Automated, High Resolution Terrain Generation for XSEDE

科学界已经获得了大量高分辨率卫星图像，这有可能改变地球科学研究。虽然这些数据是免费提供的，但它们的形式不是普通用户可以访问的。该项目将使用高性能计算设施和新软件将图像转换为研究人员和公众更容易访问的格式。通过NSF赞助的研究中心获得高分辨率(亚米级)商业图像正在推动地球观测的革命，重新定义可以从太空测量的东西。然而，目前还没有有效的系统来从海量的存档图像中制作和分发数字高程模型(DEM)，这阻碍了极地研究界的利用。该项目将利用XSEDE高性能计算框架和美国国家科学基金会资助的极地地理空间中心(PGC)数据服务，建立按需制作和分发极地DEM的服务。PGC拥有覆盖全球的大量快速增长的亚米级世界观立体卫星图像档案，从中可以免费提取和公开分发高分辨率数字高程模型。免费按需生成高分辨率和精确的数字高程模型将为广泛的科学、工程和规划活动提供强大的新能力。该项目将支持高性能科学计算方面的研究生和本科生教育，并将创建具有高教育价值的地球科学数据集。

A significant body of high resolution satellite imagery has been made available to the scientific community, which has the potential to transform Earth science research. While these data are freely available, they are not in a form that an average user can access. This project will use high performance computing facilities and new software to convert images to a format that is more easily accessible to researchers and the public. Access to high resolution (sub-meter scale) commercial imagery through NSF-sponsored research centers is driving a revolution in Earth observation, redefining what can be measured from space. Currently, however, there is no efficient system for producing and distributing Digital Elevation Models (DEMs) from the vast amount of archived imagery, and this impedes utilization by the polar research community. This project will establish a service for on-demand polar DEM production and distribution utilizing the XSEDE High Performance Computing framework and the NSF-funded Polar Geospatial Center (PGC) data services. The PGC holds a large and rapidly-growing archive of sub-meter WorldView stereoscopic satellite imagery covering the globe from which high resolution DEMs can be extracted and openly distributed at no cost. Enabling on-demand generation of high resolution and precise DEMs at no cost will offer a powerful new capability for a wide spectrum of scientific, engineering and planning activities. This project will support both graduate and undergraduate education in high performance scientific computing and will create Earth science datasets of high educational value.