Accuracy of Climate-Change Projection Research Questionable, Study Finds

August 01, 2008 - News Release

Research on the potential impacts of climate change on species may not be a reliable as we think, according to a new University of Guelph study.

Environmental biology professor Jonathan Newman and master's student Anna Mika have revealed that the climate change projections commonly used by researchers can yield contradictory predictions.

Predicting the biological impacts of climate change on individual invasive, endangered and economically important species is often accomplished using a tool called a bioclimatic envelope model. These models give predictions based on the inputed climate change projections.

The climate change projections used are themselves the output of computer models developed for the Intergovernmental Panel on Climate Change (ICCP). There are 31 models under ICCP that can be accessed by researchers and used to study the impact climate change will have on a species.

"These models are the basis on which all research on climate change is done," said Newman. "They are used to make policies and come up with mitigation strategies."

Newman and Mika recently used two such climate models for a research study to determine how climate change over the next 80 years would impact the proliferation of the swede midge in North America. The swede midge is an invasive species and a serious agriculture pest that was first around in the United States in 1996 and has since made its way to southern Ontario.

Despite using two climate models that both predicted the same future climate, they found the biological impacts were vastly different depending on which model was used. Their findings were recently published in the journal Global Change Biology.

"We basically got opposite answers when we should have gotten the same answer," said Newman. "This shows that the work being done around predicting the biological impacts of climate change can vary depending on which particular climate model is being used. It's concerning because a vast majority of the impact research conducted is based on the projections of these models."

Often researchers will only use climate projections from one model when determining the potential impact of climate change on an animal or plant so the discrepancy wouldn't even be known, he added.

The two researchers used projections from both a Canadian climate model and a British model. The impact predictions for the swede midge should have been roughly the same regardless of which model was used to generate the future climate, said Newman. But when the two researchers compared the results, they found they were completely different.

Predictions based on the Canadian model indicated the species would expand across Ontario and into northern and western regions of Canada and northern United States because these areas would provide the warm and moist conditions favourable to the swede midge. Whereas the predictions based on the British model indicated the areas suitable for the insect would significantly shrink.

"We ought to be extremely cautious about research results, particularly if the findings are based on only one model projection, before running off and making policies," Newman said. "The findings can't be considered likely unless they are run through many models."

Jonathan Newman
Department of Environmental Biology
519-824-4120, Ext. 52147

For media questions, contact Communications and Public Affairs: Deirdre Healey, Ext. 56982/

University of Guelph
50 Stone Road East
Guelph, Ontario, N1G 2W1