The upper catchment area (UCA) plays a crucial role in hydrology, geomorphology, and glacial lake hazard assessment research. While manual digitization remains the most common method for extracting the UCA in small study areas, it is often labor-intensive and time-consuming. Automated estimation techniques have been utilized at regional scales, but their accuracy remains limited in alpine terrains, particularly around glacial lakes. To address this limitation, we developed an improved approach, implemented in ArcGIS Model Builder, to extract the outlet and UCA of glacial lakes automatically using a digital elevation model (DEM) and a glacial lake dataset from circa 2020. To evaluate the method, we mapped 963 glacial lakes with a total area of 59.75 ± 6.95 km2 in four test watersheds using 18 images from Landsat 8. A total of 264 glacial lakes, varying in size from 0.02 to 5.83 km2, were identified as potentially dangerous for testing based on three filtering criteria. The calculated outlets and UCAs, validated by visually digitized references, indicate that our approach achieves an accuracy of 93.68 ± 2.12 %, significantly improving over previous methods. Our approach is more efficient and accurate than previous methods, offering researchers a reliable tool. Our developed techniques and tools can be broadly applied to glacial lake hazard assessments and related hydrological research.

中文翻译：

---

一种自动计算高山冰川湖出口和上部集水区的新方法


上流域 （UCA） 在水文学、地貌学和冰川湖灾害评估研究中起着至关重要的作用。虽然手动数字化仍然是在小型研究区域中提取 UCA 的最常用方法，但它通常是劳动密集型且耗时的。自动估计技术已在区域尺度上使用，但其准确性在高山地形中仍然有限，尤其是在冰川湖周围。为了解决这一限制，我们开发了一种改进的方法，在 ArcGIS Model Builder 中实施，使用数字高程模型 （DEM） 和大约 2020 年的冰川湖数据集自动提取冰川湖的出口和 UCA。为了评估该方法，我们使用来自 Landsat 8 的 18 张图像绘制了四个测试流域中的 963 个冰川湖，总面积为 59.75 ± 6.95 km2。共有 264 个冰川湖，大小从 0.02 到 5.83 km2 不等，根据三个过滤标准被确定为具有潜在危险的测试。通过视觉数字化参考验证的计算出口和 UCA 表明，我们的方法达到了 93.68 ± 2.12% 的准确率，比以前的方法有了显著提高。我们的方法比以前的方法更高效、更准确，为研究人员提供了一个可靠的工具。我们开发的技术和工具可广泛应用于冰川湖灾害评估和相关水文研究。