中文摘要：

首先利用光学遥感数据识别新疆玛纳斯河流域山区积雪时空分布信息,然后利用C波段合成孔径雷达图像分析非积雪期、积雪期、融雪期的积雪和无积雪覆盖地表在不同下垫面类型、不同局部入射角、不同极化条件下的后向散射系数差异、后向散射系数变化情况和干涉相干性差异,结果表明：（1）在积雪期,积雪与无积雪覆盖地表后向散射系数相近,在融雪期,积雪覆盖地表后向散射系数比无积雪覆盖地表低5~10dB;（2）从非积雪期到积雪期,HH、VV极化的后向散射系数变化较小,HV、VH极化的后向散射系数降低2~4dB;（3）从积雪期到融雪期,积雪覆盖地表HH、VV极化的后向散射系数降低约2dB,HV、VH极化的后向散射系数变化不明显,无积雪覆盖地表HH、HV、VH、VV极化的后向散射系数增加2~3dB;（4）从非积雪期到融雪期,积雪覆盖地表HH、HV、VH、VV极化的后向散射系数降低约2dB,无积雪覆盖地表HH、VV极化的后向散射系数增加1~2dB,HV、VH极化的后向散射系数变化不明显;（5）HH、VV极化方式下积雪覆盖地表相干系数明显低于无积雪覆盖地表.SAR（Synthetic Aperture Radar）图像表征分析结果对积雪及其物理状态的雷达识别研究提供科学依据.

英文摘要：

The optical remote sensing data is used to discriminate snow cover information in mountainous areas of Manasi River Basin in this study.Then by using C band synthetic aperture radar（SAR）images acquired in snow-free,snow-accumulation and snow-melt period,the backscattering coefficient,backscattering coefficient variations,interferometric coherence were analyzed combining with underlying surface type,local incidence angle and polarization state.Analysis results indicate that：（i）There is no backscattering coefficient difference between snow-free areas and snow-covered areas in snow-accumulation period.The backscatteri1 ng coefficient of snow-covered areas is 5~10dB smaller than that of snow-free areas in snow-melt period.（ii）From snow-free period to snow-accumulation period,the backscattering coefficient in HH and VV polarization showed little variation and 2~4dB reduced in HV and VH polarization.（iii）From snow-accumulation period to snow-melt period,the backscattering coefficient of snow-covered areas reduced about 2dB in HH and VV polarization.However,the variation for HV and VH polarization was not distinct.（iv）From snow-free period to snow-melt period,the backscattering coefficient of snow-covered areas reduced about 2dB in HH,HV,VH,and VV polarization,whereas backscattering coefficient of snow-free areas increased 1~2dB in HH and VV polarization.The variation for HV and VH polarization was not distinct.（v）The interferometric coherence of snow-covered areas was significantly smaller than snow-free areas in HH and VV polarizations.The analysis results about characterization of SAR image provide foundations for snow cover extraction and snow physical state identification by SAR data.