Research On Information Extraction Of Soil Parent Material Based On Multiple Data Sources

The soil is an “independent natural body” developed on the basis of the parent material.Therefore,the parent material is particularly important to the formation and development of soil,especially in the small area,the type of soil parent material is an important factor which determines the soil properties and the soil type.However,there is no clear map of the parent material currently,at the meanwhile,lack of the methods of obtaining parental information is another problem.Therefore,it is necessary to research the ways and methods of obtaining parent material information.This paper takes Ling yuan of Liaoning Province as the research area,selects geological map,soil map,DEM data and Google Earth remote sensing image as the data source.By establishing the correspondence between the legend information in the geological map and the parent material type,the lithology and material composition information contained in the common geological map is extracted to obtain the partial parent material type and its spatial distribution.According to the correspondence between soil classification and parent material,parent material information and its spatial distribution were extracted through the correspondence between soil type and parent material type in soil map.The DEM elevation data is used to obtain the slope information of the research area and the ENVI software is used to calculate the normalized vegetation index(NDVI)of the study area.Finally,the decision tree is established in the ENVI software to calculate the parent material information by using the automatic classification method.Using Google Earth remote sensing image data,the correspondence between remote sensing image and parent material type is established,and the visual interpretation method is used to discriminate the parent material.Finally,the information on the parent material obtained by using the above various data sources is synthesized,and the method of obtaining the parent material is summarized.The results show that the visual interpretation of Google Earth remote sensing image data can make full use of the multiple information provided by remote sensing data,such as texture and landform,so it is more accurate to acquire the parent materials of flood,residual and slope,and more convenient and accurate to acquire the parent materials of loess and red clay by using soil map.By using the information of slope,elevation and normalized vegetation index in the study area,the decision tree is established,and the alluvial parent material information in the study area can be obtained automatically,accurately and comprehensively by computer in ENVI software.A single data source can only obtain part of the parent material information,and the range and accuracy of obtaining parent material information can be improved by utilizing multiple data sources.