QUO VADIS PRECISION FARMING

Abstract

The agriculture sector is a unique sector due to its strategic importance for both citizens and economy which, ideally, should make the whole sector a network of interacting organizations. There is an increasing tension, the like of which is not experienced in any other sector, between the requirements to assure full safety and keep costs under control, but also assure the long-term strategic interests of Europe and worldwide. In that sense, agricultural production influences, and is influenced by water quality and quantity, ecosystems, biodiversity, the economy, and energy use and supply. The seasonality and ubiquity of agriculture make agricultural practices and production amenable to efficient synoptic monitoring. The effectiveness of each production, including agriculture, is determined by the ratio of the value of the production outputs to the value of production inputs. For agricultural production the efficiency is affected not only by the internal factors of the production process, but also by external factors (climate, subsidies, the situation in the global market, etc.). The nature of agricultural production does not simply allow pure reducing of energy intensity; in fact the production is affected by many other factors such as: Crop rotation requires several years interval between growing certain crops on the

Full text

QUO VADIS PRECISION FARMING
K. Charvát
a
, T. Řezník
b
, V. Lukas
c
, K. Charvát Jr.
d
, Š. Horáková
d
, M. Kepka
e
, M.
Šplíchal
f
a
BOSC - Baltic Open Solutions Center, Krišjāņa Barona iela 32-7, Riga, Latvia
b
Masaryk University, Faculty of Science, Department of Geography, Brno, Czech Republic
c
Mendel University, Faculty of Agronomy, Department of Agrosystems and Bioclimatology, Brno,
Czech Republic
d
Wirelessinfo, Cholinská 19, Litovel, Czech Republic
e
University of West Bohemia, Faculty of Applied Sciences, Department of Geomatics, Pilsen, Czech
Republic
f
Czech Center for Science and Society, Radlická 28, Praha, Czech Republic
A paper from the Proceedings of the
13
th
International Conference on Precision Agriculture
July 31 – August 4, 2016
St. Louis, Missouri, USA
Abstract. The agriculture sector is a unique sector due to its strategic importance for both citizens and
economy which, ideally, should make the whole sector a network of interacting organizations. There is
an increasing tension, the like of which is not experienced in any other sector, between the
requirements to assure full safety and keep costs under control, but also assure the long-term strategic
interests of Europe and worldwide. In that sense, agricultural production influences, and is influenced
by water quality and quantity, ecosystems, biodiversity, the economy, and energy use and supply. The
seasonality and ubiquity of agriculture make agricultural practices and production amenable to efficient
synoptic monitoring. The effectiveness of each production, including agriculture, is determined by the
ratio of the value of the production outputs to the value of production inputs. For agricultural production
the efficiency is affected not only by the internal factors of the production process, but also by external
factors (climate, subsidies, the situation in the global market, etc.). The nature of agricultural production
does not simply allow pure reducing of energy intensity; in fact the production is affected by many other
factors such as: Crop rotation requires several years interval between growing certain crops on the

Proceedings of the 13
th
International Conference on Precision Agriculture
July 31 – August 3, 2016, St. Louis, Missouri, USA Page 2
same land repeatedly or specific crop sequence is required, Selection of crops variety is affected by
market demand, Agricultural production is greatly influenced by the subsidy rules, Some operations
can be operatively affected by climatic conditions., The use of waste biomass energy potential is
influenced by the structure of crops and the technology used on the farm. Acquiring knowledge about
the energy and carbon intensity of different crops on different lands, how the farm processes work, and
how to take care of the variability within fields in a single farm is very demanding using traditional
approach of farming. Considering large area of agricultural lands, new technologies are demanded for
collecting sensitive data and evaluating these data, No optimization processes can be performed
without sufficient and objective knowledge.
Keywords. Precision Farming, Energy, Environment, Economy, Big data, Knowledge Management

Proceedings of the 13
th
International Conference on Precision Agriculture
July 31 – August 3, 2016, St. Louis, Missouri, USA Page 3
Preface – What is Precision Agriculture/Farming
There exist a number of definitions for what Precision Farming is. Glen C. Rains (Rains, 2009), use
the main focus on economy of production and defined Precision farming (PF) as a management
practice with the potential to increase profits by utilizing more precise information about agricultural
resources. The aspects of precision agriculture are described by Pierce et al. (Pierce & Nowak, 1999)
They defined precision agriculture as “the application of technologies and principles to manage spatial
and temporal variability associated with all aspects of agricultural production for the purpose of
improving crop performance and environmental quality”.
Similar approach was used by Gnip P. (Gnip, Charvat, Holy, & Sida, 2002), where Precision Farming
was defined as a new agriculture technology designed to monitor, analyses and control plant
production to optimize costs and ecological effects. The basic principle is positional control of
fertilization with an accuracy of a few meters. To control the process a huge amount of data has to be
collected and analysed. Important is focus not only on economy, but also on ecology of production.
On CEMA (CEMA) is summarized currently used broader approach for Precision farming, which
includes:
• High precision positioning systems (like GPS) are the key technology to achieve accuracy
when driving in the field, providing navigation and positioning capability anywhere on earth,
anytime under any all conditions.
• Automated steering systems: enable to take over specific driving tasks like auto-steering,
overhead turning, following field edges and overlapping of rows.
• Geomapping: used to produce maps including soil type, nutrients levels etc. in layers and assign
that information to the particular field location.
• Sensors and remote sensing: collect data from a distance to evaluating soil and crop health
(moisture, nutrients, compaction, and crop diseases). Data sensors can be mounted on moving
machines.
• Integrated electronic communications between components in a system for example, between
tractor and farm office, tractor and dealer or spray can and sprayer.
• Variable rate technology (VRT): ability to adapt parameters on a machine to apply, for instance,
seed or fertiliser according to the exact variations in plant growth, or soil nutrients and type.
According to the Oerke & Gerhards (Oerke, Gerhards, Menz, & Herbert, 2010) precision agriculture
relies upon:
• Intensive sensing of environmental conditions in the crop.
• Extensive data handling and processing.
• Use of decision support systems (DSS).
• Control of farm machinery in the field.
In our paper, we would like more extend his understanding of Precision faming and look on Precision
farming like a complex knowledge management combining economic and environmental aspects.
Introduction
The world population will reach 9 billion in 2050. But the land is a resource, which is strictly limited.
Moreover, in reality there is less and less land available for farming due to other anthropogenic
processes. Farmers have to care not only about their production, but also about environment, soil
protection, water protection, biodiversity protection etc. As mentioned above, PF goal is limited
incomes (chemicals, lands, water) and increased production. In order to drive sustainability and
profitability goals, farms have more and more improve their knowledge management, ultimately
increasing the power and influence of the farmer within the entire value chain. Farmers become bigger.
As a result, it is more and more difficult to manage all the required information from numbers of data
sources, which has to be integrated into decision support. Sensors, imagery and other technologies
are all working together to give farmers details about soil content, weeds and pests, sunlight and shade,

