1. bookTom 15 (2022): Zeszyt 1 (January 2022)
Informacje o czasopiśmie
License
Format
Czasopismo
eISSN
1178-5608
Pierwsze wydanie
01 Jan 2008
Częstotliwość wydawania
1 raz w roku
Języki
Angielski
Otwarty dostęp

Model based on the principles of smart agriculture to mitigate the effects of frost and improve agricultural production in the Cundiboyacense plateau

Data publikacji: 29 May 2022
Tom & Zeszyt: Tom 15 (2022) - Zeszyt 1 (January 2022)
Zakres stron: -
Otrzymano: 07 Dec 2021
Informacje o czasopiśmie
License
Format
Czasopismo
eISSN
1178-5608
Pierwsze wydanie
01 Jan 2008
Częstotliwość wydawania
1 raz w roku
Języki
Angielski
Introduction

Frost is a phenomenon in which low temperatures cause steam in the air to deposit on plants in the form of ice, it affects different regions of the World same as, limited water resources, droughts, desertification, land degradation, erosion, hail, floods, early frosts, among others (Rathore and Chattopadhyay, 2016). These phenomena affect the agricultural production of the regions where they occur. Global challenges related to food insecurity are increasing due to the impact and probability of these phenomena (van der Elst and Equipo Climandes de la Oficina Federal de Meteorología y Climatología de Suiza MeteoSwiss, 2018). This article focuses on the Colombian terrain, specifically on the Cundiboyacense plateau. In this area, frosts represent a significant problem, because they have a strong impact on the number of hectares that are affected, this phenomenon generates damage to the product itself and the plant and, therefore, the complete loss of crops.

Agriculture typically is based on experience and intuition but, due to the application of new technologies, decisions can be made instantly through the storage and analysis of large volumes of real-time data, from sensors distributed in the crops. The world is facing a revolution led by information and communication technologies, and it is also applied to agricultural fields, adopting what is called smart agriculture or precision agriculture. Its main objective is optimizing resources, increasing production, reducing environmental impact, and improving the quality of the final product (Lopez-chacón and Martínez-garcía, 2018).

Precision agriculture or “Climate-Smart agriculture” (SA) is the process of managing crops, controlling different factors that vary and affect them. Technologies such as, Global Positioning System (GPS), sensors, satellites, aerial images, along with geographic information systems, are used to capture the details required to predict climate events, it is analyzed through data analytics and blockchain tools to more accurately predict climate changes and improve crop production and quality (Pivoto et al., 2018). Currently, more technologies are used in crops and livestock farms, therefore, it is necessary to acknowledge, monitor, store, and relate data to events to anticipate and define trends. Therefore, decisions that allow taking preventive and corrective actions could be done, to improve the productivity and quality of agricultural products in the studied region. These new technologies can be used within the Cundiboyacense region to prevent the damage caused by the frost phenomenon. Likewise, an added value can be generated in agricultural production where the proper application of SA builds more competitive farmers and crops.

Description of the phenomenon
Frost definition

There are different types of frost; by radiation, by advection (the process of transporting an atmospheric property like heat or humidity, by the effect of the wind), by evaporation, black frost, and white frost (Figure 1). However, this article's objective is the study of frost presented at the Cundiboyacense plateau, which occurs mainly by radiation, the progressing and intensive cooling of the soil by heat radiation (I. de hidrología meteorología y estudios ambientales IDEAM, 2019; I. de hidrología meteorología y estudios ambientales IDEAM, 2020). The meteorological field, explains that frost occurs when water or steam in the air freezes and deposits itself in the form of ice on the earth's surface at 2 m above sea level and when the temperature reaches 0°C (I. de hidrología meteorología y estudios ambientales IDEAM, 2019). The presence of frosts is determined when the temperature decreases lower than 0°C. To help determine these phenomena, meteorological stations are used to record temperatures and rainfall, among other factors.

Figure 1

Definition and types of frost diagram. Own source, adapted from FENALCE (2018).

On the other hand, according to the agro meteorological field, frost is considered the temperature at which the plant tissues start suffering damage. It affects the leaf temperature and resistance to low temperatures within the crops in the different stages of development (FENALCE, 2018). That is why information and information networks play an essential role. In Colombia, the Hydrologic, Meteorologic and Environmental Studies Institute (IDEAM) oversees the climate forecasting process. Even though this forecast does not estimate the damage to agricultural production, it does estimate rainfall and average tem peratures which are key factors to predict the occurrence of frosts.

Description of the region
Terrain characteristics

Colombia is a country whose terrain and climate are suitable for growing different plant and fruit species. According to studies carried out by the Agustín Codazzi Geographical Institute (IGAC), the lands that have more potential for cultivation in Colombia are the Atlántico, Sucre, Magdalena, Quindío, Cundinamarca, Bolívar and Antioquia.

Within these regions, one that has special conditions is the Cundiboyacense plateau, which is formed of high and flat areas between the departments of Cundinamarca and Boyacá, extending from the Sumapaz Páramo to the foothills of the Sierra Nevada del Cocuy, and is considered one of the richest and most densely populated regions of the country (de Boyacá, 2012) The underlying climate is mainly cold thermal, in the central highlands of the region, bimodal rains are presented, highlighted by a period of continuous rain between April–June and October–November, and presenting isolated rains the rest of the year (IDEAM, 2001).

Cundinamarca and Boyacá are departments that constitute a total of 6.95% of the cultivable area of the entire country. Currently, the cultivated land in each of them are 8.44% and 23.72% (Instituto Geográfico Agustín Codazzi, 2021). Additionally, Colombia has the UPAs (Agricultural Production Units). Which are terrains used to produce agricultural, forestry, livestock and/or aquaculture goods and have a single producer, natural or legal, who assumes responsibility and risks, in addition to that, they use at least one means of production such as constructions, machinery, equipment and/or labor on the territory (DANE and MINAGRICULTURA, 2015). The departments of Cundinamarca Boyacá, Nariño, Antioquia, and Cauca are approximately 53.9% of the UPAs present throughout the country (DANE, 2014).

By 2004, Boyacá produced around 36 different species of fruit trees (ASOFRUCOL, 2006) common of the geographic and climatological characteristics of the region, being one of the departments with the highest number of crop species. The guava, orange, and plum are the most representative, not leaving aside the other species. According to the National Fruit Plan carried out in 2006, there were predictions showing that in the next 20 years most of the area available in the region would be used for livestock operations, with pasture crops, representing 42% (ASOFRUCOL, 2006).

On the other hand, according to the studies published by the Agustín Codazzi institute, about the coverage and use of the soil in the region; The municipalities of Aquitania, Sogamoso, Tota, and Cuítiva, in the region, stand out for producing about 80% of onion in the country, in addition, these regions produce; sugar cane, coffee, cocoa, honey cane, beans, cassava, corn, and black tobacco, among others. Also of the 123 municipalities of the territory, the ones that stand out are Puerto Boyacá, Saboyá, Santana, Toguí, Moniquirá, Santa Sofía, Chiquinquirá, Combita, Paipa, Tuta, Toca, Oicatá, Firavitoba and Susacón with the highest agricultural production (I. G. A. C. IGAC, 2021).

