Amir Abbas Bakhtiari and Amir Hematian – Precision Farming Technology, Opportunities and Difficulty

Abstract

Precision farming is a data‑based management and a way of agricultural production, which takes into account the in–field variability. Precision agricultural technologies, such as Global Positioning Systems, Geographic Information Systems, remote sensing, yield monitors, mapping, and guidance systems for variable rate application, made it possible to manage within–field variation on large scales. The objectives of this perusal are to collect information about precision farming technology and its opportunities, challenge and difficulty. Results of the study show that there are many opportunities and challenges for adoption of precision agricultural techniques around the world. Although the form of precision practices may be different from one place to another place, depending upon the creative mindset of farmers, practitioners, scientists and consultants local to the area of interest.

Keywords: Precision agriculture (PA), adoption, challenge, technology.

1. INTRODUCTION

Precision Agriculture (PA) or Precision Farming (PF) has witnessed unprecedented growth in the last decade, especially in countries such as the United States, Germany, Canada and others. While the rest of the world has been relatively slow in embracing precision agricultural practices, the change is coming. From Australia to Zimbabwe, PA is growing across the globe. This is clearly evident by the number and diversity of manuscripts published in the area of PA in international journals and also by the variety of papers presented at the major international conferences on PA from different countries around the world. Publications and presentations may not be a scientific metric to account for the geographical spread but it indeed is a reflection of changing times and the proliferation of PA techniques and concept (Khosla, 2010). Interestingly, there are a number of definitions and concepts that can be found in literature pertaining to PA. The one that is most commonly cited and used by practitioners is the one that consist of several "R" of Precision.

Agriculture. Robert et al. (1994) proposed three "R", the Right time, the Right amount and the Right place. Later, the International Plant Nutrition Institute added another "R" to that list, "the Right Source", and more recently, Khosla (2008) proposed an additional "R", the Right manner.

Generally, three major components of precision agriculture are information, technology and management. Base of these three principles, we can defined PA in different ways. Precision farming is information‑intense. A lot of data are required to generate treatment maps and many techniques are under development or have been developed during the last decade.

Precision Agriculture is a management strategy that uses information technologies to bring data from multiple sources to bear on decisions associated with crop production (National Research Council, 1997). It allows for the management of spatial and temporal variability within a field, reduction of costs, improvement of yield quantity and quality and reduction of environmental impacts (Reichardt and Jurgens, 2008).

In other hand, PA can be defined as a set of technologies that have helped propel agriculture into the computerized information‑based world, and is designed to help farmers get greater control over the management of farm operations (Gandonou, 2005). Precision farming is based on several separate technologies, yet interdependent parts, which together form the basis for individual management systems.

Precision farming technology (PFT) is designed to provide information and data to assist farmers when making site‑specific management (SSM) decisions. Demand for information about technologies to manage agricultural production systems increased with the advent of yield monitors and global positioning systems, etc., and improvements in computing power and data management (Swinton and Lowenberg–DeBoer, 1998; Griffin et al., 2004). The following section outlines the most prevalent systems that are used in PA.

2. SPATIAL DATA MANAGEMENT

Field is not one uniform unit. In principle, all parameters and farming practices can be a part of a site‑specific resource management.

2.1. Geographical Positioning

Precision farming, as we know it today, started in the late 1980s with the advent of the global positioning system (GPS) into the agricultural sector, when it was realized how much farming data were spatially related (Blackmore, 2003). Geographical positioning with GPS or DGPS (Differential GPS) is the “backbone” of most PF practices. It is essential for most site‑specific practices that a specific action is recorded and positioned in order to use the information for future treatments. With a positioning system it is possible to gather information site‑specifically. The global positioning system (GPS) makes possible to record the in field variability as geographically encoded data (Nemenyi et al., 2003). As an alternative to the GPS‑system is developed the Russian GLONASS satellite system, which in principle is similar to the GPS‑system and based on several satellites.

The Global Positioning System (GPS) is one of the key technologies that make precision agriculture possible. GPS receivers with sufficient accuracy for yield mapping, grid sampling, variable rate application, and other precision activities are available at moderate cost. GPS receivers, which provide accurate, geo‑referenced position information, are often used with combine yield monitors, scouting equipment, or variable rate application machinery (Stombaugh et al., 2002). Several factors, including the satellite and receiver clock limitations, ephemeris variation, satellite configuration, atmospheric interference, multipath and selective availability (SA) can cause errors in GPS position information.

