| Extracting sea ice space-time distribution information quickly and accurately is a basic issue in the study of sea ice and ecological environment.The rapid development of remote sensing satellite technology provides a well way to solve this problem.The normalized difference snow index(NDSI)is mainly based on the difference in spectral characteristics of snow and other features(high reflectivity at wavelengths 0.53 to 0.59 ?m,strong absorption at 1.57 to 1.65 ?m),and a ratio difference calculation operation.The spectral reflection characteristics of sea ice are similar to that of snow,especially in the blue-violet and blue-wave bands where the reflectance is higher than the green wavelength.In this paper we proposed the new index(ENDSI)based on adding emissivity characteristics of snow and sea ice in 0.433–0.453 ?m and 0.450–0.515 ?m by using Landsat8 OLI datas as the da ta source,according to the spectral characteristics of snow.What's more,we use this new way to extract sea ice informations.In this paper,the concept of ever frozen area is proposed,and the development trend of sea ice since 1973,the spatial distrib ution and Influence of seawater depth on diurnal variation of sea ice was analyzed by using time series,overlay analysis and correlation analysis methods in Bohai.The results show that:(1)Compared with the iced area in single day,ever frozen area is mo re sensitive to temperature and it is more balanced to reflect the actual situation during the year ice,so it can be used as a measure of the interannual variation of sea ice;(2)Since 1973,the sea ice has not weakened in the Bohai,the minimum icing area has remained unchanged,and the largest ever frozen area has been increasing.After 2000,the interannual variation of sea ice intensified,and sea ice is more serious;The probability of annual freezing in 1/2 sea area of Liaodong Bay is over 70%,which is the main icing area in the Bohai Sea.The often frozen area of the Bohai Bay and Laizhou Bay is mainly concentrated in the Southeast side where depth is less than 6m,and The annual probability of annual freezing between 60%-80%.(4)the depth of seawater is an important factor affecting the spatial variation of sea ice in the Bohai Sea.it has a different effect on the four stages of the development of sea ice in Liaodong Bay,This effect causes sea ice spatial changes in the valley line and the valley line of seawater depth tends to be consistent,and slightly different.Liaodong Bay is the most serious region of sea ice disaster in Bohai,and a precise prediction of the sea ice spatial distribution and dynamic change plays an important role in reducing the sea ice disaster risk.In this paper,the freeze information from 2010/2011 to 2014/2015 were extracted using meteorological data and remote sensing data,and we used these informations to calculate the probability of the ice in Liaodong Bay.Finally,the relation model between freezing probability and accumulated temperature of Yingkou meteorological station is established,then,it was used to predict the sea ice spatial distribution and dynamic change.Simulation and case study results show that: this model can forecast the dynamic change of sea ice in Liaodong Bay,and using the confusion matrix to test the sea ice extraction results,the overall accuracy is more than 90%,which provides a reference for the prediction of sea ice under the local conditions in Bohai and other regions. |
PDF Full Text Request