STATEMENT OF THE PROBLEM:
Crops in the Southern United States are generally produced in fields which are known to have a high degree of variability in soil type, topography, soil moisture and other major factors which affect crop production. New technologies which can enable the development of an agricultural system to effectively manage fields to account for this variability, to optimize profit, and reduce environmental impact have become available as management tools. In recent years, major advancements have been made in the technologies required to implement precision farming practices. A very high level of interest in precision farming is evident from the major agricultural sectors, including the farm machinery industry, agri-business firms, cooperatives, and farmers. Many unknown factors are related to the implementation of precision farming, including the continued development of associated technologies and economic factors. Precision farming is an emerging technology and therefore limited research is available to practitioners who adapt precision agriculture for southern soils and crops. This lack of information provides the impetuous for the proposed work.
JUSTIFICATION:
Relationship to Agriculture
Recent technological developments have paved the way for important and far-reaching changes in agricultural production practices. These technologies promise to revolutionize the production of food and fiber in the Southern United States. More specifically, these technologies can enable micro-management techniques on a site-specific basis to account for the natural and human induced variations which exist in agricultural fields; variation in soil type, moisture, topography, chemistry, physical properties, and other factors. These technologies promise the possibility of carrying agricultural production techniques to the next level, with the hope of optimizing profit and reducing the adverse environmental impact of farming.
In 1995, the Southern Association of Agricultural Experiment Station Directors sponsored the "Conference on Agroecosystems", with the objective to "re-examine the land grant vision to better deal with the broadening agenda which includes relationships between agriculture, natural resources, and the environment". In the executive summary of this conference, Clarke(1995) points out that this conference "allowed a focus on the interaction and impact of agriculture at the broader ecosystem level and has illuminated the need for using a systems approach both for research to understand complex interrelationships and for managing whole farm systems in the context of broader ecosystems". He further indicated the need for blending several disciplines in conjunction with new technological tools which can improve agricultural production and provide an integrated whole-farm management system. By its very nature, precision farming provides a whole-farm management system.
The American Society of Agricultural Engineers published their research priorities in 1995. They identified four major areas for agricultural engineering research: (1) conservation of natural resources, (2) renewable energy and bioresources, (3) safe and wholesome food supply, and (4) economic competitiveness in global markets. Some of the specific objectives for the first research area include: the development and implemention of improved water management designs, irrigation methods and controls, and sensors and technology to improve precision application of nutrients and pesticide. For the fourth research area, several of the specific objectives are to: develop sensors, instrumentation, and control systems to enable automation in food and agricultural and forest production and processing operations; develop computer-based mathematical models capable of simulating biological production and processing systems, such as those for crop growth; develop information and decision-support systems that use production and process simulation models to optimize management decisions; apply advanced engineering technologies and improved decision-making tools; and develop improved systems for efficient crop production. The research project outlined herein directly addresses both the first and fourth national agricultural engineering research priorities, and specifically addresses the objectives mentioned above for each of these priorities.
The State Agricultural Experiment Stations (SAESs) published a detailed agricultural research plan for the future, "Opportunities to Meet Changing Needs: Research on Food, Agriculture, and Natural Resources" (1994). This executive summary of the strategic agenda for the State Agricultural Experiment Stations indicated six major research initiative areas: (1) environment and natural resources, (2) nutrition, food safety, and health, (3) processes and products, (4) economic and social issues, (5) animal systems, and (6) plant systems. For the plant systems research area, three of the objectives of this area are directly addressed by this proposal: protect plants for sustained productivity, develop alternative plant management systems, and understand fundamental plant processes. The Southern Association of Agricultural Experiment Station Directors also published a similar strategic agenda (1994) specific to the Southern United States.
Relationship to Environment and Public Concerns
The Southern United States has vast land-based natural resources and available water to produce food and fiber. Refinement of appropriate management techniques is needed to minimize the potential for mis-management, which can result in soil, water, and air pollution. Precision farming technologies promise the ability to apply farm chemicals only where needed and in the appropriate amount thus reducing the potential for pollution.
Clarke(1995) pointed out that the United States Congress is moving "toward a more voluntary and less regulatory approach to environmental goals" and yet there remains an increasing public interest in the environment and the influence of agriculture on the environment. He pointed out the availability of new technologies which can provide knowledge-based management of agricultural production to reduce environmental impact. Some of the technologies identified were computer literacy, Geographic Information Systems (GIS), Global Positioning System (GPS), expert systems, and remote sensing.
This project includes some of the specific objectives listed in the SAES's national research initiative (1994) for environmental and natural resources. More particularly, the development of precision farming systems is aimed at conserving and enhancing air, soil, and water resources and enhancing resource management systems.
Extent of the Problem
In almost all agricultural fields in the Southern United States, significant variation is found in the major factors that affect crop yield. The need for the development of management technologies and processes to account for this variability is considered to extend across the entire Southern United States, and to involve virtually all crops.
The Need for and Advantages in a Cooperative Approach
The continued development of precision farming practices requires the cooperative involvement of most agricultural disciplines. The need to develop engineering technologies, agronomic practices, horticultural practices, entomological practices, and economic information exists for crops grown on major soil types in the Southern United States This will require cooperation between research disciplines within and between states to optimize development of technology and practices needed for precision agriculture in the South.
The Potential Benefit
Precision farming is a promising technology for improving crop management, offering potential for improving both the economic and environmental aspects of farming. Improved and more intensive management arising from better knowledge of the site-specific characteristics of a field, in combination with improved positioning and variable-rate input technologies, may increase revenues through higher average yields. Precision farming may also reduce costs of production and aid the environment by reducing the amounts of fertilizers and pesticides to better match the crop's biological need for the input (Christensen and Krause, 1995). Nevertheless, precision farming is in its infancy, and its full effects on crop yields, production costs, and the environment are still emerging. Little economic analysis has been completed to evaluate its benefits and costs (e.g., Shearer, et al., 1996; Christensen and Krause, 1995; Lowenberg-DeBoer and Swinton, 1995; Fiez, et al, 1994.; Hibbard, et al., 1993; Carr, et al., 1991; Roberts, et al, 1997).
While little is known about the economic benefit of precision farming practices, initial indications are that precision farming practices may be able to increase yield and reduce agricultural inputs, thereby increasing farm income and improving relative competiveness. Profit margins in Southern farming are small, thus even small increases in profit can significantly benefit farmers. In addition, the potential to minimize pollution of our soil, air, and water resources may not be easy to evaluate economically, but certainly would be of significant benefit to the Southern United States.
The potential of precision agriculture is just now being realized. For example, with respect to nutrient management, Midwest corn production studies have demonstrated that increased profits of $37 per hectare are possible. Work in Kentucky has shown that $37 per hectare can be gained with variable-rate seeding and an additional savings of $37 per hectare may be possible for variable-rate application of herbicides. Applying these savings to corn production in Kentucky alone will result in additional profits of $55 million per year. Extending the same practices to other Southern crops across the region has the potential to produce similar profits on a state by state basis.
