text size: smaller reset larger



'GLAM'ming it up for crop forecasting

Despite enormous improvements in agricultural technology since the Second World War, food production continues to remain vulnerable to year-to-year weather variations, especially in the tropics.Climate change will make anticipating the weather even harder, as regional variations in weather (intense storms, drought etc.) and climatic variations in temperature and rainfall are likely to increase. Extreme weather events, such as short periods of very high temperatures are also much more likely to occur over wider areas by the end of this century. Tools are needed to investigate the impact of climate change on crop production and help farmers reduce risk by providing them with crop-yield forecasts a season or more in advance.

A combined climate and crop model

Linking weather forecasts and climate models to crop models is difficult because of differences in scale, with weather and climate modelled over large areas and crops modelled at the field level. To bridge the gap, scientists at two UK research centres* have developed the General Large Area crop Model (GLAM). Using information from the latest models for predicting climate and weather, this combined climate-and crop-forecasting system can predict harvests in current and future climates. "These forecasts are not only useful to farmers," said Dr Tim Wheeler, director of the Plant Environment Laboratory at the University of Reading, "they can also be used to advise policymakers of potential climate change and prospective dangers to food supplies in the years to come."

Central to developing GLAM has been understanding the relationship between weather variables and crop yield and quality. Wheeler and his colleagues found that seasonal rainfall variation across India during 1966-89 explained more than half the variance in groundnut yields at a national level. A useful weather-yield relationship thus exists over several hundred square kilometres, which is close to the spatial scales used in climate models. GLAM, a unique tool based on weather-related yield-determining processes, potentially allows researchers to make seasonal forecasts of crop yields, investigate the influence of weather variability within a season (such as a sudden dry spell), and predict how climate change will affect crop yield. Researchers tested the combined forecasting system by applying it to observed weather over a past period and seeing how well it allowed them to "predict" crop yields. Using combined data from several models, they produced hindcasts of national groundnut yields in India for 1987-98 that closely correlated with actual yields, notably in extreme years when accurate forecasting is most important.

Effect of climate on crop growth...

Groundnut harvesting (Peter Craufurd, University of Reading)
Groundnut harvesting
Peter Craufurd, University of Reading

Whilst extreme years, with prolonged drought or high temperatures, occur irregularly in current climates, it is possible that crops will be prone to increased environmental stress under climate change. Rice subjected for one hour to 35ºC or more at the flowering stage becomes sterile. In groundnut, high temperatures affect how well fruit sets, and wheat and maize show similar effects of heat on reproductive processes. "Reproductive limits are crucial," said Professor John R. Porter of the Royal Veterinary and Agricultural University in Denmark, speaking on crop responses to climate variability at the Royal Society conference Food Crops in a Changing Climate. "These are much narrower than lethal limits and are very important when talking about crop models."

According to recent climate studies, groundnut in India is currently rarely affected by high temperature stress. However, a climate scenario produced by a regional model from the Hadley Centre in the UK shows that, by the end of the century, temperatures could exceed 34ºC on most days during flowering, which would seriously affect yield. Farmers can adapt by growing groundnut varieties more tolerant to high temperatures, which GLAM shows would significantly improve yield.

...and food quality

Climate variability can also affect grain quality. Drought not only reduces wheat grain size but also glutenin content, and consequently dough strength - an important consideration if the wheat is to be used for bread-making. According to Porter, insufficient attention was given to the impact of climate on food quality in the 2001 report of the Intergovernmental Panel on Climate Change (IPCC). "We have the data," he said. "We just haven't been asking the right questions."

By extending the methods beyond groundnut to rice, wheat and maize, GLAM researchers hope to ask the right questions and better understand how climate change will affect global food production. Also under investigation is the fertilisation effect that higher CO2 levels may have on photosynthesis and food production, as well as feedback systems between crop growth and climate. But GLAM is not just a crystal ball used to forecast the future. With this unique tool seasonal forecasts in current climates will benefit farmers in coming years and could be of great value in predicting crop failure for famine early warning systems.

*NCAS Centre for Global Atmospheric Modelling and Department of Agriculture, University of Reading.

Date published: July 2005


Have your say


The New Agriculturist is a WRENmedia production.

This website uses cookies to improve your experience. By continuing to browse the site you are agreeing to our use of cookies.
Read more