Abstract: The footprint of flux is effective for observation of the contribution of turbulent exchange to the atmosphere. It is closely related to the exchange of flux between the atmosphere and ecosystem. However, in practice, the complexity of the underlying surface increases the difficulty and uncertainty in calculating flux. Evaluation of flux footprint can solve the problem of spatial representativeness of flux, which makes it easy to calculate flux exchange. In order to study the variation in flux footprint for complete growth seasons of maize in semiarid area, flux footprint and source area functions were calculated from continuous flux measurements for the period from 1st January 2014 to 31st December 2014 using the eddy covariance system driven by FSAM model. The eddy covariance system was in the maize field in the Experiment Station of Agro-ecosystem in Semiarid Area (ESASA) of Lanzhou University, which is in a hilly area of the Loess Plateau. The flux measurement in the agro-ecosystem in semiarid area was spatially representative. The results showed that the prevailing wind direction in the research area was 90°180°. Wind frequencies in that direction during active growth and non-growing seasons of maize were respectively 59.11% and 55.28% of the total wind frequency. The second main wing direction was 270°360°. In the prevailing wind direction, the upwind footprint tail of the growing season was longer than that of non-growing season. However, the reverse was the case for the vertical upwind direction. The comparison between non-growing season 1 (January 1 to March 31) and non-growing season 2 (November 1 to December 31) showed a stable area of flux footprint for the non-growing season. In the prevailing wind direction, flux footprint decreased from seedling stage to tasseling stage while it increased from tasseling stage to maturity stage of maize. The farthest point of flux source area was significantly affected by aerodynamic roughness. Under stable stratification, source areas were larger than those under unstable conditions. The horizontal upwind range of source areas was 892 m and vertical upwind range of source areas was 35+35 m at 0.9 level under stable stratification. Then under unstable stratification, source area upwind range was 783 m and source area vertical upwind range was 25+25 m. There was no obvious difference between the areas of flux footprint between 90°180° direction and 270°360° direction, suggesting that flux footprint was not closely related with wind direction. Daytime flux source area was smaller than that of night-time, which was caused by different atmospheric stabilities for day-time and night-time. Comparison with other research results suggested that aerodynamic roughness and atmospheric stability influenced agro-ecosystem flux footprint by changing flux source length. Also δv/u* was the main factor driving flux source width. This study suggested that the flux footprint measured with FSAM adequately revealed the characteristics of surface flux in agro-ecosystems in semiarid areas.