Impact

Frosts occur in different parts of Colombia and the world. They mainly affect areas located over 2,500 meters above sea level, especially in the driest months of the year (Hurtado, 1996). On the other hand, in the Cundiboyacense plateau, they occur due to the decrease in temperature in the early morning hours. It has higher possibilities of happening, in the periods of drought during January–February and June–September (IDEAM, 2001).

In 1996, IDEAM, led by the meteorologist Gonzalo Hurtado Moreno, carried out a series of studies and publications regarding the climate in Colombia. Starting from 1961 to 1993, droughts in Colombia did not have homogeneous distribution. According to the dry periods presented in each decade, a non-parametric increase was exhibited until the 80s with a decrease in the 90s (Hurtado, 1996). Although the droughts did not show a homogeneous behavior, some phenomena caused by global warming and other factors greatly enhance the occurrence of these states of nature. The drought seasons presented in 2012 as a consequence of the warming of the Pacific Ocean waters, will continue and, as a consequence, the departments of Cundinamarca, Boyacá, and Nariño will have a strong presence of frosts (González and Torres, 2012).

As demonstrated in Figure 2, there are multiple communities in the Cundiboyacense plateau that have increased susceptibility to frost, these are:

Northern zone of the Andean region : Silos and Pamplona in Norte de Santander, Tona in Santander.

Central zone of the Andean region: Santa Rosa de Osos in Antioquia, Salamina in Caldas, Tunja, Sogamoso, Samacá, Paipa, Duitama, Chita, Toca, Nobsa, Tibasosa in Boyacá. Mosquera, Tabio, Zipaquirá, Subachoque, Sesquilé, Facatativá, Madrid, Sopó, Nemocón, Bojacá, Chía, Suesca, Cogua, Tenjo, Chocontá, Funza, Ubaque, Choachí in Cundinamarca (I. de hidrología meteorología y estudios ambientales IDEAM, 2020).

It is estimated that the months in which this phenomenon has the most presence is in December, January, and February with an 80% probability of occurrence (Hurtado, 1996).

Figure 2

Areas with the greatest presence of frost. Source IDEAM (I. de hidrología meteorología y estudios ambientales IDEAM, 2019).

Due to this phenomenon, which coincides with periods of drought, many farmers took measures to cope with it. They mainly created irrigation systems, which reduce frost impact on the leaves and pastures (considering that a large part of the territory is cattle rancher) (IDEAM, 2001).

Despite the measurements taken by some farmers, this phenomenon significantly affects agricultural production in the region; In 2020, from the 150.000 hectares that the Boyacá region has for food production, 43,000 hectares were affected in February, reaching temperatures as low as −6.8°C (FAO and CAF, 2013). Also, in Cundinamarca, 19.000 hectares of damaged pastures were registered, as well as 5.400 hectares of crops, with temperatures varying from −1 to 2°C (DANE, 2020). According to the CAR (Regional Autonomous Corporation), 67.000 hectares were affected in the department, near 30.000 farmers (ARGENPAPA, 2021).

Climate-smart agriculture concept review

The condition of the farmlands depends on the location and the climate in which they are, therefore, Colombia has the right conditions to be a great country to cultivate. For this reason, agriculture has had great importance over the years in terms of economy and the nation's development. However, due to climate change, and its consequences such as frost and lack of water, the advantages that Colombia has in terms of productivity, crops survival capacity, and the final quality of agricultural products have been affected.

According to the Food and Agriculture Organization of the United Nations (FAO), food security occurs when a population has permanent access in a physical, economic, and social way to sufficient and quality food that can satisfy their needs and nutritional preferences so that they can lead healthy and active lives. Considering the imminent threat of climate change and the need to maintain food security, the term climate-smart agriculture was introduced (Torquebiau et al., 2018). This concept postulates the implementation of advanced information and communication technologies, such as robotics, the internet of things, and precision agriculture in agricultural processes. All of them seek to transform, reorient and develop agricultural systems with the help of technologies to contribute to sustainable, efficient agriculture, with increasing production and quality of food.

Smart Farming or Smart agriculture is a term that encompasses ideas such as precision agriculture and Farm information management systems (FMIS), It is a system that can collect, process, store and disseminate data in specific formats to carry out operations in rural areas (Sørensen et al., 2010).

Precision agriculture as part of climate-smart agriculture aims to manage and administer agricultural plantations. It uses technologies such as (Beecham Research, 2017):

Detection technologies (sensors)

Data analysis solutions

Software applications

Communication systems

Global positioning services (GPS)

Hardware and software systems

These systems serve as a form of decision support, backed by real-time data. Therefore, it is important to determine which variables will feed the system and, based on the information obtained, define; frequency of use of fertilizers and pesticides, frequency and quantity of irrigation, sowing patterns, and expected production (J. C. FAJARDO JUNCO, 2014).

As mentioned above, one of the technologies used for smart agriculture are sensors. These are small devices that measure physical variables in different environments and those signals are information that can be read by other instruments. Among the variables that the sensors can measure are temperature, humidity, light, pressure, noise levels, presence, or absence of certain types of objects, levels of mechanical stress, speed, direction, and size of the object. Another instrument to obtain environmental information, which is widely used, is satellite and aerial images obtained with drones, these are applied mainly in large crops (Gómez et al., 2018).

These technologies are related to the Internet of Things IoT, which allows objects to be controlled remotely over existing networks. They use the infrastructure of the internet where more direct integration is created between the physical world and computer-based systems (Pivoto et al., 2018). Through sensors and satellite images, information is collected to later be analyzed. These devices are also applied for Wireless Sensor Network (WSN) that allows communication, in which sensors and different actuators, relate to the environment and share information between platforms, driven by other wireless technologies such as “tags Radio Frequency Identification (RFID) and integrated sensor and actuator nodes” (Gubbi et al., 2013).

IoT technologies have many advantages in their application, and the field of agriculture is one of those that need it most urgently. With its help, farmers can detect problems in the shortest possible time and with greater precision. Thanks to the advances, it is possible to apply sowing plans based on data that maximize their productivity and profits. According to George Adamides, in his article, “A Review of Climate-Smart Agriculture Applications in Cyprus”, in agriculture, the IoT is applied in three stages:

Data acquisition

Data processing

Analysis of data

Big data techniques and methods, which are applied in agriculture to increase feed efficiency in livestock production and manage the risk of crop loss, include; “Comparative evaluation, implementation and analysis of sensors, predictive modeling and use of better models” (Lesser, 2014; Wolfert et al., 2017). In this way, big data serves as a form of support to the farmer and the economy through predictions. Big data is used to support the decision-making process in real-time. The inclusion and application of these methods and technologies help to improve the process in the supply chain and food security (Gilpin, 2014; Wolfert et al., 2017).

