Blog icon

By  Madeleine Clarke Alison Donnellan 29 June 2023 3 min read

Key points

  • We’ve applied statistical modelling to satellite imagery to understand the impact of the war in Ukraine on global food supply.
  • Ukraine’s crop productivity significantly declined due to the war and cargo shipping activity at the major ports at Odesa and Mariupol decreased about 50 per cent.
  • The new method could support international leaders with fast comprehensive insights to inform responses to crises like war, drought and pandemics.

One country produces about 12 per cent of the grains (wheat, barley, maize and sunflower) traded globally. Developing countries in Africa and South Asia rely heavily on it. In fact, this country supplies up to 40 per cent of the grain distributed by the United Nation’s World Food Programme.

This country is Ukraine, and it is at war.

It’s a case study of how disruptions to agriculture supply chains by events such as war, drought or pandemics can have wide-ranging and complex impacts. In a globalised and increasingly interconnected world, understanding the flow-on effects of these major events is crucial.

To access data on crop productivity after the invasion of Ukraine, our researchers used remote sensors like the Google Earth Engine. Image: Google, Maxar Technologies, CNES/Airbus

Our researchers analysed images of Ukrainian cropland and cargo shipping activity after the Russian invasion taken by satellites.

Their work, published in the public access journal PLOS1, uses statistical modelling techniques to compare cropping and supply patterns in 2022 with historical data.

The results help pinpoint the impact of war on the global food supply.

Statistical modelling from space

An artist's conception of the Landsat 9 spacecraft, one of the satellites and remote sensors which can be viewed on Google Earth. Image: NASA's Goddard Space Flight Center/Conceptual Image Lab

Statistical modelling is a method that allows us to understand how and why variables change over time and space.

It’s particularly useful for making predictions and understanding impacts when access to data is limited, such as during a war.

This research is the first to apply this method to satellite imagery during a conflict. It offers fast, remote insight into unfolding events.

Dr Petra Kuhnert is an author and applied statistician from our Data61 business unit. She said these insights are fertile ground for decision-makers.

“This approach can be applied to a diverse range of events that disrupt food security, including pandemics, natural disasters, and economic downturns,” Petra said.

“Governments and international agencies often need to take immediate action and respond to food value chain disruptions quickly. Such insights delivered at speed can prove incredibly valuable.”

Michael Wellington is a PhD student at the Australian National University. He is supervised by Petra and Roger Lawes, CSIRO researchers and coauthors of the modelling on Ukraine. Michael previously used this method to measure the impact of drought on crop production in Madagascar.

The island nation has been experiencing acute food shortages since early 2021. He found the method could be adapted for diverse farming systems from smallholder irrigation schemes to broadacre grain regions.

Export logistics disruption

A decade of crop growth and loss data, and two years of grain cargo shipping movements around Ukraine, was analysed to create a baseline for comparison. The 2022 data was pulled from the Google Earth Engine and the European Marine Observation and Data Network.

The modelling showed Ukraine’s 2022 overall crop productivity significantly decreased as a result of the war, compared to baseline data from 2010-2021.

Furthermore, the impact of the war on export logistics was severe.

A comparison of the density of cargo ship routes between 2021 and 2022 revealed activity around the Ukrainian ports at Odesa and Mariupol decreased by 45 and 62 per cent respectively following the invasion.

Between 2021 and 2022, activity around the Ukrainian ports at Odesa and Mariupol decreased by 45 and 62 per cent.

Petra said these insights shed light on the importance of exploring multiple lines of evidence to understand the bigger picture when studying and responding to crises.  

"Previous studies on the impacts of conflict on food systems have focused on farm production. However, our results suggest disruption to supply and logistics may be a greater threat," she said.

"This could be due to over-reliance on logistics facilities in a few key locations, while the many enterprises comprising the farm sector are geographically dispersed.”

Our research into how diverse events disrupt food security could improve the resilience of global systems.

An early warning system for global food security  

This work demonstrates rapid insights can be gained from satellite data during global challenges to inform international responses.  

Coauthor Dr Roger Lawes is a farming systems scientist in our Agriculture and Food business unit. He said the method could be applied on a wider scale to act as an early warning system for global food security. 

"This combination of farming systems research with big data analytics has created a new way to rapidly monitor the global food supply," Roger said.  

"Quantifying the global impact of an ongoing event less than a month after it began could help manage acute food shortages.  

"This method could help enhance the preparedness and resilience of the world’s food system in the face of changing environmental, geopolitical and economic circumstances."

Contact us

Find out how we can help you and your business. Get in touch using the form below and our experts will get in contact soon!

CSIRO will handle your personal information in accordance with the Privacy Act 1988 (Cth) and our Privacy Policy.


First name must be filled in

Surname must be filled in

I am representing *

Please choose an option

Please provide a subject for the enquriy

0 / 100

We'll need to know what you want to contact us about so we can give you an answer

0 / 1900

You shouldn't be able to see this field. Please try again and leave the field blank.