The use of military vehicles results in soil disturbance and vegetation loss. While the capacity of installation lands to sustain training is a function of both the sensitivity of lands to specific activities and the natural recovery rates of vegetation, it is also a function of vehicle characteristics, the doctrine which establishes how these systems are used, and actual locations where activities are conducted. Accurate assessment of vehicle impacts is limited by the technical data available. An approach is provided for collecting vehicle impact data to more effectively support impact assessments. The approach combines historic land condition data that quantifies long-term cumulative disturbance patterns and vehicle tracking systems (VTS) that quantify short-term individual vehicle impact patterns. VTS are used to assess short-term impacts by tracking vehicle location and operating characteristics (i.e. turning radius and velocity). Controlled courses are used to establish a range of vehicle dynamic properties. Impacts associated with vehicle use are measured along the vehicle course, and statistical models are developed from the field data to predict vehicle impacts. Vehicle tracking systems are then used to track vehicles during training exercises. Vehicle property and location information are used with statistical impact models to predict the cumulative impact of training exercises. Permanent field transects and remotely sensed images are used to quantify long-term cumulative vehicle impact patterns. A conditional co-simulation algorithm is used to generate time series maps of vehicle disturbance. Vehicle training load is quantified and combined with land condition estimates to predict future vegetation cover values based on projected training loads. The proposed approach utilizing both long-term cumulative and short-term individual vehicle impact assessments provides the foundation for comparing and contrasting alternative land use scenarios.