Accurate models of millimeter-wave (MMW) radar ground clutter allow improved evaluation of algorithms that detect and identify military vehicles for smart weapons systems operating in a clutter environment. To this end, an empirical model was developed that uses radar clutter data measured at 95 GHz for desert and European-like environments measured over approximately 6 to 7 week periods. Using low-angle, frequency-averaged MMW radar cross sections per unit area ((sigma) 0) data that were collected in the spring season at Yuma, Ariz. and Grayling, Mich., I evaluated a multivariate linear regressive model of (sigma) 0 based upon least-square estimation using measured environmental parameters as independent variables. The results for the data collected at Yuma indicate that (sigma) 0 values were dependent upon the parameter of chronological time, and moderately dependent upon the parameters of soil moisture content and relative humidity. The results for the data collected at Grayling indicated that (sigma) 0 values were highly dependent on the environmental conditions. The model successfully described refrozen ground and snow, and drying ground conditions using the available environmental data, but melting, transitional, and falling snow conditions were not successfully described. For the entire Yuma data set and for a large portion of the Grayling data set, environmental parameters were identified and incorporated into a linear model that described the variations in (sigma) 0.