Ebook: Agriculture and Environment Perspectives in Intelligent Systems

For the last decade, intensive and integrative studies have been performed by applying information technology to the research fields of agriculture and environmental sciences. This research trend aims to improve the agricultural process using information technology and at the same time making this process more sustainable. Applying sensor technology and combining information technologies creates the opportunity to obtain more precise information on agriculture processes and environmental phenomena. Indeed, the Internet of Things paradigm opens new frames to collect large amounts of agricultural and environmental data through different kinds of sensor networks [1].

The agricultural control starts with analyzing the current situation using chronological data. Traditionally, growers observe plants and detect problems based on their experience and symptoms of diseases. Most of such cases, the detection critical point is usually late in sense of optimization. An automatic method to detect symptoms and provide an early warning would at least help to improve the efficiency of agricultural processes. The correlations between environmental factors are the keys for evaluating agricultural characteristics. In this sense, a large amount of heterogeneous data is essential for integration studies of automated agriculture. There is a high demand for a large quantity of accurate spatiotemporal data for automated agriculture. Whilst the digitalization of the agriculture is helping to fulfill this demand, new challenges are posed related to the intelligent management of the data gathered from sensors to help farmers and environmental managers make decisions. It is in these challenges where the Intelligent Environment (IE) paradigm comes into play to guide the design of systems, techniques and algorithms able to analyze the data and provide recommendations that can be put into practice by farmers, managers and other stakeholders.

Some examples of applications of IE in this research field range from smart farming (see for example [2–4]) to intelligent applications for ecological disaster management (see for example [5] where a social sensor is developed for detecting floods through natural language processing (NLP)). Specific examples of IE applications in agriculture can be found for resource allocation and decision making in agricultural systems [6,7], farming system design [8] or data fusion for characterizing key agricultural system attributes [9]. Likewise, other IE proposals for environmental management can be found in different areas such as waste management [10] or environmental assessment and rehabilitation [11]. It is also noteworthy the increasing number of conferences and workshops devoted to the development of IEs in agriculture: approximately 30 events were found in the WikiCFP webpage in 2018 tagged with “agriculture” and “intelligent environments” as keywords, as for example the International Workshop on Intelligent Systems for Agriculture Production and Environment Protection (ISAPEP).

From our experience, we believe there is still a need for bridging the gap among the needs of farmers, environmental managers and stakeholders in the agriculture and environmental sectors and the solutions offered by the Information and Communication Technology field. The eventual goal of applying IE to agriculture is to replace, or significantly reduce, the human labor for the agricultural production. Most of contemporary research on Artificial Intelligence (AI) technology focuses on finding methods to simulate human perception, cognition and reasoning process. However, still a long way to reach the human ability remains. When the AI technology reaches maturity, the dramatic transformation on devices, equipment and machinery of agriculture production will begin. The motivation of editing this book was to foster the collaboration among researchers from computer science, agriculture engineering and environmental science fields who are in a privileged position to provide new and valuable insights on intelligent applications for agriculture and environmental management systems. As a result, this book presents several reviews and examples of the latest developments in the application of IEs to agriculture and environmental sciences through 5 chapters.

The editors of this book would like to thank the authors who contributed articles to this initiative. The diversity of the topics included in the chapters showcases the relevance in the research and development of Intelligent Environments applied to agriculture and environmental problems. We look forward to a future where IE will make a difference in creating efficient and really helpful systems for farmers, environmental managers and the entire society.

Outline of the book

Chapter 1 explores the use of intelligent systems in Controlled Environment Agriculture (CEA) facilities as an alternative to offer low cost sensing systems to gather data on how plants evolve in these facilities. At the moment, data collected in CEAs are restricted to recording the general environmental control responses. In this chapter the author relates his experiences within a farm company when developing more fine-grained methods to gather data, especially focused on those related to growing methods such as hydroponics or aeroponics and on intelligent techniques for measuring nutrients.

Chapter 2 reviews the adoption of intelligent systems in the research field of biomonitoring. In this chapter is described the evolution of environmental information systems and their application in biomonitoring, starting from the use of biosensors to the analysis of current environmental information systems to the introduction of intelligent techniques and methods in these systems. Several examples of applications of intelligent biomonitoring systems are analyzed, from systems to monitor algae to applications for monitoring fish. The authors also include a discussion on the limitations and future perspectives of intelligent biomonitoring systems.

Chapter 3 proposes an intelligent system to acquire and preprocess data for precision agriculture applications. It is composed of two modules, the first aimed to measure different environmental factors on the target plants and the second for refining time-series raw data previously collected in order to reduce noise in the data. The main goal of this system is to facilitate the acquisition of different type of data to farmers, since it allows any analog sensor to be connected with minimal effort of calibration. In this manner, the system is presented as a user-friendly device for multi-data collection which may help in the creation of big data sources for new applications.