While precision agriculture shows promise for increasing profit margins, it also has application with respect to environmental quality. As public concern over water quality continues to escalate, producers will be required to make more prudent decisions regarding use of agricultural chemicals. The very nature of precision agriculture will require producers to keep accurate records while helping them to reduce unwarranted pesticide application and to better manage nutrients in sensitive areas. These intensive management skills will also lead to better use of animal wastes and other biosolids.
RELATED CURRENT AND PREVIOUS WORK:
A review of current research projects in the USDA CRIS (Current Research Information System) system revealed no existing regional research projects in the United States on precision farming, except for the initial request for this project. However, a number of individual Hatch and state projects on precision farming were found. A North Central Region information exchange group, NCR-180 titled SITE SPECIFIC MANAGEMENT, seeks to review current knowledge and application technology, develop technology transfer needs, and coordinate titled research. A brief review of these projects is pertinent. The following is a summary of the existing individual projects on precision farming found in the CRIS system for the Southern United States in the following format: Institution, Title, Duration, and Objectives. A brief summary of projects in other regions of the United States on precision farming is included. These projects reflect wide research interest in precision farming in the United States, and most of these projects contain some objective element of this proposal. However, these projects also reflect the need for a cooperative and integrated systems approach to precision farming research which covers all of the major elements of research needed for precision farming systems.
Southern United States
Clemson University, SITE SPECIFIC CROP MANAGEMENT FOR ENVIRONMENTAL AND ECONOMIC BENEFITS, 10/01/95-09/30/00, (1) identify and estimate the costs and potential returns of different levels of site specific farming systems, (2) develop decision making strategies for implementing different levels of site specific farming practice for given farm situations based on environmental and economic benefits, and (3) demonstrate site specific farming technologies.
University of Georgia, IMPROVING PRECISION FARMING AND POST-HARVEST SYSTEMS, 07/01/95-06/30/00, (1) to research, develop and implement precision farming techniques appropriate for the region, (2) to improve peanut quality through technology developments in post-harvest curing and sorting techniques prior to the shelling process.
Oklahoma State University, DEVELOPMENT OF EQUIPMENT AND MECHANICAL SYSTEMS TO SUPPORT AND ENHANCE PRECISION FARMING TECHNOLOGY, 10/01/96-09/30/01, (1) design and develop variable-flow valves suitable for use with precision liquid applications, (2) design and develop variable-flow granular metering devices suitable for use with precision granular applications, (3) design and develop machine elements, utilizing state-of-the-art mechanical and electrical components, to improve the enhance site-specific crop production machinery.
Other Regions of U. S.
Research projects at the University of Minnesota, Purdue University, the University of Nebraska and the University of Idaho are aimed at the development of sensors for precision farming. The University of Minnesota is also investigating the development and/or adoption of machinery and precision farming economics. A project at South Dakota State University has the goal of managing nitrogen stress to reduce environmental impact and maintain profitability. The University of Wisconsin is focusing on site specific farming systems related to hay and forage production. Research at Colorado State University has the goal of evaluating the potential for precision farming systems to protect water quality and conserve resources. The University of Missouri is working on a project to evaluate the effect of precision farming practices on surface water quality.
The proceedings of "SITE-SPECIFIC MANAGEMENT FOR AGRICULTURAL SYSTEMS", the most recent international conference on precision agriculture (Robert, et al 1996), indicates the extent of the national and international interest in precision agriculture. This conference consisted of a workshop on precision nitrogen management, and six technical sessions on the subjects of natural resource variability, managing variability, engineering technology, profitability, environment, and technology transfer. Outlining some key issues related to management of variability, Rawlins (1996) pointed out that there is a need to increase the scientific knowledge base through a coordinated effort of a team composed of government, academia, and industry members to move to the next plateau in variable-rate application of crop inputs. Stafford (1996) discussed engineering technology development and pointed out the need for technological advances in sensing and control systems. In discussing economic perspectives, Lowenberg-Deboer and Boehlje (1996) summarized 11 site specific management (SSM) economic studies. They pointed out that five of these studies showed that SSM was not profitable; four had mixed or inconclusive results; and two studies showed potential profitability. They pointed out that these results are closely related to sampling and variable-rate application costs. The potential for precision agriculture to provide environmental benefits was discussed by Baker, et al (1996). They expressed the concept that precision agriculture has the potential to reduce off-site transport of pollutants, but more information I needed on the interaction between practices and chemical transport.
At a technology symposium held in Washington, D.C. a group of scientists met to project the impact technological advances will have on agriculture. A subgroup studying the impact of crop input technologies stated the following:
"Beginning in the very near future, the farmer will utilize GPS/GIS to map problem sites in his field. Problem areas are already being identified by early adopters using yield monitors on combines and supplemented with detailed maps of weed and nematode problems. The farmer follows up with diagnostics in the problem areas. Variable-rate application equipment allows on-the-go adjustment of fertilizer, pesticide, and other inputs rates. Needless-to-say, diagnostic systems will have to evolve further to accommodate ever improving levels of mapping sensitivity. Even though treatments will be carried out on a very small area (perhaps to the point that every plant in the field will have an address, and be treated separately), planning will probably occur on a very large land area known as the agroecosystem. It is at this point that utilization of animal and municipal wastes, multi-crop sequencing, and pesticide use will be accommodated into a comprehensive watershed plan."( Amerman, Backman, and Marlow, 1997).
A subgroup evaluating livestock technologies stated that in the future, successful farmers will aggressively adopt new technologies to reduce real costs of production (Schneiderman, 1990). They reported that:
"The necessity of maintaining domestic market share, and capturing and holding a share of the international protein market has produced an environment in which only the least-cost producer will succeed. In this case, the least-cost producer is not only the individual producer but the least-cost species or nation. Success in production agriculture increasingly will be determined by the ability to compete, and competitive advantage is driven by cost of production" (Genho, 1993).
Finally, a subgroup evaluating crop machinery technologies indicated that precision farming can aid the environment by reducing the excess amounts of fertilizer and pesticides applied to a field. Instead of applying fertilizer at a constant rate across a field, variable-rate application allows changes in rates within a field to better match crop demand depending upon soil characteristics and test results. Pesticides applications can be limited only to specific problem sites in a field. (Roberts, Kemper, and Christensen, 1997) Yield increases and decreased environmental impacts projected at this conference rely in large part on continued research into the improvements of site-specific technologies.
In addition to the research activities on precision agriculture outlined above, numerous articles have appeared in almost all major farm magazines and the popular press in recent years, indicating the level of interest among farmers. Many of the major farm equipment manufacturers have either begun manufacturing variable rate equipment or yield monitors, or are seriously considering such opportunities. Agribusiness firms are making major investments in precision farming, as indicated by one of the major agricultural cooperatives in the U. S., Goldkist. Goldkist has hired a director of precision farming, and this cooperative is rapidly expanding its activities in this area. It is obvious that there is a high level of interest in precision farming technology on every level, and that there is a critical need to develop a solid knowledge-base on which the technology can move forward to provide the expected benefits.
