面向对象的青藏高原积雪累积与消融过程时空建模研究

Snow accumulation and depletion is a complex and dynamic geophysical phenomenon. In previous studies, pixel-based methods are usually employed to monitor and to analyze the spatio-temporal variation of seasonal snow. However, these pixel-based methods will induce bias in the analysis results of seasonal duration. They also cannot reveal the spatio-temporal evolutionary relationships among snow. This study proposes an object-based spatio-temporal model for analyzing snow accumulation and depletion processes. This model organizes data into snow zones, snow sequences, and snow processes based on the levels of spatio-temporal aggregation of snow. This model is not only be able to describe the snow accumulation and depletion process, but also to provide the spatio-temporal evolution and moving path. This model is conducted on Tibet Plateau, where snow is discontinuously distributed due to the complexed topography. Also, to over the limitation due to the low spatial resolution of snow depth from passive microwave data, the visible remotely sensed data and multi-topographic variables are incorporated in this model by using a non-linearly downscaling method. This study is capable of providing valuable model and method for the snow analysis on Tibet Plateau. And it is of significant importance for the sustainable use of water resources and the study of climate change.

积雪的累积与消融是一个复杂和动态的地理现象。之前的研究通常采用基于像元的统计方法从时间序列积雪遥感影像中获取积雪的时空变化特征。但是这种基于像元的方法会造成季节性时长信息偏差,不能反映积雪之间的时空演化关系。本项目提出一种基于面向对象的积雪累积与消融时空过程模型，这个模型根据积雪在时空由低到高的聚合等级，依次分为了积雪区、积雪序列和积雪过程三个层次，不仅能表达积雪累积和消融的时空过程，还能揭示积雪的时空演化关系和转移路径。项目将此模型用于了青藏高原。特别地，针对青藏高原复杂地形对积雪非连续分布的影响，以及被动微波雪深数据空间分辨率较低的局限性，本项目融合可见光遥感数据和多个地形因子，对被动微波雪深进行了非线性降尺度，提高了模型输入参数的质量。本项目能够为青藏高原积雪时空变化分析提供重要的模型和技术支撑，对于青藏高原水资源可持续利用和气候变化研究等方面具有重要的意义。

积雪的累积与消融是一个复杂和动态的地理现象。为了支撑积雪累积与消融时空过程分析，需要准确获取积雪时空分布特征，以及发展可靠的积雪累积与消融时空过程模型。本项目融合多源遥感数据，提出了基于HMRF的时空建模技术，填补了原始MODIS积雪产品中由于受到经常性云层遮挡所引起的大量数据空缺，能够提供精度和可靠性都显著改进的时空连续的积雪产品；开发了一种基于夜间灯光遥感数据的积雪识别最小误差阈值算法，该算法生成的积雪产品可作为日间积雪信息的重要补充和完善；在此基础上结合多因子开展了非线性雪深降尺度算法，为青藏高原积雪时空动态分析提供了高质量数据源。进一步构建了面向对象的积雪累积与消融过程时空模型。这个模型根据积雪在时空由低到高的聚合等级，依次分为了积雪、积雪序和积雪过程三个层次，不仅能表达积雪在其生命周期中累积和消融的时空过程，还能够揭示积雪内部和积雪之间的时空演化关系。项目将模型应用于了青藏高原地区，研究成果能够为青藏高原积雪时空变化分析提供重要的模型和技术支撑，对于青藏高原水资源可持续利用和气候变化研究等方面具有重要的意义。