Abstract: To identify truncated lithologic reservoirs associated with angular unconformity, we propose a method that integrates unconformity reflection coefficients obtained through sparse inversion with instantaneous amplitude. This method employs L1-norm constrained sparse inversion, namely, the fast iterative shrinkage-thresholding algorithm (FISTA), to reconstruct high-precision reflection coefficient sequences from band-limited seismic signals, thereby establishing the correlation between reflection coefficients of angular unconformity and truncated steeply-dipping sand packages in underlying formations. Model tests demonstrate that this method significantly outperforms the traditional root mean square (RMS) amplitude method in both noise immunity and the accuracy of identifying truncated sand packages. A case study of LZ9 block in Shengli Oilfield successfully identifies truncated steeply-dipping sand packages in the underlying Kongdian Formation by inverting the reflection coefficients of the angular unconformity and integrating them with the instantaneous amplitude attribute. The reflection coefficient attribute more accurately captures the abrupt changes at sand package boundaries. The distribution of truncated sand packages beneath the angular unconformity, mapped through multi-attribute fusion, shows good agreement with data from 17 validation wells in the study area. This method provides effective support for the exploration and development of truncated lithologic reservoirs beneath angular unconformities.