Crowdsourcing support and solutions from rural communities with collective intelligence

The tradition of using fire to clear land for agriculture has been used since humankind first gave up hunting and gathering to adopt a more settled existence around 12,000 years ago. It is not only the cheapest and fastest way to dispose of waste and to clear the land, it is also a traditional way of socializing in small groups that goes back to the ancient tradition of humans gathering around the fire.  Back then, however, the practice was practical as human settlements were small and scattered, and the landscape generally greener than it is now.

Today this practice has serious implications for our environment and our health. Each year, the total number of fires differs due to weather, people actually reporting the fire and so, the State Emergency Service of Ukraine reports  between 36,000-56,000 fires in open ecosystems burning across the country. These fires, even if started with good intentions, pollute the air, destroy entire ecosystems and cause untold economic hardship. In 2019, these fires reportedly caused 84 deaths in addition to economic losses of about UAH 3.4 billion (US$ 120.6 million).

So far this year the situation has not improved. Entire villages have been razed to the ground, and thousands of hectares of forest lost to forest fires.  Fires also have occurred in the Chornobyl exclusion zone, raising concerns of radioactive particles escaping into the atmosphere. Furthermore, and of particular concern, wildfire smoke irritates the lungs, causes inflammation, affects the immune system, and makes people more prone to lung infections, including SARS-CoV-2, the virus that cause COVID-19 [more info on that here].

Figure 1: The problem of open burning is not unique to Ukraine. Many countries are working to eliminate this destructive practice.  In Ukraine, it is much more common, however, as seen in this graphic depicting the number of fires registered by satellite in April 2020. The green line shows roughly the border area.

The search for solutions

To confront this crisis, the United Nations Development Programme, through its Kyiv-based Accelerator Lab, is exploring innovative and modern solutions to this perennial problem, while engaging with local communities for help. It has conducted several mapping exercises around the country to determine the extent and nature of the challenge. This work includes the technological and legal aspects, but also puts a lot of emphasize on the need for human behaviour change. The following chart depicts the problems space as we defined it through our research.

Figure 2: Problem Space

Having a good understanding of the complexity of the problem, we then joined the Collective Intelligence Design Studio (a global learning programme to promote application of collective intelligence to solve waste management issues (designed specifically for the UNDP Accelerator Labs by NESTA). Together with 12 other Labs from around the world, we embarked on a quest to use a combination of data, technology, people’s knowledge and imagination to discover solutions and design appropriate and targeted programmatic interventions for Ukraine. In this process, we involved more than 200 representatives of local communities and developed six GIS-analysis instruments.

Our “quest” was two-pronged: 1) for the Lab to learn and apply this collective intelligence tools on the go and; 2) to seek new knowledge about waste management using novel types of data, tracking burning in real time and combining this technology with solutions proposed by the communities.

To create new knowledge that could address the open-burning problem in Ukraine we experimented, tested and prototyped scalable solutions. To achieve this, we established a partnership with Center for Innovations Development, a Ukraine-based non-governmental organization that brought GIS expertise, experience of working with communities, and a sound understanding of collective intelligence.  Through this close collaboration we produced, in several rounds of brainstorming and , workshops, the “Don’t Burn – compost” project.

The project design is experimental in its nature and was meant to provide deeper understanding of the problem and possible solutions. The key questions we sought to address in practice through the project included:

• Data empowerment: can we empower citizens and communities with near real-time satellite data for better action against practices of open burning?
• Crowd-facilitation: can a digital collaboration process facilitate emergence of deeper knowledge about the problem and surface new solutions?  
• Crowd-mapping: can visualization of composting sites all around Ukraine help us to understand how to promote the practice of composting waste as an alternative to open burning?

This process of Collective intelligence involved applying a cognitive process involving big data to large groups. This data is then analyzed computationally to reveal patterns, trends, and associations, especially relating to human behaviour and interactions.

Collective intelligence in practical terms is related to participatory decision-making, which puts a lot of emphasis on involving groups of stakeholders in the processes through public consultations, strategic sessions, research and implementation. Until recently, this process was limited to the amount of stakeholder with whom you could engage in the “participatory” process. Now it is possible to involve thousands of credible opinions, empowering their action and helping them to co-create better solutions to development challenges. That is the power of collective intelligence.

