Retrieval Of Characteristic Parameters Of Vegetation By Remote Sensing And Its Application In Monitoring Oriental Migratory Locust

Oriental migratory locust (Locusta migratoria manilensis Meyen) is the kind of locust which has very often brought the harmful effects to many regions in China. More than 800 locust plagues have been recorded in Chinese historical documents, in which most of the plagues were caused by oriental migratory locust. In recent years, a tendency of more and more serious locust plague has emerged due to the climatic change and human activities.Remote sensing as the modern technology and with the advantages such as large-area covering and undamaged monitoring has become an important means in monitoring locust plague. As the source of food and inhabitancy of locust, vegetation is particularly important for occurrence and development of migratory locust. During the period of locust plague vegetation is suffered greatly from locust damage and this kind of damage will be sensitively reflected in variation of the characteristic parameters of vegetation. Therefore, occurrence and outbreak of locusts can be rather easily detected by monitoring the variation of some characteristic parameters of vegetation.In this thesis, the city of Huanghua in Hebei province was taken as the study area and TM images, the data collected in field were used in the study. Firstly, three kinds of characteristic parameters of vegetation which are very sensitive to locust plague were retrieved from the TM images and the ground collected data. These parameters include the leaf area index(LAI), the characteristic parameters of hyper-spectra and vegetation index(VI),respectively. Secondly, the models to reflect the relations between the characteristic parameters of hyper-spectra and LAI were developed and, based on the models, the locust damage spectra index (DSI) was set up. Finally, the classification of locust damage intensity for the study area was completed based on the DSI data.Through this study the following conclusions were obtained:1) From the point of view of the mechanism of developing the optical models and through quantitative analysis, four indirect methods used for LAI retrieval and based on the gap fraction of vegetation canopy were analyzed and compared. The results show that, among these algorithms, the LAI-2000 algorithm is most suitable for estimating LAI in the study area. In addition, a nonlinear positive correlation exists in the relation between LAI and vegetation coverage.2 ) The correlation analyses were conducted between LAI measured in field and all kinds of VI retrieved from Landsat-5 TM image data. It was found that RDVI is the best index in the total 12 kinds of vegetation index in terms of the ability of indicating vegetation growing condition.