Local primary emission, transport, and secondary formation of aerosols constitute the major atmospheric particulate matter (PM) over a certain region. To identify and quantify major sources of ambient PM is important for pollution mitigation strategies, especially on a city scale. We developed two source apportionment methods to make the first‐order estimates of local primary contribution ratio (LCR) of PM2.5 (PM with diameter less than 2.5 μm) using the high‐density (about 1/km2) network observations with high sampling frequency (about 1 hr). Measurements of PM2.5 mass concentration from 169 sites within a 20 km × 20 km domain are analyzed. The two methods developed here are mainly based on the spatial and temporal variations of PM2.5 within an urban area. The accuracy of our developed methods is subject to the assumptions on the spatial heterogeneity of primary and secondary formed aerosols as well as those from long‐range transport to a city. We apply these two methods to a typical industrial city in China in winter of 2015 with frequent severe haze events. The local primary pollution contributions calculated from the two methods agree with each other that they are often larger than 0.4. The LCR range is from 0.4 to 0.7, with an average value of 0.63. Our study indicates the decisive role of locally emitted aerosols in the urban severe haze formation during the winter time. It further suggests that reductions of local primary aerosol emissions are essential to alleviate the severe haze pollution, especially in industrial cities. Key Points Two approaches are developed to quantify the local contribution from primary emissions to haze using high‐density site observations Both methods indicate the importance of local primary emissions in haze formation in an industrial city with contribution ratios of 0.4‐0.7 Local primary emission contribution ratio increases with mean pollution level during the winter of an industrial city