A few examples of the successful use of collective intelligence include:

Farm hack - Community of farmers building and modifying tools together

iNaturalist – computer vision models empower citizen scientists to collect data on species, their distributions, and the risks posed to their survival

London Air Quality Monitoring - Primary school pupils in London carry backpacks with air quality sensors on their journey to and from school to help monitor the levels of toxic air

More examples available on this Trello board prepared by NESTA.

To design our project, we relied to a great extent on the Collective intelligence design playbook   - a guide prepared in a participatory way by CI practitioners. This manual is filled with templates, instructions, playcards, frameworks and canvases.  Our process began with a deep dive into the collective intelligence concept started at a workshop in Istanbul (February 2020) where we refined our knowledge, and sketched the framework of our future project. Most importantly, we met other teams from countries that were addressing similar challenges. We learned that, for example, colleagues in Lao PDR our colleagues are also working on open burning of waste. We consulted and advised each other, giving additional boost to our work. The UNDP Country Office in Lao PDR recently published a two-blog series on their experiences with using collective intelligence to understand the problem of burning : 1) The Journey: Collective Intelligence (CI) to understand open burning; and 2) Why do we need Collective Intelligence (CI) for solid waste management?

Taking a bird’s eye view with satellite imagery

To help assess the extent of the burning, and to communicate the urgency to local communities, we drew upon available satellite sensor data.  The satellites circle the earth approximately every 3-4 hours to register heat anomalies. Some people do not consider fires as emergency and sometimes it is not reported to the state emergency service, hence evading official statistics.  Satellites also register the exact location and times of the fires unlike reported sources of data.  

Best of all, the satellite data is freely available. Of course, satellites have limitations – a thick cloud will not allow capturing the fires, some small fires will also not be registered. But they perfectly serve our purpose by providing insights on the patterns of burning in terms of time and place.

We explained to all participants in the collective intelligence exercise how to subscribe to live fire alerts in their community and receive them directly to their smartphones (service available from NASA). We taught them how to use the GIS analysis solution that was developed by the Center for Innovations Development for this purpose, visualized the patterns of fires for previous five years.

We then asked participating communities to draw some conclusions using provided GIS data analysis tool relevant for a specific community.  For the sake of scientific integrity, this data cannot be extrapolated for whole country. Nevertheless, here are some of such conclusions that explain the value from such analytics:

The community of Zasulska determined that more than 50 percent of fires occur in corn fields. GIS analysis also allowed them to understand the dynamic of the problem: number of fires in the community was twice that of the previous year.

The community of Torchynska discovered that in 2019 more fires were observed on agricultural fields, while in 2020 (for 5 months) more than 80 percent fires were on grass-fields. They were also able to evaluate the vast areas that were on fire – for a small community of 10,000 residents, there were fires on fields totaling 2600 hectares (note: this does not mean that all 2600 hectares burned down, this rather informs about the possible risks).

To further bolster our efforts and to give the community participants a bird’s eye view on the problem, we reached out to the Emergency Management Service of Copernicus (European Space Agency) and received the geo-spatial analysis of wildfires hazards in the central Ukraine. Their analysis provided a clear picture of which type of land cover are under the highest risk of fire. Available calculations made by the Copernicus show that more than 66 percent of analyzed area has a very high fire hazard score. Stunningly, 70 percent of the population and almost 90 percent of the tangible assets (e.g. settlements) are exposed to high and very high fire hazards. Analysis confirms the anthropogenic nature of wildfires – which start as a “controlled” cleaning of an agricultural field or a hayfield and later can escalate into uncontrolled blaze.

Figure 3: Distribution of fire hazard level per land cover, EMSN-075Forest fires preparedness, Ukraine, ESA

So what did collective intelligence teach us in Ukraine?

Collective intelligence is about becoming smarter together. It combines the power of people, data and technology to solve problems, but it is not an easy or straightforward task. I’d like to share a little about the questions we wanted to answer and our experience, which might be useful for those who want to give CI a try.

