【摘要】 鄱阳湖作为典型的吞吐性通江湖泊,获取其高动态的水面覆盖变化格局信息具有重要意义.为适应鄱阳湖区水体覆盖类型复杂的特点,尝试利用大气校正有效剔除影像信号中的瑞利散射,并设计了近红外反射率(ρrc(NIR))与绿光/近红外反射率比(ρrc(Green)/ρrc(NIR))相结合的水体提取方法,有效地改善了影像水体的提取精度.利用2008年9月～2011年12月151景无云HJ-1A/1B卫星CCD传感器数据,通过改进后的新方法高精度地提取了水体覆盖信息.在此基础上,利用时空分析方法动态分析了鄱阳湖近年来水面变化的季节性特征与年际变化特征.研究结果能够为湖区旱涝灾害监测、调洪等提供科学依据.更多还原

【Abstract】 Poyang Lake is the largest freshwater lake of China.It always shows a dramatic variation of the water area in different seasons and years.The tempo-spatial dynamic hydrologic characteristics of Poyang Lake make it exist like a river in dry season and as a lake in wet season.The monitoring results from remotely sensed data with lower temporal resolution are hard to explain the season and annual inundation pattern of the lake.Several water extraction methods of HJ-1A/1B satellite CCD images were compared.The methods integrated Rayleigh scattering correction reflectance and get the best results with 97% overall accuracy.It proves that the atmospheric correction process could effectively improve the extraction results of complicated water areas.The 151 scenes HJ CCD data were collected from Sep 2008-Dec 2011,and were used to extract the water body through the novel methods introduced in the study.The tempo-spatial inundation pattern of Poyang Lake were analyzed based on the monthly average inundation area.This paper utilized the satellite data with high temporal resolution to dynamic monitor the fluctuation of the inundation area of Poyang Lake,and explain the reason of tempo-spatial inundation pattern occurs in Poyang Lake in the past three years.更多还原