The Farquhar–von Caemmerer–Berry (FvCB) model of photosynthesis has been widely used to estimate the photosynthetic C flux of plants under different growth conditions. However, the seasonal fluctuation of some photosynthesis parameters (e.g. the maximum carboxylation rate of Rubisco (Vcmax), the maximum electron transport rate (Jmax) and internal mesophyll conductance to CO2 transport (gm)) is not considered in the FvCB model. In this study, we investigated the patterns of the FvCB parameters during flag leaf development based on measured photosynthesis–intercellular CO2 curves in two cultivars of winter wheat (Triticum aestivum L.). Parameterised seasonal patterns of photosynthesis parameters in the FvCB model have subsequently been applied in order to predict the photosynthesis of flag leaves. The results indicate that the Gaussian curve characterises the dynamic patterns of Vcmax, Jmax and gm well. Compared with the model with fixed photosynthesis parameter values, updating the FvCB model by considering seasonal changes in Vcmax and Jmax during flag leaf development slightly improved predictions of photosynthesis. However, if the updated FvCB model incorporated the seasonal patterns of Vcmax and Jmax, and also of gm, predictions of photosynthesis was improved a lot, matching well with the measurements (R2 = 0.87, P < 0.0001). This suggests that the dynamics of photosynthesis parameters, particularly gm, play an important role in estimating the photosynthesis rate of winter wheat.