Atmospheric Roughness Inversion Based On Multi-source Remote Sensing Data

Tibet Plateau, as an essential indicator of Asia climate change, has an important effect in global environment and climate research, and plays an essential role in numerical simulation of the atmospheric models with various scales. As its indispensable part, atmospheric roughness is usually used in atmospheric modeling; the calculation of its value plays an important role in the correct construction of the entire atmospheric model. However, roughness algorithm used in past shows its limitations in the experiment data acquiring, algorithm implementation and inversion of process parameters, as well as application range. So some new technology and data are developing in the calculation of roughness.LiDAR and remote sensing techniques have an great development in recent years. As the new data acquisition techniques, LiDAR can quickly obtain high-resolution digital terrain model and three-dimensional information of ground features, what's for, it can obtain the vertical structure and morphology of ground objects. Remote sensing has its advantages such as simultaneous observation, timeliness and other characteristics, and widely used in the monitoring of surface characteristics change. These data were applied to the roughness calculation in order to get a better solution of roughness inversion problem.This paper select the representative area in Tibet Plateau such as Qinghai Province, western Sichuan and Naqu station as the study area. Using remote sensing image data, laser radar data and wind observation data, the roughness algorithms based on various remote sensing data are developed to extract the roughness. Of course, these parameters which used in the inversion process are extracted by the RS technology, and then calculate the roughness. These roughness results which extract by different algorithms-profile method, the dimensionless wind speed method and the remote sensing method, are compared and analyst with the factor. The results show that the remote sensing algorithms are good method, it overcomes algorithm's limits.Use of remote sensing data (MODIS), laser lidar data (GALS) and wind data in the Qinghai and western Sichuan area, the remote sensing parameterization method three is executing to obtain roughness. During extraction process, the dimensionless constant, vegetation coverage, canopy height and wind data which are related to the vertical wind characteristics are estimated to obtain roughness and displacement height of the Qinghai and western Sichuan. The analyses of results show that the remote sensing method three can describe ground features, its results have consistent with surface land types distribution. In vegetation covered area, this method has a better applicability, while in the bare and desert surface this method has its limits.Airborne laser radar data (LiDAR) combine with multi-layer wind data to extract the roughness of the Naqu. The high-density point cloud data can develops high-precision DEM, DSM, characteristic parameters of roughness element and vegetation. What's more, the regional wind field (include wind speed, wind direction) is simulate with DSM and wind data by the software WINDSTATION. In the last, the final calculation results of the August2011in Naqu area are obtained and analyzed. The result shows the remote sensing methods can resolve these questions which encountered by the method used in the past and it is successful.