Updating the New Zealand Glacier Inventory

拟议研究项目的目的是更新新西兰的冰川清单。新西兰唯一完整的冰川清单始于1978年，不再代表目前的情况。无论从科学角度还是从社会经济角度来看，更新清单都是至关重要的。它将作为模拟未来冰川行为、水资源和未来水力发电的基础，此外，还将作为探测潜在自然灾害来源的基础。新的清单将通过使用遥感和(半自动)程序进行汇编。它将基于2010年至2015年的ASTER传感器图像和15米分辨率的国家数字高程模型。自动编辑雪和干净或略脏的冰川冰是一个既定的程序。为此，将使用ASTER传感器频段ASTER 3/ASTER 4的频段比率方法，包括某个阈值。然而，绘制新西兰常见的被碎片覆盖的冰川的地图相当具有挑战性。为了半自动绘制这些冰川的地图，将进行基于像素的多光谱分类，结合数字地形分析和邻域关系分析，并通过应用基于对象的图像分析来不断改进。此外，水体，特别是冰上湖泊，将自动汇编，冰川长度将半自动。所有附有进一步属性的数据将提交给国际冰川组织和GLIMS平台，以便纳入其开放源码数据库。最后，将对旧的和更新的冰川清单进行比较，以发现潜在的冰川变化。

Aim of the proposed research project is to update the New Zealand glacier inventory. The solely complete glacier inventory for New Zealand dates from the year 1978 and does no longer represent the current situation. From a scientific as well as from a socio-economic point of view, an updated inventory is essential. It will serve as basis for modelling of future glacier behavior, water resources, and future generation of hydroelectric power and, additionally, for the detection of potential sources for natural hazards. The new inventory will be compiled by the use of remote sensing and (semi-)automatic processes. It will be based on ASTER sensor images from 2010 to 2015 and a national digital elevation model with a 15 m resolution. The automatic compilation of snow and clean or slightly dirty glacier ice is an established procedure. A band ratio approach from the ASTER sensor bands ASTER 3 / ASTER 4 including a certain threshold will be used for this purpose. Mapping of debris-covered glaciers common in New Zealand is, however, rather challenging. For a semi-automatic mapping of these glaciers, a pixel-based multispectral classification combined with a digital terrain analysis and analysis of neighborhood relations will be conducted, consecutively improved by application of an object-based image analysis. Additionally, water bodies, especially supraglacial lakes, will be automatically compiled, glacier length semi-automatically. All data accompanied by further attributes will be submitted to the international glacier organization and platform GLIMS for to be incorporated into their open-source database. Finally, a comparison between the old and the updated glacier inventory will be conducted to detect potential glacier changes.