Air pollution control in the U.S. has evolved into a comprehensive policy system spanning from the federal to the state level over time. A unified quantitative analysis of policy intensity can shed light on the policy evolution across different levels, the influence of partisan and regional factors on policy, and the relationships with emissions of major pollutants. By harnessing the policy text of the Clean Air Act (CAA) at the federal level and State Implementation Plans (SIPs) at the state governments (1955-2020), we deployed a Natural Language Processing (NLP) approach to define a policy intensity index to systematically quantify the U.S. air policy landscape. Our findings highlight that the 1970 CAA amendment carries the most vigorous intensity as it established a holistic control system for the first time. Subsequent years witnessed a general trend of partisan polarization, eventually leading to a graduate convergence between red and blue states. Blue states demonstrated a closer alignment with federal directives and a superior efficacy in pollutant reduction. Regionally, the Northeast displays the highest overall policy intensity and the West exhibits the highest coordination with the federal benchmarks, making these regions outperform others in air pollution control. Our study not only discusses policy implications for air pollutant reductions considering partisan and regional differences but also provides a novel measurement tool to quantify policies for assessing disparities and synergies.