云通过对地-气系统能量平衡和水分循环的调节，在气候变化中起到了至关重要的作用。在全球CO2浓度增加，气候增暖的背景下，由于云特性变化及其辐射反馈过程难以在模式中合理表征，造成气候模拟预测依然存在较大的不确定性。因此，在全球范围内开展准确的云观测，对理解云物理过程，约束和改进模式云参数化方案，合理预测气候变化有重要意义。卫星遥感是在全球尺度上获取云观测的唯一有效手段，CloudSat卫星搭载的94GHz雷达可以获取云特性的三维分布，被广泛应用于云的相关研究。然而CloudSat目前的云检测产品对弱回波信号存在误检率明显较高的问题，导致弱信号云无法被有效使用。本项目计划利用申请人提出的双边滤波方案，对CloudSat云检测算法进行改进，开发一套能显著降低弱信号误检和漏检率的新方法，在此基础上完成新的云检测产品，对不同气候模式模拟的云特性进行对比评估，为约束和改进模式中云的合理表征提供重要依据。

Clouds play a crucial role in climate change through the regulation of the Earth's energy balance and the water cycle. As the global warming under the increase of CO2 concentration, the climate projection still has large uncertainty due to the difficult representations of cloud properties and their feedback in climate models. Therefore, long-term accurate cloud observations are important for understanding the cloud physical process, constraining and improving the model cloud parameterization. Satellite remote sensing is a unique in obtaining cloud observations on a global scale. The CloudSat satellites 94GHz radar can obtain three-dimensional distributions of cloud properties and is widely used in cloud-related research. However, current CloudSat detections have a problem that the false positive rate of weak echo signal is too high, leading to the useless of these weak signal cloud productions. In this project, we plan to use the bilateral filtering scheme, which was proposed by the applicant, to improve the CloudSat cloud detection algorithm and develop a new method that can significantly reduce the false positives and false negatives of the weak signals. A new cloud mask product will be completed based on the improved method. The distributions of cloud properties produced by CMIP5 models will be compared and evaluated against the new CloudSat product to constrain and improve the parameterizations of clouds in climate models.