Proceedings of the 13
th
International Conference on Precision Agriculture
July 31 – August 3, 2016, St. Louis, Missouri, USA Page 4
and other factors. The challenge, however, is that not only does putting this data to use require a fair
about of synthesis; it is also only applicable to the farm on which the data was collected. Any farm
decision depends on availability of relevant information and knowledge. To improve farming decision
it is necessary to guarantee effective utilization of existing information and knowledge has to be derived
from available data sets. Large datasets are available on different levels, but they are heterogeneous:
in some cases also unstructured, hard to be analysed and distributed across various sectors and
different providers.
Future farm knowledge management systems have to support not only direct profitability of farms or
environment protection, but also activities of individuals and groups allowing effective collaboration
among groups in agrifood industry, consumers and wider communities, especially in rural domain.
When having such considerations in mind, the proposed vision lays the foundation for meeting
ambitious but achievable operational objectives that will definitively contribute to fulfil identified needs
in the long run. A farmer has to have either if able to handle easy to use tools that allow with a few
clicks to solve all these problems or (s)he gets support by new models of farm advisory systems that
are able to solve his needs. The existing extension services are only partly able to keep updated with
the farmer’s common needs. A modern farm management system offers the potential to fundamentally
alter agricultural decision-making. The use of large machinery and hired labor has caused many
farmers to think of large fields as the basic management unit. Information technologies permit the
modern grower to obtain detailed explicit information at a small scale common to farming practices of
earlier times but with considerably more information, enabling them to efficiently manage the land at
these finer scales. The basic principle of modern knowledge management allows more accurate
production.
Future Drivers agriculture production
In order to build vision for future knowledge management in arable farming, there were analyzed
examples of existing knowledge management systems and also drivers, which will have in future
potential influence on farming sector. Within the Future Farm project a trans-European investigation
has led to the definition of the key objectives needed to realize this vision of a new concept of farming
knowledge management respecting changing conditions and demands. FutureFarm (Charvat K. ,
Gnip, Vohnout, & Charvat, 2006) defines next groups of drivers, which will have influence on farm
management and which could also eventually stimulate new demand for knowledge management:
• Climate changes.
• Demographic (Growing population, Urbanisation and land abandonment).
• Energy cost.
• New demands on quality of food (Food quality and safety).
• Aging population and health problems.
• Ethnical and cultural changes.
• Innovative drivers (Knowledge based bio economy, Research and development, Information and
communication, Education, Investment).
• Policies (Subsidies, Standardisation and regulation, National strategies for rural development).
• Economy (Economical instruments, Partnerships, Cooperation and Integration and voluntary
agreements).
• Sustainability and environmental issue (Valuation of ecological performances, Development of
sustainable agriculture).
• Public opinion (Press, International Organisation, Politicians).
The results of Future Farm project analysis (FutureFarm) defines three levels of Knowledge
management on farms:
• Macro level, which includes management of external information, for example about market,
subsidies system, weather prediction, global market and traceability systems, etc.)

Proceedings of the 13
th
International Conference on Precision Agriculture
July 31 – August 3, 2016, St. Louis, Missouri, USA Page 5
• Farm level, which include for example economical systems, crop rotation, decision supporting
system.
• Field level including precision farming, collection of information about traceability and in the future
also robotics.
It is necessary to consider previous analyses for suggestion of future knowledge management system
functionalities and their interrelation. The basic principles of interrelation could be expressed by Figure
1.
Fig 1. Farming information flow
The whole process requires a big amount of data to be collected; this data enables a control of the
whole process and also introduces information about the situation outside the farm (economical
information). For an even better understanding, the whole process requires to improve access to this
data and make analysis of such data. Mathematical and statistical analyses and usage of spatial
information retrieved from this data can bring new quality to the whole process of future farming. The
real end-users of the technologies are not only farmers and agriculture managers, but also advisors
and eventually service organizations. Part of knowledge could also be available for the food industry,
market and direct consumers. The limitation of better utilization of knowledge is, unfortunately, their
limited effective sharing of knowledge. It is necessary to improve access to these data and the
possibility of using new information sources. (Charvat, Gnip, & Krocan)
Knowledge management systems for generation of homogeneous information for traceability transfer
and business as well as integration and management of such information are thus specifically complex
issues in this sector. Therefore, the challenging problem is twofold. Firstly, how to assure the full
security and safety of products when minimizing costs. Secondly, how to provide benefit to the food
sector networks of organisations enabling them to interoperate, to exchange information and data and
to fully integrate miscellaneous business functions along the value chain. These problems (partly valid
for a number of other sectors) are increasingly becoming critical and difficult in the Agri-food sector.
Future challenges
agriXchange (Charvat, et al., 2012) identified a number of future challenges; some from them are
coming from previous activities and are related to PF in broader sense:

Proceedings of the 13
th
International Conference on Precision Agriculture
July 31 – August 3, 2016, St. Louis, Missouri, USA Page 6
• Collaborative environments and trusted sharing of knowledge and supporting innovations in agri-
food and rural areas, especially supporting food quality and security - the concept of the trust
centres has to represent an integrated approach to guarantee the security aspects for all
participants in the future farm. There will be a growing importance of protection of privacy and
IPR. Trust of information is one from the priorities for all rural communities. Pan European social
networks have to support trust centres and enable such technologies as cloud applications and
which will have to guarantee knowledge security - trusted access to information is extremely
important in relation with aspects, which will be discussed in next chapter Big Data.
• ICT applications for the complete traceability of production, products and services throughout a
networked value chain including logistics - to develop world-class network management solutions
that facilitate communication and co-operation between networks of SMEs and large enterprises
in the agrifood and rural development domains. These solutions will enable the management of
food supply chains/networks, virtual and extended enterprises through collaboration and
knowledge exchange – future precision farming has cover not only filed operation, but be focused
of fuel consumption reduction, but also on quality monitoring
• New generation of applications supporting better and more effective management of agriculture
production and decision making in agriculture - Future farm knowledge management systems
have to support not only direct profitability of farms or environment protection, but also activities
of individuals and groups allowing effective collaboration among groups in agrifood industry,
consumers and wider communities, especially in rural domain. The added value chain will play
important role in future PF
• ICT applications supporting the management of natural resources - with better understanding of
the environmental relations, the necessary valuation of ecological performances will become
possible. Pilot projects and best practice samples will be the key to demonstrate for a wide
auditorium the benefits of environmental caretaking. New model of payment of the different groups
of beneficiaries have to be worked out (local, regional, national, continental and worldwide) as
well as the best practice between today’s "government owned" environment or "private owned
with social responsibilities" has to be worked out (Brynjarsson, Stokic, Sundmaeker, & de Juan).
• ICT application supporting adoption of farming production on climatic changes – it is important to
support monitoring of influences of climatic changes on farming production, analysis of influence
of climatic changes in regions and advisory systems for farmers. Monitoring of influence of climatic
changes could be provided using earth observation, sensor measurements, but mainly using
crowdsourcing methods. Important are also social spaces, where farmers will have possibility to
discuss methods of farm adoption – to be able react on climatic changes is important task for
future PF.
• ICT applications supporting agrifood logistic – the focus has to be on the transportation and
distribution of food, sharing online monitoring information from trucks during the transport of cargo,
a flexible solution for on-demand dock reservation and an integrated freight and fleet
management. In general, all the selected applications have the same practical benefits as cost
reduction, better coordination and better information for decision making, and the proactive control
of processes leading to increasing efficiency and effectiveness (smartagrifood. eu) - logistic on
farm level and also food level will influence bot economy and environment.
• ICT application supporting energy efficiency on farm level - the problem of energy efficiency in
agriculture hasn’t been taken into account. Agriculture production uses energy in two ways –
directs energy consumption is in arable farming mainly related to machinery energy consumption,
indirect energy consumption is related to production of chemicals used in agriculture.
Environmental effects of savings on direct and indirect energy use in agriculture are integrally
considered, as energy use efficiency also implies reduction of greenhouse gas emissions.
Applications has to be focused on development of hardware and software solutions supporting
complex energy and chemicals intensity reduction in crop production, operational decision about
protection of crops and logistics optimization in agriculture company (farm) based on advanced
sensor monitoring. By improving the energy and chemicals balance and better targeted protection
of crop production and optimization of logistics system the project will contribute to increasing the

