Dendrolimus tabulaeformis disaster monitoring and analysis of its influencing factors through remote sensing technology

Abstract

Considering that a large area of Pinus tabulaeformis suffers from Dendrolimus tabulaeformis in West Liaoning, this study monitored Dendrolimus. tabulaeformis disaster promptly, efficiently, and precisely through remote sensing technology. Information on the influencing factors, such as geography and meteorology, was collected to provide a reference for future prediction. This study adopted the method of Thematic Mapper (TM), Enhanced TM Plus data source, and Geographic Information System technology to examine the image gray histograms of injured Pinus tabulaeformis with various damage degrees and monitor the damage degrees by using the Ratio Vegetation Index (RVI) [near infrared (NIR)/red]. To analyze the imposing factors, the classification results from RVI analysis were overlapped with geography and meteorology statistics, with cross reference to previous studies on the biological characteristics of Dendrolimus tabulaeformis. By analyzing the image gray histograms, this study determined that in the Pinus tabulaeformis spectrum, the NIR band shows high sensitivity to mild infection, whereas the red band shows high sensitivity to severe infection. Therefore, using RVI, which combines the two bands, prompts the disaster monitoring. The monitoring results are in accordance with the biological characteristics of Dendrolimus tabulaeformis, which prefer dry and warm environments, thus proving the efficiency of the remote sensing monitoring. For the factors that influence the damage, Dendrolimus tabulaeformis prefers south gentle slopes. Areas with long sunshine time, minimal rain, and low accumulative temperature suffer from severe damage. This finding provides a basis to predict the probability of disaster outbreak. By applying a novel converse method to test the reliability and accuracy of monitoring results, this study showed that researchers could take advantage of the biological and ecological characteristics of remote sensing objects to assess the reliability of results when forestry damage field investigations are lacking and there isno access to fieldwork sample statistics to estimate the monitoring results. This procedure could reduce the load and difficulty of field investigations.