Blind Super-Resolution aims to reconstruct high-resolution images from low-resolution images degraded by complex and unknown factors. However, existing methods assume that images suffer uniform degradation, exhibit homogeneous content complexity and ignore varied interactions between various degradation and content, which limit the performance and adaptability to real-world scenarios. To address these issues, we propose a Multi-Degradation Aware Transformer model that leverages wavelet transforms to map degradation into the frequency domain, to distinguish regional degradation and content complexity to achieve degradation-content adaptive decoupling. MDAT adopts a dual-strategy framework that simultaneously enhances detail recovery in high-frequency bands and improves global structure reconstruction in low-frequency bands. The complex degradation and image content are processed within a unified stage, which is allowing the model to adapt its capacity based on the complexity of the degradation and content. Experimental results demonstrate that MDAT performs exceptionally well in BSR tasks, significantly improving robustness and reconstruction quality, especially in heterogeneous degradation scenarios. MDAT effectively handles both simple and complex degradation, which is making it highly adaptable to diverse degradation cases.