Since late 2024, large parts of Syria have been experiencing a rapidly intensifying agricultural crisis driven by severe and prolonged drought. This has placed acute pressure on rural livelihoods and accelerated broader economic vulnerability across the country. By mid-2025, irregular winter rainfall had already jeopardised up to 75% of wheat production, with projections indicating a potential 2.7-million-ton national shortfall, threatening the food security of more than 16 million people in the year ahead.
Declines in the Normalised Difference Vegetation Index (NDVI) across key agricultural zones point to widespread deterioration in vegetative health, suggesting that the 2025 season may be the worst in decades.Combined with significant rainfall deficits, these trends indicate a sharp contraction in wheat yields and a heightened risk of systemic production failure. At the same time, officials and analysts warn that poor harvests will have cascading effects, driving food insecurity, undermining fragile markets, and eroding already limited coping capacities among vulnerable communities.
Despite this, response efforts risk operating without a clear evidence base. Planning for the 2025–2026 planting and irrigation cycles remains constrained by uncertainty, limiting the ability of authorities and humanitarian actors to allocate resources effectively, stabilise markets, or mitigate emerging risks of food insecurity and displacement. Without timely and granular data, interventions are likely to remain reactive, fragmented, and insufficiently aligned with rapidly evolving conditions on the ground.
The project
The Crop Recognition and Observation Platform for Identification and Diagnostics (CROP-ID) initiative aims to address these critical information gaps by developing a remotely sensed, machine learning–enabled analysis of Syria’s agricultural landscape. By combining open-source satellite imagery with field-level training data, the project seeks to generate a cost-effective and scalable approach to understanding crop distribution, agricultural trends, and environmental stressors across the country.
Through the production of high-resolution geospatial outputs, including crop classification maps, raster datasets, and analytical reports, CROP-ID will provide a structured and interpretable evidence base to support decision-making across the humanitarian–development–peace nexus. The project will also test the operational feasibility and reliability of integrating machine learning with remote sensing in complex and data-constrained environments, contributing to broader learning on innovative approaches to agricultural analysis.
Expected impact
CROP-ID aims to enable humanitarian, governmental, and development actors to access timely and granular insights into Syria’s evolving agricultural conditions. By identifying crop types, tracking cultivation patterns, and estimating water demand, the project aims to support more targeted and efficient interventions across food security, early recovery, and climate adaptation programming. A shared and accessible evidence base also aims to support greater alignment across actors, enabling more coherent and coordinated responses in a resource-constrained environment. By strengthening local analytical capacity and integrating stakeholder engagement throughout the process, CROP-ID will also contribute to more sustainable, data-driven agricultural planning in Syria.