Please describe your proposed solution.

The solution is the ability to create technology which can automatically detect any agricultural field in the world, starting with Ethiopia, and its current productive (yield based on plant biomass) and sustainability based on historical data (up to 30+ years), e.g. whether the field has been farmed sustainably, preventing it from depleting its natural resources over time, and also creating a baseline, starting point of current biomass level (current date) to continuously on a quarterly basis monitor the sustainable development of the field going forward and providing financial incentives and monetary rewards for carbon capture (e.g. increasing natural resources) through verified carbon marketplaces.

Decentralised supply chain leads to transparency, availability and the resulting verification of all parties’ data and more sustainable farming practices. The decentralisation of data can enable access to finance and immediate payment which supports the first-mile actor financially.

Ethiopia’s economy is highly dependent on agriculture, which accounts for 40 percent of the GDP, 80 percent of exports, and an estimated 75 percent of the country's workforce. However, the population suffers a challenge where 35% of the population does not have enough nutritious food to eat.

This index will firstly enable:

1. Digitising of all field boundaries along with seeded acres across all of Ethiopia enabling smallholder farmers to claim and access digital identity of their unique agricultural fields, which is currently not publicly or freely available.

2. Enabling quickly scalable digitising of precision agricultural services and in-field analytics to smallholder farmers (through open access on mobile devices).

3. Creating baseline for assessing the sustainability and long-term productivity (based on yield/biomass) for each individual field historically over 30+ years (based on Landsat EO-data/NDVI biomass calculations) in order to assist the farmers to make decisions which will lead to improving the sustainability (increase in natural resources) over time and prevent them from running out of in-field productivity, ensuring long-term livelihood and climate-friendly agricultural practices (cover crops, reduced tillage, crop rotation etc).

4. This baseline and field-sustainability ranking will create the baseline for:

5. Enabling smallholder farmers to issue tokens on the open market for their individual parcels of land which will finance the sustainable agricultural practices (regenerative practices).

6. Corporations and individuals will be able to purchase tokens and offset their climate footprint (Co2) with the sustainable actions of the farmers.

7. Monitor the sustainability index and development over time (every quarter) to assess the effects of the activities and provide measurable and actionable results of carbon capture farming in an open and transparent method.

Screenshot attached in supporting document illustrates the current methodology and index from all the agricultural field boundaries in Hannover, Germany and the NDVI slope (vegetation) over a 35+ year period, showing decrease or increase in productivity (biomass yield) and subsequently sustainability.

Please describe how your proposed solution will address the Challenge that you have submitted it in.

Blockchain and decentralisation is key for digital transformation in agriculture and Cardano is the ideal disruptor in this industry, and is directly aligned with Charles Hoskinson’s recent tweet re speaking on this at the U.S. House of Representatives Committee (tweet link).

The solution will address the "Property Registration" section of the challenge and provide a significant pillar to help smallholder farmers in Ethiopia monetize on farming sustainable and future prosperity, in addition to holding global corporate food and beverage brands accountable for providing fair wages and working conditions to their supply network.

Especially in Ethiopia where agriculture accounts for an estimated 75% of the country’s workforce and just five percent of land is irrigated and crop yields from small farms are below regional averages, the ability to create a unique digital identity for each field boundary will create necessary autonomy, transparency and improved financial incentives for the individual farmers. The solution will enable smallholder farmers to easier access financing through applying regenerative agricultural practices and to be able to verify and validate the sustainability of their practices and

Currently, land registration and field boundaries are not publicly or freely available in Ethiopia, in combination to that the data which has been digitised (less than 30% est) are largely inaccurate and outdated.

The digital identity of individual field boundaries for smallholder farmers is crucial to their land management, ownership records, lending and microfinancing options, subsidy eligibility, taxes and for adopting digital agricultural precision agriculture services and technologies. Without a unique digital identity of each farmer's fields, the ability to increase productivity and reduce crop-input costs is not possible. This is directly aligned with the “Nation Building Apps” KPIs and objectives including: “building blocks enabling sustainable prosperity…able to provide nations with decentralised solutions to their infrastructure by eliminating a single point of failure, protecting national, corporate, and individual data, property, and assets.” including “Property Registration”.

What are the main risks that could prevent you from delivering the project successfully and please explain how you will mitigate each risk?