OBJECTIVES:
1. Adapt and develop sensors and data acquisition technologies for precision farming.
2. Adapt, develop, and assess methodology and equipment for variable-rate control of inputs to precision farming systems.
3. Develop methodologies and analytical tools for optimum utilization of inputs in precision farming systems.
4. Assess the economic and environmental effects of precision farming.
PROCEDURES:
The procedures to be followed for this project are outlined below in the order of the objectives. Each objective is clearly identified along with participating states and organizations. As stated in the "Organization" section, the participants in each objective will annually elect an objective coordinator. The objective coordinator will be responsible for coordination of work on that objective, the sharing of data and data analysis. The project chair will coordinate inter-objective activities. Statements relating coordination and data sharing and analysis for each procedure are provided. This Procedures section is followed by an outline of agribusiness, industry, and farmer cooperation given under a separate heading.
Procedures for Objective 1 (Adapt and develop sensors and data acquisition technologies for precision farming)
One major factor which has contributed to the advancement of precision farming has been the development and/or adaptation of electronic sensors to agricultural applications. For example, current combine grain monitors utilize sensors such as strain gages or potentiometers to measure grain flow rate and capacitative moisture sensors to measure grain moisture content. The continued advancement of precision farming requires the development and/or adaptation of other sensors. Activities planned under this objective can be divided into two major thrust areas which address the adaptation and/or development of sensors and technologies to measure and map: (1) crop yield and (2) soil and plant properties. Specific activities for these objectives are listed and explained below.
(1) Measurement and Mapping of Crop Yield
Techniques will be evaluated and tested for monitoring yield for cotton (TN, GA, SC, NC, MS) and peanuts (GA) to accurately document yield variability within production-scale fields. Existing sensor systems will be modified for application on a commercial peanut harvester and a cotton picker. New sensing alternatives will be potentially developed and evaluated. The primary emphases will be toward accurate monitoring of crop yield in a reasonable "management" area. Consideration will be provided for cost of technologies, adaptability to existing systems, and moving technologies rapidly toward commercialization. The researchers will develop and use a standard methodology for the calibration of yield sensor systems. Data between the various locations will be compared by the cooperators to determine which of the sensor systems provide optimum results. Procedures to develop reliable yield maps when multiple harvesters are used in a field will be developed. A sensor to determine in-orchard measurements for the size, weight, and number of fruit from each tree in the orchard on a site-specific basis will be developed (Clemson (SC)). This orchard measurement system will provide orchard managers with previously unobtainable information about the quality and quantity of fruit from individual trees. Tennessee (TN) will coordinate this effort, including comparisons of the various sensor systems.
(2) Measurement and Mapping of Soil and Plant Properties
Because soil strength is known to be an important factor in crop yield, the use of computer-controlled penetrometers and real-time soil strength sensors (with GPS) will be developed to characterize soil properties spatially within a field (GA, TN, AL, USDA-ARS(AL), NC) . These soil strength maps are expected to be related to yield maps to possibly explain one source of yield variation. In the second year of the project, the cooperators will develop standardized field procedures and techniques for the development of soil strength maps. In addition to real-time soil strength measurements, an optical-based sensor to measure soil structure as it relates to the degree of seed-furrow sidewall smearing will be developed at the National Soil Dynamics Laboratory (USDA-ARS(AL)). Variable-tillage depth operations will be investigated based on depth to a root-restricting layer (USDA-ARS (AL)). Many soils are limited in yield potential by a restrictive soil layer much like a fragipan which can be found in 55% of the upland soils. Alabama (AL) will coordinate these activities.
High accuracy real-time kinematic GPS and Ground Penetrating Radar (GPR) will be investigated as possible tools to map the surface and sub-surface soil profile (TN, GA). Development of concepts to use GPS as a tool to create soil topographic maps will include the conceptional design and development of multiple receivers for accurate work on land which has a significant degree of slope. Fields will be mapped using GPR and ground-truthed using direct soil augering to estimate the accuracy of GPR as a non-intrusive sensing technique. Depth of soil above restrictive soil layers will be compared with spatially varying yield maps using yield monitors and GPS. Tennessee will provide the GPR equipment and expertise; Georgia will provide the high accuracy GPS equipment and expertise for this cooperative development. Tennessee (TN) will coordinate this activity.
Early detection of plant nutrient deficiency is a crucial component of precision farming where corrective action is possible during the growing season. A "plant health" system using spectral reflectance for the detection of nutrient deficiency will be investigated for cotton (TN, GA) and peanuts (GA). Proposed work will implement the use of fuzzy logic and neural networks to allow the current sensor to be readily adapted for real-time detection of multiple plant conditions that may be corrected spatially within a field. This sensing system will be interfaced with a commercially-available control system for variable liquid application rates (TN). For the work in Georgia, the measurement system will separate the plant and soil components of the images where the plants can be analyzed for plant size and to soil fertility and eventually yields. This system will sense morphological changes in individual plants, particularly as caused by drought stress. Plant damage from insects will also be investigated by assessing the area of the plant leaves consumed. Specific procedures include: (1) determine the relationship between NIR image pixel intensities/spectral properties and actual water contents; (2) determine relationship between COLOR image pixel intensities, spectral derivative data and actual plant chlorophyll contents; (3) develop methods to resolve the data confounding problem due to spectral mixtures of spectra and due to aliasing; and (4) develop a field transportable sensing system by transferring technical results and design methodology of thrust areas 1 and 2. Georgia (GA) will coordinate these activities.
Development of sensors and associated systems for the development of variable rate irrigation systems will involve cooperation between a state institution (Clemson (SC)) and a federal agency (USDA-ARS (SC)). This project will focus on managing plant drought stress through both spatial and temporal irrigation control of center-pivot systems. Two center-pivot irrigation systems will be developed to deliver variable rates of water, nutrients, and pesticides to areas about 10 m by 10 m in size. Implementation of infrared thermometers in conjunction with multi-spectral data from remote sensing is planned. A simple low-cost soil water sensor is needed as a crucial component of site-specific moisture management. Relationships between soil moisture and the relatively simple capacity-type moisture probe will be developed. The procedure should result in a low-cost, continuous, water-content probe for precision application of water. USDA-ARS (SC) will coordinate the work and sharing of data related to soil water and drought stress.
The cooperators on this activity will work together to develop a standard methodology for the sharing of data which can be used to develop and compare soil and plant property maps. Since the work under Objective 1 is largely developmental in nature, the findings under this objective do not lend themselves to formal statistical designs. However, descriptive statistics commonly used to report sensor and measurement system accuracy will be used. It has been stated that the technology associated with precision farming is ahead of the science. One goal of the project is to use the advancements in measurement technologies developed in Objective 1 to help advance plant and soil science research. Advancements in electronics, optics, and other technologies will permit a much more rapid assessment of crop response in a multi variate environment than ever before.
