IoT in protecting wildlife
Protection of Wildlife Using IoT
Earth has different things that are alive (plants, animals), things that aren’t alive (lakes, rocks), these are called biotic and abiotic factors. And what we call the place that consists of these factors is an ecosystem. There are different types of ecosystems (deserts, rainforests etc.). All the different ecosystems offer different types of animal, plant and insect species. There are of course variety in ecosystem sizes too.
Researchers have always studied the way we interact with ecosystems. We as humans still interact heavily with the worlds’ ecosystems in our day to day life, even though we are in an urban environment, life is thriving with urban tree planting, between the paving stones, in a humid basement, even in your fridge. Living things are everywhere and we have interacted with Earths’ ecosystems all this time. This is why studying and understanding the ecosystems we live in, teaching them, protecting them from false information is important.
Ecosystems can be studied through a variety of approaches—theoretical studies, studies that use the help of a camera or volunteers to monitor a specific ecosystem to gather data, those that look at differences between ecosystems to understand how they work and direct manipulative experimentation. All of these can need the help of technology. Especially emerging technology, as these fields need new and better ways of research.
Technology and science has always helped each other, and they both helped us in our day to day life and also helped us understand ecosystems, then Earth, then the multiplanetary system, then the universe. Both of them are very precious and great tools for humanity.
There is a lot of ways technology is being used in science. There are centrifuge machines, telescopes and microscopes, computers, cameras, sensors; the list is very long. Mainly the use of computers has helped researchers immensely. The way computers are used in research has been very creative, and as the computing technology got better, computers helped researchers even more. Nowadays it’s almost impossible to make research that contain big datasets without computers, as the datasets are bigger because the information that can be gathered in the name of research has increased with better technology.
Internet of Things (IoT) is becoming more and more popular in wildlife research. There is a heavy use of cameras with sensors when it comes to watching and analyzing animal behavior in wildlife. Nowadays IoT is used to protect these animals. There is a long list of dangers for wildlife. Poachers, farming, deforestation are the main issues the researchers focus on.
Wildlife is a vital and a major part of the ecosystem of Earth. As humans, our survival depends on protection of the wildlife. Biodiversity provides food (pollination of plants), provides oxygen, and it has a major part in a lot of systems of earth. If we don’t start protecting it now, the end of life on earth comes closer every single day. Climate change and pollution of the air is measurably getting worse. There are a lot of different ways to protect wildlife -therefore protecting life- and IoT is one of the many techniques scientists use.
IoT is things sharing information with other things via the internet. This information is gathered with sensors and the types of information “things” provide varies. IoT processes huge amounts of information that can’t be possibly processed by a human. It can also predict complicated patterns of connections between information. Because of this, it can handle the chaotic nature of wildlife (for example rainforests). Of course there are a lot of issues that needs to be solved for IoT to be integrated into wildlife but there has been successful attempts at using them in such a complex and chaotic environment.
IoT gathers and uses sound, movement, light, heartbeat, heat, location data through many different sensor types. GPS trackers, cameras, heartbeat monitors, thermal imaging cameras are commonly used. Usage of different sensors varies from environment to environment and type of studies done with IoT. For example, if the study is about the protection of wildlife from illegal hunting, there are cameras with movement sensors to record poachers.
These useful sensors consist of a lot of electronics. And these electronics can break and malfunction as they’re put in complex environments that consist of a lot of animals that vary in size. This has been addressed in Internet of Things for Efficient Wildlife Conservation: Challenges and Opportunities, GCRF Facilitation Grant Report, “Sabah Malaysia is a high humidity region where electronic components could get damaged quite quickly due to environmental factors such as moisture. Further, in jungle terrains, insects could also get attracted to copper within electronic components.”. Later in the report, it also mentions how the cameras could be affected; “The lighting conditions, raining, dust, insects and many other factors could affect the camera.”.
It also mentions problems that arise with the data communication techniques such as 3G etc.
; “Satellite communication is costly. On the other hand, 3G communication is only available in certain spots within the sanctuary. Therefore, the most practical solution seems to be SMS.”. This tells that there is a lot of problems IoT sensors face in the complex environment of a rainforest. The researchers themselves also face many problems as they are the ones setting up the sensors.
