Analyzing the disruption of agricultural systems by conflict

A case study of sunflower production in eastern Ukraine

Journal Article (2026)
Author(s)

Junjie Qiu (Hangzhou Normal University)

Yuekai Hu (East China Normal University, TU Delft - Civil Engineering & Geosciences)

Dailiang Peng (International Research Center for Big Data for Sustainable Development Goals, Chinese Academy of Sciences)

Haijian Liu (Hangzhou Normal University)

Weichun Tao (Hangzhou Normal University)

Bin Xie (Hangzhou Normal University)

Tangao Hu (Hangzhou Normal University)

Jiake Wang (Hangzhou Normal University)

Xiao Liang (Hangzhou Normal University)

Tao Chen (Hangzhou Normal University)

Junfeng Xu (Hangzhou Normal University)

Research Group
Coastal Engineering
DOI related publication
https://doi.org/10.1016/j.agsy.2026.104636 Final published version
More Info
expand_more
Publication Year
2026
Language
English
Research Group
Coastal Engineering
Journal title
Agricultural Systems
Volume number
233
Article number
104636
Downloads counter
47
Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Abstract

Sunflower is a critical oilseed crop that underpins global food security. As the world's largest exporter of sunflower oil, Ukraine produced 6.89 million tons in 2021, accounting for approximately one-third of global production. However, the ongoing Russia–Ukraine conflict has severely disrupted the local agricultural system, and the specific impacts on sunflower production dynamics remain unclear. To address this, we constructed a comprehensive monitoring framework by integrating Sentinel-1 Synthetic Aperture Radar (SAR) and Sentinel-2 optical imagery. First, we mapped annual sunflower cultivation distributions from 2019 to 2023 using an automated sample extraction method coupled with a Random Forest model, achieving an overall classification accuracy of 94.35%. Second, we implemented grid-based production prediction to capture fine-scale agricultural productivity heterogeneity. The results reveal a 49.5% decline in sunflower cultivation area between 2021 and 2022, accompanied by severe landscape fragmentation. Notably, the loss pattern exhibited a distinct “strip-like” distribution along the conflict frontline. Regarding production, while total output collapsed, the trend in unit yields diverged from the drastic reduction in cultivated areas, suggesting a potential shift in the agricultural production mode toward concentration. Finally, analysis based on multi-source indicators confirmed that farmland destruction, personnel loss, and water source damage were the drivers of agricultural decline in the region. These findings highlight the high vulnerability of agricultural systems under armed conflict and provide critical insights for post-conflict agricultural recovery and sustainable land use policymaking.

Files

Taverne
warning

File under embargo until 13-07-2026