Procedures for Objective 2 (Adapt, develop, and assess methodology and equipment for variable-rate control of inputs to precision farming systems)
One of the major objectives of precision farming is to vary the rate of application of field inputs (seeds, fertilizer, lime, herbicides, etc.) in accordance with site-specific recommendations. While equipment for some variable-rate field operations is commercially available, there is a need to adapt and assess the methodology and equipment, as well as to develop new variable-rate technologies. Three main activities will be conducted under Objective 2 as outlined below.
(1) Variable rate methodology and equipment
A study to evaluate the potential benefits of variable-rate seeding of corn (KY) and cotton (NC) will be conducted. Corn will be seeded at rates proportional to top-soil depth (i.e. thicker soils will support higher populations and enhanced yields). Soil conductivity will be investigated as a means of assessing and mapping soil depth for the purposes of generating seeding recommendations. Population sensors will be developed for the purposes of mapping stand to assess the role of germination in generating yield variations on highly eroded soils. A study to evaluate the variable rate seeding of cotton in conjunction with automatic seed depth control will be conducted. The soil drying front will be sensed and depth of planting automatically changed on-the-go. Seed rate will be varied based on yield potential using soil series maps. This experiment will evaluate both automatic depth control for cotton seeding and variable rate seeding as site specific seeding practices. Yield will be monitored to correlate to these seeding practices in Kentucky (KY) and North Carolina (NC).
Precision farming technologies offer the potential to manage specific sites in many types of agriculture. Applying this technology in orchards can improve tree health, and fruit quality and quantity. Variable rate application equipment to apply nutrients, fumigants, lime, foliar sprays, etc. will be developed/adapted as needed to enable economical and precise application of materials on an individual tree basis. Sensors for measuring the tree-canopy volume, foliage density and foliage health of trees at specific orchard sites will be developed/adapted to vary application rates accordingly. (Clemson (SC))
Southern agriculture does not enjoy some of the economies of scale associated with mid-western agriculture, and as such, the investment in variable rate practices can not be amortized over large areas. Therefore it is essential that variable rate fertility practices be developed for southern producers. Grid sampling strategies will be explored to develop soil sampling practices for southern soils. Geostatistic and Monte Carlo simulation techniques will be utilized to guide southern producers in the adoption of appropriate sampling grid or cell size. Modeling of potential returns to the producer through increased yields and reduced inputs will provide the basis for assessing the profitability of site specific application of soil amendments and sampling grid size. Equipment limitations such as spread patterns, system delays, GPS latency and other sources of application error will be considered in this analysis. (KY, GA)
Large multiple-bin, variable-rate fertilizer application equipment developed for mid-western agriculture may not be well suited to the needs of southern producers. High capital costs make this equipment largely unavailable to producers who farm small fields in remote locations. Work will be conducted to assess the potential of upgraded fan spreader boxes for these areas. Lag times, distribution patterns, GPS receiver latency and application efficiencies will be characterized and compared. Recommendations will be provided regrading the potential use and benefits of upgrading existing application equipment in contrast to the high end commercial systems in use in the Midwest. (KY)
A multiple-probe soil cone penetrometer with GPS capability will be used to examine differences in yield resulting from root-restricting layers. This resulting information will be used to examine the potential to control depth of tillage tool operations to determine yield differences when this additional knowledge is available. Existing information on depth of root-restricting layers and established depth of tillage appropriate to ameliorate this profile will be used. Tillage depth will be chosen to analyze if increases in efficiencies can be obtained by tilling sufficiently deep to remove the hardpan. Additional tillage studies will be conducted to assess the influence of tool design and configuration on soil structure and residue cover. A study of tillage quality assessment, from the perspective of adjusting the tool configuration and operation to achieve the desired soil condition change, will be conducted. This will be done on-the-go as a function of changes in initial soil conditions as they deviate from desired conditions. A study of the effects of tool configuration (chisel plow) on draft/energy, residue flow/incorporation, and soil "disturbance" will be conducted, hopefully leading to "variable-rate tillage control" in future designs. The three-depth soil mechanical impedance sensor (Objective 1 - NC) will be georeferenced in relation to the average chisel plow draft measured on-the-go to automatically vary tillage depth and look at this tillage depth in relation to yield for corn and cotton. Auburn (AL) will coordinate these efforts. (Auburn, AL and USDA/ARS (AL), AL A&M, NC).
Field-scale studies employing variable-rate technology (VRT) will be conducted (LA, MO) to compare cotton, corn, and small-grain yields from VRT fields with those of conventional fields. VRT technology to be applied includes fertilizer application (nitrogen on cotton) based on soils and yield maps (LA).
Overall coordination of this variable-rate methodology and equipment activity will be coordinated by Kentucky (KY). One aspect of this objective involves assessing two different management strategies, conventional and variable-rate application of crop inputs. Meta-analysis statistical techniques will be used on the combined data set between states to analyze the effectiveness of variable-rate technology. Pooling of this data between states will enhance the statistical analysis due to the low number of replications (case studies) that any one state is able to provide. This analytical technique will permit the evaluation of two or more different management systems even though different decision rules will be applied to the variable-rate management system on a case-by-case basis.
(2) Site-specific management of pesticides
The potential to minimize water and soil pollution with site-specific application of herbicides will be evaluated on selected watersheds. These isolated watersheds will be instrumented with flumes, automatic sampling equipment and data logging equipment to collect runoff from ditches, surface channels and springs. A detailed geological survey will be made of the local watershed to determine the underlying geology and origination of ground water flowing from spring and wet weather seeps. Intensive soil sampling and mapping will be conducted in all locations initially to establish reference soil conditions. During the first year, water quality sampling equipment will be installed and all sites will be monitored to establish background conditions with current production practices. "Site-specific" management of herbicides will begin during the second year and continue through the third cropping season. Water samples will be analyzed for suspect chemicals (i.e. triazines and alachlors) following standard practices. Grain-yield maps will be generated during each season of the investigation to be used as input in generating application rate recommendations, and as a means to assess the economic impact of precision farming practices. (KY) Technologies will be developed for site-specific pesticide applications to reduce the cost of application and the environmental loading of pesticides (MS).
Center pivot irrigation systems designed to deliver variable rates of water, nutrients, and pesticides to areas about 10 m by 10 m in size will be utilized to study variable rate irrigation systems. The system will contain a variable-rate pesticide system that is independent of the water delivery system. The low-volume pesticide system will be used to apply some pesticides and will be evaluated with respect to distribution uniformity, both within the control element and among elements along the truss length. The variable-rate application system on the second center pivot will utilize a different metering/delivery system, and expertise gained from the first center pivot on fixed-boundary plots will be used to implement variable-rate management on areas with irregularly shaped boundaries. All application and management systems will be evaluated with regard to efficiency, distribution uniformity within and among control elements, crop response, alleviation of plant stress, and environmental quality. USDA-ARS (SC) will coordinate this activity.
