Mapping Diverse and Dispersed Smallholder Irrigation in Sub-Saharan Africa through Remote Sensing

This thesis explores the challenges and methodologies of mapping smallholder irrigated agriculture in sub-Saharan Africa using remote sensing. Despite growing interest in irrigation for agricultural development and food security, smallholder farmers’ initiatives often go unnoticed due to their fragmented nature and technical biases in defining irrigation. The research examines four study areas in Mozambique with diverse agroecological characteristics, addressing methodological considerations in remote sensing classification including algorithm selection, composite temporal length, training sample size, and model transferability between regions. Key findings highlight the impact of methodological choices on irrigation mapping accuracy, emphasizing the need for transparency in reporting classification methods. The study contributes to understanding how remote sensing can better represent farmer-led irrigation development while acknowledging the limitations and potential improvements in mapping approaches.

Thesis

PhD Thesis, Wageningen University