In aeromagnetic survey or ground magnetic survey(whether in total fieldsurvey or vertical gradient survey), full sampling seems extremely importantfor further image processing and image enhancement processing as well asinterpretation of magnetic anomalies. The reason is that all the data obtainedare discrete data set of continuous anomalous field. By applying typical two-dimensional dyke model to conduct theoretical calculation, we used the dis-tance between the field source and the probe to replace buried depth of fieldsource. Based on a series of theoretical calculations and analyses, the authorholds that in large-scale high-precision aeromagnetic survey, the samplinginterval between stations along the survey line should be shorter than orequal to half of the f;ying height, whereas the distance between the surveylines should not exceed the f;ying height.Two application examples are given in this paper.