Proceedings of the 13
th
International Conference on Precision Agriculture
July 31 – August 3, 2016, St. Louis, Missouri, USA Page 7
competitiveness of farms not only through cost reduction, but also by increase in labor
productivity.
Big data
What’s the role of Big Data in the farming ecosystem? Farmers need to measure and understand the
impact of a huge amount and variety of data which drive overall quality and yield of their fields. Among
those are local weather data, GPS data, ortophotos, satellite imagery, soil specifics, soil conductivity,
seed, fertilizer and crop protectant specifications and many more. Being able to leverage this data for
running long and short term simulations in response to “events” like changed weather, market need or
other parameters is indispensable for farmers in terms of maximizing their profits. From a regulatory
perspective tracking and tracing products throughout the supply chain or Country of Origin labelling
provides additional Big Data challenges.
Technologies for data gathering and statistics to support the new needs (could be at policy level or
farm level) already exists, applications will be extended with the objective (depending on spatial and
time frequency of the observation) of monitoring / control (indicators-administrative) / land management
/ statistics. Satellite remote sensing depending on the spatial/time resolution will deliver applications
with lower costs and more effectiveness than in the past PF technologies (drones, micro sensors,
positioning systems, portable communication technologies…etc.) will be used more at local level to
improve productivity, decrease input costs and mitigate GHG (Greenhouse gas) emissions.
Now with developing of new satellites (Copernicus Sentinel data) earth observation data available with
higher frequency (e.g.3 days revisit in Europe), higher spatial coverage and high resolution. This data
are also often open and free at world level for everyone. Already now amount of data is at the edge of
big data. There is growing data processing needs in terms of multi-temporal geospatial analysis and
also growing number of external data for Geospatial analysis
Second area generates future big data in Agriculture are IoT (Internet of Technology) including field
sensors and machinery monitoring. The experimentation in FarmTelemetry project demonstrates that
one average Czech farm (i.e. around 1’000 hectares) could generate daily 20 MegaBytes of data. This
could be only for Czech Republic something between 30 and 50 GB per one day. We may easily reach
Terabytes of data a day from agricultural basic monitoring by sensors in Europe. Together with satellite
data agriculture will need to manage extremely large amount of data.
Big data are not only a subject of interest for farmers. Such data are interesting for agribusiness
companies, suppliers and manufacturers of agricultural machinery, weather stations and laboratories,
traders and industry partners and technology and solution providers.
On one side there is growing whole ecosystem with a strong need to secure Big Data from different
repositories and heterogeneous sources. In some cases, sharing of data could be common interest,
but on other side, there are also different interests and data could help to one part of value chain to
take bigger part of profit. From this reason Big data are sensitive topics and trusting of producers about
data security is essential. The producers of seeds and chemicals want to maximize their business with
farmers. Manufacturers of agricultural machinery are collecting data for their purposes. They are
focused on machine optimization monitors, the productivity of the machines and try to figure out how
the machines can be made more efficient and simultaneously farming logistics data helps farmers
control the growing farms and the ever-expanding machine fleet. But this data could be also used by
food industry against farmers.
So there is necessary to build such strategy for Big data in Agriculture, which will support collaboration
in the whole chain, but which will also be trusted by producers. There is necessary to build strategy for
Big data in Agrifood chains: a perspective, which prepared recommendation for future utilisation of Big
data. Big data in Agrifood chain has guarantee a more complex role. It includes food safety and
security, energy, waste management, recycling and the environment. This may lead to increased
information collection and flow and in the needs to use ICTs in a more optimal form. There will be calls

Proceedings of the 13
th
International Conference on Precision Agriculture
July 31 – August 3, 2016, St. Louis, Missouri, USA Page 8
for greater integration of information flows that include data and information on biomass, water and
other natural resources across the agrifood chains to achieve whole systems efficiencies. Current
technologies indicate how the new ICTs and information flows would emerge in this perspective around
use of sensors, sensor networks, Big Data at plot, field, farm, village, region levels on large data clouds,
advanced analytics, the Internet of Things etc. An emerging need in this perspective comprises more
open data an requires Big Data repositories, trust centres for open data and information, cloud based
Big Data and applications and greater bandwidth and speed for connectivity as common infrastructure.
Data privacy will be one from key aspects of future Big Data strategies. When private data will be
managed on Big Data platforms, only a data owner will have the right to decide, who and under which
conditions will have access to the data. To achieve the acceptation of platform by farmers, the data
privacy needs to be guaranteed. For such a purpose we allocate specific task to data privacy. Data
ownership and data privacy is a major issue with regard to data used and produced on the farm, and
forest. Multiple sources of data need to be integrated in order to be useful, and ownership and control
become major issues, quite apart from the technical challenge of allocating value in such complex
scenarios. Data produced by farmers and forest owners needs to be suitably protected both for reasons
of privacy and ownership. This task will combine legal, ethical and technical approaches to make data
usage both possible and attractive to the relevant actors. The starting point for our approach will be to
use semantic architectures for data provenance and trust to enable data to be appropriately tagged
and associated privacy and restrictions to be allocated to data.
Social organization of farmers’ decision-making
Social organization of farmers’ decision-making follows analyses of farming structures in different
countries and the way, how adoption of precision farming is running in these countries. In many regions
Precision Farming was considered an actual, but not a topic of increasing interest. It was stated that
political will and support of these technologies is not demonstrated yet and therefore their full potential
is not exploited.
It was also recognized that agricultural technology firms and private consultants are considered as the
main engine of Precision farming adoption. Interviewees pictured that typical Precision Farming farms
are rather large in size and to be operated by relatively young and highly skilled farm managers.
Important is role of consultants in site specific crop management, who could be consider as
intermediates or partners facing high expectations and pressure. European farmers still prefer phone
to e-mail, but web pages play an increasing role. Farm data is considered sensitive and farmers still
prefer personal contact to their consultants.
The big problems for collaboration are non-compatible solutions. They have forced customers into
purchasing only products of one provider. Compatibility problems have delayed the adoption of site
specific crop management and can still be considered as a most important barrier towards investment.
We therefore assume that when the technology of Precision Farming works trouble-free and economic
benefits can be clearly demonstrated according to the kind of client (cooperative, farm, contractor etc.)
the technology will develop and spread similar to smartphones and become common standard.
Farmers are not searching for hyper-mechanization. Their premise is to register and administer what
is useful and to report the inevitable. Precision Farming is adopted, when economic reasons such as
high input prices or environmental regulations are in favour and/or certain barriers are removed.
Introduction of site specific technology also happens by evolutionary replacement of old machines.
New machinery is increasingly equipped with site specific on-board technologies. Integrative and easy
to handle solutions are needed. The critical discussion on the possible ecological benefits of PF and
its practicability should be deepened. (Charvat K. e., 2010)
The study provided by Ganicky at al. (Ganicky) solve question, when precision farming could be
profitable. This question is difficult to answer by any published profitability review because there are:
• incompatible approaches to economic analyses;
• costs that are often overlooked, and

