Image credit and thanks: Nine Koepfer
“With AI and machine learning we can start to put together the signature of healthy and unhealthy hives. The holy grail would be to identify early warning indicators of problems.”
Simon G. Potts, Professor of Biodiversity, University of Reading, BBC Radio 4
Click each question below for more information:
1. What is the origin of The World Bee Project?
“In 2012 millions of people concerned about the declines in global bee populations backed by EFSA were campaigning for a ban on widely used toxic pesticides that were threatening bees and the entire food chain. Around the same time, I noticed that beekeepers were beginning to use commercial hive sensors to manage their hives.
The two events merged in my mind – I saw that the sensors were excellent news for hive management but also saw that they weren’t going to solve the global pollinator crisis. It occurred to me that to improve pollinator health, sensors – enhanced by AI and emerging technologies – could be used to collect bee and environmental data across varying global regions. I felt that to help solve the global crisis we – all of us – had to take a global view and I must– as Carl Sagan said – also take my stand as an individual. That is why I founded The World Bee Project in 2014.”
2. Did you have any mentors at that time?
“No, I didn’t have any mentors, but I am grateful to be blessed with imagination. I find all the mentoring I ever need in books, in poetry and myth, in the arts, in physics and technology. I am continuously mentored by what I see – the spirituality, beauty and power of Earth as a superorganism, as a single dynamic, self-regulating, inter-connected, inter-related system with the human species as one of its many wondrous species.”
3. What drives you?
“I care about justice. Environmental. Social. Economic. This is why I designed The World Bee Project to merge social, environmental and economic benefits. Our World Hive Network© collects global bee health data and in parallel we support smallholder farmers and beekeepers with our World Bee Education© programmes. Of course, we cannot do any of this alone. People, communities, the private sector, and local and national governments all need to participate in improving the health of bees and other pollinators because all of us depend on pollinators for our own wellbeing.”
4. Why is the World Hive Network a unique innovation?
The World Bee Project is unique in multiple ways.
- The World Hive Network© is an Artificial Intelligence (AI), Internet of Things (IoT) and Data Analytics driven global monitoring programme generating fresh insights and new evidence to help find solutions to global pollination decline, food insecurity and smallholder farmer/beekeeper poverty.
- We take an integral approach by looking at pollination in the context of the Earth as one system of interconnected elements with every single social and economic system embedded in nature’s system.
- We take a socially based approach from smallholder farmers right up the global food system. Our work provides opportunities for social connection, identity and entrepreneurship, particularly in challenged and more excluded areas and communities.
- We are creating an opening to a new community of Data Science analysts and academics who find ecological data interesting and important. For example, in 2020 at the Data Science Festival (5000+ members worldwide) ran hackathons at Oracle offices using our World Hive Network bee data.
5. What impact is the World Bee Project aiming for?
The World Bee Project is using IoT and AI and Big Data driven technology to create a dynamic knowledge base of pollination services and agricultural biodiversity in different regions. The World Hive Network© database can expand to generate millions of rows of data points every day. This width and depth of insight can arm communities and governments with the knowledge to determine strategies that can improve pollinator health and create the sustainable ecosystems that all life depends on.
Our innovation can play a significant role in enabling researchers, scientists and governments to better mitigate the threats to food security, smallholder farmer and beekeeper livelihoods and to create predictive models to ameliorate potential future stresses.
We are bringing this data to life with analytics and visualisation fuelled by AI, Machine Learning and Natural Language Processing. This critical area of scientific research is also helping businesses to gain new insights into their own data too.
6. What is the greatest value The World Bee Project can bring?
Better global understanding of bee and pollinator health and its relationship with its local environment is a value in itself. The greatest potential for benefit from the work of The World Bee Project CIC is the value of collective intelligence that the World Hive Network© will increasingly bring with the expansion of its network. It is only when the data from hives in different environments in regions across continents can be captured securely and integrated with third-party data – and made freely available to beekeepers, scientists and researchers around the world – that the world will be ready to make valuable advances in understanding the relationship between the health of bees and the health of our environments.