Review of patents related to smart agriculture

Using Intelligo search engine to find WIPO (World Intellectual Property Organization) patents, though Boolean equation analysis, climate-smart agriculture patents were explored. The equation identifies the elements, models, and existing developments where intelligent technologies are applied to agricultural systems. Different searches are carried out, as can be seen in Table 1. At first, the term “SMART AGRICULTURE” was used as the base, finding a large number of articles. Later filtering this information using keywords that are shown in different articles and are relevant to the present study.

Patent Search at WIPO. Own source.

WIPO patent search through Boolean equations
Boolean equation Findings
Smart agriculture OR precision agriculture 42980
Smart agriculture 19540
Smart agriculture AND Agricultural implement 6639
Smart agriculture AND crop 996
Smart agriculture AND sensor 412
Smart agriculture AND Latin America 1

In Figure 3, the result of the Boolean equation for “Smart Agriculture AND Sensor” is obtained. As visible, in the WIPO patents the most frequently repeated words are: crop, agriculture, agricultural field, agricultural equipment, soil, and agricultural machine. Concepts are very important when proposing a model since, for the most part, they have a high probability of being part of it, either as agents or as essential complements. Indeed, it is difficult to talk about intelligent systems, leaving aside the human factor such as harvesters and farmers as well as the crops where several developments are applied, such as the soil and the agricultural field.

Figure 3

Related concepts diagram, Intelligo (OEI Organización de estado iberoamericano para la educación la ciencia y la cultura).

On the other hand, the following graphs (Figures 4 and 5) show the countries that have filed invention patents in this field and the years of publication of the patents. Thus, it is interesting that the countries that have filed the greatest number of patents in this regard are: the United States, Germany, Japan, China, and France. And, several of these countries, such as the United States, China, and Germany (as part of the European Union) are within the ranking published by the Think tank in 2021 (Information and Technology and Innovation Foundation-ITIF). These are within the 30 countries with the most potential in terms of Artificial Intelligence. However, although these countries are the ones that have generated the most technological developments focused on agriculture, they do not base their economy on it. For example, in the ranking of territory used for agricultural activity, developed by the world bank in 2020, the United States occupies first place in development but is in 12th place with 44.4% in the usage of territory for this activity. While other countries such as Uruguay, occupying the first place with 82.6% of land usage for agricultural activities, have not generated outstanding advances in new technologies applied to said activity. Colombia is in position 13 with 40.3% of land use and has no patented developments (Vita Mesa, 2021).

Figure 4

Patent inventing countries, Intelligo (OEI Organización de estado iberoamericano para la educación la ciencia y la cultura).

Figure 5

Patent Publication Year, Intelligo (OEI Organización de estado iberoamericano para la educación la ciencia y la cultura).

Furthermore, 2021 is the year in which the most patents for smart agriculture have been published, as evidenced in Figure 5. When reviewing developments in recent years, it is possible to relate the increase in patents to globalization; for example, easy access to information has demonstrated the importance and benefits of generating and protecting new ideas.

Although the majority of Latin American countries such as Colombia, have not presented patents related to smart agriculture yet and are far below the list, it is expected that they will not be left behind and with investment in technology, they will achieve the publication of patents in smart agriculture. In 2019 Oxford published the world ranking of pro-artificial intelligence countries, it evaluated the capacity of governments and sectors of the economy to incorporate and make correct use of these technologies, making the most of them in their product development. Colombia ranked 44th out of the 194 countries evaluated, it can be concluded that, although it is not in the top positions, it still has considerable development potential. Salas (2021)

Materials and methods

Starting from the need of the agricultural sector in the Cundiboyansence plateau in Colombia to become more technical and to increase the quality, efficiency, and productivity of crops, based on the control of the variables that affect them, a solution is proposed so that small and large producers can register the environmental conditions involved in their process simply and immediately and support them in making decisions.

Reference models

The following models were used as a basis for this article's proposed model:

Model 1. In this model (illustration 6), the benefits of the use of small technological devices such as sensors linked with the internet and artificial intelligence are presented, and their application in the supply chain. Although this model was developed as a complement to “smart cities” and “smart homes” where agriculture plays an essential role in providing food, it also presents the benefits that it can bring at the scientific level.

This model presents (Figure 6) an intelligent agriculture system in which crops are monitored through sensors and airships, collecting, and processing this information through a variety of different and complex technological tools, later, said information is shared in a cloud platform to be received and analyzed by end-users.

Figure 6

IOT model taken from Perera et al. (2014).

Model 2. This model (Figure 7) is divided into four phases: the first phase of sensors, which includes different types of sensors and technologies such as thermal cameras, for the collection of information (these can be installed directly on crops or unmanned aerial vehicles). The second phase is the connection to the network, this includes those devices that allow the constant transfer of data in real-time from the sensors to the internet. The third phase is the server, in which the collected data is processed to convert it into useful information for the correct use of pesticides and disease prevention. And a fourth and final phase of the application, in which the information is transmitted to the farmer so that he can make the most appropriate decisions in his cultivation.

Figure 7

Architecture of the precision agriculture system. Taken from Triantafyllou et al. (2019).

As can be seen, the models presented differ in several aspects; In the first place, these models were designed with various purposes, the first was developed with a general application being part of the food supply chain to smart cities and homes, and the second for the prevention of diseases and the correct application of pesticides in the crops. The first model does not have a specific development, in this, it only raises the benefits that could be obtained by applying sensors and technologies to crops, while in the second model its development specifically covers the operation of each of its phases.

Proposed model

This model (Figure 8) is diagramed based on the “design principles of the intelligent agriculture system based on IoT” that Chunling Li and Ben Niu define in their article “Design of smart agriculture based on big data and Internet of things”. There are three stages: data acquisition, data processing, and data analysis, and based on this, a model is developed starting from the crops, abstracting the information from them and their environment, to the decision-making agents (farmers and actuators).

Figure 8

Proposed smart agriculture model for frost mitigation in the Cundiboyacense highlands. Own source.

The proposed model consists of 3 phases:

Phase I. information gathering

In the first phase, the information that is collected will be used to make the corresponding predictions, therefore the variables that feed the model must be carefully selected so that these predictions can be made as accurate as possible.

Given that the model represented seeks to reduce or avoid the effect of frost, it should be taken into account that these occur mainly in areas located more than 2500 m above sea level in the driest times of the year, usually in the early morning hours. The temperature can go below 0°C, which causes the water or steam in the air to freeze and settle on the ground. To avoid the consequences of frost, farmers use irrigation systems, which prevent damage to leaves and plants.

Analyzing this phenomenon and the effect it has on crops in the Cundiboyacense plateau, it can be concluded that the main variables that predict the phenomenon are temperature and humidity among others. Therefore, these are the variables that will feed the model.