This report is focused on detection of poachers and tagging of animals. Main goal is to protect wildlife and biodiversity. They used artificial intelligence to trap the specific animal they need to catch and tag. Another problem they’ve faced was the detection of animals when catching them; “These traps are not perfect and mostly have weight-based or bait base
activation triggers. Therefore random animals could also get trapped in these traps (due to similar weight or attracted to similar baits)”
This problem led them to use AI technology, which is a great tool when integrated with IoT; “…trap door activation could be detected by using different types of sensors such as vibration, noise, light, motion, and camera and so on”, “…we could use low-cost edge computing technologies to detect and identify the animal trapped within a given trap. State of the art deep learning is sophisticated enough today to be able to use in such identification.” This tells
us that IoT is very integratable with AI and its practices. It can open a big door when it comes to detection of different types of animals, and it can help researchers do less identification of animals and focus more on the bigger picture the data IoT will present them.
In this report they also focus on detection of poachers and it’s interesting that they’ve focused on vehicle and boat detection. The poachers in the area use four-wheel vehicles and boats, and it’s prohibited to be in the area of research after 7 pm, therefore the vehicles who are in the area in a prohibited time throughout the day, they report the authorities about the vehicle.
They also mention how boat detection techniques can parallel with car detection techniques; “…combine the data we gather from vehicle detection sensors and combine that knowledge with boat tracking data.”. The report then explains how the boats and the cars work together, as poachers have become more professional at illegal hunting and they work as a team. It’s devastating to see how illegal hunting and illegal logging are big industries.
An article by Carsten Rhod Gregersen “How IoT protects animals, reveals behaviors on World Wildlife Day” starts off with alarming information, the wildlife has lost 500 vertebrate species in the 20th century but he says that we will lose the same amount in the next 20 years. This signals how things became worse and it’s clear that we need to protect these species.
He then shines a light on the rhino horn industry ;“The rhinoceros, for example, is one of the most endangered species in the world”, then he says “…roughly 1,000 rhinos are killed by poachers every year for their horns, which can fetch between $30,000 to $100,000 per kilogram on the black market.” The article then explains how the personal guardian system for rhinos is a very expensive and a weak solution. He suggests using IoT for a more cheap and better working solution to protecting rhinos.
Image from WWF, “Poachers walk free as assault on Zimbabwe rhinos escalates”
It’s clear that IoT devices (sensors etc.) are getting cheaper and better as they are more commonly used in the world; “It’s exciting to think that this is only the start of using our best technologies to protect our most vulnerable species. All variations of IoT sensor types. from image to pressure and proximity sensors. are becoming cheaper and smaller with scale. and there is plenty of scope to enable the tracking animals that are also critically endangered. such as the Pangolin.”
There is more information on protection of Pangolin populations on the article “Video technology from Hikvision aids endangered pangolin populations”.
This article talks about the ancient creatures called pangolins and their importance when protecting wildlife in the forest. They are mostly victims of illegal hunting.
This species are very strong, and has many different tactics to run away or shield themselves. They have scales that are made from keratin which makes them very strong even compared to teeths of predators. And on top of all that the article mentions “The Pangolins’ unique features play an essential part in balancing their delicate ecosystem. Pangolins use their long curved claws to tear apart ant and termite nests. Their tongues which can be up to 16 inches long, gather insects burrowed underground. Feasting on termites and other insects keeps the gentle creatures well-fed, but it also mitigates the destruction of the forests (which includes wildlife and biodiversity) that the insects would otherwise cause.“.
But the article says that the future looks positive. governments and international organizations are providing significant resources to stop poaching of pangolins. Also, it mentions that there is a IoT technology for protecting pangolins; “The system integrates professional video software with dedicated functions including AI-equipped cameras that can identify this unique animal automatically. The system further analyses nearby animal activity and monitors human contact with wildlife”
Image from “Video technology from Hikvision aids endangered pangolin populations”
This technology provide real-time monitoring of pangolins, and it provides great information about their behavior in wildlife. And of course, this technology can be used to detect poachers.