(3) Investment analysis of variable-rate technology
Investment analysis of variable-rate technology, including a systems feasibility analysis, will be conducted. Comparative profitability of variable-rate technology to current average-rate production technologies will be evaluated. This will require close collaboration among the agricultural economists and the primary participants in Objective 2. Data requirements for economic analysis will be provided by participants from all stations within a year of project initiation. An inventory of variable rate application projects from all stations will be conducted. The agricultural economists will review ongoing and proposed projects and develop a list of data needs for successful economic evaluation. Multi-year data and multi-site information will be required for risk analysis. Therefore, the need for analysis over time and across state boundaries will be considered as experimental designs are developed. Since many of the traditional approaches toward the analysis of field-plot data are not readily applicable in this type of study, non-traditional statistics, possibly rotated factor analysis, will be needed in many of the analyses. Accomplishing this task is key to the success of the entire project.
Procedures for the comparative profitability analysis will be to compare (TN, LA, MO) average-rate technology to variable-rate technology. To do this, yield response functions are required for the inputs studied within Objective 2 for various sub-field areas. Data gathered for this analysis are expected to include soil physical and chemical properties, weather, insect and weed pressure, plant population, and other pertinent information that might affect variable-rate application. Tennessee (TN) will coordinate the economic studies, and will be responsible for the collection and analysis of data from other institutions. The Tennessee Experiment Station Statistician's Office has already been, and will continue to be consulted relating to appropriate statistical analysis techniques for Objectives 1, 2, and 3. Particular effort will be directed toward soliciting the cooperation of all researchers participating under Objective 2.
Procedures for Objective 3 (Develop methodologies and analytical tools for optimum utilization of inputs in precision farming systems)
Under this objective, we will develop new methods and tools to enhance the economic and environmental benefits of precision farming. To accomplish this, we will pursue new procedures to evaluate the variability of soils and their ability to produce crops, to assess how variability of soil properties can be used to modify the application rates of fertilizers and agricultural chemicals needed for crop production inputs, to develop procedures to assess the variability of pest incidence on the landscape, and to modify pest control inputs to provide the control needed with the minimum quantities of pest control chemicals. The new methods to be tested for their applicability include the use of crop simulation models and remote sensing to assess the properties of soils on the landscape, and the use of GIS for creating maps of the important attributes that might be used for decision making and modeling the relationships between attributes. These studies will be organized and conducted under three thrust areas.
(1) Evaluation of available soil-water storage capacity, plant nutrients, and plant pests through integration of databases and models.
This approach is especially promising in the southeast Atlantic Coastal Plain where water availability for crop growth is often a major factor limiting crop growth and yield (Bruce et al., 1980). In most of this region, mean monthly rainfall exceeds monthly evapotranspiration during much of the year (Bruce et al., 1980). However, evapotranspiration exceeds rainfall during summer months, and rainfall is often as high-intensity, short duration convective storms. Thus, the ability of the soil to accept and retain water from rainfall has a considerable impact on crop growth and yield. In this region, the soil's ability to accept and retain water varies considerably. Many soils in the Coastal Plain part of the southern region are sandy to considerable depths (Quarzipsamments and Arenic and Grossarenic subgroups of Kandiudults and Paleudults), and the texture of the argillic or kandic horizon varies considerably (coarse-loamy to fine families). Though surface textures are commonly sand or loamy sand, surface crusting is common (Chiang et al., 1993) which increases surface runoff. In addition, redistribution of water from high to low landscape positions through shallow subsurface flow is widespread (Hubbard et al., 1983).
Efforts will be taken to predict yield potential based on simulation models that use weather inputs and the storage capacity of plant available water for different soil. Accomplishing this will enable N applications to be tailored to achieve optimum yields. (GA, NC, TN, USDA/ARS(SC), AL A&M, and AL)
Through a systematic process of grid sampling for available plant nutrients, the application of variable rates of fertilizer across fields with variation in available plant nutrients, and from yield maps for these fields, developing large data bases for soil-test calibration will be possible. These data bases will be especially useful in evaluating calibration changes with changes in crop varieties, soil type, and crop management practices (TN , GA, MO).
Many crop production fields have been intensively sampled, often using a grid sampling pattern, for plant available nutrients. More fields will be grid sampled in the future. The maps developed from grid sampling will be evaluated for patterns in soil nutrient availability as related to soil series, slope positions, topsoil depth and other quantitative physical features of the landscape. Similar strategies will be used in the collection of plant samples to determine nutrient deficiency or sufficiency. Plant pests that may be affected by soil type or landscape position will also be studied to determine optimum sampling strategies (KY and GA).
A subset of agricultural economists (Objective 4) working on the project will provide assistance in the development of required economic terminology for incorporation into the data dictionaries. The terms to be defined will evolve from the decision rules and other economic analytical tools developed for this and other objectives in the project (TN).
Georgia (GA) will coordinate the pooled database and related model studies and analyses. Because of the substantial variability among the soils, climates, and crops of the six participating states/locations, non-traditional statistical techniques, rotated factor analysis, e.g., will be used to assess relationships and correlations between and among methodologies and analytical tools.
(2) GIS (Geographic Information Systems), spatial modeling, and mapping
Establishing relationships between spatially variable attributes will allow the development of new understanding that can be used in precision farming. First, the impacts of spatial field parameters (elevation, slope, etc), soil properties (chemical, physical, biological) on spatial distribution of crop yield, and yield potential will be evaluated and quantified. Spatial soil and crop data (yield monitor and remotely sensed) will be collected for major soil types and for major crops in the Coastal Plain and Peidmont regions. All data will be input into a field-scale GIS (ArcInfo), and interlayer data analytical tools will be utilized to quantify spatially dependent relationships.
Then remote sensing of biomass and N content of growing crops (wheat, corn, cotton, etc.) will be employed to develop variable N recommendations. Initial N recommendations will be estimated by establishing spatially variable yield goals determined from crop-yield-monitor data. Pre-plant or at planting N will be uniformly applied to each crop (approx. 10% of total N recommendation). Subsequent top dress or lay by N will be variably applied and based on calibrated remote sensed maps of biomass and tissue N content (NC, TN, USDA/ARS(SC), and GA). North Carolina (NC) will coordinate the work, including the collection and analysis of data from other locations, on this procedure.
Remote sensing of crop production fields can often provide valuable information about the rate of crop growth and development, primarily through analysis of electromagnetic radiation (EMR) images. These images will be used (GA) in the research outlined under the thrust above to help assess the variation in plant-water stress in crop production fields, thus useful maps of soil productivity might be developed that will aid in the variable application of nitrogen and other fertilizer elements. EMR may also be used to assess fields for infestation by crop production pests to provide an early warning about impending crop damage. Pests might then be controlled by spatially variable applications of crop protection chemicals. Analysis of EMR images might also be useful in assessing variations in soil properties, like percent clay, that could be useful in spatially variable applications of herbicides, for example. These and other opportunities to use image analysis will be utilized in the research programs (GA).
