Leaf functional traits are important bridge between plants and their external environment. They are not only indexes to explore the adaptation of plants to the environment and to study global climate changes, but also important input variables for eco-hydrological models. We selected dominant shrub species Dasiphora fruticosa and Caragana jubata in the middle section of Qilian Mountains as our study objects. The change characteristics of leaf functional traits were studied by collecting lead sample in high time resolution. Results showed that the leaf area (LA) changed significantly from early June to early July 2020, The specific leaf area (SLA) of Dasiphora fruticosa and Caragana jubata was between 76-157 cm²/g and 120-217 cm²/g, respectively. The mean SLA of Dasiphora fruticosa (123 cm²/g) was less than that of Caragana jubata (176 cm²/g). The leaf area index (LAI) of Dasiphora fruticosa was significantly higher than that of Caragana jubata during the growing period. The changes of leaf area index of them with time was optimally expressed by the cubic function. The optimal relationship between biomass of shrub and the product of base diameter and height was power function. Expressions of aboveground biomass per plant and leaf functional parameters in different growing phases were different, the optimal model selected between aboveground biomass per plant and leaves biomass was S function for Dasiphora fruticosa and power function for Caragana jubata. Aboveground biomass per plant of Dasiphora fruticosa with of the SLA in different stages could be simulated by logarithmic function, while other leaf functional traits could be simulated by cubic functions.