Scattering from biological entities can be a significant source of reverberation and clutter for active sonar systems operating in deep or shallow water. Resonant scattering from the swim bladders of fish is a dominant source of this biologically induced reverberation and clutter for midfrequency active sonar (1–10 kHz), which can obscure targets, result in false targets, and otherwise overload operators of tactical systems. Though significant questions remain regarding the fisheries oceanography needed to predict clutter and reverberation, the basic physics of scattering from individual fish and groups of fish is well understood. This motivates a data-driven, clutter adaptive approach to mitigation of biologically induced reverberation and clutter that utilizes physical models without requiring unavailable databases of fish-species abundance and local spatiotemporal distributions. In this work, physics-based models relating physical parameters of fish and their spatiotemporal behaviors to spectral properties and spatiotemporal distributions of scattering form the basis of in situ parameter estimation. In this presentation, adaptive processing developed for the early stages of the active-sonar processing chain using these in situ estimates is described, including waveform and matched-filter design and within and across-beam normalization. [Work supported by a NAVSEA Phase I SBIR award.]