Site-specific agriculture, precision farming, prescription farming -- all of these terms are used to describe the same discipline. However, producers, consultants, agribusiness personnel and educators use the terms to signify different things. Some refer to precision agriculture as just "gadgets," such as yield monitors, GPS receivers and computer-controlled application devices. Others think of it as information and/or the technology required to collect information. These are two separate issues that need to be combined in order to help the producer, who is the ultimate end user, make informed decisions about his/her farming operation. Site-specific agriculture, as defined by this paper, is a management strategy that uses information technologies to bring spatial data from numerous sources, which can influence decisions associated with crop production (Brase, 2006; Srinivasan, 2006; NRC, 97). With this said, let's define the term information in general and determine what it implies in terms of agricultural site-specific management practices.
Information (knowledge), information technologies (spreadsheets, databases, etc.) and site-specific agriculture (GPS, GIS and RS) have four components:
- Data capture.
- Data access.
- Data interpretation and analysis.
- Implementation of a management scheme (NRC, 97).
Unfortunately, each of these components has its own scale of variability. Factors such as soil fertility and weed management can vary significantly at the subfield level and temporally over a growing season; however, the management of insects and weather are examples of management decisions, which encompass areas much larger than individual fields. The implementation of site-specific management plans most often is dictated by the available equipment in terms of the scale or resolution of the equipment itself. For example, a producer may soil sample on a 2 1/2-acre grid (330 ft between sampling points); however, the fertilizer spreader used for application may have a spreading width of 45 ft. This scenario means that either there is an over application of fertilizer at the edges of the grid cell or an area is relegated to little or no application. This example is just one of many when talking about the different resolutions involved in site-specific agriculture. Others are yield monitor data vs. remote sensing images vs. soil types vs. weather patterns vs. weed infestation vs. drainage issues vs. topography, etc.
Tremendous amounts of spatially referenced data on individual fields are being generated by yield monitors, on-the-go sensors, remote sensors and direct observations from producers and crop consultants. This site-specific data will have value for use within individual fields but will also have some value in interpreting crop variability within certain regions or geographic areas. For example, seed hybrids can be evaluated on a site-specific basis as to the performance of separate varieties within individual fields or how well they perform in a certain temperature or geographic region. However, caution should be exercised when trying to interpolate across large spans of geographic distances because different interpolation methods can create opposing results. Current agricultural data warehousing enterprises promise that the aggregation of data is for the common good of the industry. This may be true, but unfortunately, many spatial correlations are at the best difficult to prove.
One of the major difficulties with site-specific data collection and analysis is how to appropriately combine the data collected at different spatial scales to help make better management decisions. A number of scales or spatial resolutions characterize crop production systems of today. These scales might be viewed as a continuum ranging from individual plants in a field to plant populations, fields, farmsteads and regions. With the advent of remote sensing data to site-specific management schemes, spatial resolution or scale becomes more important. Remote sensing systems now have the capability to produce very high-resolution images which aid in the detection of crop anomalies and physical soil properties; however, the data volume increases in several orders of magnitude. The sheer volume and size of these data sets can systematically bring any PC based computer system "to it's knees." Aside from the volume, the aggregation of different spatial resolutions creates difficulties for even the most-experienced information technologist. In terms of data volume, a one-meter resolution image will have up to 10 thousand data points per hectare; yield monitor data collected at 1 second intervals can have 4 to 5 thousand points per hectare depending on speed and header width of the combine.
With this magnitude of data, large and small producers, agribusinesses and crop consultants all struggle with the fact that "more information does not always mean more control." The "sorting through" of multiple data files with several different scales or resolutions, all in distinct geographic projections, is a daunting task. This "information overload" more often than not leads to decisions being made out of desperation because of the pressure or need to perform using the technology that has been purchased. Even more often, site-specific decisions are made without the producer, consultant, etc. having the "knowledge" to make an informed decision. The "knowledge" as stated here means the ability to discriminate among all of the variables in a given area and re-aggregate those variables into meaningful management decisions. The determination of the most-limiting factors is currently both difficult and expensive, and these costs are considered by decision makers. All of these concerns point to the need for analytical systems and technologies that can determine the important factors and decision-support systems that can use available data.