Chapter 4 illustrates the use of intelligent algorithms to make a more efficient use of a scarce resource such as water. These algorithms predict the intake of pharmaceutical contaminant elements such as diclofenac in reclaimed water for irrigation of lettuces. As a result, this proposal allows farm managers to automatically control the quality of the reclaimed water and effectively use it when it is safe. The authors compare several machine learning algorithms to this end, proving that the Random Forest algorithm can reach a 97% of fitness when predicting the presence of contaminants in the water.

Chapter 5 focuses on the generation of intelligent models to predict frosts in crops in Southeastern Spain. These models not only use data from physical sensors placed in the land or trees, but they also rely on the utilization of open data from weather stations. By using different rule and decision trees techniques, the authors are able to identify the most relevant features for predicting frosts, such as dew point, vapor pressure deficit and maximum relative humidity. The results show that the proposed model is able to predict a frost with 98% of confidence and obtain an error of less than 0.5°C. in the prediction of the minimum temperature.

Andrés Muñoz (Andrés Muñoz would like to thank the Spanish Ministry of Economy and Competitiveness for its support under the project TIN2016-78799-P (AEI/FEDER, UE)) and Jaehwa Park

References

[1] Ray PP. Internet of things for smart agriculture: Technologies, practices and future direction. Journal of Ambient Intelligence and Smart Environments 9(4). 2017, 395–420. doi: 10.3233/AIS-170440

[2] Aziz IA, Ismail MJ, Haron NS, Mehat M. Remote monitoring using sensor in greenhouse agriculture. In 2008 International Symposium on Information Technology, Vol. 4, IEEE. 2008, 1–8. doi: 10.1109/ITSIM.2008.4631923

[3] Aquino-Santos R, González-Potes A, Edwards-Block A, Virgen-Ortiz RA. Developing a new wireless sensor network platform and its application in precision agriculture. Sensors 11(1). 2011, 1192–1211. doi: 10.3390/s110101192

[4] Muñoz A, Soriano-Disla JM, Janik LJ. An Ontology-Based Approach for an Efficient Selection and Classification of Soils. In Analide C. and Kim P. (editors), Intelligent Environments 2017: Workshop proceedings of the 13th International Conference on Intelligent Environments, Vol. 22, IOS Press. 2017, 69–79. doi: 10.1016/j.biosystemseng.2018.11.016

[5] Periñán-Pascual C, Arcas-Túnez F. Detecting environmentally-related problems on Twitter. Biosystems Engineering 177. 2019, 31–48. doi: 10.1016/j.biosystemseng.2018.10.001

[6] Jiang Y, Hao K, Cai X, Ding Y. An improved reinforcement-immune algorithm for agricultural resource allocation optimization. Journal of computational science 27. 2018, 320–328. doi: 10.1016/j.jocs.2018.06.011

[7] Kaim A, Cord AF, Volk M. A review of multi-criteria optimization techniques for agricultural land use allocation. Environmental Modelling & Software 105. 2018, 79–93. doi: 10.1016/j.envsoft.2018.03.031

[8] Robert M, Dury J, Thomas A, Therond O, Sekhar M, Badiger S, Ruiz L, Bergez JE. CMFDM: A methodology to guide the design of a conceptual model of farmers’ decision-making processes. Agricultural Systems 148. 2016, 86–94. doi: 10.1016/j.agsy.2016.07.010

[9] Jiayu Z, Shiwei X, Zhemin L, Wei C, Dongjie W. Application of intelligence information fusion technology in agriculture monitoring and early-warning research. In 2015 International Conference on Control, Automation and Robotics, IEEE. 2015, 114–117. doi: 10.1109/ICCAR.2015.7166013

[10] Anagnostopoulos T, Zaslavsky A, Kolomvatsos K, Medvedev A, Amirian P, Morley J, Hadjieftymiades S. Challenges and opportunities of waste management in IoT-enabled smart cities: a survey. IEEE Transactions on Sustainable Computing 2(3). 2017, 275–289. doi: 10.1109/TSUSC.2017.2691049

[11] Sun AY, Zhong Z, Jeong H, Yang Q. Building complex event processing capability for intelligent environmental monitoring. Environmental Modelling & Software 116. 2019, 1–6. doi: 10.1016/j.envsoft.2019.02.015

The need to increase the supply of food for larger and increasingly urbanised populations has created what many see as an entirely new sector based around reducing the distance salad crops travel. This usually involves growing delicate crops in enclosed spaces where the environment is controlled to give what is thought of as ideal growing conditions. In reality, since plants have evolved over many millennia, the process of growing a plant needs many of the variations nature delivers. We are only just beginning to learn what this means for the human diet. It is now apparent growing conditions cannot be changed substantially without changing some known and unknown qualities. This chapter looks first at what is being offered by those investing in Controlled Environment Agriculture (CEA) and then how, if done properly, the use of Intelligent Systems can unlock much greater benefits and then go further by informing some aspects of traditional agriculture. CEA needs to look further at why plants evolved as they did and use Intelligent Systems to build that knowledge and make effective use of it. Some aspects of traditional agriculture are using technology to deliver far greater granularity of data than those who grow in environments more usually found in a laboratory. The overwhelming majority of CEA practioners talk of data but then do little more than record environmental setpoints and rely on measuring Electrical Conductivity (EC) as a proxy for assessing nutrient levels. This chapter shows how the current array of low cost sensing options offer enormous improvement for growers and researchers. Introducing truly intelligent use of contemporary technology can create valuable outcomes to improve many areas of agriculture.

