Sensitivity Analysis

Exploring the relative significance of inputs on the value of outputs is called sensitivity analysis. Systematically altering the inputs and observing changes in outputs allows an analysis of what inputs are causing the most variation in outputs. Large variation in some inputs cause little variation in model results. On the other hand, small changes in other inputs can cause big changes to model outputs. This is why a firm understanding of model inputs such as inventory data, growth and yield data, response to treatments, and succession is important. The steps for carrying out a sensitivity analysis are as follows:

  1. Alter input datasets (e.g. targets) for each factor. A factor could be harvest yield, habitat, or disturbance patch size for example.

  2. Run same sets of analyses. The scenarios are loaded with the same factors, however with some of them adjusted.

  3. Compare results. Each of the factors can be compared with the use of reports and comparison reports.