The most common way to counteract GPS errors is by using Differential GPS or DGPS. The Differential Global Positioning System (DGPS) is an integration of space‑based and ground‑based segments that together comprise a radio‑navigation facility (Shearer et al., 1999). In a DGPS system, a GPS receiver is placed at an accurately known location. This base station receiver will calculate GPS errors by comparing its actual location to the location computed from the GPS signals. This error information is sent to the rover receiver, which uses it to correct the position information it computes from the GPS signals (Stombaugh et al., 2002). DGPS corrections can be broadcast by tower‑based or satellite‑based systems (Figure 1).

Figure 1. DGPS tower‑based (left) and satellite‑based systems (right)

By using the geographical positioning system it is possible to locate the information for further analysis and make a visual presentation in a GIS‑system.

2.2. Geographical Information System

In order to save your geographical position and site‑specific field characteristics it is an indispensable to have a Geographical Information System (GIS) platform to save and handle spatial information. The GIS created by computing background makes possible to generate complex view about our fields and to make valid agro‑technological decisions (Pecze, 2001). Most GIS‑systems are reliable but also fairly expensive depending on the features and capabilities of the program.

A GIS is a set of computer tools that allows one to work with data that are tied to a particular location or spatially mapped area on the earth. A GIS is a database that is specifically designed to work with map data (Price, 2006). GIS is the merger of attribute and a geographic database. This allows for multiple detailed data to be graphically depicted on a map and utilized for decision making. Farmers have long since utilized maps for data collection and decision making. However, the difference with applying the advanced technology of GIS is that the GIS map exhibits “intelligence” where you can ask a question and get an answer. GIS technology application is influencing decision making because it avoids the shortcomings of traditional maps, allows for the rapid computer analysis, and applies sophisticated data structures. GIS is used as a data synthesizing and decision‑support tool in many fields (Mickle, 2009). GIS is a spatial decision support system not a decision making system.

A GIS software program enables the farmer to assemble and organize different sources of site‑specific and geographical data information in various layers. Each layer in GIS is termed “coverage” and consisting of topologically linked geographic features and associated data.

2.3. Remote Sensing

Remote sensing is the science and art of acquiring information about the earth's surface without actually coming in contact with it. This is done by recording energy, which is either reflected or emitted from the earth's surface. The information recorded is then processed and analyzed, and the information is used to develop a prescription map that can be used in a variable rate application (Grisso et al., 2011). This technology can be used to obtain various layers of information about soil and crop conditions. It allows detection and/or characterization of an object, series of objects, or the landscape without having the sensor in physical contact with the soil. It uses aerial or satellite imaging to sense crop vegetation and identify crop stresses and injuries, or pest infestation. So far satellite images have mainly been used to produce maps and predict yield potentials on large areas. With higher resolution and near‑infrared and red‑sensing it is possible to map the relative reflectance from the crop in a small scale (Adamchuk et al., 2003).

Technological advances are broadening the use of remote sensing in PA, as follows:

(a) Satellites such as Quick Bird (QB) provide high spatial resolution images of around 2.0–2.4 m in multi‑spectral (DIGITAL GLOBE Corp, 2010);

(b) Conventional airplanes images can be used to map the spatial variability of biotic and abiotic parameters on agricultural plots at spatial resolutions of 0.25–0.50 m (STEREOCARTO, 2010);

(c) Unmanned aerial vehicles (UAV) flying at low altitude have been developed by commercial companies (UAV, 2010) to provide high spatial resolution images, with the advantage of autonomous management and ability to work in cloudy days. Satellite images are in principle commercially available but the images require a further processing into a GIS‑format.

The satellite images can be provided with resolutions between 1 and about 100 meters per pixel, depending on the image and number of colors. In that respect they are not as detailed as aerial photos (Gomez–Candon et al., 2011). The size of the satellite image are typical relatively large, implying that it is very costly for the individual medium sized farm holding to utilize the images.

