Understanding Agricultural Systems in Zambia
Sub-Saharan Africa (SSA) has undergone significant transformation over the last 25 years, reshaping where households choose to live, how they earn income, and how food is grown and consumed in a world experiencing rapid climate change. This interconnected system is best understood through the climate-food-urbanization (CFU) nexus, which examines the linkages of climate processes and shocks, food production and consumption, and urbanization pathways. This research project specifically focuses on the CFU in Zambia, a landlocked country in SSA; chosen for its diverse agronomic conditions, reliance on a key staple crop (maize), and varying levels of agricultural investment by state and private investors, as well as foreign actors across the country.
Kyra and Gemma researched the relationship between measures of agricultural investment and crop yields using a highly detailed survey of ~1500 households in Zambia conducted during 2024. The survey covered household demographics, migration, assets, food security, and detailed crop production. They combined the survey results with gridded remotely sensed measures of precipitation, demographic change, and other agro-climatic variables for a more complete picture of how households choose to invest their wealth in crop production, given the climatic conditions. Lastly, they looked at how different conservation agriculture (CA) practices could mediate climate shocks to household yield and food security measures to build a more resilient agricultural system for smallholder communities in SubSaharan Africa.
Kyra and Gemma’s Impacts:
- Conducted an in-depth literature review on agriculture in Zambia and Sub-Saharan Africa, climate shocks, food security, and the efficacy of different conservation agriculture practices in Zambia for a robust understanding of the geographic context
- Participated in and presented work to an All-Hands project meeting with researchers from different universities (UC Santa Barbara, Clark University, University of Illinois -Urbana Champaign) working on this topic in July
- Dove into the smallholder household survey – from drafting a survey codebook and ID keys, detailed data cleaning in R, to selecting relevant variables used in later statistical model analyses
- Processed gridded geospatial data (precipitation, population change, land classes, water availability, etc.) in R to compute zonal statistics and link to the household survey responses

