Remote Sensing Inversion And Spatiotemporal Variation Of Black Carbon And Snow Grain Size In Typical Snow Regions Of China Based On MODIS Data

Snow cover is one of the natural elements with high surface reflectivity and an important part of the global energy and hydrological cycle.Its spatial distribution and characteristics are very sensitive to climate change.Changes in black carbon concentration and snow grain size in surface snow can significantly reduce snow albedo,resulting in the absorption of solar radiation by the snow layer,which in turn has a feedback effect on the regional hydrological cycle and climate change.Quantitative assessment of black carbon and snow grain size in seasonal snow using remote sensing technology can obtain the change of snow black carbon concentration and snow grain size of continuous systems in space and time,which also provides the necessary and accurate input parameters for many climate models and hydrological models.Based on the reflectance data of the 3(0.47 ?m),2(0.86 ?m)and 5(1.24 ?m)bands of MODIS data,this paper uses the SGSP(Snow Grain Size and Pollution amount)algorithm developed by the AART(Analytical Asymptotic Radiative Transfer)model to retrieve the black carbon concentration and snow grain size of the three stable seasonal snow regions in China and attempts its application in mid-latitude seasonal snow regions.The products of snow black carbon concentration and snow grain size in China from 2000 to 2018 are generated,and the accuracy of the products is evaluated based on the ground observation data.Based on the products,the temporal and spatial variation characteristics of black carbon concentration and snow grain size in typical snow cover areas in China are analyzed.The results show that:(1)Based on MODIS data,the SGSP algorithm can be used to invert the snow black carbon concentration and snow grain size in the three major snow regions in China.The inversion results show that it has high precision in pure snow pixels and can obtain the snow black carbon concentration and snow grain size products with a long-term series and a high spatial and temporal resolution.The verification results show that the correlation coefficient between the retrieved snow black carbon concentration and the measured snow black carbon concentration is 0.83,and the root mean square error and the average absolute error are 116.24ng/g and 100.45ng/g,respectively.The correlation coefficient between the retrieved snow grain size and the measured snow grain size is 0.78,and the root mean square error and the average absolute error are 40.23?m and 33.91?m,respectively.Therefore,the inversion results of the SGSP algorithm have a good correlation with the measured results,the inversion accuracy is higher,and the obtained products are of higher quality.(2)Based on the analysis of MODIS long-term series snow black carbon concentration and snow grain size products,it is shown that the overall snow black carbon concentration in China shows a downward trend from 2000 to 2018,with an average annual value of 600 ng/g to 800 ng/g,a decrease of 3.44 ng/g and a cumulative average annual black carbon concentration of 694.27 ng/g.The black carbon concentration of snow in Northern Xinjiang and Qinghai Tibet Plateau shows no significant upward trend,while that in Northeast China shows a significant downward trend,with an average annual decrease of 10.51 ng/g.The snow black carbon concentration varies from month to month in the winter snow period,with the highest snow black carbon concentration in December and the lowest snow black carbon concentration in March.In terms of spatial variation,the areas where snow black carbon concentration has decreased are mainly distributed in the Gurbantunggut Desert in the southwest of Altay in northern Xinjiang,the Junggar Basin,the western marginal area of the Qinghai-Tibet Plateau,and the industrial capitals of Harbin,Changchun and Shenyang in the northeast in economically developed regions,the reduction is 10 ng/g ~ 50 ng/g.The increase areas are mainly concentrated in the Tianshan economic belt in northern Xinjiang,Qiqihar in Heilongjiang in the northeast,and northern Jiamusi,with an increase of 10 ng/g to 20 ng/g.In addition,using atmospheric aerosol optical thickness data,nighttime lights data,wind speed and other auxiliary data to explore the possible causes of black carbon concentration distribution and changes in snow,it was found that snow black carbon is affected to a certain extent by population,economic development level and atmospheric stability.In northern Xinjiang and northeastern regions,affected by local pollution sources,the high-value areas of snow black carbon concentration are mainly concentrated in densely populated and industrially developed urban belts,while the Qinghai-Tibet Plateau has high altitudes and sparse population,the snow black carbon concentration is mainly affected by atmospheric stability,and black carbon pollution mainly comes from emissions from low altitude areas outside the plateau.(3)Based on the analysis of MODIS long-term series snow black carbon and snow grain size products,the annual average snow grain size in China fluctuates from 120 ?m to 130 ?m in 2000-2018,with an average value of 124.67 ?m,an annual reduction of 0.04 ?m and a coefficient of variation of 0.019.Those results show that there is no significant change in the annual average snow grain size in China,and the change trend is stable.The change of monthly average snow grain size shows that the cumulative monthly snow grain size in December is the largest,while it in March is the smallest.The spatial variation range of snow grain size is 0 ~ 10 ?m.In most areas,the snow grain size tends to increase,and the distribution of change rate are not obvious.