Keywords

1 Introduction

Historically, the concept of IoT has its roots in the ancient Beacon Tower system. This system was employed in China and other countries to rapidly warn about unexpected fires on hills or towers as an early alert for potential enemy invasions (Zhao et al., 2013). IoT is an emerging and innovative technology, and many definitions and frameworks have been studied, analyzed, developed, and proposed. Generally, the architecture framework of IoT includes four key layers: (a) a device layer, (b) a network layer, (c) a data analytics layer, and (d) an application layer (Sahal et al., 2021; Singh et al., 2021). IoT-based technologies find applications in various aspects and areas of life and science (Maksimovic, 2018; Nižetić et al., 2020). The primary objective of this revolutionary technology is to enhance the quality of life by transforming processes in different fields. Figure 1 illustrates the main IoT-related applications related to forest topics.

Fig. 1
Three radial diagrams on the applications of the Internet of Things, Enviormental Internet of Things, and the Forest Internet of Things. Each component has sub-components.

Applications of Internet of Things (IoT), Environmental Internet of Things (EIoT) and Forest Internet of Things (FIoT) (Asghari et al., 2019; Bayne et al., 2017; Bo & Wang, 2011; Choudhry & O’Kelly, 2018; Hock et al., 2016; Li et al., 2019; Zhao et al., 2005)

Full size image

According to the Food and Agriculture Organization (FAO), a forest is defined as “a land area of more than 0.5 ha, with a tree canopy cover of more than 10%, which is not primarily used for agricultural or other non-specific non-forest purposes” (Foundation, C.D.R., 2020). As of 2016, the total global forest area was estimated at 39,958,245.9 km2 (Forest Area, 2022).

Forests are recognized as dynamic providers of ecosystem services (Spanos et al., 2021), and any significant disruption to their existing structure can have various impacts on different aspects of human well-being. Forest fires, along with insect infestations and forestland fragmentation, are among the major causes of unsustainability. Conventional forest fire management (FFM) often lacks coherence and efficiency, hindering the development of an integrated approach and complicating decision-making processes, resulting in the inefficient use of human and material resources. The primary objectives of integrated FFM are twofold: (a) to minimize the damage caused by unintentional fires and (b) to maximize the benefits of intentional, prescribed fires (Nyongesa & Vacik, 2018). Prescribed fires offer several positive effects, such as controlling insect pests, reducing competition for nutrients, and providing habitats for nesting birds, making them an integral component of FFM due to the ecological benefits that often outweigh their negative effects.

It is widely acknowledged that an integrated, sustainable, human-centered, and proactive approach to FFM should address the root causes of extreme forest fires (wildfires) (Wunder et al., 2021). Many international initiatives have been launched in relation to forest land restoration, which are highly connected with the main goals of an IoT ecosystem:

  • The Bonn Challenges aim to bring 150 million ha of degraded and deforested landscape into restoration by 2022 and 350 million ha by 2030 following the global agenda for sustainable development.

  • New European Union (EU) Forest Strategy for 2030 recognizes the crucial and catalytic role of forests in making Europe the first neutral continent by 2050.

  • The EU Biodiversity Strategy for 2030 sets out a pledge to plant at least 3 billion additional trees by 2030 following the ecological principles (right tree in right place and for right purpose).

IoT is a new technology around the globe and has the potential to contribute to socially and ecologically sustainable forestry. Regarding the agriculture sector, the social impacts and inequalities linked to digital agriculture need to become a key concern of policy makers so that the above-mentioned potential will be determined (Hackfort, 2021). Agriculture is undergoing high transformation due to the technological assets. On the other hand, forestry is running behind the majority of the industries that have already adopted that practice (Choudhry & O’Kelly, 2018). This has recently started to change. In forestry, digital technologies confront a system which has been established for a long time—76% of forests globally are public-owned with the remainder being held by holders of average less than one-hectare land. Management of state or public forests is usually conservative and private forests lack scale and expertise needed for adoption of new technologies (Choudhry & O’Kelly, 2018). Forest Internet of Things (FIoT) is a category of IoT and refers to smart devices distributed in forests mainly for monitoring, management, fire detection, and prevention (Sahal et al., 2021).

Figure 2 illustrates an FFM framework that highlights the primary facets and associated activities (pre-, at-, and post-fire management) where IoT can serve as a valuable tool. Numerous suggestions emphasize the urgent need for a digital forest system. Choudhry and O’Kelly (2018) have emphasized that there is no time to waste, and action should begin immediately, rather than waiting for technology to fully mature. Therefore, it is imperative to develop a roadmap by initiating small-scale projects initially and subsequently expanding them on a larger scale. However, it is important to consider a structure that can mitigate operational and financial risks. Additionally, Nassef (2022), a leading figure in climate adoption policy at the United Nations, noted in an interview that the substantial transformation of our actions over the next decade, including the adoption of digital technologies, is crucial to preventing human extinction.

