From what I understand, global warming will make growing food a lot harder in areas like the American Midwest because temperatures will become too high for most crops. At the same time, Arctic (and Antarctic) regions will heat up and permafrost will melt. Would areas like Alaska, Siberia, and northern Canada then become suitable for large-scale agriculture? And if so, would this be enough to make up for the lost production in warmer areas? Or are there other factors in those areas that would make farming impractical?
Much of the unfarmed area is the Canadian Shield (shades of red below):
The current surface expression of the Shield is one of very thin soil lying on top of the bedrock, with many bare outcrops. This arrangement was caused by severe glaciation during the ice age, which covered the Shield and scraped the rock clean. The lowlands of the Canadian Shield have a very dense soil that is not suitable for forestation; it also contains many marshes and bogs (muskegs). The rest of the region has coarse soil that does not retain moisture well and is frozen with permafrost throughout the year.
I think the "coarse soil" is Glacial Till.
CLIMATE CHANGE ADAPTATION AND CANADA’S CROPS AND FOOD SUPPLY (2013) (122 pages) discusses warming, precipitation, soils and other factors.
There is definitely potential for a northern expansion of agriculture as a result of changes in heat and moisture in [the western prairies], although most of these lands will remain only marginally suitable for farming because of poor soil conditions.
High resolution version from Arctic tundra is 80 per cent permafrost. What happens when it thaws? - The Narwhal:
Edit, Oct., 2020:
In this map from the study, areas that transition from no current suitability for major commodity crops to suitability for one or more crops are depicted in blue, while currently uncultivated areas that transition to suitability for multiple major commodity crops are shown in red. (Hannah et al, 2020)
KC acknowledged the study looked only at climate as a factor when determining where crops could be grown in the future, and didn't take into account things like soil quality.
Soil scientist David Burton of Dalhousie University, who wasn't involved in the study, says that could have a big impact on how much of the potential new farmland is actually arable. "The soils of Southern Canada have taken 10,000 years to develop," he said. "Rapid change wouldn't necessarily make soils in Northern Canada suitable for production."
Johanna Wandel, an associate professor of geography at the University of Waterloo who edited a book called Farming in a Changing Climate, agrees the soil in much of Canada would likely be unsuitable for growing crops, as it's too rocky or acidic.
It may be beneficial to look at what we know about agriculture during the Medieval Warm Period to start to get an answer to this question. During this time period, c950-c1250, there was an agricultural revolution in Europe and there is some evidence of barley being produced in Greenland by the Vikings.
Granted there may still be some debate about how warm the Medieval Warm Period was, especially in comparison to what many are expecting will happen in the future, but it does seem like there is a good deal of evidence that during the Medieval Period there was a good deal more agriculture occuring at higher latitudes. This doesn't necessarily answer whether or not the arctic areas will be suitable for large scale agriculture in the future, but if history is a guide, addition northern lattitudes with suitable soils will become more productive.