Modis Data Products, Four-dimensional Assimilation Of The Prototype System Design And Implementation Of The Ensemble Kalman Filter

MODIS (Moderate Resolution Imaging Spectro-radiometer) is an important remote sensing sensor on the American EOS(Earth observing system series of satellites) series satellites. MODIS remote sensing data is a new generation source of satellite remote sensing. It has the extensively applied foreground in many researches such as ecology study,the environment monitor, global weather variety and the agriculture resources inquisition etc. However, the quantity of the MODIS data product is not good enough. Using these data to the land ecosystem will produce a great deal of indetermination, and some results will even be false.On the basis of reviewing existing filter methods and assimilation algorithm (including S-G filter, ensemble Kalman filter, wavelet transformation, asymmetric Gauss transformation, etc), we assimilate MODIS data products with 4 dimensions, and integrate these kinds of assimilation algorithm into the 4 dimensions assilmilation prototype system, besides attached other functions, such as algorithm appraisal and format transformation and so on.The paper detailed on 4 dimensions assimilation of ensemble Kalman filter on MODIS data products and 4 dimensions assimilation prototype system on MODIS data products.So we design a 4 dimensions assimilation prototype system on MODIS data products, composed of 4D assimilation algorithms and 4D assimilation algorithm appraisals. This paper is detailed about architectural design of 4 dimensions assimilation prototype system on MODIS data products.We apply ensemble Kalman filter algorithm to the index of MODIS(LAI). It makes full use of the data information of MODIS(LAI), and obtains better qualified data. Based on the result, we analyze drawbacks of ensemble Kalman filter algorithm, and improve it, and appraise it at the last section of this paper.