In the radiometric sensing of soil moisture through a forest canopy, knowledge of canopy attenuation is required. Active sensors have the potential of providing this information since the backscatter signals are more sensitive to forest structure. In this paper, a new radar technique is presented for estimating canopy attenuation. The technique employs details found in a transient solution where the canopy (volume-scattering) and the tree-ground (double-interaction) effects appear at different times in the return signal. The influence that these effects have on the expected time-domain response of a forest stand is characterized through numerical simulations. A coherent forest scattering model, based on a Monte Carlo simulation, is developed to calculate the transient response from distributed scatterers over a rough surface. The forest transient-response model for linear copolarized cases is validated with the microwave deciduous tree data acquired by the Combined Radar/Radiometer (ComRAD) system. The attenuation algorithm is applicable when the forest height is sufficient to separate the components of the radar backscatter transient response. The frequency correlation functions of double-interaction and volume-scattering returns are normalized after being separated in the time domain. This ratio simply provides a physically based system of equations with reduced parameterizations for the forest canopy. Finally, the technique is used with ComRAD L-band stepped-frequency data to evaluate its performance under various physical conditions.