Remote Sensing of deep convective clouds - A new satellite view

云及其在气候变化中的演变是预测未来气候的一个重要因素，因为它们影响到地球的辐射收支。同时，它们在时间和空间上的可变性是不确定因素的一个来源。它们的全球分布和演化只能通过卫星遥感来观察。因此，标准遥感方法的不确定性以及新方法的发展将是未来十年的一个重要研究课题。在一定程度上研究了低边界层云的(亚传感器分辨率)不均匀对现有方法的影响。然而，还没有关于垂直方向上更复杂的云类型的遥感研究，例如深对流，尽管它们独特的三维结构可能会给标准的遥感方法带来很大的误差。在这个项目中，我们将考察不同的、标准的和新颖的遥感方法在深对流云情况下的潜力和不确定性。将模拟与已知云结构相对应的卫星观测，并将测试模拟数据的遥感技术。为了实现这一目标，将使用来自物理云解析建模(CRM)的复杂云结构。辐射传输的三维模拟将使用最先进的蒙特卡罗程序来模拟辐射场。在这种情况下，蒙特卡罗代码将被优化，以减少3D蒙特卡罗模拟所需的大量计算时间，特别是在红外光谱范围内。

Clouds and there evolution in a changing climate are a factor of major importance in the prediction of future climate due to their impact on the radiation budget of the earth. At the same time their variability in time and space is a source of uncertainties. Their global distribution and evolution can only be observed by satellite remote sensing. Thus, the uncertainties of standard remote sensing methods as well as the development of novel methods will be an important research topic for the next decade. The effects on existing methods due to (sub sensor resolution) cloud inhomogeneity have been investigated to some extent for low boundary layer clouds. However, no studies have yet been conducted concerning remote sensing of vertically more complex cloud types, e.g. deep convection, although their distinct three-dimensional structure is likely to introduce large errors into standard remote sensing methods. In this project the potential and the uncertainties of different, standard and novel, methods of remote sensing will be examined for deep convective cloud cases. The satellite observations corresponding to known cloud structures will be simulated and remote sensing techniques will be tested for the simulated data. To achieve this objective, complex cloud structures from physical cloud resolving modeling (CRM) will be used. A state-of-the-art Monte Carlo code for the three-dimensional simulation of radiative transfer will be used to simulate the radiance field. In this context, the Monte Carlo code will be optimized to reduce the extensive computing time necessary for 3D Monte Carlo simulations, in particular in the infra-red spectral range.