Study on the construction of multi-dimensional Remote Sensing feature space for hydrological drought

35th International Symposium on Remote Sensing of Environment

地球科学;生态环境科学

Xiang, Daxiang^1,2 ; Tan, Debao^1 ; Cui, Yuanlai^2 ; Wen, Xiongfei^1 ; Shen, Shaohong^1 ; Li, Zhe^1

Changjiang River Scientific Research Institute, Changjiang River Water Resources Commission, No.23, Huangpu Street, Wuhan, China^1

College of Water Resources and Hydroelectric Engineering, Wuhan University, No.129, Luoyu Road, Wuhan, China^2

关键词: Collaborative model;    Field observation data;    Hydrological droughts;    Remote sensing imagery;    Spectral characterization;    Surface temperatures;    Temperature differences;    Vegetation productivity;   

Others  :  https://iopscience.iop.org/article/10.1088/1755-1315/17/1/012095/pdf
DOI  :  10.1088/1755-1315/17/1/012095

学科分类：环境科学（综合）

来源: IOP

【 摘 要 】

Hydrological drought refers to an abnormal water shortage caused by precipitation and surface water shortages or a groundwater imbalance. Hydrological drought is reflected in a drop of surface water, decrease of vegetation productivity, increase of temperature difference between day and night and so on. Remote sensing permits the observation of surface water, vegetation, temperature and other information from a macro perspective. This paper analyzes the correlation relationship and differentiation of both remote sensing and surface measured indicators, after the selection and extraction a series of representative remote sensing characteristic parameters according to the spectral characterization of surface features in remote sensing imagery, such as vegetation index, surface temperature and surface water from HJ-1A/B CCD/IRS data. Finally, multi-dimensional remote sensing features such as hydrological drought are built on a intelligent collaborative model. Further, for the Dong-ting lake area, two drought events are analyzed for verification of multi-dimensional features using remote sensing data with different phases and field observation data. The experiments results proved that multi-dimensional features are a good method for hydrological drought.