Models have been developed to predict disturbance caused by off-road vehicles in a previous investigations. However, the effect of input data on model quality and performance and the adequate structure of models have not been investigated. To improve model quality and performance, a study was conducted. In the experiment, vegetation cover and disturbance intensity, two types of model structure, and three subsets of data are considered and tested. R-square and residuals from validation data are used to represent model quality and performance, respectively. Analysis of variance (ANOVA) is used to identify factors which have significant effects on model quality and performance. The results show that subsets of data have significant effect to both model quality and performance for both measures of disturbance. Although model structure is not a significant factor based on ANOVA tests, models containing interaction terms increase R-squares for all tested conditions and the maximum improvement can reach 30%.
See more of Symposium---Twenty Years of Military Environmental Research
See more of A02 Military Land Use & Management
See more of The ASA-CSSA-SSSA International Annual Meetings (November 6-10, 2005)