Research On Glacier Extraction And Long Time Series Change Based On Remote Sensingin Bogda Region,Tianshan

The high-density time-series analysis of glacier area can more accurately grasp the characteristics of glacier change and its fluctuation law,predict the possible changes in the future,and provide more scientific basis for the sustainable use of water resources and the prevention of natural disasters.Based on Landsat image data from 1987 to2019,this study used a new method combining Gaussian Mixture Model and Normalized Difference Snow Index to extract the glacier boundary in Bogda area in 33 years.The ARIMA model is also known as summation autoregressive moving average model,which was used to predict the time series of glacier changes.By reserving data for 10 years from 2010 to 2019 as validation data for ARIMA model prediction,it is concluded that the ARIMA model is reliable and universal in predicting glacier changes.This paper calculates and analyzes the change information of 28 glaciers per year,and obtains the change situation and law of the glaciers in 33 years.The ARIMA time series analysis model is used to predict the change of 28 glaciers in bogda region for 5 years,so as to provide a basis and reference for the water resource utilization and the prevention of natural disasters in this region.The main research work of this paper is as follows:(1)Research on glacier extraction method.In view of the traditional manual selection threshold which is low automation,repeated trial operation,and subject to subjective factors,Gaussian Mixture Model is established for normalized difference snow index of local glacier region by the expected maximum algorithm,and remove the mixed pixel metaclasses in the region.Then,the GMM was used to simulate the NDSI distribution of purified glacier and non-glacier.According to the distribution of the improved gaussian mixture model,the glacier extraction threshold was calculated automatically in the region.The algorithm experiment is carried out in the research area,and then the glacier extraction boundary is compared with the glacier inventory data for verification.The results show that the results obtained by the method are reliable and accurate.(2)Research on intensive time series model analysis.ARIMA model is called summation autoregressive moving average model,which is a time series prediction model.In this paper,the ARIMA model is constructed for the time series of glacier change.By reserving data for 10 years from 2010 to 2019 as validation data for ARIMA model prediction,it is concluded that the ARIMA model is reliable and universal in predicting glacier changes.(3)The temporal and spatial characteristics of glaciers in the study area are analyzed.Based on the time series analysis of the information extracted from 28 glaciers in the Bogda Main Peak Region,the overall change characteristics of glaciers,the change characteristics of glaciers on the south and north slopes,the change characteristics of glaciers of different sizes,the change characteristics of glaciers at different(terminal)elevations,and the space-time change characteristics of three typical glaciers are obtained respectively.ARIMA model was constructed to predict the area of28 glaciers for the next five years,and it was concluded that Bogda glacier would advance slightly in some years,but generally retreat.