Detection of poachers is also used when it comes to IoT research that focuses on deforestation caused by illegal logging.“IOT Based Anti-Poaching Sensor System for Trees in Forest” focuses heavily on detection of poachers. They say; “ Poaching of monetarily valuable trees has turned into a noteworthy danger to the estate of these trees…” which means that it’s a threat for the wildlife as lack of these trees hurt biodiversity.
The study mentions how the Indian government is trying to restrain exportation and sales of sandalwood, this signals how important the issue is; “The Indian sandalwood tree has turned out to be uncommon as of late, trying to control its conceivable misfortune, the government is attempting to restrain the exportation of sandalwood.”.
They explain their IoT based system in the paper, and it uses 3 main sensor types; tilt sensor (to distinguish the tendency of tree when its being cut), temperature sensor (for forest fires), sound sensor (for recognition of illicit logging, even the sounds produced while cutting out the tree is additionally detected). Data is checked with an app; “Information produced from these sensors is consistently checked with the guide of Blynk App.”, “Blynk App was planned for the Internet of Things. It can control hardware remotely.”.
They use a website for the analysis of results. They gave different trees different IDs and they can check which trees have fallen from this website with the tree ID system.
Another study focusing on trees presents a very interesting term “Internet of Trees”. The study “Forest 4.0: Digitalization of forest using the Internet of Things (IoT)” aims to drastically improve the forest’s ecosystem.
This study explained the term deforestation very well, they say deforestation is “…cutting down trees permanently for agricultural expansion, forestry mining, growth of infrastructure…”. They also mention forest fires, as they are one of the devastating reasons for deforestation. The study then says how deforestation affects wildlife, therefore biodiversity, therefore Earth as a whole.
They explain the term internet of trees with the words “…the trees are embedded with internet-based modules for sensing, communicating, and monitoring the environmental parameters in the forest…” then they mention how this technology uses the internet to share this data therefore making this technology an IoT based system.
An article that focuses on trees, In-Vivo Vibroacoustic Surveillance of Trees in the Context of the IoT, says “Wood-boring insects such as termites, weevils and certain beetles can directly affect the health and growth of specific tree species. There is a wide bibliography on piezoelectric sensors, microphones, accelerometers, laser vibrometry and optical methods used to detect locomotion and feeding sound of larvae or adult pests inside the tree trunk.”.
Their goal with their work is making automated insect surveillance everywhere in the world a common practice. They are also developing a new accelerometer-based sensor that transmits short vibration clips, that help us understand if the tree is infected or not. Another point to consider about trees is the cost of maintaining them in urban environments. Urban trees provide important things essential for the ecosystem such as regulating temperature, filtering air therefore the breath we take is excellent, and they control humidity. Although these are great things, there is another aspect of urban trees and that is the cost. Huge numbers of urban trees create problems that affect public safety. The maintaining of trees consist of inspection, watering and spraying (disinfection), and it is very costly.
A study focusing on this is; “An IoT Tree Health Indexing Method Using Heterogeneous Neural Network”. They use sensors to inspect tree health.
This study says this to emphasize the cost of maintaining urban trees; “In the US, the number of urban trees exceeds 3.5 billion. The total compensatory value of the trees, consisting of removal cost, replacement cost, costs of settling tree damages and deaths, insurance claim, and loss of property value etc., is more than $2 trillion USD.” Of course they realize how it positively affects our environment and human health “The literatures prove that the urban trees are beneficial to human mentality and physical health, such as reducing stress level and blood pressure. Besides, the urban trees provide foods and habitats to animals and other plants, thus enhancing urban biodiversity”. “The inspections and maintenances of urban trees should be conducted regularly to monitor tree health and ensure public safety.
They then explain how they used IoT to protect trees therefore wildlife; “The Internet-of-things (IoT) is a concept of mega integration and it proliferates the deployment of wireless sensors for various applications. Smart applications can be realized by employing artificial intelligence to analyze the data generated from the IoT network. An IoT tree is a tree equipped with various types of sensors. The sensors are connected to the IoT network and provide data for tree health analysis. Such method greatly increases the efficiency of tree health assessment and facilitates the urban tree planning.”