3. SOIL SAMPLING AND MAPPING

A variety of different soils exist across any given field. Also, a field may contain a low level of one nutrient and a high level of another nutrient. Soil sampling and testing is the best way to identify these differences and to adjust liming and fertilization practices. Both grid and directed sampling are used to describe soil properties for management of variable rates of fertility. Grid soil sampling involves sampling at points on a square grid throughout a field. An alternative to grid sampling is directed sampling, which involves dividing a field into regular or irregular management zones based on features such as soil type, topography, or past yield performance. In either system, collect at least five or six subsamples from each cell or zone and mix them into one sample container (Stombaugh et al., 2001). The soil samples are collected from each section and then forwarded to a laboratory for analysis often guided by the advisory service. It is absolutely necessary to take care to assure that the soil sample you send to the laboratory accurately represents the area sampled (Thom et al., 1997). When soil fertility data is returned from the soil testing lab, you can enter the data into your mapping program to create fertility maps.

Gathering of soil data from soil samples to create a soil map is a relatively expensive and time consuming task but have also valuable information for later treatments with fertilizers and design of decision support systems. A soil analysis can identify some of the limiting factors regarding yield and growth potential for the specific crop on a specific spot in the field. Soil test results should be included in a record system for each production field on a farm, along with the amounts of lime and fertilizer applied each year, the crops grown, and the yields obtained (Thom et al., 2000). In an effective sampling program, each production field should be tested at least every three to four years.

3.1 Soil Electrical Conductivity Mapping

Soil conductivity mapping is a simple and cheap method to quickly characterize soil differences within a field. Soil conductivity correlates to soil properties, which affect crop productivity such as soil texture, drainage conditions, organic matter and salinity. The spatial soil conductivity mapping may enable the farmer to explain yield variations, pH‑levels and water holding capacity in sub–areas of the field. In this respect, conductivity mapping could enable the farmer to reduce costs for liming, irrigation for certain crops or nitrogen application. Moreover soil conductivity mapping could be valuable information to direct soil samples on the field in management units to minimize the total number of samples on the field (Nehmdahl and Greve, 2001). There are currently two methods to measure soil conductivity; electromagnetic induction (EMI) and contact electrode. Both systems show similar results (Doerge et al., 2002). Corwin and Lesch (2005) provide a comprehensive review of the historical development of EC measurements in agriculture as well as a discussion of the basic principles, including general theory and factors influencing the EC measurement; different geophysical techniques for measuring EC; mobilized EC measurement equipment and applications to site‑specific crop management (SSCM).

4. YIELD MONITORING AND MAPPING

Yield monitors and maps form a very important part of a PF system. Yield monitors are a logical first step for those who want to begin practicing SSCM. The yield monitor which predicting yield from other variables, is intended to give the user an accurate assessment of how yields vary within a field. With a yield monitor, a producer also can conduct on‑farm variety trials or weed control evaluations without the need of a weigh wagon. Such on farm comparisons help producers fine‑tune crop production practices to their soils. Yield monitors are a combination of several components. They typically include several different sensors and other components, including a data storage device, user interface (display and key pad), and a task computer located in the combine cab, which controls the integration and interaction of these components. The sensors measure the mass or the volume of grain flow (grain flow sensors), separator speed, ground speed, grain moisture, and header height. The sensors are interfaced with both analog to digital and direct digital inputs. Yield is determined as a product of the various parameters being sensed (Shearer et al., 1999).

Grain volume flow sensor is prone to errors on slopes as the grain can slip to one side and block the beam. By adding slope sensors, this has been accounted for (Moore 1998). Yield monitoring values can be exported to a personal computer and stored in nonvolatile memory for further analysis or printing via specialized software packages. Yield data were recorded and analyzed over ten years, resulting in a rich data set for both spatial and temporal trend analysis and management information.

According to Blackmore et al. (1996) six main sources of errors of yield measurement that have been identified are unknown crop width entering the header during harvest, time lag of grain through the threshing mechanism, the inherent "wandering" error from the GPS, surging grain through the combine transport system, grain losses from the combine, and sensor accuracy and calibration. Yield mapping is one application of data analysis where the process of observation, interpretation, evaluation and implementation can be applied. Yield mapping software can be purchased as a part of an overall investment in a combine harvester or separately.

To display the maps as contours the (effectively) randomized combine data are converted onto a regular grid. As these grid points rarely coincide with existing data points, a technique called interpolation is used to estimate the value from surrounding values. There are many forms of interpolators, but the two most commonly used are called Inverse Distance to a Power (usually Inverse Squared) and Kriging (Blackmore, 2003).