Proceedings of the 13
th
International Conference on Precision Agriculture
July 31 – August 3, 2016, St. Louis, Missouri, USA Page 9
• benefits with ill-defined value.
To see what price a farmer has to pay to switch from traditional (conventional) to Precision Farming,
let's analyse how effective are the investments to Precision Farming and which of them are
unavoidable. Economically effective management of within-field variability means in other words that
well trained farm manager does correct decisions based on complex information and these decisions
are precisely implemented. High quality information is the basis for effective management. Therefore
initial investments into information like boundary mapping, soil sampling and management zones
identification, GIS mapping etc. is also unavoidable. These investments into information should be
viewed as durable and their costs should be amortized as a fixed cost over a number of years.
Implementing the decisions of farm management means in fact to operate the fields. All costs of this
type are considered to be variable. These costs are unavoidable. To operate fields, appropriate
Variable Rate Technologies (VRT) and other types of technology are required, like e.g. GPS-receiver,
yield monitor, computer, GIS and other software, VRT application equipment etc. All this equipment
represents durable capital investments. Furthermore, there are other fixed costs like depreciation,
interest on investment, insurance costs associated with durable capital (like the abovementioned
equipment). These investments are avoidable. The investments and fixed costs associated with
purchasing VRT application farm equipment usually constitute substantial part of all investments and
costs encountered by a farmer when adopting Precision Farming. In any case, effective use of PF
management may require the development of the knowledge base, experience and accumulating
information about fields and their productivity over several years. In all cases mentioned above a farmer
may elect to custom hire the VRT equipment, yield monitor and other technology like GPS together
with the consulting services from specialized companies. Such companies in general are better
equipped with modern VRT machines, have highly qualified specialists and can offer full service (for
example GPS field boundary mapping, soil sampling and management zone establishing, fertiliser
recommendation, fertiliser prescription and VRT application). Such operating leases are offered on a
variable cost basis - priced per hectare or per day of operation. For smaller farms, and in any case for
a novice to PF management, this way of operating fields offers the optimal and at the same time the
least expensive possibility. The model, which can reduce fixed cost for farmers and make PF profitable,
is model based on outsourcing. Farmers buy services from service organisation, fixed cost are changed
on variable cost. Technology on side of service organisation used for longer period and more
effectively, so it reduces cost for the process.
Standardization
It is important to focus on standardisation for a wider adoption of precision farming. Standardisation
importance will grow for interconnectivity of different levels of farming knowledge systems. It is
important to support the development of machine-readable protocols, and standards to integrate
management information systems with policy tools. Major priorities for future knowledge systems will
be the integration and orchestration among services based on semantic integration of collaborative
activities, including semantic compatibility for information and services, as well as ontologies for
collaboration. (Charvat, Gnip, & Mayer) (Charvat K. , Gnip, Krocan, Spyros, & Mayer)
More information on standardisation within the Precision farming domain may be found in a separate
paper. (Řezník, et al.)
Contribution of FATIMA and FOODIE towards precision farming
Fatima
The challenge for sustainable crop production is to achieve optimized yield (in quantity and quality)
and farm income with a minimum of inputs (nutrients, water, energy, pesticides, herbicides, labor,
money), while preserving and protecting the environment and social fabric. We focus here on nutrients

Proceedings of the 13
th
International Conference on Precision Agriculture
July 31 – August 3, 2016, St. Louis, Missouri, USA Page 10
and water, while maintaining an integrated perspective of all factors. FATIMA to this challenge is plant-
and people-centered at the same time and both are linked.
“People-centered” means that we first set out to learn and understand how people (farmers, rural
communities, agribusinesses) grow food and how people (consumers, land and water managers,
decision makers, policy makers) act and react in the given governance structures. Only then can we
co-create and co-develop (together with them) technological tools, socio-economic models, and policy
instruments that can assist them in their tasks. Only this way leads to acceptance by and co-ownership
with users and other actors, which is the basis for wide uptake of any technology.
“Plant-centered” means that we look at how the plant/crop grows at the center of a complex web of
external drivers (like climate, weather, soil), inputs (nutrients, water, also energy, pesticides), internal
growth factors (cultivar, variety, soil fertility), and outputs (yield quantity and quality), all of these
spatially distributed and evolving with time. The accurate knowledge of the actual plant status in each
moment and at each location is the key to drive optimum management and to apply more effectively
and efficiently the knowledge that the scientific-technical basis puts in our hands.
The main target of FATIMA is to develop Earth observation (EO) and wireless sensor network (WSN)
assisted pre-operational tools and services for effective and efficient precision farming and agri-
environmental management. Methods are target on increasing nutrient efficiency means reducing
external inorganic inputs, as well as water in the case of irrigated crops, while maintaining or increasing
yield (both in quantity and quality). FATIMA focused on improvement of ground and surface water
quality will be achieved by reducing inorganic inputs and saving water thus reducing diffuse pollution
to a minimum and keeping recharge at stable levels. This direct (local) improvement will be potentiated
by the non-local effect of reduction of the primary energy embedded in inorganic inputs and in water
delivery. It supports reduction of soil contaminations with toxic compounds and heavy metals will
equally be achieved by reducing inorganic inputs and by additionally following good practices on
selecting “clean” source of these inputs (depending on where they are manufactured, they may contain
more or less heavy metals). Important is also conservation of biodiversity and wildlife will depend on
healthy ecosystems. Saving water in crop production will leave the necessary water for ecosystems
maintenance; healthier soils will also protect biodiversity. Reduction of soil erosion and improvement
of soil quality and structure will be achieved again by the reduction of inorganic inputs and by the
measures derived from. E.g. Fine-tuned spatial and temporal coupling of N-cycles will help maintain
soil physico-chemical properties; less contamination (in combination with soil remedial measures) will
restore soil water retention capacity and reduce external nutrient and water needs. Spatial and
temporal coupling of the N cycle means protection of the environment by drastically reducing i) nitrate
leaching below the rooting depth and thus ground-water contamination, ii) greenhouse (nitrous oxide)
emissions to the atmosphere, iii) disturbance of background values of soil chemical properties (nitrates,
electrical conductivity, pH) that can otherwise increase oxidation of soil organic matter and offset soil
fauna and microbial communities. Last goal is improving human health, through the reduced release
of pollutants and GHGs, will again be achieved by reducing inorganic inputs. There is a double
feedback in here: firstly the reduction of inorganic inputs directly reduces the contamination of the
environment and the agricultural products; secondly, through the energy embedded in these inorganic
inputs it leads to a reduction in primary energy (oil, gas) consumption and the pollutions and GHG
emissions associated with primary energy production.
FOODIE
The FOODIE project focused on building an open and interoperable agricultural specialized platform
hub on the cloud for the management of spatial and non-spatial data relevant for farming production;
for discovery of spatial and non-spatial agriculture related data from heterogeneous sources;
integration of existing and valuable European open datasets related to agriculture; data publication
and data linking of external agriculture data sources contributed by different public and private
stakeholders al lowing to provide specific and high-value applications and services for the support in
the planning and decision-making processes of different stakeholders groups related to the agricultural

