Project Story

Weather alerts for crop monitoring

Weather alerts for crop monitoring image

In the ALERT project we worked with a UK supermarket, a large global food company and two commercial growers to develop a proof-of-concept software tool to demonstrate how weather could affect crops across a range of test locations in the UK and overseas.

Within agriculture, weather risks can operate over a wide variety of timeframes. Events can be “gradual onset” (such as a drought, or a spell of sunny weather) or “acute” (such as a hailstorm, or a storm causing structural damage), occurring with relatively little warning. Within a season, the favourability of weather conditions can impact production by affecting crop development or the agricultural process, for example, excessive moisture can hamper planting or harvesting activity. The definition of favourable or adverse conditions changes as the plants enter different development stages (for example stone fruit orchards may be unaffected by hail when in the budding stage, but at risk when fruit are setting). 

Weather conditions may also affect the onset and distribution of pests and diseases. For example, wet conditions may encourage the contraction of viruses, or prevailing winds may affect the spread of pests. The impacts of adverse weather may include: (i) a reduction in yield; (ii) a reduction in quality of product or (iii) the complete destruction of a crop. There are potentially several risks for every combination of crop, location, method of production and development stage. Given that there are many thousands of growing locations in some supply chains this could represent hundreds of thousands or millions of risks to monitor and manage. 


The project started with a series of client workshops to refine the problem, to map the different components of the fresh produce supply chain, to identify the variables to be predicted and estimate the required levels of timeliness and accuracy required for the forecast information to be delivered to produce managers. The IEA weather experts then built a data repository and correlated the list of past incidents with reported weather events. 


The IEA technical team developed an early stage proof-of-concept software application to monitor a limited number of crops in specific locations using the weather data repository that had been created. This was then trialled over several months with users drawn from across the 5 clients involved in the trial. 


    The proof-of-concept was successful in demonstrating the potential usefulness of giving advanced warning of supply chain disruption to our commercial partners in the project. Based on historic data, examples were found of the ability to anticipate impacts from days to months ahead of occurrence, and the opportunity to activate probability driven responses to severe weather events within days of occurrence.

    Back to project stories

    Tell us about your data challenge

    Get in touch