Type of risk: Technological

Description of the risk:

• Inability for deep neural network models to not accurately detect and predict field and seeded acres at a high accuracy in smallholder markets such as Ethiopia, with a main focus on the major crops including: teff, wheat, maize, sorghum and barley.
• Inability to access historical optical Satellite imagery data from Sentinel-2 and Landsat (up to 30 years) due to potential consistent cloud cover, especially from June to August when it’s monsoon season.
• Inability to accurately assess long-term historical productivity (yield) based on NDVI over 30 years due the lack of pixel resolution from Sentinel-2 and Landsat (10 and 30 metres).

Effect of the risk? Effect of risk includes that the field boundary detection model does not reach high enough accuracy in order to be commercially attractive to clients, i.e. not high enough accuracy compared to Cadastral map data or manual delineation in addition to not being able to capture sufficient imagery for developing a reliable sustainability index.

Mitigation methods? Solution; Our solution will be based on leveraging super-resolved Sentinel-2 at 1m per pixel from available archived data (back to 2015) and cloud-removal algorithm (for the seasons where there is not sufficient imagery from standard L2A Sentinel-2 at 10m per pixel). Additionally, this will also mitigate the risk of not having enough imagery through the season to create the sustainability index. Additionally, as we have extensive experience in delineating field boundaries in other smallholder markets including India, Vietnam, Thailand and Tanzania this will ensure the successful completion of the project.

Type of risk: Cost

Description of risk: computational complexity and costs of data processing too high for planned price-point and practical market applications, limiting uptake and market segment.

Effect of the risk: if data processing costs are too high which will reduce our profit margins, this will risk the long-term sustainability of DigiFarm’s business model and operational capacity, as DigiFarm has identified the price-point with extensive market research to determine the (lowest possible price), currently 25% lower than competitors for 10x higher resolution and accuracy.

Mitigation methods: Data engineering and algorithmic design is continuously focusing on computational efficiency and scalability. DigiFarm will set up its own HPC infrastructure with local GPU-instances in order to reduce costs of data processing in cloud providers ecosystems where costs are ~$3 per hour per GPU, compared to $0.30 per hour. This has only been made possible due to: (a) recent advances in hardware (Graphic Processing Units) by industry (lead by NVIDIA) where performance of GPUs have tripled since 2015, (200 TFLOPS to 1500) combined with the affordability of the units (approx. 300% reduction in performance/$ ratio).

Type of risk: Commercial

Description of risk: uptake of launched services does not reach required target for long-term sustainability of business model.

Effect of the risk: if there is low commercial traction for the product and service developed in Ethiopia, both among smallholder farmers (B2C) and agribusinesses (B2B/B2G) this will incur delays on reaching targeted profitability and prolonged scalability.

Mitigation methods: Prior to proposal DigiFarm has conducted significant market research in Ethiopia, through its discussions with IFC, World Bank and UNCDF regarding the needs internally in Ethiopia and the market landscape. Additionally, as our business model does not rely on the ground-data collection and marketing, we intend to use our partner network and to secure commercial local partners for the last-mile implementation. Lastly, as our business model is SaaS and only based on consumption on a per hectare basis (starting from 0.03 EUR with a 70% margin) the services we can offer (fully automated) will be affordable for even the smallest farmers in Ethiopia (down to 0.02 hectares) at the same time as offering actionable advice on optimisation; e.g. reduce input cost by up to 10-15% and increase yield by up to 10%.

Please provide a detailed plan, including timeline and key milestones for delivering your proposal.

M1-M3

1. Identify AOIs in Ethiopia where ground truth data is available (from small section of small holder farmers)
2. Process Sentinel-2 imagery for 2022 data (3-4 cloud free images, L2A) across the AOIs
3. Deeply resolve Sentinel-2 10m per pixel imagery to 1m per pixel (4-bands, RGB + NIR)
4. Process additional data (30+ years) from Landsat and NDVI (vegetation index)
5. Start model training (deep neural network) based on the training data

M3-M5

1. Start collecting training data (manual delineation of field boundaries across the AOI)
2. Proof-of-concept: field boundaries and seeded acres model production ready results
3. Register field boundaries with metadata on blockchain
4. Re-train model if necessary in order to achieve accuracy of IoU of 0.95+
5. Technical architecture documentation posted at <https://digifarming.readme.io/>
6. Automatically detect long-term productivity Zones (35 years) on the AOI
7. Coordinate with local financial institutions and credit bureau's to assess credit portfolio of smallholder farmers - in order to build a model where farmers can access affordable and reliable financing pre-season and during-season
8. Proof-of-concept: web-app DigiFarm
9. Formal partnership with legal professional(s) to liaise with Ethiopia (IFC) land management office
10. Blog post: write-up
11. Blog post: pilot scenario

