A novel method to estimate Q from seismic data based on rock-physics modeling

Abstract

Although the mechanisms causing seismic wave attenuation ([Formula: see text]), the inverse of the quality factor, are always frequency-dependent, most Q estimation methods are still based on the assumption that Q is constant across frequencies. Therefore, existing rock-physics Q models have not been effectively applied to bridge frequency-dependent Q estimations and rock properties such as porosity and fluid saturation. An alternative way of dealing with frequency-dependent Q estimation is to use the mechanical response of viscoelastic models in the log spectral ratio method. Nevertheless, viscoelastic models are phenomenological models that lack physical meaning resulting in errors in Q estimation. Through the process of suitable rock-physics modeling and scaling, a new frequency-dependent Q estimation method, called the two-parameters spectral ratio method, is developed based on the wave-induced gas exsolution and dissolution and differential Kuster-Toksöz approach, appropriate for evaluating attenuation in saturated rocks. Compared with conventional Q estimation methods, our algorithm builds a link between rock-physic models, logging, and seismic data, providing more accurate and reliable Q results for better interpretation of subsurface properties.