Other variables that are important to the farmer are: the intensity of light that the crop receives and the composition of the soil (including information such as; soil water content and its conductivity, soil moisture, and even its pH). This data also provides information that helps make decisions such as the location of the crop, the use of pesticides or fertilizers, among others.

Based on the above, it is determined that the data that will feed the model are:

Temperature, taking into account that the temperature at which frost is generated is 0°C or below, it is expected that the system will generate an alert or activate the actuator, which will start irrigation depending on the crop needs, before reaching this temperature, that is, around 10°C. For this model, this will be the first factor that the system will take into account to activate the actuator.

Humidity, for damage to plants or crops to be generated, it is necessary that when it reaches 0°C there is a certain humidity in the environment. Therefore, it is determined that depending on the amount of humidity or water vapor in the environment, the alert is generated, or the actuator is activated. If the relative humidity is less than 50%, activation is not necessary, as there will not be enough water vapor in the atmosphere to settle on the surface and affect the crop. This will be the second factor in the activation. It is necessary that both temperature and humidity are within the ranges to continue the process.

The intensity of light, this factor is important, because if there is greater intensity of light the sky will be clear, this indicates less presence of clouds. Clouds prevent radiation from moving freely towards space, which generates a loss of temperature, and accelerates the cooling of the soil (Lasso Espinosa, 1987). This will be the third factor to be consider in the activation of the alert and actuator system. As mentioned before all of the factors have to fulfill the conditions. IDEAM uses four categories to describe cloudiness (Torres and Rangel, 2005):

Category 1: Clear or slightly overcast sky

Category 2: Partially cloudy sky

Category 3: Overcast sky

Category 4: Dark sky

Composition of the soil, knowing the composition of the soil allows determining the areas that are most appropriate for the growth and quality of the crop. Therefore, this is a factor that supports the farmer's decision making.

The data collection is carried out using high-precision sensors for agricultural activities, which obtains the indicated parameters in real-time. It is considered that they must be connected to a computer that will need a power source to operate and there are two options for connecting to the internet. The first is that the equipment should have an internet connection through a SIM card and the second is that the equipment can directly access a connection with a Wi-Fi module. Once the data are uploaded to a server through the internet connection, the brain of the device -in addition to the distribution of energy to the different sensors and the connection module- is the one in charge of obtaining the data and transferring them so that they can be stored in the cloud, this process can be carried out in real-time allowing complete control and analysis of different types of crops, either outdoors or those crops that are carried out in closed spaces (greenhouses).

In addition, to complement the information received, the historical data, in real-time and forecasts of climate, rainfall, and droughts published by IDEAM can be used to determine when frosts are most likely to occur, as well as its relationship with rainfall and droughts, to have more accurate data to feed the system and improve decision making.

In the case of the Cundiboyacense plateau, as shown in Figure 9, the information obtained through the sensors and historical records of IDEAM and the system, obtained in Phase I, are used as input to the process.

Figure 9

Process description.

Phase II. processing and synthesis of the information collected

After obtaining information through sensors and from the cloud (IDEAM databases), it must be processed so that it can be used to make decisions. Through data mining, the data obtained is converted into knowledge that can be adequately synthesized and analyzed.

The data mining process for this case consists of 5 stages (Hernández Orallo et al., 2004):

Sampling and data selection: After obtaining data, the first stage consists of sampling and selecting those specific data that can be very useful.

Sensors: Relevant information on the waves and samples perceived by the sensors.

IDEAM databases: Relevant information regarding maximum and minimum temperatures, rainfall, and droughts.

In this phase, data from sensors and IDEAM databases is stored and classified (Figure 10). This process is daily developed with enough frequency to establish decision-making criteria, that is, that the data adequately respond to the necessary tests and algorithms used later for data analysis.

Data cleaning: After the previous stage, the system has the information it needs for decision-making. But it is possible that there are unrelated data or that it was mistakenly captured, which is why it is necessary to perform a data cleaning, discarding those that:

Have a non-existent value: intervals in which the measurements could not be taken, therefore they do not exist. For example, if the system loses internet and cannot collect data, if one of the sensors is damaged and it is necessary to change or repair it or if the system does not receive power, among others.

Unclassified values after the identification of extremes: temperatures, humidity, light, among others, that have very high or very low values concerning the values admitted in the region. For example, if the temperature exceeds 25°C (maximum temperature recorded in the region) this value should be discarded unless there is any real climatic phenomenon.

Data transformation: At this stage, characteristics are established between the data to begin to identify them. In addition, the amount of information is reduced through different processes such as those presented below:

Definition of characteristics that make each type of data useful: The aim is for the data to reduce its size for later graphs presentation; For example, given the large amount of data that sensors can collect in a single day, week, or month, it is necessary to determine an appropriate scale to present them, in such a way that they are of great importance for the farmer, as well as for the actuator and its decision.

Similarly, by having a clear definition of the characteristics of each variable that enters the system (temperature, humidity, light intensity, soil composition, rainfall, and drought forecasts), it is possible to determine the correlation between the variables presented.

Transformation of the data in such way that it is useful and manageable: It is intended to have a basis to present the data in such a way that it would be easily understood by the farmer. Hence the type of variables interacted with must be determined and, if necessary to normalize this data, to make exact use of maximums, minimums, and averages.

Use of Data mining algorithms (Beltran Martinez, 2003): After the transformation of data, the next step is to make use of algorithms for; On the one hand, to determine behavior patterns of the data to make a forecast of future situations that may arise and that may harm the sowing, and on the other hand for the presentation of the information.

For the presentation of information with these algorithms, it is intended not only to show the variables collected from the sensors and IDEAM, but also to discover associations between actions and independent events (Beltran Martinez, 2003) as well as their evolution in time. Recognizing the time that usually elapses between the occurrence of different events. For example, the effect on the composition of the soil, after the mitigation measures for the phenomena that may occur.

Verification of results and obtaining complements: It seeks to verify the consistency of the results obtained. That way, it is checked if the information collected from the source (IDEAM sensors and weather stations) is correct and if this information was not modified in its treatment. On the other hand, the necessary information will also be complemented to make the appropriate recommendations to the farmer according to the phenomena that may occur.

Figure 10

Flowchart phase 2.

Phase III. information analysis and decision making

To finalize the process, based on the knowledge obtained in the previous phase, the information is organized in a way that is useful for the farmer and, for the actuator to make the decision regarding its activation, as follows:

Modeling: In this stage, knowledge is presented to the farmer in two ways;

Descriptive modeling, the variables of temperature, pressure, humidity, light intensity, rainfall, and droughts are presented in bar and dispersion diagrams that present the historical data of the variable and the information in real-time. In addition, a calendar with the lunar phases, the cycles of plants, and the cycles of rain are presented.

Predictive modeling, since in descriptive modeling the historical and present variables are presented, in this stage, it is intended that using the information, the temperature forecast is generated and finally, the knowledge is presented in the form of alerts that will be determined by the farmer according to their needs. For example, if the temperature or humidity reaches a certain value in which application of fertilizers, pesticides, or the activation of irrigation systems must be applied an alert will be sent to the farmer. This will be useful to the farmer as a form of support in the decision-making process.