Fig. 2
A circular flow diagram illustrating the steps to achieve goals in forest fire management, including pre-fire management, at-fire management, and post-fire management.

A conceptual fire management framework (Daskalakou et al., 2014; Mavsar et al., 2011; Nyongesa & Vacik, 2018; Salam, 2020)

Full size image

2 IoT Technologies in FFM—Forest Fire Internet of Things (FFIoT)—Socioeconomic Aspects

Wildfires are climate-related events with significant impacts on both climate change and human well-being. Technology plays a crucial role in understanding the causes of “green swan” events, providing innovative solutions to address them. Emerging technologies like the Internet of Things (IoT) and artificial intelligence (AI) have become valuable and powerful tools for responding promptly, efficiently, and innovatively to these situations (Nassef, 2022).

Remote participation in forest management can give rise to specific dynamics of power and politics, as forests are increasingly viewed as global resources (Gabrys, 2020). Figure 3 presents a visual example of Forest Fire Internet of Things (FFIoT) in a forest landscape. Various smart devices are collecting various types of data and communicating wirelessly through a gateway. The messages received by the base gateway are stored and then transferred to an online database. This collected information is readily accessible for further analysis and informed decision-making. Figure 4 provides a detailed breakdown of the four layers of FFIoT for end-users.

Fig. 3
A diagram showing the process of using a forest fire internet of things (FFiot) in a forest landscape, with various smart devices collecting data and communicating wirelessly through a gateway. The messages received by the base gateway are stored and transferred to an online database for further analysis and informed decision-making.

A diagram of architecture of FFIoT (Icons were inserted from painting and draw.io software)

Full size image
Fig. 4
A diagram showing the four layers of a forest fire internet of things (FFiot): data analytics layer, forestry information perception layer, application layer, and administration and decision-making layer.

A conceptual framework of FFIoT

Full size image

3 Related Work—Review

We conducted this review and analysis based on the principles of forest fire management (FFM) and IoT (Internet of Things). We compiled an indicative list of Forest Fire Internet of Things (FFIoT) articles published from the beginning of 2017 through May 2022, using various databases. The search was concluded in May 2022. The keywords employed for the search included “Forest Internet of Things,” “Smart Forests,” “Digital Forestry,” and “Forestry 4.0.” We focused on studies primarily related to forest fires. While this list is not exhaustive, it provides comprehensive coverage of the topic in terms of geography and ecology, highlights key findings, addresses gaps in the literature, and outlines future challenges in the field.

Addressing the critical issue of air pollution resulting from wildfires, Mahajan et al. (2017) developed a framework utilizing IoT for predicting the concentration of particulate matter with a size of 2.5 µm or less (PM2.5), a significant air pollutant released into the atmosphere following a fire. These particles have severe adverse effects on human health, and this innovative study aims to achieve highly accurate predictions to mitigate potential health risks. The study utilized data from 119 Airbox Devices stations in Taiwan for experimentation and evaluation. Prior studies had only presented a generic model applicable to all stations. This limitation was overcome by introducing the concept of clustering monitoring stations into grids based on their distances from each other. The results have the potential to enhance air quality monitoring systems, support local and national environmental objectives, and contribute to meeting sustainable development goals (SDGs).

Dubey et al. (2018) introduced a cluster-based approach for forest fire prediction in Spain, leveraging sensor network technology. This approach involved three key steps: (a) Collecting forest fire images from sensors distributed across various geographical areas; (b) Analyzing these images using a clustering algorithm designed for forest fire prediction. This algorithm grouped clusters based on different colors and textures corresponding to specific locations; (c) Implementing a system for early forest fire detection, minimizing response time by calculating the average value of cluster trees within the sensor network.

By upgrading the existing wireless sensor network (WSN) fire detection approach, this system achieved a reduction in false alarms. The hardware components of this system included:

  • A gas humidity sensor for measuring humidity and temperature.

  • A gas sensor for detecting gases like carbon dioxide, carbon monoxide, methane, hydrogen, propane, etc.

  • A flame sensor converting detected light into infrared light.

  • A microprocessor/microcontroller, compact in size, capable of simulations, and responsible for controlling all the sensors.

  • A buzzer alarm for alerting.

This application offers several advantages:

  • Early detection of fires with the ability to promptly inform authorities.

  • High accuracy of 96.7% on test data.

  • A straightforward and simple model.

New findings from this approach include the implementation of a highly accurate fire prevention and detection system, significantly reducing false alarms. This method outperformed satellite-based forest fire detection in terms of effectiveness. Early and accurate wildfire detection plays a crucial role in limiting the fire’s extent, and the user-friendly model enhances coordination efforts.

In the same year, Toledo-Castro et al. (2018) developed a forest fire controller based on a fuzzy logic model. The primary objective was to analyze environmental information, including meteorological data, oxygen levels, and polluting gases, to rapidly assess potential forest fire risks and detect ongoing fires. They also introduced decision-making methods aimed at reducing real-time analysis of environmental data. The proposed wireless sensor network (WSN) had the capability to conduct environmental monitoring, including parameters such as temperature, humidity, wind speed, rainfall, carbon dioxide and monoxide levels, and oxygen levels. This system was built upon a prototype of an IoT device distributed across various forest areas. This innovative approach to designing an effective environmental monitoring network, capable of providing scientific data essential for ecosystem management and decision-making processes, addresses one of the key challenges in environmental monitoring (Li et al., 2019). Moreover, there is potential for improvement by incorporating new sensors into the system.

Choudhry and O’Kelly (2018) highlighted that fire monitoring is among the 15 promising practices within the technological landscape of precision forestry, encouraging the participation of forestry companies. To illustrate how forest fires are perceived by society and the resulting chain reactions, they focused on the 2017 wildfire in South America, which resulted in forest owners losing at least four percent of their forests. In response to these losses, owners are now exploring ways to enhance detection and control systems using digital technologies. Regarding national parks, Peinl (2020) introduced the design and implementation of a novel, flexible, and open architectural system called ASPires, aimed at early prevention and detection of forest fires. ASPires evaluated the capabilities of a new low-power LoRa wireless technology, capable of covering long distances of up to 15 km. The system was tested for viability, accuracy, and effectiveness in three national parks: (a) Pyrin, Bulgaria; (b) Mavrovo, North Macedonia; and, (c) Pelister, North Macedonia. The system included the following components: (a) sensors for measuring temperature, air and soil humidity, carbon dioxide and monoxide levels, dust particle concentrations, wind speed and direction, and noise levels; (b) a power regeneration component or rechargeable battery; and, (c) an optional local data storage facility and wireless transmission technology. The key advantages of this system were:

  • Reduction in costs and damage caused by forest fires.

  • Overcoming existing drawbacks.

  • Utilization of cost-effective sensors.

  • Low energy consumption, whether powered by batteries or other energy sources.

  • Long-range coverage and operation on free frequencies using wireless technologies.

  • Deployment of flying gateways instead of field-based gateways, which are prone to damage.

These advancements represent significant progress in forest fire prevention and mitigation efforts.

In the same year, Srividhya and Sankaranarayanan (2020) recognized the need for periodic monitoring and surveillance of forest fires in vulnerable areas due to the accelerating impact of global warming on the frequency and severity of forest fires. In their work, they proposed an IoT-Fog-based framework for forest fire monitoring systems, with the goal of addressing limitations in other wireless sensor networks. The IoT-Fog framework allows for the balancing of computing and data analysis operations, distributing these functions across aggregator and central cloud layers. The cloud component plays a pivotal role in managing all fire-related notifications, issuing alerts to forest offices and individuals in the surrounding areas. Some of the main advantages of this system include low bandwidth usage, reduced latency, and the ability to handle heterogeneous data computation effectively. Future challenges in this sector include optimizing energy consumption and strategically placing Fog nodes to enhance the system’s efficiency and coverage.

Post-fire forest management can benefit from the adoption of an alternative approach to reforestation, complementing the traditional method of manual planting (Gabrys, 2020). While the traditional method remains significant and widely practiced, emerging digital technologies offer the potential to plant billions of trees annually through aerial techniques involving drones and the mass planting of polycultures (mixed forests) using smart machines capable of separating various species. Additionally, planting at scale from the sky is being explored (Nassef, 2022), among other innovations. The idea that forests can transform into digital environments managed through technology to address mitigation, climate change, and environmental challenges has the potential to shift the paradigm in forest management. This vision is not distant and may become a routine practice in future forest management, accelerating the adoption of alternative reforestation methods.

A novel methodology for forest fire warning based on IoT was designed and implemented in the Jizhou forest district of China by Han (2021). This forest plays a crucial role as the primary provider of ecosystem services for the nearby major cities of Beijing and Tianjin. The system consisted of three key components: sensing, transmission, and application. The application layer of the system enabled rapid and effective communication with the fire-affected region. In general, the forest fire warning system was complex and required professional expertise and scientific input. The technical aspects involved the use of smart devices for data acquisition, transmission, control, display, monitoring, and management. This innovative system has the potential to be implemented not only in remote forest areas but also in urban and peri-urban forests, enhancing the capacity to detect and respond to forest fires effectively.