Experiments will be undertaken to determine the feasibility of relating remotely sensed data to crop yield, biomass production, and crop water stress as determined by canopy temperature and subjective assessments. Satellite-based multi spectral data will be gathered on several dates on fields of three major crops. Ground-truth data will be collected with commercial and experimental crop-yield sensors, in situ weather stations, and technicians with GPS equipment for geo-referencing subject assessments. The data will be incorporated into a GIS. Statistical classification procedures will be applied to the overlaid data. The accuracy of the classifications will be determined, and the implications of the expected improvement in management capability will be examined (MS, LA). MS will coordinate the sharing and analyses of the imagery obtained by GA, MS, and LA.
This procedure will be led by the coordinator for objective 3. The committee will work toward the establishment of standard mapping procedures that will allow transfer of data between cooperators on this project as well as among users of the technologies that will be developed. In an attempt to document in-field variability and the variables that influence yield, site-specific field evaluations will be analyzed using multiple regression analysis for continuous variables while data sets with discrete variables will be analyzed using a mixed-model approach. Statistical analysis will be performed on a common data set compiled from all states with like dependent variable (crop yield) and independent variables (soil physical and chemical variability, moisture, pest pressure, etc.).
(3) Develop methodologies and analytical tools for optimum utilization of inputs in precision farming systems.
GIS/GPS systems will be developed that will include the inputs from various sensors or measurement systems, (yield, soil analysis, soil type, irrigation, pest pressure, weed control, tillage practices, etc.) into a database for decision making in future operations. The feasibility of future actions such as spraying or irrigating will be the output from the GIS. The methodology of handling such large databases and decision making capability will be a fundamental part of this research. This work will include sensors and/or methods developed as part of Objectives 1 and 2.
Production economics will be incorporated (SC, TN, GA, TX) into the analytical tools and decision rules developed for precision agriculture. These tools will include economic analysis to determine the optimum levels of input application within a field. Break-even analysis will be conducted to determine the minimal variability required for precision farming to be profitable for a given farm or field size. Break-even analysis will be conducted to determine the minimum acreage required for profitable use of precision farming technology for a given level of potential yield variability within a field. Break-even analysis will also be conducted to determine the price of precision farming technology required for profitable use for given variability and field size.
Research will also examine in a dynamic framework the influence which revenue uncertainty has on adoption of technologies. Optimal rules to guide investment decisions and adoption rates for precision farming systems will be derived. The optimal revenue and yield thresholds necessary for the adoption of a new technology or the conventional cropping system will be derived using dynamic programming. These differences in thresholds will then be compared with any changes in environmental degradation associated with a new technology and conventional cropping. Assuming a technology leads to a lessening of environmental degradation, this comparison will provide a cost associated for environmental improvements (SC, TN, GA).
Tennessee (TN) will coordinate this procedure. The Tennessee Agricultural Experiment Station's statistical consultant(s) will assist in selecting and applying appropriate analysis techniques/models for the studies outlined under thrust areas 1, 2, and 3 of Objective 3.
Procedures for Objective 4 (Assess the economic and environmental effects of precision farming)
While important research efforts have taken place during recent years in the Midwestern United States in developing the technological side of precision farming, many questions related to the economic and environmental effects of this technology remain to be answered. Under this objective, we will attempt to assess the economic and environmental effects of precision farming through six activities. Economic research procedures are designed to facilitate the generation of critical economic data which will be used to evaluate these effects. Care will be taken throughout the study to take data such that accurate economic analysis may be assured. Tennessee (TN) will coordinate collection and analysis of data for this objective. All states participating in Objectives 1 and 2 will be responsible for transmitting their experimental findings to one of the states participating in this objective for analysis as soon as they become available, and informing Tennessee (TN) of that transmittal by sending Tennessee (TN) a copy of the findings. In most cases, the statistical analysis of the field data will be performed by those individuals participating in Objectives 1 and 2. Most of the following procedures use non-statistical modeling methods, such as budgeting and simulation modeling, but rely on the statistical findings obtained under those objectives. The following procedures are used to address the six activities under Objective 4.
(1) Analysis of profitability, investment potential, and risk of using precision farming technologies
Investment in precision farming technologies must be profitable in the long run for wide-spread adoption by farmers (Lee et al., 1980; Swinton and Ahmad, 1996). The most important factors that affect profitability are 1) the variability in yield potential within a field, 2) the proportion of that yield variability controlled by farmers, 3) the number of acres being farmed, and 4) the price of the technology. These factors are of interest to farmers as they ask the questions: How variable does yield potential have to be on my farm for investment in precision farming technology to be profitable? What acreage is required for the technology to be profitable? How much should the price of precision farming technology fall before it becomes profitable on my farm? Researchers and machinery dealers also might ask the latter question as they attempt to produce equipment for sale to farmers. Can they produce and sell it to farmers for a price that allows both the dealer and the farmer to make a profit? If not, how can they redesign the equipment to lower the price to farmers?
Collaborative economic research will be established among several participating states to evaluate the profitability of sensor and variable-rate precision farming technologies and management alternatives. Costs of technologies will be determined, including sensors, yield monitors, retrofitted harvesting equipment, variable-rate fertilizer and chemical applicators, variable-rate planters and drills, variable-rate tillage equipment, GPS units, and computer technology, including GIS. Results for these precision farming technologies obtained from replicated experiments under Objectives 1 and 2, including crop and hay yields, fertilizer and pesticide requirements, variable seeding rates, and variable tillage rates, will be used to estimate the economic costs and benefits of precision farming versus traditional cropping practices. The expected annual net revenue, including annualized investment costs, will be calculated for each technology tested. Break-even thresholds will be calculated for yield variability, acreage, and equipment prices that equate net revenue from precision farming with net revenue from conventional farming practices (Osburn and Schneeberger, 1983). The resulting break-even thresholds will be used to evaluate the investment potential for the precision farming technology tested in Objectives 1 and 2 (Osburn and Schneeberger, 1983). This analysis will be accomplished using standard budgeting methods (Boehlje and Eidman, 1984). Budgeting methods, combined with data on yield variability obtained from the replicated experiments under Objectives 1 and 2, will also be used to estimate variability in income and costs, allowing the assessment of production and economic risk (Barry, 1984). The effects of income and cost variability on farm survival probabilities and stochastic profits will be analyzed with FLIPSIM (Richardson and Nixon, 1986), a farm level policy and financial simulation model. These break-even and risk analysis methodologies and subsequent results will be made available to farmers, researchers, and agribusinesses for their use in answering the aforementioned questions.
