The most prevalent global surface models are derived from space-based technologies. The spatial and temporal coverage of radar and imaging systems have significant advantage over other sensors and platforms. However, these systems are often are challenged in certain environmental conditions to produce accurate elevations as compared to what might achieved with laser altimetry. Radar derived DEM (Digital Elevation Model) elevation accuracy is often less in vegetated regions, as the wavelengths associated with radar mapping missions do not fully penetrate the canopy. Elevation errors are also substantial over dynamic topography. Correction models using airborne lidar reference datasets can be effective for localized studies to improve the DEM but often the data is not readily available or seasonally/temporally irrelevant. This work presents an automated method for correcting digital terrain models derived from the Shuttle Radar Topographic Mission (SRTM) elevation products using NASA's newest Earth observing laser altimeter, the Ice, Cloud and Land Elevation Satellite-2 (ICESat-2). ICESat-2 elevations, in concert with Landsat 8 (global imagery), create a model based correction strategy for the SRTM derived elevations using geographically correlated canopy cover and surface slope information. The results are validated at the study site using high-resolution, high fidelity airborne lidar datasets as a reference surface. The application of the correction model on the radar measurements results in nearly 50% improvement in elevation accuracy for this region. Additionally this established proof of concept provides a starting point for further research in how this method, with ICESat-2 data, can be extended to other environmental regions, other radar or image derived elevation products and inform future techniques for similar application at the global scale.

最流行的全球地表模型源自天基技术。雷达和成像系统的空间和时间覆盖范围比其他传感器和平台具有显着优势。然而，这些系统在某些环境条件下经常面临挑战，以产生与激光测高相比可能实现的准确高度。雷达衍生的 DEM（数字高程模型）高程精度在植被区域通常较低，因为与雷达测绘任务相关的波长不能完全穿透树冠。动态地形上的高程误差也很大。使用机载激光雷达参考数据集的校正模型对于改进 DEM 的局部研究可能是有效的，但数据通常不易获得或与季节/时间无关。这项工作提出了一种使用 NASA 最新的地球观测激光高度计冰、云和陆地高程卫星 2 (ICESat-2) 来校正从航天飞机雷达地形任务 (SRTM) 高程产品衍生的数字地形模型的自动化方法。ICESat-2 高程与 Landsat 8（全球影像）配合使用地理相关的冠层覆盖和地表坡度信息，为 SRTM 派生的高程创建基于模型的校正策略。结果在研究现场使用高分辨率、高保真机载激光雷达数据集作为参考面进行了验证。将校正模型应用于雷达测量结果使该区域的高程精度提高了近 50%。