Spectral Characters Analysis Of Snow In Snovmelt Period In The Northern Slope Of Tianshan Mountains

Snow is covering snow layer formed above the ground or ice, it is one of themost sensitive elements of the cryosphere, having a significant impact on globalatmospheric and ocean thermal conditions as well as regional climate. Subtle changesof the Earth's surface albedo will affect the land-air system energy balance, therebycausing climate change. As the most sensitive elements of the cryosphere, the changefor our country, especially the western region's economic construction and people'sdaily lives is more extensive and direct. Simulation of the snowmelt runoff, itsdirected impact on the accuracy of the results.Simulation of the snowmelt runoff, snow albedo its main impact factor, its directimpact on the accuracy of the results. Thus, the snow spectral characteristics isparticularly important, reflecting properties of the snow surface and snow particle size,density, moisture and other physical properties will affect its albedo, albedo changeswith these physical properties of snow changed. Therefore,based on the study of theabsorption characteristics of snow reflectance and snow physical characteristicsprediction model.evolution snow and the Calibration of spring snowmelt floodmodel will have important practical and theoretical value.In this study, the use of spectral changes proceed from the actual determinationof the snow spectrum, the spectral characteristic parameters, the spectral absorptionindex means and methods of analysis of the characteristics and changes in thecharacteristics of the snow spectrum, and snow spectra of principal componentanalysis, analysis of the band contribution rate and the snow spectral curves and snowphysical properties of the correlation analysis to select the most sensitive spectralband of snow physical properties (grain size, density, moisture content), theestablishment of consolidated snow physical properties prediction model using themeasured snow physical properties data for the prediction model validation andanalysis of results showed that:(1) snow spectral curve of580nm,800nm,900nm, 1025nm,1250nm,1500nm,2000nm,2250nm feature point parameters, characteristicpoints of the trend slope consistency high, the same trend of change in slope betweenthe different samples, and remained horizontal.(2) The curve of spectral absorptionpeak absorption position at the bottom of the trough wavelength reflectance,absorption width, depth, SAI have shown good approximation, minor changes in theabsorption index at1025nm,1255nm wavelength position of the sample spectralcurve about1in the1500nm near the absorption index average of about10.(3) Theanalysis shows that the daily variation of measured snow spectra: the snow spectralchanges with time, a gradual decline in reflectivity. Observation, the day11:00snowreflection rate of the highest,15:00reflectance minimum value of the reflectivity ofthe snow the day, the albedo17:00rebounded, as the snow freezing and thawingobservation provides a new idea.(4) sunny snow density forecast model of complexdetermination coefficient (R2) was0.774, shady snow density prediction model ofcomplex determination coefficient R2is0.866, sunny snow water content predictionmodel complex determination coefficient R2of0.866, shady snow water contentprediction model complex determination coefficient R2is0.879, the snow grain sizeprediction model complex determination coefficient R2of0.802, measured snowphysical properties data verification and analysis on the prediction model, theprediction model NS efficiency coefficient is very high, the results indicate that: ableto get a better prediction model on the forecast of snow physical properties.