Towards Fully Automated Decision-Making Systems for Precision Agriculture: Soil Sensing Technologies – “The Missing Link”

Towards Fully Automated Decision-Making Systems for Precision Agriculture: Soil Sensing Technologies – “The Missing Link”

Marios Sophocleous (University of Cyprus, Cyprus)
DOI: 10.4018/978-1-7998-5000-7.ch004
OnDemand PDF Download:
No Current Special Offers


As the world population increases, food demands continue with it. This puts the challenge to feed the world in a continuously rising priority position, which requires to be timely and properly addressed. Precision agriculture (PA), a concept that has been around for decades with the potential to address this challenge by maximizing agricultural efficiency and effectiveness, hasn't yet seen worldwide adoption due to missing enabling technologies. Soil sensing technologies able to monitor soil fertility/quality can close the loop in maintaining optimum soil conditions for maximizing crop yield and quality through the implementation of agricultural automated decision-making systems (ADMS). This chapter provides an overview of the existing soil sensing technologies towards the implementation of ADMS and highlights the major challenges in the development of such systems. The chapter continues to give an insight on how different technologies could be combined to form sophisticated systems for agricultural applications.
Chapter Preview


Current and future generations face an ever-increasing challenge concerning food supply. “As the population increases, there is an escalating demand for food, especially fruits and vegetables” (M. Sophocleous & Georgiou, 2017). The world population is projected by the Food and Agriculture Organization (FAO) of the United Nations to reach more than 9 billion people by 2050(FAO, 2009). At the same time, it is agreed by analysts that inter-country and inter-regional inequalities within the developing world would become more pronounced (Zarco-Tejada, Hubbard, & Loudjani, 2014). Furthermore, the projected global economic growth (2.9% annually) would lead to a major reduction of absolute economic poverty in developing countries; however, this will be far from solving the problem of economic deprivation and malnutrition of large parts of the global population (FAO, 2009). These trends suggest that the market demand for food will continue to grow(Alexandratos & Bruinsma, 2012). The rise of biofuels has the potential to alter some of these trends and cause an eruptive upsurgeon food demand that will mainly depend on energy prices and governmental policies(Stafford, 2000). Nowadays, that rural labor force is decreasing and the rise of a huge upcoming bioenergy market is on the horizon, farmers are given the burden to produce more food and fiber to feed the continuously increasing population. Furthermore, agriculture is now required to adapt to new, more efficient, sustainable and environment friendly production methods. The projections show that with a world population exceeding 9 billion people in 2050, overall food production must rise by approximately 70% comparing to 2005/07 production (FAO, 2009). Especially in developing countries, the production would need to double implying major increases in key commodities production. FAO expects that 90% of the global growth in crop production would come from higher yields and increased cropping intensity. Only 10% increase will be achieved by land expansion. Additionally, irrigable land would expand by 11% and harvested irrigated land by 17% with the increase happening entirely in developing countries.Fresh water resources follow a similar trend to the land availability(FAO, 2009; Zarco-Tejada et al., 2014). Overall, the total amount of water resources is sufficient but it is distributed in certain regions. There is an increasing number of countries reaching alarming levels of water scarcity whilst other countries have enough water to provide for more people than the country’s population. However, there are still plenty of opportunities to increase water use efficiency. The potential to improve crop yields, even with the existing technologies is still significant but socio-economic reasons do not encourage the farmers to invest significant amount of money to bridge the gaps.Unfortunately, the same projections show that production increases alone would not be enough to ensure food for everyone(Bongiovanni & Lowenberg-Deboer, 2004). Governments need to ensure food access even to their most vulnerable groups. These projections emphasize the importance of establishing effectivepoverty reduction strategies, safety netsand rural development programs. The problem is further accentuated by healthier diet trends and an increase in veganism. A significant increase in demand of agricultural products puts an even heavier burden on farmers to improve their crop yields (Zhang, Wang, & Wang, 2002).

In the past centuries, in many surviving writings, it is obvious that especially for larger fields the spatial and temporal differences within the same field were highly appreciated. The first seeds of a concept to adopt to these variations were planted back in the 1930s by Linsley & Bauer (Linsley & Bauer, 1929) and possibly also by Eden & Maskell(Eden & E.J., 1928). The technology that actually initiated the progress of this concept was the development of the Global Positioning System (GPS) back in the 1970s. In 1983, the concept of custom prescribed tillage was developed, paving the way to use technology and automation for increased crop production (Stafford, 2000).

Key Terms in this Chapter

Internet of Things (IoT): IoT is defined as the interconnection via the internet of computing devices embedded in everyday objects, enabling them to send and receive data.

In-Situ Sensors (ISS) for Soil: Soil ISS are defined as sensors that are buried in the soil, which are able to monitor soil parameters directly in order to assess soil quality.

Precision Agriculture (PA): The concept of using sensing technologies and automation to respond to special and temporal variations within a field.

Automated Decision-Making Systems (ADMS): Automated decision-making systems (ADMS) are systems, which have the ability to operate a farm/field without the need of human presence or human input. In this context, ADMS are able to monitor soil and plant parameters, assess soil and plant condition and make the right decisions as to maintain those parameters at optimum levels by controlling various inputs.

Machine Learning (ML): ML is the scientific study of algorithms and statistical models that computer systems use to perform a specific task without using explicit instructions, relying on patterns and inference instead.

Soil Quality Index: Soil quality index is calculated by mathematical or statistical frameworks that use a range of soil parameters or indicators to assess soil quality taking into account multiple goals, such as environmental quality, agronomic sustainability, and socio-economic viability.

On-the-Go Sensing Systems (OTGS): On-the-go sensing systems are defined as sensing systems that are installed on moving agricultural equipment, such as tractors that in this context are used to monitor soil parameters in order to assess soil quality.

Sodium Absorption Ratio (SAR): SAR is defined as the ratio between sodium concentration and the multiplication of calcium and magnesium concentrations in soil.

Satellite Imaging (SI): Satellite imaging (SI) is an approach that uses images from satellites to monitor environmental conditions. In this context, SI is used for monitoring soil quality through the image processing of multi-spectral images of specific fields to assess several soil parameters.

Remote Sensing (RS): Remote sensing (SI) is defined as the acquisition of information about an object or phenomenon without any physical contact. This concept is mostly used for observing Earth’s crust and several phenomena on Earth’s atmosphere. Usually, this is achieved be processing of images acquired by satellites.

Soil Quality Indicators (SQIs): SQIs are defined as physical, chemical, and biological properties of soil that is sensitive to disturbance and represents performance of ecosystem function in the soil of interest. Indicators are dynamic soil properties.

Artificial Intelligence (AI): AI is defined as the ability of a digital computer or computer-controlled robot to perform tasks commonly associated with intelligent beings. The term is frequently applied to the project of developing systems endowed with the intellectual processes characteristic of humans, such as the ability to reason, discover meaning, generalize, or learn from experience.

Complete Chapter List

Search this Book: