Drones and Blockchain Integration to Manage Forest Fires in Remote Regions

Round 1

Reviewer 1 Report

Summary:

The manuscript proposes the use of blockchain technology to manage the actions of drones for the task of delivering firefighting materials and monitoring forest fires in remote areas. Specifically, the drones are used as nodes in the blockchain network, and the drones are authenticated through smart contracts to ensure the authenticity of data sources. Share and record the process of drone mission execution and mission status to facilitate tracking drones and keep track of firefighters, wind direction and fire status through drones. However, there are some issues of concern.

 

Issues:

1. L274: “As mentioned before, the image data are not transmitted to the blockchain; ”. How to ensure that a centralized storage solution with old image/video caching problems is not a disaster dispatch. How to ensure that fire suppression tasks are not affected when a single point of failure problem occurs. 

2. There is no relevant performance testing done on the proposed smart contract, and whether it meets the application scenario is open to discussion.

Author Response

Reviewer 1

1. a) The manuscript proposes the use of blockchain technology to manage the actions of drones for the task of delivering firefighting materials and monitoring forest fires in remote areas. Specifically, the drones are used as nodes in the blockchain network, and the drones are authenticated through smart contracts to ensure the authenticity of data sources. Share and record the process of drone mission execution and mission status to facilitate tracking drones and keep track of firefighters, wind direction, and fire status through drones. However, there are some issues of concern.

 

Thank you very much for reviewing our manuscript and providing constructive comments.

 

1. b) L274: “As mentioned before, the image data are not transmitted to the blockchain; ”. How to ensure that a centralized storage solution with old image/video caching problems is not a disaster dispatch. How to ensure that fire suppression tasks are not affected when a single point of failure problem occurs.

 

Thanks for your comment. The authors have made relevant amendments to the paper and hope that the revised version is to your satisfaction.

 

Proposed framework solution

When combating fires in remote regions, the integration of smart contracts based on blockchain and UAVs collaboration makes operations easy to control and more robust to a single point of failure. Blockchain can efficiently control drone performance, support collaboration to avoid an operation failure, secure it, and ensure drones remain in the mission area. Data in a blockchain-based network is public for each node, and therefore not only does each node has access to other nodes' data, but it also controls its attack detection. As a result, the single point of failure challenge is mitigated [61]. For ceasing wildfire in remote regions, the current paper focuses on examining the achieved solutions by BC and identifying decentralized drone mission challenges. Figure 1 presents the process of uploading mission status reports to the blockchain network.

.

.

.

.

Managing the operations: The drone needs to report the firefighters' location, wind direction, and fire status. As mentioned before, the image data are not transmitted to the blockchain. The InterPlanetary File System (IPFS) is used as decentralized storage to store the hash values of image data imported from the Decentralized Application (DApp) [69]. Back to the previous example, the function "checkFirefighters" causes the drone to send pictures of each firefighter. If one of them is in trouble, the station and other firefighters can be informed of an alarm. As mentioned in the previous section, when the battery's capacity is almost finished, the drone needs to return to the station; a new drone will continue the mission. Figure 6 presents the algorithm for monitoring the framework of the firefighting operation with blockchain technology.

 

 

 

 

2. c) 2. There is no relevant performance testing done on the proposed smart contract, and whether it meets the application scenario is open to discussion.

 

Thanks for this suggestion. The highlights have been amended and now read as the following:

 

 

The nodes based on joint planning need to reach a common aim of the operation. Since decision-making in a distributed network is complex, autonomous solutions can solve the challenge [63]. Besides, unit testing is used to analyze the correctness of individual components in smart contracts. A unit test can provide an obvious reflection of mistakes if the test fails [64]. Blockchain technology offers an effective solution to ensure that all users are connected to share their data decentralized. The proposed framework enables the drone's performance under control through the smart contract. In this procedure, the drone controller and the manager in the fire station follow the terms of the smart contract security system.

 

 

Author Response File: Author Response.docx

Reviewer 2 Report

The paper’s subject is interesting and relevant. In general, the paper has a good presentation. I’d like to propose some recommendations:

-        Could you consider other methods for fair detections, for example, remote sensing based:

o   Zaitseva, E., Stankevich, S., et al. Assessment of the risk of disturbance impact on primeval and managed forests based on earth observation data using the example of Slovak Eastern Carpathians, IEEE Access, 2021, 9, pp. 162847–162856

o   Naderpour, M., Rizeei, H.M., Khakzad, N., Pradhan, B. Forest fire induced Natech risk assessment: A survey of geospatial technologies, Reliability Engineering and System Safety, 2019, 191,106558

o   Soubry, I., Doan, T., Chu, T., Guo, X. A systematic review on the integration of remote sensing and gis to forest and grassland ecosystem health attributes, indicators, and measures, Remote Sensing, 2021, 13(16),3262

-        Could you modify Fig.5? It is not understanding.

-        TFig.6 is not informative and can be deleted

Author Response

Reviewer 2

1. a) The paper’s subject is interesting and relevant. In general, the paper has a good presentation. I’d like to propose some recommendations

 

 

We would like to thank Reviewer 2 for the positive evaluation of our manuscript. The authors have made relevant amendments to the paper and hope that the revised version is to your satisfaction

 

 

1. d) Could you consider other methods for fair detections, for example, remote sensing based:

 

o   Zaitseva, E., Stankevich, S., et al. Assessment of the risk of disturbance impact on primeval and managed forests based on earth observation data using the example of Slovak Eastern Carpathians, IEEE Access, 2021, 9, pp. 162847–162856

 

o   Naderpour, M., Rizeei, H.M., Khakzad, N., Pradhan, B. Forest fire-induced Natech risk assessment: A survey of geospatial technologies, Reliability Engineering and System Safety, 2019, 191,106558

 

o   Soubry, I., Doan, T., Chu, T., Guo, X. A systematic review on the integration of remote sensing and gis to forest and grassland ecosystem health attributes, indicators, and measures, Remote Sensing, 2021, 13(16),3262.

 

Thank you for this comment. The explanation from these papers has been added.

 

1. Introduction

    Different ecosystems such as forests and grasslands provide a natural resource and maintain resilience and natural cycles. These environments are characterized by a wide variety of attributes. Specific attributes of each ecosystem need to be monitored to properly assess their health. These certain attributes are aligned with their ecosystem stressors. The main stressor for forests is “climate change”, and the second most important is “Fire disturbance” [1]. Since traditional methods in ecosystem health monitoring had limitations, the combination of Geographic Information System (GIS) with Remote sensing became more and more popular for monitoring various Spatio-temporal scales. These technologies allow up-to-date monitoring and prediction of forest disturbance risks [2].

    Forest fires in remote regions pose severe threats to ecosystems and are considered important drivers of climate change with adverse impacts on the environment [3]. There is a need for an emergency response to detect the exact location of fires in remote regions and prevent it from turning into a disaster [4]. However, in the field of firefighting, departments are slow to implement novel technologies due to restrictive protocols and uncertainty about value addition [5]. Therefore, many studies have been conducted to improve scientific methods to save the environment, ecosystems, and risks to the public [6]. In this regard, to reduce the destructive effects of forest fires in remote regions, researchers have used wireless sensor networks and machine learning models [7]. Furthermore, operational models and forest fire simulations have been created to predict fire behavior [8]. The other technique for forest fire detection is using satellite imagery and forest fire modeling considering spatial parameters [9]. Recently, deep learning frameworks have been utilized to predict forest fire progression and protect human lives [10]. Due to the dangerous nature of firefighting operations, using robots in extinguishing fires in remote regions has been of particular interest [11].

    Unmanned Aerial Vehicles (UAVs) and drones are able to fly autonomously without a human pilot and are remotely controlled [12]. One of the drone usages is in a warehouse to scan items and products and communicate to managers for any appropriate action [13]. Recently, drones have achieved an important role in the logistics sector since they have some advantages such as time-saving by getting away from the traffic, and environmental friendliness by reducing carbon footprint. However, since drone management relies on machine-learning techniques, there are some limitations for example employing a highly skilled person that increases costs. Another issue is a limited battery problem that can be handled by drone-charging facilities, but it can lead to a significant initial cost. Accordingly, there is necessary to define the factors that may play a remarkable role in the adoption of drones in different industries [14].

 

 

 

1. b) Could you modify Fig.5? It is not understanding.

 

 

Thank you for this comment. The sub-pictures of Figure 5 have been numbered to be more understandable.

Figure 5. The steps of operation of fire suppression using a drone

 

 

1. c) TFig.6 is not informative and can be deleted.

 

 

Thank you for this comment. Figure 6 has been removed.

 

Author Response File: Author Response.docx

Reviewer 3 Report

The manuscript is well written, understandable, interesting and from the standpoint of English is ok. Also, I find the general topic of the manuscript entitled “Drones and Blockchain Integration to manage forest fires in remote regions” related to the profile of the journal. The authors should consider some comments.

The” introduction” is divided in three sections, and I find it quite useless. I propose to merge these sections in one structure introduction or at least merge “1.1” and “1.2”. Furthermore, I will suggest adding more discussion. The limitations of drones like the cost and the experienced staff that is needed to manage the drones are important at least to be mentioned. Addittionally, the characteristics of the area (vegetation, density of the forest etc.), the wireless communication in such remote areas should be taking account in the discussion. Overall, I should recommend discussing a little bit the proposed solution in a more environmentally friendly way and considering the importance of the people work in the field. Technology and drones give rise to manage forest fires but always combine with the knowledge of the area and foresters experienced in the field.

Line 11: Does not exist in the names

Line 172: Table 1: I found it pointless

Line 210: (a) powder; (b) balls

Line 241: Number the images correct

Author Response

Reviewer 3

1. a) The manuscript is well written, understandable, interesting and from the standpoint of English is ok. Also, I find the general topic of the manuscript entitled “Drones and Blockchain Integration to manage forest fires in remote regions” related to the profile of the journal. The authors should consider some comments.

 

 

We would like to thank Reviewer 3 for the positive evaluation of our manuscript. The authors have made relevant amendments to the paper and hope that the revised version is to your satisfaction

 

 

1. b) The” introduction” is divided into three sections, and I find it quite useless. I propose to merge these sections in one structure introduction or at least merge “1.1” and “1.2”. Furthermore, I will suggest adding more discussion. The limitations of drones like the cost and the experienced staff that is needed to manage the drones are important at least to be mentioned. Additionally, the characteristics of the area (vegetation, the density of the forest, etc.), and the wireless communication in such remote areas should be taking account in the discussion. Overall, I should recommend discussing a little bit the proposed solution in a more environmentally friendly way and considering the importance of the people working in the field. Technology and drones give rise to managing forest fires but always combine with the knowledge of the area and foresters experienced in the field.

 

Thanks for your comment.

The sections of the introduction are merged.

The limitations of drones are considered.

The characteristics of the area have been added to the introduction.

An explanation has been added to explicit the role of drones in reducing carbon footprint.

1. Introduction

    Different ecosystems such as forests and grasslands provide a natural resource and maintain resilience and natural cycles. These environments are characterized by a wide variety of attributes. Specific attributes of each ecosystem need to be monitored to properly assess their health. These certain attributes are aligned with their ecosystem stressors. The main stressor for forests is “climate change”, and the second most important is “Fire disturbance” [1]. Since traditional methods in ecosystem health monitoring had limitations, the combination of Geographic Information System (GIS) with Remote sensing became more and more popular for monitoring various Spatio-temporal scales. These technologies allow up-to-date monitoring and prediction of forest disturbance risks [2].

    Forest fires in remote regions pose severe threats to ecosystems and are considered important drivers of climate change with adverse impacts on the environment [3]. There is a need for an emergency response to detect the exact location of fires in remote regions and prevent it from turning into a disaster [4]. However, in the field of firefighting, departments are slow to implement novel technologies due to restrictive protocols and uncertainty about value addition [5]. Therefore, many studies have been conducted to improve scientific methods to save the environment, ecosystems, and risks to the public [6]. In this regard, to reduce the destructive effects of forest fires in remote regions, researchers have used wireless sensor networks and machine learning models [7]. Furthermore, operational models and forest fire simulations have been created to predict fire behavior [8]. The other technique for forest fire detection is using satellite imagery and forest fire modeling considering spatial parameters [9]. Recently, deep learning frameworks have been utilized to predict forest fire progression and protect human lives [10]. Due to the dangerous nature of firefighting operations, using robots in extinguishing fires in remote regions has been of particular interest [11].

    Unmanned Aerial Vehicles (UAVs) and drones are able to fly autonomously without a human pilot and are remotely controlled [12]. One of the drone usages is in a warehouse to scan items and products and communicate to managers for any appropriate action [13]. Recently, drones have achieved an important role in the logistics sector since they have some advantages such as time-saving by getting away from the traffic, and environmental friendliness by reducing carbon footprint. However, since drone management relies on machine-learning techniques, there are some limitations for example employing a highly skilled person that increases costs. Another issue is a limited battery problem that can be handled by drone-charging facilities, but it can lead to a significant initial cost. Accordingly, there is necessary to define the factors that may play a remarkable role in the adoption of drones in different industries [14].

    Drones can effectively monitor operations to avoid undesirable results when natural disasters occur. They can be a tool to get “Aerial photography” used for the investigation of disaster management [15]. Aerial pictures of operations facilitate detailed descriptive analysis of equipment [16] and give helpful clues for scene understanding [17]. Moreover, these robotic vehicles can carry payloads and fly to remote regions [18]. Multi-drones need interaction to share data such as locations, paths, tasks, and purposes. Complex interactions between drones and stations can be supported by artificial intelligence (AI), the Internet of Things (IoT), and Cloud/Edge computing [19]. Modern technologies can minimize casualties and damages, enable quick responses, and prevent false alarms. However, information systems confront a primary challenge since they require trust in centralized management that cannot effectively protect data from unauthorized access [20]. Therefore, there is a need for distributed networks to access information and facilitate data exchange in a secure procedure [21].

 

 

 

1. i)

 

Line 11: Does not exist in the names

 

Thank you for this comment. The names have been corrected.

 

Line 172: Table 1: I found it pointless

 

Thank you for this comment. Table 1 has been removed.

 

 

Line 210: (a) powder; (b) balls

 

Thank you for this comment. The comment has been considered.

 

Line 241: Number the images correct?

 

Thank you for this comment. The sub-numbers of Figure 5 have been corrected.

 

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

I have no other comments and recommendations. Authors modified the manuscript according to my recommendations.