Data-driven Insights for Sustainable Agriculture (DISA)

An ambitious, multi-stakeholder effort to promote resilient and sustainable farming practices in Africa. 

Logo of the project and photo of a DISA collaborator speaking with Rwandan farmers about their farming practices.

Data-driven Insights for Sustainable Agriculture (DISA) is an interdisciplinary project that leverages the power of machine learning, satellite imagery, local knowledge and human-centred collaborations to promote an alternative approach to farming called regenerative agriculture.

Beyond the accrued benefits that regenerative agricultural practices bestow on the planet, they are also vital to the viability of smallholder farms and farmers. In this regard, the practices enhance food security, healthy diets, crop yields, employment prospects and income.

Background

Conventional, resource-intensive agricultural practices pose severe consequences for people and the environment. In Rwanda, these practices have caused over 40% of the soils to be degraded, with productivity continuing to decline. This is occurring while the demand for food is increasing. Most smallholder farmers in Rwanda (over 60% of which are women) live on less than $5 a day. These farmers are disproportionately finding it difficult to access needed inputs and other resources.

The implementation of policies and strategic incentives are crucial in driving the transition to regenerative agriculture. DISA project is designed to provide data-driven evidence to policy makers about the merits of regenerative agriculture. It is hoped that, with more substantive evidence targeted to meet the needs of Rwandan policy-makers, policies will be devised to incentivize these practices. This, in turn, will encourage farmers to transition from resource-intensive, extractive farming models to knowledge-intensive, nature-positive, net-zero models that drive economies.

Read more

DISA combines high-resolution satellite imagery with algorithms that can identify regenerative and non-regenerative agricultural practices at scale. This approach will allow us to demonstrate the significance of regenerative agriculture for high impact outcomes such as soil fertility, soil erosion and susceptibility of farms to extreme weather events, among others.

The goal is to have the project serve as a catalyst, inspiring confidence in regenerative agriculture to generate policy action and grassroots change. While our scope is initially focused on the Rwandan context, the objective is to scale the project in multiple countries across the continent.

The Origin of DISA

DISA originated in response to the pressing need for data-driven strategies aimed at accelerating the transformation of food systems. DISA strives to be a powerful force for the promotion and acceptance of sustainable agriculture, starting with the east-central African nation of Rwanda, where land-degradation problems are especially severe and where most of the farmers are smallholders.

The project is a collaborative effort among five organizations: Sustainability in the Digital Age (SDA), Future Earth, Planet, CIFOR-ICRAF and Mila. Future Earth and SDA are responsible for building local partnerships within Rwanda among experts in the field, policy-makers and farmers. Planet is providing access to its high-resolution satellite imagery, which the Mila team is using to create algorithms that offer insight into the long-term impact of sustainable agricultural practices in Rwanda. CIFOR-ICRAF is providing novel approaches of monitoring soil health and collecting data at scale.

The DISA team also includes consultants in Rwanda whose expertise lies at the intersection of geography, sustainability and data analysis. These include members of the Regional Research Center for Integrated Development (RCID) as well as Kaspar Kundert and Connie Schmidt Kundert.

Planting the Seeds of Environmental and Economic Sustainability

Our focus on Rwanda is the first step in a shift toward sustainable, evidence-based farming that we would like to see throughout Africa—a transition that can help to secure the livelihoods and futures of smallholder farmers. 

The project strives to empower farmers to build the long-term economic resilience that comes from reliable yields and support them in gaining access to carbon markets (while ensuring they are compensated for reductions in emissions and carbon sequestered). The aim is to reduce and, in some cases, prevent poverty, particularly among female smallholder farmers. Such farmers represent the majority of the Rwandan population and 60-80% of the farming labour market.

Making steady progress 

The DISA team is currently forging partnerships with local organizations and securing funding from philanthropic donors. They are also working diligently on defining the problem from a machine-learning perspective in ways that are contextually meaningful and technically feasible. To this end, we are drafting, and plan on publishing, lessons learned on the topic of Human-Centred AI.

Resources

 

Land Use Change Impacts on Water Erosion in Rwanda
Nambajimana et al., 2019. In MDPI Open Access Journals.
Agricultural Household Survey 2020
This report by the National Institute of Statistics of Rwanda (NISR) presents the results of the Agricultural Household Survey carried out from September 6 to October 8, 2020.
Regenerative agriculture and integrative permaculture for sustainable and technology driven global food production and security
McLennon et al., 2021. In American Society of Agronomy.
Regenerative agriculture – the soil is the base
Schreefel et al., 2020. In Science Direct - Global Food Security.
A farm in Rwanda.

Meet the Team

The technical portion of DISA’s work is being led by Mila, under the management of Allison Cohen, Senior Manager, Applied Projects. The team is composed of machine learning specialists including researchers.

Mila Members
Portrait of Allison Cohen
Senior Manager, Applied Projects
Portrait of Benjamin Prud'Homme
Vice President, Policy, Safety and Global Affairs, Leadership Team
Portrait of Gaétan Marceau Caron
Senior Director, Applied Machine Learning Research
Other Members and Collaborators
Mélisande Teng (PhD student, Mila)
Connie Schmidt Kundert (Specialist Consultant)
Kaspar Kundert (Specialist Consultant)
Kinsie Rayburn (Planet)
Melissa Rosa (Planet)
Athanase Mukuralinda (CIFOR-ICRAF)
Eliane Ubalijoro (CIFOR-ICRAF)
Leigh Winowiecki (CIFOR-ICRAF)
Tor—Gunnar Vagen (CIFOR-ICRAF)
Erin Gleeson (Future Earth)
Ernest Habanabakize (Future Earth)
Jennifer Garard (Future Earth)
Poonam Maskeri (Future Earth)
Rosette Savanna (Leapr Labs)
Jules Kazungu (Regional Research Centre for Integrated Development)
Rosine Ndayishimiye (Bridge2Rwanda)

Partners

This project has been supported thanks to generous contributions from the ClimateWorks Foundation and McGill Mastercard Foundation Scholars’ Program.