1. How many historical scenarios should a forecast be compared against to demonstrate a high level of confidence in predicting disruptions?*
(Choose as many as you like)
  1. How accurately should a forecasting model be able to reproduce historic trends in order to give you confidence that future predictions will be useful?*
e.g. If a population forecast can approximate the changing global demographics over the past 20 years, and whether these populations are increasing or decreasing overall, is this accuracy sufficient (in your opinion) to use as a basis for future decisions on population-related environmental policies (such as legislation on resource consumption)?