| Remote sensing technology and spatial analysis technology are the basis of geographic informatization.The efficient and timely acquisition of water resources dynamic information,the establishment of scientific and accurate water resources supervision and management mechanism,the reasonable solution of water resources in irrigation,flood control and other aspects of the problem,and then to achieve the sustainable use of water resources are inseparable from the support of remote sensing and GIS technology.This paper took Suining City and three images(Landsat-8,Sentinel-2 and GF-1)as the research area and data sources respectively.The object-oriented classification method was used to study water extraction.Based on that,the spatial topography distribution characteristics and dynamic change of water in Suining City were analyzed.The main research contents and results of this paper were displayed as follows:(1)From aspects in terms of spectral features,texture features,geometric features and band combination features,various surface features acquired from different remote sensing images had been examined.Additionally,water extraction rule set was built by using e Cognition intelligent image analysis software to realize water extraction in the study area.The accuracy of the water extraction results of the three images was evaluated by the confusion matrix and Kappa coefficient.The results showed that the overall accuracy of water extraction results from Landsat-8remote sensing images was 85.62%.And the overall accuracy of water extraction from Sentinel 2 remote sensing image was 88.05%.While,the overall accuracy of water extraction from GF-1 remote sensing image was 91.89%.The water extraction result of GF-1 remote sensing image with spatial resolution of 2 meters was the best.(2)After the comprehensive analysis of the water extraction results of three remote sensing images,it was proved that three kinds of remote sensing images all could extract obvious water well,while for the extraction of linear water and small patch-like water,the use of GF-1 data had obvious advantages.Sentinel-2 and Landsat-8 remote sensing images were relatively convenient and had abundant band information when acquiring data and processing and,which could perform data characteristics to meet research needs in applied researches with longer timescales and wider range.(3)In order to explore water distribution in different districts of the study area,elevation,slope and topographic potential index were examined by spatial analysis function of GIS.The results showed that in terms of elevation,the water in the study area mainly distributed within that range from 253 m to 383 m.Also,the water mainly distributed on the gentle slope with less than 15°.The topographic potential index of Suining city shifted between 0.51 and 2.83.With the topographic index increasing,the water area showed a trend of decrease after growing first.When the topographic index was larger than 2.36,which is located after the fifth grade of topographic index,the water distribution in the study area was relatively dispersed.Therefore,it should be crucial to carry out the construction and renovation of small water conservancy projects in the areas with high topography and insufficient water.(4)Compared with the water extraction results of Fujiang(Suining Section)in2002 and 2018,the water area changed from 122.70 hm~2 to 1825.41 hm~2,with a net increase of 1702.71 hm~2.After 2002,the development of water conservancy construction in Suining city had been accelerated.The data showed that scientific and rational development of water resources had improved local ecological environment of Suining city,living security of residents as well as local economic sustainable development.Meanwhile,engineering water shortage of the Suining city and seasonal water shortage caused by the uneven distribution of water resources could be alleviated. |
PDF Full Text Request