Abstract: The peak pressure estimation methods of field measurement-based wind loads on low-rise buildings were investigated for the design of wind resistance on low-rise buildings. Based on pressure data measured on the roof of the test building during monsoon climate at the wind engineering field measurement base constructed on seashore beside Pudong International Airport of Shanghai by the State Key Laboratory of Disaster Reduction in Civil Engineering of Tongji University, including a full-scale low-rise building with adjustable roof pitches and two anemometer towers respectively with a height of 10 and 40 m, the distribution characteristics of wind pressures were analyzed, and the validity of the peak value estimation method to use the extreme pressure of short time interval samples to estimate that of long time interval samples from field measurement data was discussed. The research result shows that as for the flow separation region, the probability distributions of wind pressure coefficient have non-Gaussian features and a good agreement with the Gamma and lognormal distributions. When the time-history curve of wind pressures on the roof are divided into several sub-sections and the relationships of typeⅠEVD (extreme value distribution) parameters between parent section and sub-sections are used to estimate peak pressures, the estimated peak pressures are more accurate than observed extreme values, and the optimal length of a sub-section can be determined by the auto-correlation analysis of the parent section. For the full-scale data the extreme value distribution has a good conformity with type Ⅰ EVD when sub-section data are divided by the optimal length, and the peak pressure estimation method based on type Ⅰ EVD for short time interval samples has reliable results.