Creating first smallholder farmer's digital identity with NMKR and Cardano Foundation

Overview

We have been working with the Cardano Foundation over the last couple of years and this has been a long time in the making but we're very excited about this project as it's close to our hearts and truly believe this project and solution can have long-lasting impact in the smallholder farming market.

Here at DigiFarm we have always been passionate about field boundaries, as it's often the "enabler" for not only solving some of the challenges in agriculture including lack of digitisation and fragmentation but also for being able to break down barriers for farmers to access precision ag services and technologies as well as also giving some power back to the people who deserve and need it the most, farmers.

We were first introduced to the Cardano Foundation in 2022 and learned about the impact the technology could have through decentralisation and building an ecosystem around digital identities, so this got us excited of course, thinking of field boundaries and agriculture.

Furthermore, this lead to us partnering up with NMKR to tokenise and create the very first NFT for the individual field boundaries and farmers in Kenya and Tanzania on a small dataset for piloting. This will enable wide-scale adoption, exposure and an entire eco-system available and open to the Cardano community to build upon and provide additional value, creating the very first digital identity for smallholder farmers, in the world, including over 40+ years of in-field crop production (biomass/yield) to support farmers achieve better terms on credit lines, insurance premiums, agronomical advice and better decisions in their quest to become financially viable.

DigiFarm will provide the most fundamental data-layer in digital agriculture, i.e. field boundary and planted area, historical and in-season crop-yield (productivity) data, on the ledger, which will drive the ecosystem for building services and solutions directly to both the farmers and agri-value chain stakeholders, such as insurance companies, financial lending institutions, crop-input suppliers (fertiliser, seeds, crop protection), grain-mill, ag-traders, governments among others. The solution will be through a paid-version (to support the cost of creating + delivering the data continuously) but the output data will have open-license, i.e. creating derivate products and service layers on top, which is where the value will be derived.

Why are we doing this?

Smallholder farmers face significant challenges worldwide, including improper infrastructure, lack of funding, gender biases and economic difficulties. The broader agricultural value chain system has been developed in a way that the farmer has been marginalised and left with no real options of escaping economic hardship. This project will revolutionise how we can help smallholder farmers own their own data, independently from the supply chain, and use this data to negotiate better financing, crop-input supply and insurance (among other) terms, this will give farmers more control, autonomy, power and control and be able to leverage this towards the broader system operators.

Additionally, one of the core reasons why this has not been done before is that still 83% of the world's agricultural field boundaries are smaller than 2 hectares, i.e. traditional publicly available Satellite data does not have sufficient spatial resolution to be able to detect these field boundaries, i.e. the alternative is commercial grade SatEO data, but the problem is that this data is very expensive and typically in these regions, i.e. APAC and Africa there is no budget for being able to justify the cost of this imagery, even if there is a project which is being funded by NGO or non-profit the struggle always seems to be able to sustain a viable business model post-project phase into a commercial setting, this is why we believe we have the correct tools in the toolbox as we leverage the highest publicly freely available SatEO data source in the world, i.e. Sentinel-2 from European Space Agency (ESA) which is 10 meter per pixel resolution across 12 spectral bands and through the wonderful technology and ability to super-resolve this data from 10 meter to 1 meter, we're able to visually see and hence then automatically detect the field boundaries and associated production area, i.e. in terms of vegetation and biomass, and to be able to sustain the post-project phase with a commercial offering in the region.

An example can be seen above which highlights the AOI in Kenya, where 10 meter per pixel resolution from Sentinel-2 is visible on the left and deeply-resolved Sentinel-2 on the right at 1 meter per pixel resolution

What will the solution be?

The solution will be a digital ledger and unique ID for every single agricultural field boundary and it's "crop-performance" of those fields historically over a 40 year period by using Satellite data as well as compared to other fields in the region, the ability to compare performance and yield over time is the most important base-line for optimising production, the ability to make this public and digital, will empower the entire value chain agricultural value chain. The solution will cover the extended areas we're building on from Fund 11, i.e. in Kenya and Tanzania, this will be the very first MVP and prototype of the unique technology.

