摘要
积雪作为气候系统中重要的环境影响因素受到了普遍的重视,特别是在高纬度、高海拔地区,积雪面积和积雪厚度是积雪研究的两个重要因子。利用2010年l 1月8日、14日和25日三天的MODIS LlB数据,根据在可见光波段,地表、云和雪在该波段的反射率都比较高的特点,计算得出的积雪覆盖面积空间分布图。积雪深度受降雪量、坡度、坡向等多因素的影响表现为不连续分布,单纯利用气象监测站点无法获得积雪厚度的空间连续分布,而遥感影像也受到局地特征的影响而精度不够。因此,首先基于经验公式,利用MODIS L1B数据计算了积雪覆盖面积,在此基础上,基于气象监测点的积雪深度数据和MODIS LlB波段的相关性,选择影像数据的最优通道,通过建立多波段的积雪深度回归模型,并且对积雪深度进行空间连续插值,通过将模拟的积雪深度与监测点插值结果进行对比,发现统计回归的方法有效的提高了积雪深度、覆盖度等雪情监测信息精度。
Snowcover depth and snowcover data are two important parameters of regional and global climate change and basin-scale water storage in the mountainous areas. Therefore, it is important to ensure that they are accurate and as free as possible of any known biases. Compared with AVHRR, EOS/MODIS data has a higher spatial reso- lution and more channels. It is proved that there is a good liner relationship between snowcover depth and snowcov- er surface reflectance in visible to short-infrared window channels when snowcover has a 30 cm depth which makes it possible to retrieve snowcover depth using EOS/MODIS data and field data. This paper mainly introduces the principle theory and process to establish a snowcover depth retrieval model within 30 cm using EOS/MODIS visible to short infrared channel data and field measured data. The northeast of Inner Mongolia Autonomous Region （ 111 ° 909′-126°29′E,39°658′ ~53°20′N） was taken as study area. Through three days（Nov. 8th,11th and 25th, 2010） in-situ snowcover depth data of 118 in-situ observations and MODIS L1B images over study area were used for analysis. Based on the characteristics of visible light with high retlectivity from the surface, clouds and snowcov- er, the NDSI（ Normalizing Deference Snow Index）was processed and analyzed to show the snowcover coverage and depth. Through the analysis of snowcover, based on snowcover mapping, the paper dealed with the methods of inte- gration of multi-data, grid calculation and snowcover calculation under the support of GIS spatial analysis tech- niques, produced the snowcover products of observed area, and set up the snowcover depth retrieval model of EOS/ MODIS snowcover remote sensing monitoring and GIS application. This paper also simulated the snowcover depth by Kriging interpolator. The results are as follows： the statistical regression method is effective to improve monito- ring accuracy of snowcover depth, snoweover coverage and other conditions ~ information. The paper compares the field data
出处
《干旱区地理》
CSCD
北大核心
2012年第6期890-896,共7页
Arid Land Geography
基金
国家水体污染控制与治理科技重大专项"海河流域水生态功能三级四级分区研究"(编号:2012ZX07501002002)