Actuator decision: For the actuator to decide whether to activate the irrigation system automatically, it will consider the temperature in real-time. For this model, it is determined that a frost occurs when there are temperatures below 0°C, relative humidity is over 50% and there is no presence of clouds, therefore, the irrigation system must be activated when the three factors are within the mentioned ranges or the ones that the farmer previously defines, thus avoiding the formation of ice on the plants. For this decision, the information provided by IDEAM regarding droughts and rainfall must also be taken into account, since the presence of any can affect the decision, that is, if precipitation has occurred, it is not necessary to activate the system, but, if on the contrary, the day before there is a drought with humidity, there is a high probability of frost and the irrigation system must be activated. For example, Table 2 shows the decision process based on the value of the main variables, in this situation, the actuator or the alarm sets off or not, depending on them. It is important to know that the actuator should activate before the frost starts so necessary actions are taken, but also an alarm should be set off once the frost begins in case other actions need to be done.

The knowledge that must reach the farmer is presented using software that can be viewed in a mobile application or on a website on the computer and, with complete information, decisions can be made.

Decision variables example for decision making. Own sourse.

Decision variables
Option 1
  Temperature <0 Alarm sent to the farmer
  Humidity >50%
  Cloudiness Category 1
Option 2
  Temperature < 10 Activation of the actuator and alarm
  Humidity > 50%
  Cloudiness Category 1
Option 3
  Temperature < 10 No activation
  Humidity < 50%
  Cloudiness Category 1

The model (Figure 11) is presented in a summarized way:

Figure 11

Proposed model architecture Own source.

Benefits of the model and innovation

The main objective of the proposed model is to mitigate frost in the Cundiboyacense plateau. Although its use can be extended to determine the need for pesticides and continuous monitoring of other factors that can affect crops, the system focuses mainly on frost. The basis models (see models 1 and 2), which make use of intelligent technologies for crop monitoring and maintenance, seek a broader objective and therefore their elements differ both in the data that enter the system, as well as in their processing and presentation.

The model considered the affected population (farmers), their social, economic, demographic and cultural context, as well as the access to the required technologies and the acquisition of the essential and useful data presented in such a way that could be understanding is simple allowing the farmer and the actuator to make the most appropriated decisions for the efficient solution of the detected problem.

Although the aim is to mitigate the effect of frost on the plateau's crops, the actuator must make decisions that respond to the problem but at the same time generate savings in water resources.

At present, farmers make use of tools within their reach to mitigate these effects, such as irrigation systems or the use of plastics; however, the use of new technologies represents a social and economic advance that can benefit them to a great extent. In this sense, the model is innovative in terms of the use of these technologies on a large scale and widely, to generate a higher competitive level at the national and international level.

In recent decades, there has been great progress worldwide in the use of intelligent technologies and the internet of things; however, in countries such as Colombia, these technologies are not easily accessible because they represent a large investment by the client. The proposed model not only seeks easy access to these technologies by farmers making use of open access data (IDEAM) and intelligent sensors but also its application in agriculture can generate a great impact on the level of agricultural production but also on the level of research in this field.

The model takes into account the tropical climate of the country, which is essential due to the difficulty of climate prediction that this represents, which implies the use of stochastic statistics to predict with greater accuracy the elements that can have negative or positive effects on crops. In addition, by using the two proposed tools (IDEAM forecasts and sensors) the data can be analyzed with greater precision and the use of data science not only represents an aid to the farmer but also for other areas, as well as to feed climate prediction models widely used in the country.

If the model is applied, it could benefit the farmer and the economy of the Cundiboyacense plateau because it supports the decision-making process, since it could take into account the real-time needs of the crop, avoiding the generation of losses due to the low temperatures in the area and humidity at certain times of the year. Likewise, by implementing this model, the farmer will have a way to access all the useful information in a simple, complete, and understandable way, in addition to providing adequate recommendations for the care of their crops. From the use of Information and Communication Technologies (ICT) and data processing.

Taking into account that many of the farmers have long experience and knowledge regarding their land, it is intended to provide additional information support in the face of new changes that affect the soil, the environment, and therefore the agricultural production.

Limitations of the model

Economic: It will be crucial to make a large investment. In the case of farmers, generally, they do not have the capacities to apply the necessary technology. Therefore, economical support is needed from the government or private entities to allow this sector to use them and in this way, not only reduce losses but also increase productivity and improve product quality.

Cultural: Many farmers may have adverse reactions to the use of technologies as they are not familiar with them and consider that their application could be difficult. Even in some sectors, they do not use tools such as smartphones or computers and, therefore, it is necessary to train these people in certain technologies.

Accessibility: In some sectors of the country access to the Internet is still difficult and in consequence the application of certain technologies applied in the model may present difficulties. Internet access plays a fundamental role and it is a fact that in the region studied this factor can affect the interaction of the entire proposed system. For its correct application, permanent and continuous access must be ensured.

An example of this, was given in 2009 in a study in the Puno-Peru region, (INEI, 2012) (as cited in (van der Elst and Equipo Climandes de la Oficina Federal de Meteorología y Climatología de Suiza MeteoSwiss, 2018)). Among its main crop species of food, quinoa is affected by frost. It is determined that although there were sources of information available, there were restrictions that prevented its use because it was incomplete, confusing, insufficient, and difficult to access (SENAMHI and MeteoSwiss, 2018).

Many projects, like the one in Peru, have not been successful enough. For this reason, one focused on responding to the effects of frost in the Cundiboyacense plateau is proposed, applying what is known as smart agriculture.

Conclusions

This model could be used as a basis to the development of a system to mitigate the effect of frost in the Cundiboyacense plateau. It is proposed contemplating the characteristics of the area and it is also considered that it must be simple and applicable corresponding to the economic and technological capabilities of the region. This model differs from those that were taken as a basis, on the one hand in the applied technology and the specifications of the model, and on the other hand with the decision-making agent. It is proposed that using artificial intelligence, autonomous decisions are made based on provided information to mitigate frost consequences.

It is proposed that the model, in addition to working with the information from the sensors, uses the IDEAM data that, even if it is not 100 percent accurate, could complement the system and determine with greater accuracy the actions to be taken to avoid adverse effects on crops, this taking into account that intelligent systems need large amounts of data to process, so the decision making is carried out by the system automatically. New technologies have proven to be an immense source of support to man and their constant use has become even more common in various parts of the world providing solutions to both large industries (process automation) and end consumers (internet of things). Although Colombia has not developed these technologies, this study shows how their use has become indispensable for sustainable development, because it prevents the unnecessary usage of resources.

In addition to the application of technologies, farmers must be trained so that they get to know the new technologies that are available to them, that they understand how they are used, what their purpose is, and how they benefit from them due to their role of supervising that frost does not have any effect on their crops and act against other phenomena based on the alerts, graphs, and information provided by the system. Likewise, the application of these new technologies and the scope they have in the field change how these activities are generally carried out and, therefore, this can be a form of motivation for the younger generations to get involved and apply their knowledge and trends.