5. VARIABLE RATE APPLICATION AND TECHNOLOGY

The concept of PF has the potential of aiding farmers to use nutrients and chemicals in an efficient way according to crop needs at a given site on the field and thereby make farming more economically competitive and sustainable. Variable Rate Application (VRA) of nutrients has the potential to improve the specific crop quality in a number of rotations, for example to obtain certain protein content in cereals; in particular wheat for bakery flour and barley for malting. Variable Rate Technology (VRT) is the equipment with the ability to change the application rate and mixture as it moves across the field. VRT adjust pesticide, herbicide and seeding application rates to match soil potentials or problem areas (Adamchuck and Mulliken, 2005). Robert s et al. (2001) reported that N losses to the environment were lower with VRT than with uniform rate application (URA), except on fields with little spatial variability. Gandonou (2005) showed that variable rate fertilizer application results in an increased yield.

The two basic technologies for VRA are map‑based and sensor‑based. Map‑based VRA adjusts the application rate based on an electronic map, also called a prescription map. Using the field position from a GPS receiver and a prescription map of desired rate, the concentration of input is changed as the applicator moves through the field. Canopy management can be conducted with a combination of crop sensors and real time modeling. Sensor‑based VRA requires no map or positioning system. Sensors on the applicator measure soil properties or crop characteristics "on the go". Based on this continuous stream of information, a control system calculates the input needs of the soil or plants and transfers the information to a controller, which delivers the input to the location measured by the sensor (Grisso et al., 2011). Because map‑based and sensor‑based VRA have unique benefits and limitations, some SSCM systems have been developed to take advantage of the benefits of both methods.

6. SECTION CONTROL

Precision agriculture technology has evolved in such a manner that it provides farmers with new and innovative ways to possibly improve profitability. One of these ways is a new approach to the application of liquid chemicals and other inputs known as automatic section control. Automatic section control on a sprayer has the ability to selectively manage input application across the spray boom. This technology utilizes a global positioning system (GPS) to locate the position of the sprayer within the field, and then records the areas covered. If the sprayer traverses an area previously covered it can automatically turn the appropriate section off, therefore eliminating over application. In addition, automatic section control can manage chemical application in undesirable areas such as point rows, waterways, and during headland turns. The largest benefit associated with automatic section control is the reduction in overlapped areas especially prevalent on irregular shaped fields. As a result, this new technology has the potential to increase profits due to reducing input costs. Environmental benefits are also possible due to the ability to manage buffer zones and protect sensitive areas in and around the field. Furthermore, improved efficiency can occur if coupled with a navigational aide such as lightbar or auto‑steer (Shockley, 2010). Once a prototype was created and farm trials conducted, Dillon, et al. (2007) examined the economic implications of utilizing automatic section control. Mooney, et al. (2009) conducted the most recent economic analysis and determined automatic section control became profitable at input saving levels of 11% or above.

7. GUIDANCE SYSTEMS

There are two classifications of guidance systems for agricultural production: guidance aides, and autonomous systems. Guidance aides are devices that provide guidance information to the operator and still require the operator to fully control the machines operations. The intent of the autonomous system is to free up the operator of the machine from the guidance task and thus improve operating efficiency.

7.1. Guidance Aid

The term guidance aide refers to devices that provide guidance information to the operator but do not attempt to replace the operator (Ima and Mann, 2004). One of the most broadly adopted and more recently developed methods to reduce guidance error is the lightbar technology. The main function of lightbar is to assist the machine operator in driving; thus decreasing operator fatigue and minimizing application errors of overlaps and skips. Most lightbar systems include a Differential GPS (DGPS) receiver and antenna, some kind of computer or microprocessor, and a lightbar or graphics display (Stombaugh, 2002).

From the study of cost maps Kayrouz (2009) indicated that inaccuracy of machinery movement, whether in the application stage or the harvesting stag e is very costly. Lightbar gives a visual guide that changes colors as the operator veers off the desired path. Previous technologies were developed with the idea of achieving the same goal, for example, planter markers and foam markers. Both of these previous technologies are becoming shadowed by lightbar and its effectiveness in improving farm profitability. As a result, foam and planter markers are becoming an obsolete technology (Kayrouz, 2009).