In their comprehensive review, Sahal et al. (2021) outlined the transition from Industry 4.0 to Forestry 4.0, facilitated by advanced technologies like the Internet of Things (IoT). This transition is expected to bring about social, environmental, and economic benefits, and the review provided a roadmap for this transformation. Forest fires are recognized as one of the major risks negatively impacting the environment and are a significant dimension of Forestry 4.0. Given the limited practical studies available, the review emphasizes the research gap regarding the potential adoption of the Internet of Forest Things (FIoT) for environmental sustainability. It particularly focuses on forest fire management and detection, considering it a central component of Forestry 4.0.

Singh et al. (2021) introduced a novel IoT system designed for forest fire detection and prediction in India, aiming to address previous limitations such as connectivity, edge device fire detection, and power consumption. The proposed framework improved the short-term estimation of forest fire risks and the early detection of fires through the utilization of LoRa communication and edge computing-based gateways. To detect the emergence of fires, sensor nodes with integrated sensors were deployed. These sensors measured various parameters, including temperature, humidity, light intensity, rainfall, wind speed, and infrared values. The system was powered by a battery, and local storage was available for data backup. Additionally, a computer vision node (CVN) worked in tandem with the sensor node to provide precise information.

Ntinopoulos et al. (2022) emphasized the connection between the Fire Weather Index (FWI) and climate conditions in Greece. They utilized datasets derived from daily wind speed, precipitation, relative humidity, minimum, mean, and maximum temperature, as well as solar irradiance. Regions with high FWI, indicating drier and hotter conditions, exhibited a higher likelihood of experiencing more forest fires. Meteorological information was instrumental in assessing the fire danger classification in these areas. Maintaining, expanding, and connecting databases could prove to be a valuable tool for predicting forest fires at regional, national, and global levels.

In a recent analysis, the focus was on biodiversity products that can be retrieved from space using available space assets. Among these products, those describing the abiotic drivers of ecological disturbance, including the biological effects of fire disturbance, were ranked as some of the most important products in terms of relevance, feasibility, accuracy, and maturity (Skidmore et al., 2021). The study highlights the importance of detecting the planet’s dramatic changes over decades rather than relying solely on periodic (annual or seasonal) observations to align with operational and research-oriented management and reporting needs. Ensuring the continuity of free and open data, affordability, and establishing links between in-situ and remote sensing observations are essential for this purpose.

4 Conclusions and Challenges

The potential for high-value creation through improved forest fire management is one of the key environmental advantages of Forest Fire Internet of Things (FFIoT). It has the potential to enhance operational ecosystem monitoring, contributing to national and global ecological security (Li et al., 2019). The development of a monitoring system that combines remote sensing observations, in-situ measurements, and model simulations ensures high accuracy. These digital archives are crucial for monitoring planet-wide changes, including forest fires (Kays et al., 2020). Additionally, FFIoT can improve health and safety by helping to avoid dangerous locations (Hock et al., 2016). From a societal perspective, FFIoT applications shift participation from local actors to a more global stage, promoting democratic engagement and including all relevant stakeholders, thereby contributing to diverse and inclusive forest management (Gabrys, 2020; Gabrys et al., 2022). This collaborative approach fosters the development of sustainable and durable databases, and digital technologies can serve as a catalyst for environmental initiatives and regulations, transforming and expanding previous practices and technologies (Gabrys et al., 2022). Moreover, technology adoption in environmental issues can motivate the business sector to become more involved due to economic benefits.

However, there are drawbacks to adopting these new technologies from an environmental perspective. IoT can accelerate the depletion of natural resources and increase the environmental footprint due to the production of electronic devices, leading to higher waste generation and emissions (Nižetić et al., 2020). Effective recycling methods and advanced e-waste management are essential for a greener future. High energy consumption also poses financial challenges, necessitating technical advancements like energy-efficient communication and long-term batteries. Conducting life cycle assessments (LCAs) or environmental impact analyses (EIAs) is critical to evaluating the environmental footprint of these new products.

From a socio-political perspective, digital technologies can exacerbate environmental injustices and inequalities, with privileged actors having easier access. The specialization of workers in response to changing job requirements can lead to social impacts, potentially affecting mental well-being, necessitating an emphasis on the development of soft skills. Preparing the new generation for the transition into the era of big data requires an educational structure. The next steps should focus on further developing technology for data collection, statistical tools for analysis, and suitable infrastructure for data management. Creating a list of selected parameters that can be adopted universally is crucial. Starting with small-scale FFIoT projects and gradually scaling up is advisable. Collaboration across disciplines, better-established research, and a focus on avoiding socio-political inequalities and undemocratic governance are essential (Gabrys, 2020). The alignment of legal frameworks and the development of digital forest legislation are also necessary.

Lastly, in the face of significant planetary changes in which wildfires play a crucial role, the urgency for institutions to recognize the value of “born digital data” and invest in standards for informative data records that contribute to the management and conservation of natural resources is paramount.