Figure 1

Definition and types of frost diagram. Own source, adapted from FENALCE (2018).
Definition and types of frost diagram. Own source, adapted from FENALCE (2018).

Figure 2

Areas with the greatest presence of frost. Source IDEAM (I. de hidrología meteorología y estudios ambientales IDEAM, 2019).
Areas with the greatest presence of frost. Source IDEAM (I. de hidrología meteorología y estudios ambientales IDEAM, 2019).

Figure 3

Related concepts diagram, Intelligo (OEI Organización de estado iberoamericano para la educación la ciencia y la cultura).
Related concepts diagram, Intelligo (OEI Organización de estado iberoamericano para la educación la ciencia y la cultura).

Figure 4

Patent inventing countries, Intelligo (OEI Organización de estado iberoamericano para la educación la ciencia y la cultura).
Patent inventing countries, Intelligo (OEI Organización de estado iberoamericano para la educación la ciencia y la cultura).

Figure 5

Patent Publication Year, Intelligo (OEI Organización de estado iberoamericano para la educación la ciencia y la cultura).
Patent Publication Year, Intelligo (OEI Organización de estado iberoamericano para la educación la ciencia y la cultura).

Figure 6

IOT model taken from Perera et al. (2014).
IOT model taken from Perera et al. (2014).

Figure 7

Architecture of the precision agriculture system. Taken from Triantafyllou et al. (2019).
Architecture of the precision agriculture system. Taken from Triantafyllou et al. (2019).

Figure 8

Proposed smart agriculture model for frost mitigation in the Cundiboyacense highlands. Own source.
Proposed smart agriculture model for frost mitigation in the Cundiboyacense highlands. Own source.

Figure 9

Process description.
Process description.

Figure 10

Flowchart phase 2.
Flowchart phase 2.

Figure 11

Proposed model architecture Own source.
Proposed model architecture Own source.

Decision variables example for decision making. Own sourse.

Decision variables
Option 1
  Temperature <0 Alarm sent to the farmer
  Humidity >50%
  Cloudiness Category 1
Option 2
  Temperature < 10 Activation of the actuator and alarm
  Humidity > 50%
  Cloudiness Category 1
Option 3
  Temperature < 10 No activation
  Humidity < 50%
  Cloudiness Category 1

Patent Search at WIPO. Own source.

WIPO patent search through Boolean equations
Boolean equation Findings
Smart agriculture OR precision agriculture 42980
Smart agriculture 19540
Smart agriculture AND Agricultural implement 6639
Smart agriculture AND crop 996
Smart agriculture AND sensor 412
Smart agriculture AND Latin America 1

ARGENPAPA. 2021. Colombia: 30.000 productores afectados por las heladas en Boyacá y Cundinamarca. https://www.argenpapa.com.ar/noticia/8440-colombia-30-000-productores-afectados-por-las-heladas-en-boyaca-y-cundinamarca (Accessed Apr 6). ARGENPAPA 2021 Colombia: 30.000 productores afectados por las heladas en Boyacá y Cundinamarca https://www.argenpapa.com.ar/noticia/8440-colombia-30-000-productores-afectados-por-las-heladas-en-boyaca-y-cundinamarca (Accessed Apr 6). Search in Google Scholar

ASOFRUCOL. 2006. Desarrollo de la Fruticultura en Boyaca. ASOFRUCOL 2006 Desarrollo de la Fruticultura en Boyaca Search in Google Scholar

Beecham Research. 2017. Smart Farming The sustainable way to food. Beecham Research 2017 Smart Farming The sustainable way to food Search in Google Scholar

Beltran Martinez, B. 2003. Mineria de datos, Benemérita Univ. Autónoma Puebla. doi: 10.1016/0032-0633(82)90071-X. Beltran MartinezB. 2003 Mineria de datos Benemérita Univ. Autónoma Puebla 10.1016/0032-0633(82)90071-X Otwórz DOISearch in Google Scholar

DANE. 2014. [Online]. Uso, coberturay tenencia del suelo 3.er Censo Nacional Agropecuario 2014. p. 74. Available https://www.dane.gov.co/files/CensoAgropecuario/entrega-definitiva/Boletin-1-Uso-del-suelo/1-Presentacion.pdf. DANE 2014 [Online]. Uso, coberturay tenencia del suelo 3.er Censo Nacional Agropecuario 2014 74 Available https://www.dane.gov.co/files/CensoAgropecuario/entrega-definitiva/Boletin-1-Uso-del-suelo/1-Presentacion.pdf. Search in Google Scholar

DANE. 2020. Informe especial precios mayoristas y abastecimiento. pp. 67–69. DANE 2020 Informe especial precios mayoristas y abastecimiento 67 69 Search in Google Scholar

DANE and MINAGRICULTURA. 2015. Censo Nacional Agropecuario 2014 Las Unidades de Producción Agropecuaria (UPA): Infraestructura, asistencia técnica y financiamiento. DANE and MINAGRICULTURA 2015 Censo Nacional Agropecuario 2014 Las Unidades de Producción Agropecuaria (UPA): Infraestructura, asistencia técnica y financiamiento Search in Google Scholar

de Boyacá, G. 2012. Aspectos Geográficos. Available https://www.boyaca.gov.co/aspectos-geograficos/ (Accessed Apr 6, 2021). de BoyacáG. 2012 Aspectos Geográficos Available https://www.boyaca.gov.co/aspectos-geograficos/ (Accessed Apr 6, 2021). Search in Google Scholar

FAO and CAF. 2013. [Online]. Colombia Nota de Análisis Sectorial Colombia Colombia p. 68. Available http://www.fao.org/3/a-ak167s.pdf. FAO and CAF 2013 [Online]. Colombia Nota de Análisis Sectorial Colombia Colombia 68 Available http://www.fao.org/3/a-ak167s.pdf. Search in Google Scholar

FENALCE. 2018. BOLETIN DE HELADAS DENTRO DEL ALTIPLANO Puesta del Sol Entre las 5: 50 p. my las 5: 55 p.m.,” p.2. FENALCE 2018 BOLETIN DE HELADAS DENTRO DEL ALTIPLANO Puesta del Sol Entre las 5: 50 p. my las 5: 55 p.m. 2 Search in Google Scholar

Gilpin, L. 2014. How big data is going to help feed nine billion people by 2050 – TechRepublic. May https://www.techrepublic.com/article/how-big-data-is-going-to-help-feed-9-billion-people-by-2050/ (Accessed April 6, 2021). GilpinL. 2014 How big data is going to help feed nine billion people by 2050 – TechRepublic May https://www.techrepublic.com/article/how-big-data-is-going-to-help-feed-9-billion-people-by-2050/ (Accessed April 6, 2021). Search in Google Scholar