Buick and White (1999) explained several reasons why lightbar is replacing foam markers. They mentioned, lightbar is more reliable and more accurate than foam markers, lightbar allows accuracy at higher speeds, lightbar is a possibility with spinner spreaders, lightbar is easy to use, lightbar provides effective guidance over growing crops, lightbar allows operation when visibility is poor, lightbar is less affected by weather, and lightbar has lower recurring costs. Also Griffin et al. (2005) indicates that lightbar is commercially available and promises increased efficiency of field operations.

7.2. Autonomous Field Machinery

According to Haapala et al., (2006) as human capacity to handle simultaneous information is limited new intelligence has been developed between the user and the system to be controlled. The intelligent layer decides which kind of information is passed on for the user and which part is used for other purposes. Development stages of the Human‑Machine Interface are shown in Figure 2. Introduction of small, light‑weight robots that can perform agricultural tasks autonomously may prove to be a realistic option for farmers in the future. These robots will likely operate in fleets and utilize intelligent controls to cooperate with each other to perform tasks such as scouting for weeds and diseases, yield and field mapping, and plant specific operations like sowing and fertilizing (Shockley, 2010). The role of automation should be designed so that automation helps the individual user to simultaneously avoid stress, and increase efficiency of work.

Figure 2. Human‑Machine Interface development stages (Haapala et al., 2006)

Recently, engineers have developed various autonomous machines capable of agricultural production. Several studies have investigated the mechanization and the design of autonomous robots (Blackmore and Blackmore, 2007; Vaugioukas, 2007; Vaugioukas, 2009). The majority of studies have focused on the advancement of autonomous platforms with regard to accuracy, steering, and performance (van Henten et al., 2009; Bak, 2004). Many research conducted in different methods on automated guidance and its algorithm. Bakhtiari et al. (2011) formulated an ant colony optimization (ACO) algorithm which can successfully generate routes that can be followed by farm machinery equipped with automation systems such as auto‑steering navigation systems. Some other advancements on robotic equipment navigation perception technology researches include the work reported by Subramanian et al. (2006) who used a monocular camera to observe citrus grove alleyways, and Nara and Takahashi (2006) who applied a vision system to detect obstacles. Lee and Ehsani (2008) investigated the accuracy of two common laser range finder units. Other studies have concentrated on autonomous weed detection and management (Griepentrog et al., 2009; Gottschalk et al., 2009; Pederson et al., 2007). An overview of the activities in weed control is given in Slaughter et al. (2008). Also the study of agricultural robot application for plant production presumably started with a tomato harvesting robot (Kawamura et al., 1984).

8. DISCUSSION

The use of new precision farming technology allows growers to micro‑;manage individual grids or management zones in a specific field according to its unique production capabilities. In summary, the implementation of PF schemes offers the possibilities for farmers to:

Be more economically competitive
Improve the sustainability of farming
Improve technical functionality
Improve the crop quality and produce more homogeneous qualities
Secure traceability of products through documentation by the system
Be better adapted to comply with regulations regarding use of nutrients and chemicals.

The farming operation is one that involves a significant level of risk and uncertainty. Reduction in government price supports, weather variability, and other uncontrollable environmental factors such as insect and disease infestation contribute to the increased concern of risk management. The term risk refers to the variability of the outcomes of some uncertain activity. Consequently, risk management involves choosing among alternatives to reduce the effects of risk. It is hypothesized by Lowenberg–DeBoer (1999) that precision agriculture technologies are useful in risk reduction. The application of such technologies provides producers with more and better information and increased control of crop growing conditions. Several studies have demonstrated the economic and ecological benefits of PF tools over conventional techniques (Sylvester–Bradley et al., 1999; Silva et al., 2007; Takacs–Gyorgy, 2008). Although, McBratney et al. (2005) conclude that drive assist systems are the only success products in PA since they are not requiring the farmer to make additional decisions but help her/him to reduce them. Drive assist systems give benefits for the farmer since they are easy–to–use and they solve actual important problems, and they return investment costs immediately.

Nowadays environment protection is more and more in focus. Most of the papers reviewed indicate that PA can contribute in many ways to long‑term sustainability of production agriculture, confirming the intuitive idea that PA should reduce environmental loading by applying fertilizers and pesticides only where they are needed, when they are needed (Wang et al., 2003; Bonham and Bosch , 2001; Timmermann et al., 2001; Hatfield, 2000).