7. Bees have been managing well without technology – why is technology now an important aspect of saving the bees?
The devastating truth is that right now the bees are not managing well. We are seeing a growing number of pollinator species worldwide that are being driven toward extinction by diverse pressures, many of them human-made. Scientists report that bee and pollinator declines are caused by poorly managed human activity, including intensive agriculture, destruction of flowering habitats, widespread use of toxic pesticides and changing climate patterns.
Using Artificial Intelligence (AI) and remote monitoring technology allows us to listen to and understand how honeybees are interacting with each other, how they behave in varying environments and we can better understand the factors affecting their health – these factors vary from one environment to the next which is why wide-scale monitoring is a critical need.
By closely monitoring and detecting patterns in bee behaviour we can learn to protect honey bees and wild bees, and ultimately protect our planet and ourselves.
8. How is The World Bee Project using Artificial Intelligence?
The World Hive Network© uses systems driven by Artificial Intelligence to interpret the bee health data generated by remotely monitored bee hives in the World Hive Network. This is why we can offer researchers and stakeholders real time information on the impact of factors such as agricultural practices, use of pesticides, habitat loss, diseases, and parasites on the health of bees.
We carried out a number of proofs of concept with Oracle to discover what can be done using technology with embedded Artificial Intelligence and Machine Learning. What we learned is that when we overlay freely available data such as pollution, habitat and farming practices to the data from our World Hive Network and apply Artificial Intelligence and Machine Learning we find new associations and insights that aren’t readily apparent in the baseline data.
9. What is so smart about The World Bee Project Hive Network?
The real “smart” bit is what happens when those monitored hives are connected to the World Hive Network©, which is the world’s first globally coordinated honeybee monitoring initiative. By pulling all the information that is coming from hives into one big, global database i.e. The World Hive Network©, we can generate new insights about bee health and its relationship with forage, weather patterns, diseases, parasites, predator species and pesticides. The goal is for beekeepers, farmers, researchers, governments and stakeholders to be able to access a source of data and begin to work together to help protect our bees and our planet.
10. Where do sensors and data come into play?
A single monitored hive can generate a million data points per day!
There is inherent value in the data that is currently gathered from the hives. For example, bees are extremely good at maintaining the temperature of the hive around the brood frame during the months in the year where they are foraging for nectar and pollen and while the queen is actively laying eggs. They maintain this temperature at around 34.5 degrees regardless of the temperature outside of the hive. If the sensors pick up a significant change in this temperature that cannot be explained easily, it can mean that there is a problem that needs investigating. Analytics can be used to recognise this change in temperature and alert beekeepers and other stakeholders.
Another example is the monitoring of acoustic data to predict when a bee colony might swarm. A swarm is a natural occurrence that happens when the bees have outgrown their hive and involves roughly two-thirds of the bees leaving with their queen to start
a new colony, leaving a new queen to support the existing one. It can be a problem, though, if the bees swarm at the wrong time of the year. Even if they do swarm at the right time of the year, a beekeeper risks losing two-thirds of the bees from the colony in the hive. Experienced beekeepers are very good at manually monitoring their hives and recognising when they are becoming over-crowded. Unpredicted swarms do still occur, however, and local beekeeper associations are called to deal with those large formations of bees settled on trees and branches. The World Hive Network© system uses its acoustic sensors and analytics to predict when bees might swarm. The bees’ sound changes in frequency as they get ready to swarm and by recognising that change in the sound, a possible swarm can be predicted up to three weeks in advance. With Oracle Stream Analytics we have explored ways of enhancing this and looked at new ways to alert beekeepers and supporting organisations.
The World Hive Network© acoustic sensors can also detect hornets, which are a huge threat to bee populations. The sound from the wing of a hornet is different from that of bees because the wings flap at a different frequency and the sensors can pick this up and alert beekeepers to the hornet threat. Oracle has used video analysis techniques running on Oracle’s High-Performance Computing to recognise hives coming under attack from Asian Hornets. While this trial was technically successful, it remains impractical as a solution at this point as the cost of a high definition camera focused on a single hive would currently be prohibitive.
11. What other Oracle technologies are involved? (Like Oracle Cloud tech?)
The areas The World Bee Project has explored with Oracle include the following:
- To initially replicate what TWBP can already do with their existing interface but to demonstrate how the data can be further visualised and analysed using Oracle Analytics Cloud (OAC).
