Linear mixed effects models for the effects of year, fire (two levels), grazing (3 levels) and topography (4 levels) on grass height across the three surveys.

The models assumed a normally distributed response with grass height log-transformed. Transect was treated as a random effect in all models and therefore the constant model included transect as a random effect. K is the number of parameters in each model including the intercept and residual variance, Delta AIC is the difference between each model and the model with the lowest AIC. The model with the lowest AIC is the best model.