The very nature of precision farming and the GPS‑system is to trace each action on the field. As a first objective this feature has enabled the farmer to conduct measurements of variable application of inputs like fertilizers and pesticides. In theory, PF also enables retailers and the final consumer to trace and control each action on the field.

8.1. Adoption

PF is comprised of numerous component technologies that farmers may adopt as a system. Adoption is a learning process where information needs to be collected, integrated and evaluated (Pannell et al., 2006). Organizational arrangements (Leeuwis, 2004) and networks (Allaire and Boiffin, 2004) are also addressed in adoption theory. Also the agricultural advisory centers have an important role in implementation and adoption of PF.

To date, we are in a stationary state between the early adopters and the early majority, mainly since yield increases arent well enough documented to cover the cost of equipment. The adoption of PFTs is likely to follow a normal distribution with the innovators and early adopters as the first to adopt the technology and then later on will the majority of farmers follow up (Lamb et al., 2008). The adoption of PA is currently in a stationary phase between the early adopters and the majority.

In other hand, Kutter et al. (2011) explained that the adoption of PA tools is related to farm size, forms of communication and co‑operation. Also Torbett et al. (2007) found that land tenure, age, and computer use ability had a significant impact on farmer„s perceptions of the importance of PFTs. Larger farms are able to spread PFT costs over more production acres (Solano et al., 2003). Older farmers may be less likely to invest resources in obtaining PF information without the certainty of receiving returns on their investment in the short run (Banerjee et al., 2008). Higher educational levels give farmers higher analytical ability to use information and translate it into a useful input for their decision‑making processes (Just et al., 2002). Daberkow and McBride (2003) reported that mainly young, well‑educated full‑time farmers operating large farms are interested in PA.

Information plays an important role in the adoption of PFTs. According to Jenkins (2009), producers learning about PF tend to use multiple sources of information to increase their knowledge about precision agriculture. Arnholt (2001) found that the single most important motivational factor driving PF adoption is to increase profits. A common conclusion is the fact that the farmers want to use the information and data gained from PF adoption to make better informed management decisions thereby increasing their farm business profits.

Finally, a quodlibet that needs to be considered is mentioned by Olson (1998) who argues that the adoption process for PA is difficult to predict. It is not a single technology but a suite of management strategies, technologies, and practices used to improve agricultural decision making that can be chosen in many different combinations of products and services. Farmers will use them in various combinations depending on variations in geography, production systems, and the farmers themselves.

8.2 Challenge

Precision farming seems to be a logical management tool for farmers, and several manufactures and research groups have put a lot of effort into this technology. Despite the big effort and many promises amongst manufacturers many farmers are still reluctant for the following reasons: PFT has the potential to improve production efficiency but adds complexity to the decision making processes because of the large amount of information to be processed. Thus, the large amount of information available to farmers through different PFTs may require guidance on how this information is incorporated into actual management plans (Griffin and Lambert, 2005). A general concern among farmers is the agronomic decision support, which so far has been unable to cope with the comprehensive agronomic data available today. The decision algorithms in most decision support systems rely on one or few information sources. However, to cope with the complexity of the agronomic process it is necessary to use many complementary sources of information in order to carry out precise and site‑specific application of nutrients. Thus farmers who gather data based on site‑specific tools (GPS, yield mapping and sensors) have a limited number of agronomic models to evaluate this spatial information and thereby adapt their decisions within the field (Thomsen, 2001 as quoted in Pedersen, 2003; Acock and Pachepsky, 1997). Consequently, the positive impact of PF on farm economics has not yet been demonstrated.

Profitability of PF continues to be difficult to predict and uncertain (Atherton, 1999). Some studies have recognized that PF has the potential to be economically profitable, but the profitability depends heavily on the degree of spatial variability within the field according to such attributes as soil types and yield response variability (Roberts et al., 2000). Although, the PA profitability review conducted by Lambert and Lowenberg–DeBoer (2000) found that 73 percent of the studies done on the profitability of PA concluded that adoption of the technology was profitable. But in other view, according to Haapala et al. (2006) usability issues have not been a central issue in electronics development in agriculture. Poor experiences of unacceptable operation could be one reason for the customers not relying on new electronic control systems such as those of PFT.

