Abstract: Numerical simulations to study the concentration of particulate matter (PM, both PM_2.5 and PM₁₀) play a crucial role in recognizing the spatiotemporal characteristics and further adopting emission control strategies. In this paper, emissions of various sectors such as industry, power, agriculture, residents and transportation in Gansu Province were calculated based on combination of the MEIC inventory and the second national investigation on pollution emissions, and the spatiotemporal characteristics of pollution emissions were subsequently analyzed. Then, WRF-Chem simulations were conducted to obtain the PM concentrations in Gasnsu Province in January 2019, and the daily outputs were validated against available observations from 33 national control environment monitoring stations. Finally, the spatiotemporal characteristics of PM concentrations were analyzed. The results indicated that: (1) The emissions of SO₂, NO_x, PM₁₀, PM_2.5, VOCs, NH₃ and CO in January were 2.12×10⁴, 2.96×10⁴, 2.97×10⁴, 2.43×10⁴, 3.18×10⁴, 1.27×10⁴ and 3.04×10⁵ t, respectively. The largest pollutant emissions, except for NH₃, occurred in developed industrial areas, such as Lanzhou City and Jiayuguan City. (2) The simulated outputs were in good agreement with the observations, and the correlation coefficients between the modeled and observed PM₁₀ and PM_2.5 were 0.544 and 0.597, respectively. The study revealed that Lanzhou City had higher PM concentration values, followed by Tianshui City and Qingyang City, while Gannan Tibetan Autonomous Prefecture and Hexi Regions had the lowest PM concentrations. These characteristics were attributed to the industry distribution, topographical features, as well as meteorological and diffusion conditions. The results show that the WRF-Chem simulation satisfactorily represent the spatial and temporal characteristics of PM concentration in Gansu Province.