Recognition of shelterbelt continuity using remote sensing and waveform recognition

Shelterbelt systems play an important and long-term role in ecological stability and food security. Today, shelterbelts are threatened by anthropogenic destruction, plant diseases, and insect pests. Determining the continuity of shelterbelts is essential for the management and maintenance of complete and stable shelterbelt systems. In this paper, we acquired SPOT 5 images with a 10 m × 10 m spatial resolution, and, based on a GIS method, we generated belt waveforms using a waveform data collector. We then developed a waveform recognition model for belt continuity based on waveform recognition theory. Finally, we identified a minimum threshold value for the identification of belt gaps, and conducted belt continuity recognition in the study area. The results were compared with data calculated from aerial images with a 0.5 m × 0.5 m spatial resolution. When belt gaps were <10 m, recognition was poor, but when belt gaps were between 10 and 20 m, the mean absolute error was 3.8 %, and when belt gaps were >20 m, the mean absolute error was 2.7 %. We propose that this method provides a powerful new tool for the investigation and management of shelterbelt systems.