Search for: All records

Mapping agricultural fields using high-resolution satellite imagery and deep learning (DL) models has advanced significantly, even in regions with small, irregularly shaped fields. However, effective DL models often require large, expensive labeled datasets, which are typically limited to specific years or regions. This restricts the ability to create annual maps needed for agricultural monitoring, as changes in farming practices and environmental conditions cause domain shifts between years and locations. To address this, we focused on improving model generalization without relying on yearly labels through a holistic approach that integrates several techniques, including an area-based loss function, Tversky-focal loss (TFL), data augmentation, and the use of regularization techniques like dropout. Photometric augmentations helped encode invariance to brightness changes but also increased the incidence of false positives. The best results were achieved by combining photometric augmentation, TFL, and Monte Carlo dropout, although dropout alone led to more false negatives. Input normalization also played a key role, with the best results obtained when normalization statistics were calculated locally (per chip) across all bands. Our U-Net-based workflow successfully generated multi-year crop maps over large areas, outperforming the base model without photometric augmentation or MC-dropout by 17 IoU points. 

Abstract Agricultural supply chains play a crucial role in supporting food security in Africa. However, high-resolution supply chain information is often not available, which hinders our ability to determine which interventions in food supply chains would most enhance food security. In this study, we develop a high-resolution supply chain model for essential staple crops in Zambia, aiming to estimate how improvements in transportation infrastructure would impact food security. Specifically, we simulate district-level monthly consumption, trade flows, and storage for maize and cassava in Zambia. We then conduct a counterfactual case study with low transportation costs, discovering that reducing transaction costs leads to higher aggregate net agricultural revenue and aggregate net expenditure. These results indicate that transportation investments are more beneficial to suppliers than to consumers, with implications for household food security in smallholder agriculture. Our study highlights the potential for infrastructure investments to improve food security.