M5-M7

1. Extend AOI to multiple application use-cases
2. Open the database to smallholder farmers to access this data and to communicate data to the financial and credit lending facilities
3. Blog post: "DigiFarm in Ethiopia - Results and findings"
4. DigiFarm Whitepaper (overview of solution concept and technical architecture)
5. Securing commercial partnerships with local partners including IFC to broaden scope of project and reach wider stakeholders in agricultural value chain for maximum impact

Applications which can be built on top of the baseline service:

1. Credit underwriting and micro-financing assessment index
2. Yield prediction
3. Crop classification and run assessments in conjunction with long-term productivity data
4. Provide subsidy-declaration service to national agricultural authority to verify total seeded acres and field boundaries declared against open ledger
5. Analyse historical data, e.g. deforestation as a result of crop-production back to 2015 across entire Ethiopia

End of Fund 9 grant

M7-M12

1. DigiFarm Whitepaper v2: Decentralised registry of field boundary productivity in Ethiopia
2. DigiFarm Initial Stake Pool Offering (ISPO)
3. DigiFarm beta Dapp on Testnet
4. 250 Ethiopia property owners field testing DigiFarm
5. Secure partnerships with local institutions to support the smallholder farmers and develop standard for certification and monitoring of sustainability

M12+

1. DigiFarm Level 3 Certified Dapp
2. DigiFarm launch on Cardano mainnet
3. 2,500 registered smallholder farmers in Ethiopia
4. DigiFarm's platform fully operational

Screenshot attached in supporting documents shows the phased introduction and launch of services in Ethiopia: starting with the green labelled areas, blue, dark green and lastly orange.

Please provide a detailed budget breakdown.

Budget breakdown:

Hosting for the project website and code repositories are provided free of charge via Github. Community outreach will be done via (free) Linkedin, Facebook and YouTube accounts along with Project Catalyst communication channels.

Detailed roadmap above for descriptions of the tasks and work products that will be delivered in three, four-week sprints

1. Month 1-2 (Tasks #1-5): 250 hours x $65 = $16,250
2. Includes cloud costs involved in processing imagery data, e.g. AWS open bucket and GPUs running (V/P100s) for approx. 20 tiles (20 million hectares of land data)
3. Build and develop script for processing 30+ year of Landsat data with the assumption field boundaries will be small (e.g. less than 0.5 hectares on average), will require optimisation of processing.
4. Field boundaries delineated: includes running current model developed for Tanzania in Fund 8 across the Tanzania AOIs in order to assess accuracy based on IoU and to identify the required manual training data for optimisation
5. Month 2-4 (Tasks #1-9): 500 hours x $65 = $32,500
6. Includes creation of training data, estimated at 3,500 hectares per 1 mill. hectares of delineated field boundaries (e.g. 0.35%), which will include approx. 350 different areas distributed across the 20 million hectares of initial AOIs, this will include manually drawing and digitising field boundaries in QGIS (GIS-platform), including two reviews.
7. Model training commences; e.g. start training the updated deep neural network models for delineating field boundaries based on the additional training data, run accuracy assessment report based on three distinct factors: val_IoU, extent_IoU and contour_IoU, assess if above 0.94 accuracy (extent) then can inference model with first updated and improved results, ready for live production environment.
8. Develop blockchain technology for putting field boundaries and initial ledger on BC
9. Coordinate and secure partnership with IFC (International Finance Corporation) in Ethiopia, based on existing conversations and solidify agreement to include.
10. Month 4-6 (Tasks #1-4): 530 hours x $65 = $34,450
11. Includes securing commercial contracts with local partners in Ethiopia
12. Coordinating and converting smallholder farmers to the platform, including training, onboarding and retention.
13. Develop methodology and technology for conducting Crop classification on existing field boundary dataset for 2022 season, focusing on major crops: barley, wheat, surgum
14. Coordinate with local partner including smallholder farmers and IFC to collect manual training data, e.g. crop types planted for each plot (distributed across the 20 million hectares) preferably across 2+ years, seeding and harvesting date (if possible to collect).
15. Create whitepaper and case study with both local partners (IFC etc) as well as smallholder farmers onboarded.
16. Professional fees: including legal, accounting, and communications: $8,000

SUBTOTAL: $91,200

Please provide details of the people who will work on the project.

