The assumptions made by some of the authors were not based on the best data, or were just a little bit too convenient, and had a strong impact on the results.
Dan Kammen