It is always best to work with the communities that are most affected by the problem. These people experience negative effects (in our case hazards of open burning of waste), they often tried to solve the problem on their own and have the heuristic knowledge of what solutions might work. Since disposing of organic waste with fire is common for the whole country, we involved more than 200 representatives of local communities from all around Ukraine, volunteers and experts. We engaged with them in more than 10 webinars to go deeper in group thinking and to capture their knowledge and experience.  As part of their follow-up, project participants coordinated actions to inform their communities about the perils of burning waste, analyzed the patterns of fires and integrated this knowledge into decision making in their communities. 

Figure 4: A snapshot of a digital brainstorming board with project participants, Accelerator Lab Ukraine and the NGO “CID”

We discovered that overall, most people are not aware about the health hazards of smoke from burning organic waste and the risks to personal property of widespread fires.  Most public awareness campaigns to date have only focused on the fines or punishment.  Future campaigns should be targeted to raise awareness of the widespread dangers and personal risks that come with the practice of open burning.

There is, however, a high degree of locally-organized campaigns by those people who were aware of the dangers of open burning.  These people created chats, organized volunteer patrols, and installed air quality sensors to facilitate collective action against the practice.

In those areas where people were motivated to make a change, we found that oftentimes the case was a lack of reliable services to support a change in behaviour. In many communities, for example, waste collection services do not cover “green” household waste, and thus this is the first step that a community could make to reduce open burning.

Barriers to composting at scale

Most participants in the collective intelligence exercise embraced more effective composting as a solution, but confess to a lack of understand on how to do it.  Others confided they did not want to get one’s hands dirty, or create a "smelly dump" in the yard. Nevertheless, we have managed to find a considerable number of home composting grounds in the country, which shows that such solutions are spreading the fastest among neighbors.

For the smaller communities, it is not economically viable to create a centralized composting site, hence priority сould be given to distributed composters (private or collective), small biogas production units etc.

Enforcing something that contradicts tradition is a not an easy task. Disposing of fallen leaves, crop residue and other organic waste is a practice that spans through generations. This is socially acceptable behaviour and those officials that can issue fines are often members of that community. They need to provide support and feel that the community supports them.

To address the challenge of assessing current composting capacities and needs, we chose to facilitate connection between those who have organic waste (food waste, crop residues) and those who are already composting and willing to take more composting material, or those interested in getting organic fertilizers.  We then mapped out the location of these composting 

Figure 5: Snapshot of composters map, Accelerator Lab Ukraine and NGO “CID”

sites to determine their accessibility. We were interested in all types of composters: for private gardening, home, community and municipal, those that are part of agro-production, etc. So far we have recorded 367 composters.

Although the map is far from covering whole Ukraine at the moment, there are some hints about the composting ecosystem that we draw already:

Out of all mapped composters, only 19 are open to the public, a person can bring organic waste and it will be processed there. Nearly half of them are privately owned, which means that some people are willing to open their own composters to the members of community. This is an inspiring finding that speaks about the potential to create shared composters, one per few households.

Out of those who shared the information 7percent percent (11 composters, all type of ownership: private, municipal, state) are interested in sharing/supplying the compost for needs of others. And only few of map composters are actively looking for waste to produce compost. Although the data here is anecdotal, I still believe that there is a potential in connecting those who have and those who need organic waste.

Next steps

The practice of setting fires is a centuries-old tradition in countries around the world. It is thus important to avoid stigmatizing household owners.  Blaming them for just doing what they and they ancestors have already done risks further excluding them from public dialogue – and we need to engage them in any solutions, for those solutions to be viable and sustained. We need to work with them to discover, develop and implement real alternatives to burning.  Furthermore, they need to “get” the value of this change and see the value this change will bring to their communities.

The Accelerator Lab brings to the ecosystem new knowledge, finding solutions, insights or leverage points that were not known before. Although our interventions are short term and small in terms of resources, they are prototypes of a large-scale programmatic response of UNDP and can be scaled to achieve positive impact on the state level (and beyond).

In this case, in just a few months, we were able to create scalable GIS solutions that can be further introduced into the Disaster Risk Management projects, sustainable agriculture and overall support policy making for better air quality, and better security of communities.  Our plan is to upscale this effort with the support of partners to a level where it can have a lasting impact and bring long-term benefits to communities across the country.