Proceedings of the 13
th
International Conference on Precision Agriculture
July 31 – August 3, 2016, St. Louis, Missouri, USA Page 11
and environmental domains. The
main idea
of FOODIE project is to:
•
build an open and interoperable agricultural specialized platform hub on the cloud for the
•
management of spatial and non-spatial data relevant for farming production
•
discovery of spatial and non-spatial agriculture related data from heterogeneous sources
•
integration of existing and valuable European open datasets related to agriculture
•
data publication and data linking of external agriculture data sources contributed by different
public and private stakeholders allowing to
•
provide specific and high-value applications and services for the support in the planning and
decision-making processes of different stakeholders groups related to the agricultural and
environmental domains.
which is conceptualized in the following architecture diagram of FOODIE service platform (Figure 2).
(Charvat, et al., 2014, May).
Fig 2. FOODIE service platform hub in cloud
Examples of potential new types of Precision Farming services
Tactical planning
Main research is currently focused on basic precision farming methods, which are typically related to
operative decision on the field level. On other side it was documented, that right decision on the base
of farm level and level of external (prices, cost, worldwide yield, subsidies) information has the main
influence on profitability of farm. The experiences from previous years demonstrates that using
precision farming methods a famer could increase his profitability approximately from 15 to 25 percent.
The decision about optimal production mix and used methods of production could increase farmer’s
profit from 100 to 300 percent. From this reason it is necessary to introduce new knowledge
management and access to information on all levels, which has to support such decision.
Tactical planning was a concept introduced as part of COIN project (Krivanek, Charvat, Gnip, Vohnout,
& Charvat) and it was focused on introducing new methods oriented on tactical planning for next
seasons. This concept is till now not fully operational, because it requires access to large amount of
data, which are now not available, but this could well demonstrate potential advantage of Big data

Proceedings of the 13
th
International Conference on Precision Agriculture
July 31 – August 3, 2016, St. Louis, Missouri, USA Page 12
approach. The main purpose of tactical planning is to recommend optimal production and land use for
next seasons to maximize expected profit. The problem could be described such as a farmer has a
number of parcels with different properties available for his production. It is possible to grow different
kinds of products on these lands. On each of these lands it is possible to grow maximum one product
using one of two methods. Due to many reasons, for example crop rotation and field localization, it is
not possible to grow each of the products on each of the lands in next season. For formulation and
solving the problem we need various input data. The factors affecting the profit could be:
•
prices at which the farmer will be able to sell his/her products,
•
subsidies for plant production or different use of land,
•
yields / hectare of particular crops,
•
costs including several items:
o The costs for growing different crop kinds depending on land properties and land area. A
lot of items like seed, fertilizer, operating costs and labour costs can be included into these
costs.
o Fixed costs connected with each kind of crop.
o Fixed costs of using variable fertilizer application, which have to be included, if farmer uses
variable application on any of his lands.
o Fixed costs of variable fertilizer application on particular lands.
The goal of this problem is to create recommendation saying which crop should be sowed on each
land and which way of their growing should be chosen. The optimal production plan for the given input
data is described by optimal solution of this problem. In terms of mathematical programming solving
this problem means finding optimal values of all variables, which are used in this model. Especially in
case of farmers, who were using precision agriculture in the past, it will be possible to estimate many
data values with quite high accuracy. Most of the information will not exist in the farm information
system in the form requested by the model. To get the values of some data it will be necessary to
create procedures, which will calculate the values from available data.
Sensor Observations and Measurements
Importance of sensors in the field of Precision Farming is still growing. Utilization of sensors in PF can
be divided into two basic groups. First group contains static in-situ sensors mainly used for observing
agro-meteorological data in the fields. Another group can be called as mobile and contains sensors
connected to agricultural machinery and used for tracking. Example of the utilization of mobile sensors
will be described in next paragraph FarmTelemetry. Important issue for usability of sensor data in PF
is data exchangeability between solutions or systems. There is difference between standardized and
used protocols or formats.
Observing of agro meteorological phenomena on the level of individual farm provides more detailed
information and accurate description of weather, water and soil conditions than results from
observations on regional or more global level. Providing this kind of information to farmers continuously
helps them to make substantiated decisions. Example of client for agro meteorological observations is
shown on Figure 3. Continuous collecting of data produces large volume of measurements where it is
necessary effective storing and proper analysing. There are two types of information from static in-situ
sensors. First type is current state of conditions on the farm and on fields that can be constructed from
last observed values or short-term series. Another type is seasonal state of conditions that demands
on long term series of observations. Providing both types requires well designed data model on one
side and suitable visualization methods on another side. Integration of data produced by in-situ sensors
with machinery tracking data and LPIS block information during Foodie project provides complex
analyses with high added value results.