They used a system for evaluating the health of trees, so that they can have a systematic way of assessing the health of trees. “An automated tree health assessment is demanded to improve the efficiency and it can be realized by incorporating sensor monitoring network with AI-based assessment algorithm. In this paper, an automated assessment scheme is developed to evaluate and index the health of urban trees, namely Urban Tree Health Index (UTHI).”
They then note that different cities’ standards are different; “In different cities, the standards of tree health assessments can be slightly different due to the customizations with respect to the climates, land conditions, and other environmental factors.”
This study also use AI like one of the ones mentioned before; “…the indexing problem in this paper is a non-linear problem which requires an advanced algorithm to solve. Machine learning is a widely adopted method for pattern recognition, classification and prediction.”
Study then says that they developed a great tool for efficient tree health assesment with the power of IoT. This is one of the many studies, again underlines the power of IoT integrated with AI. These systems can deal with the complex issues ecosystem has, because they are very though-out and smartly designed data collection and data analization techniques.
Wildlife monitoring is extremely important for many researches and studies. Making sense of behaviour of living creatures and figuring out their habits, repeating activities is beneficial for biodiversity research. Also we need to consider dangerous human-wildlife encounters, which can hurt us and the animals in the wildlife.
There is a paper focusing on this problem, and it uses cloud computing with IoT which is very exciting. It’s “Where’s The Bear?- Automating Wildlife Image Processing Using IoT and Edge Cloud Systems” and it has very interesting conclusions.
They explain what their goal with their paper is; “In this paper, we address these challenges with the design and implementation of a new wildlife monitoring system that leverages recent advances in the Internet-of-things (IoT) and in open source image processing and deep learning for image recognition, to automate image classification and analysis. Our system, called Where’s The Bear (WTB), is an end-to-end, distributed data acquisition and analytics system that implements an IoT architecture consisting of sensors (cameras), “edge clouds,” and a back-end public or private cloud.”
To train their AI with data, they use not only the images of animals, they also empty images of the forest “…our approach combines a small number of empty images from our camera traps at different times of the day with images of animals of interest taken from Google Images.”.
“We use one camera trap (called Main) in this study to evaluate our system.”, “The cameras collect images continuously and to date they have been in use intermittently over the past 3 years to collect over 1.12M images (716.3 GB).” This tells us that these studies always use a lot of GBs of digital storage. Which is very expected, yet 716 GBs is a huge number that can’t be skipped carelessly without being addressed.
Image from ““Where’s The Bear?- Automating Wildlife Image Processing Using IoT and Edge Cloud Systems”
Here’s one of the examples of the pictures the camera captured. The deers look beautiful drinking water from a natural source, this is a great example of how wildlife is so beautiful.
In “IoT Beehives and Open Data to Gauge Urban Biodiversity”, there is a great point that was made on how urbanization is dangerous for the environment. But they also did a counter argument to that which is very interesting; “It is estimated that by 2050 about twothirds of the global population lives in cities. On the one hand, the consequence of urbanization is a large-scale loss of natural habitats, affecting people’s health and well-being”, “…on the other hand, if planned well, ‘green and blue’ cities can maintain nature-friendly habitats and provide ample opportunities for a diverse flora and fauna ecology.”.
Urbanization is considered inevitable in these times, therefore we should focus on protecting wildlife and biodiversity while urbanization is taking place in our society.
They say that their goal is a long-term objective, they want new cities to consider protecting biodiversity as a major decision in planning of cities. “Our objective is to build a
first-of-a-kind supervised machine learning model, that is able to predict urban biodiversity, based on data that can be collected real-time.”
They mention using AI in their IoT system; “Our envisioned AI aims for a system producing ecological quality measures, rather than detailed species counts. These aggregate measures express aspects of ecosystem health, focusing on species groups (insects, flowers, birds, etc.).”, “…the AI model for urban biodiversity and its underlying data engineering pipeline enable an Urban Biodiversity Monitor as a real-time service. An Interactive Bee Tracker, that shows the flight patterns as indicated by the waggle dances projected on an actual city map, and a Bee Health Dashboard might be considered as informative visualizations.”
