Modis Lai Product Authenticity Verification Space Dimension Conversion Method Research

With the development of remote sensing technology, a large number of different resolution remote sensing data provide a sufficient source of information for the Earth’s surface and even a certain depth of quantitative remote sensing. Leaf area index (LAI) is an important surface parameter, which quantifies the unit area of leaf surface area and is closely related to net primary productivity. And the MODIS LAI products in the crop growth monitoring and yield estimation, simulation of the Earth’s surface processes, global change research has wide application areas, so the authenticity of the global MODIS LAI product validation is very important.In this paper, the study area is the oasis area in the Heihe River Basin of Northwest of China, the experimental data includes ground data, TM image data of high resolution and the MODIS LAI product data of low resolution, which is used to establish the empirical model of the LAI inversion. This paper proposes a new method of spatial scale conversion which is combining with the MODIS product geometry information and the decomposition based on NDVI pixel with the PSF, and is compared with the traditional method. The main results are as follows:(1) The TM images data of the study area has a high correlation between use of the empirical formula for the LAI inversion of surface cover class, which is mainly due to the empirical formula of the model is to establish ground truth value based on the region’s land cover class. Although it is not as good as the physical model which has a wide range of applicability of the space, is fully applicable in the region, and the model is simple and highly targeted.(2) In the experience inversion model, LAI-NDVI linear fitting model for the linear correlation of data is very high in the region, which fits the requirements of the inversion accuracy and has a good agreement with the method of the spatial scale conversion.(3) Directly using the ground measured LAI values or30m TM image inversion of LAI to validate the MODIS of LAI products has a lot of large errors, the main reason is the influence of surface heterogeneity and the use of different resolution data for mutual authentication effected by the spatial scale.(4) Because of the TM pixel and the MODIS pixel sensor corresponding with ground instantaneous field of view of the irregularities, if we take use of the traditional scale method to validate the MODIS LAI product and without considering the influence of geometric factors, which will lead to the lower the credibility of the verification results.(5) This article uses a combination of spatial scales based on the PSF of the pixel decomposition of large-scale NDVI and MODIS imaging geometric principles to convert research methods, compared with traditional methods, it is the pixel correspondence, the autocorrelation of the image and surface heterogeneity of both the combination of validation of the MODIS products of LAI, the experimental results significantly better than traditional spatial scale conversion method, thereby improving the accuracy and credibility of the MODIS of LAI product validation.