The reason for addressing the challenge and building this solution comes from extensive experience in the agricultural industry, farmers, who are individuals are being taken advantage of by larger corporate organisations, I believe the only way to solve this is to take back control, and this starts with data and digital identity which creates transparency, i.e. farmers would be able to negotiate contracts, prices when there is complete transparency of what other farmers are being offered by these suppliers. The ability to share data among farmers and suppliers in the value chain, whether it’s for micro-financing, insurance, crop input such as buying seeds, fungicide, herbicide, fertiliser or real estate transactions and valuations, farmers desperately need more (a) transparent pricing from suppliers (b) control and bargaining power (c) more choice to buy crop-input products and services as well as more options for selling their crops (d) additional revenue streams: typically and currently larger agribusinesses are the ones benefiting from sustainable initiatives and ESG reporting while leaving farmers behind, this project and solution will completely change this and give the power back to the ones who need and deserve it, the farmers. The solution will engage the last mile, i.e. smallholder farmers in Kenya and Tanzania, across the entire 60 million hectares of cropland area. The solution will also engage with the agricultural industry value chain, i.e. lending facilities, insurance companies, grain traders etc. to be able to negotiate better conditions and terms from various parts of the value chain, the ability to cooperate with not only farmers directly but creating a decentralised "marketplace" for farmers and the entire value chain to negotiate and interact will be critical for the future empowerment of farmers.

Project timeline and milestones

We kicked off the project in October (2024) and have already managed to reach some key milestones with NMKR including the tokenisation of field boundaries and associated meta-data into NFTs, enabling the very first tests to go live and in production, we've included a draft timeline below so we can keep reporting on our progress as we're advancing through the project.

Project Milestones

We have also included a break down of the workflow and logic surrounding the tokenisation of the field boundaries, initially we had some challenges related to what data we could include in the NFT and CSV file, as some data is raster (GIS-based such as productivity zones, sustainability index, crop classification etc) and based on text such as field boundaries which we could convert to this in the CSV and eventually we managed to find a solution, and decided on the details which needed to be in the NFT, and hence the metadata itself, there were several iterations of CSV files and data layers.

Sample Field Boundaries Delineation in Kenya 2023/2024 years.

The challenge with this first version was that the fields in the CSV had more than 64 characters, which is the limit on Cardano per line, we could make an array but not in this instance as we give the JSON values in CSV, so we needed to find a solution.

We also needed to find a solution on storing the data, i.e. either in the token or in a 3rd party storage, accessible from a link shared in the token, besides the numerical columns most of the data will be larger than 64 characters, which still posed a challenge. Having this in mind we prepared another CSV file with an ID column and each PNG is named using that ID column as name, and then we would need a solution for the additional information.

The decision was made to make the “tokenname” as the ID column, i.e. to match the PNG files, all names after the ID column, and then the other columns in the CSV would be meta-data for the NFT. 

CSV file with the field boundaries and associated meta-data to be minted and tokenised

Additionally, we've included an example of the field boundaries delineated and their geometrical shapes which represents the unique vertices and digital identity of the field boundary and agricultural land, as can be seen in the screenshot below every field boundary is different, representing a unique digital identity and footprint linked back to a decentralised ledger which can be used to add additional value and data such as historical productivity, crop information, soil data, yield etc.

NFT Field Boundaries Preview

Furthermore, we then decided on the token-gating to leverage Vault3, a trusted partner of NMKR as well, and we would need unlimited amount of space (as each asset will have specific data), and here is the final output and prototype data, the minting of each asset can be found in this dataset, where we set a price of 0 ADA so we could test on mainnet, the zip files were publicly attached to the NFTs and then the next step and milestone will be the Vault3 and toke-gating. Additionally, in this project we also have minted an asset, and here is the url link:

https://pool.pm/asset1zsqw5epzclynwcyev9phfgwc8agn880709ue0n.

NFT Field Boundary minted example

———————————————————————————————

Lastly, below we can see the very first results from the field boundary delineation model in the first pilot AOI (Area of Interest) in Kenya, where the field boundaries have been delineated using super-resolved Sentinel-2 from 10 meter to 1 meter per pixel resolution and based on vegetation data (i.e biomass) which then represents the actual, correct and in-season planted area, which is the baseline for all analytics and is the backbone of any assessment and ag-derived services including crop insurance premium, microfinance lending, seed, fertiliser and crop protection input (incl. VRT).

Field boundary example across the first AOI in Kenya with delineated field boundaries and associated meta-data

Once the field boundaries and associated meta-data has been tokenised and digitised, the ability to layer in historical vegetation data, i.e. meaning biomass from previous crop seasons, all the way back to 1980s will enable the project to create a strong baseline for assessment on productivity, potential and benchmarking against other field boundaries with similar crop-types, this will provide a unique dataset and enable not only the farmer to plan optimisation but also to negotiate better terms on their financing, crop insurance for example at the same time as leveraging other revenue opportunities, specifically related to carbon market.

Field boundaries and associated vegetation data from sample dataset in Kenya

Field boundary sample data in Kenya showing the legends and coloring of field boundaries based on their productivity over time, i.e. the greener the field the higher average biomass and vegetation over a 35+ year period and the red fields also serves as a great potential for restoring less productive land.

Want to find out more and stay posted, check out the full project details here.