Backscattering in forests, how does it interact?

I study the relationship between backscatter and biomass in temperate forests of the Araucanía region in Chile. On the saturation point of the L band I have obtained correlations (r2), from 0.6 - 0.65 …, forests between 300 and 500 Mgha-1. They are forests of Araucaria and Nothofagus with different degrees of degradation.

I expected the L band to be completely saturated on that volume, but I have a different result.

Query: Araucaria forests, composed trees with different architecture and degraded forests allow a greater sensitivity of the L band, over the saturation limit for this band?

I would appreciate any comments and if you can recommend a study to find out how the radar signal interacts with the forests, especially the degraded forests.