Multi-scale analysis of single-point wind signals in outdoor near-surface environments

The odor/gas dispersion in atmospheric boundary layer is dominated by airflow. The analysis of the wind speed/direction properties is helpful for understanding the mechanism of odor/gas dispersion. In this paper, twenty groups of wind speed/direction time series are collected using ultrasonic anemometer. The wind speed statistics (including variance, turbulent intensity, skewness, kurtosis and turbulent kinetic energy) and wind-direction changes on different time scales are analyzed based on multi-resolution wavelet transform. The results indicate that intermittence of wind speed occur on small scales; the variance of wind speed and turbulent kinetic energy first rises and then decays with an increase in time scale; the probability-density distribution of wind speed is almost symmetric on each scale; the consistency and the standard deviation of the wind direction are mainly affected by large scale motion; and the saltation of wind direction is caused by small scale motion. In use of the wind information, it´s better to consider the distinct properties on different scales, or extract the information on particular scales to eliminate the disturbance on other scales.