Proceedings of the 13
th
International Conference on Precision Agriculture
July 31 – August 3, 2016, St. Louis, Missouri, USA Page 13
Fig 3. Client for agro-meteorological observations
Farm Telemetry
The Farm Telemetry system is focused on monitoring of activities and utilization of individual tractors.
Data collected by sensor units on tractor allow possibility of analyses and obtaining overall overview
for individual farmer’s fields (blocks). Application currently supports following tractor-oriented analyses:
•
Cultivated blocks – provides list of all farmer’s blocks, where the selected tractor was working
during the selected day. There is provided information for each of the blocks about spent time and
about used passive machinery. Information about average fuel consumption (litres per hour) can
be provided, if current settings of active unit supports reading this information.
•
Utilization – provides information about time the tractor spent by working on farmer's blocks, by
moving on other places and by standing in place during selected day. The sum of these times is
always 24 hours.
•
Activities log – provides detailed information about activities of the selected tractor during the
selected day. The log contains start time and end time of each activity, location of the activity,
used passive machinery and sum of fuel consumption during the activity. Level of detail of the log
can be customized by user. User can adjust values of three attributes:
o
Minimal work time – if the sum of movement times on farmer’s block doesn’t reach at
least this value during selected day, the movements on this block are considered as
“other movements”, not work.
o
Minimal pause time – shorter stops aren’t listed explicitly in the log. They are only
counted as sum of delays during movement or work.
o
Allowed outside field time – if the time outside field (for example during turning at the
edge of the field) doesn’t reach at least this value, leaving the field (block) is not listed
in the log.

Proceedings of the 13
th
International Conference on Precision Agriculture
July 31 – August 3, 2016, St. Louis, Missouri, USA Page 14
•
Overview of activities at field (block) – provides monthly overview for selected field. The overview
shows information about sum of times each tractor has spent on the selected field including used
passive machinery and sum of fuel consumption.
The field trials during season demonstrate that Farm Telemetry will deal with real Big data. This
required optimization on the level of database structures and communication protocols. The main
problem is that data from every farm has to be stored minimally for full season, to guarantee complex
analysis of costs.
Measuring unit with connected sensors is mounted on active machinery. Current values of sensors are
read in case of position change or every 2 seconds of running engine of machinery. If measuring unit
is connected to the Internet, sequences of observed values are sent to the server. If the connection is
not available, values are stored to the RAM on the unit.
Number and type of collected data depends on control units of specified active machinery (e.g.
CANBUS) what values allow to be read (fuel consumption, engine load, speed etc.). From this reason
each measuring unit is producing different number of values, but in most cases the basic set of values
contains same phenomena. Example of analyses results can be seen in FarmTelemetry application
client on Figure 4.
Fig 4. Farm Telemetry client with examples of analyses results
Delineation of management zones based on satellite remote sensing
Yield potential zones are areas with the same yield level within the fields. Yield is the integrator of
landscape and climatic variability and therefore provide useful information for identifying management
zones (Kleinjan, Clay, Carlson, & Clay, 2007). This presents a basic delineation of management zones
for site specific crop management, which is usually based on yield maps over the past few years.
Similar to the evaluation of yield variation from multiple yield data described by Blackmore et al.
(Blackmore, Godwin, & Fountas, 2003), the aim is to identify high yielding (above the mean) and low
yielding areas related as the percentage to the mean value of the field. Also the inter-year spatial
variance of yield data is important for agronomists to distinguish between areas with stable or unstable

Proceedings of the 13
th
International Conference on Precision Agriculture
July 31 – August 3, 2016, St. Louis, Missouri, USA Page 15
yields. The presence of complete series of yield maps for all fields is rare, thus remote sensed data
are analysed to determine in field variability of crops thru vegetation indices.
The estimation of yield potential zones from multi-temporal satellite data is established as the general
model in FOODIE platform. As the main data source, ESPA repository of LANDSAT satellite images
is used, which offers surface reflectance products, main vegetation indices (NDVI, EVI) and clouds
identification by CFmask algorithm. A selection of scenes from recent 8 years is made for specific farm
area to collect cloud-free data related to second half of vegetation period. Yield potential is calculated
for separate scenes as the relation of each pixel to mean value of whole field. In last step all scenes
are combined and median value of yield potential is calculated. After fully operation of Sentinel 2A/B
satellites, calculation of yield potential will be enhanced by these vegetation products.
Fig 5. Map of yield potential delimitated from multi-temporal Landsat imagery
Conclusions
As visible as a red line throughout the whole paper, we are standing at the beginning of a new era for
precision farming. Technological as well as personal shifts enable to obtain more data than ever before
which is the real challenge of the Big data concept. The discussion is not about having floods of data,
but even more about how to obtain more valuable information from such data flood than from poorer
but clearer data sources.
Current approaches in precision agriculture have a common denominator, which is standardized and
interoperable geospatial information. Geospatial information, i.e. knowing where something happened,
is exactly the kind of information that stimulates all the current approaches, no matter whether it is
variable rate application or an automated steering system. Only standardized and interoperable data