It is true that other industries have struggled with the "information overload" that agriculture is now experiencing. Any industry or business which has gone from paper records to electronic files has experienced the feeling of "now what?" A current trend in industry is to hire the appropriate personnel (usually called information technologist) to manage the data and disseminate it to the masses. These information technologists are typically trained in the computer sciences and strictly deal with databases and how to effectively incorporate data collection into their given profession (Moore, 1999). This works well with most businesses, but not in agriculture!
Farmers, past and present, are an independent group of people. They usually do not want anyone to have access to their records or data. The decisions they make affects their bottom line, and they have very little room for error. Farmers understand that the world in which they work is a biological system that is not easily understood. Therein lies the problem. In most non-agriculture-related industries, data are binary -- either "black or white" numbered in rows and columns; agriculture is a biological system in which each row and column interacts with other rows and columns. These causes and effects are not yet very well understood, and the premise of companies which deal in site-specific agricultural hardware and software is that their product will help the producer understand what is happening within his field. Unfortunately, this is not always the case because data collected do not always equal data understood.
For the use of site-specific agricultural management to become widespread, producers, consultants and agribusinesses will need general computing skills and technical literacy. Specific skills such as GPS and/or GIS could be taught through traditional four-year programs or two-year vocational training programs. The question is: do these educational programs teach basics of agricultural sciences to students who are already familiar with the technologies used in site-specific agriculture or technology to students versed in agricultural sciences? In many cases, several university departments provide the expertise for site-specific management, but cross-disciplinary instruction is difficult not only because instructors are not versed in the disciplines of other departments but also because the students resist the fact that they need to study areas outside of their principle interest. For example, an agricultural systems management student may resist the need for the agronomy and/or economics of site-specific agriculture; he/she is usually just interested in the technology side. An agronomy student does not understand why he/she needs to know about combine dynamics or sprayer speeds. The need for a strong understanding of the different disciplines is helpful for both instructors and students. Educational materials will have to be developed for both the agriculturists needing more training in technology and the technologically sophisticated students needing more training in the agricultural sciences.
Site-specific management technologies are new and largely unproven. Before widespread adoption can occur, the economic benefits must outweigh the cost of the technology. However, the environmental benefits need to be capitalized upon and will require a fundamental shift in agricultural paradigms. The ability to document what, when and where inputs are applied will play a large part into how the producer approaches his/her operation. As the 21st century approaches, producers, consultants, agribusiness and students must be increasingly aware of the fact that the environment is a closed system. The inputs they apply do not disappear; they are only redistributed. Environmental consciousness must not be considered a nuisance but a prerequisite for survival (Morris, 1997).
Information technologies have the potential to provide considerable amounts of useful information for decision making in site-specific agriculture. All of the above examples contribute to the "information overload" that exists in today's information-rich "agriculture." Educational programs and user-friendly tools such as farm-based GIS and GPS systems need to be developed before across-the-board implementation can occur. We must go from an "agricultural revolution" mentality in regards to site-specific agriculture to more of an evolution mentality -- building upon past management practices using information technologies as a management tool and nothing else. Site-specific agriculture will produce enormous data sets on crops, soils, pest management and their interactions with the environment. The challenge is to convert these data sets into useful suggestions to aid in the decision-making processes facing the producer as he/she progresses into the 21st century.
Works Cited
Brase, T. 2006. Precision Agriculture. Thomson Delmar Learning, Clifton Park, NY.
Moore, G.A., 1999. Crossing the Chasm: Marketing and Selling High-Tech Products to Mainstream Customers. HarperCollins Publishers, Inc. New York, NY.
Morris, D.K., 1997. Development of a Precision Applications System for Liquid Animal Manures. Masters Thesis, Purdue University.
National Research Council, 1997. Precision Agriculture in the 21st Century: Geospatial and Information Technologies in Crop Management, National Academy Press, Washington, D.C.
Srinivasan A. 2006. Precision Agriculture: An Overview. In Handbook of Precision Agriculture. Ed. A Srinivasan, 3-15. Haworth Press Inc.