree fruit industries.

Publications

  • Burks, T. F., Jordan, J. D., and Lee, W. S. 2002. Early detection of citrus tree diseases using machine vision. ASAE-CIGR Meeting Paper No. 021109. St. Joseph, Mich.: ASAE.


Progress 09/15/01 to 12/31/01

Outputs
This is a new project and there are no results reported yet.

Impacts
(N/A)

Publications

  • No publications reported this period