Farming can often be a volatile business, but NASA researchers this week published a new study showing that climate data could help predict crop failures months in advance.

The study, published in the journal Nature Climate Change, linked temperature and soil moisture for a large segment of global rice and wheat crops. A computer simulation was then able to predict crop failures three months before they happened for around one-fifth of global crops. The computer simulation was based on temperature and rain forecasts, as well as satellite data taken from 1983 to 2006.

“You can estimate ultimate yields according to the climatic condition several months before,” said Molly Brown, a researcher in the Biospheric Sciences Laboratory at NASA’s Goddard Space Flight Center. “From the spring conditions, the preexisting conditions, the pattern is set.”

The study also looked at corn and soybean simulations, but found that the model was less accurate for those crops. For rice and wheat, though, the simulations were able to predict even minor crop yield changes. The study’s authors believe that the model can be used to help farmers in poorer countries better predict and invest in their crops.

“We can make a new framework that would allow much greater exploitation of satellite data and climate prediction models,” said Brown. “If you knew you were going to have a good year, you could plan, you could give out loans, you could do other things to boost food production to be prepared for bad years.”