Non-statistical modeling approaches, such as budgeting and the use of FLIPSIM for farm level risk analysis, will be used for these economic analyses. Yield response models for each homogeneous sub-fields, when required for economic analysis, will be estimated from the experimental data using standard regression methods or through mechanistic simulation models (crop growth and yield, nutrient movement, water quality, N transformations) such as EPIC (Williams, et al., 1990). The Tennessee Agricultural Experiment Station Statistician's Office will be consulted relating to the application of appropriate statistical analysis techniques.
Research studies to be conducted under this procedure will be directed toward developing decision-support tools to help farmers assess the technical, economic, and environmental effects of precision farming. This effort will be oriented toward helping farmers make decisions about their future use or non-use of precision farming technologies (NC, KY, TX, OK, GA, TN, MO). Tennessee (TN) will coordinate this procedure.
(2) Dynamic analysis of investment in precision farming technology
The adoption of precision farming in comparison with conventional cropping practices will be investigated considering the irreversibility and flexible timing characteristics of precision farming technology. The adoption of precision farming involves expenditures which are irreversible in the short term; they are a sunk cost. However, the investment in precision farming can be delayed, possibly indefinitely. When investments are irreversible and can be delayed, the investment decision becomes extremely sensitive to uncertainty over future costs, which are likely to fall over time, and returns associated with the technology. The ability to choose the timing of adoption adds a dynamic component in the adoption of precision farming, which can be treated as a financial call option. A dynamic programming framework similar to that developed by McDonald and Siegal (1986) and Dixit and Pindyck (1994) can be used for determining the value of delaying investment.
The influence irreversibility and/or flexibility has on adoption of precision farming will be examined in a dynamic framework. Revenue uncertainty associated with precision farming and conventional cropping practices will be modeled utilizing stochastic calculus. The optimal revenue and yield thresholds necessary for the adoption of precision farming or the conventional cropping system will be derived using dynamic programming. These differences in thresholds will then be compared with changes in environmental conditions attributed to precision farming and conventional farming. Assuming precision farming leads to fewer environmental problems, this comparison will provide a cost associated with environmental improvements (GA, TN). Georgia (GA) will coordinate the work on this procedure.
(3) Analysis of economic variability across a field emphasizing modeling for farmer decision aids.
Management during the growing season depends upon combining historical information with real-time identification of developing problem areas. How a management decision could impact the crop can be assessed using computer models describing the crop's response to weather and soil variables (Boote, et al. 1996). Among others, the Decision Support System for Agrotechnology Transfer (DSSAT) crop and soil models (Tsuji, 1994) will be used to simulate crop growth and yield using weather, crop, and soil variables. Remote sensing, center pivot irrigation, and yield monitoring data obtained from replicated experiments performed under Objectives 1 and 2 will be incorporated into these modeling efforts. Software programs will be written to help farmers map economic variability across a field and to make variable-rate input decisions. These programs will account for variation in both yield and cost, allowing the farmer to enter into the models geo-referenced yields and fixed and variable costs to assist in making economic decisions in farming the field (KY, Clemson (SC), USDA-ARS (SC), TX). Clemson (SC) will coordinate these efforts.
(4) Evaluation of both the physical and economic impacts of precision farming on soil, water, and air quality and wildlife.
Precision farming can potentially reduce the effects of agricultural production on the environment (Larson, et al., 1997). Replicated experiments performed under Objectives 1 and 2, where possible, will be monitored for runoff water volume and quality, water and wind erosion, and soil nutrient loading to assess environmental implications regarding soil and water quality. Evaluation of landscapes with respect to environmental impacts will be made using hydrologic and plant processing models (English, 1995). Analysis of site-specific technologies will be compared to the more traditional technologies currently employed (Bell, et al.). Mechanistic simulation models (crop growth and yield, nutrient movement, water quality, N transformations) such as EPIC (Williams, et al., 1990) will be used, after careful validation, to assess long-term potential for safeguarding the environment. Watershed models such as SWAT (Arnold and Allen, 1992) will be employed to assess long-term in-stream water quality implications of precision farming practices. When possible, the economic benefits of precision farming will be coupled with environmental improvements to measure the environmental improvement per dollar for each technology (TX, OK, GA, KY, NC, TN). North Carolina (NC) will coordinate work on this procedure.
(5) Analysis of the impacts of precision farming adoption on agribusiness and rural communities
The current structure of agribusiness will likely change with wide-spread adoption of precision farming. Different inputs and services will be required. Cropping patterns will shift. Environmental impacts resulting from fertilizer, tillage, and chemical applications will change. If wide-spread adoption of precision farming occurs, subsequent structural changes could have important consequences for the long-term health and vibrancy of non-farm rural areas. A number of important questions relating to these possible changes will be addressed in this project: What are the impacts of precision farming on farm input suppliers in local communities? Will precision farming lead to reduced demands for inputs, adversely affecting the business of input suppliers and the rural communities where these businesses are located? Will precision farming methods lead to changes in the structure of agriculture in rural areas that would not have otherwise taken place? What are the implications of these structural changes on the survivability of rural communities? What are the implications of precision farming adoption for farm labor? How is the competitive position of farmers employing precision farming technology altered relative to those who choose not to employ it, and what are the broader implications for non-farm rural areas? Estimates of the impacts of precision farming adoption on our rural landscape will help us answer these questions as we plan for the changes that will occur. In areas where precision farming would be most profitable, an input-output analysis using IMPLAN will be conducted to measure the impacts of precision farming technology on the rural economy (Huarachi, et. al., 1994).
Previous research in this area has dealt with: 1) the impacts of new energy and water-saving technologies on farm size and the structure of agriculture (Fujimoto, 1977; Heaton and Brown, 1982), 2) the sustainability of rural communities and the sustainability of agriculture (Flora, 1995 ; Flora and Flora, 1988; Voth and Moon, 1997), and 3) tying the structure of agriculture to the survivability of rural communities (Goetz and Debertin, 1996; Hayes and Olmstead, 1984; Marousek, 1979). Similar research evaluating the impacts of precision farming adoption on rural communities does not exist. However, the aforementioned research will give guidance in evaluating those impacts (TN, KY). Kentucky (KY) will coordinate this procedure.
(6) Analysis of the macro impacts of precision farming adoption and evaluation of policy tools that affect its adoption and diffusion, including impacts on farm prices, cash receipts, net returns, and international competitiveness.
Precision farming has the potential to affect U.S. agriculture at the farm-level by reducing production costs, increasing yields, and reducing the off-farm environmental impacts associated with agricultural chemical use. To the extent that the adoption of precision farming technologies becomes wide spread, the availability of these technologies could pose a variety of significant impacts on U.S. agriculture. Higher yields could result in significant changes in the level of production and prices for some commodities and have impacts on livestock feed costs and agricultural producer income. The unknown potential of precision farming to initiate shifts in regional crop mixes also may result in structural changes for U.S. agriculture (e.g., increasing farm size and vertical integration). After determining the economic benefits and costs of precision farming technologies and management alternatives (under Procedure 1), along with their associated environmental impacts (under Procedure 4), socioeconomic impacts of environmental policy initiatives can be evaluated.
