Plants absorb atmospheric carbon dioxide and convert it to carbohydrates during photosynthesis. This removal of carbon dioxide could be enhanced when we reforest areas that we have cleared of forests in the past. Due to this effect, reforestation has been suggested as a tool to mitigate global warming. Reforestation is a key issue, for example in international negotiations in the United Nations Framework Convention on Climate Change.
However, forests and reforestation affects climate not only by removing carbon dioxide from the atmosphere, but also by other effects. In particular, forest is darker than grass and cropland. So the reflectivity of forest is lower, it absorbs more solar radiation, and therefore often leads to a warming of the land surface.
Previous studies have concluded that in some regions, in particular in the temperate and boreal regions of the world, this warming effect due to the decrease in reflectivity could actually be larger than the cooling effect from carbon dioxide removal. And therefore, reforestation in the boreal and temperate latitudes might not be an effective tool to mitigate the global warming.
Now, these studies have greatly advanced our knowledge that it’s really crucial to include not only carbon dioxide effects but also the other effects of reforestation. And we build on this methodology in our study, but we add substantial geographical detail. Those previous studies performed numerical model simulations in which they always assume that entire latitudinal bands would be deforested or reforested.
But we now ask the question what would happen if we reforest specific locations that have actually been deforested in the past, so they would be available for potential reforestation. And we try to really pinpoint the carbon dioxide and the reflectivity effects to each specific location of past deforestation and therefore potential future reforestation.
The way we do it is that we run a global climate model, which uses a reconstruction of historical land cover change, and then calculate the effects of reflectivity and carbon dioxide changes over time. And we use a measure called radiative forcing to pinpoint these effects to each specific location.
What we find is that the carbon dioxide effects in the past have, of course, led to a warming, as we would expect, because deforestation has led to emissions. The reflectivity effects have indeed led to a cooling, transforming dark forests to bright grass and cropland. But the net effect of the two is that the carbon dioxide effect is stronger than the reflectivity effect.
When we now infer from the past deforestation to future reforestation – and we can assume that the past is a good indicator for future effects of reversing the process – then we would say that the net effect of reforestation in most regions in the world would be a cooling. And this includes the temperate and boreal regions.
Now, this seems to be in disagreement with what previous studies found. The reason for this lies in the assumptions about the vegetation changes in those previous studies as compared to ours. While the previous studies looked at large-scale latitudinal deforestation/reforestation, we looked at historical deforestation. We find that past land use decisions have increased the mitigation potential of reforestation.
In the past, farmers chose to use those forest areas for agriculture that were most suitable for agriculture, and this had two effects. On the one hand, these areas were highly productive, so planting forest on these areas would remove more carbon dioxide than an average forest across this latitude. On the other hand, the reflectivity effects are smaller in those suitable areas for agriculture. The reason for that is that these areas are not in regions of long-lasting snow cover – those are not suitable for agriculture. But snow is a key point in leading to reflectivity changes going from one type of vegetation to another one. So the preferential choice of suitable areas in the past for agriculture has increased the mitigation potential for reforestation.
So the implications of this finding are that, when we conclude from idealized studies to actual reforestation projects in the future, we are likely underestimating the effects in the temperate and boreal regions, and they may indeed be higher than previously thought. This becomes relevant because, in particular in North America and in Europe, agricultural area becomes abandoned and is potentially available for reforestation. And we find that reforestation in these areas may indeed be effective to mitigate global warming, and thus may be useful beyond aspects such as preservation and restoration of natural vegetation.