Patterns in Arctic Vegetation
Problem being addressed
Climate change has the potential to severely impact the types of vegetation that grow in the Artic region. How can statistical methods be used to understand what the Arctic will look like in the 2050s?
Climate warming is expected to impact the Arctic through multiple feedback loops that go between atmosphere and biosphere. These loops are expected to change the composition of pan-arctic vegetation types and to disrupt existing ecology and biomes. In this research, machine learning methods are applied to predict the changing composition of the Arctic biosphere. It is anticipated, based on the analysis that over half the composition of the vegetation will shift to a different physiognomic class, and that approximately 52%of the Arctic Tundra will shift to woodland ecosystems. Further, it is expected that since this vegetation will change the albedo (reflected sunlight) profile of the Arctic, that the change will create a positive feedback loop that exacerbates climate warming.
Advantages of this solution
This research provides key insight into the changes that will be brought about by climate change in the Arctic.
Solution originally applied in these industries
Possible New Application of the Work
This research looks at how changing external constraints affects the composition of an environment. In the energy sector, we can anticipate that regulatory effects will play a part in changing the composition of the sector as climate issues move to the fore. Perhaps similar research could be used to understand the impact of regulations on energy.
As pressure for local food production rises, and there is less of an emphasis on food that has been transported long distances, it can be expected that the variety of food available to consumers will change. It would be interesting to understand, and map out, the impact of rising transport costs on food production and variety as carbon emissions embedded into products becomes more of an issue.
Warming climate also raises concerns for the types of diseases that can survive and be spread globally. The range and variety of waterborne and airborne diseases in developing world could increase; and from an epidemiology perspective machine learning would be a useful tool for understanding the distribution of and spread of certain diseases (e.g. malaria) in the next decades.
Source DOI: #############
Source URL: #############