Gómez, J. E., Castaño, S., Mercado, T., Fernandez, A. and Garcia, J. 2018. Sistema de internet de las cosas (IoT) para el monitoreo de cultivos protegidos. Ingeniera e Innovación 5(1). Doi: 10.21897/23460466.1101. GómezJ. E. CastañoS. MercadoT. FernandezA. GarciaJ. 2018 Sistema de internet de las cosas (IoT) para el monitoreo de cultivos protegidos Ingeniera e Innovación 5 1 10.21897/23460466.1101 Otwórz DOISearch in Google Scholar

González, O. C. and Torres, C. F. 2012. [Online]. Actualización Nota Técnica Heladas 2012. Available http://www.ideam.gov.co/documents/21021/21147/Documento+FINAL+actualizacion+nota+tecnica+heladas.pdf/e10a0183-62e6-410a-8e96-7e0739f6f06b. GonzálezO. C. TorresC. F. 2012 [Online]. Actualización Nota Técnica Heladas 2012 Available http://www.ideam.gov.co/documents/21021/21147/Documento+FINAL+actualizacion+nota+tecnica+heladas.pdf/e10a0183-62e6-410a-8e96-7e0739f6f06b. Search in Google Scholar

Gubbi, J., Buyya, R., Marusic, S. and Palaniswami, M. 2013. Internet of Things (IoT): A vision, architectural elements, and future directions. Future Generation Computer Systems 29(7): 1645–1660. GubbiJ. BuyyaR. MarusicS. PalaniswamiM. 2013 Internet of Things (IoT): A vision, architectural elements, and future directions Future Generation Computer Systems 29 7 1645 1660 10.1016/j.future.2013.01.010 Search in Google Scholar

Hernández Orallo, J., Ramírez Quintana, M. J. and Ferri Ramírez, C. 2004. No Title Introducción a la mineria de datos. Hernández OralloJ. Ramírez QuintanaM. J. Ferri RamírezC. 2004 No Title Introducción a la mineria de datos Search in Google Scholar

Hurtado, G. 1996. Estadisticas de sequías y heladas meteorologicas en Colombia.pdf. p. 87. HurtadoG. 1996 Estadisticas de sequías y heladas meteorologicas en Colombia.pdf 87 Search in Google Scholar

IDEAM. 2001. El medio ambiente en Colombia. IDEAM 2001 El medio ambiente en Colombia Search in Google Scholar

I. de hidrología meteorología y estudios ambientales IDEAM. 2019. Probabilidad de heladas en el altiplano cundiboyacense. vol. Comunicado, p. 3. I. de hidrología meteorología y estudios ambientales IDEAM 2019 Probabilidad de heladas en el altiplano cundiboyacense. vol. Comunicado 3 Search in Google Scholar

I. de hidrología meteorología y estudios ambientales IDEAM. 2020. ESTA SEMANA SE PRESENTA ALTA PROBABILIDAD DE HELADAS EN EL. vol. Comunicado, p. 3. I. de hidrología meteorología y estudios ambientales IDEAM 2020 ESTA SEMANA SE PRESENTA ALTA PROBABILIDAD DE HELADAS EN EL. vol. Comunicado 3 Search in Google Scholar

I. G. A. C. IGAC. 2021. Suelos de Boyacá no son tan productivos como se piensa. https://igac.gov.co/es/noticias/suelos-de-boyaca-no-son-tan-productivos-como-se-piensa-igac (Accessed Mar 29). I. G. A. C. IGAC 2021 Suelos de Boyacá no son tan productivos como se piensa https://igac.gov.co/es/noticias/suelos-de-boyaca-no-son-tan-productivos-como-se-piensa-igac (Accessed Mar 29). Search in Google Scholar

INEI. 2012. IV Censo Nacional Agropecuario. [Online]. Result. Defin. IV Censo Nac. Agropecu. p. 62. Available http://proyectos.inei.gob.pe/web/DocumentosPublicos/ResultadosFinalesIVCENAGRO.pdf. INEI 2012 IV Censo Nacional Agropecuario. [Online]. Result. Defin. IV Censo Nac. Agropecu. 62 Available http://proyectos.inei.gob.pe/web/DocumentosPublicos/ResultadosFinalesIVCENAGRO.pdf. Search in Google Scholar

Instituto Geográfico Agustín Codazzi. 2021. En dónde están los mejores suelos para cultivar en el país? Available https://igac.gov.co/es/noticias/en-donde-estan-los-mejores-suelos-para-cultivar-en-el-pais (Accessed Mar 29). Instituto Geográfico Agustín Codazzi 2021 En dónde están los mejores suelos para cultivar en el país? Available https://igac.gov.co/es/noticias/en-donde-estan-los-mejores-suelos-para-cultivar-en-el-pais (Accessed Mar 29). Search in Google Scholar

J. C. FAJARDO JUNCO. 2014. [Online]. Apoyo a la agricultura de precisión en Colombia a partir de imágenes adquiridas desde vehículos aéreos no tripulados (UAV’S). pp. 1–112. Available https://repository.javeriana.edu.co/bitstream/handle/10554/16484/FajardoJuncoJuanCamilo2014.pdf?sequence=1. J. C. FAJARDO JUNCO 2014 [Online]. Apoyo a la agricultura de precisión en Colombia a partir de imágenes adquiridas desde vehículos aéreos no tripulados (UAV’S) 1 112 Available https://repository.javeriana.edu.co/bitstream/handle/10554/16484/FajardoJuncoJuanCamilo2014.pdf?sequence=1. Search in Google Scholar

Lasso Espinosa, L. D. 1987. Anotaciones sobre el fenómeno de las heladas. Lasso EspinosaL. D. 1987 Anotaciones sobre el fenómeno de las heladas Search in Google Scholar

Lesser, A. 2014. Big data and big agriculture. Available https://gigaom.com/report/big-data-and-big-agriculture/ (Accessed Apr 6, 2021). LesserA. 2014 Big data and big agriculture Available https://gigaom.com/report/big-data-and-big-agriculture/ (Accessed Apr 6, 2021). Search in Google Scholar

Lopez-chacón, R. and Martínez-garcía, H. 2018. Sistema de Bajo Coste para la Medida y Monitorización en Agricultura Inteligente. pp. 250–256. Lopez-chacónR. Martínez-garcíaH. 2018 Sistema de Bajo Coste para la Medida y Monitorización en Agricultura Inteligente 250 256 Search in Google Scholar

OEI Organización de estado iberoamericano para la educación la ciencia y la cultura. Intelligo - Patentes. Available http://patentes.explora-intelligo.info/ (Accessed Mar 04, 2022). OEI Organización de estado iberoamericano para la educación la ciencia y la cultura Intelligo - Patentes Available http://patentes.explora-intelligo.info/ (Accessed Mar 04, 2022). Search in Google Scholar

