Accurately quantifying methane emissions from cities and understanding the processes that drive them are important for reaching climate mitigation goals. Methane emissions from New York City metropolitan area (NYCMA), the most populous urban area of the United States, have consistently been underestimated by emission inventories compared to aircraft and satellite observations. In this study, we used continuous rooftop measurements of methane over 6 winter-to-spring transitions (January–May, 2019–2024) to examine the variability of city-scale methane enhancements (ΔCH₄) and estimate methane emissions from the NYCMA. We found large variability in the 10-day mean observed ΔCH₄ (~50–250 ppbv) and monthly afternoon methane emissions rates (10.1–30.4 kg s^–1) within and between the years of our study period. A recently released high-resolution regional methane emission inventory developed for the NYCMA performed better than other global and national inventories against the rooftop observations but still underestimated methane emissions, especially in winter. The estimates of methane emissions correlated with those of carbon monoxide (CO) emissions, determined from coincident measurements, suggesting a common city-scale incomplete combustion source for both methane and CO. Our analysis of these continuous measurements also implies a consistent diurnal cycle in urban methane emissions from the NYCMA, which reveals a potential bias in traditional afternoon-only approaches in this domain. This work highlights the usefulness of a long term, multi-species approach to constrain urban greenhouse gas emissions and their sources.