高寒山区积雪状态遥感识别及其衰退过程模拟研究

The distribution range, the state spatial heterogeneity, and its spatial and temporal changes of snow cover in alpine mountains during the ablation period are not only essential for the development of snow and ice remote sensing, but also an important basis for water resources management in the arid region of China. However, there exist problems such as the low accuracy of snow parameters identification in the alpine mountains, the uncertainty of the ablation rules and the response mechanism. Therefore, based on the rapid identification for the aging state of snow using optical remote sensing data, this project aims on to evaluate the recognition accuracy of snow aging index (SAI) and to obtain the topographic adjusted SAI (TASAI) with high precision by means of correcting the influences from the undulating terrain, the zenith effect, and shady errors in the alpine mountains, and to expand the applicability of TASAI in multi-source and multi-resolution remote sensing data. Further, exploring the spatio-temporal recession rules of snow cover in alpine mountainous by precisely capturing the spatio-temporal characteristic nodes. Also, the spatial correlation method is used to measure multiple influencing factors, to select the optimal spatio-temporal econometric model, to establish the spatial-temporal decay response model of snow cover, and then explore the applicability of the models. Finally the understanding and recognition of snow recession process and its complex response mechanism in the alpine are deeply studied. The series of key technologies and related conclusions proposed by this project can provide theoretical basis for inversion and monitoring of snow during the period of ablation in cold alpine regions and provide technical support and scientific reference for basin water resources management with high theoretical and practical application value.

高寒山区消融期积雪分布范围、状态空间异质性及其时空变化等信息不仅是冰雪遥感发展的需要，也是我国干旱区水资源管理的重要依据。针对目前存在的山区积雪参数识别精度低，衰退规律和响应机制不明等问题，本项目拟以光学遥感数据实现积雪新老状态的快速表征为基础，评价老化指数的识别结果，通过修正起伏地形、方位角和阴影等造成的误差，构建具有较高识别精度的地形调节老化指数，并拓展其在多源、多分辨率遥感数据上的适用性；精准捕捉可表征积雪衰退特征的时空节点，探求山区积雪时空衰退规律；利用空间关联法度量多重影响因子，筛选最优时空计量模型，构建积雪时空衰退响应模型，探讨模型的适用性，深入对高寒山区积雪衰退过程及其复杂响应机制的理解和认知。本项目提出的系列关键技术及相关结论可为高寒山区消融期积雪反演与监测提供理论依据，为流域水资源管理提供技术支持和科学参考，具有很高的理论与实际应用价值。

高寒山区消融期积雪分布范围、积雪衰退状态的空间异质性及其时空变化等信息不仅是冰雪遥感发展的需要，也是我国干旱区水资源管理的重要依据。针对目前基于光学卫星遥感数据在山区积雪状态参数识别精度低，积雪衰退规律和影响机理不明等问题，本项目以天山中段为研究区，利用光学遥感数据，构建了综合多重老化指标的积雪衰退状态快速表征的技术方法，通过修正起伏地形、阴影等造成的误差，拓展其在高寒山区及多源、多分辨率遥感数据上的适用性。综合积雪衰退状态信息与影响因子（海拔、坡度、坡向、辐射、风速、气温、地表温度、河谷距离、降水、粗糙度）数据，建立了长时间序列积雪衰退状态参数数据集以及影响因子数据库。利用地理探测器和时空面板模型等方法，在多重时空尺度下判定了引起积雪衰退的关键影响因子，分析了多因子的交互作用、多因子影响差异以及关键因子区间对积雪衰退的影响差异等，实现了对山区积雪时空衰退规律的深入探索。进一步，利用多尺度的流域及栅格划分方法，进行积雪衰退的时空同质区的划分，界定进行山区积雪衰退过程模拟的最优尺度，选择最优空间权重矩阵，构建了基于空间面板模型的积雪衰退时空变化模型，实现了高寒山区积雪衰退过程的模型与预测。本项目提出的系列关键技术及相关结论可为高寒山区消融期积雪信息反演与监测预报提供理论依据，为流域融雪径流模拟与预测、积雪水资源利用与管理提供技术支持和科学参考。