GM pollen: it gets around

Rebecca Tyson has developed a mathematical model to determine the appropriate isolation distance between genetically modified and conventional crops.

“Numbers tend to get thrown around and until our paper came out there wasn’t a lot of very good research saying what that distance should be,” said the associate professor of mathematics at the University of British Columbia.

Tyson was challenged by W. David Lane, an Agriculture Canada biologist working at Summerland, B.C., to see if she could devise a way to scientifically determine how far GM pollen from apple trees travels.

Her light bulb moment occurred when Lane’s son mentioned that GM pollen and regular pollen were like hot and cold.

“You model movement of heat by the diffusion equation and I knew how to do that,” said Tyson.

Other stories in The 2017 Innovation Issue:

She knew the pollen drift study would be similar to determining how far embers travel from a fire.

What she needed was data to plug into the model and Lane supplied her with all she required.

For two years, he had conducted an experiment where he had a row of GM apple trees planted in an orchard. He harvested seeds from the surrounding trees and figured out how many of them carried the transgene.

Tyson plugged the data into her fractional diffusion model and was able to calculate a safe distance to plant a conventional orchard next to a GM apple orchard with a cross-pollination rate of less than 0.9 percent.

It depends on the relative size of the crops. If the conventional crop is two times the size of the GM crop, the ideal distance is 51 metres. Crops of identical size should be separated by 72 metres and if the GM crop is twice as big as the conventional crop the safe distance is 88 metres.

“Our results suggest that separation distances of several hundred metres proposed by some European countries is unnecessarily large but separation by 40 metres is not sufficient,” she said.

Lucy Sharratt, co-ordinator with the Canadian Biotechnology Action Network (CBAN), said it’s great to have a new tool for establishing buffer zones but she thinks there is a missing ingredient.

“Certainly, we would hope that if there’s to be a discussion of appropriate buffer zones for the GM apple, for example, that local farmers would also be consulted,” she said.

Growers have unique knowledge of things like the density and types of orchards, as well as bird and bee populations in the area. The model should also take into account unpredictable things like human behaviour and equipment use.

“It’s tempting to think that we could map out biodiversity in an area and predict behaviour but our experience shows us that contamination is particularly an issue because of unpredictable events,” said Sharratt.

She said the Canadian Food Inspection Agency has established buffer zones for field testing various types of GM crops, but there are no buffer zones required once the crop is commercialized.

Tyson’s model also provided a mathematical answer to a conundrum that had previously puzzled biologists.

Put a GM plant in the midst of a lot of conventional plants and see how far the pollen travels and the answer is not very far at all.

But put a GM plant all by itself and place a conventional trap plant a long distance away and, lo and behold, the pollen spreads all the way to the trap plant.

“Mathematically, we were able to show that these are consistent,” said Tyson.

That’s because pollen spread is like a bell curve. There is very little pollen at the tail end of the curve and when that tail is in the midst of a field full of conventional plants the GM pollen can’t be detected because there is so much conventional pollen around.

But once outside the conventional plot the conventional pollen drops off quickly as well and suddenly the amount of GM pollen goes up proportionally and can be detected.

So far, the model has been used only for GM apple orchards.

The Arctic apple, non-browning Golden Delicious and Granny Smith apples developed by Okanagan Specialty Fruits, received Health Canada approval in 2015.

Groups like CBAN and some British Columbia orchardists protested the field-testing of the Arctic apple in Canada.

So, the company established orchards in the U.S. and began test-marketing its sliced apples in U.S. grocery stores this year.

Tyson said her mathematical model could be adapted to determine isolation distances for other GM crops such as canola and alfalfa but the model would have to be tweaked because apple trees rely strictly on bees for pollination, while canola pollen can be spread by wind as well.

About the author


Stories from our other publications