Abstract: Abstract: As a major part of agroforestry ecosystem, windbreaks play an important role in agroforestry ecosystem. Porosity is one of the most important structural parameters and usually used to the study of the wind protection of windbreaks. However, a small amount of investigation data and the high cost limit the extension of the related studies. It is desired to have a rapid and cost-effective method of identifying and mapping structural parameters of windbreaks on a large field scale. With the characteristics of high spatial resolution and repetition observation, remote sensing technology helps to fill in gaps of field investigation. In this paper, the porosity was described and estimated by using high-resolution satellite imagery. First, 11 parameters describing windbreaks structure characteristics were summarized. They were average tree height, average tree crown length, average crown breadth, width of windbreak, aboveground biomass, volume, density of biomass volume, leaf area index, gap fraction, the product of average tree crown length and leaf area index, and the product of average tree crown length, leaf area index and width of windbreak. A variable with the highest correlation with the porosity was selected by using statistical analysis and then the porosity estimation model was established. The results indicated that variable combining the average tree crown length, leaf area index and width of windbreaks was highly related to the porosity and the correlation coefficient was -0.941. The average prediction accuracy of statistical model was 85.413%. First, the average tree crown length, leaf area index and width of windbreaks were estimated by using remote sensing data, respectively. The average tree crown length was estimated by using the first principal component of GF-1 data with the average prediction accuracy of 82.788%. The leaf area index was estimated by using the ratio vegetation index with the prediction accuracy of 84.159%. The width of windbreaks was identified by using the object-based image analysis (OBIA) approach. Then, the porosity was calculated by using these 3 variables obtained from the high-resolution satellite imagery. The estimated results indicated that there was a good agreement between the estimated values and the measured data. Following this method, the porosity of windbreaks was estimated, and at the same time its spatial distribution in the study area was obtained. Generally, the windbreaks with a high porosity have more pores, and it indicates that more wind can pass through the windbreak and wind reduction is less effective. According to this, the porosity as a good indicator was used to evaluate the barrier shelter effect. According to the estimated porosity value, the windbreaks were grouped into 3 types, i.e. dense (porosity less than 20%), medium (porosity ranging from 20% to 50%) and sparse (porosity larger than 50%), and the efficiency of wind protection was evaluated. The result showed that the main type of windbreaks was the medium and its area occupied 60.733% of the total area of the windbreaks in study area. Medium type windbreaks usually have the optimum efficiency of wind protection. This study can provide useful guide to wind protection of windbreaks based on remote sensing technology.