Intelligent environmental systems are emerging as one of the most efficient technologies to monitor the effects of contaminants and environmental changes on various species of organisms. As these systems are computer supported, the retrieval of information is very easy and time saving compared to conventional methods involved in biomonitoring. This chapter gives an overall idea of environmental biomonitoring with the inclusion of new advancements in this field. In this chapter the intelligent environmental systems and their importance is given focus with a critical analysis of this new technology in environmental biomonitoring research and analysis. The evolution of intelligent environmental systems from basic lab analysis to incorporation of smart technologies in biomonitoring are discussed. The history of development of intelligent environmental systems shows that environmental biomonitoring is now a highly sophisticated field and has the potential to provide a large volume of information with the aid of modeling and Geographical Information Systems (GIS). As biomonitoring technologies become more and more sophisticated, more studies can be done in the future on plants, animals and human beings without wastage of time and human energy. The chapter mainly targets researchers, students and the people who are interested to know about the new trends in environmental research and development.

Increasing agricultural productivity is a global concern as food security is expected at the risk in the near future. Lots of information technology based studies have shown positive effects in analysing and streamlining the agricultural work. However due to frequent data shortage, it is difficult to facilitate precision agriculture. Then an easy deploying, preprocess-integrated data acquisition system may help to solve these problem. This paper presents a system that generates various temporal data streams and refines them automatically. The system has a server and one or more station. The station is dedicated to a plant and collecting a heterogeneous data periodically. The number of station is easily extendable to gather more crop’s information. For effective data collection, all sensors have the same sampling period, 1 min. This sampling period is based on the daylight and its mathematical foundation is presented too. Data preprocess is conducted with low-pass filter and resampler. A method to find an optimal filter based on root-mean-square-error is proposed and analysed.

Currently, due to the global shortage of water, the use of reclaimed water from the Wastewater Treatment Plants (WWTPs) for the irrigation of crops is an alternative in areas with water scarcity. However, the use of this reclaimed water for vegetable irrigation is a potential entry of pharmaceutical products into the food chain due to the absorption and accumulation of these contaminants in different parts of the plants. In this work we carried out an analysis of five machine learning techniques (Random Forest, support vector machine, M5 Rules, Gaussian Process and artificial neural network) to predict the uptake of carbamazepine and diclofenac in reclaimed water-irrigated lettuces with the consequent saving of environmental and economic costs. For the different combinations of input and output, the prediction results using the of machine learning techniques proposed on the pharmaceutical components in reclaimed water-irrigated lettuces are satisfactory, being the best technique the Random Forest that obtains a model fit value (R2) higher than 96.5% using a single input in the model and higher than 97% using two inputs in the model.

The concept of precision agriculture tries to integrate the problems of agriculture with new technologies, in order to provide effective, feasible and efficient solutions, always trying to obtain a higher yield from farmers. Agriculture is an area of high economic relevance in many places. Specifically in this study we will focus on the Spanish province of the Region of Murcia, where agriculture accounts for more than 20% of its economy. Frosts in crops are posing as a serious problem for farmers in this area due to climate change. In this paper we address the problem of frosts suffered by farmers in southeastern Spain with frosts in their crops. At the end of the winter season, temperatures vary by as much as 20 degrees Celsius from midday to night. These variations provoke the anticipation of the blossoming in stone fruit trees, having the risk of frosts at night.Thus, in this paper, the Intelligent Data Analysis have been applied to create predictive models of minimum temperature in plots. In addition, a selection process of the most relevant characteristics to predict the minimum temperature will be presented using the information provided by the models. The data necessary to carry out this study will be collected from the different weather stations of the Institute of Agricultural and Food Research and Development of Murcia. Specifically, data from forty weather stations have been studied, with the aim of finding local or global models that predict the temperature one hour in advance. The data analysis techniques used for the prediction models have been the M5 rule technique for predicting the minimum temperature and the C4.5 decision tree for classifying whether frost will occur or not. The results have identified the most relevant attributes both for predicting and characterize temperature and for classifying whether frost: dew point, vapour pressure deficit and maximum relative humidity occur. The results obtained indicate that both a local classification model and a local prediction model fit perfectly to the resolution of the problem obtaining on the one hand an error of less than 0.5 degrees Celsius for the prediction of the minimum temperature and on the other hand a precision of 98% for the classification of whether frost will occur or not.