Although yield maps is a well‑established and relatively low cost method to gather data about field variability, the findings of Goodwin et al. (2002) indicate that farmers and crop advisors should be cautious about using yield maps solely to develop nitrogen application maps. Other complementary field related data are required to optimize the application of fertilizers in plants. PF will not make farming less complex. A management system like PF which heavily depends on data, maps and images is likely to create new concerns about how to communicate this information between the acting parties and presumably also the ownership of data and responsibility of different tasks (Olesen, 2002). Another general concern among farmers is hardware and software compatibility and to choose the right technical systems for conducting PF (Rehnberg, 2002; Pedersen et al., 2001). It is important that different technologies, especially hardware devices, are compatible with other electronic components and systems. The PFTs calibrations are very important and crucial. For example in VRT any change in the equipment performance (boom pressure or etc.) or vehicle speed from that of the calibration results in an application rate different from the planned rate. Moreover, the environmental impacts of precision farming are not yet clear (Stafford, 2000).

There is more to PA than just buying the tools and technologies. As mentioned, PA has the potential to improve efficiency but also adds complexity to the decision making processes. In all steps of precision agriculture technology adoption, decision management is a process or set of processes from incorporation of technical and economical components. Moreover, according to Stombaugh et al. (2001) PA's effectiveness is highly dependent on how much variability exists within your fields and your ability as a producer to identify and put into use the best management practices for each field's sub‑area.

Finally as discussed above, data collected from soil sampling, yield monitoring, crop scouting, remote sensing, and satellite imaging are used to create maps. Many of these maps can be overlaid to look at interactions between yield and topography or yield and soil N content for example. It is the specific ability to process multiple layers of spatial data (yield maps, soil maps, or topography maps) that makes PA a powerful management and decision tool. The availability of historical data combined with multiple layers of information for a farmer engaged in PA improves the quality of inputs recommendations and management decisions.

9. CONCLUSION

Precision farming is information–intense and geographical positioning is the backbone and an essential tool to record all the site‑specific information about the field, weed patches, crop canopy, soil texture and previous yields. This study provides an investigation into various precision agriculture technologies, their benefits challenges and difficulties. Benefits of PA are widely agreed. Today, PA enables farmers to increasingly integrate and take control of the production process in order to improve the profitability of the farm operation and reduce production risk. In spite of its great potential, there are still a significant number of obstacles obstructing the full development of the PA technology and its adoption by a majority of farmers. Profitability of PA is the largest concern listed by farmers, equipment developers, and researchers. In near future, the replacement of large manned machines with small autonomous robots will be a paradigm shift in agricultural production to a small scale precision farming approach.

REFERENCES

Acock, B. & Pachepsky, Y. 1997. Holes in precision farming: Mechanistic crop models. In: Precision Agriculture 1997. Ed. J.V. Stafford. Oxford: BIOS Scientific Publishers LTD., 397–404.
Adamchuck, V.I. & Mulliken, J. 2005. Site – Specific Management of Soil pH (FAQ). University of Nebraska–Lincoln, Extension EC05 – 705.
Adamchuk, V.I., Perk, R.L., & Schepers, J.S. 2003. Application of remote sensing in site–specific management. University of Nebraska, Cooperative Extension. Institute of Agriculture and Natural Resources.
Allaire, G., & Boiffin, J. 2004. Innovation and development: Intensification/disintensification paradigms. Reflections from French experience. In A. Cristovao (Ed.), Proceedings of the 6th European IFSA symposium, Vila Real, Portugal, 2 p.517–525.
Arnholt, M.W. 2001. Evaluating adoption and uses of precision farming technologies. Master's Thesis, The Ohio State University.
Atherton, B.C., Morgan, M.T., Shearer, S.A., Stombaugh, T.S., & Ward, A.D. 1999. Site– specific farming: a perspective on information needs, benefits and limitations. Journal of Soil and Water Conservation, 54(2): 455–460.
Bak, T. & Jakobsen, H. 2004. agricultural robotic platform with four wheel steering for weed detection. Biosystems Engineering 87(2): 125– 136.
Bakhtiari, A.A., Navid, H., Mehri, J., & Bochtis. D.D. 2011. Optimal route planning of agricultural field operations using ant colony optimization. Agricultural Engineering International: CIGR Journal. Vol.13, No. 4. Manuscript No. 1939.