It is interesting that the term biodiversity isn’t only a countable amount of species. It’s also being researched on DNA levels; “Biodiversity is studied at various levels, from DNA to species and even entire ecosystems.”
They say that the city map they’ve created will create awareness among citizens, becuase it will show the urgentness of the biodiversity crash; “The interactive visualization tools create for citizens awareness and a sense of urgency for the biodiversity crash.
Then they mention how it will affect city designs fo the better as AI will predict what will happen with the old style city designs; “The city digital twin combined with predictive AI tools enable policy makers to come up with city layouts that are nature friendly.”
And of course they mention how the open data their system will provide will help researchers and biologists; “…biologists might use the wealth of data that will be collected (and published as open data) to study bee behavior.
In “Emerging Technologies to Conserve Biodiversity” they mention how new technologies are being developed; “Technologies to identify individual animals, follow their movements, identify and locate animal and plant species, and assess the status of their habitats remotely have become better, faster, and cheaper as threats to the survival of species are increasing.“
In this article they talk about the new problems the new technologies create, and how some habitats are not protected because researchers generally focus mostly on the same habitats. This can be said for IoT too, as it’s getting better everyday and as the technology progresses new problems emerge because new data is giving us new information about our ecosystem, and we will be aware of the problems we didn’t know existed before. Of course this is a good thing, because we will be able to act before these problems get even bigger.
As IoT develops more and more, research on wildlife is going to get more efficient and overall better. In conclusion, it’s clear that IoT is a great tool, and hopefully IoT will continue developing. The researchers who are trying to find ways of protecting earth, wildlife, ecosystem, biodiversity and humanity from climate changes’ effects are and will be using this technology for years to come.
Literature:
- Hikvision, Video technology from Hikvision aids endangered pangolin populations, April 21 2022
- IOT Based Anti-Poaching Sensor System for Trees in Forest, Parthiban M, Dharani M, Kathiga S, Keruthika M (International Journal of Innovative Technology and Exploring Engineering (IJITEE) ISSN: 2278-3075, Volume-8, Issue-6S4, April 2019)
- Forest 0: Digitalization of forest using the Internet of Things (IoT), Rajesh Singh , Anita Gehlot , Shaik Vaseem Akram , Amit Kumar Thakur ,⇑ , Dharam Buddhi , Prabin Kumar Das (Journal of King Saud University – Computer and Information Sciences)
- In-Vivo Vibroacoustic Surveillance of Trees in the Context of the IoT, Ilyas Potamitis, Iraklis Rigakis, Nicolaos-Alexandros Tatlas, Stelios Potirakis
- An IoT Tree Health Indexing Method Using Heterogeneous Neural Network, CHUNG KIT WU), KIM FUNG TSANG, YUCHENG LIU1 , HAO WANG1, HONGXU ZHU1, CHEON HOI KOO, WAI HIN WAN,, AND YANG WEI1
- Where’s The Bear?- Automating Wildlife Image Processing Using IoT and Edge Cloud Systems, Andy Rosales Elias, Nevena Golubovic, Chandra Krintz, Rich Wolski (Computer Science Dept of California, Santa Barbara, California )
7-IoT Beehives and Open Data to Gauge Urban Biodiversity, Gerard Schouten, Mirella Sangiovanni, Willem-Jan van den Heuvel
8-Emerging Technologies to Conserve Biodiversity, Stuart L. Pimm, Sky Alibhai ,Richard Bergl, Alex Dehgan, Chandra Giri, Zoë Jewell, Lucas Joppa, Roland Kays, Scott Loarie
9- Internet of Things for Efficient Wildlife Conservation: Challenges and Opportunities, Charith Perera (PI), Omer Rana, Benoît Goossens, Pablo Orozco-ter Wengel
10-Carsten Rhod Gregersen, “How IoT protects animals, reveals behaviours on World Wildlife Day”, March 08, 2021