Validating a cassava production spatial disaggregation model in sub-Saharan Africa

Area of Science:

• Agricultural Science
• Geospatial Analysis
• Food Security

Background:

• Cassava is a vital staple crop in sub-Saharan Africa, adaptable to diverse environmental conditions.
• Previous research indicates a correlation between rural population distribution and cassava cultivation density.
• Accurate cassava production mapping is crucial for food security and agricultural planning.

Purpose of the Study:

• To investigate the relationship between cassava production disaggregation models (CassavaMap, MapSPAM) and rural population density in sub-Saharan Africa.
• To identify country-specific spatial trends in cassava production.
• To validate and improve the reliability of cassava production mapping models.

Main Methods:

• Analysis of survey data from 69 locations in Côte d'Ivoire and 87 in Uganda.
• Examination of relationships between cassava cultivation proportion and rural population/settlement data using generalized additive models.
• Aggregation of rural settlement data within 2, 5, and 10 km buffers around survey locations.

Main Results:

• No significant correlation found between rural population and cassava production using original survey data in both models.
• CassavaMap demonstrated an ability to capture large-scale cassava production variations when settlement buffers were aggregated.
• Identified distinct country-specific spatial trends associated with higher cassava production areas.

Conclusions:

• Aggregated settlement data improves the large-scale accuracy of cassava production models like CassavaMap.
• Validated cassava production disaggregation models enhance confidence in subsequent analyses, such as disease spread and nutrient availability.
• Improved cassava production estimates benefit researchers, policymakers, and the general population by ensuring greater reliability.