Remote Sensing Monitoring Of Typical Pests And Diseases In Forestry

If we cannot get effective control after the forest suffers from pests and diseases,Pathogens may to be a source of infection mostly and further damage to the forest can not be estimated. Accurately and timely monitoring of forest pests and diseases is fundamental to control the diseases. However, traditional monitoring methods are generally conducted by pest experts through artificial ground sampling. It isn't reflect the occurrence of pests and diseases timely. It is unable to meet the needs of forest pest monitoring. The remote sensing just can make up the shortage of artificial investigation.It has the advantages of the macroscopic nature, obtains information fast, short repetition period and low cost. In this paper, we use Landsat OLI 8 image and forestry ground survey data to monitor typical forest pests and diseases(Lophodermium piceae and Pinus armandii funorea Desm) of LiXian and Maoxian. The main work of this paper is as follows:(1)First, the spatial distribution of the research tree species in the study area was extracted. This is to avoid the impact of other features on the monitoring results. Then,according to the specific situation of the study area, we choose the time series model.This is use the two phases of the study area at different times image(before and after the disaster). A one-way regression model between the pest and the NDVI rate of change was established based on the ground survey data. The correlation coefficient of the model is R = 0.860, R2 = 0.740 and the model accuracy is 82.61%. This model basically meets the needs of this paper. Based on the established model, the entire study area was monitored and analyzed for pests and diseases. The results are basically consistent with the occurrence of pests and diseases. This paper uses the ground investigation of the victim data to verify the situation. The results show that the monitoring results are in good agreement with the actual situation of the ground in spatial distribution and victimization.(2)According to the pathogen of life activities which based on Lophodermium piceae in Lixian, we select annual average speed annual sunshine hours, annual average temperature, annual average relative humidity and annual total rainfall as meteorological factors. The relationship with meteorological factors and area of spruce disease was analyzed by stepwise regression analysis.The result shows that the annual average temperature and wind speed have great influence on spruce disease. Moreover,using two factors to establish regression model can estimate damaged area of spruce.Based on the analysis of stand, soil and topographic factors by Canonical Correspondence analysis(CCA) method, it indicated that the damaged degree is consistent with ash land, altitude and soil layer. The results show that the general rule of the distribution trend. Moreover, the damage is more serious if the altitude lower, the soil of ashes higher and thinner. These influencing factors can reveal the general occurrence regularity of pests and diseases. It is of great reference value to guide the prevention and control of damage and provide the theoretical basis for effective control of damage.