Epidemiological studies have suggested that particulate matter (PM) pollution seriously affects human health, particularly it is closely associated with respiratory diseases. The aim of this study is to quantitatively evaluate the effect of PMs (PM₁₀ and PM_2.5) on emergency room (ER) visits for respiratory diseases in Lanzhou, a valley basin city in northwest China. Based on the data of the ER visits, daily concentration of particulate matters and daily meteorological elements from January 1, 2013, to July 31, 2017, we used a generalized additive model (GAM) of time series to evaluate the exposure–response relationship between PMs and respiratory ER visits. Seasonal modified effects of PM_2.5 and PM₁₀ on different age and gender groups were also performed. Results showed that the highest incidence of respiratory diseases occurred in winter. Respiratory ER visits for the total were significantly associated with PM_2.5 (at lag 0 day) and PM₁₀ (at lag 3 days), with relative risks (RRs) of 1.042 (95%CI: 1.036 –1.047) and 1.013 (95%CI: 1.011–1.016), respectively. Effects of PM pollutants on respiratory diseases are different among different age and gender groups. Children under 15 years and the elders over 60 years were the most sensitive to PM pollution, and males were more sensitive than females. The results obtained in the current study would provide a scientific evidence for local government to make policy decision for prevention of respiratory diseases.