In LUCAS, scenarios describe particular land-use policies to be simulated. As an example, suppose that a natural resource manager in the LTRB would like to determine the impact of not logging any trees for 25 years on the habitat of the Wood Thrush ( Hylocichla mustelina). The scenario is formally defined to use the historical transition probabilities based on existing map layers from 1975, 1980 and 1986 along with the restriction that once a grid cell of land is forested, it will remain forested. For example, the land manager may run LUCAS with 10 replicates for 5 time steps each to simulate the change over 25 years. She can quickly examine the graphical statistics plotted on the screen or more carefully analyze the statistics saved to a SAS [30] file. Other scenarios with different constraints can be investigated and their results compared. In this way the investigator can better understand the effects of potential land-use decisions. The simulation and resulting statistics produced by LUCAS are discussed in Section 3.
To validate the LUCAS model, historical data are compared against the simulated data. Starting with the oldest existing map, the period of time up to the year for which the newest map exists must be simulated. The degree to which the statistics for the simulated and historical land cover layers agree determines the accuracy of the model for this period. Section 5 addresses this validation procedure and explains the results.