Supporting Food Security through Emerging Technologies
Leveraging technology and local expertise for comprehensive food security solutions.
FEWS NET, the Famine Early Warning Systems Network, was established in 1985 by USAID in response to devastating famines in East and West Africa to address the critical need for better and earlier warning of potential food security crises. It is often lauded as one of the most successful programs in the agency’s history. FEWS NET provides timely, evidence-based, transparent information and analysis of current and future acute food insecurity all over the world – from Central America to East Africa to Ukraine, and everywhere in between. This, in turn, informs decisions on humanitarian planning and response in the world’s most food-insecure countries.
Chemonics International’s Stephen Browne leads the FEWS NET 7 Pillar 3 Task Order 1, which is focused on livelihoods and food security research. His team’s mission is to expand knowledge on food insecurity by developing Household Economy Analysis baseline profiles, livelihood zone maps, and analysis tools in targeted geographies. They seek to strengthen local capacity, support evidence-based food security efforts, and collaborate with FEWS NET’s Early Warning Team, USAID, and other partners to improve resource targeting and allocation.
As evolving market and trade dynamics, geopolitics, climate, and conflict make tracking and analyzing food insecurity ever more complex, Stephen reflects on the importance of innovation, local expertise, and strong partnerships in achieving FEWS NET’s mission.
At a time where new technologies are plentiful, how does FEWS NET balance their integration with human expertise?
FEWS NET’s Livelihoods Team welcomes innovation and actively explores methods to integrate technology and traditional data collection processes to address information gaps.
Technology can prove particularly valuable in contexts of conflict and inaccessibility, where traditional data collection may be difficult or impossible. For example, in Northeast Nigeria, our data collection teams could not access areas around Lake Chad due to ongoing conflict. Alongside consortium partners, the Livelihoods Team piloted an approach that used mobile phone data collection technology to interview the area’s residents. Integrating this technology is promising, but further refinement is required to explore the full potential of its application.
Many of the results of new data collection approaches – artificial intelligence, mobile data collection, and other computer-assisted technologies – are quantitative. However, traditional data collection methods can capture the story and context behind the numbers. For example, if households report not meeting our baseline of 2,100 kilocalories per person per day, it indicates a potential issue – either in how we articulated the line of questioning or in the respondents’ recollections. Manual, human-driven data collection allows us to remain flexible and adjust our questions during the interview to ensure accurate reporting on energy needs and incomes or expenditures. However, AI can review our database of global baselines and identify patterns in the datasets of geography, agroecology, market basins, and livelihood patterns, enabling us to form assumptions about un-reported areas and develop proxy baselines. AI-supported proxy baselines could further support the flexibility of this process and contextualize our analysis. Both approaches – technological and human-driven, manual ones – should complement each other to create the most complete set of evidence possible to inform food insecurity analyses.
What does this balance and integration look like in areas where implementation may be more challenging?
At the country-level, we start broad, collecting data through various methods, including livelihood zoning and mapping exercises. Then, we move to smaller, more targeted sampling within each livelihood zone.
In the Democratic Republic of the Congo (DRC), for example, teams collected qualitative data on seasonality, production, market prices, and socio-economic profiles through focus group discussions, corroborating information gathered across multiple regional-level livelihood zoning exercises. This information is still collected via paper and pencil at the territory, market trader, and community leader levels.
On the other hand, implementing partner REACH identified an opportunity to streamline some of our data collection approaches, particularly in conflict-affected areas, to minimize the time data collectors stay in insecure environments. By adopting digital versions of our traditional paper and pencil tools, such as the longer, quantitative household level representative interviews, teams could more quickly and safely collect, store, and share this data.
What role does local expertise play in addressing food security?
Our project has had great success tapping into the expertise of local government stakeholders. We recognize the real value in strengthening existing technical capacity and collaboratively reinforcing the current food security analysis infrastructure.
In Mali and the DRC, we included government stakeholders in planning and executing the fieldwork and analysis, ensuring transparency and open communication. Their insights on the operating environments were crucial and greatly enhanced our planning, reduced inefficiencies, and provided first-hand context to the analysis results.
We also worked alongside government institutions, UN agencies, and NGOs in the DRC to understand how newly developed baselines can inform existing food security monitoring structures. Now, these groups are positioned to independently conduct food security analyses nationwide alongside FEWS NET’s early warning team and food security analysts from multiple agencies, which reduces the need for external consultants or international experts.
In your experience, how can the public and private sectors collaborate to lead innovation in food security?
Private sector partnerships have untapped potential to complement public sector work, especially around data collection and disaggregation.
Many of the world’s rural poor do not produce sufficient food from their land, and therefore must work for others to earn the cash used to purchase food and other items for survival and to maintain their livelihood. In Guatemala, for example, private companies in the coffee, sugar, and fruit sectors monitor wages, seasonal peak periods, frequency of work, and other labor-related information. FEWS NET could use the data to communicate needs to decision-makers, allowing them to more accurately target their support to populations most in need of assistance.
Collaboration with the private sector to develop and implement innovative solutions around data collection, local infrastructure, market access, and capacity strengthening can enhance current methods and fill gaps that will add value to food security analysis. FEWS NET looks forward to future partnerships with organizations in both the public and private sectors to better serve decision-makers and the communities where we work.