Proceedings of the 13
th
International Conference on Precision Agriculture
July 31 – August 3, 2016, St. Louis, Missouri, USA Page 16
are those that may be combined in space (from farm to farm) and time (from season to season).
Standardized and interoperable data are required for operative decisions, i.e. ad hoc management.
Such requirement is even more important for the long-term visions. Economic and ecologically sound
long-term decisions require integration of much more data sources than in the case of ad hoc
management. Otherwise, obstacles of manual integration of large amounts of data exceed its benefits.
It is clear that not all the precision farming stakeholders may benefit from such progress. For instance,
minimizing the environmental burden also means using less chemicals as incomes. This is certainly
positive for a farmer and also the whole society, however negative for a producer and reseller(s) of
such chemicals who are used for expanding precision farming market. Shift to the new era of precision
farming comes at a price that is for some stakeholders bigger than for others.
An open issue lies in the area that affects Big Data in all its forms. Farmers in particular commonly
distrust the companies aggregating data. Farmers are afraid that their sensitive data may be misused.
Future efforts should, therefore, focus on the technological as well as the personal level in order to
ensure that the FOODIE platform remains useful in daily life. In other words, the geospatial technology
is ready to help minimise the environmental burden while maximising the economic benefits; the main
obstacles now consist in insufficient policies and people.
Acknowledgements
The research reported in this paper has been supported by the EU FArming Tools for external nutrient
Inputs and water Management - FATIMA, Horizon 2020 project, H2020-SFS-2014-2 EU, project
number 210177428 and Farm Oriented Open Data In Europe - FOODIE project (http://foodie-
project.eu/, CIP-ICT-PSP-2013-7, Pilot B no.621074)
References
Blackmore, S., Godwin, R. J., & Fountas, S. (2003). The Analysis of Spatial and Temporal Trends in Yield Map Data over Six
Years. Biosystems Engineering.
Brynjarsson, B. A., Stokic, D., Sundmaeker, H., & de Juan, J. (n.d.). Collaborative Work Environments – ICTs supporting Agri-
food businesses and Rural development. DRAFT Strategic Research Agenda.
CEMA. (n.d.). Retrieved from Precision Farming: key technologies & concepts: http://cema-agri.org/page/precision-farming-
key-technologies-concepts
FutureFarm. (n.d.). Retrieved from http://www.futurefarm.eu
Ganicky, a. a. (n.d.). WP2 Process Model Development, PREMATHMOD, IST–2000-28177. STATISTICAL AND
MATHEMATICAL MODELLING, DATA ANALYSIS, SIMULATION AND OPTIMALISATION METHODOLOGIES FOR
PRECISION FARMING.
Gnip, P., Charvat, K., Holy, S., & Sida, A. (2002). Precision farming trough Internet and mobile communication. 6th International
Conference on Precision Agriculture and Other Precision Resources Management.
Charvat, K. e. (2010). Vision Statements and Road-Map Methodology for Knowledge Management Adoption. AGRIS on-line
Papers in Economics and Informatics 2.4 (2010): 47., p. http://online.agris.cz/.
Charvat, K., Esbri, M. A., Mayer, W., Campos, A., Palma, R., & Krivanek, Z. (2014, May). FOODIE - Open data for agriculture.
IST-Africa Conference Proceedings, 2014 (pp. 1-9). IEEE.
Charvat, K., Gnip, P., & Krocan, M. (n.d.). Integration of Farm Management Information Systems to support real-time
management decisions and compliance of management standards. Deliverable 1.2.1 Revision: Final Knowledge
Management Methods.
Charvat, K., Gnip, P., & Mayer, W. (n.d.). FutureFarm vision. AGRIS on-line Papers in Economics and Informatics, p.
http://online.agris.cz/.
Charvat, K., Gnip, P., Krocan, M., Spyros, F., & Mayer, W. (n.d.). D1.2.3 Visions and recommendations for knowledge
management. Report from FUTUREFARM project.
Charvat, K., Gnip, P., Vohnout, P., & Charvat, K. J. (2006). VISION FOR A FARM OF TOMORROW. IST Africa. Gabarone,
Botswana.
Charvat, K., Horakova, S., Wolfert, S., Holster, H., Schmid, O., Pesonen, L., . . . Mildorf, T. (2012, 11 28). Common Basis for
policy making for introduction of innovative approaches on data exchange agri-food industry. Final Strategic Research

Proceedings of the 13
th
International Conference on Precision Agriculture
July 31 – August 3, 2016, St. Louis, Missouri, USA Page 17
Agenda (SRA).
Kleinjan, J., Clay, D. E., Carlson, C. G., & Clay, S. A. (2007). Productivity zones from multiple years of yield monitor data. In F.
J. Pierce, & D. C. Clay, GIS applications in agriculture. CRC Press, Boca Raton.
Krivanek, Z., Charvat, K. J., Gnip, P., Vohnout, P., & Charvat, K. (n.d.). New Tools for Data Accessibility and Decision Support.
IST Africa.
Oerke, E. C., Gerhards, R., Menz, G., & Herbert, G. W. (2010). Precision crop protection - the challenge and use of
heterogeneity. Dordrecht; Heidelberg [u.a.]: Springer.
Pierce, F. J., & Nowak, P. (1999). Aspects of Precision Agriculture. Advances in Agronomy.
Rains, G. C. (2009). Precision Farming An Introduction. p.
http://athenaeum.libs.uga.edu/xmlui/bitstream/handle/10724/12223/B1186.pdf?sequence=1.
Řezník, T., Lukas, V., Charvát, K., Charvát, K. J., Horáková, Š., & Kepka, M. (n.d.). FOODIE DATA MODEL FOR PRECISION
AGRICULTURE.
smartagrifood. eu. (n.d.). Retrieved from Smart Food and Agribusiness: Future Internet for Safe and Healthy Food from Farm
to Fork, D700.2: http://www.smartagrifood.eu/sites/default/files/content-files/downloads/SAF_D700-
2_V011_Final_0.pdf