DigiFarm’s team is the ideal fit for the project as our core team has extensive experience in (a) developing agricultural technology for crop-monitoring using AI and remote sensing (Satellite data) to the agribusinesses market (B2B/B2G) using SaaS-models. Successfully built commercial agricultural technological solutions using remote sensing (Satellite-data) and AI across 100 million hectares: >90% accuracy in crop Detection and >85% accuracy in yield-prediction in soybean and corn (US/Brazil) (b) core team has over 15+ years of on-the-ground crop-producing (farming) experience and close partnership withs Felleskjøpet (largest ag-coop in Norway, NLR (Norwegian Agricultural Advisory Organisation) and University of Life Sciences (NMBU) (c) commercial and corporate Ag-market: over 20+ years combined corporate agriculture leadership experience (d) over 40+ experience in agronomy academic research internationally.

Additional qualifications in DigiFarm’s core team and founders (10) include technical and agronomical experience: (a) over 40 years combined international work experience in precision-Ag projects in Canada, USA, Germany, Switzerland, Brazil, Australia, Russia and Ukraine (b) successfully filed 5 patents (AI-based technologies) in agriculture/biology (e) developed technology for Zoner.ag (one of first geospatial web-platforms for analyzing agricultural fields) successfully acquired by Bayer to become the geospatial engine of Xarvio digital-farming platform (owned by BASF). Management capacity: led and managed the Bayer CropScience division as Global Technology Lead with the Digital Farming Division, overseeing expansion Xarvio to over 100 employees, serving over 3.4 million farmers and agronomists worldwide (b) founded and grew AI-based Gamaya (Swiss-based) agtech startup, managed team growth to 45 employees in under 24 months and secured $20 million in VC funding from Mahindra.

1. Nils Helset, 15th generation farmer in Norway. Extensive experience with precision agriculture services and agronomy - over 8 years experience in crop-producing, managing over 40k decares of pilot farmers with Felleskjøpet.
2. Linkedin profile
3. Konstantin Varik - built advanced AI models in agriculture using remote sensing for over 8 years including AI model for crop-yield prediction for maize and soybeans for USA, Brazil and Argentina with 98-99% yield-prediction accuracy 1-1.5 months prior to harvesting (data provided to the USDA).
4. Linkedin profile
5. Alex Melnichouck founded B2B ag-tech startups Zoner.ag in 2012 (acquired by BASF in 2015), led and managed the global BASF digital farming team for over 4 years.
6. Linkedin profile
7. Yosef Akhtman founded and grew Gamaya, Swiss-based ag-tech startup focused on remote sensing and AI funded by Mahindra. Yosef is a scientist and inventor with over 5 registered patents in agriculture and biology.
8. Linkedin profile

If you are funded, will you return to Catalyst in a later round for further funding? Please explain why / why not.

Yes, DigiFarm has already one on-going project which was successfully funded from Fund 8 "Open ledger for agricultural land" for Tanzania. DigiFarm firmly believes it's unique technology: super-resolution Satellite data coupled with deep neural network models will be a crucial fundamental layer of digital agricultural identity, worldwide, and intend to keep growing our commitment to the Cardano vision and community.

Please describe what you will measure to track your project's progress, and how will you measure these?

1. Github commits for each sprint (4 weeks)
2. Reports and deliverables at each milestone (M1-M7)
3. Prototype and web-application
4. Collect user feedback from MVP and beta users through digital forms and frequent communication with pilot users and additional stakeholders (partners and clients)
5. Number of onboarded users and feedback looped turned into product design improvements
6. Number of applications built on top of the technology stack across different industry verticals

What does success for this project look like?

1. Ability to create the fundamental layer of agricultural digital identity for where scalable applications and products from the Cardano community can be built upon to provide additional value, assisting Cardano attracting users
2. Ability to reach target IoU accuracy of delineation of permanent crops across all arable land in Ethiopia at above IoU of 0.95
3. Ability to accurately assess and calibrate with local farmers and ground truth data, e.g. yield-data and productivity zones, e.g. low/medium and high zones will be assigned predictive yield-values, in-season and historically above 90% accuracy which can be used as leverage from securing financing for the smallholder farmers.
4. Ability to secure 500+ farmers within first 9 months and secure 5 lending institutions as pilot users and further 2,500 users within 12 months
5. Successful sustainable commercial model after the grant-period has ended

Please provide information on whether this proposal is a continuation of a previously funded project in Catalyst or an entirely new one.

Entirely new one.