| Sudden flood disasters can cause serious damage to agricultural production.Rapidly extracting information on the extent of flooding on agricultural land and capturing the impact of flooding on crops is important for estimating the affected area,promoting post-disaster farmland restoration and providing auxiliary decisionmaking basis for flood prevention and relief departments.Traditional flood information collection methods are time-consuming and difficult to meet the realistic needs of rapid and real-time flood information analysis.With the development of satellite remote sensing technology,the cost of data has been greatly reduced,and remote sensing technology has become a regular means to obtain real-time change monitoring of surface information.And the development of Google Earth Engine platform in recent years has provided a technical platform for processing and analyzing massive remote sensing data rapidly and in large quantities.In this paper,taking the flood events in Henan Province as an example,we propose a method to extract the range information of flood inundation area and farmland affected area according to the change characteristics of radar data and optical data before and after the disaster.The method is established based on the flood event in Henan Province and validated in the flood event in Shanxi Province.The method consists of change detection,threshold extraction and superposition analysis,which weakens the adverse effects of speckle noise of radar data and cloud contamination of optical data on the extraction of farmland inundation area to a certain extent.Finally,based on the results of flood affected area distribution,the growth condition of crops affected by flooding was further analyzed.The main research findings are as follows:(1)This paper provides a low-cost,fast and convenient method for flood information based on Google Earth Engine(GEE)platform with multi-source remote sensing data.Based on the GEE cloud platform,a large amount of remote sensing images and other data resources required for the study can be accessed and collected easily and quickly,and the pre-processing of multi-source image data covering the whole Henan Province can be completed with its fast and high performance cloud computing capability.(2)The results generated by this method have relatively clear boundaries and accurate ranges,and the overall accuracies of flood inundation area and flood affected area extraction in Henan Province are 0.87 and 0.92,respectively.the proposed method combines the advantages of radar and optical remote sensing data to extract flood information,and can derive the specific ranges of flood inundation area and flood affected area on a large spatial scale.It is also verified in the flood events in Shanxi province,where the overall accuracy of flood inundation area extraction is0.85 and the overall accuracy of flood affected area is 0.96.(3)Based on the results of the distribution of flood-affected areas,the extent of autumn crop extinction can be extracted conveniently and quickly;as well as analyzing the effect of late sowing on the growth of winter wheat.2021 Henan 7.20 flood caused an area of 3792.8 km~2 of autumn crop extinction,and the flooded fields were waterlogged resulting in delayed wheat sowing.this paper analyzes the distribution of the extent of late sowing of wheat in the flood-affected areas,and provides scientific guidance for post-disaster wheat plant management and This paper analyzes the distribution of late sowing of wheat in the affected areas to provide scientific guidance for post-disaster wheat plant management and breeding. |
PDF Full Text Request