Krutika Ravishankar is the co-founder and executive director at Farmers for Forests (F4F), a non-profit that works on protecting and increasing India’s biodiverse tree and forest cover in collaboration with communities vulnerable to climate change.
Krutika has over a decade of experience working in rural India in the agriculture, forestry, food security and public health sectors. She has had stints with the World Bank and the Abdul Latif Jameel Poverty Action Lab (J-PAL).
An Elevate prize winner, an Ashoka fellow and a Mulago fellow, she has a bachelor’s degree in economics and environmental studies from Mount Holyoke College, USA.
Krutika spoke to indianexpress.com on agroforestry, forest protection and restoration, use of technology, the importance of drones and the increasing use of AI in forest monitoring. Edited excerpts:
Venkatesh Kannaiah: Tell us about Farmers for Forests, its focus and the journey.
Krutika Ravishankar: Farmers for Forests was founded in 2019 with the idea that for environmental action to be sustainable in the long run, the communities or farmers involved should be financially remunerated.
We have an agroforestry programme and a sustainable forestry programme. Under our forest protection and restoration programme, we work with 20,000 households directly, and across five lakh acres of land. And for our agroforestry programme, we work with 6,000 farmers spread around 7,000 acres of land.
In our agroforestry model, farmers grow two main fruit trees, some shrub trees specifically for pollinators, a ground-cover crop like black gram, harbhara, and grow timber trees on the boundary.
In this agroforestry model, farmers start getting money from the first three to four months and see a significant increase from the fourth or fifth year onwards, once the trees start bearing fruit. Farmers can earn around Rs 2 lakh to Rs 2.5 lakh per acre, which at times can go up to Rs 6 lakh per acre.
In the sustainable forestry programme, we work with communities to reduce the incidence of forest fires, remove invasives in the forest, increase biodiversity, and prevent unnecessary logging of trees.
The forest protection programme is primarily in eastern Maharashtra, specifically in Gadchiroli, and our agroforestry programme is in Gujarat, Maharashtra, Madhya Pradesh, and Jharkhand.
Venkatesh Kannaiah: Tell us about how your agroforestry model works on the ground.
Krutika Ravishankar: We provide farmers with high-quality saplings, drip irrigation systems, and organic fertiliser material for mulching to improve soil fertility and retain moisture.
A portion of the total cost is borne by the farmer while the rest is covered by philanthropy and carbon finance. Once the farmer has planted the trees and the irrigation systems are in place, we fly drones over the land. The drones collect data on the number of trees, their height, species, and overall health. It also tells us how much carbon each tree is removing over time.
The drone imagery is then fed into an AI model, which can even measure how many metres of drip irrigation pipe have been installed. If there are invasive species on the land, it can detect them and identify their location. So, if farmers dig pits before planting trees, the AI model can identify and count the number of pits.
Venkatesh Kannaiah: How important is tech in your overall work?
Krutika Ravishankar: In the long run, our model is funded based on how much carbon the trees are removing. So, we need to have accurate data.
When we first started collecting data, we were doing it manually. We would take a measuring tape or put a QR code on a tree, and then someone would go and scan every QR code and take a picture of every tree. And that was fine till we had around 5,000 trees in the programme. Now, we are monitoring more than a million trees.
Without tech, it would just be impossible to measure it accurately.
Venkatesh Kannaiah: Tell us about your data collection using tech, and its impact.
Krutika Ravishankar: We collect multiple types of data from farmers. This includes the location of their land, the activities they undertake on it, their current income levels and asset ownership, soil health, and their dreams, hopes, or aspirations for the land.
For each farm or farmer, we have close to 200 or more data parameters or variables. This data is not specific to agri inputs; it also delves into social and economic metrics.
We collect this data using a mobile phone app. We also look at satellite imagery for that farmer’s land.
We try to understand what has happened on the land over the past 10 years, including land degradation, nearby water sources, current baseline levels of the soil, and farmers’ income levels to help us plan the right intervention. We then decide on the tree species to be planted, the market linkages, and how many different types of species should be planted.
The drone part comes after the trees are planted. We fly drones to assess how the trees are doing and what the actual intervention looks like., This data can help us secure carbon-offset funding. For example, drone data for a single acre of land might have around 12,000 images taken from many different angles.
The challenge is that most of the people funding carbon offsets are sitting thousands of miles away, in the US or Europe. For them, it is important to know what is happening on the ground. Drones and AI have been a great way to provide that information.
Under the carbon-offset model we work with, corporate buyers pay us upfront so that we can deliver services to farmers that will lead to carbon capture in the future because the trees remove carbon over a period of 40 years. We also work with independent verification bodies and auditors who check the data.
In our forest protection work in eastern Maharashtra, we also use drones to fight forest fires. While forest fires are sometimes considered beneficial, the excessive dryness in the region causes multiple fires, and they get out of control.
As a pilot project, we have deployed drones in around 10,000 acres of land to detect these fires, identify their exact location and the intensity, and help us send patrol teams to put them out. We were able to reduce the intensity and incidence of forest fires by nearly 50- 60%.
Venkatesh Kannaiah: Tell us about your dashboards for impact and forest protection.
Krutika Ravishankar: Using all the data we collect, we have built multiple dashboards for various stakeholders to help them make informed funding decisions.
Our software stack is open source. When we started in 2019, there was nothing available in the market that met our needs, so we developed the software and tools in-house. Now, we recognise that these tools could be useful to others who are working in this space.
There is our TreeLens dashboard, which allows anyone to upload a drone image of a tree, and it will analyse it and share the results. It is online and free to use. We have also open-sourced all of our code and made it available on GitHub.
Venkatesh Kannaiah: In the areas that you work in, tell us about technologies globally that are changing the nature of your work.
Krutika Ravishankar: The most impactful tech intervention has been the use of high-resolution satellite imagery to understand deforestation or forest fires. It has transformed the way we plan our interventions and measure impact.
There is drone tech that has transformed the way data is collected, and there are digital data collection tools on mobile. Using mobile phones, our team members can map the boundaries of the land using geofencing.
The last one is bioacoustics. We’ve been recording the sounds of birds in our areas to understand what kind of biodiversity these land parcels are supporting
Venkatesh Kannaiah: How is AI changing the nature of your work?
Krutika Ravishankar: I would say the biggest help has been in analysing millions of trees. If not for AI, we would have been analysing all of these trees manually or analysing the photographs manually.
AI has made it possible for us to tell farmers the exact value of natural assets on their land. We are also able to predict how those trees are likely to grow and what that farmer’s income could look like.
Venkatesh Kannaiah: What are your challenges?
Krutika Ravishankar: Our biggest challenge is the operational complexity due to logistics and procurement. Then you also have to deal with local political influence.
This year, in June, there was no rainfall at all in our areas of operation. So, it was quite a challenge.
There is also the challenge with government land, from a carbon credits angle. All of our forestry programmes happen on government land. The Government of India has not yet clarified if it will own the carbon credits that are generated on government land. There is also a lack of clarity on whether the carbon credits can be sold. Hence, we have not done any carbon offsets on such land.
Venkatesh Kannaiah: Tell us about the carbon offset space and the ecosystem in India.
Krutika Ravishankar: In India, the carbon offset space is primarily dominated by for-profits. I wish there were more non-profits, as they could help create public and digital infrastructure that the entire sector can use and benefit from. There are some non-profit pioneers in this space, but it is not enough.
