| Soil moisture is an important indicator used to monitor land degradation anddrought. It is related to the production, the growing of vegetation and the growthconditions of crop etc. It is also an important parameter in the hydrology, climate,agriculture and ecological fields. The fast and efficient development of remotesensing makes it possible to monitor soil water status in real time and dynamic andassess the soil moisture condition of large area,which overcomes the shortcomings oftraditional methods of soil moisture monitoring.This paper principally studies the RS inversion of soil moisture based on MODISData in the ZhaoKou irrigation. Based on the introduction of soil moisture and severalmethods of remote sensing inversion used to soil moisture, the apparent thermalinertia approach was chosen to inverse the distribution of the soil moisture in the baresoil and the initial stages of crop growth. The vegetation water index method wasselected to calculate the distribution of the soil moisture,In the area covered by highvegetation. The NDVI values was used to divide the surface cover types. In short, wemerge these two ways to inverse the distribution of the soil moisture in the ZhaoKouIrrigation. The main contents:(1) This paper studies the methods to inverse the soil moisture used the remotesensing technology in the domestic and the abroad. The traditional methods can’tmeet the requirements of in the job, the combination of the new method, which canachieve the high efficiency, rapid large-scale monitoring. Then the advantages anddisadvantages of soil moisture remote sensing inversion methods were introduced andwe make the comprehensive comparative analysis in these methods. The apparentthermal inertia method is relatively simple, the inversion accuracy of the soil moistureis relatively high in the bare soil and the soil on the initial growth of the crop isrelatively high. But, this way can’t achieve the high accuracy about the soil moisturein the high vegetation coverage area. The vegetation and water index (VSWI) issuitable for areas of high vegetation coverage. But, in low vegetation coverage areas,it will exaggerate the influence of vegetation. Therefore, this study combined the apparent thermal inertia of water and vegetation indices to invert the soil moisture.(2) The introduction of remote sensing images. Because MODIS Data is fast inmaking imaging, high spectral resolution, easily accessible, wide in the range ofimaging, these advantages is the key why we choose MODIS image to invert soilmoisture. Then the MODIS sensor and images were introduced,finallyMOD021KM, data with the resolution of1000m,was chosen.Then we carried out onthe image data pretreatment using the ENVI remote sensing image processingsoftware, which includes mostly removing noise bands, bow-tie correction andgeometric correction, we clip the image with the vector boundary map of theZhaoKou irrigation generated by the ArcGIS.(3) When the MODIS was used to invert soil moisture, it is necessary todetermine the type of surface covering the study area to improve the accuracy of theinversion, which is namely to determine the area is bare ground/low vegetationcoverage area or high vegetation coverage area. For the former, we use the apparentthermal inertia model, for another we use vegetation water index. Then we use theNDVI (normalized difference vegetation index) to divide cover the types, we take thecorrespondence method to different kinds of vegetation coverage area in order toimprove the invention accuracy of soil moisture using the remote sensing. We invertthe key parameters in the model and generate the corresponding calculated resultsfigure of the remote sensing images. At last the result will be combined with themeasured data to analysis comparatively for getting the soil moisture distributionconditions of the ZhaoKou Irrigation. The results show that soil moisture remotesensing inversion model constructed in the paper improves the accuracy of soilmoisture monitoring in the ZhaoKou Irrigation. |
PDF Full Text Request