Background, aim, and scope Ice nucleating particles (INPs) play a crucial role in aerosol-cloud interactions and significantly influence cloud microphysical properties. Urban areas, with their high aerosol particle emissions, are recognized as important potential sources of INPs. However, due to the complex and diverse composition of aerosols, the impact of anthropogenic emissions on atmospheric INPs in urban environments remains poorly understood. Quantifying the contribution of specific pollution sources to INP activity presents a major challenge. This study aims to measure the ice nucleation activity and PM_2.5 mass concentration of atmospheric aerosols in Shijiazhuang, a pollution-prone city, to assess the activation of INPs, examine the influence of aerosol chemical composition on INP concentration (N_INP), and to quantitatively evaluate the contribution of different pollution sources to ice nucleation. Materials and methods The PM_2.5 sampling site was located in the yard of Hebei Sailhero Environmental Protection Technology Co., Ltd. in Shijiazhuang City, Hebei Province (38.04°N, 114.65°E). PM_2.5 samples were collected from December 7, 2021 to January 8, 2022, using a flow sampler at a flow a rate of 1.13 m³/min. The collected samples were processed by clipping the membranes and placing them in centrifuge tubes containing 10 mL of deionized water, followed by sonication for 30 min. An aliquot of the suspension was transferred to a cold stage and cooled at 1 ℃/min until complete freezing of the droplets. Carbon fractions (organic carbon (OC) and elemental carbon (EC)), water-soluble ions, and inorganic elemental components were analyzed using a DRI 2001 thermo-optical carbon analyzer, an ion chromatograph, and an energy-dispersive X-ray fluorescence analyzer, respectively. Pollution sources in Shijiazhuang were assessed using the positive matrix factorization (PMF) model combined with multiple linear regression (MLR). Results Throughout the observation campaign, daily mean PM_2.5 mass concentrations ranged from 30.8 μg/m³ to 155.5 μg/m³, with an overall average of (84.0±35.4) μg/m³, exceeding the Grade Ⅱ daily limit of the Chinese National Ambient Air Quality Standard (75 μg/m³). Secondary inorganic aerosols (SIA, dominated by sulfate‑nitrate‑ammonium (SNA: SO , NO , and NH )) and organic matter (OM) accounted for 38.9% and 27.2% of PM_2.5 mass, respectively. The median freezing temperature (T₅₀) of PM_2.5 suspensions averaged (-15.0±1.3) ℃, substantially higher than that of deionized water (-25.0 ℃), indicating the presence of ice-active constituents in ambient PM_2.5. The onset temperatures for ice nucleation ranged from -12.5 ℃ to -6.5 ℃, with a mean value of (-8.4±1.3) ℃ under both clean and polluted conditions. Over the temperature interval from -20.0 ℃ to -6.5 ℃, N_INP exhibited pronounced variability, spanning from 0.0016 L⁻¹ to 0.6721 L⁻¹ with a range covering two orders of magnitude. Correlation analyses revealed that at -15 ℃, N_INP was significantly and positively correlated with the metallic elements Ca (r = 0.48, p < 0.05), Ti (r = 0.47, p < 0.05), and Fe (r = 0.50, p < 0.05). In contrast, water-soluble ions (SO , NO , NH , K⁺, Cl^-), OM, and EC showed no significant correlation with N_INP. Discussion Although PM_2.5 mass concentration did not exhibit a significant relationship with INP activity, marked disparities emerged among individual chemical components in their influences on ice nucleation. The metallic elements Ca, Ti, and Fe emerged as the principal chemical determinants of INP activity at -15 ℃, implicating fugitive dust as a key source of INPs at this temperature regime. In contrast, no statistically significant correlations were observed between water-soluble ions, carbonaceous fractions, and N_INP, likely attributable to their intrinsic chemical properties and diminished ice nucleation efficiency under relatively warm freezing conditions. In the low-temperature region (below T₅₀), N_INP on clean days exceeded that on polluted days, a phenomenon consistent with the elevated mineral dust fraction during cleaner atmospheric conditions. Quantitative source apportionment indicated that fugitive dust constituted the predominant INP source (46.1%), followed by vehicle emissions (31.1%) and industrial emissions (13.4%), with secondary sources making only minor contributions. Conclusions At -15 ℃, fugitive dust (46.1%) was identified as the dominant source of atmospheric INPs, followed by vehicle emissions (31.1%) and industrial emissions (13.4%). These source contributions were closely associated with the metallic tracers Ca, Ti, and Fe. Secondary aerosol sources contributed marginally to INP formation, presumably due to the relatively high subzero temperature, which limits the ice nucleation activity of SNA components. This study provides a quantitative observational basis for understanding the sources of INPs in urban atmospheres and highlights the predominant role of primary particulate emissions over secondary aerosol formation in governing urban INP concentrations. Recommendations and Perspectives This study only examined the influence of individual chemical components on INP concentrations. However, the effects of internal and external mixing states of aerosols on their ice nucleation ability remain unclear, and the microscopic formation mechanisms of INPs in urban aerosols have not been fully explored. Future research should combine laboratory simulations with single-particle analysis to investigate the role of mixing states at the molecular level. In addition, physical properties such as particle size and surface area should be incorporated alongside chemical composition to systematically evaluate the key factors affecting ice nucleation activity. On this basis, a physiochemical parameterization scheme could be developed to improve the representation of urban INPs in regional climate models and enhance the simulation accuracy of cloud and precipitation processes.