Using climate models to drive seed innovation.
There is a saying in agriculture among farmers, control the controllables and manage for the uncontrollable. Now, the biggest uncontrollable factor in raising a crop is the weather and there is no question that the climate patterns being experienced today are different from those experienced 10-15 years ago. That also means that in 10-15 years, the climate patterns will be different than today. This leads to the question of how current crop varieties will perform in future weather patterns.
Weathering the Future with Climate Models
Abigail Swann, University of Washington atmospheric and climate sciences and biology professor is tackling this question. Swann is applying her experience with climate science and investigating plant physiology in changing climates to agricultural systems, specifically corn production.
“A lot of my work is at the global scale, looking at various land systems and plants, and how climate matters for the functioning of those plants,” Swann says.
She has focused her research on studying how various plants function in current climate systems, observe how their functions have changed as the climate has changed and to predict how changes in climate will affect plant functions and the surrounding ecosystem. She and her team have now taken that concept and applied it to corn plants in a field.
The goal is to take the established ability to simulate a multitude of plant trait combinations — in this case corn trait combinations that do not currently exist in the varieties available — and investigate how those trait combinations will perform in future climate conditions. For those prediction models, they can identify the trait combinations that seed producers may want to start breeding for.
The first step in this research is to develop climate models and predictions for future climate changes.
Simulating the Seasons
“We started with meteorological station observations, actual measured meteorology,” said Swann. “We did that to get as good of information as possible of the correlation between different environmental variables.”
For example, the collected data included how much sunlight was coming at a specific time, what temperature the sunlight was coming in and what the humidity was at that time.
By starting with measured data, Swann’s team was able to remove some of the biases that can come in when developing weather models. They then were able to start adjusting the weather model for known climate change conditions.
“We stuck to the things we think we know well; we know that temperatures will increase,” she says. “Then we made simple assumptions about other environmental variables.”
They took the present-day observations and increased the temperatures. They also explored possible reductions in rainfall, staying consistent with other climate prediction models.
