Research On Remote Sensing Identification Method Of Shelterbelts Structural Based On ZY-3 Image

Farmland shelterbelt, as one kind of shelter forest, plays an important role in improving the ecological environment and field microclimate, and also offers the potent guarantee for a high and stable crops yield. In resent 40 years, the construction of shelter forest in our country, especially the construction in “Three-north Forest Protection Project”, has made remarkable achievements. But it also faces the problems that the shelter forest is gradually entering into the stage of mature and over-mature. The renewal of the farmland shelterbelt and continuing operation is increasingly important. In addition, with the rapid development of remote sensing and GIS technology, it can be possible to quantify the structure information of the shelterbelts. Therefore, we need to study and master the structure state of the shelterbelts, to evaluate its structure integrity so as to provide important basis for managing and establishing perfect shelterbelt systems.At present, there are lots of literatures about structure, function and management of the shelterbelt. In the early time, the researches was mainly based on the ground experiments and statistical methods. The researchers then gradually apply the remote sensing technique to monitor the shelterbelts. However, it is not enough mature and perfect to compute some the belt parameters. So, this paper carried out recognition method of belt structural information from the belt scale based on the previous basis. The main contents include the following aspects:The paper explored the method of extracting the shelterbelt information by using remote sensing and GIS method based on the main data source of ZY-3 multi-spectral image, which has a resolution of 5.8 m. Then we extracted the belt vector data by using morphology and object-oriented method. The belt fraction was calculated by using the method of the decomposition of mixed pixels based on ZY-3 image. Then it has been verified by the ground experimental data. And, the extraction method of belt waveform was established to extract the belts’ continuity, survival rate, crown width and lines based on the belt vector data and the belt fraction data. Next, we put forward the belt integrity index to evaluate the integrity of the shelterbelt structure based on analysis of the length, continuity and survival rate of the shelterbelt.Finally, according to the integrity of the forest in the study area to get the following conclusion: The shelterbelts which have a good integrity above accounted for 56.29% of the total forest, proportion of more than half; Part of the integrity shelterbelts and incomplete part were 11.31% and 2.67%; The very incomplete shelterbelts accounted for 30.82%. As a result, the overall structure is relatively perfect in the study area, but nearly 31% belts in a very incomplete state which need to be updated or reconstructed.