- To further transform data in real time to provide alerts when certain patterns are recognised using Oracle Stream Analytics (e.g. potential swarm alerts).
- A voice controlled chatbot using Oracle Digital Assistant that can guide Beekeepers through a hive inspection to automatically collect and store the data for additional monitoring and analysis.
- We have also explored the use of Oracle Data Science with Oracle High Performance Computing and GPU for image and video analysis for use in monitoring biodiverse farming practices. We did this by analysing photographs of fields to monitor if there is above a certain percentage of plants that are pollinator friendly and flower throughout the summer months or even just if fields are surrounded by a hedge rather than a fence.
- Similar techniques and technologies were also used to analyse high definition images of honey samples to identify the plant from which the nectar came to make the honey. This information could be used to support TWBP’s development of a “Bee Mark” ecolabel to prove that honey on sale in shops has come from a sustainable source.
- Oracle has also briefly looked at the use of Oracle Blockchain to securely share this monitored data between different parties in a supply chain to facilitate any potential certification.
- While both the above have shown what is technically possible, any production solution will depend on an evidence-based standard of certification to be agreed and adopted.
As previously mentioned, these innovations make up a kit bag of capabilities that The World Bee Project can and is applying in specific programmes.
12. What does The World Bee Project offer to technology and data companies?
The World Bee Project’s innovation of the World Hive Network© is important for data companies because we offer:
- A model for a range of services that customers can empathise with.
- A rare opportunity to showcase the entire breadth of any technology platform, including IoT, block chain, and AI.
- A chance to talk about a real-time, self-driving, intelligent data network which can truly help to save the planet.
- A framework that defines any corporate narrative of Doing Business to Do Good.
- An example for corporate social purpose, particularly in developing and frontier markets.
- A platform to shift perceptions on brand for current customers and prospective customers.
13. Why is pollination vital to life?
Pollinators are economically, socially and culturally important. There are biological, moral and cultural arguments for pollinator conservation – we need to protect pollinators to protect wild plant diversity, pollinator diversity and resilience, and the intrinsic value of pollinators and plant-pollinator relationships.
In 2016, the Thematic Assessment of Pollinators, Pollination and Food Production, the first-ever issued by IPBES, estimated:
- 20,000 – Number of species of wild bees. There are also some species of butterflies, moths, wasps, beetles, birds, bats and other vertebrates that contribute to pollination.
- 75% – Percentage of the world’s food crops that depend at least in part on pollination.
- US$235 billion–US$577 billion – Annual value of global crops directly affected by pollinators.
- 300% — Increase in volume of agricultural production dependent on animal pollination in the past 50 years.
- Almost 90% — Percentage of wild flowering plants that depend to some extent on animal pollination.
- 1.6 million tonnes – Annual honey production from the western honeybee.
- 16.5% — Percentage of vertebrate pollinators threatened with extinction globally.
- +40% – Percentage of invertebrate pollinator species – particularly bees and butterflies – facing extinction.
- Between US$235 billion and US$577 billion worth of annual global food production relies on direct contributions by pollinators.
- Chocolate, for example, is derived from cacao tree seed (annual world cocoa bean crop value, US$5.7 billion. Cecidomyiid and ceratopogonid midges are essential for its pollination.
- The volume of agricultural production dependent on animal pollination has increased by 300 per cent during the past 50 years, but pollinator-dependent crops show lower growth and stability in yield than crops that do not depend on pollinators.
- Nearly 90 per cent of all wild flowering plants depend at least to some extent on animal pollination.
- In addition to food crops, pollinators contribute to crops that provide biofuels (e.g. canola and palm oils), fibres (e.g cotton), medicines, forage for livestock, and construction materials. Some species also provide materials such as beeswax for candles and musical instruments, and arts and crafts.
- Pollinators, especially bees, have also played a role throughout human history as inspirations for art, music, religion and technology. Additionally, they improve quality of life, globally significant heritage sites and practices, symbols of identify, aesthetically significant landscapes. Sacred passages about bees occur in all major world religions.