Perera, C., Zaslavsky, A., Christen, P. and Georgakopoulos, D. 2014. Sensing as a service model for smart cities supported by internet of things. Transactions on Emerging Telecommunications Technologiesl 25(3): 1–12. PereraC. ZaslavskyA. ChristenP. GeorgakopoulosD. 2014 Sensing as a service model for smart cities supported by internet of things Transactions on Emerging Telecommunications Technologiesl 25 3 1 12 10.1002/ett.2704 Search in Google Scholar

Pivoto, D., Waquil, P. D., Talamini, E., Finocchio, C. P. S., Dalla Corte, V. F. and de Vargas Mores, G. 2018. Scientific development of smart farming technologies and their application in Brazil. Information Processing in Agriculture 5(1): 21–32. PivotoD. WaquilP. D. TalaminiE. FinocchioC. P. S. Dalla CorteV. F. de Vargas MoresG. 2018 Scientific development of smart farming technologies and their application in Brazil Information Processing in Agriculture 5 1 21 32 10.1016/j.inpa.2017.12.002 Search in Google Scholar

Rathore, L. S. and Chattopadhyay, Nabansu 2016. Servicios meteorológicos y climáticos para los agricultores de la India | Organización Meteorológica Mundial. Bulletin no 65(2) https://public.wmo.int/es/resources/bulletin/servicios-meteorológicos-y-climáticos-para-los-agricultores-de-la-india (Accessed Mar 25, 2021). RathoreL. S. ChattopadhyayNabansu 2016 Servicios meteorológicos y climáticos para los agricultores de la India | Organización Meteorológica Mundial Bulletin no 65 2 https://public.wmo.int/es/resources/bulletin/servicios-meteorológicos-y-climáticos-para-los-agricultores-de-la-india (Accessed Mar 25, 2021). Search in Google Scholar

Salas, F. 2021. Ranking Mundial de Países pro Inteligencia Artificial - IA Latam. Available https://ia-latam.com/2019/08/19/ranking-mundial-de-paises-pro-inteligencia-artificial/ (Accessed Jun 7). SalasF. 2021 Ranking Mundial de Países pro Inteligencia Artificial - IA Latam Available https://ia-latam.com/2019/08/19/ranking-mundial-de-paises-pro-inteligencia-artificial/ (Accessed Jun 7). Search in Google Scholar

SENAMHI and MeteoSwiss. 2018. Designing user-driven climate services. SENAMHI and MeteoSwiss 2018 Designing user-driven climate services Search in Google Scholar

Sørensen, C. G., et al 2010. Conceptual model of a future farm management information system. Computers and Electronics in Agricutlure 72(1): 37–47. SørensenC. G. 2010 Conceptual model of a future farm management information system Computers and Electronics in Agricutlure 72 1 37 47 10.1016/j.compag.2010.02.003 Search in Google Scholar

Torquebiau, E., Rosenzweig, C., Chatrchyan, A. M., Andrieu, N. and Khosla, R. 2018. Identifying Climate-smart agriculture research needs. Cah. Agric. 27(2). Doi: 10.1051/cagri/2018010. TorquebiauE. RosenzweigC. ChatrchyanA. M. AndrieuN. KhoslaR. 2018 Identifying Climate-smart agriculture research needs Cah. Agric. 27 2 10.1051/cagri/2018010 Otwórz DOISearch in Google Scholar

Torres, L. and Rangel, E. 2005. [Online]. Protocolo para el control de calidad de la información meteorológica en las etapas de obtención, evaluación, verificación, cálculo y procesamiento. 1(1): 1–55. Available http://documentacion.ideam.gov.co/openbiblio/bvirtual/020777/PROTOCOLOPARAELCONTROLDECALIDADDEINFORMACIONMETEOROLOGICA/Protocolo_control_meto_.PDF. TorresL. RangelE. 2005 [Online]. Protocolo para el control de calidad de la información meteorológica en las etapas de obtención, evaluación, verificación, cálculo y procesamiento 1 1 1 55 Available http://documentacion.ideam.gov.co/openbiblio/bvirtual/020777/PROTOCOLOPARAELCONTROLDECALIDADDEINFORMACIONMETEOROLOGICA/Protocolo_control_meto_.PDF. Search in Google Scholar

Triantafyllou, A., Tsouros, D. C., Sarigiannidis, P. and Bibi, S. 2019. An architecture model for smart farming, Proc. - 15th Annu. Int. Conf. Distrib. Comput. Sens. Syst. DCOSS 2019 no. May 385–392, doi: 10.1109/DCOSS.2019.00081. TriantafyllouA. TsourosD. C. SarigiannidisP. BibiS. 2019 An architecture model for smart farming Proc. - 15th Annu. Int. Conf. Distrib. Comput. Sens. Syst. DCOSS 2019 no. May 385 392 10.1109/DCOSS.2019.00081 Otwórz DOISearch in Google Scholar

van der Elst, A., Equipo Climandes de la Oficina Federal de Meteorología y Climatología de Suiza (MeteoSwiss). 2018. El Proyecto Climandes: compartiendo experiencias en el diseño de servicios climáticos adaptados a los usuarios | Organización Meteorológica Mundial, Bulletin no. 67(2). Available https://public.wmo.int/es/resources/bulletin/el-proyecto-climandes-compartiendo-experiencias-en-el-diseño-de-servicios (Accessed Mar 25, 2021). van der ElstA. Equipo Climandes de la Oficina Federal de Meteorología y Climatología de Suiza (MeteoSwiss) 2018 El Proyecto Climandes: compartiendo experiencias en el diseño de servicios climáticos adaptados a los usuarios | Organización Meteorológica Mundial Bulletin no. 67 2 Available https://public.wmo.int/es/resources/bulletin/el-proyecto-climandes-compartiendo-experiencias-en-el-diseño-de-servicios (Accessed Mar 25, 2021). Search in Google Scholar

VitaMesa, L. 2021. Uruguay lidera ranking de países más agropecuarios, con 82,6% de tierra dedicada al agro. Available https://www.agronegocios.co/agricultura/uruguay-lidera-ranking-de-paises-mas-agropecuarios-con-826-de-tierra-dedicada-al-agro-2969466 (Accessed Jun 06). VitaMesaL. 2021 Uruguay lidera ranking de países más agropecuarios, con 82,6% de tierra dedicada al agro Available https://www.agronegocios.co/agricultura/uruguay-lidera-ranking-de-paises-mas-agropecuarios-con-826-de-tierra-dedicada-al-agro-2969466 (Accessed Jun 06). Search in Google Scholar

Wolfert, S., Ge, L., Verdouw, C. and Bogaardt, M. J. 2017. Big Data in Smart Farming—a review. Agricutural Systems 153: 69–80. WolfertS. GeL. VerdouwC. BogaardtM. J. 2017 Big Data in Smart Farming—a review Agricutural Systems 153 69 80 10.1016/j.agsy.2017.01.023 Search in Google Scholar

Polecane artykuły z Trend MD

Zaplanuj zdalną konferencję ze Sciendo