Given data (e.g., variable costs, net returns, yields for various crops on various soil types) on precision farming from the other phases of the project, both regional and macro-level economic and environmental impacts of precision farming will be estimated using models developed by the Agricultural Policy Analysis Center at The University of Tennessee (De La Torre Ugarte, 1997) and by the Policy Research Institutes at the University of Missouri and Texas A&M University. Using the project's anticipated production cost and yield data, these models will be used to evaluate changes in national prices, production, income, and other important economic indicators of agriculture, as well as regional and macro-level environmental impacts. With this analytical potential and capacity, proposed and emerging legislation can be analyzed with more science supporting those results given to the Congressional and Executive branches of government. Prompt response to legislative requests regarding impacts of precision farming using environmental and economic analyses can sometimes mean the difference between bad and good lawmaking (TX, OK, GA, MO, TN). Georgia (GA) will coordinate the work on this procedure.
AGRIBUSINESS, INDUSTRY, AND FARMER COOPERATORS:
In practically all states, there will be close cooperation with agri-business, industry and farmers, who are briefly identified below. The detailed nature of their cooperation is not identified in this research proposal for sake of brevity; because some of these relationships are in development; and in some cases, to protect patent rights. Most field experiments will be conducted on farms, not on University or USDA experiment stations. It is expected that additional cooperation with industry and farmers will occur as the project progresses. Georgia (GA) will be working with industry cooperators including John Deere, Microtrak, Kelley Manufacturing Co., and Zycom Corp.; the GoldKist Farmer Cooperative; NASA Space Grant; and with a number of farmers. North Carolina (NC) will be working with Southern States & Top-Soil and Open Grounds Farm. Field studies in Kentucky (KY) will occur on the Worth and Dee Ellis Farms, Luck Farms, and Peterson Dairy Farms; Kentucky (KY) will be working with Agrem Inc and with John Deere. Tennessee (TN) has already developed, and will continue to maintain working relationships with, two commercial farms in north central Tennessee. Mississippi (MS) is in the process of selecting farmer cooperators. Clemson (SC) will be conducting yield monitor studies with Greens Farms Inc and farmer Warren Bozard; and also other farmer cooperators are being selected. Louisiana State (LA) will be working with Ag-Chem Corporation. USDA (AL) will be working with Allen Inc., a John Deere dealer. Mr. Tommy Valco of Cotton Incorporated is a participant in the project and is working closely with multi-disciplinary teams in several states to design best management practices for precision farming of cotton; including the use of remote sensing to identify crop stress and the measurement of yield variability parameters for cotton.
EXPECTED OUTCOMES:
An important starting point to implement precision farming is the development of yield maps. The equipment and procedures are commercially available today for yield mapping with any crops that can be harvested with a grain combine. However, two important southern row crops (cotton and peanuts), and tree fruit do not yet have yield monitoring equipment available, and the work on this project will hopefully develop the technology to make possible the production of this equipment. Also, the work on tree fruit will include mapping of fruit quality as well as quantity, which can possibly lead to variable-rate harvesting techniques related to fruit quality. Sensors to map the physical and chemical properties of soils will be adapted or developed to lead toward the development of maps which can be related to variable-rate application equipment and yield maps. Sensors, which can enable the mapping of crop stress and nutrient levels, will be developed to enable decision making related to spraying, irrigation, and nutrient application schedules.
The second objective considers the application of existing variable-rate equipment, or development of new variable rate application equipment specific to Southern crops and soils. The work will include the development of a knowledge base for decision making processes on variable-rate seeding, tillage, application of fertilizers and herbicides, and other farm chemicals. The economics of soil sampling or real-time measurement of soil or plant properties, and the variable-rate application equipment will be included to determine the crops and soils for which variable rate applications are economical.
One of the important aspects of precision farming is the development of historic data bases that form an important part of the micro-management aspects of the technology. The third objective will address this issue by looking at the major factors that affect crop yield, such as soil-water storage capacity, soil type, and soil chemistry, leading toward a knowledge-based decision making process. For example, the optimum process to be used to obtain soil samples for nutrient analysis, and then how to apply this information to variable-rate application of fertilizer is not presently understood. Interpretation and application of aerial photographs and space imagery to farming processes will be enhanced through the studies outlined under objective 3. The development of standard mapping procedures will enhance sharing of data among locations and states.
Finally, the bottom line of precision farming is economics. Objective 4 will address this issue with the collection of economic data across locations and states, a combined analysis to examine the economic issues will be of great value in identifying the crops, soils, and other major factors which indicate the profitable application of precision farming from an economic perspective.
ORGANIZATION:
A regional technical committee will be organized upon project approval. Operational procedures to be followed will be according to those outlined in the CSREES Manual for Cooperative Regional Research as revised in October 1992. The voting members of the regional technical committee will include one representative from each cooperating agricultural experiment station or institution appointed by the director, and a representative of each cooperating USDA-ARS research unit. The administrative advisor and the CSREES representative will be considered nonvoting members. All voting members of the technical committee will be eligible for office.
The offices of the regional technical committee will be the chair, vice-chair and secretary and they will serve as the executive committee. Officers for the first year will be elected at the organizational meeting of the technical committee. In subsequent years the officers will be elected annually and may succeed themselves. The chair, in consultation with the executive committee, may appoint subcommittees to facilitate the accomplishment of the various research and administrative tasks involving the cooperating institutional representatives. Such tasks may include, but are not limited to, research planning and coordination, development of specific cooperative research procedures, assimilation and analysis of data from contributing scientists, and publication of regional bulletins. The chair will be responsible for the overall coordination of the project activities.
The participants within each of the four objectives shall annually elect an objective coordinator. Each objective coordinator shall be responsible for research planning and coordination, and assimilation and analysis of data from contributing scientists, and for publication of regional bulletins. The objective coordinators will also provide appropriate inter-objective coordination of research activities, sharing of data, analysis of results, etc. The objective coordinators shall be responsible for the coordination and presentation of activities under their respective objectives at the annual meeting.
The chair, in accord with the administrative advisor, will notify the technical committee of the time and place of meetings, prepare meeting agendas and preside at meetings of the technical committee and the executive committee. The chair is responsible for preparing the annual progress report and coordinating the preparation of regional reports. The vice-chair assists the chair in all functions. The secretary shall record and distribute the minutes and performs other duties assigned by the technical committee or the administrative advisor.
During the organizational technical committee meeting of this new project, the administrative advisor will arrange to have a qualified consulting statistician address the technical committee concerning pooling of data, appropriate statistical analysis techniques, and related statistical considerations.
Annual meetings, organized by the technical committee, will be held for the purpose of conducting project business. Considerable time will be devoted to research coordination, data sharing, and progress reports by objective. The objective coordinators shall organize the presentation of these reports.
Attachment I: Project Leaders
Attachment II: Resources