National level satellite-based crop field inventories in smallholder landscapes

2025-07-14

AI Summary

In smallholder farming systems, the lack of spatially explicit information on actively cultivated land and field-level boundaries hinders national-scale agricultural census and evidence-based policy formulation. To address this, we propose a low-reference-data-dependent field boundary segmentation method that integrates 1.5-meter-resolution satellite imagery with a deep transfer learning model. Applied across Mozambique, it produced the first nationwide vector dataset of 21 million individual fields. Our approach overcomes the limitations of conventional land cover products in detecting fragmented smallholder plots, achieving an overall classification accuracy of 93% and a median field-level Intersection-over-Union (IoU) of 0.81. Spatial analysis reveals that 83% of fields are smaller than 0.5 hectares, and we quantitatively demonstrate their significant associations with socioeconomic and environmental covariates. This work establishes a novel paradigm for smallholder-oriented sustainable agriculture monitoring and governance, underpinned by high-precision, field-scale geospatial data.

Technology Category

• Computer Vision: Remote Sensing / Geospatial AI
• Machine Learning: Matrix & Tensor Methods
• Search and Optimization: Distributed Search

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

The design of science-based policies to improve the sustainability of smallholder agriculture is challenged by a limited understanding of fundamental system properties, such as the spatial distribution of active cropland and field size. We integrate very high spatial resolution (1.5 m) Earth observation data and deep transfer learning to derive crop field delineations in complex agricultural systems at the national scale, while maintaining minimum reference data requirements and enhancing transferability. We provide the first national-level dataset of 21 million individual fields for Mozambique (covering ~800,000 km2) for 2023. Our maps separate active cropland from non-agricultural land use with an overall accuracy of 93% and balanced omission and commission errors. Field-level spatial agreement reached median intersection over union (IoU) scores of 0.81, advancing the state-of-the-art in large-area field delineation in complex smallholder systems. The active cropland maps capture fragmented rural regions with low cropland shares not yet identified in global land cover or cropland maps. These regions are mostly located in agricultural frontier regions which host 7-9% of the Mozambican population. Field size in Mozambique is very low overall, with half of the fields being smaller than 0.16 ha, and 83% smaller than 0.5 ha. Mean field size at aggregate spatial resolution (0.05°) is 0.32 ha, but it varies strongly across gradients of accessibility, population density, and net forest cover change. This variation reflects a diverse set of actors, ranging from semi-subsistence smallholder farms to medium-scale commercial farming, and large-scale farming operations. Our results highlight that field size is a key indicator relating to socio-economic and environmental outcomes of agriculture (e.g., food production, livelihoods, deforestation, biodiversity), as well as their trade-offs.

Problem

• Mapping national-level crop fields in smallholder landscapes
• Separating active cropland from non-agricultural land use
• Analyzing field size variations and their socio-economic impacts

Innovation

• Uses high-resolution Earth observation data
• Applies deep transfer learning techniques
• Generates national-level crop field datasets